Neurologic Clinics Pediatric Neurology Part I

Neurol Clin N Am 20 (2002) xixii
Preface
Pediatric Neurology, Part I
John B. Bodensteiner, MD James F. Bale Jr., MD Guest Editors
Faced with editing an issue of Neurologic Clinics on Pediatric Neurology and charged with the task of covering topics in which there have been signicant advances since the last Clinics issue on Pediatric Neurology (1990), we quickly realized that the advances in Pediatric Neurology have been so substantial that there were few areas that could be excluded. With this in mind we approached the publisher with the concept of preparing either a two-part issue or two separate issues devoted to Pediatric Neurology. Fortunately, it was apparent to all that this was the only reasonable approach, and permission was given to produce two issues, one in 2002 and a second in 2003. Even so, we had to carefully select the topics to be presented. As a result we have attempted to limit the contributions to those in which there have been the greatest or most important changes over the last decade. The rst of our issues includes topics in which there have been revolutionary changes in our understanding. Foremost among these are the genetics of brain development, the neurocutaneous syndromes, and Rett syndrome. Also included in the rst issue are articles describing infections of the nervous system. These consist of AIDS in children, a topic which has changed remarkably in the last decade, meningitis and encephalitis in the post HiB immunization era, and an expanded view of congenital infections. Neuroradiology has continued to advance in the last decade, and the newer technologies have changed the ways we evaluate and manage our patients. Advanced MRI techniques, a signicant part of the new developments in pediatric neuroradiology, represent important additions, not only improving our ability to evaluate our patients clinically but also providing an unique set of techniques that enhance our understanding of basic pathogenetic mechanisms.
0733-8619/02/$ - see front matter 2002, Elsevier Science (USA). All rights reserved. PII: S 0 7 3 3 - 8 6 1 9 ( 0 2 ) 0 0 0 2 7 - 0
xii
J.B. Bodensteiner, J.F. Bale / Neurol Clin N Am 20 (2002) xixii
The rst volume also includes updates regarding childhood stroke and epilepsy. Cerebral vascular disease in children is remarkably common particularly when one considers that most cerebral palsy results from cerebral vascular disease in one form or another. Our understanding of the causes and prevention of childhood stroke has changed considerably in the last decade. Equally important advances have occurred in the evaluation and management of childhood epilepsy. New anticonvulsants, non-medicinal therapies, and surgical management of intractable seizures have revolutionized the approaches to children with epilepsy. The nal articles in this issue address movement disorders in children and a new class of conditions called neurotransmitter disorders. Movement disorders continue to cause concern. We do not know basic pathogenetic mechanisms, our therapies have limitations, and the co-morbid conditions can be dicult to address, as well. The inuence of streptococcal infection is a fascinating, novel aspect of the pathogenesis of movement disorders as well as a variety of psychiatric conditions. Knowledge regarding neurotransmitter disorders, a group of rare but important disorders, has evolved in the last decade and reects a new way of viewing these enigmatic conditions. Topics that will be included in the second volume include neuromuscular disease, autism, learning disabilities, ADHD, neuro-oncology, neurorehabilitation, sleep disorders in childhood, depression in children, neuroimmunology, headache, and selected neurometabolic conditions. Due to space limitations an the desire to update areas with new developments, there will be many topics that cannot be included. We believe, however, that two volumes are better than one; having two volumes allows us to cover many areas in which the new developments have most important implications for the practice of neurology and pediatric neurology. We hope the readers will nd these articles as entertaining and informative as we have. John B. Bodensteiner, MD Guest Editor Barrow Neurological Institute Pediatric Neurology 500 W. Thomas Road, Suite 930 Phoenix, Arizona 85031 James F. Bale Jr., MD Guest Editor Primary Childrens Medical Center Pediatric Neurology 100 North Medical Center Drive Salt Lake City, Utah 84113
Neurol Clin N Am 20 (2002) 917939
Brain development and the genetics of brain development

Gary D. Clark, MD
Departments of Pediatrics, Neurology, and Neuroscience, Cain Foundation Laboratories, 6621 Fannin Street, MC 3-6365, Houston, TX 77030-2399, USA
The brain is a seemingly nonsegmented organ that is, however, formed in a segmented fashion by the overlap of genes that dene anatomic and probably functional components of the brain. Other genes and their encoded proteins regulate the processes of cell proliferation and migration; many of these genes have been identied based upon discoveries of human and mouse disease-causing genes. Human brain developmental disorders represent clinical challenges for the diagnosing clinician as well as for the treating physician. Some disorders represent well-dened clinical and genetic entities for which there are specic tests; others have ill-dened genetic causes, while others can have both genetic and destructive causes. In most cases the recognition of a disorder of brain development portends certain developmental disabilities and often seizure disorders that can be very dicult to treat. In addition, it now bears upon the treating physician to recognize the genetic causes, and to properly advise patients and their families of the risks of recurrence or refer them to the proper specialist who can do so. The genetics of some of these disorders are not all well dened at present, and the recognition of some disorders is variable; what is known is presented herein. The genetics and signaling utilized in brain development is briey reviewed to provide the framework for the understanding of human brain developmental disorders. The well-dened genetic disorders of brain development are discussed, and a brief suggested algorithm for evaluation and for counseling of patients is provided.
E-mail address: gclark@bcm.tmc.edu (G.D. Clark). 0733-8619/02/$ - see front matter 2002, Elsevier Science (USA). All rights reserved. PII: S 0 7 3 3 - 8 6 1 9 ( 0 2 ) 0 0 0 2 4 - 5
918
G.D. Clark / Neurol Clin N Am 20 (2002) 917939
Brain development Overview General mechanisms tend to recur in all phases of brain development, and these include induction, cell proliferation, cell fate determination (dierentiation), cell process formation and targeting (synapse formation), and cell movement (migration). Induction is the process by which one group of cells or tissue determines the fate of another by the release of soluble factors or inducers. Cell fate or dierentiation is dependent upon this process of induction, and probably can best be understood as the initiation of a genetic program by the recognition of an inducing molecule and/or expression of a transcriptional regulator. In general, it is rare that a cell in the nervous system is born and dierentiates in the same location that it nally resides. Rather, cells migrate over long distances to reach their nal locations. Similarly, cells in the nervous system must extend processes over long distances to reach their synaptic targets. Neural tube formation The human brain is formed from the neuroectoderm, a placode of cells that are induced to dierentiate from the surrounding ectoderm by the presence of the notochord at about 18 days gestation. Candidate inducing factors include the retinoids, follistatin, and Noggin [14]. The neuroectoderm develops folds in the lateral aspects that begin to approximate in the region of the future medulla and fuse at 22 days gestation. This closure is known as neurulation, and results in the formation of a tube termed the neural tube [5]. The anterior neural tube closes by about 24 days gestation and serves as the foundation for further brain development; the posterior neural tube closes by about 26 days gestation and serves as the foundation for further spinal cord development. Defects in the closure of the neural tube lead to encephaloceles or myelomeningoceles. Nervous system segmentation At the rostral end of the newly closed neural tube exures delineate the primary vesicles, which are designated as the hindbrain (rhombencephalon), mesencephalon, and forebrain (prosencephalon). The primary vesicles can be further subdivided into secondary vesicles that will form adult brain structures. The hindbrain can be divided into the metencephalon and myelencephalon, which will become the pons, cerebellum, and medulla oblongata of the adult. The mesencephalon will be the midbrain, and the prosencephalon divides into the telencephalon (two telencephalic vesicles) and diencephalon. The telencephalic vesicles will become the cerebral hemispheres; the diencephalon will become the thalamus and hypothalamus. Regional specication of the developing telencephalon is an important step in brain development, and is likely under control of a number of genes
919
that encode transcription regulators. In the fruit y, Drosophila, these genes are involved in segmentation of this animal and dene structures such as hair-like spiracles. Not surprisingly, the role of these genes in human brain development diers, yet it appears that the general role of these proteins is that of regional specication of clones of cells destined to form specic brain structures. Homeobox and other transcription genes encode some of these transcriptional regulators and these turn on genes by binding to specic DNA sequences, and in so doing initiate genetic programs that lead to cell and tissue dierentiation. EMX2, a transcriptional regulator, has a homolog in Drosophila that denes the hair spiracles and has been implicated in human brain malformations. Disorders of segmentation Schizencephaly (cleft in brain) has been regarded by many as a migration abnormality; however, it is best understood as a disorder of segmentation because one of the genes that is abnormal in the more severe and familial forms is EMX2 [6,7]. Thus, this developmental disorder, at least in the more severe cases, appears to the result of failure of regional specication of a clone of cells that are destined to be part of the cortex. Clinically, these patients vary depending upon the size of the defect and upon whether bilateral disease is present [8]. The clefts extend from the pia to the ventricle and are lined with a polymicrogyric gray matter (see the discussion in Polymicrogyria) [9]. The pia and ependyma are usually in apposition, especially in severe cases. The defect is termed open-lipped if the cleft walls are separated by cerebrospinal uid, and closed-lipped if the walls are in contact with one another. Bilateral schizencephaly is associated with mental retardation and spastic cerebral palsy; aected patients often are microcephalic. Seizures almost always accompany severe lesions, especially the open-lipped and bilateral schizencephalies. The exact frequency of seizures in patients with the less severe lesions is uncertain. Most patients in whom schizencephaly is diagnosed undergo neuroimaging because of seizures. Therefore, a bias in favor of a universal occurrence of seizures in this disorder is noted. Hence, patients with schizencephaly who do not have epilepsy might exist, but the malformation remains undetected because no imaging is performed. Seizure type and onset may also vary in this disorder. Patients may experience focal or generalized seizures, and some will present with infantile spasms. The onset varies from infancy to the early adult years. Seizures may be easily controlled or may be recalcitrant to standard anticonvulsant therapy. Improvements in neuroimaging have enhanced the recognition of schizencephalic lesions [913]. The lesions may occur in isolation or may be associated with other anomalies of brain development such as septo-optic dysplasia (see Disorders of prosencephalic cleavage; Fig. 2) [14]. Disorders of segmentation likely represent a heterogeneous set of abnormalities of varying etiologies. One theory holds that an early (rst-trimester)
920
destructive event disturbs subsequent formation of the cortex. Another theory is that segmental failure occurs in the formation of a portion of the germinal matrix or in the migration of primitive neuroblasts. Certainly, the nding of mutations of the EMX2 gene in some patients with the openlipped form of schizencephaly supports the latter hypothesis. Prosencephalon cleavage At about 42 days of gestation, the prosencephalon undergoes a division into two telencephalic vesicles that are destined to become the cerebral hemispheres. The anterior portion of this cleavage is induced by midline facial structures and the presence of the notochord. Abnormalities of this process are thought to result in holoprosencephaly, septo-optic dysplasia, and agenesis of the corpus callosum [15]. One of the important molecules responsible for the induction of this cleavage is Sonic hedgehog [16]. This protein is produced by the notochord, ventral forebrain, and the oor plate of the neural tube [17]. It interacts through at least one receptor, PTCHa human homolog of patched, and alters the expression of transcription factors [18,19]. Furthermore, in an interesting link between these ventral inductive events and segmentation, Sonic hedgehog can alter the expression of the transcriptional regulating genes when applied to proliferating cells at critical times in development [21]. This ties the inductive proteins to the expression of transcriptional regulating genes and gives a hint as to the mechanisms involved in inductive processes. Other molecules of interest in this inductive process are the retinoids, which are lipids capable of crossing membranes and that have been shown to exist in posterior to anterior gradients across embryos [3,20]. Retinoic acid can alter the pattern of transcriptional factors in neuroepithelial cells [3] and can downregulate Sonic hedgehog, perhaps explaining some of the head defects seen in retinoid embryopathy [17,22]. Disorders of prosencephalon cleavage Holoprosencephaly Holoprosencephaly is a heterogeneous disorder of prosencephalon cleavage that results from a failure of the prosencephalic vesicle to cleave normally. Three forms of this disorder have been described: alobar, semilobar, and lobar [23,24]. In the alobar form, the telencephalic vesicle completely fails to divide, producing a single horseshoe-shaped ventricle, sometimes with a dorsal cyst, fused thalami, and a malformed cortex (Fig. 1C). In the semilobar form, the interhemispheric ssure is present posteriorly, but the frontal and, sometimes, parietal lobes, continue across the midline [25]; in some cases just ventral fusion is noted (Fig. 1A and B). In the lobar form, only minor changes may be seen: the anterior falx and the septum pellucidum usually are absent, the frontal lobes and horns are hypoplastic, and the genu of the corpus callosum may be abnormal.
921
Fig. 1. Holoprosencephaly. (A, B) Images from the same patient with semilobar holoprosencephaly. Arrowheads point to lack of ventral interhemispheric cleavage in (A) and to fused thalami in (B). Note in (B) that the septum pellucidum is absent. (C) Alobar holoprosencephaly. Note the horseshoe or mushroom-shaped single ventricle designated by *. The arrowhead points to the anterior midline where no interhemispheric ssure is noted.
Holoprosencephaly is associated with a spectrum of midline facial defects. These include cyclopia, a supraorbital proboscis, ethmocephali, in which the nose is replaced by a proboscis located above hypoteloric eyes; cebocephaly, in which hypotelorism and a nose with a single nostril are seen; and premaxillary agenesis, with hypotelorism, a at nose, and a midline cleft lip [26]. Only children who have the lobar and semilobar forms are known to survive for more than a few months. An infant aected with the severe form is microcephalic, hypotonic, and visually inattentive [25]. In infants with the less severe forms of holoprosencephaly, myoclonic seizures frequently develop and, if the infant survives, autonomic dysfunction, failure to thrive, psychomotor retardation, and atonic or spastic cerebral palsy often are present. Some infants with the lobar form may be only mildly aected and, for example, present as a relatively mild spastic diplegia. Pituitary defects may be associated with these malformations, and may result in neuroendocrine dysfunction [27]. One, therefore, has to wonder how much
922
genetic overlap exists between this condition and septo-optic dysplasia to be described below. Holoprosencephaly has been associated with maternal diabetes [28], retinoic acid exposure, cytomegalovirus, and rubella [29]. Chromosome abnormalities associated with this disorder include trisomies 13 and 18; duplications of 3p, 13q, and 18q; and deletions in 2p, 7q, 13q, and 18q [30]. Of particular concern to the clinician is the existence of an autosomal dominant form in which mutations in Sonic Hedgehog lead to variable expression of holoprosencephaly. In the mildest form of this genetic disorder, patients may have a single central incisor, a choroid ssure coloboma, or simply attention decit disorder; a parent of a child with holoprosencephaly manifesting these features should be considered to be at high risk for recurrence of holoprosencephaly in their children (up to 50% risk) [31,32]. A number of other genes (HPE1 (21q22.3), HPE2 (2p21), HPE3 (7q36), HPE4 (18p), ZIC2, SIX3 (2p21), and PATCHED) have been associated with holoprosencephaly, and although potentially inherited in an autosomal recessive fashion, most occurrence seems to be random [33,34]. Septo-optic dysplasia Septo-optic dysplasia (de Morsier syndrome) is a disorder characterized by the absence of the septum pellucidum, optic nerve hypoplasia, and hypothalamic dysfunction (Fig. 2A, B). It may be associated with agenesis of the corpus callosum. This disorder should be considered in any patient who exhibits at least two of the above abnormalities and perhaps even solely hypothalamic dysfunction [35]. Septo-optic dysplasia also appears to involve prosencephalic cleavage and development of anterior telencephalic structures [36]. About 50% of patients with septo-optic dysplasia have schizencephaly [14] (Fig. 2C, D). Patients may present with visual disturbance, seizures, mental retardation, hemiparesis (especially if associated with schizencephaly), quadriparesis, or hypothalamic dysfunction. Endocrine abnormalities may include growth hormone, thyroid hormone, or antidiuretic hormone function or levels. The consideration of septo-optic dysplasia necessitates an evaluation of the hypothalamicpituitary axis because as many as 60% of the children with this disorder might exhibit evidence of a disturbance of endocrine function [37]. This evaluation can include thyroid function studies and electrolytes; these patients are at high risk for growth retardation. The recent identication of patients with this condition that harbor mutations in the transcriptional regulator gene HESX1, suggest that the mechanism of this disorder is likely genetic and a patterning or segmental abnormality [38]. Even though the genetic abnormality has been identied for a minority of patients, there exists the possibility that this may not represent an entirely genetic disorder because associations have been made with young maternal age, diabetes, the use of anticonvulsants, phencyclidine, cocaine, and alcohol [39].
923
Fig. 2. Septo-optic dysplasia with associated schizencephaly. All images are from the same patient. (A, B) Axial and coronal images, respectively, with arrows pointing the midline where a septum pellucidum should be present, but is not. In (C) and (D) axial and coronal images show a superior open lipped schizencephaly designated by the arrow in (C) and by the black arrowhead in (D). Additionally, areas of polymicrogyria are noted and are designated by white arrowheads.
Cell proliferation Following telencephalic cleavage, a layer of proliferative pseudostratied neuroepithelium lines the ventricles of the telencephalic vesicles. These cells will give rise to the neurons and glia of the mature brain. The generation of the proper complement of cells is a highly ordered process that results in the
924
generation of billions of neurons and glia. Neuroepithelial processes extend from the ventricular surface to the pial surface, and the nuclei of the primitive neuroepithelial cells move from the cortical surface in a premitotic phase to a mitotic phase near the ventricle. Cells divide at the most ventricular aspects of the developing telencephalon, and after division move back toward the pial surface. The pial processes of neuroepithelial cells near the ventricle often will detach from the cortical surface before a new cycle begins. Neuroepithelial cells divide in so-called proliferative units such that each unit will undergo a specic number of divisions resulting in the appropriate number of cells for the future cortex. Abnormalities in the number of proliferative units or in the total number of divisions can lead to disorders of the brain manifested by abnormal brain size and, therefore, an unusually small or large head circumference. Two such disorders resulting in small head sizeradial microbrain and microcephaly veraare believed to result from abnormalities of this phase of neurodevelopment [40]. Disorders in which too many cells are generated in the proliferative phase result in megalencephaly (large brain) or, if proliferative events go awry on only one side of the developing cortex, hemimegalencephaly. The genes and molecules involved in regulating the proliferative cycles in human brain formation are likely similar to those involved in other species. This cell cycle in the brain can be divided into a number of distinct phases: mitosis (M), rst gap (G1), deoxyribonucleic acid synthesis (S), and second gap (G2) [41]. These phases appear to be regulated by key molecules to check the advancement of proliferation. Some cells enter a resting state (G0) that they maintain throughout life. Others temporarily enter this phase to await a specic signal to proliferate later. Probably the G1S transition regulation determines the number of cell cycles and, therefore, the complement of cells that will make brain [40]. Cyclins are proteins that appear to be involved in cell cycle control. These proteins are activating subunits of cyclin-dependent kinases. Cyclins Dl, D2, D3, C, and E seem to control the key transition of a cell to the G1S interface; this transition is regarded as important because it commits a cell to division [4244]. Cyclin E seems to be the gatekeeper for this transition, and is essential for movement from the G1 to the S phase [42,45]. The number of cells that nally make up the mature nervous system is less than that generated during proliferation. Cells appear not only to be programmed to proliferate during development but to contain programs that lead to cell death [46,47]. The term apoptosis (from the Greek, meaning a falling o ) has been applied to this programmed loss of cells [48]. Disorders of neuronal and glial proliferation Microcephaly Although primary microcephaly may be a normal variant, in the classic symptomatic form, clinical and radiologic examinations reveal a receding
925
forehead, at occiput, early closure of fontanelles, and hair anomalies such as multiple hair whirls and an anterior cowlick. Neuroimaging may show small frontal and occipital lobes, open opercula, and a small cerebellum [24]. The cortex may appear thickened and the white matter reduced. Histologic examination may show a reduction of cell layers in some areas and an increase in others [49]. Neurologic ndings also vary. Only mild psychomotor retardation may be noted, sometimes associated with pyramidal signs, or more severe retardation, seizures, and an atonic cerebral palsy might be evidenced. Primary microcephaly is seen in many genetic syndromes and, in its isolated form, may be autosomal recessive, autosomal dominant, or X-linked [5052]. Microcephaly vera is the term most often applied to this genetic form of microcephaly. Aected children present with a head circumference that is usually more than 4 standard deviations below the mean, hypotonia, and psychomotor retardation. They later show mental retardation, dyspraxias, motor incoordination and, sometimes, seizures. On histologic examination, neurons in layers II and III are depleted [53]. Destructive lesions of the forming brain, such as those caused by teratogens and by infectious agents, also may result in microcephaly. Teratogens of note are alcohol, cocaine, and hyperphenylalaninemia (maternal phenylketonuria) [54]. Intense radiation exposure (such as that from a nuclear explosion) in the rst trimester, can cause microcephaly [55]. Microcephaly and intracranial calcications are likely due to well-recognized in utero infections caused by cytomegalovirus, toxoplasmosis, or the human immunodeciency virus. Megalencephaly and hemimegalencephaly The terms megalencephaly and hemimegalencephaly refer to disorders in which the brain volume is greater than normal (not owing to the abnormal storage of material); usually, the enlarged brain is accompanied by macrocephaly, or a large head. Although considered by some to be a migration disorder, the increase in brain size in these disorders appears to be attributable to errors in neuroepithelial proliferation, as the microscopic appearance of the brain is that of an increase in number of cells (both neurons and glia) and in cell size [5659]. Typically, patients are noted to have large heads at birth, and may manifest an accelerated head growth in the rst few months of life [60,61]. Children with megalencephaly or hemimegalencephaly may come to medical attention when presenting with seizures, a developmental disorder (mental retardation), hemihypertrophy, or a hemiparesis (opposite the aected hemisphere). Seizures vary both in onset and in type, and usually are the most problematic symptom, sometimes necessitating hemispherectomy or callosotomy [58]. Approximately 50% of patients with linear sebaceous nevus syndrome have hemimegalencephaly [62,63]. Many patients with hypomelanosis of Ito also
926
have hemimegalencephaly [64]. The neuropathologic and clinical pictures of these associations appear to be identical to the isolated hemimegalencephalies. Neuronal dierentiation Normal dierentiation At the time of neuronal dierentiation the neural tube consists of four consecutive layers: (1) the ventricular zone, the innermost layer, which gives rise to neurons and all of the glia of the central nervous system; (2) the subventricular zone, which is the adjacent, more supercial layer and is the staging area from which postmitotic neurons begin to dierentiate and to migrate; (3) the intermediate zone, which is the contiguous, more supercial zone, and which is destined to become the cortical plate and the future cerebral cortex; and (4) the marginal zone, which is the outermost zone and is composed of the cytoplasmic extensions of ventricular neuroblasts, corticopetal bers, and the terminal processes of radial glia (which, at this time, are completely spanning the neural tube). Dierentiation of neuroepithelial cells begins in the subventricular layer at approximately gestational day 26. The older, larger pyramidal cells are the rst cells to be born and probably dierentiate early to act as targets in the migration of the nervous system. Disorders of dierentiation Disorders such as tuberous sclerosis, in which both tumor development and areas of cortical dysplasia are seen, might be a dierentiation disorder. The brain manifestations of this disorder include hamartomas of the subependymal layer, areas of cortical migration abnormalities (tubers, cortical dysgenesis), and the development of giant-cell astrocytomas in upwards of 5% of aected patients. Two genes for tuberous sclerosis have been identied: TSC1 (encodes for Hamartin) has been localized to 9q34 [65], and TSC2 (encodes for Tuberin) has been localized to l6pl3.3 [65]. Neuronal migration Normal migration At the most rostral end of the neural tube in the 40- to 41-day-old fetus, the rst mature neurons, Cajal-Retzius cells, begin the complex trip to the cortical surface. Cajal-Retzius cells, subplate neurons, and corticopetal nerve bers form a preplate [66]. The neurons generated in the proliferative phase of neurodevelopment represent billions of cells poised to begin the trip to the cortical surface and to form the cortical plate. These neurons accomplish this task by attaching to and migrating along radial glial in a process known as radial migration or by somal translocation in a neuronal process [67]. The radial glia extend from the ventricle to the cortical surface. In the process of migration, the deepest layer of the cortical plate migrates and
927
deposits before the other layers. Therefore, the rst neurons to arrive at the future cortex are layer VI neurons. More supercial layers of cortex then are formedthe neurons of layer V migrate and pass the neurons of layer VI; the same process occurs for layers IV, III, and II. The cortex therefore is formed in an inside-out fashion [6769]. A possible mode of movement in neuronal migration on glia would be the attachment of the neuroblast to a matrix secreted by either the glia or the neurons. The attachment of the neuron would be through integrin receptors, cytoskeletal-linking membrane-bound recognition sites for adhesion molecules. That attachment serves as a stronghold for the leading process and soma of the migrating neuron. Neuron movement on radial glia involves an extension of a leading process, a neural outgrowth having an orderly arrangement of microtubules. Shortening of the leading process owing to depolymerization or shifts of microtubules may result in movement of the soma relative to the attachment points. This theory of movement of neurons also must include a phase of detachment from the matrix at certain sites, so that the neuron can navigate successfully along as much as 6 cm of developing cortex (the maximum estimated distance of radial migration of a neuron in the human). Finally, the movement of cells must stop at the appropriate location, the boundary between layer I and the forming cortical plate. Therefore, some stop signal must be given for the migrating neuron to detach from the radial glia and begin to dierentiate into a cortical neuron. Perhaps that signal is REELIN, a protein that is disrupted in the mouse mutant Reeler and is expressed solely in the Cajal Retizius cells at this phase of development [66,7073]. Migration disorders Advanced neuroimaging techniques, particularly magnetic resonance imaging, have allowed the recognition of major migration disorders and of the frequency of more subtle disorders of migration. Some of these disorders are associated with typical clinical features that might alert the clinician to the presence of such malformations even before imaging is obtained. In other disorders, the clinical features are so varied that a strong correlation between imaging and the clinical presentation points to a specic genetic syndrome. Lissencephaly Lissencephaly (smooth brain) refers to the external appearance of the cerebral cortex in those disorders in which a neuronal migration aberration leads to a relatively smooth cortical surface. One should not consider only agyria in making this diagnosis, rather, the full spectrum includes agyria and pachygyria. Gyri and sulci do not form in this disorder because the lack of corticalcortical attractive forces owing to improper axon pathways. At least two types have been identied: classic lissencephaly, and cobblestone
928
lissencephaly. The distinction is based upon the external appearance and upon the underlying histology, and can be made with neuroimaging. Classic lissencephaly Classic lissencephaly may occur in isolation, owing to LIS1 or Doublecortin aberrations or in combination with somatic features and LIS1 deletions in the Miller-Dieker syndrome. The hallmarks on imaging are a lack of opercularization (covering of the sylvian ssure), large ventricles or colpocephaly (dilated posterior horns), and agyria or pachygyria (Fig. 3). The corpus callosum is almost always present, and the posterior fossa is usually normal, although a form of lissencephaly does exist that includes cerebellar hypoplasia. LIS1 genetic syndromes The Miller-Dieker phenotype consists of distinct facial features that include bitemporal hallowing, upturned nares, and a peculiar burying of the
Fig. 3. Neuronal migration disorders: classic lissencephaly spectrum. A normal T1-weighted axial image is shown for comparison. Images (A) ILS 176, (B) ILS165, and (D) LP94-051 are gifts from William Dobyns and the Lissencephaly Project. ILS176 and ILS165 are images from patients with deletions of the LIS1 gene. (C) ILS087 and (D) ILS94-051 are images from patients with point mutations detected in sequencing of the LIS1 gene. ILS176, ILS165, and ILS087 represent grade 2 or 3 lissencephaly with the typical thickened cortex, heterotopic cells below surface, and an abnormally smooth cortical surface. LP94-051 represents a patient with band heterotopia who had a mutation in LIS1; the band is worse posteriorly.
929
upper lip by the lower lip at the corners of the mouth. The lissencephaly is usually more severe than isolated lissencephaly, and the prognosis is worse. Most aected patients die in the rst few years of life. By both molecular and cytogenetic techniques, deletions in the terminal portion of one arm of chromosome 17 can be found in approximately 90% of Miller-Dicker lissencephalv cases [74]. The deletions of the terminal part of chromosome 17 in these cases have included microdeletions [74], ring 17 chromosome [75], pericentric inversions [76], and a partial monosomy of 17p13.3 [77]. The most appropriate genetic test is a uorescent in situ hybridization (FISH) for LIS1; this test involves marking chromosome 17 at the centromere and LIS1 with uorescent probes. The greatest risk to future ospring exists when a parent harbors a balanced translocation involving this region of chromosome 17. In families that are aected in this manner, screening by amniocentesis can be performed in subsequent pregnancies. Therefore, it is recommended that both parents have screening for chromosome 17 rearrangements by FISH for LIS1. Should a translocation be present in a parent, then the LIS1 uorescence will be on another chromosome. Isolated lissencephaly Classic lissencephaly without somatic or facial features represents a distinct genetic syndrome from Miller-Dieker, but it involves the same gene LIS1. Approximately 40% of patients with isolated lissencephaly have FISH detectable deletions of LIS1, and about 20% of additional patients harbor mutations of this gene [7880]. The remaining patients may have mutations involving promoter regions of LIS1 or abnormalities of other genes such as Doublecortin or the involvement of other genes that have not been recognized. The genetic risk of recurrence is highest when rearrangements of chromosome 17 exist in one parent. This is rare in isolated lissencephaly, but could occur. Therefore, it would be prudent to perform FISH for LIS1 in the parents of children with isolated lissencephaly who have FISH proven deletions for the LIS1 region. The prognosis for this disorder is better than that for Miller-Dieker syndrome, but it is not consistent with long-term survival. These patients typically present in the rst few months of life with hypotonia, lack of visual xation, and often seizures. Patients with lissencephaly will uniformly have seizures and profound mental retardation. Often, seizures are very dicult to control and require multiple anticonvulsant drugs. X-linked lissencephaly The imaging of X-linked lissencephaly looks nearly identical to the images of lissencephaly involving LISI. Patients have classic lissencephaly, and the neurologic presentation described above. However, the skeletal and other
930
anomalies seen in the Miller-Dieker are not noted in this form of lissencephaly. When viewing the images from patients with lissencephaly owing to abnormalities of LIS1 and of Doublecortin it is apparent that dierences in an anterior to posterior gradient of severity exists [8183] (Fig. 4). Doublecortin mutations result in anterior greater than posterior severity, whereas LIS1 mutations result in posterior greater than anterior severity [84]. In addition, X-linked lissencephaly occurs mostly in boys; girls who are heterozygous for Doublecortin mutations have band heterotopia [80,81,85] (Fig. 4). Women with band heterotopia have been known to give birth to boys with lissencephaly. In female patients, the less severe phenotype probably is attributed to random lyonization of the X chromosome, such that in a variable number of cells, normal gene expression is seen and, in the remaining cells, the Doublecortin mutation-containing X chromosome is expressed. It is presumed that those cells expressing the abnormal X chromosome will be arrested in the migration to the surface of the brain and reside in a subcortical band. When viewing images from patients with this disorder, a thick band of tissue that is isointense with cerebral gray matter is seen within what should be the white matter of the hemispheres (Fig. 4). The overlying gyral appearance may vary from normal cortex to a pachygyria. A brain biopsy performed in a patient demonstrated well-preserved lamination in cortical layers IIV [86]. Layers VVI were not clearly separated and merged with underlying white matter. Beneath the white matter was a coalescent cluster of large, well-dierentiated neurons. Lissencephaly with cerebellar hypoplasia The association of lissencephaly with cerebellar hypoplasia represents a distinct malformation from both a genetic and clinical standpoint to those described above. The cerebellar hypoplasia is usually extreme, and the brainstem may be small. Patients may or may not have an associated micro-
Fig. 4. Subcortical band heterotopia. (A, B, C) Images from the same patient showing dierent views of extensive subcortical band heterotopia owing to a Doublecortin mutation. Arrowheads point to subcortical band in axial (A), coronal (B), and lateral saggital (C) views. The subcortical bands have the same signal intensity of overlying gray matter.
931
cephaly. This disorder is often inherited in an autosomal recessive fashion and may be due to mutations in REELIN in some families [87]. Cobblestone lissencephaly Cobblestone lissencephalies are disorders in which a smooth conguration of cortex is noted, but the distinction from classic lissencephaly is made based upon the clinical association of eye abnormalities, muscle disease, and progressive hydrocephalus (Fig. 5). The term cobblestone refers to the appearance of the cortical surface upon pathologic examination. In these disorders, cells pass their stopping point and erupt over the surface of the cortex into the subarachnoid spaces. This results in a cobblestone street appearance to the surface, and therefore, the name. The Walker-Warburg, muscleeyebrain, and HARD E syndromes are likely all varying degrees of the same entity. Abnormalities that may or may not be seen in these disorders include muscular dystrophy, ocular anterior chamber abnormalities, retinal dysplasias (evidenced by abnormal electroretinogram and visual evoked responses), hydrocephalus (usually of an obstructive type), and encephaloceles. The Walker-Warburg syndrome might be diagnosed even if the ocular examination and muscle biopsies are normal if on MRI, an abnormal white matter signal and a thickened falx suggest the diagnosis. Neuroimaging of the muscleeyebrain disorders often reveals focal white matter abnormalities.
Fig. 5. Cobblestone lissencephaly. Note extensive hydrocephalus, abnormal white matter, and relatively smooth cortical surface.
932
Fukuyama muscular dystrophy is distinguished from the WalkerWarburg-like syndromes by the severity of the muscular dystrophy [8891]. This disorder is seen more often in Japan than in the Western hemisphere, probably because it is the result of a founder mutation. Patients typically present with evidence of a neuronal migration defect, hypotonia, and depressed reexes. Recent identication of Fukutin as the causative gene in this disorder should provide insight into the pathogenesis of the cobblestone lissencephalies [92]. This disorder is inherited as an autosomal recessive disorder. The cobblestone lissencephalies often have an associated cerebellar and brainstem hypoplasia, and therefore may be dicult to distinguish from lissencephaly with cerebellar hypoplasia described above. The presence of eye abnormalities, elevated CPK, or other evidence for the presence of muscle disease and progressive hydrocephalus distinguish this disorder. These disorders may be inherited in an autosomal recessive manner. Polymicrogyria Polymicrogyria (many small gyri) is a disorder often considered to be a neuronal migration disorder, but alternate theories exist regarding its pathogenesis. The microscopic appearance of the lesion is that of too many small abnormal gyri. The gyri may be shallow and separated by shallow sulci, which may be associated with an apparent increased cortical thickness on neuroimaging (Fig. 6). The multiple small convolutions may not have intervening sulci, or the sulci may be bridged by fusion of overlying molecular layer, which may give a smooth appearance to the brains surface. The interface of white matter with gray matter is not distinct and often this observation serves as the conrmation of the presence of polymicrogyria. Polymicrogyria has also been associated with genetic and chromosomal disorders. It is found in disorders of peroxisomal metabolism such as Zellweger syndrome and neonatal adrenal leukodystrophy. Familial bilateral frontal polymicrogyria and bilateral perisylvian polymicrogyria have been reported. Therefore, if no identiable cause of the polymicrogyric malformation is found, the recurrence risk may be that of an autosomal recessive disorder. A bilateral parasagittal parieto-occipital polymicrogyria has also been described. The clinical picture varies depending on the location, extent, and cause of the abnormality. Microcephaly with severe developmental delay and hypertonia may result when polymicrogyria is diuse. When polymicrogyria is unilateral, focal decits might be seen. Epilepsy often is present, characterized by partial complex seizures or partial seizures that secondarily generalize. The age at presentation and severity of seizures depends on the extent of the associated pathology. Bilateral perisylvian dysplasia is a disorder of perisylvian polymicrogyria resulting in an uncovered sylvian ssure on neuroimaging and on saggital imaging an extension of the sylvian ssure to the top of the convexity
933
Fig. 6. Polymicrogyria. (A, B) Axial and saggital images from the same patient that demonstrate the ndings typical of bilateral perisylvian dysplasia. Black arrowheads in (A) point to uncovered sylvian ssures (lack of opercularization) and in (B) to areas of polymicrogyria lining the sylvian ssure. Gray arrow in (B) points to an extension of the sylvian ssure to the top of the convexity. (C) Imaging from a patient with extensive polymicrogyria designated by black arrowheads. (D) Imaging from a patient with frontal and perisylvian polymicrogyria (black arrowheads).
(Fig. 6A, B). Patients with bilateral perisylvian dysplasia have a pseudobulbar palsy, and often dysphagia can impair proper nutrition. The majority of patients have epilepsy with early onset; infantile spasms are common. The cause of this syndrome remains unknown, although hints of a genetic mechanism exist. Detailed chromosomal analyses have revealed deletions of chromosomes 1, 2, 6, 21, and 22 [93], and an X-linked form has been also described [94,95]. Monozygotic twins and siblings with this disorder have been described, suggesting a possible autosomal recessive mechanism. Some speculate that this is a disorder of regional specication, given the bilateral, symmetric nature of the lesions. Heterotopias Heterotopias are collections of normal-appearing neurons in an abnormal location, presumably secondary to a disturbance in migration. The exact mechanism of the migration aberration has not been established, although various hypotheses have been proposed. These include damage to the radial glial bers, premature transformation of radial glial cells into
934
Fig. 7. Bilateral periventricular nodular heterotopia. Arrowheads point to nodules with MR signaling characteristics that are similar to gray matter lining the lateral aspects of the lateral ventricles.
astrocytes, or a deciency of specic molecules on the surface of neuroblasts or of the radial glial cells (or the receptors for those molecules) that results in disruption of the normal migration process [96]. Heterotopias often occur as isolated defects that may result in only epilepsy. However, when they are multiple, heterotopias might also be associated with a developmental disorder and cerebral palsy (usually spastic). In addition, if other migration defects such as gyral abnormalities are present, the clinical syndrome may be more profound. Usually, no cause is apparent. Occasionally, heterotopias may be found in a variety of syndromes, including neonatal adrenal leukodystrophy, glutaric aciduria type 2, GM1 gangliosidosis, neurocutaneous syndromes, multiple congenital anomaly syndromes, chromosomal abnormalities, and fetal toxic exposures. Heterotopias may be classied by their location: subpial, within the cerebral white matter, and in the subependymal region. When subependymal, one must consider the X-linked dominant disorder associated with Filamin mutations (Xq28) (Fig. 7). Leptomeningeal heterotopias often contain astrocytes mixed with ectopic neurons and may resemble a gliotic scar. They may be related to discontinuities in the external limiting membrane and often are associated with cobblestone lissencephaly. These subarachnoid heterotopias are responsible for the pebbled appearance of the surface of the brain. White matter heterotopias may be focal, subcortical, or diuse. They may cause distortion of the ventricles and may be associated with diminished white matter in the surrounding area. Summary The progress made in the understanding of the genetics of human brain malformations has lead to insight into the formation of brain and into
935
mechanisms of disease aecting brain. It bears upon neurologists and geneticists to recognize the patterns of diseases of brain formation, to properly diagnose such disorders, to assess the recurrence risk of these malformations, and to guide families with appropriate expectations for outcomes. This article may serve as a guide to neurologists in their approach to these disorders. Because this area is one of rapid progress, the clinician is advised to seek more current information that may be available through on-line databases and other sources. References
[1] Brockes J. Reading the retinoid signals. Nature 1990;345(6278):7668. [2] Lemire R, Loeser J, Leech R, et al. Normal and abnormal development of the human nervous system. New York: Harper & Row; 1975. [3] Scotting PJ, Rex M. Transcription factors in early development of the central nervous system. Neuropathol Appl Neurobiol 1996;22:46981. [4] Thaller C, Eichele G. Isolation of 3,4-didehrdroretinic acid, a novel morphogenetic signal in the chick wing bud. Nature 1990;345(6278):81522. [5] Geelan JA, Langman J. Closure of the neural tube in the cephalic region of the mouse embryo. Anat Rec 1977;189:62540. [6] Brunelli S, Faiella A, Capra V, et al. Germline mutations in the homeobox gene EMX2 in patients with severe schizencephaly. Nat Genet 1996;12:946. [7] Hillburger AC, Willis JK, Bouldin E, et al. Familial schizencephaly. Brain Dev 1993;15: 2346. [8] Barkovich AJ, Kjos BO. Schizencephaly: correlation of clinical ndings with MR characteristics. Am J Neuroradiol 1992;13:8594. [9] Barth P. Disorders of neuronal migration. Can J Neurol Sci 1987;14:116. [10] Barkovich A, Chuang S, Norman D. MR of neuronal migration anomalies. Am J Neuroradiol 1987;8:100917. [11] Barkovich A, Kjos B. Schizencephaly: correlation of clinical ndings with MR characteristics. Am J Neuroradiol 1992;13:8594. [12] Brodtkorb E, Nilsen G, Smevik O, et al. Epilepsy and anomalies of neuronal migration: MRI and clinical aspects. Acta Neurol Scand 1992;86:2432. [13] Kuzniecky R. Magnetic resonance imaging in developmental disorders of the cerebral cortex. Epilepsia 1994;35(Suppl 6):S4456. [14] Aicardi J, Goutieres F. The syndrome of absence of the septum pellucidum with porencephalies and other developmental defects. Neuropediatrics 1981;12:31929. [15] Leech R, Shuman R. Midline telencephalic dysgenesis: report of three cases. J Child Neurol 1986;1:22432. [16] Smith J. Hedgehog, the oor plate, and the zone of polarizing activity. Cell 1994;76:1936. [17] Helms JA, Kim CH, Hu D, et al. Sonic hedgehog participates in craniofacial morphogenesis and is down-regulated by teratogenic doses of retinoic acid. Dev Biol 1997;187:2535. [18] Johnson RL, Rothman AL, Xie J, et al. Human homolog of patched, a candidate gene for the basal cell nevus syndrome. Science 1996;272:166871. [19] Stone DM, Hynes M, Armanini M, et al. The tumour-suppressor gene patched encodes a candidate receptor for Sonic hedgehog. Nature 1996;384:12934. [20] Conlon RA. Retinoic acid and pattern formation in vertebrates. Trends Genet 1995;11:3149. [21] Nakagawa Y, Kaneko T, Ogura T, et al. Roles of cell-autonomous mechanisms for dierential expression of region-specic transcription factors in neuroepithelial cells. Development 1996;122:244964. [22] Lammer EJ, Chen DT, Hoar RM, et al. Retinoic acid embryopathy. N Engl J Med 1985; 313:83741.
936
[23] Barkovich AJ. Imaging aspects of static encephalopathies. In Miller G, Ramer JC, editors. Static encephalopathies of infancy and childhood. New York: Raven Press; 1992. p. 4594. [24] Dobyns WB. Cerebral dysgenesis: causes and consequences. In Miller G, Ramer JC, editors. Static encephalopathies of infancy and childhood. New York: Raven Press; 1992. p. 23548. [25] Nyberg DA, Mack LA, Bronstein A, et al. Holoprosencephaly: prenatal sonographic diagnosis. Am J Radiol 1987;149:10518. [26] Souza JP, Siebert JR, Beckwith JB. An anatomic compairson of cebocephaly and ethmocephaly. Teratology 1990;42:34757. [27] Romshe CA, Sotos JF. Hypothalamic-pituitary dysfunction in siblings of patients with holoprosencephaly. J Pediatr 1973;83:108890. [28] Barr M, Hansen JW, Currey K, et al. Holoprosencephaly in infants of diabetic mothers. J Pediatr 1983;102:5658. [29] Cohen MMJ, Lemire RJ. Syndromes with cephaloceles. Teratology 1983;25:16172. [30] Muenke M. Clinical, cytogenetic, and molecular approaches to the genetic heterogeneity of holoprosencephaly. Am J Med Genet 1989;34:23745. [31] Heussler HS, Suri M, Young ID, et al. Extreme variability of expression of a Sonic Hedgehog mutation: attention diculties and holoprosencephaly. Arch Dis Child 2002;86:2936. [32] Roessler E, Belloni E, Gaudenz K, et al. Mutations in the human Sonic hedgehog gene cause holoprosencephaly. Nat Genet 1996;14:35760. [33] Ming JE, Kaupas ME, Roessler E, et al. Mutations in PATCHED-1, the receptor for SONIC HEDGEHOG, are associated with holoprosencephaly. Hum Genet 2002;110: 297301. [34] Wallis D, Muenke M. Mutations in holoprosencephaly. Hum Mutat 2000;16:99108. [35] Brickman JM, Clements M, Tyrell R, et al. Molecular eects of novel mutations in Hesx1/HESX1 associated with human pituitary disorders. Development 2001;128:518999. [36] Ouvier R, Billson F. Optic nerve hypoplasia: a review. J Child Neurol 1986;1:1818. [37] Costin G, Murphree AL. Hypothalamicpituitary function in children with optic nerve hypoplasia. Am J Dis Child 1985;139:24954. [38] Dattani MT, Martinez-Barbera JP, Thomas PQ, et al. Mutations in the homeobox gene HESX1/Hesx1 associated with septo-optic dysplasia in human and mouse. Nat Genet 1998; 19:12533. [39] Patel H, Tze WJ, Crichton JU, et al. Optic nerve hypoplasia with hypopituitarism: septooptic dysplasia with hypopituitarism. Am J Dis Child 1975;129:17580. [40] Caviness VS, Takahashi T. Proliferative events in the cerebral ventricular zone. Brain Dev 1995;17:15963. [41] Ross ME. Cell division and the nervous system: regulating the cycle from neural dierentiation to death. Trends Neurosci 1996;19:628. [42] Matsushime H, Quelle DE, Shurtle SA, et al. D-type cyclin-dependent kinase activity in mammalian cells. Mol Cell Biol 1994;14:206676. [43] Meyerson M, Harlow M. Identication of G1 kinase activity for cdk6, a novel cyclin D partner. Mol Cell Biol 1994;14:207786. [44] Pardee AB. G1 events and regulation of cell proliferation. Science 1989;246:6038. [45] Sherr CJ. G1 phase progression: cycling on cue. Cell 1994;79:5515. [46] Clarke PG. Developmental cell death: morphological diversity and multiple mechanisms. Anat Embryol (Berl) 1990;181:195213. [47] Schwartz LM, Smith SW, Jones MEE, et al. Do all programmed cell deaths occur via apoptosis? Proc Natl Acad Sci USA 1993;90:9804. [48] Kerr JFR, Wyllie AH, Currie AR. Apoptosis: a basic biological phenomenon with wideranging implications in tissue kinetics. Br J Cancer 1972;26:23957. [49] Warkany J, Lemire RJ, Cohen MM. Mental retardation and congenital malformations of the central nervous system. Chicago: Year Book Medical; 1981. p. 1340. [50] Haslam R, Smith DW. Autosomal dominant microcephaly. J Pediatr 1979;95:7015.
937
[51] Optiz JM, Holt MC. Microcephaly: general considerations and aids to nosology. J Craniofac Genet Dev Biol 1990;10:175204. [52] Renier WO, Gabreels FJM, Jasper TWJ, et al. An xlinked syndrome with microcephaly, severe mental retardation, spasticity, epilepsy and deafness. J Ment Dec Res 1982;26: 2740. [53] Evrard P, De Saint-Georges P, Kadhim H, et al. Pathology of prenatal encephalopathies. In French JH, Harel S, Casaer P, editors. Child neurology and development disabilities. Baltimore: PH Books; 1989. p. 15376. [54] Lenke RR, Levy HL. Maternal phenylketonuria and hyperphenylalaninemia. N Engl J Med 1980;303:12028. [55] Yamazaki JN, Schull WJ. Perinatal loss and neurological abnormalities among children of the atomic bomb. Nagasaki and Hiroshima revisited, 1949 to 1989. JAMA 1990;264: 6059. [56] Barkovich AJ, Chuang SH. Unilateral megalencephaly: correlation of MR imaging and pathologic characteristics. Am J Neuroradiol 1990;11:52331. [57] DeRosa MJ, Secor DL, Barsom M, et al. Neuropathologic ndings in surgically treated hemimegalencephaly: immunohisotchemical, morphometric and ultrastructural study. Acta Neuropathol (Berl) 1992;84:25060. [58] Renowden S, Squier M. Unusual magnetic resonance and neuropathological ndings in hemi-megalencephaly: report of a case following hemispherectomy. Dev Med Child Neurol 1994;36:35769. [59] Rintahaka P, Chugani H, Messa C, et al. Hemimegalencephaly: evaluation with positron emission tomography. Pediatr Neurol 1993;9:218. [60] DeMyer W. Megalencephaly in children. Clinical syndromes, genetic patterns and dierential diagnosis from other cases of megalencephaly. Neurology 1972;22:63443. [61] DeMyer W. Megalencephaly: types, clinical syndromes and management. Pediatr Neurol 1986;2:3218. [62] Hager BC, Dyme IZ, Guertin SR, et al. Linear nevus sebaceous syndrome: megalencephaly and heterotopic gray matter. Pediatr Neurol 1991;7:459. [63] Pavone L, Curatolo P, Rizzo R, et al. Epidermal nevus syndrome: a neurologic variant with hemimegalencephaly, gyral malformation, mental retardation, seizures, and facial hemihypertrophy. Neurology 1991;41:26671. [64] Jelinek JE, Bart RS, Schi GM. Hypomelanosis of Ito (incontinentia pigmenti achromians): report of three cases and review of the literature. Arch Dermatol 1973;107: 596601. [65] Haines JL, Short MP, Kwiatkowski DJ, et al. Localization of one gene for tuberous sclerosis within 9q329q34, and further evidence for heterogeneity. Am J Hum Genet 1991; 49:76472. [66] Ogawa M, Miyata T, Nakajima K, et al. The reeler gene-associated antigen on CajalRetzius neurons is a crucial molecule for laminar organization of cortical neurons. Neurons 1995;14:899912. [67] Sidman R, Rakic P. Neuronal migration with special reference to developing human brain, a review. Brain Res 1973;62:135. [68] Aicardi J. The place of neuronal migration abnormalities in child neurology. Can J Neurol Sci 1994;21:18593. [69] Angevine J, Sidman R. Autoradiographic study of cell migration during histogenesis of cerebral cortex of the mouse. Nature 1961;192:7668. [70] DArcangelo G, Miao GG, Chen S-C, et al. A protein related to extracellular matrix proteins deleted in the mouse reeler. Nature 1995;374:71923. [71] DArcangelo G, Miao GG, Curran T. Detection of the reelin breakpoint in reeler mice. Brain Res Mol Brain Res 1996;39:2346. [72] Hirotsune S, Takahara T, Sasaki N, et al. The reeler gene encodes a protein with an EGFlike motif expressed by pioneer neurons. Nat Genet 1995;10:7783.
938
[73] Rommsdor M, Gotthardt M, Hiesberger T, et al. Reeler/disabled-like disruption of neuronal migration in knockout mice lacking the VLDL receptor and ApoE receptor 2. Cell 1999;97:689701. [74] Ledbetter SA, Kuwano A, Dobyns WB, et al. Microdeletions of chromosome 17p13 as a cause of isolated lissencephaly. Am J Hum Genet 1992;50:1829. [75] Sharief N, Craze J, Summers D, et al. Miller-Dieker syndrome with ring chromosome 17. Arch Dis Child 1991;66:7102. [76] Greenberg F, Stratton R, Lockhart L, et al. Familial Miller-Dieker syndrome associated with pericentric inversion of chromosome 17. Am J Med Genet 1986;23:8539. [77] Dobyns WB, Stratton RF, Parke JT, et al. Miller-Dieker syndrome: lissencephaly and monosomy 17p. J Pediatr 1983;102:5528. [78] Chong SS, Pack SD, Roschke AV, et al. A revision of the lissencephaly and Miller-Dieker syndrome critical regions in chromosome 17p13.3. Hum Mol Genet 1997;6:14755. [79] Lo Nigro C, Chong SS, Smith ACM, et al. Point mutations and an intragenic deletion in LIS1, the lissencephaly causative gene in isolated lissencephaly sequence and Miller-Dieker syndrome. Hum Mol Genet 1997;6:15764. [80] Pilz DT, Matsumoto N, Minnerath S, et al. LIS1 and XLIS (DCX) mutations cause most classical lissencephaly, but dierent patterns of malformation. Hum Mol Genet 1998;7: 202937. [81] Gleeson JG, Luo RF, Grant PE, et al. Genetic and neuroradiological heterogeneity of double cortex syndrome. Ann Neurol 2000;47:2659. [82] Leventer RJ, Pilz DT, Matsumoto N, et al. Lissencephaly and subcortical band heterotopia: molecular basis and diagnosis. Mol Med Today 2000;6:27784. [83] Matsumoto N, Leventer RJ, Kuc J, et al. Mutation analysis of the DCX gene and genotype/phenotype correlation in subcortical band heterotopia. Eur J Hum Genet 2000;9(1):512. [84] Dobyns WB, Truwit CL, Ross ME, et al. Dierences in the gyral pattern distinguish chromosome 17-linked and X-linked lissencephaly. Neurology 1999;53:2707. [85] Gleeson JG, Allen KM, Fox JW, et al. Doublecortin, a brain-specic gene mutated in human X-linked lissencephaly and double cortex syndrome, encodes a putative signaling protein. Cell 1998;92:6372. [86] Palmini A, Andermann F, Aicardi J, et al. Diuse cortical dysplasia, or the double cortex syndrome: the clinical and epileptic spectrum in 10 patients. Neurology 1991;41: 165662. [87] Hong SE, Shugart YY, Huang DT, et al. Autosomal recessive lissencephaly with cerebellar hypoplasia is associated with human RELN mutations. Nat Genet 2000;26:936. [88] Fukuyama Y, Ohsawa M, Suzuki H. Congenital progressive muscular dystrophy of the Fukuyama type-clinical, genetic and pathologic considerations. Brain Dev 1981;3:129. [89] Takashima S, Becker L, Chan F, et al. A Golgi study of the cerebral cortex in Fukuyamatype congenital muscular dystrophy, Walker-type lissencephaly, and classical lissencephaly. Brain Dev 1987;9:6216. [90] Toda T, Miyake M, Kobayashi K, et al. Linkage-disequilibrium mapping narrows the Fukuyama-type congenital muscular dystrophy (FCMD) candidate region to less than 100 kb. Am J Hum Genet 1996;59:131320. [91] Toda T, Yoshioka M, Nakahori Y, et al. Genetic identity of Fukuyama-type congenital muscular dystrophy and Walker-Warburg syndrome. Ann Neurol 1995;37:99101. [92] Kobayashi K, Nakahori Y, Miyake M, et al. An ancient retrotransposal insertion causes Fukuyama-type congenital muscular dystrophy. Nature 1998;394:38892. [93] Worthington S, Turner A, Elber J, et al. 22q11 deletion and polymicrogyriacause or coincidence? [In Process Citation]. Clin Dysmorphol 2000;9:1937. [94] Leventer RJ, Lese CM, Cardoso C, et al. A study of 220 patients with polymicrogyria delineates distinct phenotypes and reveals genetic loci on chromosome 1p, 2p, 6q, 21q and 22q. Am J Hum Gen 2001;70(Suppl):Abstract.
939
[95] Villard L, Nguyen K, Winter RM, et al. X-linked bilateral perisylvian polymicrogyria maps to Xq28. In Am J Hum Gen 2001;70(Suppl):Abstract. [96] Raymond AA, Fish D, Sisodiya S, et al. Abnormalities of gyration, heterotopias, tuberous sclerosis, focal cortical dysplasia, microdysgenesis, dysembryoplastic neuroepithelial tumour and dysgenesis of the archicortex in epilepsy: clinical, EEG and neuroimaging features in 100 adult patients. Brain 1995;118:62960.
Neurol Clin N Am 20 (2002) 941964
Tuberous sclerosis complex and neurobromatosis type 1: the two most common neurocutaneous diseases
Raymond S. Kandt, MD*
Johnson Neurological Clinic, High Point, NC Department of Pediatrics, Wake Forest University School of Medicine, Winston-Salem, NC, USA
Tuberous sclerosis complex (TSC) and neurobromatosis type 1 (NF1) are autosomal dominant disorders. They are the prototypes of the neurocutaneous diseases. The involvement of multiple tissues and organs, the similar locations of the macular skin lesions of TSC and NF1, the variable clinical expressivity, and similarities in their biochemical pathologic ndings cause these two disorders to be considered more closely related than the other neurocutaneous syndromes. Clinically, however, the two disorders are easily dierentiated. Other disorders classically considered as neurocutaneous diseases include neurobromatosis type 2 (NF2) [1], ataxia telangiectasia, von Hippel Lindau syndrome, and Sturge-Weber syndrome. This article focuses on TSC and NF1, particularly with regard to salient clinical aspects of childhood presentation and genetic aspects. Tuberous sclerosis complex TSC is an autosomal dominant disorder that involves multiple organs and tissues. The major impact of the disorder is on the brain, skin, and kidneys, but it also aects the eyes and heart. The prevalence of TSC is 1:10,000. Lesions are caused by hamartomas and hamartias [2]. A major point regarding symptoms and signs of TSC is the variable clinical expression of the disorder, even among patients from families with multiple aected generations. The major manifestations of TSC include skin lesions in more than 95%, mental retardation in approximately 50%, autism, seizures in approximately 85%, kidney disease in approximately 60% or more, and uncommon cerebral
* Correspondence. 606 N. Elm Street, High Point, NC. 0733-8619/02/$ - see front matter 2002, Elsevier Science (USA). All rights reserved. PII: S 0 7 3 3 - 8 6 1 9 ( 0 2 ) 0 0 0 2 3 - 3
942
R.S. Kandt / Neurol Clin N Am 20 (2002) 941964
giant cell astrocytomas, but the severity ranges from asymptomatic patients to severe disability. Although the neurologic symptoms most often raise suspicion for the diagnosis of TSC, the relatively unique skin lesions and cranial MRI characteristics are the most helpful for conrming the diagnosis. Diagnosis requires the presence of two or more hamartomas. Neurologic symptoms of tuberous sclerosis complex The neurologic symptoms of TSC are the ones that often call attention to the diagnosis but are not adequate to conrm the diagnosis. TSC patients may have epilepsy, mental retardation, and autism as well as learning disorders. Although neurologic symptoms provide much of the morbidity, TSC patients do not demonstrate unique manifestations of these neurologic problems. For example, even though TSC is one of the major diagnosable causes of infantile spasms, there are many other causes. By contrast, cutaneous symptoms are more helpful for diagnosis but cause no signicant morbidity, with the exception of occasional disgurement. Seizures Typical absence seizures do not occur more frequently in TSC patients than in the general population, but all other seizure types are more common in TSC patients. The most common seizure types are infantile spasms or partial seizures, often with rapid secondary generalization. Seizures are the most common neurologic symptom of TSC, occurring in approximately 85% of patients [2]. The prevalence of seizures and mental retardation is not related to subependymal nodules, but the prevalence of both is higher in patients with more cortical tubers. The most serious seizures for TSC patients are infantile spasms [3], which peak between 4 and 6 months of age. Neonatal onset of seizures in TSC is uncommon. The diagnosis of TSC should be considered in all infants who have infantile spasms. The more common features that raise suspicion for TSC in patients who have infantile spasms include hypomelanotic macules of the skin, cortical tubers on cranial MRI, and cardiac rhabdomyomas by echocardiography. For older children who have seizures, the suspicion for TSC is raised by additional dermatologic manifestations. If there is no family history for TSC and no cranial MRI or skin ndings for TSC, further evaluation for TSC in an epileptic patient is usually unnecessary. Seizures are evaluated and treated medically in TSC patients in the same manner as in other patients with epilepsy; that is, electroencephalograms (EEGs) are performed as part of seizure diagnosis and are repeated as indicated by the course of the seizures. Some antiepileptic agents have been specifically tested in TSC and found to have similar utility as in non-TSC patients. For example, lamotrigine is a useful antiepileptic medication for TSC patients and is more likely to decrease seizures in those TSC patients who have partial seizures only and no history of infantile spasms [4]. Infantile spasms are
943
typically treated with corticotropin or prednisone. In general, antiepileptic medications for TSC patients are chosen based on seizure type [4]. Still controversial is the use of vigabatrin for the treatment of infantile spasms in TSC [5]. Evidence has indicated that vigabatrin is especially eective for the infantile spasms of TSC, leading to rapid cessation of infantile spasms in 95% of TSC patients [6]. This is a success rate that is arguably not achieved by the more standard treatments of corticotropin or prednisone or by alternative antiepileptic agents, such as topiramate, zonisamide, lamotrigine, divalproex sodium, or others. As yet, no randomized trial has conrmed the superiority of vigabatrin. Evidence of side eects of vigabatrin, such as constriction of visual elds in human beings and vacuolation of cerebral white matter in animals, has derailed eorts for US Food and Drug Administration (FDA) approval of vigabatrin in the United States and has raised serious concerns regarding the safety of vigabatrin [5,7]. Therefore, if parents in the United States desire that their infants receive vigabatrin treatment for infantile spasms, they must obtain it from foreign countries. Surgical treatment of intractable seizures in TSC is receiving more attention. Previously, it was argued that the multiple cortical tubers of TSC rendered futile the surgical excision of epileptic foci; that is, removal of one epileptic tuber would probably uncover the epileptogenic activity of another tuber. However, several series have demonstrated the utility of tuberectomy in selected TSC patients whose major seizure activity arises from a single tuber [8,9]. Localization of epileptogenic tubers can be improved with multimodality imaging, including combinations of positron emission tomography (PET) scanning, single photon emission computed tomography (SPECT) scans, and MRI [10]. When seizures remain medically intractable and tuber resection is not tenable, the ketogenic diet, corpus callosotomy, or vagus nerve stimulation is a viable alternative treatment [8]. Mental retardation The second most common neurologic symptom of TSC is mental retardation, which occurs in approximately 50% of patients. When generalized seizures, including infantile spasms, start in the rst 2 years of life, most TSC patients are mentally retarded, autistic, or both. All mentally retarded children with TSC have seizures. TSC patients who have normal intelligence may or may not have seizures, but the seizures are usually not as severe and usually have later onset (Fig. 1). TSC patients who have the most tubers are more likely to have both mental retardation and seizures [2]. The most disabled are those TSC patients who have both infantile spasms and severe to profound mental retardation. Behavior problems, such as hyperkinesis, attention problems, and aggression, are often associated with mental retardation or autism. Autism Among patients who have autism, the cause is unknown for many. Nevertheless, there are several recognizable causes of infantile autism that
944
Fig. 1. A human gure drawing and printed name by a 5-year-old normally intelligent girl with tuberous sclerosis complex.
stand out, including TSC, Rett syndrome, and fragile X syndrome. As many as 50% to 60% of TSC patients may have autism, but the true prevalence is probably lower. Autism may be more prevalent in TSC patients whose tubers predominate in the temporal lobes or cerebellum. Although autism may occur in TSC patients who have neither seizures or mental retardation, the presence of seizures in an autistic patient increases the likelihood of TSC. Therefore, as with patients who have infantile spasms, it is important to consider and evaluate for the diagnosis of TSC for autism, particularly for autistic patients who have seizures. Autism is a communication disorder that constitutes a syndrome composed of a triad of features: impairment of reciprocal social interaction, impairment in verbal and nonverbal communication, and a markedly restricted repertoire of activities and interests that may manifest as stereotypic activities. It is important to note that all these features are common in patients who have mental retardation and that they become more common as IQ decreases. Therefore, to make a diagnosis of autism, the impairments must be out of proportion to those that are appropriate for the patients intellectual level. In a study using PET scans, compared with TSC patients who were retarded or had normal intelligence, autistic TSC patients were more likely to show glucose hypometabolism in the lateral temporal cortices,
945
increased uptake of a-methyl-tryptophan in caudate nuclei, and glucose hypermetabolism in deep cerebellar nuclei [11]. Either a prior history of infantile spasms or the presence of temporal lobe hypometabolism in TSC patients was associated with a communication disorder [11]. The overall number of cortical tubers does not correlate with a greater likelihood of autism, but an increased number of cerebellar tubers does [12]. Increased intracranial pressure Increased intracranial pressure in TSC is caused by the subependymal giant cell astrocytoma, a lesion relatively unique to TSC [13] and typically nishing its growth by the end of the second decade. The giant cell astrocytoma of the brain characteristically arises at either foramen of Monro from a subependymal nodule and typically enhances after intravenous contrast on CT or MRI [14]. If it grows large enough to obstruct one or both foramina of Monro, the astrocytoma may cause hydrocephalus, which is often unilateral. Baseline cranial MRI is useful to determine if the ventricle was abnormally shaped or enlarged before the giant cell astrocytoma. Indications for surgical removal of this relatively benign mass include progressive enlargement of the giant cell astrocytoma, progressive ventricular enlargement, or symptoms of elevated intracranial pressure. If the giant cell tumor is rst diagnosed in an individual near or older than 20 years of age who has hydrocephalus that is not progressive, the risk of surgery may outweigh the benet. Early symptoms of hydrocephalus typically include morning headache and vomiting, and sometimes sixth cranial nerve palsy. The giant cell astrocytoma can be surgically removed with a minimal chance of recurrence. Chemotherapy or radiation therapy is unnecessary and is not administered. Dermatologic signs of tuberous sclerosis complex The skin lesions are most helpful in recognizing TSC. There are multiple skin lesions that are characteristic of TSC. The most common skin lesions are hypomelanotic macules (also called hypopigmented macules or ash-leaf spots). Hypomelanotic macules Probably 90% or more of TSC patients have hypomelanotic macules, often 0.5 to 2 cm but sometimes larger, occurring on the face, trunk, and extremities (Fig. 2). The macules do not disturb the contour of the skin; they cannot be detected by an examiner with closed eyes. They are pale but do not totally lack pigment as opposed to vitiligo. Frequently, they have an ash-leaf shape, but equally as often in my experience, they have the shape of a thumb print. Occasionally, the hypomelanosis manifests on the scalp as white hair-poliosis. On sun-exposed skin, particularly in older individuals who have tanned their skin repeatedly, areas of skin atrophy may show a similar decrease of pigment as the hypopigmented macules. These areas should
946
Fig. 2. A hypomelanotic macule of tuberous sclerosis complex.
be discounted as providing diagnostic evidence for TSC. Of particular diculty is the dierentiation between small areas of skin atrophy and the confetti lesions of TSC in these sun-exposed individuals; the confetti lesions, as the name suggests, manifest as numerous small hypopigmented spots. Hypomelanotic macules rarely may occur in individuals without TSC, but usually are single or few in number. In TSC, hypomelanotic macules are valuable to suggest the diagnosis because they are usually present at birth, a time when there may be few other manifestations. Because the macules are pale, they can be dicult to detect, especially in pale-skinned individuals. Skin examination with the ultraviolet Woods light makes the macules easy to detect but is not always necessary in patients with dark complexions. Angiobromas Facial angiobromas are the classic skin lesions of TSC. Perhaps 50% of TSC patients have facial angiobromas; in the past, they were misnamed as adenoma sebaceum. The angiobromas typically do not occur until near puberty, a time when the facial bumps can easily be confused with acne. Angiobromas do not have comedones (whiteheads or blackheads), however. The facial angiobromas may occur in a limited distribution when rst developing, and there may only be three or four, most often on the cheeks and quite close to the nose or in the depression between the lower lip and chin (Fig. 3), and they may progress near and after puberty. In older people, facial angiobromas can be misdiagnosed as acne rosacea. In people with lighter complexions, the facial angiobromas are often esh colored, but the red angiomatous component may show through, giving the impression of a red papular rash. In younger children destined to develop angiobromas, a fever may cause excessive reddening of the cheeks. Angiobromas, even when not numerous, may cause facial disgurement because of their redness or prominence. They may also be prone to easy bleeding even from such minor trauma as contact with a pillow case, as occurred with the teenager in Figure 4.
947
Fig. 3. A small number of esh-colored tuberous sclerosis complex angiobromas on the cheek near the nose and a small number in the depression between the lip and chin.
Other skin lesions: forehead plaques, shagreen patches, ungual bromas, and gingival bromas The forehead plaques and shagreen patches have the same angiobromatous pathologic appearance as facial angiobromas but appear less vascular and are most common in other areas. Forehead plaques are often darker than surrounding skin, and their location is implicit in the name.
Fig. 4. More extensive tuberous sclerosis complex angiobromas. These may bleed with mild trauma. In this boy, the angiobromas are bright red. Note the gingival bromas.
948
Identical-appearing lesions may be present on the scalp or cheeks, however. The shagreen patches are often in the region of the lower back, either close to the midline or on the sides of the back. Sometimes they are in groups of small papules that are only several millimeters large, or they can give the appearance of the surface of an orange peel, but they are usually esh colored or slightly darker or redder than surrounding skin. One of my patients had multiple large shagreen patches, including one on his forearm (forearm shagreen). Several patients have had shagreen patches on their thighs or buttocks. In most cases, no treatment is necessary for shagreen patches. The ungual bromas have the same pathologic characteristics as the facial angiobromas and are generally esh colored. Some have a red core from the angiomatous component (Fig. 5). Rarely, trauma to the nail can cause an ungual broma (Fig. 6), or the chronic trauma from tight shoes can cause the appearance of an ungual broma, especially on the lateral side of the fth toe. Gingival bromas between the teeth are less common (see Fig. 4). Other clinical ndings in tuberous sclerosis complex Renal involvement is relatively common in TSC and includes renal cysts, angiomyolipomas, and, rarely, renal carcinoma (<2%). The original localization of the TSC2 gene was facilitated by the occurrence of polycystic kidneys in TSC, because it led to the use of DNA probes related to autosomal dominant polycystic kidney disease [15,16]. A contiguous gene syndrome in which there are deletions of both the TSC2 gene and the adjacent PKD1 gene manifests both TSC and massively enlarged polycystic kidneys at birth [17]. Signicant cystic renal disease in TSC more often occurs with the TSC2 gene. Renal cysts occur in approximately 10% to 20% of individuals with TSC. Of patients with angiomyolipomas, approximately 10% have TSC [18]. By contrast, as many as 75% of TSC patients develop angiomyolipomas [19]. Angiomyolipomas are more common in women and are a common cause of morbidity in TSC. Angiomyolipomas have fateasily seen on CT,
Fig. 5. A esh-colored ungual broma with a central red core.
949
Fig. 6. A traumatic ungual broma on the great toe of a patient with back pain and no clinical features of tuberous sclerosis complex. His toe was traumatized with a hammer when he was 6 years old, and it was traumatized again years later with a cast iron skillet that removed the nail.
a feature that dierentiates them from renal cell carcinomas. Renal insuciency can occur with both cysts and angiomyolipomas, and serious hemorrhage may occur with larger angiomyolipomas (>3.5 cm). Renal disease is a leading cause of death in TSC [18]. Cardiac rhabdomyomas are usually asymptomatic but may cause obstruction of ow through the heart; congestive heart failure; cardiac arrhythmias, including Wol-Parkinson-White syndrome; and sudden death. Rarely, they are associated with stroke, which is more likely related to associated thrombotic material than to tumor fragments [20]. Although TSC pulmonary disease is rare, it can cause severe lung problems. It occurs in adult women (<1%). Aneurysms of cerebral and other vessels rarely occur, but most have been reported in children or young adults [2]. Retinal hamartomas rarely cause visual dysfunction but may be useful to conrm the diagnosis. Other ndings are numerous [2] and sometimes helpful for diagnosis, but unless symptomatic, they rarely require clinical attention. Neuroimaging of tuberous sclerosis complex Radiologic signs in the brain, such as by MRI, most commonly include cortical tubers in the cerebrum or cerebellum. Other ndings include
950
subependymal nodules, dysplastic heterotopic neurons that are seen as migration lines, and subependymal giant cell astrocytomas. In infants, the cortical tubers show dierent signal characteristics; that is, tubers have high T1 signal in infants, whereas the tubers do not begin to show high T2 signal until the child reaches the age of 12 months. Because MRI T2-weighted sequences show the contrast between the hypomyelination often associated with tubers and normal myelination, tubers are more readily imaged by MRI after myelination has matured to a greater extent (ie, from 1218 months and older). Typical MRI T2-weighted sequences detect many tubers, but MRI more readily detects tubers using uid-attenuated inversion recovery (FLAIR) sequences. Tubers are typically seen at the gray-white junction in the cerebral cortex (Fig. 7). Subependymal nodules are present in more than 80% of TSC patients and are commonly calcied. On MRI, they have relatively high T1 signal (isointense to white matter) and are found throughout the outer walls of the lateral ventricles (Fig. 8), often adjacent to caudate nuclei. On T2-weighted images, subependymal nodules are sometimes isointense to white matter but are often hypointense, particularly if they are calcied (Fig. 9). Calcied subependymal nodules are easy to see on CT, but they are now readily recognized by MRI;
Fig. 7. Coronal uid-attenuated inversion recovery MRI demonstrating a cortical tuber with a migration line extending to the edge of the ventricle.
951
Fig. 8. Axial T1-weighted MRI demonstrating subependymal nodules.
therefore, MRI is more commonly used for cranial imaging in TSC because it is superior to CT for detection of cortical tubers and migration lines. Genetic aspects of tuberous sclerosis complex Unique to neurocutaneous diseases, inactivating mutations in either of two distinct TSC genes (TSC1 and TSC2) cause one basic syndrome. Before DNA linkage studies, it was not suspected that TSC could be caused by
Fig. 9. Axial T2-weighted MRI demonstrating the same subependymal nodules as in Figure 8.
952
more than one gene [15,21]. This contrasts with the situation in neurobromatosis. Although NF1 and NF2 are caused by two dierent genes, the two genes cause two distinct disorders, and they were recognized as such before the genes were discovered. Of the two TSC genes, the TSC1 gene was the rst to be localized (in 1987) to chromosome 9, but the gene was not sequenced until 1997 [22]. By contrast, after TSC2 was mapped to chromosome 16 in 1992 [16], the gene was sequenced the next year [23]. Although a third TSC locus is possible, it is unlikely [23]. The product of the TSC1 gene is called hamartin, and that of the TSC2 gene is called tuberin [24]. Autosomal dominant inheritance occurs with TSC. In TSC families, the cause is almost evenly divided between TSC1 and TSC2 mutations. As many as two thirds of TSC cases occur as a result of sporadic (not inherited) mutations, however, and most sporadic TSC cases are caused by mutations in TSC2. Because of gonadal mosaicism in TSC, a recurrence risk of 2% is quoted to parents who are themselves unaected but have a child with TSC. Clinical dierences between TSC1 and TSC2 are now being recognized, but they are typically dierences of degree and not of character. For example, in one series of 224 patients, the clinical syndrome caused by TSC2 was more severe and more often associated with mental retardation [25]. By contrast, another large series of 225 patients found no phenotypic dierences between TSC1 and TSC2 [26]. Although dierences between TSC1 and TSC2 families are recognized, the symptoms of TSC are still highly variable in onset and severity. Even in a TSC1 or TSC2 family in which multiple members have TSC, there may be great variabilityone patient may have skin lesions, seizures, and mental retardation, whereas others may have skin lesions or seizures but no other problems. The kidney may have cysts, benign tumors, or, rarely, renal cell carcinoma (<2%). Part of the variability of TSC manifestations stems from the mode in which the defective genes have their deleterious eect. Although their cellular functions remain to be elucidated, several facts have been discovered. Tuberin and hamartin interact in a protein complex [27], which perhaps explains why a deciency of either one can produce essentially the same clinical syndrome. Both hamartin from TSC1 and tuberin from TSC2 are thought to function, at least in part, as tumor suppressor genes, by which they regulate cell growth and development. It is thought that both alleles of a TSC gene must be defective for the development of a hamartoma or hamartia. This model seems valid for most manifestations of TSC but has not been proven for cortical tubers. As an example of this model, the TSC2 allele on one member of the chromosome 16 pair may harbor a mutation inherited from one of the parents (ie, germline mutation). At the TSC2 allele on the other chromosome 16, a somatic mutation (eg, deletion) may occur. The absence of both normal alleles produces a lack of tuberin, culminating in one of the manifestations of TSC, such as a renal angiomyolipoma [28].
953
Diagnosis of tuberous sclerosis complex Blood tests based on gene analysis are available but have a high falsenegative rate. No single sign is present in all patients. The diagnostic features constituting hamartomas or hamartias have been divided into major and minor categories [29]. For diagnosis as denite TSC, the patient should have 2 of 11 major features or 1 major feature plus 2 of 9 minor features. Genetic blood tests for TSC1 and TSC2 are available, but perhaps 30% of patients with TSC have negative results. Major diagnostic features include the following: Facial angiobromas or forehead plaque Nontraumatic ungual broma Three or more hypomelanotic macules Shagreen patch Multiple retinal nodular hamartomas Cortical tuber (dysplasia plus migration tracts count as one feature) Subependymal nodule Subependymal giant cell astrocytoma Cardiac rhabdomyoma (single or multiple) Lymphangiomyomatosis Renal angiomyolipoma (when both lymphangiomyomatosis and renal angiomyolipomas are present, they count as one feature) Minor diagnostic features include the following: Multiple dental pits Hamartomatous rectal polyps (histologic conrmation is suggested) Bone cysts (radiographic diagnosis is sucient) Cerebral white matter radial migration lines Gingival bromas Nonrenal hamartoma Retinal achromic patch Confetti skin lesions Multiple renal cysts Additional diagnostic categories include probable TSC with one major plus one minor feature and possible TSC with one major or two or more minor features. When performing the diagnostic workup, a careful history and skin examination are the rst steps. If infantile spasms or autism is present, further workup to diagnose TSC includes MRI of the brain. If either skin lesions or brain MRI ndings are present but inadequate for a diagnosis of TSC, ophthalmologic examination for retinal hamartomas or renal ultrasonography may provide conrmatory evidence. Echocardiography is usually not helpful past the age of 2 years, because cardiac rhabdomyomas regress with advancing age [30]. When TSC is diagnosed, the following
954
studies are performed if not already completed: neurodevelopmental testing, ophthalmologic examination, electrocardiography, renal ultrasonography, and MRI of the brain [31]. Recommendations for follow-up testing after diagnosis focus on treatment or prevention of problems [31]. If there are no seizures, EEG is not needed. Neurodevelopmental testing is repeated at school entrance. Giant cell astrocytomas of the brain typically do not grow after the second decade, whereas renal angiomyolipomas often enlarge during early adulthood. Therefore, MRI of the brain is repeated every 1 to 3 years during childhood and adolescence, and renal ultrasonography is repeated every 1 to 3 years. For either brain MRI or renal ultrasonography, symptoms may require more scans. Chest CT is performed at adulthood in women for the rare complication of pulmonary lymphangiomyomatosis and is repeated if pulmonary dysfunction occurs. Other tests, such as EEG, ophthalmologic examination, electrocardiography, and echocardiography, are repeated only if clinical ndings suggest the need [31]. Treatment of tuberous sclerosis complex After diagnosis, TSC patients require treatment of seizures, educational treatment for mental retardation and autism, and pharmacologic treatment for behavioral disorders related to mental retardation and autism. The major symptom that can be treated is the epilepsy. Various antiepileptic medications, and sometimes surgical approaches, are used as discussed previously. Facial lesions are treated with laser therapy, and bothersome ungual bromas can be surgically removed. For patients with learning problems, mental retardation, or autism, educational intervention is helpful, but medications can also be useful, including dextroamphetamine and other psychostimulants for hyperactivity and inattention, uoxetine for autistic symptoms, clonidine for hyperarousal, and risperidone for aggression. Giant cell astrocytomas of the brain are treated surgically as discussed previously. Renal angiomyolipomas, particularly those larger than 3.5 cm, can often be treated with nephron-sparing surgery. For renal failure, TSC patients may undergo kidney transplantation and do not have excessive risk from immunosuppression.
Neurobromatosis type 1 NF1 is an autosomal dominant disorder with variable expressivity; patients who have the abnormal NF1 gene may manifest (express) dierent numbers or intensities of features, even within NF1 families. The prevalence is approximately 1 per 4000, the same order of magnitude as Duchenne muscular dystrophy. In contrast to TSC, mental retardation or seizures are relatively uncommon, but nerve sheath and central nervous system tumors
955
cause the major morbidity. The diagnostic suspicion is usually raised by the presence of multiple cafe au lait spots. Less commonly, in infants, the diagnosis is suspected because of facial malformations caused by bone dysplasia. Occasionally, children present with impaired vision from optic nerve glioma. Onset with other types of neurologic dysfunction is uncommon. Diagnosis requires two or more of the diagnostic features. Diagnostic criteria for neurobromatosis type 1 Two or more of the following criteria are required for diagnosis: Six or more cafe au lait spots 1.5 cm or larger in postpubertal individuals and 0.5 cm or larger in prepubertal children Two or more neurobromas of any type or one or more plexiform neurobromas Freckling in the axillary or inguinal regions Optic nerve glioma (optic pathway tumor) Two or more Lisch nodules (iris hamartomas) A distinctive osseous lesion, such as dysplasia of the sphenoid bone and dysplasia or thinning of long bone cortex (pseudoarthrosis) A rst-degree relative with NF1 according to the preceding criteria Of note, the absence of cafe au lait spots, axillary freckling, cutaneous neurobromas, and Lisch nodules by the age of 5 years excludes the diagnosis of NF1 with greater than 95% certainty. Blood tests based on gene analysis are available but have a high false-negative rate. Dermatologic features of neurobromatosis type 1 Cafe au lait spots and skinfold freckling The hallmark sign of NF1 is the cafe au lait spot (Fig. 10), named because of its color, similar to coee with milk. Most children with NF1 have cafe au lait spots and axillary freckling, and most adults also have cutaneous
Fig. 10. Cafe au lait spot.
956
neurobromas and Lisch nodules of the iris. Other than cosmesis, cafe au lait spots have no clinical consequences and no need for treatment. One or two cafe au lait spots are common in individuals without NF1, but six or more cafe au lait spots, occurring in more than 95% of people with NF1, rarely occur in people without NF1. Cafe au lait spots are macular; are distributed on the trunk, face, and extremities; and are darker in sun-exposed areas. A regular border dierentiates cafe au lait spots of NF1 from the cafe au lait spots of McCune-Albright syndrome (polyostotic brous dysplasia with precocious puberty). Rarely, a child may have isolated autosomal dominant cafe au lait spots. Cafe au lait spots are typically present by 2 years of age, but other manifestations of NF1 often do not develop until later. Among children with six or more typical cafe au lait spots, most (73%) develop other ndings for NF1 or segmental NF1 by the age of 5 years [32]. Axillary freckling occurs in most children with NF1 but may not have developed by the time the cafe au lait spots are rst diagnosed. By 5 years of age, however, as many as 84% of the children with NF1 have axillary freckling, with a smaller proportion ($50%) having inguinal freckling. Submammary freckling in women with NF1 occurs less commonly. Skinfold freckling is diagnostically helpful because freckling in individuals without NF1 is rare in areas that are not sun exposed. Neurobromas Cutaneous (dermal) neurobromas. Neurobromas arise from the peripheral nerve sheath. Neurobromas of the skin are rare in young children but may develop in preadolescence in up to 14% of children less than 10 years old. When they rst appear, the neurobromas often have a slight irregularity of skin with a combined small elevation and depression, sometimes with reddening of skin caused by dilated or proliferated capillaries. Larger neurobromas protrude from the skin, are often pink or violaceous, and frequently have an appearance that suggests a hard surface, but they are typically soft, and some even feel velvety. Cutaneous neurobromas typically enlarge and increase in number at certain times of hormonal change (puberty in boys and girls and pregnancy in girls and women). Although 95% of adults with NF1 who are more than 30 years old have neurobromas, less than half of adolescents with NF1 have neurobromas. Typically, an early age at the time of appearance of neurobromas portends larger and more numerous neurobromas in adulthood. Some neurobromas, particularly subcutaneous neurobromas or nodular neurobromas on nerve trunks, may be painful or pruritic, and they can be removed surgically. Plexiform neurobromas. Compared with dermal neurobromas, these skin tumors have ill-dened borders and are more diuse and highly vascular and do not follow tissue planes. The skin overlying the tumor is typically hyperpigmented and often hairy. Associated structures can be hypertrophied. Some feel nodular with palpable nerve trunks. Plexiform neurobromas that occur
957
in deep structures (eg, mediastinum and retroperitoneal compartment) may be dicult to diagnose and may cause various complications relating to the structures involved. Plexiform neurobromas occur in approximately 25% of NF1 patients and are present at birth or detected in early childhood; however, they may enlarge, especially near adolescence. They may be disguring, particularly when they involve the head (eg, eyelids) or neck. The plexiform neurobromas are dicult to treat surgically, regrow if not completely removed, and do not respond to radiation therapy. If they are disguring or located in areas that cause signicant morbidity, they should be removed. The best results occur with early removal, but because of their nature, the surgical morbidity may be substantial, making the decision for surgery dicult. In contrast to dermal neurobromas, which have no malignant potential, plexiform neurobromas may transform in 1% to 4% of individuals into neurobrosarcomas, rarely before the age of 10 years. Areas of abrasion (eg, the belt line) are more susceptible to malignant transformation. Rapid growth is an indication for biopsy [33]. Lisch nodules of the iris Lisch nodules are benign iris hamartomas, often visible only by slit-lamp examination. They cause no clinical problems but are useful for diagnosis. They are uncommon in young children (22% by 5 years of age), but if only cafe au lait spots are present, the discovery of Lisch nodules can conrm the diagnosis of NF1. They occur in 70% of children by 10 years of age [34] and are present in almost all adults (96%) with NF1. Neurologic features of neurobromatosis type 1 Tumors The most common tumors of the central nervous system in NF1 occur in the optic pathways, and are mostly slow-growing low-grade astrocytomas. Rarely, meningiomas occur in the same location in NF1 patients and may show identical ndings. Although imaging studies demonstrate that approximately 15% to 20% of NF1 patients have optic pathway tumors, no more than half of the tumors become symptomatic. In general, the presence of an optic nerve glioma in a child should raise the suspicion for NF1; at least 50% of children with optic nerve gliomas have NF1. Because chiasmatic tumors may involve the hypothalamus, the occurrence of precocious puberty in NF1 children should signal the likelihood of an optic glioma. Most children with symptomatic gliomas develop decreased visual acuity and optic pallor/atrophy, but depending on the location of the tumor in the visual pathways, they may present with visual eld defects, restricted eye movements, proptosis, headache, or hypothalamic dysfunction. Tumors that cause symptoms are typically diagnosed by the age of 6 years, and more than 90% show no progression of symptoms after diagnosis. A more recent study of 1893 NF1 patients less than 21 years old found that symptomatic
958
optic gliomas are usually diagnosed by the age of 3 years [34]. Therefore, imaging as a screen for optic pathway tumors in the absence of symptoms has limited value. Treatment for progressive tumors usually includes surgery and chemotherapy [35] or radiation therapy. Nevertheless, treatment of lesions that are questionably progressive should be tempered by reports of spontaneous regression [36] as well as by treatment complications, such as radiation-induced vasculopathy [37], cognitive problems, endocrine dysfunction, and the possibility, particularly after radiation therapy, of transformation of low-grade tumors into more malignant tumors. Whether treated initially or not, patients with visual pathway tumors are followed closely for any deterioration of vision or endocrinologic dysfunction. Symptomatic parenchymal brain tumors occur infrequently in NF1 ($1% 2% of patients), and they are usually low-grade astrocytomas. Brain tumors in NF1 generally become symptomatic by the time the patient reaches the age of 20 years. With the exception of brainstem gliomas, the outcome of these tumors is typically the same as in children without NF1. Brainstem gliomas, however, show slower progression or even regression in NF1 and may not need treatment [38]. Hydrocephalus, which may occur with brainstem tumors (eg, tectal gliomas), may require ventriculoperitoneal shunting. Other tumor types are much less common, including ependymomas, meningiomas, medulloblastomas, and primitive neuroectodermal tumors, and these tumor types are treated as they would be in children without NF1. If brain tumors are symptomatic and they are low-grade hemispheric or cerebellar astrocytomas, they are often removed surgically, particularly if they are not in areas that are critical for nervous system function. Those that are recurrent or not surgically accessible are treated with chemotherapy [35] or radiation therapy. In young children, the morbidity from radiation therapy is high. Similar to optic nerve gliomas in NF1, parenchymal brain tumors are not treated on discovery. They should be monitored before treatment for either lack of progression or the possibility of spontaneous regression [36,38]. Spinal nerve root neurobromas and, less commonly, spinal cord gliomas may cause symptoms and are typically treated surgically. Spinal meningoceles, often thoracic, are caused by dural ectasia and can be confused with neurobromas. Spinal meningiomas are probably no more common in NF1 than in the normal population; rather, they more commonly complicate NF2. Cognitive problems Mental retardation occurs only slightly more commonly in NF1 (4%8%) than in the normal population (3%). Specic learning disabilities have a major impact, however, and occur in 30% to 60% of children with NF1. Psychologic testing shows visual-spatial problems, language disorders, and memory dysfunction. Some of the children have attention-decit hyperactivity disorder. It remains unsettled as to whether cranial MRI T2
959
hyperintensities are correlated signicantly with the learning problems of NF1. Awareness of the possibility of learning problems in NF1 children can lead to proper educational treatment. Other complications of neurobromatosis type 1 NF1 may cause problems with almost every organ system because it aects both neural crestderived cells and mesodermal tissues. There is only a slight increase in the prevalence of epilepsy, perhaps 2% to 5%. Somewhat less than half of NF1 children have macrocephaly without hydrocephalus or any denite abnormality. Short stature occurs in approximately one third of such children. Scoliosis is frequent in NF1 ($15%). Scoliosis occurs earlier, and it more commonly needs surgical correction than does idiopathic scoliosis. Although it may have a long length similar to idiopathic scoliosis, vertebral dysplasia is often related to the NF1 scoliosis and then may show an acute angulation. Causes of scoliosis include spinal meningoceles and nerve root neurobromas (often with a dumbbell conguration). Sphenoid wing dysplasia occurs congenitally in less than 1% of children with NF1 but is relatively unique for NF1. Thinning of long bones, especially with medial bowing of the tibia, may predispose to fractures and pseudoarthrosis. Glaucoma occurs rarely and is usually congenital. It is associated with photophobia and enlargement of the corneal diameter. Neurobromas can occur in multiple areas, leading to unexpected complications, including constipation and prostate involvement [39]. Some tissues may be enlarged, leading to clitoral enlargement, macrodactyly (with or without a plexiform neurobroma), and macroglossia. Hypertension may occur in childhood but is more common in adults with NF1. A neurobroma may compress the renal artery, or the vascular dysplasia of NF1 may cause renovascular hypertension. NF1 may be the most common cause of renovascular hypertension in children [40]. Hypertension may also be caused by a pheochromocytoma. In addition to pheochromocytoma, NF1 is rarely associated with other neoplasms, such as carcinoid of the duodenum, nonlymphocytic leukemia, adenocarcinoma of the ampulla of Vater, and possibly rhabdomyosarcoma. The vascular dysplasia of NF1 is also associated with stroke in childhood, usually caused by the moyamoya phenomenon associated with occlusion of the supraclinoid carotid artery or proximal vessels of the circle of Willis [41]. Neuroimaging in neurobromatosis type 1 Compared with TSC, neuroimaging in NF1 does not have as much importance for diagnosis of the disorder but is more important for monitoring complications. There are no neuroimaging ndings included in the
960
diagnostic criteria for NF1. When the consensus criteria for diagnosis of NF1 were formulated in 1987, there was not much experience with MRI, especially in children. MRI T2 hyperintensities may have diagnostic signicance, however. In NF1, most young patients have unidentied bright objects (UBOs) on cranial MRI (Fig. 11), whereas the prevalence of UBOs goes down in older children. The nature of the UBOs remains unclear, and they are postulated to represent hamartomas, dysmyelinated areas, or spongiosis. Most adults with NF1 either do not have UBOs typical for NF1 or have confounding lesions caused by vascular or other disease. DeBella et al [42] used cranial MRI to examine 19 children with NF1 and 19 controls. Of the 19 control children, 11 had UBOs. Of note, the control children all had neurologic disorders. When only the UBOs typical for NF1 were considered (ie, those located in the basal ganglia, cerebellum, and brainstem), their presence was correlated with NF1 to a high degree, yielding a diagnostic sensitivity of 97% and a specicity of 79%. In terms of diagnosis, UBOs are more helpful in young children who have only one criterion for diagnosis (usually multi ple cafe au lait spots). They are less helpful later, because by the age of 5 years, diagnosis can usually be made based on other criteria. Further studies are necessary before typical UBOs can be considered a diagnostic criterion. Because of the morbidity of tumors in NF1, the follow-up of optic nerve gliomas, brain tumors, and spinal lesions is the most important use of neuroimaging in NF1. On T2-weighted MRI, optic nerve gliomas either demonstrate an enlarged optic nerve with a core of low signal surrounded by high
Fig. 11. Axial T2-weighted MRI of unidentied bright object in the cerebellar white matter.
961
signal, or there is a tubular enlargement of aected optic nerves. Optic nerve sheath dysplasia occasionally is mistaken for an optic nerve tumor. As discussed previously, imaging of asymptomatic patients to screen for optic nerve gliomas or brain tumors is rarely useful. Most clinicians obtain a baseline cranial MRI. If asymptomatic lesions are discovered that might represent tumors, they are commonly monitored every year or so. Cranial MRI T2 hyperintensities in the visual pathways, brainstem, or cerebellum often represent benign UBOs. After discovery of MRI T2 hyperintensities, however, it is helpful to perform a study with contrast. An enhancing lesion is more likely to represent an astrocytoma or other tumor. Although lack of enhancement is reassuring, some astrocytomas do not enhance. A follow-up scan is usually indicated to conrm lack of progression. Genetic aspects of neurobromatosis type 1 NF1 is an autosomal dominant disorder. For an aected parent, the risk of aected ospring is 50% with each pregnancy. The spontaneous mutation rate is high; thus, perhaps 50% of aected patients represent sporadic cases. Those with a new sporadic mutation then transmit NF1 as an autosomal dominant disorder. The gene for NF1 was initially localized to chromosome 17 in 1987, the same year that TSC1 was localized to chromosome 9. The NF1 gene is large and was described in 1990. There is only one gene that causes NF1. Most NF1 mutations cause premature truncation of the neurobromin protein. Similar to the TSC1 and TSC2 gene products, the normal NF1 gene product, neurobromin, is thought to function as a tumor suppressor [24]. Neurobromas are composed of multiple cellular elements, but the Schwann cell seems to be the target for NF1 gene inactivation. An individual with NF1 inherits an inactivating mutation of one of the NF1 alleles. This mutation is present in all cells. When a somatic inactivating mutation of the NF1 gene occurs in the other allele, a neurobroma or other tumor may develop. The prevalence of learning problems and other symptoms unrelated to tumor formation has led to the speculation that this large gene may have additional functions that may play a role in the heterozygous state. Germline mosaicism, documented in only one clinically normal father of two ospring with NF1, is thought to be rare enough that it is not important for genetic counseling [43]. By contrast, some patients have segmental or anatomically limited signs of NF1 consistent with somatic mosaicism. If patients with segmental NF1 have mosaicism of germ cells, they transmit NF1 as an autosomal dominant disorder. Management of neurobromatosis type 1 Patients with NF1 have multiple possible complications and need access to multiple specialists. A clinician who has a broad familiarity with NF1, often a neurologist or geneticist, coordinates the care. Patients are seen yearly and are evaluated for complications, such as focal neurologic dysfunction,
962
abnormal puberty, optic pathway tumor, scoliosis, cognitive/learning problems, and hypertension. Counseling includes supportive information, genetic aspects, surveillance of plexiform neurobromas for malignant transformation, and reassurance that less than half of NF1 patients (40%) develop medical problems from NF1. The similarity of a small portion of the neurobromin molecule to the GTPase-activating proteins raises the possibility of pharmacologic treatment (eg, farnesyl transferase inhibitors), which might prevent or ameliorate the neoplastic complications of NF1. Summary TSC and NF1 are the most common of the neurocutaneous diseases, and both are autosomal dominant with a high spontaneous mutation rate. For diagnosis, two features are necessary for each disease. Skin ndings for each are especially helpful for diagnosis, as is neuroimaging in TSC. For NF1, neuroimaging is not yet reliable for diagnosis. In children, brain symptoms cause most of the morbidity in TSC, and nerve sheath and nervous system tumors as well as learning disabilities cause major morbidity in NF1. Renal disease becomes a serious problem for adults with TSC. The TSC1, TSC2, and NF1 genes function as tumor suppressor genes and have other functions that are being investigated. Blood tests for diagnosis have a high falsenegative rate. Therapies for TSC and for NF1 are both medical and surgical. References
[1] MacCollin M, Mautner VF. The diagnosis and management of neurobromatosis 2 in childhood. Semin Pediatr Neurol 1998;5:24352. [2] Gomez MR, Sampson JR, Whittemore VH. Tuberous sclerosis complex. New York: Oxford University Press; 1999. [3] Riikonen R, Simell O. Tuberous sclerosis and infantile spasms. Dev Med Child Neurol 1990;32:2039. [4] Franz DN. Diagnosis and management of tuberous sclerosis complex. Seminars in Pediatric Neurology 1998;5:25368. [5] Riikonen RS. Steroids or vigabatrin in the treatment of infantile spasms? Pediatr Neurol 2000;23:4038. [6] Hancock E, Osborne JP. Vigabatrin in the treatment of infantile spasms in tuberous sclerosis: literature review. J Child Neurol 1999;14:714. [7] Prasad AN, Penney S, Buckley DJ. The role of vigabatrin in childhood seizure disorders: results from a clinical audit. Epilepsia 2001;42:5461. [8] Guerreiro MM, Andermann F, Andermann E, et al. Surgical treatment of epilepsy in tuberous sclerosis: strategies and results in 18 patients. Neurology 1998;51:12639. [9] Koh S, Jayakar P, Dunoyer C, et al. Epilepsy surgery in children with tuberous sclerosis complex: presurgical evaluation and outcome. Epilepsia 2000;41:120613. [10] Asano E, Chugani DC, Muzik O, et al. Multimodality imaging for improved detection of epileptogenic foci in tuberous sclerosis complex. Neurology 2000;54:197684. [11] Asano E, Chugani DC, Muzik O, et al. Autism in tuberous sclerosis complex is related to both cortical and subcortical dysfunction. Neurology 2001;57:126977.
963
[12] Weber AM, Egelho JC, McKellop JM, et al. Autism and the cerebellum: evidence from tuberous sclerosis. J Autism Dev Disord 2000;30:5117. [13] Shepherd CW, Scheithauer BW, Gomez MR, et al. Subependymal giant cell astrocytoma: a clinical, pathological, and ow cytometric study. Neurosurgery 1991;28:8648. [14] Nabbout R, Santos M, Rolland Y, et al. Early diagnosis of subependymal giant cell astrocytoma in children with tuberous sclerosis. J Neurol Neurosurg Psychiatry 1999;66: 3705. [15] Kandt RS. Tuberous sclerosis: the next step. J Child Neurol 1993;8:10710. [16] Kandt RS, Haines JL, Smith M, et al. Linkage of an important gene locus for tuberous sclerosis to a chromosome 16 marker for polycystic kidney disease. Nat Genet 1992;2: 3741. [17] Sampson JR, Maheshwar MM, Aspinwall R, et al. Renal cystic disease in tuberous sclerosis: role of the polycystic kidney disease 1 gene. Am J Hum Genet 1997;61:84351. [18] Neumann HP, Schwarzkopf G, Henske EP. Renal angiomyolipomas, cysts, and cancer in tuberous sclerosis complex. Semin Pediatr Neurol 1998;5:26975. [19] Ewalt DH, Sheeld E, Sparagana SP, et al. Renal lesion growth in children with tuberous sclerosis complex. J Urol 1998;160:1415. [20] Kandt RS, Gebarski SS, Goetting MG. Tuberous sclerosis with cardiogenic cerebral emboli: magnetic resonance imaging. Neurology 1985;35:12235. [21] Kandt RS, Pericak-Vance MA, Hung W-Y, et al. Absence of linkage of ABO blood group locus to familial tuberous sclerosis. Exp Neurol 1989;104:2238. [22] van Slegtenhorst M, de Hoogt R, Hermans C, et al. Identication of the tuberous sclerosis gene TSC1 on chromosome 9q34. Science 1997;277:8058. [23] The European Chromosome 16 Tuberous Sclerosis Consortium. Identication and characterization of the tuberous sclerosis gene on chromosome 16. Cell 1993;75:130515. [24] Gutmann DH. Parallels between tuberous sclerosis complex and neurobromatosis 1: common threads in the same tapestry. Semin Pediatr Neurol 1998;5:27686. [25] Dabora SL, Jozwiak S, Franz DN, et al. Mutational analysis in a cohort of 224 tuberous sclerosis patients indicates increased severity of TSC2, compared with TSC1, disease in multiple organs. Am J Hum Genet 2001;68:6480. [26] van Slegtenhorst M, Verhoef S, Tempelaars A, et al. Mutational spectrum of the TSC1 gene in a cohort of 225 tuberous sclerosis complex patients: no evidence for genotypephenotype correlation. J Med Genet 1999;36:2859. [27] Nellist M, van Slegtenhorst MA, Goedbloed M, et al. Characterization of the cytosolic tuberin-hamartin complex. Tuberin is a cytosolic chaperone for hamartin. J Biol Chem 1999;274:3564752. [28] Au KS, Hebert AA, Roach ES, et al. Complete inactivation of the TSC2 gene leads to formation of hamartomas. Am J Hum Genet 1999;65:17905. [29] Roach ES, Gomez MR, Northrup H. Tuberous sclerosis complex consensus conference: revised clinical diagnostic criteria. J Child Neurol 1998;13:6248. [30] Jozwiak S, Kawalec W, Dluzewska J, et al. Cardiac tumours in tuberous sclerosis: their incidence and course. Eur J Pediatr 1994;153:1557. [31] Roach ES, DiMario FJ, Kandt RS, et al. Tuberous Sclerosis Consensus Conference: recommendations for diagnostic evaluation. National Tuberous Sclerosis Association. J Child Neurol 1999;14:4017. [32] Korf BR. Diagnostic outcome in children with multiple cafe au lait spots. Pediatrics 1992;90:9247. [33] Neville HL, Seymour-Dempsey K, Slopis J, et al. The role of surgery in children with neurobromatosis. J Pediatr Surg 2001;36:259. [34] DeBella K, Szudek J, Friedman JM. Use of the National Institutes of Health criteria for diagnosis of neurobromatosis 1 in children. Pediatrics 2000;105:60814. [35] Packer RJ, Ater J, Allen J, et al. Carboplatin and vincristine chemotherapy for children with newly diagnosed progressive low-grade gliomas. J Neurosurg 1997;86:74754.
964
[36] Venes JL, Latack J, Kandt RS. Postoperative regression of opticochiasmatic astrocytoma: a case for expectant therapy. Neurosurgery 1984;15:4213. [37] Grill J, Couanet D, Cappelli C, et al. Radiation-induced cerebral vasculopathy in children with neurobromatosis and optic pathway glioma. Ann Neurol 1999;45:3936. [38] Pollack IF, Shultz B, Mulvihill JJ. The management of brainstem gliomas in patients with neurobromatosis 1. Neurology 1996;46:165260. [39] Chung AK, Michels V, Poland GA, et al. Neurobromatosis with involvement of the prostate gland. Urology 1996;47:44851. [40] McTaggart SJ, Gelati S, Walker RG, et al. Evaluation and long-term outcome of pediatric renovascular hypertension. Pediatr Nephrol 2000;14:10229. [41] Andrews I, Ryan MM, Kandt RS. Genetic causes of pediatric stroke. In: Alberts MJ, editor. Genetics of cerebrovascular disease. Armonk, NY: Futura Publishing; 1999. p. 273. [42] DeBella K, Poskitt K, Szudek J, et al. Use of unidentied bright objects on MRI for diagnosis of neurobromatosis 1 in children. Neurology 2000;54:164651. [43] MacCollin M, Kwiatkowski D. Molecular genetic aspects of the phakomatoses: tuberous sclerosis complex and neurobromatosis 1. Curr Opin Neurol 2001;14:1639.
Neurol Clin N Am 20 (2002) 965981
Neuroradiology of the central nervous system in childhood

Gary L. Hedlund, DO*
Department of Pediatric Medical Imaging, Primary Childrens Medical Center, University of Utah, Salt Lake City, UT, USA
The introduction of clinical magnetic resonance scanners in the late 1970s provided a new window into understanding the CNS. Subsequent renements in computer hardware, software, and gradient coils led to improved image quality, faster scan times, and advanced capabilities, such as magnetic resonance angiography (MRA), magnetic resonance venography (MRV), magnetic resonance spectroscopy (MRS), diusion and perfusion imaging, and cortical activation. Although MRI and advanced magnetic resonance applications provide the backbone for CNS imaging, other techniques such as CT angiography, cranial sonography, and catheter angiography have important roles in the diagnosis and treatment of pediatric CNS disease. In the child with seizures refractory to medical management, nuclear medicine single photon emission computed tomography (SPECT) and positron emission tomography (PET) scans are integral in evaluation and preoperative planning. The purpose of this article is to familiarize readers with new imaging applications, describe the strengths and weaknesses of medical imaging technology, and emphasize the practical aspects of imaging the pediatric CNS. Because of the advances in MRI, most of this article is devoted to the role of MRI in evaluating children with diseases of the CNS. Advanced applications of magnetic resonance imaging Diusion-weighted magnetic resonance imaging Diusion-weighted magnetic resonance imaging (MRI) is a powerful tool for noninvasive neuroimaging. The signal intensity in a diusion-weighted
* Correspondence. Department of Pediatric Medical Imaging, Primary Childrens Medical Center, 100 North Medical Drive, Salt Lake City, UT 84113, USA. E-mail address: pcghedlu@ihc.com (G.L. Hedlund). 0733-8619/02/$ - see front matter 2002, Elsevier Science (USA). All rights reserved. PII: S 0 7 3 3 - 8 6 1 9 ( 0 2 ) 0 0 0 2 0 - 8
966
G.L. Hedlund / Neurol Clin N Am 20 (2002) 965981
image is a function of the random translational motion of water molecules (Brownian motion). Neurologic conditions that produce cell membrane depolarization, such as acute ischemic stroke, result in cytotoxic edema,one of the underlying mechanisms responsible for diusion-weighted changes in the brain [1,2]. In cytotoxic edema, the apparent diusion coecient (ADC) of water in ischemic tissue is reduced relative to that of normal brain water and allows the ischemic territory to be visualized as a hyperintense region. Water ADC values are also aected by the presence and orientation of biologic barriers to translational water motion, such as cell membranes and myelin bers [3]. Because diusion-weighted imaging is mapped to a T2weighted image, any lesion with a prolonged T2 value may appear hyperintense on diusion-weighted images. ADC maps reveal an initial restrictive diusion coecient within minutes of infarction, and the ADC remains abnormally low for approximately 4 to 6 days. ADC values normalize at 7 to 10 days, but become high after approximately 10 to 14 days [3,4]. Diusion-weighted imaging has an essential role in the management of acute clinical stroke. Cerebral ischemia arises from a reduction in the brain tissue blood supply because of an obstruction of inowing arterial blood. When cerebral blood ow falls below a critical threshold, metabolic energy failure disrupts water and electrolyte homeostasis and leads to cytotoxic edema (Fig. 1). A similar perfusion-decit threshold has been shown to coincide with the onset of ADC changes, and a correlation exists between ADC reduction and cerebral cytotoxic edema [35]. Because diusion-weighted imaging normalizes by 10 to 14 days after infarction, this technique has proven useful for distinguishing recent stroke from remote infarction. The utilization of diusion-weighted imaging in the neonate, infant, and child provides important insights into such pathologic processes as anoxic ischemic encephalopathy, clinically signicant hypoglycemia, status epilepticus, and
Fig. 1. Diusion-weighted MRI in a 15-year-old girl with acute lymphocytic leukemia receiving L-asparaginase, who suered a devastating clinical stroke 10 hours before this scan. Cytotoxic edema is seen as bright signal within the right hemisphere.
967
neuronal necrosis secondary to encephalitis. In large part, T2-weighted images and uid attenuated inversion recovery (FLAIR) images have proven to be relatively insensitive in the early detection of these pathologic states. Diusion-weighted imaging has become an integral part of brain imaging when evaluating the possibility of early infarction and in dierentiating early infarction from chronic infarction in children with sickle cell disease, congenital heart disease, bacterial endocarditis, hypercoagulable states, vasculitides, arterial dissection, near drowning, hypoglycemia, mitochondrial encephalopathy with lactic acidosis and stroke-like episodes (MELAS), and other disorders predisposing to stroke [1]. In children with traumatic brain injuries, diusion-weighted imaging may reveal areas of restricted diusion secondary to cellular injury. Diusion restriction may be observed in contusional, hemorrhagic, and penetrating injuries of the brain. Diusion-weighted imaging has become a particularly powerful tool in detecting superimposed posttraumatic infarction that may be silent on early CT and standard MRI scans of the abused child. Explanations for these diusion changes after abuse include profound or prolonged hypoxia, hypotension, hypoventilation, seizures, suocation, strangulation, and cerebral edema [1,2]. Encephalitis, meningitis, vasculitis, and demyelinating diseases may have overlapping clinical and MRI ndings. In bacterial meningitis, diusionweighted imaging has a role in dierentiating reversible processes, such as inammatory vasogenic edema, from cytotoxic edema, and brain infarctions. In acute disseminated encephalomyelitis (ADEM), most lesions appear isointense to normal brain on diusion-weighted images, suggesting that the demyelinating process has not caused permanent cell injury [1,2]. In addition to the low ADC values observed in acute stroke and severe hypoglycemia, low ADC values can reect the presence of viscous material, such as the contents of a brain abscess. Therefore, diusion-weighted imaging can play a role in the evaluation of a cavitary ring-enhancing lesion, distinguishing abscess from CNS tumor. Some chemotherapy and transplant-associated medications, such as cyclosporin, are associated with CNS toxicity manifested by T2 hyperintensity on brain MRI (the reversible posterior leukoencephalopathy syndrome). Diusion-weighted imaging has been helpful in dierentiating irreversible or cytotoxic-associated abnormalities from those changes that are likely to be reversible. Several technical factors and artifacts can compromise diusion-weighted imaging. Fat has a low ADC value and therefore is bright on diusionweighted imaging. Magnetic susceptibility eects, such as those seen with shunt connectors, surgical clips, and dental metallic hardware, can create obstacles to performing clinical diusion-weighted imaging. Additionally, motion artifacts, such as patient head motion, breathing, and cardiac driven cerebrospinal uid (CSF) pulsation, can compromise diagnostic information.
968
Noninvasive vascular imaging: magnetic resonance angiography and magnetic resonance venography In the diagnostic evaluation of pediatric CNS disease, MRA and MRV represent valuable alternatives to catheter angiography and venography. From a practical standpoint, MRA adds no more than 5 to 7 minutes to the MRI examination. MRA yields information regarding signal from protons in owing blood. Flow direction, volume, and velocity aect the ow signal and apparent size of the vessel being studied. The MRA techniques most commonly used in the clinical setting of pediatric neuroradiology are time-of-ight (TOF) MRA and phase contrast angiography (PCA) (Fig. 2). TOF techniques take advantage of the dierences in signal amplitude between stationary tissue and owing blood (ow-related enhancement), whereas PCA exploits dierences in signal phase between owing and stationary spins [6]. Physiologic and anatomic factors inuence the image quality and interpretation of MRA/MRV studies. Relevant factors include blood ow direction relative to the imaging plane, geometry of the vessel, velocity of ow, complex patterns of ow, and T1 relaxation times of stationary tissue (eg, parenchymal hemorrhage). Maximum ow-related enhancement occurs in
Fig. 2. Three-dimensional time-of-ight magnetic resonance angiography shows occlusion of the right supraclinoid carotid artery (arrow).
969
TOF MRA when blood is owing perpendicular to the plane of section. An example would be ow within the sagittal sinus detected with coronal MRV. Flow appearing obliquely or parallel to the plane of section results in loss of signal and can suggest vessel narrowing or thrombosis. Vessels that have complex geometry or turbulent ow are subject to areas of signal loss. Another potential pitfall in the interpretation of TOF MRA is the presence of fat or methemoglobin in the region of interest. This can lead to short T1 relaxation times and shine through that may mimic pathologic changes. We avoid this pitfall by close scrutiny of the source images or by implementing a phase contrast technique. Because PCA is a true background subtraction technique, it should be used in cases of suspected vessel thrombosis, such as subacute dural venous sinus thrombosis or subacute parenchymal hemorrhage [6]. Both two-dimensional (2D) and three-dimensional (3D) techniques can be applied to TOF and PCA. The 3D techniques are best for high-resolution imaging of small vessels having complex geometry, allowing the arteries to be viewed in multiple oblique projections without a loss in spatial resolution. In 3D techniques, thin nonoverlapping slices (1 mm in thickness) are also used. In 2D TOF and 2D PCA, the slice thickness ranges from 2 to 5 mm and slices are overlapped by approximately 20% to 30%. Using these thicker slices allows coverage of larger anatomic regions, such as the pediatric neck for carotid and vertebral artery assessment [6]. Clinical applications of MRA include investigation of suspected anatomic arterial variants (eg, embryonic basilar-carotid connections), evaluation of unusual vessel turns that may mimic an aneurysm, intracranial vessel relation to tumor (displacement versus encasement), and suspected arteriovenous malformations (Fig. 3). Studying the venous side of the intracranial circulation using MRV is useful in evaluating children with suspected intracranial dural venous thrombosis, evaluation of pseudotumor cerebri, preoperative evaluation of encephalocele, and evaluation of tumors that are juxtaposed to venous structures (Figs. 4 and 5). Magnetic resonance spectroscopy Magnetic resonance spectroscopy (MRS) oers the capability to study brain metabolism in vivo. The relatively short acquisition times make it possible to perform MRS in conjunction with conventional MRI studies even in young sedated children. MRS is proving to be helpful in the diagnostic evaluation and management of pediatric patients with brain tumors, metabolic and heritable disorders, epilepsy, demyelinating disease, hypoxic-ischemic encephalopathy, head trauma, developmental delay, and hypotonia as well as in the evaluation of T2 signal abnormalities in patients with symmetric basal ganglia signal alterations [7,8]. The primary choice in parameters with conventional MRS software is between short and long echo times (TEs). Long TE acquisitions produce
970
Fig. 3. Three-dimensional time-of-ight magnetic resonance angiography in a 12-year-old girl with headache shows an arteriovenous malformation fed by the left posterior cerebral artery (arrow).
spectra that include N-acetyl aspartate (NAA), choline (Cho), creatine/phosphocreatine (Cre), and possibly lactate (Lac) peaks. Short TE acquisitions include these same neurometabolites, myoinositol (mI), glutamate (Glu), glutamine (Glx), and possibly alanine (Ala) as well as proteins and lipids. NAA, which is present in relatively large quantities in the brain, is the largest peak in the normal spectra found at 2.02 ppm. NAA, generally believed to be a neuron-specic metabolite, has also been found in immature oligodendrocytes and astrocyte progenitor cells. Reduced NAA is found early in disorders that cause a decline in the number of neurons, impairment in neuronal metabolism, or replacement of normal neurons. In normal brain development, there is a progressive increase in NAA, particularly in the cerebral gray matter, from a ventral to dorsal direction and from the cerebral hemispheres to the level of the spinal cord (Fig. 6) [9,10].
971
Fig. 4. Normal phase contrast magnetic resonance venography in sagittal projection shows normal ow signal within the sagittal sinus.
Cho, found at 3.2 ppm, primarily represents phospho-/glycerophospho-/ phosphatidyl-choline. An elevation in total Cho may reect increased membrane turnover, such as one might see with cell proliferation secondary to gliosis, cerebral neoplasm, or in the normal actively myelinating infant brain. Cho can also rise with increased membrane breakdown in demyelinating and dysmyelinating disorders. Regional variation in Cho is found between white matter and gray matter [7]. Creatine (Cre) is found at two locations, 3.0 ppm and 3.9 ppm. These peaks, indicate energy stores of the brain, reecting a combination of Cre and phosphocreatine. There is a relatively high concentration of Cre in the brain with a progressive increase from white matter to gray matter and an increase from the cerebral hemispheres to the cerebellum. The Cre peak is generally accepted as an internal standard of reference [7,9,10]. Lactate is undetectable in the normal spectra. The lactate peak can be visualized, however, in any condition that causes anaerobic glycolysis. This includes energy metabolism disorders, hypoxia, stroke, neoplasms, or seizures. Lactate can be detected in macrophages and therefore may be seen with acute inammation. Lactate also accumulates in necrotic and infarcted tissue. When present, we note it as a doublet at 1.3 ppm [11,12]. Myoinositol (mI) (at 3.56 ppm) is not present in neurons and may be glia specic, because variations are often linked to myelin formation and myelin
972
Fig. 5. Coronal phase contrast magnetic resonance venography shows thrombus within the left transverse sinus (arrow).
breakdown. This is not surprising, because mI represents a component of the membrane phospholipid. mI, an important osmotic regulator, serves a reserve pool for inositol diphosphate, which acts as a second messenger [13]. Glutamate (Glu), glutamine (Gln), and gamma-amino glutaric acid (GABA), which are generally inseparable at 1.5 T, result in a complex peak seen just to be the left of the NAA peak (between 2.1 and 2.5 ppm). Glu, an excitatory amino acid found in neurons, participates in the regulation of fatty acid synthesis and in the Krebs cycle. Glu also has a role in ammonia regulation but is restricted to the cerebral astrocytes. GABA, a product of Glu, functions as an inhibitory neurotransmitter [12]. Lipids produce peaks at 0.9 to 2.0 ppm, reecting methyl protons of lipids (macromolecules, methylene protons of neutral lipids, lactate, and cytosolic protein macromolecules). Lipid peaks may be found in the setting of a necrosis, high-grade tumors, meningiomas, myelin breakdown, and certain inborn errors of metabolism. The developing brain displays variations in the content of certain neurometabolites. The NAA content, low at birth, increases rapidly for the rst 2 years of life. Because most neurons form during gestation, the postnatal changes presumably reect changes in the number and size of neuronal dendrites, the number and complexity of dendritic/axonal connections, and the organization and dierentiation of neurons and their connections. Choline
973
Fig. 6. Normal white matter spectra (magnetic resonance spectroscopy) (echo time 288) in a 2year-old boy being evaluated for developmental delay. NAA N-acetyl aspartate; Cr creatine and phosphocreatine; Cho choline compounds. X axis parts per million (ppm).
shows a relative decline in amplitude in the rst 2 to 3 years of life, reecting progressive myelination. Creatine rises to adolescent or young adult levels by approximately 4 months of age. A prominent myoinositol peak is seen at birth but rapidly declines in the rst 3 to 4 months, reaching adult levels by the end of the rst year [7]. Clinical applications of magnetic resonance spectroscopy Neoplasms Considerable interest has been generated regarding the utilization of MRS in the evaluation of intracranial neoplasms. Most studies suggest certain trends, a low NAA/Cre ratio, and an elevated Cho/Cre ratio. Some tumors show elevated lactate and lipids as well. The NAA content is reduced, because the tumor replaces normal neurons or the tumor cells produce lower amounts of NAA than normal neurons. Cre can be reduced if energy stores are reduced by the high metabolic demand of tumor. The Cho content, elevated secondary to increased membrane turnover, rapid tumor cell growth, and proliferation of the tumor, tends to be more prominent in aggressive or anaplastic tumors. Although lactate can be observed in the setting of cerebral neoplasms, its presence is not predictive of tumor grade. The presence of lactate depends on the level of glycolytic activity. Elevated lipid levels in the context of elevated
974
lactate are more suggestive of an anaplastic or aggressive tumor. Additional clinical questions that are raised in the context of a cerebral neoplasm include distinguishing residual tumor from postoperative changes, assessing tumor progression from radiation necrosis, determining the nature of an unknown mass (a primary/secondary tumor, infectious lesion, or tumefactive demyelinating disease), and assessing if a lesion is benign or malignant. MRS can be used to evaluate a cystic brain mass when the considerations include abscess or necrotic tumor. Not only can lactate and lipids be present, but abscesses usually contain products of fermentation and leukocyte lysis, such as acetate and succinate, and amino acids, such as valine and leucine, and products of proteolysis. A number of unknown marker peaks for infection have been described at 2.2, 2.9, 3.2, 3.4, and 3.8 ppm [14,15]. Epilepsy The preoperative evaluation of children with epilepsy includes analysis of seizure semiology, neuropsychologic testing, video electroencephalographic (EEG) monitoring, MRI, and occasionally SPECT or PET. Accurate localization of a structural abnormality by MRI and MRI results that are concordant with other clinical information increase the likelihood that the child will be seizure-free after epilepsy surgery [16]. PET serves as an alternative screening procedure in patients who have discordant clinical data [34]. PET has a high sensitivity and specicity for identifying seizure foci. The cost, regional localization, and limited availability, however, make PET a somewhat less attractive modality [17,18]. MRS has been shown to be helpful in the lateralization of epileptogenicity in children with suspected mesiotemporal lobe seizures. The MRI features in children with mesiotemporal sclerosis consist of more subtle volume loss and/or signal abnormalities in the hippocampal formations than might be seen in adults. At least one study has shown an advantage in sensitivity in MRS compared with PET, presumably reecting the enhanced capability of MRS to detect subtle neuronal dysfunction [16,19]. Metabolic diseases and white matter diseases The MRS ndings in leukodystrophies and metabolic disorders are variable. Many of these conditions are associated with a reduction in NAA as a result of neuronal/axonal degeneration and loss of brain parenchyma. Metachromatic leukodystrophy (MLD), a dysmyelinating disorder, has a nonspecic CT and MRI appearance. In addition to showing reductions in NAA, the spectra of children with MLD identify mild increases in lactate, large reductions in Glu, and striking elevations in mI. Myoinositol, a distinguishing nding on MRS in MLD, may actually play a role in the pathophysiology (myelin membrane instability) of the disorder [20]. MRS plays an important role in conrming the diagnosis of Canavan disease; the underlying deciency in aspartoacylase, and resultant accumulation of NAA leads to striking elevations in the NAA/Cho and NAA/Cre ratios [21].
975
MRS has been shown to be more sensitive than MRI for detection of abnormal white matter in children with adrenoleukodystrophy (ALD). Consequently, MRS can be used to monitor the progression and ecacy of therapy in these patients. In ALD, MRS shows a reduction in the NAA/ Cre ratio; a reduced NAA/Cho ratio; an elevation in the Cho/Cre ratio; and an increase in Glx, mI, and lipids. Lactate may be present as well [8,12]. MRS frequently identies elevated lactate content in patients with disorders of energy metabolism. Lactate is thought to be an indicator of impaired mitochondrial oxidative metabolism. Some of the disorders in which MRS has been used in this regard include mitochondrial myopathy, MELAS, myoclonus, epilepsy with ragged red bers (MERFF), Leigh syndrome, and Kearns-Sayre syndrome. MRS can also be used to detect certain amino acidopathies and to monitor disease activity. The enzyme deciency in children with maple syrup urine disease causes an accumulation of branch chain amino acids and their 2-oxyacids. The latter can be recognized as a small inverted peak at 0.9 ppm and additional pathologic peaks at 2.4 and 4.2 ppm. In nonketotic hyperglycinemia, a glycine (Gly) peak is detected at 3.55 ppm [8]. Global cerebral insult There is considerable interest in the application of MRS in the evaluation of neonatal asphyxia, near drowning, and traumatic brain injuries. Although CT answers urgent management questions and MRI provides additional imaging details, neither CT or MRI detects extensive cellular damage that may be evolving. In traumatic brain injuries, the spectrum of MRS ndings includes reduced NAA/Cre ratios reecting neuroaxonal injury; elevated Cho content indicating diuse axonal injury and membrane turnover; low NAA, Cho, and mI in the setting of inappropriate diuretic hormone secretion (SIADH); and increased Cho, Cre, mI, or NAA in hyperosmolar states after trauma. The presence of lactate correlates with poor neurologic outcome in children who have experienced perinatal asphyxia or near drowning. Lactate appears in the rst 24 hours and persists for at least 48 hours in the setting of asphyxia. If the child or infant is not studied within the rst 2 or 3 days after the event, a reduction in the NAA may be a more useful predictor of poor outcome (Fig. 7) [11,22]. Cortical activation (functional MRI) imaging in children The ability to combine brain function with detailed anatomy is a powerful tool that can be used to understand physiologic changes in the pediatric patient. Activation functional MRI (fMRI) utilizes the blood oxygenation leveldependent (BOLD) technique to examine changes in the oxy-/deoxyhemoglobin ratio after activities designed to stimulate a particular area of the brain. fMRI in children is more challenging than in adults owing to a
976
Fig. 7. Intermediate echo time (144) magnetic resonance spectroscopy of the occipital cortex in a 4-year-old boy 4 days out from drowning. There is reduction of N-acetyl aspartate, elevation of choline, and the presence of lactate. X axis parts per million (ppm).
need for cooperation in performing tasks, such as motor and language paradigms. Most of the current paradigms do not allow for sedation, although there are novel techniques to perform fMRI in sedated infants and children [23]. A common indication for fMRI is to noninvasively map functional brain tissue in localized areas of eloquence with respect to pathologic areas of the brain. This has tremendous utility in surgical planning. Finger-thumb opposition can be used to outline the sensorimotor cortex and dene displacement of the central sulcus in relation to space-occupying lesions. The utilization of liquid crystal display (LCD) goggles stimulates the visual cortex. Utilization of covert letter and word generation paradigms identies hemispheric dominance for language before temporal lobectomy in children with intractable complex partial seizures. In selected patients, utilization of a covert word generation technique permits the determination of hemispheric dominance for language; when combined with a memory paradigm, this technique may eliminate the need for more invasive Wada testing in some children before surgery [23,24].
Advanced computerized tomography applications Conventional CT remains an important clinical tool in the assessment of acute head injury, acute neurologic decline, cranial-facial and cerebral
977
malformations, investigation of infectious intracranial processes, and the follow-up of patients with CSF diversionary shunts. Computerized tomography angiography CT angiography (CTA) is a less invasive alternative to conventional radiographic angiography. There is a wide clinical application of CTA in the adult population for the investigation of head and neck vessels, the aorta, and retroperitoneal arteries and veins. Overall, there is more limited experience with the use of CTA in young children. CTA of the head and neck has an advantage compared with other body locations in that it is not limited by respiratory artifacts; therefore, high-resolution images can be obtained [25]. Indications for CTA in the pediatric population include patients who are not candidates for the MRI environment, such as children with pacemakers or patients who have aneurysms and/or other surgical clips juxtaposed to areas of vascular interest. CTA can also rapidly provide highly detailed information regarding vascular anatomy before surgery, such as in the case of a patient with acute neurologic decline, suspected arterial dissection, or cerebral hematoma, in whom CTA provides a useful presurgical map of vascular anomalies or an aneurysm nidus (Fig. 8) [25]. Advanced ultrasound techniques for pediatric cranial imaging Transcranial Doppler sonography in children Transcranial Doppler (TCD) sonography is a noninvasive technique that uses a 2- to 2.5-MHz Doppler transducer to measure the velocity and pulsatility of blood ow within the intracranial arteries of the circle of Willis and the vertebrobasilar system. The clinical indications for TCD in children include evaluating children with various vasculopathies, such as sickle cell disease and moya moya syndrome; evaluating arteriovenous malformations; conrmation of clinical brain death; evaluating children with asphyxia and cerebral edema; monitoring children with hydrocephalus and subdural eusions; evaluating and following patients with vasospasm, especially after subarachnoid hemorrhage; monitoring children during cerebrovascular and cardiovascular interventional and surgical procedures; and investigating migraine headaches [2628]. Nuclear medicine single photon emission computed tomography/positron emission tomography brain imaging Functional brain imaging refers to a set of techniques used to derive images reecting biochemical, physiologic, or electrical properties of the CNS. The most practical of these nuclear medicine techniques are SPECT and PET. Of these two, SPECT may be the more widely available technique.
978
Fig. 8. Coronal CT angiography in a 15-year-old boy with acute onset right hemiplegia after a neck injury shows diminished caliber of the left supraclinoid carotid consistent with nonocclusive dissection (arrow).
Although the diversity of radiopharmaceuticals available for brain SPECT is not as great as those for PET, the variety of SPECT tracers is expanding rapidly [29,34]. SPECT tracers, such as Technetium-99m hexamethylpropylene amine oxime (99m Tc-HMPAO), is extracted by brain tissues on the rst arterial pass after intravenous injection and is retained within the brain for several hours. Typically, there is symmetric distribution of the radiotracer within both hemispheres. The basal ganglia, occipital cortex, and cerebellum often seem to have slightly increased normal uptake compared with other regions of the brain. Reduced tracer uptake may be seen in association with cerebral ischemia. Conversely, regions of increased radiotracer uptake reect a hypermetabolic focus, such as one might see with seizure activity or luxury perfusion of the brain [2931]. PET imaging requires a cyclotron in which an isotope target is bombarded with a beam of protons (hydrogen nuclei stripped of electrons). The most commonly administered PET radionuclide, 18F, is used for the production of 18F-2-deoxy-2-uor-D-glucose (18FDG). PET imaging has been limited by the availability of cyclotron facilities, cost of instrumentation, and short half-lives of the radioisotope [34].
979
Both SPECT and PET have become indispensable in evaluation of medically intractable epilepsy. Essential to the success of seizure surgery is the accurate localization of the seizure focus. SPECT/PET imaging can be performed at a seizure-free baseline (interictal) or be performed during a seizure (ictal) (Fig. 9). The choice of options in imaging procedures (SPECT or PET) depends, in part, on the institutional logistics in performing the studies and the stability of the available radiotracers. Tracers that do not have good stability over time or have short half-lives are dicult to prepare and deliver to a convulsing child. During the interictal period, the seizure focus is identied as an area of reduced perfusion/metabolism. Baseline interictal scans can have a normal appearance, however. During a seizure, there is an area of increased perfusion/metabolism identifying the seizure focus [17,18,3033].
Catheter angiography Although vascular imaging, such as MRA and MRV, has often supplanted the necessity for catheter angiography in the infant or child with CNS disease, there remain several important indications for catheter angiography. These include the diagnosis, characterization, and follow-up of arteriovenous malformations; the evaluation of nonocclusive arterial dissection
Fig. 9. Ictal Tc-HMPAO (Technetium-99m hexamethylpropylene amine oxime99m TcHMPAO) single photon emission computed tomography scan (coronal images) shows increased tracer uptake in the left temporal lobe (dashed arrow) and right cerebellar hemisphere (solid arrow) (crossed cerebellar diaschisis).
980
in the patient with stroke; preoperative evaluation of certain cerebral tumors (eg, meningioma); Wada testing; and inferior petrosal venous sampling in patients with Cushing syndrome and normal pituitary imaging. Major therapeutic applications of CNS catheter angiography/venography include embolization of tumors and clot removal in children with intracranial venous sinus thrombosis. Advances in spiral and 3D angiography have diminished contrast requirements and examination times for catheter cerebral angiography, which, in turn, diminishes the complications of catheter angiography in the pediatric patient.
References
[1] Gray L, MacFall J. Overview of diusion imaging. Neuroimaging Clin North Am 1998;6:12538. [2] Le Bihan D, Turner R, Douek P, et al. Diusion MR imaging: clinical applications. AJR Am J Roentgenol 1992;159:5919. [3] Kucharczyk J, Vexler ZS, Roberts TP, et al. Echo-planar perfusion-sensitive MR imaging of acute cerebral ischemia. Radiology 1993;188:7117. [4] Lutsep AR, Albers GW, DeCrespigny A, et al. Clinical utility of diusion-weighted magnetic resonance imaging in the assessment of ischemic stroke. Ann Neurol 1997; 41:57580. [5] Sorenson AG, Buonanno FS, Gonzalez RG, et al. Hyperacute stroke: evaluation with combined multisection diusion-weighted and hemodynamically weighted echo-planar MR imaging. Radiology 1996;199:391401. [6] Sheppard MS. Basic concepts of magnetic resonance angiography. Radiologic Clin North Am 1995;33:91113. [7] Cecil KM. Technical aspects of performing and interpreting proton MRS. In: Ball WS, editor. Atlanta, GA: American Society of Neuroradiology Advanced Imaging Symposium; 2000. p. 2936. [8] Pouwels PJW, Frahm J. Regional metabolite concentrations in human brain as determined by quantitative localized proton MRS. Magn Reson Imaging Med 1998;39:5360. [9] Birken DL, Oldendorf WH. N-acetyl-L-aspartic acid: a literature review of a compound prominent in 1H-NMR spectroscopic studies of brain. Neurosci Biobehav Rev 1989;13: 2331. [10] Urenjak J, Williams SR, Gadian DG, et al. Specic expression of N-acetyl aspartate in neurons, oligodendrocyte-type 2 astrocyte progenitors, and immature oligodendrocytes in vitro. J Neurochem 1992;59:5561. [11] Dubowitz DJ, Bluml S, Arcinue E, et al. MR of hypoxic encephalopathy in children after near drowning: correlations with quantitative proton MR spectroscopy and clinical outcome. AJNR Am J Neuroradiol 1998;19:161727. [12] Ross BD. Biochemical consideration in 1H spectroscopy. Glutamate and glutamine; myoinositol and related metabolites. NMR Biomed 1991;4:5963. [13] Lee J, Arcinue E, Ross B. Organic osmolytes in the brain of an infant with hypernatremia. N Engl J Med 1994;33l:43942. [14] Dowling C, Bollen AW, Noworolski SM, et al. Preoperative proton MR spectroscopic imaging of brain tumors: correlation with histopathologic analysis of resection specimens. AJNR Am J Neuroradiol 2001;22:60412. [15] Tzika AA, Vigneron DB, Dunn RS, et al. Intracranial tumors in children: small single-voxel proton MR spectroscopy using short- and long-echo sequences. Neuroradiology 1996; 38:25463.
981
[16] Aasly J, Silfvenius H, Aas TC, et al. Proton magnetic resonance spectroscopy of brain biopsies from patients with intractable epilepsy. Epilepsy Res 1987;35:2117. [17] Markland ON, Salanova V, Worth R, et al. Comparative study of interictal PET and ictal SPECT in complex partial seizures. Acta Neurol Scand 1997;95:12936. [18] Ryvlin P, Phiippon B, Cinotti L, et al. Functional neuroimaging strategy in temporal lobe epilepsy: a comparative study of 18 FDG-PET and 99mTc-HMPAO-SPECT. Ann Neurol 1990;27:1626. [19] Petro OAC, Pleban LA, Spencer DD. Symbiosis between in vivo and in vitro NMR spectroscopy. The creatine, N-acetyl aspartate, glutamate, and GABA content of the epileptic human brain. Magn Reson Imaging 1995;13:1196211. [20] Van der Knaap MS, Breiter SN, Naidu S, et al. Dening and categorizing leukoencephalopathies of unknown origin: MR imaging approach. Radiology 1999;213:12133. [21] Matalon R, Michals K, Sevesta D, et al. Aspartoacyclase deciency and N-acetylaspartic aciduria in patients with Canavans disease. Am J Med Genet 1988;20:46371. [22] Haseler LJ, Arcinue E, Danielsen ER, et al. Evidence from proton magnetic resonance spectroscopy for a metabolic cascade of neuronal damage in shaken baby syndrome. Pediatrics 1997;99:414. [23] Belliveau JW, Kennedy DN, McKinistry RC, et al. Functional mapping of the human visual cortex by magnetic resonance imaging. Science 1991;254:7169. [24] Hunter JV, Brown L, Haselgrove J, et al. Comparison of hemispheric dominance for language as identied by Wada testing versus functional MRI in a pediatric population. Presented at the American Epilepsy Society Meeting. Baltimore, 1995. [25] Harbaugh RE, Schlussberg DS, Jeery R, et al. Three dimensional computerized tomography angiography in the diagnosis of cerebrovascular disease. J Neurosurg 1992;76:40814. [26] Lupetin AR, Davis DA, Beckman I, et al. Transcranial Doppler sonography. Part I. Principles, technique, and normal appearances. Radiographics 1995;15:17991. [27] Seibert JJ, McCowan TC, Chadduck WM, et al. Duplex pulsed Doppler US versus intracranial pressure in the neonate. Clinical and experimental studies. Radiology 1989; 171:1559. [28] Taylor GA. Current concepts in neonatal cranial Doppler sonography. Ultrasound Quarterly 1992;4:22344. [29] Alavi A, Hirsch IJ. Studies of central nervous system disorders single-photon emission computed tomography and positron emission tomography: evolution over the past 2 decades. Semin Nucl Med 1991;21:5881. [30] Devous MS Sr, Leroy RF, Homan RW. Single photon emission computed tomography in epilepsy. Semin Nucl Med 1990;20(4)32541. [31] Harvey AS, Hopkins IJ, Bowe JM, et al. Frontal lobe epilepsy clinical seizure characteristics and localization with ictal 99m Tc-HMPAO SPECT. Neurology 1993;43: 196680. [32] Chugani HT. The use of positron emission tomography in the clinical assessment of epilepsy. Semin Nucl Med 1992;22:24753. [33] Gaillard WD, White S, Malow B, et al. FDG-PET in children and adolescents with partial seizures: role in epilepsy surgery evaluation. Epilepsy Res 1995;20:7784. [34] Welch MJ, Seigel BA (guest editors). The coming of age of PET (part 1). Sem Nucl Med 1998;28:247267.
Neurol Clin N Am 20 (2002) 9831011
HIV-1 infection and AIDS

Anita L. Belman, MD*
Departments of Neurology and Pediatrics, School of Medicine, State University of New York Stony Brook, NY, USA
The clinical and immunologic features of pediatric AIDS were described in 1983 within a short time after the disease was recognized in adults (homosexual men, injecting drug users [IDUs], and women whose only risk factors were sexual relations with men who were bisexual or IDUs or both) [14]. The following year, neurologic involvement was reported in children. A progressive encephalopathy (PE) syndrome was described, as was a more stable neurologic impairment [5,6,47]. It was recognized that by the time HIV-1 had advanced to AIDS, neurologic complications were frequent [6,813]. Over the next 10 years, the numbers of persons with AIDS escalated. By the end of the decade, AIDS had reached epidemic proportions. AIDS had become one of the leading causes of childhood morbidity and mortality in this country and worldwide [4]. Since then, much has been learned about the biology of HIV-1 and the cells it infects. Much has been learned about viral transmission and the natural history of HIV-1 infection. Key studies led to a better understanding of maternal-infant transmission and strategies for interrupting transmission, resulting in a signicant reduction in vertically transmitted disease. More procient diagnostic techniques made early diagnosis and identication of HIV-1infected infants possible during asymptomatic or mildly symptomatic disease stages. Major advances in treatment of primary HIV-1 infection with highly active antiretroviral therapy (HAART) and immunomodulating therapies along with successful prophylaxis for life threatening opportunistic infections (OIs), and more aggressive and successful medical management of HIV-1/AIDSassociated conditions have resulted in prolonged survival of HIV-1infected persons [14]. In developed countries (and in countries where these therapies are readily available), there has been a dramatic decline in the numbers of infants born
* Correspondence. Department of Neurology, HSC T 12-020, School of Medicine, State University of New York at Stony Brook, Stony Brook, NY 117948121, USA. E-mail address: abelman@notes.cc.sunysb.edu (A.L. Belman). 0733-8619/02/$ - see front matter 2002, Elsevier Science (USA). All rights reserved. PII: S 0 7 3 3 - 8 6 1 9 ( 0 2 ) 0 0 0 1 9 - 1
984
A.L. Belman / Neurol Clin N Am 20 (2002) 9831011
HIV-1 infected and markedly improved length of survival of those infected [1520], Children with HIV-1/AIDS are living longer. A signicant and increasing percentage of children are surviving into adolescence and young adulthood [15,19,20]. Pediatric HIV-1/AIDS has become a chronic disease. Unfortunately, this is not the case worldwide, where the complexity of medical care required and expensive drug therapies are limited [4,20,21]. The clinical, virologic, and immunologic features of pediatric HIV-1 infection and AIDS have recently been reviewed in the Pediatric Clinics of North America [14]. This article covers neurologic disorders. Salient features of pediatric HIV/AIDS (Tables 13) (Box 1) and its epidemiology are briey summarized, because neurologic aspects need to be viewed in the context of this complex and progressive illness, whose epidemiology is diverse and changing, as is the availability of treatment. A list of diagnoses that indicate AIDS is included in the box on this page. Epidemiology Now into the third decade of the epidemic, more than 60 million people worldwide are estimated to have been infected with HIV-1. By the end of 1999, 18.8 million people had died of AIDS; and of these persons, 3.8 million were children. HIV/AIDS is now the fourth leading cause of death worldwide [21,22]. A changing epidemic At the end of 2001, an estimated 40 million persons worldwide were living with HIV/AIDS. Of these persons, 37.2 million were adults and 2.7 million
Table 1 Centers for Disease Control and Prevention pediatric HIV classication (1994)a Clinical signs/symptoms Immunological No evidence of suppression Evidence of moderate suppression Severe suppression
a
N: No N1 N2 N3
A: Mild A1 A2 A3
B: Moderate B1 B2 B3
C: Severe C1 C2 C3
Centers for Disease Control and Prevention (CDC) revised classication system for HIV infection of children less than 13 years of age. Initially developed for surveillance purposes, the CDC classication for pediatric AIDS is based on three parameters: (1) infection status, (2) immunological status, and (3) clinical status. Clinical conditions are specied (see Table 3 and Box 1) and pediatric age-related denitions of immune suppression (see Table 2), and more recently, HIV-1 RNA copy numbers (viral markers) are delineated. The scheme reects the childs disease stage, establishes mutually exclusive categories and is useful for monitoring therapy and assessing disease progression. From Centers for Disease Control and Prevention. 1994 Revised classication system for human immunodeciency virus infection in children less than 13 years of age; ocial authorized addendahuman immunodeciency virus infection codes and ocial guidelines for coding and reporting ICD-9-CM. MMWR weekly 1994;43: 110 [122].
A.L. Belman / Neurol Clin N Am 20 (2002) 9831011 Table 2 Immunological categoriesa,b No/lL (%) Immune category Category 1 No suppression (N1) Category 2 No suppression (N2) Category 3 No suppression (N3) <12 months of age !1500 (!25%) 7501499 (15%24%) <750 (<15%) 15 years !1000 (!25%) 500 (15%24%)999 <500 (<15%)
985
612 years 500 (!25%) 200499 (15%24%) <200 (<15%)
a CD4 T-Lymphocyte Counts: The CD4 absolute number that identies a specic level of immunosuppression changes with age. Generally, healthy non HIV-1 infected infants have considerably higher counts than do non-infected adults. The counts slowly decline to adult values by approximately age 6 years. CD4 T-lymphocyte counts or percentages are used in conjunction with HIV-1 RNA copy numbers [122]. b HIV-1 RNA copy numbers: HIV-1 RNA copy numbers in infants and young children dier form those of infected adults. Infants have a prolonged period of high HIV-1 RNA copy numbers, perhaps as a result of the lower eciency of an immature but developing immune system in containing viral replication, or a greater number of HIV-1susceptible cells, or both. RNA copy number slowly declines during the rst several years after birth, with the most rapid decline occurring during the rst year of life. A slower decline continues until about 4 to 5 years of age. High HIV-1 RNA levels [>299,000 copies per milliliter] in infants younger than the age of 12 months appear to correlate with rapid disease progression and death [30]. HIV-1 RNA copy number is used as a guide to initiate, monitor, and change therapy.
were children. During 2001, 3 million HIV-1infected persons died; approximately 580,000 were children [21]. Of the 5 million people newly infected with HIV-1 during 2001, 1.8 million were women and 800,000 were children; approximately 2000 HIV-1infected infants were born per day [22]. As grim as these statistics are, the World Health Organization (WHO) indicates that they still underestimate the enormity of the pandemic. In many parts of the world, most new infections occurred in young adults, with young women of childbearing age being especially vulnerable. The WHO reports a striking diversity of HIVs spread. For example, Eastern Europe and Central Asia are experiencing a fast-growing epidemic, with the number of new HIV infections rising steeply because of IDUs and sexually transmitted infection in young people. An overall adult HIV prevalence of 10% is reported in some sub-Saharan African countries. In contrast, 119 countries of the world have an adult prevalence of less than 1% [22]. Heterosexual HIV-1 transmission predominates in some regions (subSaharan Africa, Asia, and the Caribbean) where the male-to-female ratio of infection is approximately 1:1. In Southeast Asia, India, and South Africa, the number of persons infected continues to escalate. The increase in the number of HIV-1infected women of childbearing age is paralleled by the potential increase in infants born HIV-1 infected [4,22].
986
Table 3 Pediatric HIV classication: clinical categoriesa Category N: not symptomatic Category A: mild symptomatic Children who have no signs or symptoms considered to be the result of HIV infection or who only have one of the conditions listed in category A Children with 2 of the following conditions but none of the conditions listed in categories B and C Lymphadenopathy (0.5 cm at more than two sites; bilateral one site) Hepatomegaly Splenomegaly Dermatitis Parotitis Recurrent or persistent upper respiratory infection, sinusitis, or otitis media Children who have symptomatic conditions other than those listed for category A or C that are attributed to HIV infection. Conditions in clinical category B include but are not limited to Anemia (<8 g/dL), neutropenia (<1000/mm), or thrombocytopenia (<100,000/mm) persisting for 30 days Bacterial meningitis, pneumonia, or sepsis (single episode) Candidiasis, oropharyngeal (ie, thrush) persisting for >2 months Cardiomyopathy Cytomegalovirus infection with onset before the age of 1 month Diarrhea, recurrent or chronic Hepatitis HSV stomatitis, recurrent (ie, >2 episodes within 1 year) HSV bronchitis, pneumonitis, or esophagitis with onset before the age of 1 month Herpes zoster (ie, shingles) involving at least two distinct episodes or more than one dermatome Leiomyosarcoma LIP or pulmonary lymphoid hoperplasia complex Nephropathy Nocardiosis Fever lasting >1 month Toxoplasmosis with onset before the age of 1 month Varicella, disseminated (ie, complicated chicken pox) Children who have any condition listed in the 1987 surveillance case denition for AIDS, with the exception of the LIP (which is a category B condition).
Category B: moderately symptomatic
Category C: severely symptomatic
a Modied from Centers for Disease Control and Prevention. 1994 Revised classication system for human immunodeciency virus infection in children less than 13 years of age. MMWR Weekly 1994;43:110 [122].
AIDS orphans The number of infected and uninfected children born to mothers with HIV-1/AIDS who have already or will become AIDS orphans continues to rise. In the United States alone, there are more than 80,000 uninfected AIDS orphans. Worldwide, an estimated 13 million children less than 15
987
Box 1. Diagnoses that indicate AIDS Serious bacterial infectionsc, multiple or recurrent Candidiasis of the trachea, bronchi, or lungsa Candidiasis of the esophagusa,b Coccidioidomycosis, disseminated or extrapulmonaryc Cryptococcosis, extrapulmonarya Cryptosporidiosis, chronic intestinala CMV (other than liver, spleen, nodes) onset at age >1 montha CMV retinitis (with loss of vision)a,b HIV encephalopathyc Chronic herpes simplex virus ulcer (>1 month duration) or pneumonitis or esophagitis onset at >1 month of agea Histoplasmosis, disseminated or extrapulmonarya Isosporiasis, chronic intestinal (>1 month duration)c Kaposi sarcomaa,b Lymphoid interstitial pneumonitisa,b Lymphoma, primary braina Lymphoma (Burkitt or immunoblastic sarcoma)c Mycobacterium avium complex or M. kansasii, disseminated or extrapulmonarya M. tuberculosis or acid-fast infection (species not identied), disseminated or extrapulmonaryc or pulmonaryd Pneumonia, recurrentd Pneumocystis carinii pneumoniaa,b Progressive multifocal leukoencephalopathya Toxoplasmosis of brain, onset at age >1 montha,b Wasting syndrome caused by HIVc
a If indicator disease is diagnosed denitively (eg, biopsy, culture) and there is no other cause of immunodeciency, laboratory documentation of HIV infection is not required. b Presumptive diagnosis of indicator disease is accepted if there is laboratory evidence of HIV infection (1987 addition). c Requires laboratory evidence of HIV infection (1987 addition). d Requires laboratory evidence of HIV infection (1993 addition; adults and adolescents). Modied from 1994 Revised classication system for human immunodeciency virus infection in children less than 13 years of age. MMWR; Recommendations and reports 1994/43(RR-12).
years of age have lost their mother or both parents to AIDS. Most of these children live in a sub-Saharan African country. The total number of children orphaned by the epidemic is forecasted to more than double by the year 2010 [4,22].
988
HIV-1 transmission Mother-to-infant transmission Mother-to-infant infection (in utero, intrapartum, or postpartum via breast-feeding) accounts for 90% or more of childhood HIV-1 infection, with most perinatal transmission occurring close to the time of or during birth. A 25% rate (range: 16%45%) of mother-to-infant transmission is estimated from studies conducted in the United States, Europe, Asia, and Africa. Increased transmission risks include low maternal CD4 counts, high viral loads, advanced HIV-1 disease [AIDS], low vitamin A, placental membrane inammation, premature rupture of membranes, increased infant exposure to maternal blood, premature delivery, and breast-feeding [16,23,24]. In 1994, zidovudine (ZDV) chemoprophylaxis was documented to reduce perinatal HIV-1 transmission by almost 70% [25]. Antiretroviral therapy was then recommended for HIV-1infected pregnant women [20]. In developed countries, this resulted in a dramatic decline in the rate of mother-to-infant transmission. The estimated number of pediatric AIDS cases diagnosed each year among children in the United States declined consistently, from a peak of 949 in 1992 to 105 in 2000 [26]. Dierent antiretroviral agents and combinations continue to be studied in clinical trials. Worldwide, unfortunately, this has not been the case, and it is worldwide where the epidemic predominates. Heretofore, access to these medications has been limited [4,16,22]. Trials of shorter courses of orally administered antiretroviral agents, now in progress, in developing countries are promising, and continued clinical trials are ongoing. HIV-1 transmission infection through breast-feeding continues to be a signicant concern, because the rates of infection from breast-feeding (estimated at $15%) may oset the benets of treatment during pregnancy and the perinatal period [16,21]. This is a major problem in some developing countries, where safe alternatives to breast-feeding (access to formula milk powder, clean water, and sanitation) are lacking. Blood/blood product transfusion or contaminated injection equipment Iatrogenic viral transmission through blood/blood product transfusion or reuse of contaminated injection equipment occurred in the early years of the epidemic before HIV-1 blood donor screening and is no longer a major problem [21]. Viral transmission in IDUs who are adolescents, however, remains a major concern in this country and worldwide [22,27]. Sexual transmission Of the 5 million new HIV-1 infections in 2001, 50% were in the 15- to 24year-old age group. HIV-1/AIDS prevalence among this age group is 11.8 million (7.3 million girls, 4.5 million boys). Injection drug use and sexual
989
exposure were the most common risk factors [22]. Behavioral and educational interventions continue to be explored and instituted [28]. A recently acknowledged and major issue is sexual abuse of children and adolescents (in the home and community and commercial sexual exploitation) [21]. Most sex industry children are 13- to 18-year-old girls. It is unknown how many child sex workers there are in the world because of the clandestine nature of this multibillion dollar world industry. In India, for example, approximately 20% of the 2 million sex workers are less than 15 years of age and 50% are less than 18 years of age. In Cambodia, 30% of sex workers ranging from 13 to 19 years of age are HIV-1 infected [22]. Pediatric HIV-1/AIDS Natural history Vertically acquired HIV-1 infection occurs at a time when the immune and nervous systems are immature. The viral load of untreated infants is high and persistent, estimated at a median of 700,000 copies, and can stay this high for approximately 2 years [29,30]. In contrast, the initial viral load in untreated initially infected adults is approximately 600,000 copies; by 8 to 12 weeks after the initial infection, this decreases signicantly as a result of the host immune response [31]. The adverse eects of the virus on the developing nervous and immune systems often result in more rapid onset of clinical symptoms and progression to death. Before the widespread use of antiretroviral agents, a bimodal evolution of symptomatic HIV-1 disease and rate of survival was observed [14,33,34]. Approximately 20% of infants developed clinical and immunologic manifestations of AIDS within the rst years of life (rapid progressors). These infants often had high levels of detectable virus at birth and died of AIDS-related complications between 1 and 4 years of age. Not infrequently HIV/AIDS encephalopathy was the presenting AIDS indicator condition, estimated in up to 67% of infants with AIDS [3436,134]. The second group of children had a less rapid form of the disease (slow progressors). Approximately 75% of these children survived to the age of 5 years, with 40% to 50% progressing to AIDS or death by the age of 6 years and a mean survival time of 9 to 10 years of age [36]. Higher mortality rates of HIV-1infected children were reported from developing geographic areas such as Africa (45%50% of infected children died by the age of 2 years, and 62% to 66% died by the age of 3 years) [20,23,37]. In reality, a trimodal evolution exists, with the third hump (albeit the smallest) occurring in the preadolescent and adolescent years [33,3841]. As in adults, latency to onset of moderately symptomatic disease in this group may be 10 years or more. What factors underlie the dierence between these groups remains unclear but may relate to timing of HIV-1 infection (gestation versus parturition), stage of maternal HIV-1 disease during pregnancy and delivery (maternal viremia), neonate viral load, virulence of the infecting virus, and as yet undened host-related factors [29].
990
Pediatric HIV-1 infection in the highly active antiretroviral therapy era HIV-1 infection can now be diagnosed by the polymerase chain reaction (PCR) in most infants by the age of 1 month and in almost all infants by the age of 6 months [42,43]. Early identication enables early institution of prophylactic therapy for OIs (namely, Pneumocystis carinii pneumonia [PCP]) and early antiretroviral treatment of primary infection [42,44]. Clinical studies charting the course of children born after 1995 are in progress. Results to date show AIDS-free survival to be signicantly longer for these children than for those born before 1995. For those children born before 1994 who survived to that time, HAART therapy has been shown to stabilize disease and delay progression to AIDS [15,1820,30].
Neurologic disorders Neurologic disorders associated with HIV-1 infection can occur during every disease stage. Certain complications and syndromes characteristically occur with greater frequency during certain disease stages. HIV-1/AIDS neurologic disorders can be classied as those caused by (1) primary HIV-1 infection (eg, HIV-1/AIDS encephalopathy), (2) secondarily by complications of immunosuppression (infections, neoplasms, and strokes), (3) complications of systemic HIV-1 disease (metabolic and endocrinologic derangements associated with systemic HIV-1related disorders and its therapies), and (4) toxic/metabolic complications of antiretroviral and antimicrobial therapies [9,45,46]. These conditions are not mutually exclusive; coexisting pathologic conditions can exist. The diagnosis of an HIV-1 neurologic disorder requires the skill of a neurologist, a careful medical and developmental history, an HIV-1 systemic disease history, assessment of current immunologic and virologic status, a medication history, neurologic examination, neuroimaging studies, and neuropsychologic assessment [47]. Neuropsychologic evaluation determines the childs cognitive abilities, characterizes domains of strengths and weaknesses, and tracks progressive dysfunction (should this occur) as well as response to therapy. In the era of HAART therapy, HIV-1/AIDS encephalopathy is an infrequent and often treatable complication that seems to respond well to eective antiretroviral control (see section on changing neuroepidemiology). Central nervous system (CNS) complications of immunodeciency and AIDS-related conditions are also less frequent because therapy has eectively delayed advanced HIV-1 disease, profound immunodeciency, and AIDS-related systemic conditions [44]. It is not known whether HIV-1/AIDSassociated CNS disease manifestations dier, should they develop, in children exposed perinatally to antire troviral agents and subsequently treated with HAART regimens. It is possible that early signs may be subtle; may not at rst be apparent on
991
motor examination; and may present as changes in behavior, mood, or cognition. HIV-1/AIDSassociated central nervous system disease Progressive encephalopathy HIV-1/AIDSassociated PE childhood, like its adult counterpart (AIDS dementia complex [ADC]) [48,85], is a syndrome complex with cognitive, motor, and behavioral features linked to primary HIV-1 CNS infection [48]. Cognitive impairment, acquired microcephaly, and progressive corticospinal tract (CST) signs are the hallmark manifestations [5,6,8,9,11,48]. Rigidity (superimposed on spasticity) and cerebellar signs may also develop [8,10]. Mood and behavioral problems are common [8,10,49]. Associated neuroimaging features are listed in Table 4. PE usually develops in children with advanced HIV-1 disease and is associated with a poor long-term outcome [35,36,4953]. The neurologic course is variable from individual to individual [5,9,10]. Progression in some infants is rapid, resulting in quadriparesis and severe cognitive impairment within months of onset, whereas other children have a stepwise course with periods of decline followed by relatively stable periods, which may or may not be followed by further deterioration. Cognitive and motor impairment can also be discordant. Some young children develop disabling motor decits yet maintain relatively stable (albeit impaired) cognitive function, whereas other children have more impaired cognitive than motor function [510,13,45,47,49,50].
Table 4 Neuroimaging features and ndings Magnetic resonance imaging computerized tomography (MRI CT) Cerebral atrophy Serial studies often show progressive atrophy Basal ganglia calcication Serial CT studies may show progressive calcication frontal white matter calcication White matter Serial studies often show progressive patchy or diuse periventricular white matter changes CT: hypodensity-rarefaction MRI: abnormal high signal on T2-weighted images A correlation appears to exist between progressive changes and severity of encephalopathy, cognitive dysfunction and behavioral changes [5,811,32,45,50,123126]. MR spectroscopy (MRS) Basal Ganglia/White Matter (Centrum Semiovale) Decreased N-acetyl aspartate, (NAA) Increased choline (cho) and myoinositol MRS oers a potentially valuable in vivo approach to study dynamic metabolic changes at dierent stages of HIV-1 associated CNS disease and may be useful in evaluation of disease progression and response to therapy [127131,137,138].
992
Signs of CNS disease manifest according to age of onset [47]. Motor involvement is usually more prominent in younger children than in school-aged children. Newborns are usually well at birth with no recognizable neurologic features of HIV-1associated CNS disease, although a slightly smaller than anticipated head circumference may be seen [55]. It is during infancy, however, that the most easily recognized and devastating PE syndrome occurs. Characteristic features are (1) progressive CST signs with loss of previously acquired motor milestones or a markedly deviant rate of attaining motor skills, (2) decline in mental development or marked delays in cognitive development, and (3) acquired microcephaly. A less severe CNS syndrome occurring within the rst 2 years of life manifests as hypotonia with delays in attainment of motor milestones associated frequently with delays in acquisition of language [10,13,54]. A change of gait is often the rst sign of HIV-1 CNS disease in toddlers and young children. They begin to toe walk, are hyperreexic, and develop spasticity. The rate of progression is variable. Some children are wheelchair bound within months, others require orthotics, and others maintain independent ambulation, although they have a spastic gait. Impairment of ne motor skills and dexterity is common as are cognitive decits ranging from mild mental retardation to low average or borderline intelligence [5,810,45,47,50]. Behavioral changes, loss of interest in school performance, decreased attention, declining cognitive abilities, psychomotor slowing, decreased verbal output, emotional lability, social withdrawal, and, more rarely, psychotic manifestations are reported in school-aged children. Hyperreexia, clumsiness, and poor ne motor ability and dexterity usually ensue. Progressive CST signs, movement disorders, cerebellar signs, and myelopathy may arise as disease advances, as does cognitive decline. At end stage, the child is apathetic and abulic [810,45,47,49]. Neurodevelopmental status The frequency of developmental problems is high, as noted in the initial descriptions of children with HIV/AIDS and subsequently corroborated by other studies [10,13,47,50,51,5458]. Almost half of the antiretroviral-naive symptomatic children entered in one study had deviant development (involving motor function [23%], reexes [21%], and behavior [13%]). Almost 50% of the children in the youngest age group had motor dysfunction, and 21% of the cohort had cognitive deciencies, again, with the highest percentages in the youngest age groups (312 months [33.7%], >1230 months [19%], >30 months to 6 years [21%]) [59]. At follow-up, those children with the lowest IQ scores and those with motor dysfunction had the highest risk for disease progression [52]. It also must be kept in mind that neurodevelopmental status can be aected by non-HIV-1related comorbid conditions, which, unfortunately, occur with signicant frequency in this population (eg, those related to
993
maternal high-risk factors during pregnancy, premature birth, complications in the neonatal period, and innumerable postnatal psychosocial stressors) [10,45,50,51,5458]. HIV-1associated minor motor and cognitive impairment Minor motor impairment, with or without cognitive decits, remains a frequent nding in school-aged children, many of whom had delays in motor and language development as infants [13,47,49]. The term denotes minor degrees of cognitive and motor impairment that are not sucient for the diagnosis of PE. Manifestations include mild clumsiness and learning disabilities. Although composite IQ scores range from average, low average, and borderline to mild mental retardation, selective decits in visual-spatial and organizational skills, cognitive exibility, and expressive language problems are reported [49,58,6067]. This suggests that HIV-1 may compromise many aspects of cognitive and motor development and results in a mild but stable encephalopathy [9,10]. Neuropsychiatric problems Psychosis. Acute psychosis, confusion, agitation, delirium, mania, and catatonia can occur in advanced HIV-1 disease. Changes in mental status may be caused by HIV-1 CNS disease, complications of CNS OIs (eg, cytomegalovirus [CMV] encephalitis), nutritional deciencies, or toxic eects of antiretroviral and other therapies [49]. Mood, aect, and depression. Emotional lability, mood swings, agitation, new onset of extreme impulsiveness, and attention problems can occur [8,9,49]. Other children may show attened aect, lack of social responsiveness, withdrawal, and declining interest in the environment. HIV-1associated CNS disease can cause or contribute to an organic depression, as can coexisting HIV-1/AIDS conditions (eg, endocrinologic and metabolic disturbances). Chronic disease, frequent hospitalizations, illness and death of family members, changing caretakers, psychosocial stressors, and stigmata attached to HIV disease itself may also lead to major depression and other behavioral or aective changes. Changing neuroepidemiology The incidence and prevalence of pediatric HIV-1/AIDS PE has signicantly decreased in the past several years. Whereas a 16% to 30% incidence of pediatric HIV-1/AIDS PE had been reported from various cohort studies conducted in the rst decades of the epidemic [47], a recent report estimated a yearly incidence (January 2000January 2001) of 4% and an overall prevalence of 10% [68].
994
Neuroepidemiologic changes can be attributed to (1) an overall decline in the numbers of infants born HIV-1 infected, and thus in the actual numbers of infants with potentially rapidly progressive disease and PE; (2) early identication and treatment of those infants born HIV-1 infected, resulting in delay of advanced disease; and (3) HAART therapy for those children who have survived, with stabilization of disease, delay of disease progression, and reduced risk of developing PE [16,1820,24,30,43,53,65]. Neuropathology of HIV-1associated central nervous system disease HIV-1 The etiologic agent was isolated in 1983 [69,70], demonstrated in the CNS shortly thereafter [7276], and named human immunodeciency virus in 1986 [77]. HIV-1 belongs to the Lentivirinae subfamily of nononcogenic cytopathic retroviruses [77,78]. A subfamily well recognized to cause neurologic disease, lentiviruses are species specic, have long periods of clinical latency and mechanisms to evade immune clearance, cause persistent infection and multisystem disease, and characteristically invade the CNS soon after primary infection [78]. Retroviruses are enveloped RNA viruses that use reverse transcriptase (RT) to transcribe virion RNA into linear double-stranded DNA with subsequent integration into the host genome. They are approximately 100 nm in diameter and have two single strands of RNA, which permits recombination between the strands (and potential for great genetic diversity). The genome is 10 kilobases in size and contains three major structural genes, gag, pol, and env, as well as extra genes that are essential to viral replication. The retroviral life cycle involves two forms, a DNA provirus and an RNA-containing infectious virion. Retroviruses have the advantage of latency because they have a DNA intermediate in their replication cycle and because the DNA provirus is integrated into the chromosomal DNA. Additionally, HIV-1 is a CD4 T-cell and macrophage-tropic virus and has the advantage of reducing the eectiveness of host immune attacks [46]. Neuropathologic ndings Neuropathologic features of HIV-1/AIDS encephalopathy are listed in the Box 2. HIV-1 is localized in macrophages, microglia, and multinucleated giant cells found predominantly in the basal ganglia, subthalamic nucleus, substantia nigra, dentate nucleus, and white matter (HIV-1 is highly neurotropic but not neuronotropic) [46,75,7982,136,139]. These cells express both CD4 and b-chemokine receptors (eg, CCR5 and CXCR4), which permit HIV-1 entry and mediate fusion with macrophage tropic strains. It is these cells that support productive retroviral infection. HIV-1 has also been demonstrated in endothelial cells and glial cells (although glial cell infection is thought to be a nonproductive infection) [7]. In adults, high levels of HIV-1 mRNA and expression of viral proteins in microglia and macrophages seem to correlate with the severity of clinical disease [46,67,83,84].
995
Box 2. Neuropathological features and ndings Gross brain studies Cerebral atrophy Ventricular enlargement; widening of sulci; attenuation of deep cerebral white matter Microscopic studies HIV-1 encephalitis Foci of inammatory cells, microglia, macrophages, multinucleated giant cells HIV-1 leukoencephalopathy Diffuse staining pallor of myelin; diffuse damage to white matter; myelin loss, reactive astrogliosis; multinucleated giant cells, macrophages HIV-1 poliodystrophy Calcic vasculopathy Mineralization detected in walls of vessels, basal ganglia, and white matter white matter changes and gliosis are common may have associated HIV encephalitis most characteristic neuropathologic nding in infants and children Spinal cord studies Corticospinal tract pathologic ndings: striking myelin pallor restricted to the corticospinal tracts Axonopathy type Myelinopathy type Myelitis Vacuolar myelopathy: older children
Data from [5,8,9,12,79,9497,102,132,133,135].
Neuropathophysiologic mechanisms The mechanisms underlying HIV-1associated CNS disease are not fully understood and are the focus of ongoing research. Processes possibly contributing to the pathophysiology include toxicity of HIV proteins; macrophage factors; locally synthesized cytokines; and viral- and host-related factors [83]. HIV-1infected macrophages/monocytes in blood are thought to be recruited to the brain by upregulation of chemoattractant chemokines and adhesion molecules on endothelial cells, enabling transendothelial migration of these cells into the CNS compartment. Activated macrophages release multiple proinammatory cytokines as well as other factors that may impair
996
Box 3. Possible factors in the pathogenesis of HIV-1associated central nervous system disease Viral proteins Blocking of transmitters Interference with neurotrophic factors and cell function Toxicity of viral polypeptide Cytokines Direct toxicity Neuroimmune-mediated central nervous system dysfunction and injury Stimulation of astrocytes Excitotoxin Toxicity of other soluble factors Autoimmunity Molecular mimicry Immune response to HIV-1 antigens T-cell deregulation Increased HIV-1 viral load in central nervous system Transactivation Cytokine induced Emergence of neurovirulent strains Host factors
Data from Johnson RT, McArthur JC, Narayan O. The neurobiology of human immunodeciency virus infections. FASEB J 1988;2:29702981, [78].
neuronal and glial function, interfere with the release of neurotransmitters, and alter synaptic transmission [46]. The current consensus is of a cytokinemediated cascade leading to neuronal dysfunction [21] and ultimately to a neurodegenerative pathway [83]. Excitotoxicity (chronic hyperactivation of excitatory amino acid pathways) may also contribute to neuronal injury via N-methyl-D-aspartate receptor activation and calcium inux [84]. Neuropathologic ndings and the developing brain In adults, HIV-1 infection occurs in a fully developed organism. In contrast, vertically transmitted HIV-1 infection occurs in an immature developing organism. HIV-1associated dementia (formally referred to as ADCAIDS dementia complex) arises in a mature and completely myelinated nervous system and in a fully developed CNS immune system [48]. Although HIV-1 invades the CNS soon after primary infection in both adults [71,73] and children [8689], signs of CNS disease in adults, adolescents, and older children usually do not occur until the disease is in an advanced stage, when the patient is severely immunosuppressed. In contrast, in some infants
997
and young children, AIDS encephalopathy is frequently the AIDS-dening illness, even at a time when the infant is not severely immunosuppressed [35,36]. The late fetal and postnatal periods of brain development are a dynamic time characterized by elaboration of dendritic and axonal ramications, establishment of synaptic contacts, selective elimination of neuronal processes and synapses, programmed cell death, and proliferation of glia and myelination [90]. Astrocytes perform a myriad of complex and vital functions, including neuroendocrine, nutritive, and supportive functions in relation to neuronal homeostasis as well as in the reaction to metabolic and structural insults [90]. Many of these developmental events are controlled, in part, by amino acid transmitters (as well as other trophic factors), which are reected in receptor ontogeny. In the early postnatal period, there is overexpression of receptors in some regions (eg, glutamate receptors in the basal ganglia and thalamus). In the immature brain, it is these regions that seem to be especially vulnerable to various perturbations and insults [90,91]. Perturbations in these and other developmental processes may lead to disturbances in proliferation, myelination, neuronal function, and even cell death [90,91]. The maturational stage of the nervous and immune systems at the time of viral exposure may explain dierences in the eects of HIV-1 in infants versus adults, may account for age-specic vulnerabilities, and may contribute to age-of-onset dierences in clinical manifestations and progression [10,92,93]. Impaired brain growth, manifested as acquired microcephaly, is a characteristic nding. A signicant reduction in brain weight and volume of cortical regions, subcortical gray matter, and cerebral white matter is seen at autopsy [9497]. There are no gross or microscopic malformations, evidence of impaired neurogenesis, or disordered neuronal migration (although as in adult neuropathologic studies [98], neuronal loss and damage to the dendritic arbor are observed in some cases) [95]. HIV-1 infection of the immature brain may aect brain cell proliferation and account, in part, for acquired microcephaly [53,94]. It is also tempting to speculate that interference with synaptogenesis could partly explain developmental delays and cognitive impairment. Factors like CNS viral load, viral strain dierences, and evolution of neurotropic and neurovirulent strains as well as mechanisms underlying viral protein and cytokine-mediated eects and toxicities and as yet undened viralhost interactions may aect the disease course and outcome, however [47]. These issues are the focus of continued investigations. Secondary central nervous system complications As the name of the disease implies, AIDS ultimately culminates in the development of profound immunosuppression [14]. The child becomes susceptible to infections, neoplasms, and cerebrovascular complications. Manifestations include the new onset of mental status changes, headache, seizures, or focal neurologic decits [10].
998
CNS infections CNS infections caused by common pathogens and OIs develop the latter especially in advanced disease when the child is immunosuppressed (Box 3). Before HAART therapy, CMV encephalitis and Candida albicans meningitis and microabscesses were the most commonly identied organisms. Congenital CNS infections (eg, toxoplasmosis, CMV) were described infrequently, as was measles encephalitis. Reactivation of previously acquired infection (eg, toxoplasmosis, herpes simplex virus [HSV], varicella zoster virus [VZV]), as in adults, did occur in children with advanced disease, as did progressive multifocal leukoencephalopathy (PML), an almost unheard of complication of immunosuppressed children before the HIV/AIDS epidemic [9,94,99109]. Neoplasms Primary CNS lymphoma (PCNSL) and systemic lymphoma metastatic to the CNS were the most frequently reported neoplasms and occurred in immunodecient children with advanced disease [9,94,110112]. PCNSL, an aggressive B-cell tumor considered to be part of the spectrum of B-cell proliferative disorders associated with pediatric HIV-1/AIDS, is related to Epstein-Barr virus. PCNSL, the most frequent cause of mass lesions in pediatric AIDS, can occur as a single mass lesion or as multicentric lesions [9,110,111]. CT characteristics include hyperdense or isodense mass lesions with variable contrast enhancement; diusely inltrating contrast-enhancing lesions; or periventricular contrasting lesions. The basal ganglia, corpus callosum, and periventricular white matter are the most common locations. Denitive diagnosis is made by biopsy, although magnetic resonance spectroscopy (MRS) studies may be able to dierentiate lymphoma from other lesions. Reduction in tumor size is achievable with radiation therapy [112]; however, long-term survival rates are grim because of tumor recurrence as well as advancing HIV-1 disease. Patients most often succumb to other AIDS-related illnesses, OIs, or both. Cerebrovascular complications Cerebrovascular complications include intracerebral hemorrhage and infarction. Clinical presentation is variable, reecting the severity and location of the hemorrhage. Strokes may be catastrophic and fatal or clinically silent [9,113]. Intracerebral hemorrhage occurs in the setting of immune-mediated thrombocytopenia, as subarachnoid hemorrhage associated with aneurysmal arteriopathy of major cerebral arteries, or as hemorrhage into tumor [113]. Infarctions are associated with pathologic changes of cerebral blood vessels, meningeal infections, or cardiomyopathy or coagulopathies [94,113]. Cerebral vascular ectasia and aneurysmal arteriopathy of major cerebral arteries and thrombosis of these arteries or of small cortical vessels are
999
reported [9,81,94,113115]. Fusiform aneurysmal dilation of the branch vessels of the circle of Willis (diuse and fusiform or focal and saccular) is an increasingly recognized complication of pediatric AIDS (and has occurred in children who were not severely immunosuppressed) [114,115]. The mechanism by which HIV-1 causes CNS arterial damage is unclear. Gp41-positive cells in arterial walls and positive macrophage markers of leukocyte common antigen (but not endothelial markers) have been demonstrated with immunohistochemical staining, suggesting that HIV-1 may be directly involved in vascular pathologic changes (possible direct HIV endothelial invasion and exposure to toxic cytokines) [81]. An underlying VZV vasculitis is also hypothesized, however. A brosing inammatory vasculopathy is described as well and thought to be related to primary HIV-1 CNS infection [94,113]. Pathologic examination shows sclerotic vessels with obliterated or markedly stenosed lumens. A subacute necrotizing encephalopathy with cystic encephalomalacia seems to be associated with dilated cardiomyopathy and may be related to an acquired mitochondrial cytopathy or hypoxicischemic damage [94]. Multiple ischemic infarcts may also result from leptomeningitis associated with infectious etiologies such as VZV, Mycobacterium tuberculosis, and Treponema pallidum) [47,101,116]. Peripheral nervous system involvement Distal sensory or sensorimotor axonal neuropathy (DSP) described by children as pain and burning in the feet and legs, demyelinating neuropathy, lumbosacral polyneuropathy, mononeuritis multiplex, inammatory polyneuropathy, median nerve compression at the carpal tunnel, and neuropathy related to antiretroviral therapy are described, although far less frequently than in adults. Similar to the case in adults, however, DSP and CMV polyneuropathy occur in the setting of advanced HIV-1 disease (and HIV-1 CNS disease) and peripheral neuropathy occurs as a toxicity of antiretroviral therapy [42,117,118].
Management and therapy Management of the HIV-infected child continues to evolve and is increasingly complex [14,42,44]. Greater understanding of the biology of HIV-1, factors related to viral entry, viralhost interaction, rates of viral production, associated uctuations and progressive T-cell depletion, emergence of viral variants, viral distribution in dierent cellular compartments, and issues of latency have led to changes in therapeutic strategies, and these studies are ongoing [29]. Therapeutic strategies have expanded greatly from treatment with a single drug to combination therapy that includes up to three dierent classes of antiretroviral agents [29,30,42,44]. Early initiation of this combination therapy to target specic components unique to the
1000
virus is currently recommended to suppress HIV-1 replication without disruption of normal cellular function, preserve immune function, and reduce the development of resistant organisms [30]. Whenever possible, a team approach supervised by an expert in this eld is recommended [44]. The mainstays of HIV-1 management include early identication of the HIV-1infected infant, antiretroviral therapy targeted to the retrovirus, prophylaxis and treatment of OIs, and management of HIV/ AIDSrelated conditions. Guidelines for management and therapy, with updates, are provided by the US Department of Health and Human Services Panel of Clinical Practices for Treatment of HIV Infection (USDPH) [42]. In parallel are habilitation/rehabilitation programs as well as attention to pain management, psychosocial stressors, and quality of life. Antiretroviral agents As of February 2001, there were 11 antiretroviral drugs approved for use in HIV-1infected children and 15 approved for use in adolescents. These drugs fall into three major classes, nucleoside analogue reverse transcriptase inhibitors (NRTIs), nonnucleoside analogue reverse transcriptase inhibitors (NNRTIs), and protease inhibitors (Table 5). NRTIs and NNRTIs act at the early stage of viral replication, and inhibitors of viral protease work in the later stage after viral integration [30,42]. Some potential neurologic complications are listed in Box 4. Detailed pediatric antiretroviral drug information, mechanisms of action, and toxicities are detailed and updated by the USDPH [42]. Nucleoside analogue drugs can induce mitochondrial dysfunction because of their anity for mitochondrial gamma DNS polymerase and interfere with mitochondrial replication. This can result in mitochondrial DNA depletion and dysfunction. In vitro studies show the relative potency is highest for didoxycytidine (ddC), followed by didanosine (ddI), stavudine (d4T), lamivudine (3TC), ZVD, and Abacavir (ABC). Toxicity related to mitochondrial dysfunction is reported in patients receiving long-term treatment with these agents. In general, resolution occurs with discontinuation of the drug(s). A possible genetic susceptibility to these toxicities is suggested [42]. New classes of antiretroviral drugs are being studied (eg, fusion inhibitors that inhibit viral binding or fusion to host target cells). In addition to antiretroviral drugs, cytokine blockers, calcium channel blockers, antioxidants, and immunomodulating agents are being evaluated [42]. Vaccine development is ongoing [119]. The time to initiate therapy for asymptomatic children, the time to consider a drug holiday for children with undetectable viral titers, and when to change therapy and to which regimen for symptomatic children who have evidence of virologic, immunologic, or clinical disease progression or toxicity or intolerance or adherence to their current drug regimen continue to be investigated [42].
1001
Management for HIV-1/AIDSassociated neurologic disorders HIV-1associated progressive encephalopathy HIV-1/AIDS PE should be treated with at least one antiretroviral agent with substantial CNS penetration (CSF-to-plasma ratios >0.2 [eg, ZVD or Neverapine (NVP)]). HAART therapy has been shown to stabilize disease and to reverse HIV-1associated CNS signs in some children when treated in the early stage [42,44]. Neurobehavioral complications Psychosis Acute psychotic episodes respond to neuroleptic medications. Children with HIV-1 CNS disease may require lower than normal doses. Psychotherapy should be a component of treatment [44]. Depression Antidepressant medications in conjunction with psychotherapy can be benecial. Interactions between protease inhibitors and psychotropic medications may occur, and dosages may need to be adjusted [44]. Short-term individual psychotherapy helps the child to cope with issues like death in the family. A childrens group (longer term therapy) helps to enable children to deal with the issues of chronic illness, death, and social stigmata associated with HIV-1 infection. Attention-decit disorder/attention-decit hyperactivity disorder Methylphenidate is of benet for some HIV-1infected children with attention-decit disorder/attention-decit hyperactivity disorder (ADD/ ADHD). Appetite suppression with stimulants can be an undesirable side eect in HIV-1infected children and needs to be monitored. Behavioral modication techniques are also strongly recommended and should be instituted before and in conjunction with a medication trial when possible [44]. Pain syndromes Children with HIV-1/AIDS (like children with other chronic diseases) encounter painful medical procedures; many have pain related to their underlying condition. A combination of pharmacologic and nonpharmacologic therapies is recommended, as is an aggressive approach to pain management, including management of painful procedures [120]. Analgesic therapy should be initiated after a thorough clinical assessment, which includes assessment of pain severity as well as identication and treatment of the underlying cause of pain. The WHO analgesic ladder provides a useful guideline for treatment strategies [121].
1002
Table 5 Antiretroviral agents and potential neurological toxicitiesa Nucleoside Analogue Reverse Transcriptase Inhibitors (NRTI) Zidovudine (ZVD) Headache; less commonly myopathy, myositis (Retrovir; AZT; ZVD) ZVD concentration may increase with valproate Phenytoin concentrations may increase or decrease Didanosine (Videx; ddI) Peripheral neuropathy (dose related) Stavudine (Zerit; d4T) Headache; less commonly peripheral neuropathy Lamivudine (Epivir; 3TC) Headache; less commonly peripheral neuropathy Zalcitabine (Hivid; ddC) Headache; peripheral neuropathy Not recommended with concomitant use of DD1 because of increased risk of neuropathy Abacavir (Ziagen ABC) Headache Nonnucleoside reverse transcriptase inhibitors (NNRTI) Neverapine Headache; less commonly myalgias or arthralgias (Viramune; NVP) Serum concentrations of antiepileptic drugs and psychotropics should be monitored Delavirdine Headache (Rescriptor; DLV) Decreases metabolism of certain drugs, resulting in increased drug levels Not recommended for concurrent use with antihistamines, sedative-hyponotics, calcium channel blockers, amphetamines Efavirenz (Sustiva; EFV) Mental status changes: Scomnolence, insomnia, confusion, abnormal dreams, abnormal thoughts, impaired concentration, amnesia, agitation, depersonalization, hallucinations, euphoria (reported more frequently in adults) EFV highly plasma bound; Potential for drug interactions with other highly protein bound drugs (phenobarbital, phenytoin) Protease inhibitors (PI) Saquinavir (Fortovase; Zavirase; Invirase; SQV) Headache, paresthesias Not recommended for concurrent use with antihistamine, sedative-hypnotics May decrease metabolism of drugs resulting in increased drug levels Headache; dizziness Not recommended for concurrent use with antihistamine, sedative-hypnotics May decrease metabolism of drugs resulting in increased drug levels Headache; less commonly circumoral paresthesias Not recommended for concurrent use with analgesics (eg, meperidine, piroxicam, proposcyphene, antihistamines, sedative-hypnotics, psychotropic drugs; bupropion hydrochloride, clozapine pimozide) Antiepileptic drugs (carbamazepine, phenytoin, phenobarbital) should be monitored; antiepileptic drugs may lower levels of RTV
Indinavir (Crixivan; IDV)
Ritonavir (Norvir; RTV)
A.L. Belman / Neurol Clin N Am 20 (2002) 9831011 Table 5 (continued ) Kaletra (Lopinavir)
1003
Nelnavir (Viracept; NFV)
Amprenavir (Agenerase; APV, VIX478)
Headache Not recommended for concurrent use with antihistamine, sedative-hypnotics; neuroleptics, St. Johns wort Antiepileptic drugs (carbamazepine, phenytoin, phenobarbital) may lower levels of lopinavir antiepileptic drugs levels should be monitored Not recommended for concurrent use with antihistamine, sedative-hypnotics May decrease metabolism of drugs, resulting in increased drug levels Paresthesias Inhibitor of cytochrome P450 isoenzyme CYP3A4; drug reaction possible with tricyclic antidepressants
a Toxicities listed are not all-inclusive. More complete and detailed prescribing and toxicity information is available [42] and from the drug companies. These should be reviewed before prescribing these drugs. Nucleoside analogue drugs can induce mitochondrial dysfunction (see text).
Neurologic and neurodevelopmental decits Habilitation/rehabilitation services are required by many HIV-1infected children. As noted previously, the rate of neurodevelopmental, motor, and cognitive deciencies is high. HIV-1infected children (like non-HIV-1 infected children) benet from habilitation/rehabilitation services, including early intervention, therapeutic nursery schools, physical and occupational programs, and special educational classes and services. When indicated, habilitation/rehabilitation programs should be an integral part of the childs clinical regimen and should be individualized to the needs of the child. The child with a stable diplegia syndrome often benets from gait analysis, orthotics, and, when indicated, Botox therapy and/or orthopedic procedures [47,61]. Summary Since the initial descriptions of AIDS in the late 1970s, much has been learned about the biology of HIV-1 and the cells it infects. Much has also been learned about mother-to-infant viral transmission and the natural history of HIV-1 infection. Key studies led to strategies for interrupting motherto-infant transmission, resulting in a signicant decline in neonatal HIV-1 infection. More procient diagnostic techniques made early diagnosis of HIV-1infected neonates and infants possible during asymptomatic or mildly symptomatic disease stages. Major advances in treatment led to the control of viral replication and thereby altered the course of disease progression. HIV-1/AIDSassociated neurologic disorders declined in parallel. In countries where these therapies are readily available, a dramatic decline in
1004
Box 4. Treatment of HIV-1/AIDSassociated neurologic disorders HIV-1 associated progressive encephalopathy One or more central nervous system penetrating antiretroviral agent having cerebrospinal uid/plasma ratios >0.2 (eg, zidovudine or Neverapine (NVP)) Neurobehavioral complications Psychosis Neuroleptics* Psychotherapy Depression Antidepressants Psychotherapy Individual* Group* Attention-decit hyperactivity disorder/Attention-decit disorder Psychostimulants Methylphenidate* Behavioral modication Pain management Painful medical procedures Pain related to HIV-1 related conditions Combination of pharmacologic and nonpharmacologic therapies Analgesic therapy should be initiated after a thorough clinical assessment Assessment of pain severity Identication and treatment of the underlying cause of pain WHO analgesic ladder provides a useful guideline for treatment strategies Habilitation/rehabilitation Therapeutic nursery schools Special education classes Physical and occupational therapy
* See text. Data from Committee on Pediatric AIDS. (1) Antiretroviral therapy and medical management of pediatric HIV infection. (2) A report on the prevention of opportunistic infections in persons infected with human immunodeciency virus. Pediatric 1998;102:100586 [44].
1005
the number of infants born HIV-1 infected has been realized as has a markedly improved survival rate of those infected. Many questions remain, however. The long-term eects of prenatal exposure to antiretroviral agents are not yet known and continue to be studied. Just exactly how HAART therapy may aect early signs of pediatric HIV-1/ AIDSassociated CNS disease, should they develop, is unclear. As new antiretroviral agents are developed and new combination drug regimens are instituted, the potential for neurologic complications, toxicities, and adverse drug interactions (eg, with antiepileptic drugs (AEDS)) exists and needs to be identied and monitored.
References
[1] Ammann AJ, Cowan M, Wara DW, et al. Acquired immunodeciency in an infant: possible transmission by means of blood products. Lancet 1983;1:9568. [2] Oleske J, Minnefor A, Cooper R, et al. Immune deciency syndrome in children. JAMA 1983;249:23459. [3] Rubinstein A, Sicklick M, Gupta A, et al. Acquired immunodeciency with reversed T4/T8 ratios in infants born to promiscuous and drug addicted mothers. JAMA 1983;249:23506. [4] The Global AIDS Policy Coalition Mann JM, Tarantola DJM, editors. AIDS in the World II. Global dimensions, social roots, and responses. New York: Oxford University Press; 1996. [5] Belman AL, Ultmann MH, Horoupian D, et al. Neurologic complications in infants and children with acquired immune deciency syndrome. Ann Neurol 1985;8:5606. [6] Epstein LG, Sharer LR, Joshi VV, et al. Progressive encephalopathy in children with acquired immune deciency syndrome. Ann Neurol 1985;17:48896. [7] Ultmann MH, Belman AL, Ru HA, et al. Developmental abnormalities in infants and children with acquired immune deciency syndrome (AIDS) and AIDS-related complex. Dev Med Child Neurol 1985;27:56371. [8] Belman AL, Lantos G, Horoupian D, et al. AIDS: calcication of the basal ganglia in infants and children. Neurology 1986;36:11929. [9] Belman AL, Diamond G, Dickson DW, et al. Pediatric AIDS: neurological syndromes. Am J Dis Child 1988;142:2935. [10] Belman AL. AIDS and pediatric neurology. Neurol Clin North Am 1990;8:571603. [11] Epstein LG, Sharer LR, Oleske JM, et al. Neurologic manifestations of human immunodeciency virus infection in children. Pediatrics 1986;78:67887. [12] Epstein LG, Sharer LR, Goudsmit J. Neurological and neuropathological features of HIV in children. Ann Neurol 1988;23(Suppl):S1923. [13] Ultmann MH, Diamond G, Ru HA, Belman AL, et al. Developmental abnormalities in infants and children with acquired immune deciency syndrome (AIDS): a follow up study. Int J Neurosci 1987;32:6617. [14] HIV/AIDS in infants, children, and adolescents. Pediatr Clin North Am2000;32:47. [15] Abrams EJ, Weedon J, Bertolli J, et al. Aging cohort of perinatal human deciency virus infected children in New York City. Pediatr Infect Dis J 2001;20:5117. [16] Bulterys M, Fowler MG. Prevention of HIV infection in children. Pediatr Clin North Am 2000;47:24160. [17] Centers for Disease Control and Prevention. Notice to readers. World AIDS Day December 1, 2001. MMWR Weekly 2001;50:10623. [18] DeMartino M, Tovo PA, Balducci M, et al. Reduction in mortality with availability of antiretroviral therapy for children with perinatal HIV-1 infection. JAMA 2000;284:1907.
1006
[19] Gray L, Newell ML, Peckham C, Levy J. European collaborate study. Fluctuations in symptoms in human deciency virus-infected children. Pediatrics 2001;108:11622. [20] Lindegren ML, Steinberg S, Byers RH. Epidemiology of HIV/AIDS in children. HIV/ AIDS in infants, children, and adolescents. Pediatr Clin North Am 2000;47:120. [21] Joint United Nations Program on HIV/AIDS. World Health Organization (UNAIDS/ WHO): report of the global HIV/AIDS epidemic. June 2000, Geneva, Switzerland. Available at: http://www.unaids.org/epidemic_update/report.htm. Accessed January 2002. [22] Joint United Nations Program on HIV/AIDS. World Health Organization (UNAIDS/ WHO): World AIDS Day 2001. Available at: http://www.unaids.org. Accessed January 2002. [23] Castetbon K, Leroy V, Spira R, Dabis F. The transmission of HIV-I from mother to child in Africa in the year 2000. Sante 2000;10:10313. [24] Fowler MG, Simons RJ, Roogpisuthipong A. Update on perinatal HIV transmission. HIV/AIDS in infants, children, and adolescents. Pediatr Clin North Am 2000;47:2138. [25] Connor EM, Sperling RS, Gelber R, et al. Reduction of maternal-infant transmission of human immunodeciency virus type 1 with zidovudine treatment. N Engl J Med 1994; 331:117380. [26] Centers for Disease Control and Prevention. HIV/AIDS surveillance report. MMWR Weekly 2001;58:13. [27] Futterman D, Chabon B, Homan ND. HIV and AIDS in adolescents. HIV/AIDS in infants, children, and adolescents. Pediatr Clin North Am 2000;47:17188. [28] HIV Prevention Trials Network. Available at: http://www.niaid.nih.gov/factsheets/ hptn.htm. Accessed January 2002. [29] Luzurianga K, Sullivan JL. Viral and immunopathogenesis of vertical HIV infection. HIV/AIDS in infants, children, and adolescents. Pediatr Clin North Am 2000;47: 6578. [30] Palumbo PE. Antiretroviral therapy of HIV infection in children. HIV/AIDS in infants, children, and adolescents. Pediatr Clin North Am 2000;47:15569. [31] Ho DD. Viral counts and HIV infection. Science 1996;272:11245. [32] Blanche S, Tardieu M, Duliege AM, et al. Longitudinal study of 94 symptomatic infants with maternofetal HIV infection: evidence for a bimodal expression of clinical and biological symptoms. Am J Dis Child 1990;144:12105. [33] Ehrnst A, Lindgren S, Dictor M, et al. HIV in pregnant women and their ospring: evidence for late transmission. Lancet 1991;338:2037. [34] Scott GB, Buck BE, Letterman JG, et al. Acquired immunodeciency syndrome in infants. N Engl J Med 1984;310:7681. [35] Cooper ER, Hanson C, Diaz C, Mendez H, Abbound R, Nugent R, et al. Encephalopathy and progression of HIV disease in a cohort of children with perinatally acquired HIV infection. J Pediatr 1998;132:808905. [36] Lobato MN, Caldwell MB, Ng P, Oxtoby MJ. Encephalopathy in children with perinatally acquired immunodeciency syndrome virus infection. Pediatric Spectrum of Disease Clinical Consortium. J Pediatr 1995;126:1705. [37] Taha TE, Graham SM, Kumwenda NI, et al. Morbidity among HIV-I infected and uninfected African children. Pediatrics 2000;106:e77 (Electronic article). [38] Auger I, Thomas P, De Gruttola V, et al. Incubation periods for pediatric AIDS patients. Nature 1988;336:5757. [39] Burger H, Belman AL, Grimson R, et al. Long HIV-1 incubation periods and dynamics of transmission within a family. Lancet 1990;336:1346. [40] Grubman S, Gross E, Learner-Weiss N, et al. Older children and adolescents living with perinatally acquired human immunodeciency virus infection. Pediatrics 1995;95:65763. [41] Persaud D, Chandwani S, Rigaud M, et al. Delayed recognition of human immunodeciency virus infection in pre adolescent children. Pediatrics 1992;90:688691. [42] Guidelines for the use of antiretroviral agents in pediatric HIV infection. Available at: http://www.hivatis.org. Accessed December 14, 2001.
1007
[43] Nielsen K, Bryson YJ. Diagnosis of HIV infection in children. HIV/AIDS in infants, children and adolescents. Pediatr Clin North Am 2000;47:3964. [44] Committee on Pediatric AIDS. 1) Antiretroviral therapy and medical management of pediatric HIV infection. 2) A report on the prevention of opportunistic infections in persons infected with human immunodeciency virus. Pediatrics 1998;102:100586. [45] Belman AL. AIDS and the childs central nervous system. Pediatr Clin North Am 1992;39:691714. [46] Belman AL, Maletic-Savatic M. HIV and AIDS. In: Goetz C, Papper EJ, editors. Textbook of clinical neurology. 2nd edition. Philadelphia: WB Saunders (in press). [47] Belman AL. Pediatric neuro-AIDS. Update. Neuroimaging Clin North Am 1997;7: 593613. [48] American Academy of Neurology AIDS Task Force. Nomenclature and research case denitions for neurologic manifestations of human immunodeciency virus type 1. Neurology 1991;41:77885. [49] Brouwers E, Belman AL, Epstein L. Central nervous system involvement: manifestations and evaluation. In: Pizzo PA, Wilfert CM, editors. Pediatric AIDS: the challenge of HIV infection in infants, children and adolescents. 2nd edition. New York: Williams & Wilkins; 1994. p. 43355. [50] Belman AL, Muenz LR, Marcus JC, et al. Neurologic status of human immunodeciency virus 1-infected infants and their controls: a prospective study from birth to 2 years. Pediatrics 1996;98:110919. [51] Nozyce M, Hittleman J, Muenz L, et al. Eect of perinatally acquired human immunodeciency virus infection on neurodevelopment in children during the rst two years of life. Pediatrics 1994;94:88389. [52] Pearson DA, McGrath NM, Nozyce M, et al. Predicting HIV disease progression in children using measures of neuropsychological and neurological functioning. Pediatric AIDS Clinical Trials 152 Study Team. Pediatrics 2000;106:e76 (Electronic article). [53] Tardieu M, Le Chanadec J, Persoz A, Meyer L, Blanche S, Mayaux MJ. HIV-1 related encephalopathy in infants compared with children and adults. French pediatric HIV infection study and SEROCO group. Neurology 2000;54:108995. [54] Brouwers P, DeCarli C, Civitello AL. Correlation between CT brain scan abnormalities and neuropsychological function in children with symptomatic HIV disease. Arch Neurol 1995;52:3944. [55] Aylward EH, Butz AM, Hutton N, et al. Cognitive and motor development in infants at risk for human immunodeciency virus. Am J Dis Child 1992;146:21822. [56] Chase C, Ware J, Hittelman J, et al. Early cognitive and motor development among infants born to women infected by human immunodeciency virus. WITS. Pediatrics 2000;106:e25 (Electronic article). [57] Gay CL, Armstrong FD, Cohen D, et al. The eects of HIV on cognitive and motor development in children born to HIV-seropositive women with no reported drug use: birth to 24 months. Pediatrics 1995;96:107882. [58] Havens J, Whitaker AH, Feldman JF, Ehrhardt AA. Psychiatric morbidity in schoolaged children with congenital human immunodeciency virus infection: a pilot study. J Dev Behav Pediatr 1994;15(Suppl):S1825. [59] Englund JA, Baker CJ, Raskino C, et al. Clinical and laboratory characteristics of a large cohort of symptomatic, human immunodeciency virus-infected infants and children. Pediatr Infect Dis J 1996;15:102536. [60] Cohen SL, Mundy T, Kaarassik B, et al. Neuropsychological functioning in children with HIV-1 infection through neonatal blood transfusion. Pediatrics 1991;88:5868. [61] Condini A, Axia G, Cattelan C, DUrso MR, Lavera AM, Viero F, et al. Development of language in 18-30 month old HIV-1 infected but not ill children. AIDS 1991;5:7359. [62] Diamond GW, Kaufman J, Belman AL, et al. Characterization of cognitive functioning in a subgroup of children with congenital HIV infection. Arch Clin Neuropsychol 1987;2:16.
1008
[63] Levenson RL, Mellins CA, Zawadzki R, Kairam R, Stein Z. Cognitive assessment of human immunodeciency virus exposed children. Am J Dis Child 1992;146:147983. [64] Preston T, Demarco R, Milazzo M, et al. Visuoconstructive and motor skills in children with vertical human immodeciency-1 infection. Ann Neurol l996 (Abstract). [65] Raskino C, Person DA, Baker CJ, et al. Neurologic, neurocognitive and brain outcomes in human deciency virus-infected children receiving dierent nucleoside antiretroviral regimens. Pediatric AIDS Clinical Trials Group 152 Study Team. Pediatrics 1999;104:e32 (Electronic article). [66] Tardieu M, Mayaux M-J, Seibel N, et al. Maternally transmitted HIV-1 infection: cognitive assessment of school age children. J Pediatr 1995;126:3759. [67] Wolters PL, Brouwers P, Moss HA, Pizzo PA. Dierential receptive and expressive language functioning of children with symptomatic HIV disease: comparison with the language of sibling controls and CT scan brain abnormalities. Pediatrics 1995;1:1129. [68] Chirboga CA, Champion S, Worrell C, Abrams EJ. Changing face of human immunodeciency virus encephalopathy in the post-highly active antiretroviral therapy era. Ann Neurol 2001 (Abstract). [69] Barre-Sinoussi F, Cherman JC, Rey F, et al. Isolation of a T-lymphotrophic virus from a patient at risk for acquired immunodeciency virus (AIDS). Science 1983;220:86871. [70] Gallo RC, Salahuaddin SV, Popovic M, et al. Frequent detection and isolation of cytopathic retroviruses (HTLV-III) from patients with AIDS and at risk for AIDS. Science 1984;224:5003. [71] Levy JA, Homan AD, Kramer SM, et al. Isolation of lymphotrophic retroviruses from San Francisco patients with AIDS. Science 1984;225:8402. [72] Ho DD, Rota TR, Schooley RT, et al. Isolation of HTLV-III from CSF and neural tissues of patients with AIDS related neurologic syndromes. N Engl J Med 1985;313: 14937. [73] Koenig S, Gendelman HE, Orenstein JM, et al. Detection of AIDS virus in macrophage in brain tissue from AIDS patients with encephalopathy. Science 1986;233:108993. [74] Shaw GM, Harper ME, Hahn BH, et al. HTLV-III infection in brains of children and adults with AIDS encephalopathy. Science 1985;227:17781. [75] Wiley CA, Schrier RD, Nelson AS, et al. Cellular localization of human immunodeciency virus infection within the brains of acquired immune deciency syndrome patients. Proc Natl Acad Sci USA 1986;83:708993. [76] Tornatore C, Meyers K, Atwood W, Conant K, Major E. Temporal patterns of human immunodeciency virus type 1; transcripts in human fetal astrocytes. J Virol 1994;68: 93102. [77] Con J, Haase A, Levy JA, et al. Human immunodeciency viruses. Science 1986;232: 697. [78] Gonda MA, Wong-Stall F, Gallo RC, et al. Sequence homology and morphologic similarity of HTLV-III and visna virus, a pathogenic lentivirus. Science 1985;227:1737. [79] Johnson RT, McArthur JC, Narayan O. The neurobiology of human immunodeciency virus infections. FASEB J 1988;2:297081. [80] Budka H, Wiley CA, Kleihues P, et al. HIV-1 associated disease of the nervous system: review of nomenclature and proposal for neuropathology-based terminology. Brain Pathol 1991;1:14352. [81] Dickson DW, Lee SC, Hatch W, Mattiace LA, Broshan C, Lyman WD. Macrophages and microglia in HIV-related CNS neuropathology. In: Price RW, Perry S, editors. HIV; AIDS and the brain. New York: Raven Press; 1993. p. 99118. [82] Kure K, Park YD, Kim TS, et al. Immunohistochemical localization of an HIV epitope in cerebral aneurysmal arteriopathy in pediatric AIDS. Pediatr Pathol 1989;9:65562. [83] Kure K, Llena JF, Lyman WD, et al. Human immunodeciency virus-1 infection of the nervous system: an autopsy study of 268 adult, pediatric and fetal brains. Hum Pathol 1991;22:70010.
1009
[84] Johnson RT. The pathogenesis of HIV infections of the brain. Curr Top Microbiol Immunol 1995;202:310. [85] Lipton SA, Gendelman HE. Dementia associated with the acquired immunodeciency syndrome. Seminars in medicine. The Beth Israel Hospital. N Engl J Med 1995;332:934. [86] Navia BA, Jordon BD, Price RW. The AIDS dementia complex: I. Clinical features. Ann Neurol 1986;19:51724. [87] Jovaisas E, Koch MA, Schafer A, et al. LAV/HTLV-III in 20-week fetus [letter]. Lancet 1985;2:1129. [88] Lyman WD, Kressy Y, Kure K, et al. Detection of HIV in fetal central nervous system tissue. AIDS 1990;4:91720. [89] Soeiro R, Rubinstein A, Rashbaum WK, Lyman WD. Maternofetal transmission of AIDS: frequency of human immunodeciency virus type-1 nucleic acid sequences in human fetal DNA. J Infect Dis 1992;2:2889. [90] Sprechter S, Soumenko G, Puissant F, DeGueldra M. Vertical transmission of HIV in a 15 week fetus. Lancet 1985;2:2889. [91] Volpe J. Neurology of the newborn. 2nd edition. Philadelphia: WB Saunders; 1994. [92] Epstein LG, Gelbard HA. HIV-1-induced injury in the developing brain. J Leukoc Biol 1999;65:4537. [93] Kourtis AP, Ibegbu C, Nahmias AJ, et al. Early progression of disease in HIV-infected infants with thymus dysfunction. N Engl J Med 1996;335:14316. [94] Nahmias AJ, Lee FK, Kourtis A, Ibegbu C. Immunology of HIV infection in the fetus and newborn. HIV and women and pregnancy. Immunol Allergy Clin North Am 1998; 18:40119. [95] Dickson DW, Belman AL, Park YD, et al. Central nervous system pathology in pediatric AIDS: an autopsy study. APMIS 1989;8(Suppl):4057. [96] Kozlowski PB, Brudkowsks J, Kraszpulsi M, Sersen EA, Wrzolek MA, Anzil AP, et al. Microencephalopathy in children congenitally infected with human immunodeciency virusa gross-anatomical morphometric study. Acta Neuropathol (Berl) 1997;93: 13645. [97] Vazeux R, Lacroix-Ciaudo C, Blanche S, et al. Low levels of human immunodeciency virus replication in the brain tissue of children with severe acquired immunodeciency syndrome encephalopathy. Am J Pathol 1992;140:13744. [98] Wiley CA, Belman AL, Dickson D, Rubinstein A, Nelson JA. Human immunodeciency virus within the brains of children with AIDS. Clin Neuropathol 1990;1:16. [99] Wiley CA, Masliah E, Morey M, et al. Neocortical damage during HIV infection. Ann Neurol 1991;29:6517. [100] Berger J, Albrecht J, Belman AL, et al. Progressive multifocal leukoencephalopathy in children with HIV infection. AIDS 1992;6:83741. [101] Biggemann B, Voit T, Neuen E, et al. Neurological manifestations in three German children with AIDS. Hippokrates 1987;1:99106. [102] Frank Y, Pavlakis SG, Lu D, et al. Childhood AIDS and varicella vasculopathy. J Child Neurol 1977;12:4646. [103] Kozlowski PB, Sher JH, Rao C, et al. Central nervous system pathology in pediatric AIDS registry. Ann NY Acad Sci 1993;693:2956. [104] McAbee GN, Belman AL, Knapik M, Solitare G, Ciminera P, Dickson D. Rapid and fatal neurological deterioration due to CNS candida infection in an HIV-1 infected child. J Child Neurol 1995;10:4057. [105] Mitchell CD, Erlich SS, Mastrucci MT, et al. Congenital toxoplasmosis occurring in infants perinatally infected with HIV-1. Pediatr Infect Dis J 1990;9:5128. [106] Shanks GD, Redeld RR, Fischer GW. Toxoplasma encephalitis in an infant with acquired immunodeciency syndrome. Pediatr Infect Dis 1987;6:701. [107] Sharer LR, Fischer B, Blumberg BM, Epstein LG. CNS measles virus infection in AIDS. J Neuro-AIDS 1994 (Abstract).
1010
[108] Silliman CC, Tedder D, Ogle JW, et al. Unsuspected varicella-zoster virus encephalitis in a child with acquired immunodeciency syndrome. J Pediatr 1993;123:41822. [109] Taccone A, Fondelli MP, Ferra G, Marzoli A. An unusual CT presentation of congenital cerebral toxoplasmosis in an 8 month-old boy with AIDS. Pediatr Radiol 1992;22:689. [110] Van der Steenhoven JJ, Dbaibo G, Boyko OB, et al. Progressive multifocal leukoencephalopathy in pediatric acquired immunodeciency syndrome. Pediatr Infect Dis J 1992;3:232237. [111] Biggar RJ, Frisch M, Goedert JJ. AIDS-Cancer March Registry Study Group. Risk of cancer in children with AIDS. JAMA 2000;284:2059. [112] Epstein LG, Dicarlo F, Joshi V, et al. Primary lymphoma of the central nervous system in children with acquired immunodeciency syndrome. Pediatrics 1988;82:35563. [113] Goldstein J, Dickson DW, Rubinstein A, et al. Primary CNS lymphoma in a pediatric patient with acquired immunodeciency syndrome treatment with radiation. Cancer 1990;66:25038. [114] Park Y, Belman AL, Dickson DW, et al. Stroke in pediatric acquired immunodeciency syndrome. Ann Neurol 1990;28:30311. [115] Husson RN, Saini R, Lewis LL, Butler KM, Patronas N, Pizzo PA. Cerebral artery aneurysms in children aected with human immunodeciency virus. J Pediatri 1992; 121:92730. [116] Philippet P, Blanche S, Sebag G, Rodesch G, Griscelli C, Tardieu M. Stroke and cerebral infarct in children infected with human immunodeciency virus. Arch Pediatr Adolesc Med 1994;148:96570. [117] Frank Y, Lim W, Kahn E, et al. Multiple ischemic infarcts in a child with AIDS, varicella zoster infection and cerebral vasculitis. Pediatr Neurol 1989;5:647. [118] Araujo AP, Nascimento OJ, Garcia OS. Distal sensory polyneuropathy in a cohort of HIV-infected children over ve years of age. Pediatrics 2000;106:E35 (Electronic article). [119] Raphael SA, Price ML, Lischner HW, et al. Inammatory demyelinating polyneuropathy in a child with symptomatic human immunodeciency virus infection. J Pediatr 1991; 118:2425. [120] Challenges in designing HIV vaccine. NIAID fact sheet. Available at: http://www.niaid. nih.gov/factsheets/challvacc.thm. Accessed January 2002. [121] Czarniecki LM, Oleske JM. Pain in children with HIV disease. PAAC Notes 1993;5: 4925. [122] Schechter NL, Berde CB, Yaster M, editors. Pain in infants, children, and adolescents. Baltimore: Williams & Wilkins; 1993. [123] Centers for Disease Control and Prevention. 1994 Revised classication system for human immunodeciency virus infection in children less than 13 years of age; ocial authorized addendahuman immunodeciency virus infection codes and ocial guidelines for coding and reporting ICD-9-CM. MMWR Weekly 1994;43:110. [124] Chamberlin MC. Pediatric AIDS: a longitudinal comparative MRI and CT brain imaging study. J Child Neurol 1993;8:17581. [125] DeCarli C, Fugate L, Falloon J, et al. Brain growth and cognitive improvement in children with HIV induced encephalopathy after 6 months of continuous infusion therapy. AIDS 1991;4:58592. [126] DeCarli C, Civitello AL, Brouwers P, Pizzo PA. The prevalence of computed tomographic abnormalities of the cerebrum in 100 consecutive children symptomatic with the human immune deciency virus. Ann Neurol 1993;34:198205. [127] States LJ, Zimmerman RA, Rutstein RM. Imaging of pediatric central nervous system HIV infection. Neuroimaging Clin North Am 1997;7:32139. [128] Depas G, Chiron C, Tardieu M, et al. Functional brain imaging in HIV-1-infected children born to seropositive mothers. J Nucl Med 1995;36:216974. [129] Gabis L, Belman AL, Huang W, et al. Clinical and imaging study of HIV infected school aged children on HAART therapy. Pediatr Res 2000 (Abstract).
1011
[130] Pavlakis SG, Lu D, Frank Y, et al. Magnetic resonance spectroscopy in childhood AIDS encephalopathy. Pediatr Neurol 1995;12:27782. [131] Salvan AM, Lamoureux S, Michel G, et al. Localized and proton MRS of the brain in children infected with HIV with and without encephalopathy. Pediatr Res 1998;44:75562. [132] Brustels O, Spiegel H, Leib SL, et al. Distribution of HIV in the CNS of children with severe HIV encephalopathy. Academic Neuropathology 1992;84:2431. [133] Burns DK. The neuropathology of pediatric acquired immunodeciency syndrome. J Child Neurol 1992;7:33246. [134] Dickson DW, Belman AL, Kim TS, Horoupian D, Rubinstein A. Spinal cord pathology in pediatric acquired immunodeciency syndrome. Neurology 1989;39:22735. [135] Sharer LR, Dowling PC, Michaels J, et al. Spinal cord disease in children with HIV-1 infection: a combined molecular and neuropathological study. Neuropathol Appl Neurobiol 1990;16:31731. [136] Kure K, Weidenheim KM, Lyman WD, Dickson DW. Morphology and distribution of HIV-1 qp 41 positive microglia in subacute AIDS encephalitis. Acta Neuropathol (Ber1) 1990;80:393400. [137] Levy JA, Shimabukuro J, Hollander H, et al. Isolation of AIDS associated retrovirus from cerebrospinal uid and brain of patients with neurological symptoms. Lancet 1985; 2:5868. [138] Lu D, Pavlakis SG, Frank Y, et al. Magnetic spectroscopy of the basal ganglia in normal children and children with AIDS. Radiology 1996;199:4238. [139] Wiley CA, Masliah E, Achim CL. Measurement of CNS HIV burden and its association with neurologic damage. Neuroimmunology 1994;4:31925.
Neurol Clin N Am 20 (2002) 10131038
Meningitis and encephalitis in children An update

Daniel J. Bonthius, MD, PhDa,*, Bahri Karacay, PhDb
Departments of Pediatrics, Neurology, and Anatomy and Cell Biology, University of Iowa College of Medicine, Iowa City, IA, USA b Department of Pediatrics, University of Iowa College of Medicine, Iowa City, IA, USA
a
A discussion of meningitis and encephalitis requires the use of superlatives. The clinical presentation of these diseases can be among the most dramatic in all of pediatrics. The disorders, among the most devastating in all of neurology and pediatrics, can strike and kill rapidly. Immunizations aimed at preventing these diseases and antibiotics used to treat them represent some of the most spectacular successes of modern medicine. At the same time, the infectious agents responsible for many cases of these disorders are ones against which modern medicine struggles but remains virtually helpless. Meningitis and encephalitis lack nothing for drama.
Denitions and overview Meningitis is inammation of the two meningeal membranes (arachnoid and pia mater) that surround the brain and spinal cord. The space between these membranes is the subarachnoid space, which contains cerebrospinal uid (CSF). In the process of meningitis, inammatory cells spill into the CSF from the meninges, producing an increased cell count and a diagnostic hallmark of the disease. The classic signs and symptoms of meningitis are headache, fever, and nuchal rigidity.
The authors are supported by Public Health Service grant KO8 NS02007-01A2 from the National Institute of Neurological Disorders and Stroke, by research grant 1-01-217 from the March of Dimes Birth Defects Foundation, by a Carver Medical Research Initiative Grant, by the John Martin Fund for Neuroanatomic Research, and by a Young Investigator Award from the Child Neurology Society. * Corresponding author. Department of Pediatrics, University of Iowa Hospital, 200 Hawkins Drive, Iowa City, IA 52242, USA. E-mail address: daniel-bonthius@uiowa.edu (D.J. Bonthius). 0733-8619/02/$ - see front matter 2002, Elsevier Science (USA). All rights reserved. PII: S 0 7 3 3 - 8 6 1 9 ( 0 2 ) 0 0 0 1 6 - 6
1014
D.J. Bonthius, B. Karacay / Neurol Clin N Am 20 (2002) 10131038
Encephalitis is inammation of brain parenchyma. Evidence of this inammation may be gained from clinical history, patient examination, and laboratory testing. The clinical hallmarks of encephalitis include fever, headache, and brain dysfunction, which may manifest as a diminished level of consciousness, focal neurologic decits, or focal or generalized seizure activity. Although meningitis and encephalitis are considered separate entities, in reality, inammation of the meninges often spreads to involve adjacent brain parenchyma, and inammation of the brain parenchyma often involves the meninges. It is probably because of this spread of inammation that patients with meningitis often also have seizures and encephalopathy, whereas patients with encephalitis often have a CSF pleocytosis. In this sense, most cases of meningitis and encephalitis are actually cases of meningoencephalitis. Nevertheless, despite this neuropathologic reality, it remains useful to consider meningitis and encephalitis separately. The signs and symptoms of either meningitis or encephalitis usually predominate in any given case, and the agents responsible for the diseases overlap only minimally. Both meningitis and encephalitis can be acute, subacute, or chronic disorders, and both can be either infectious or noninfectious in etiology. Within the pediatric population, however, most of these cases are acute and caused by infectious agents. Therefore, this article focuses on the infectious causes of acute meningitis and encephalitis in children. New information regarding the pathophysiology, etiology, treatment, and prevention of infectious meningitis and encephalitis is discussed. Subacute and chronic meningitis and encephalitis and the noninfectious causes of these disorders have recently been reviewed elsewhere [1,2]. Meningitis Pathophysiology of bacterial meningitis Until recently, it was assumed that damage to the central nervous system (CNS) accompanying bacterial meningitis was directly induced by the infecting pathogens. Within the past decade, much has been discovered regarding the pathogenesis of bacterial meningitis, including the key fact that much of the brain injury accompanying bacterial meningitis is inammation mediated. Outside the neonatal period, most bacteria that cause meningitis initially colonize the nasopharynx by attaching to nasopharyngeal epithelial cells [3]. (In the neonate, colonization of the nasopharynx probably occurs only rarely, and the infection is blood borne from an unknown initial site.) To facilitate attachment to the nasopharyngeal mucosal epithelial cells, bacteria secrete IgA proteases that break down the mucus barrier and extend cellular processes (pili) that adhere to host cell surface receptors. Once the bacteria have successfully attached to the nasopharyngeal mucosa, they travel across the epithelial membrane to invade the adjacent
1015
intravascular space. The bacteria traverse the epithelial membrane either within intracellular vacuoles or by creating separations in the apical tight junctions of the columnar epithelial cells [3]. Bacterial invasion of the intravascular space induces a bacteremia and triggers the hosts initial complement-mediated defense. Some bacteria evade the complement pathway and avoid neutrophilic phagocytosis via the presence of a polysaccharide capsule. Most bacterial strains that cause meningitis have a capsular polysaccharide and are referred to as encapsulated bacteria [4]. Bacteria that survive in the circulation may then enter the CSF. The mechanism and site at which bacteria invade the CSF are not known for certain, but transmission across the choroid plexus and cerebral capillaries is most likely. Cells of the choroid plexus and cerebral capillaries possess surface molecules that bacteria use as receptors for adherence to specic bacterial structures, thus allowing for attachment and invasion [5]. Once pathogens reach the CSF, they have an excellent chance of continuing to survive, principally because the subarachnoid space is a region of impaired host defense. Before infection, the CSF contains insucient immunoglobulin concentrations, neutrophils, and complement components to inhibit the rapid multiplication of bacteria [6,7]. Eventually, the host does generate an immune response to the bacteria in the CSF. This immune response may eectively kill bacteria, but the killing process releases bacterial cell wall components that trigger an inammatory cascade which often culminates in severe brain injury. Liberation of subcapsular surface components, including the bacterial cell wall and lipopolysaccharide, stimulate endothelial cells and CNS macrophages to produce numerous proinammatory molecules, including tumor necrosis factor (TNF), interleukin (IL)-1, platelet activating factor, and IL-6 [8,9]. In addition, the complement cascade is activated, triggering further inammation and the activation of granulocytes and platelets. A key event in the inammatory cascade accompanying bacterial meningitis is the migration of neutrophils into the CSF and a subsequent breakdown of the bloodbrain barrier. A three-phase model has been proposed to explain these events [3]. The rst phase begins with the release of inammatory cytokines within the CSF in response to bacterial replication and lysis. These inammatory cytokines bind to receptors on the surfaces of endothelial cells, generating the local production of thrombin, which induces the surface expression of selectin molecules (CD62 and endothelial leukocyte adhesion molecule-1). These selectin molecules enhance the binding of neutrophils to the endothelial surface. The prolonged inammatory cytokine stimulation triggers the second phase, in which vascular endothelium releases IL-8. Release of this cytokine induces the binding of neutrophils to endothelial intercellular adhesion molecules (ICAMs), which leads to neutrophil diapedesis and entry into the CSF. In the third phase, cytokines within the CSF degranulate and release vasoactive lipid autocoids, including
1016
leukotrienes and prostaglandins, along with toxic oxygen intermediates. These agents impair the bloodbrain barrier, thus allowing the leakage of albumin into the CSF. The migration of neutrophils into the CSF in bacterial meningitis may produce a thick exudate that encases the brain and spinal cord in pus. Bacterial infection of the meninges and the immune response to the infection can lead to brain injury through at least three broad mechanisms. First, chemical agents released by the bacteria or by cells of the immune system may be directly toxic to neurons. For example, lipopolysaccharide, TNF, platelet activating factor, and nitric oxide are all present in high concentrations in bacterial meningitis and can all damage or kill neurons [1014]. Second, brain injury in bacterial meningitis can be induced through vascular mechanisms. Meningitis, sepsis, and the proinammatory cytokines accompanying them may induce septic shock and global cerebral hypoperfusion [15]. The purulent exudate through which arteries and veins pass in the subarachnoid space may lead to vasospasm and vasculitis with secondary thrombosis, ischemia, and infarction [16]. Bacterial meningitis can also cause loss of cerebral autoregulation with subsequent inability of the vasculature to respond to tissue blood ow demands [17]. Occasionally, necrotizing arteritis may induce subarachnoid hemorrhage [18]. The third mechanism by which bacterial meningitis induces brain dysfunction and injury is through increased intracranial pressure. The toxic, metabolic, and vascular changes accompanying meningitis induce a combination of vasogenic, cytotoxic, and interstitial edema. Endotoxin and fragments of the cell wall of gram-positive organisms induce the release of IL-1 and TNF from macrophages and other sources [19]. These agents act on endothelial cells to increase the permeability of cerebral vasculature [20]. The intercellular junctions, which are normally tightly closed, are opened, thus increasing permeability to circulating albumin and inducing vasogenic edema [21]. Interstitial edema is principally caused by a diminished resorption of CSF. The accumulation of inammatory cells, protein, and other material within the CSF interferes with functioning of the arachnoid villi and blocks the resorption of CSF from the subarachnoid space [22]. Similarly, the ventriculitis that often accompanies meningitis may occlude the cerebral aqueduct and lead to hydrocephalus, further exacerbating interstitial edema and elevating intracranial pressure. Early and common ndings on CT in children with bacterial meningitis are ventriculomegaly and enlargement of subarachnoid spaces [23,24]. Increased intracranial pressure in meningitis induces brain pathologic changes via at least two mechanisms. First, elevation of intracranial pressure reduces cerebral perfusion pressure and can lead to cerebral ischemia [25,26]. Second, if the intracranial pressure rises substantially, as it often does in bacterial meningitis, cerebral herniation may occur. Cerebral herniation is commonly the ultimate cause of death in fatal cases of bacterial meningitis in children [27,28].
1017
The use of corticosteroids in bacterial meningitis In the preantibiotic era, more than 90% of children with bacterial meningitis died from the disease. The introduction of antibiotics and the development of pediatric intensive care have reduced the mortality rate to approximately 5% (although the mortality rate remains higher for some pathogens). Despite the reduction in mortality, substantial morbidity remains. As many as 20% to 30% of the survivors have long-term neurologic sequelae, the most common of which is hearing impairment [29,30]. The discovery that much of the brain injury accompanying bacterial meningitis is inammation mediated sparked the hypothesis that modulation of the immune response may ameliorate the brain damage. Studies employing animal models of bacterial meningitis provided strong evidence supporting this hypothesis. The anti-inammatory steroid dexamethasone substantially reduced levels of the cytokines IL-1 and TNF as well as prostaglandin E2 within the CSF of infected animals [31,32]. These reductions in the chemical mediators of inammation were accompanied by reductions in intracranial pressure, brain edema, and CSF lactate concentrations [33,34]. As these laboratory values were improved by the dexamethasone therapy, so were the outcomes for the experimental animals. Administration of dexamethasone decreased mortality and clinically evident neurologic sequelae in the animals with experimental bacterial meningitis [31]. These positive ndings in animal studies led to clinical studies of dexamethasone in human patients with bacterial meningitis. Since 1988, reports of a dozen randomized and controlled trials of dexamethasone therapy in childhood meningitis have been published [3538]. The results of most of these studies have suggested that dexamethasone can diminish meningeal inammation and decrease the incidence and severity of brain injury in bacterial meningitis. Nevertheless, the issue of whether to use dexamethasone in the treatment of bacterial meningitis remains controversial. Numerous ndings from the clinical studies support the use of dexamethasone in bacterial meningitis. Dexamethasone can improve CSF proles in meningitis by reducing CSF levels of the proinammatory cytokines TNF, IL-1, and platelet-activating factor; can reduce CSF lactate and protein levels; and can increase CSF glucose concentrations [37,38]. Although the steroid reduces CSF inammation, it does not delay CSF sterilization [39]. In the initial hours and days after the administration of dexamethasone for bacterial meningitis, the steroid can reduce intracranial pressure, improve cerebral perfusion pressure and the acute clinical condition of the patient, and reduce the number of days of fever and the number of seizures while the patient is hospitalized [37,38,40]. Most importantly, administration of dexamethasone to children with bacterial meningitis can reduce fatality as well as the frequency and severity of both acute and long-term neurologic and audiologic sequelae [3740]. Despite these positive ndings, the enthusiasm of many experts for the use of dexamethasone in bacterial meningitis is dampened by several facts.
1018
First, the eect of dexamethasone seems to be pathogen specic. Dexamethasone may improve outcome in pneumococcal meningitis and Haemophilus inuenzae meningitis; however, in meningitis caused by other pathogens, there is no compelling evidence that dexamethasone is of any benet. Second, H. inuenzae was the predominant pathogen in most of the clinical studies; however, as discussed below, this pathogen is now responsible for only a small portion of bacterial meningitis. Third, most of the studies were performed before the widespread problem of penicillin-resistant Streptococcus pneumoniae emerged. Thus, the eectiveness of dexamethasone in the treatment of this critically important pathogen is unclear. Fourth, although adverse eects of dexamethasone in the treatment of meningitis have not been common, several patients have developed gastrointestinal bleeding requiring blood transfusions [39,41]. Finally, the anti-inammatory property for which dexamethasone is employed may have its own negative eects by decreasing CSF penetration of some antimicrobials, such as vancomycin, and by interfering with the ability to interpret clinical response, such as resolution of fever [4244]. This latter problem necessitates that all infants and children treated with dexamethasone for meningitis undergo a repeat lumbar puncture after 24 to 48 hours of therapy so as to verify CSF sterilization. The role of corticosteroids in the treatment of bacterial meningitis remains controversial. The strength of the animal research and clinical studies persuaded the American Academy of Pediatrics to recommend the use of dexamethasone for the treatment of meningitis caused by H. inuenzae [45] and to consider dexamethasone if the meningitis is caused by S. pneumoniae [46]. At the time of the initial diagnosis of meningitis, however, the etiologic organism is unknown, and many experts in the treatment of meningitis do not necessarily agree with or follow the American Academy of Pediatrics recommendations [4749]. Despite the controversy, three principles regarding the use of dexamethasone enjoy wide agreement. First, if dexamethasone is used, the rst dose ought to be administered before or with the rst dose of antibiotics and not after antibiotics have already been started [50]. The biologic role of the corticosteroid is to block the upregulation of the inammatory mediators induced by the products of bacterial lysis [51]. Administration of antibiotics before the corticosteroid would lyse the bacteria, trigger the inammatory cascade, and render the corticosteroid less eective. Second, if dexamethasone is used, the dose should be 0.6 mg/kg/d in four divided doses given intravenously for the rst 4 days of antibiotic treatment [140]. Lower doses are less eective, and higher doses have not been studied systematically. Third, dexamethasone in the treatment of meningitis should be considered only for patients who are 6 weeks of age or older. The ecacy of dexamethasone has not been studied in younger infants, where the etiologic organisms and pathophysiology of meningitis are dierent than in older children.
1019
The changing etiology and epidemiology of bacterial meningitis in children Impact of Haemophilus inuenzae vaccination. Until recent times, bacterial meningitis was a disease primarily of children, and the main causative organism was H. inuenzae. During the recent past, H. inuenzae meningitis or invasive disease developed in nearly 1 in 200 children by the time they were 5 years of age [52]. An epidemiologic study in 1986 revealed that the median age of children with bacterial meningitis was 15 months and that H. inuenzae was responsible for 45% of all cases of bacterial meningitis and for 70% of the cases in children between 1 month and 5 years of age [53]. Just 9 years later, however, another epidemiologic study showed that the median age of persons with bacterial meningitis had risen to 25 years and that H. inuenzae was a relatively rare cause of bacterial meningitis in any age group, including children [54]. With the decline in H. inuenzae, pneumococci and meningococci are now the principal causes of bacterial meningitis in children from 1 month to 18 years of age. In neonates, group B streptococcus, Escherichia coli, and Listeria monocytogenes remain the primary culprits [55,56]. Development and administration of eective vaccines for these pathogens could make bacterial meningitis in childhood largely a relic of the past. This dramatic change in the epidemiology of bacterial meningitis was a result of the introduction in 1988 of the conjugated H. inuenzae type b vaccine for children more than 2 years of age and in 1990 for infants 2 months of age and older. Routine administration of this vaccine to infants and children has decreased the incidence of bacterial meningitis in children between 1 month and 5 years of age by 87%. Because this age group previously represented almost two thirds of the cases of bacterial meningitis, protection of children against H. inuenzae meningitis has resulted in a 55% decline in all cases of bacterial meningitis [54]. Unquestionably, development of the conjugated H. inuenzae vaccine represents one of the most important events in the history of pediatric infectious disease. Impact of penicillin-resistant Streptococcus pneumoniae. The second recent development of great importance in the etiology of bacterial meningitis is the emergence of penicillin-resistant S. pneumoniae. Antibiotic resistance is of particular concern because of the prevalence of this pathogen as a cause of meningitis. S. pneumoniae, the leading cause of bacterial meningitis in the 1-month to 2-year-old age group and in adults, is second only to Neisseria meningitidis in the 2- to 18-year-old age group [54]. Indeed, S. pneumoniae is the principal etiologic cause of bacterial meningitis in human beings. In the microbiology laboratory, penicillin susceptibility testing classies S. pneumoniae isolates as either susceptible (minimal inhibitory concentration [MIC] < 0.06 lg/mL), intermediate resistant (MIC 0.11.0 lg/mL), or highly resistant (MIC > 2.0 lg/mL) [57]. Until recently, virtually all isolates of S. pneumoniae were susceptible to penicillin and to the cephalosporins.
1020
This susceptibility allowed for the simple and eective use of monotherapy with a single antibiotic (usually a third-generation cephalosporin or a penicillin derivative) for the empiric and specic treatment of most children with bacterial meningitis outside the neonatal period. National studies in the United States have revealed that the prevalence of both intermediate-resistant and highly resistant clinical isolates of S. pneumoniae have increased at alarming rates for the past decade. A study conducted in 1999 through 2000 revealed that 34.2% of S. pneumoniae clinical isolates were penicillin resistant and that 21.5% were highly resistant [58]. The authors of this study had conducted an earlier national study in 1994 through 1995. When they compared the earlier and latter studies, the authors found that the prevalence of penicillin resistance in S. pneumoniae isolates in 1999 through 2000 had increased by 10.6% in 5 years. The molecular basis of penicillin resistance is mutation in the bacterial genes encoding for high-molecular-weight penicillin-binding proteins. These proteins normally function as enzymes that play important roles in the synthesis and modication of bacterial cell walls [59,60]. Amino acid substitutions within these proteins diminish the anity of the proteins for penicillin. The more substantial the change in these proteins, the greater is the resistance to penicillin. Because all b-lactam antibiotics interact with penicillin-binding proteins, modication of the proteins can lead to crossresistance among penicillins, cephalosporins, and carbapenems [57]. Luckily, alteration in the penicillin-binding proteins does not confer resistance to vancomycin. Resistance of S. pneumoniae to vancomycin has not yet been reported. The emergence of penicillin resistance may have been caused, in part, by genetic drift, but the actions of physicians have undoubtedly played an important role. The overuse of antibiotics, particularly the administration of antibiotics to children with fever, has led to an environment in which mutations encoding penicillin resistance have been naturally selected. A newly transformed pneumococcus may be as viable as the original nontransformed strain and demonstrates a selective advantage in an environment containing penicillin. The transformed organism may proliferate and colonize the original host and the hosts contacts. In this way, penicillinresistant clones have emerged at focal locations in the world and have subsequently disseminated across the globe. Within the United States, day care centers seem to serve as major sites of pneumococcal spread. One study documented the extensive spread of a particular penicillin-resistant pneumococcal serotype within a day care center [61]. After a child is colonized with S. pneumoniae in a day care setting, the pathogen is often introduced into the household and subsequently spreads to others within the household and beyond [62]. In the very recent past, these obstacles to development of an eective pneumococcal vaccine have been overcome. The issue of inadequate immunogenicity was resolved by conjugation of the polysaccharide antigens
1021
to a protein carrier. The problem of multiple pneumococcal serotypes was largely overcome by taking advantage of the fact that seven serotypes are responsible for the vast majority of invasive disease in children. The increased prevalence of penicillin-resistant S. pneumoniae has dictated important changes in laboratory testing and in antibiotic therapy whenever meningitis caused by S. pneumoniae is suspected. No longer can it be assumed that penicillin or a third-generation cephalosporin alone can adequately treat the infection. All S. pneumoniae isolates from CSF and all other normally sterile body uids should be tested for in vitro antimicrobial susceptibility to determine the MIC to penicillin, cefotaxime, and ceftriaxone. Isolates found to have intermediate or high resistance should have additional susceptibility testing to vancomycin, rifampin, and meropenem. Children 1 month of age and older with denite or suspected bacterial meningitis should initially receive combination therapy with vancomycin and ceftriaxone or cefotaxime. This empiric therapy is particularly indicated if the Gram stain of CSF reveals gram-positive diplococci, a nding strongly suggestive of pneumococcal meningitis. If a child has a severe allergy to the b-lactam antibiotics, treatment should initially consist of combination therapy with vancomycin and rifampin. Antibiotic susceptibility results should guide subsequent treatment. If the pneumococcal isolate is susceptible to penicillin, cefotaxime, or ceftriaxone, one of these three antibiotics should be continued and vancomycin should be promptly discontinued. Discontinuation of vancomycin minimizes the risk of vancomycin toxicity in the patient and vancomycin resistance in S. pneumoniae. If the isolate has intermediate- or high-level resistance to penicillins and to the cephalosporins, vancomycin should be continued along with ceftriaxone or cefotaxime. In vitro experiments and animal studies of penicillin- and cephalosporin-resistant meningitis have demonstrated that the combination of vancomycin and ceftriaxone has greater antimicrobial activity than either agent used alone [63]. Vancomycin as a single agent for the treatment of bacterial meningitis is not recommended because of its relatively poor penetration into the CSF. Attempts to treat pneumococcal meningitis with vancomycin alone have resulted in clinical failure [64]. If the isolate has an unusually high level of resistance to cephalosporins (MIC > 4 lg/mL) or if the clinical course or repeat lumbar puncture suggests that cephalosporin and vancomycin are not adequately eective, addition of rifampin or meropenem to vancomycin should be considered [65]. The increasing prominence of pneumococcal infections and antimicrobial resistance provided a strong impetus for development of an eective vaccine to reduce the risk of pneumococcal disease in children. Development of a pneumococcal vaccine was complicated by the existence of more than 90 different serotypes of pneumococcus, many of which cause invasive disease. Serotyping of S. pneumoniae is based on antigenic dierences of the capsular polysaccharides [66]. Pneumococcal vaccine development was further complicated by the fact that puried capsular polysaccharide antigens do not
1022
generate an adequate immune response in young children. Thus, the 23valent pneumococcal vaccine, which was composed of puried capsular polysaccharide antigens of 23 pneumococcal serotypes, was ineective in preventing invasive pneumococcal infections in young children [67]. The heptavalent pneumococcal conjugate vaccine (PCV7) was evaluated for ecacy, immunogenicity, and safety in a large trial in northern California between 1995 and 1999 [68]. The results were spectacularly positive. The trial demonstrated a vaccine ecacy of 97.4% and reduced the total invasive pneumococcal disease burden in children by 89.1%. The vaccine induced a substantial immunologic response to pneumococcal polysaccharide of all seven serotypes included in the conjugate. Safety monitoring revealed no severe adverse events related to vaccination. In 2000, the PCV7 vaccine was approved by the US Food and Drug Administration and recommended by the American Academy of Pediatrics for all children from 2 to 23 months of age and for children from 24 to 59 months of age who are at increased risk of developing invasive pneumococcal disease [50]. Since then, shortages of PCV7 have necessitated revision of recommendations for use of this vaccine [69]. It is anticipated that the incidence of disease caused by the serotypes included in the vaccine will be greatly diminished. In addition, the vaccine seems to induce cross-protection among related serotypes and may reduce the incidence of infections caused by serotypes not included in the vaccine [68]. Furthermore, because the strains included in the PCV7 vaccine are the same strains that have developed resistance to penicillin and other antibiotics [57], the reduction in disease induced by these strains may reduce the use of broad-spectrum antibiotics and slow the development of antibiotic resistance. Lymphocytic choriomeningitis virus Lymphocytic choriomeningitis virus (LCMV), rst isolated in 1933 from a patient with meningoencephalitis, was named for the intense lymphocytic inltration that it induced in the choroid plexus and meninges of infected patients and laboratory animals [70]. LCMV is a member of the Arenaviridae family of viruses, and like all arenaviruses, it utilizes rodents as its principal reservoir. The common house mouse, Mus musculus, is both the natural host and reservoir for the virus, which is vertically transferred from one generation of mice to the next via intrauterine infection. Mice that acquire the virus transplacentally often remain asymptomatic despite heavy infection, because LCMV is not cytolytic and because congenital infection in rodents provides them with immunologic tolerance for the virus. Throughout their lives, mice congenitally infected with LCMV shed large quantities of the virus in urine, saliva, and other bodily secretions [71]. A substantial proportion of wild mice harbor the virus, and large numbers of people become infected [7275].
1023
Children and adults acquire LCMV by inhalation of aerosolized virus or by direct contact with the contaminated fomites. Most commonly, infection of people occurs during the autumn and winter months, when mice cohabit with human beings. Congenital infection with LCMV occurs when a pregnant woman acquires the virus and transmits LCMV to the fetus, presumably during maternal viremia. The virus can also be acquired during the intrapartum period, probably by exposure to maternal blood or vaginal secretions during maternal viremia [76]. During the past decade, LCMV was identied as an important and underrecognized cause of neurologic birth defects in the United States [7780]. The clinical spectrum of postnatal LCMV infection is broad; in as many as one third of acquired infections, infection is asymptomatic. When disease occurs, infection typically manifests as a self-limited febrile illness. Because the symptoms are usually suciently mild, many patients with acquired LCMV infection do not seek medical attention, and most cases are never formally diagnosed. Patients occasionally have an aseptic meningitis with headache, nuchal rigidity, vomiting, fever, photophobia, and malaise. The duration of the illness is usually only 1 to 3 weeks, and patients recover fully [81]. LCMV infection can sometimes be more severe and produce transverse myelitis, hydrocephalus, and encephalitis. Rarely, an acquired LCMV infection is fatal [82]. The hallmark laboratory abnormality of LCMV infection is a CSF pleocytosis. The CSF may contain hundreds or thousands of white blood cells per cubic millimeter, most of which are lymphocytes. CSF eosinophilia has also been reported [83]. Hypoglycorrhachia and mild elevations of CSF protein are commonly observed. Whereas most postnatal LCMV infections in children and adults are self-limited and benign, LCMV infection in utero can be devastating. Infection of the human fetus can induce spontaneous abortion and fetal death [84]. Among those fetuses that survive, hallmarks of congenital LCMV infection are vision impairment and brain dysfunction [85,141]. Vision impairment is caused principally by chorioretinitis and the formation of chorioretinal scars [86] that may mimic those of congenital toxoplasmosis [87]. Other ocular abnormalities associated with intrauterine infections include optic atrophy, nystagmus, vitreitis, strabismus, micro-ophthalmia, and cataracts [78]. Intrauterine infection with LCMV often leads to either macrocephaly or microcephaly. When macrocephaly develops, it is almost invariably caused by noncommunicating hydrocephalus. Microcephaly accompanying congenital LCMV infection is caused by a virus-induced failure of brain growth. In addition to hydrocephalus and microencephaly, other abnormalities of brain structure commonly induced by congenital LCMV infection include periventricular mineralizations, cortical dysplasia, and cerebellar hypoplasia (Figs. 1 and 2) [77,79,80,86]. (The reader is referred to additional discussion of congenital LCMV virus infection elsewhere in this issue).
1024
1025
The cellular and molecular mechanisms underlying LCMV-induced damage to the developing human brain are unknown. LCMV is not cytolytic in most cell types, including glia and neurons [71]. Thus, unlike herpes simplex virus and several other pathogens that induce brain damage by directly killing host cells, LCMV teratogenesis must have some other underlying pathogenesis. Studies in our laboratory using a developing rat model indicate that LCMV produces neuropathologic changes closely mimicking those found in human beings (see Figs. 1 and 2) [88,89]. These pathologic changes include microcephaly, cerebellar hypoplasia, chorioretinitis, neuronal loss, and neuronal migrational disturbances [9092]. Developing rats infected with LCMV have learning decits, ataxia, movement disturbances, and seizure disorders, as do children with congenital LCMV infection [93,94]. Within the developing rat brain, glial cells play a critical role in the pathogenesis of the disease [88]. Glial cells, including astrocytes and radial glia, are the initial targets of the virus within the brain parenchyma. In addition, glial cells are the principal cell types in which the virus replicates and are the conduit through which the virus reaches susceptible neurons. Because radial glia cells constitute the scaolding along which neurons migrate to reach their nal destinations, it is likely that infection of these radial glia and disruption of their function underlie the neuronal migration disturbances evident in congenital LCMV infection (see Fig. 2). Mitotically active cells, such as the periventricular cells that give rise to neurons and glia (see Fig. 1), are particularly vulnerable to LCMV infection [89,9597]. Periventricular calcications observed in human patients with congenital LCMV infection are a reection of the preferential infection and subsequent death of the neuronal precursors residing there. The noncommunicating hydrocephalus commonly observed in children with congenital LCMV is likely caused by ependymal inammation within the ventricular system, particularly at the cerebral aqueduct, producing blockage of CSF egress. Acute LCMV infections in children and adults can be diagnosed by isolating the virus from CSF. By the time of birth, however, infants infected with LCMV prenatally may no longer harbor the virus. Thus, congenital
b Fig. 1. Periventricular pathologic ndings in congenital lymphocytic choriomeningitis virus (LCMV) infection. (A) Head CT scan from a 3-day-old baby with congenital LCMV infection demonstrates periventricular mineralization (arrows) adjacent to the lateral ventricles. Mineralization in this location is common in congenital LCMV infection and likely reects the preferential infection of mitotically active neuronal precursors within the subependymal zone. (B) Horizontal section through the forebrain of a 25-day-old rat infected as a neonate with LCMV. The section is stained with a polyclonal antibody for LCMV. LV, lateral ventricle. Note the specic infection of cells (arrows) adjacent to the lateral ventricle (magnication 15). (C) Inset from part B. The infected cells (arrows) have the morphology of neurons. Thus, within the developing human and rat brains, the periventricular neurons are particularly susceptible to LCMV infection (magnication 70).
1026
1027
LCMV infection is usually diagnosed serologically. The immunouorescent antibody test detects both IgM and IgG and has greater sensitivity than the more widely available complement xation method [74]. An LCMV-specic ELISA, which measures titers of LCMV IgG and IgM, is performed at the Centers for Disease Control and Prevention. The polymerase chain reaction (PCR) has also been used to detect LCMV RNA sequences in an infected infant [79]; however, because LCMV does not induce persistent infections in human beings, PCR may have low sensitivity in detecting intrauterine infections.
Encephalitis West Nile virus encephalitis West Nile virus (WNV) rst appeared in the Americas in the summer of 1999 when it caused an epidemic of human neurologic disease in and around New York City [98]. Since then, the virus has become enzootic within reservoir and vector species in the United States, has spread geographically, and has continued to cause neurologic disease in human beings. Although the frequency and severity of disease induced by the virus tend to be greater in the elderly, children are vulnerable to the infection and its consequences. WNV, an arbovirus (arthropod-borne virus) in the Flaviviridae family, was initially isolated in 1937 from a woman with a febrile illness in the West Nile district of Uganda. Since then, the virus has been found throughout much of the African continent and in the Middle East and Europe along major bird migration yways [99]. The geographic distribution of the virus correlates with the migratory patterns of birds in that part of the world, because birds are the main reservoir of the virus [100]. Mosquitoes of the Culex family, the principal vectors of WNV, transmit the virus between the natural bird hosts. Culex mosquitoes feed primarily on birds but bite human beings and other mammals. Unlike birds, people do not develop a viremia sucient to continue the transmission cycle if bitten
b Fig. 2. Lymphocytic choriomeningitis virus (LCMV) induces a neuronal migrational disturbance in the developing brain. (A) MRI scan from a 3-year-old child with congenital LCMV infection. Note the abnormal gyral pattern (arrow) and paucity of underlying white matter (arrowheads), which are strongly suggestive of disrupted neuronal migration. (B) Normal cerebellar cortex from a control (uninfected) adult rat consists of a molecular layer (M), Purkinje cell layer (P), and granule cell layer (G). Within the molecular layer, a few basket cells and stellate cells (arrows) are normally present. Granule cells have migrated through the molecular layer but no longer reside there (magnication 100). (C) Cerebellar cortex from an adult rat infected as a neonate with LCMV. Many granule cells (arrows) remain abnormally placed within the molecular layer. As a result of infection with LCMV, these ectopic neurons have failed to migrate properly to their normal location within the granule cell layer (magnication 100).
1028
by another mosquito. Thus, although people may become infected and ill, they are a dead-end host for the virus. Other mammals, including horses, may also develop symptomatic infections from the bite of an infected mosquito, but they are also dead-end hosts [99]. In the summer of 1999, all human WNV cases were localized to the metropolitan area of New York City [101]. Since then, the infection has been diagnosed in residents of many other states along the Atlantic seaboard. Avian surveillance has indicated a substantial northward and southward expansion of epizootic activity, a pattern consistent with viral spread by migrating birds. Furthermore, WNV has been detected in an increasing number of mosquito species, some of which more readily feed on human beings and other mammals than do Culex mosquitos. These epidemiologic ndings indicate amplication of the virus in the United States and suggest that the importance of WNV as a cause of disease in North America is likely to continue to increase [102]. Most WNV infections in human beings are clinically silent, with 120 to 160 asymptomatic infections for every overt infection [103,104]. Among those individuals who do develop disease, many have only mild signs and symptoms, including fever, malaise, periocular pain, lymphadenopathy, and myalgia. A ne maculopapular rash may be present and occurs more often in children than in adults [105]. Unfortunately, some individuals infected with WNV become critically ill, virtually always with neurologic signs and symptoms. Of patients hospitalized with WNV infection, approximately two thirds have encephalitis and approximately one third have meningitis without encephalitis [101]. The signs and symptoms in WNV encephalitisfever, nausea, vomiting, headache, altered mental status and sti neckgenerally mimic those observed in meningitis or encephalitis caused by other pathogens. By contrast, encephalitis caused by WNV only rarely manifests with seizures but commonly includes cranial nerve dysfunction. These ndings suggest that the brain stem may be a principal target of WNV. Many patients with WNV develop an erythematous rash, an unusual feature in other forms of viral encephalitis. Many patients with WNV develop patterns of weakness that are unusual in other forms of encephalitis or meningitis. This weakness may manifest as pronounced proximal muscle weakness or as accid paralysis of the extremities. Many patients with weakness have evidence of transverse myelitis or Guillain-Barre syndrome [106]. Among patients hospitalized with WNV, the infection has a mortality rate of 10% to 15%. Patients with encephalitis and weakness have a mortality rate of 30%. The risk of death from WNV infection is much greater in the elderly than in children or young adults. The CSF ndings of WNV encephalitis typically include a pleocytosis and a moderately elevated protein concentration. CT of the brain typically shows no abnormalities, whereas MRI may show enhancement of the meninges and periventricular regions [101]. Among patients with weakness, electrophysiologic testing often reveals evidence of either an axonal
1029
polyneuropathy, demyelinating polyneuropathy, or both. Decreased nerve conduction velocity and diminished compound muscle action potentials and brillation potentials are common ndings on electromyography of weak patients infected with WNV [101,107]. The diagnostic test of choice is the WNV ELISA. This assay, the most sensitive of the available tests, yields useful results from blood or CSF. A problem with serologic diagnosis of WNV is the potential for cross-reactivity among the aviviruses. False-positive results for WNV may arise when the infecting virus is St. Louis encephalitis or dengue fever or when the patient has been immunized with yellow fever or Japanese encephalitis vaccines. If positive, the ELISA results should be veried by the more specic plaque reduction neutralization test [108]. Separate ELISA tests can be performed for IgG and IgM antibodies [109]. Another potential problem is that IgM antibodies for WNV can persist for up to 1 year, thus causing confusion as to whether the antibodies reect an acute or prior infection. Infection with WNV can also be demonstrated by isolating the virus in cell culture. Molecular techniques using real-time PCR have recently been developed as an assay for WNV [110]. The advantage of real-time PCR lies in the rapidity with which an etiologic diagnosis can be made. An important disadvantage of this technique, however, lies in its relatively low sensitivity [111]. If specic therapy for WNV encephalitis becomes available, there should be an increased impetus for sensitive molecular diagnosis. Because no specic eective therapy exists for WNV encephalitis, treatment is supportive. Ribavirin inhibits WNV replication and the cytopathic eect of the virus on cultured neural cells but has not been studied in controlled trials [112]. Because the bite of a WNV-infected mosquito is a crucial step in the development of WNV encephalitis, personal and community protective measures can be taken to reduce the risk of infection by reducing the risk of mosquito bites. One community measure may include surveillance to monitor the prevalence of infection in mosquitoes, birds, and horses [113]. When human-biting mosquitoes are infected and present in sucient numbers, the mosquito populations may be eectively reduced, under proper conditions, by the application of insecticides [114]. Individuals can reduce their own risks by minimizing exposure to mosquito vectors, by covering exposed skin with repellants, and by eliminating mosquito breeding grounds by removing containers capable of holding water. Human herpesvirus-6 infection, febrile seizures, and encephalitis Another recent important development has been the identication of human herpesvirus-6 (HHV-6) and the elucidation of its role in human disease. HHV-6, a member of the Herpesviridae family of viruses, was discovered in 1986. Although biologically similar to cytomegalovirus (CMV), the virus was recognized as serologically and genetically distinct from other human herpes viruses [115]. Soon thereafter, HHV-6 was noted to be
1030
ubiquitous and a principal cause of the common childhood disease exanthem subitum (also referred to as roseola or sixth disease) [116]. From the neurologic perspective, HHV-6 is of importance because of its linkage to febrile seizures and encephalitis. Exanthem subitum typically occurs between 6 months and 2 years of age and is characterized by an abrupt rise in temperature, often to approximately 40C, followed several days later by a rapid defervescence that coincides with the emergence of an erythematous maculopapular rash that can persist for several days. Fever is a prominent component of the disease, and it has been recognized for many years that febrile seizures occur commonly during the early (febrile) stage of exanthem subitum. The linkage of exanthem subitum to febrile seizures, along with the identication of HHV-6 as the causative pathogen for exanthem subitum, led naturally to the hypothesis that HHV-6 infection may underlie many cases of febrile seizures. This has proven to be true. Multiple studies have demonstrated that a substantial proportion of febrile seizures occur in children with a primary HHV-6 infection [117119]. The risk of febrile seizures is high during primary infection with HHV-6, whether or not the child develops the rash of exanthem subitum. In addition to increasing the risk of febrile seizures, a primary infection with HHV-6 may increase the severity of the febrile convulsion. Partial seizures, prolonged seizures, and repeated seizures are more common when the febrile seizures are associated with HHV-6 infection than when the seizures are caused by other pathogens [119]. Furthermore, reports have suggested that HHV-6 may persist within the CNS [120] and that reactivation of HHV-6 may be associated with recurrent febrile seizures [118]. This remains controversial, however, and at least one study has found that children with an initial febrile seizure induced by HHV-6 infection have a lower risk of febrile seizure recurrence than do children whose febrile seizures were triggered by other pathogens [121]. Although the association of HHV-6 infection with febrile seizures is well documented, the mechanism underlying this association remains unclear. One possibility is that HHV-6 directly invades brain parenchyma during a primary infection and that the febrile convulsion is triggered by brain infection. This hypothesis is supported by the ndings that HHV-6 is a neurotropic virus that can replicate in glial cells and in brain endothelial cells [122125]. Identication of HHV-6 DNA in CSF of patients with febrile seizures further supports the notion that the brain is invaded during HHV-6 infection [118,123]. Other studies have not found HHV-6 DNA in the CSF of patients with febrile seizures [126,127], however, and the question of brain invasion in typical cases of HHV-6 infection remains controversial. A second possibility is that the seizures are triggered not by direct viral invasion of the CNS but by the production of a toxin or inammatory molecule liberated as part of the viral infection. Cytokines, for example, are
1031
produced at high levels by circulating monocytes during HHV-6 infection and can act on neurons to lower the seizure threshold [128130]. A third possibility is that the febrile convulsions are not directly related to the HHV-6 infection per se but to a characteristic of the fever induced by the infection. Febrile seizures occur almost exclusively in children, presumably because the immature brain has a lower seizure threshold to the destabilizing eects of temperature elevation. A hallmark of exanthem subitum is a rapid onset of high fever. Thus, children infected with HHV-6 may be at greater risk for seizures than children with other infections because of the rapidity of high fever onset with HHV-6. Arguing against this possibility, however, are the ndings that neither the maximal temperature nor the duration of fever before the seizure diered between febrile children who developed seizures and those who did not [117,131]. In a few patients, HHV-6 produces encephalitis. Encephalitis caused by HHV-6 has been observed both in immunocompromised patients and in children whose immune function was apparently normal [132,133]. As might have been anticipated from the association of HHV-6 with convulsions, seizures, and status epilepticus are common presenting signs of HHV-6 encephalitis [119,134]. Confusion and headache, common symptoms of encephalitis in general, are also commonly observed in encephalitis caused by HHV-6. When a patient develops encephalitis in association with exanthem subitum, the signs of encephalitis usually arise during the febrile phase of the illness before the onset of rash [119,134]. Whether invasion of the brain by HHV-6 is necessary to produce encephalitis is unresolved. In some cases of HHV-6 encephalitis, viral antigen has been detected within brain astrocytes and neurons [132], the location of viral antigen has been correlated with the sites of neuropathologic changes [135], and HHV-6 DNA has been detected in CSF. These ndings strongly suggest that viral invasion occurs and that the presence HHV-6 within the brain plays a key role in the pathophysiology. Nevertheless, other studies have noted the distinct absence of viral antigen from the brain [136] and of viral DNA from the CSF [134] in cases of HHV-6 encephalitis. These results suggest that in at least some cases of HHV-6 encephalitis, the brain dysfunction may be caused by an extrainfectious process. Most children who develop HHV-6 encephalitis have a good outcome with no neurologic sequelae. Some children are left with serious permanent neurologic problems, however, including mental retardation, hemiplegia, and epilepsy [137139]. Occasional cases of HHV-6 encephalitis have been fatal [125,132,136].
Acknowledgments The authors thank Dr William Bell and Dr Christine Odell for their careful review of the manuscript and their helpful suggestions.
1032
References
[1] Coyle PK. Overview of acute and chronic meningitis. Neurol Clin North Am 1999;17: 691710. [2] McNamara JO, Whitney KD, Andrews PI, et al. Evidence for glutamate receptor autoimmunity in the pathogenesis of Rasmussen encephalitis. Adv Neurol 1999;79:54350. [3] Quagliarello VJ, Scheld WM. Bacterial meningitis: pathogenesis, pathophysiology, and progress. N Engl J Med 1992;327:86472. [4] Tunkel AR, Wispelwey B, Scheld WM. Pathogenesis and pathophysiology of meningitis. Infect Dis Clin North Am 1990;4:55581. [5] Parkkinen J, Korhonen T, Pere A, et al. Binding sites in the rat brain for Escherichia coli S mbriae associated with neonatal meningitis. J Clin Invest 1988;81:8605. [6] Simberko MS, Moldover NH, Rahal J. Absence of detectable bactericidal and opsonic activities in normal and infected human cerebrospinal uid. A regional host deciency. J Lab Clin Med 1980;95:36272. [7] Zwahlen A, Nydegger UE, Vaudaux P, et al. Complement-mediated opsonic activity in normal and infected human cerebrospinal uid: early response during bacterial meningitis. J Infect Dis 1982;145:63546. [8] Quagliarello VJ, Wispelway B, Long WJ, et al. Recombinant human interleukin-1 induces meningitis and blood brain barrier injury in the rat. Characterization and comparison with tumor necrosis factor. J Clin Invest 1991;87:13606. [9] Ramilo O, Saez-Llorens X, Mertsola J, et al. Tumor necrosis factor alpha/cachectin and interleukin 1 beta initiate meningeal inammation. J Exp Med 1990;172:497507. [10] Bogdan I, Leib SL, Bergeron M, et al. Tumor necrosis factor-alpha contributes to apoptosis in hippocampal neurons during experimental group B streptococcal meningitis. J Infect Dis 1997;176:6937. [11] Dawson VL, Dawson TM. Nitric oxide in neuronal degeneration. Proc Soc Exp Biol Med 1996;211:3340. [12] Kornecki E, Ehrlich YH. Neuroregulatory and neuropathological actions of the etherphospholipid platelet-activating factor. Science 1988;240:17924. [13] Nathan C. Inducible nitric oxide synthase: what dierence does it make? J Clin Invest 1997;100:241723. [14] Pster HW, Scheld WM. Brain injury in bacterial meningitis: therapeutic implications. Curr Opin Neurol 1997;10:2549. [15] Dinarello CA. The proinammatory cytokines interleukin-1 and tumor necrosis factor and treatment of the septic shock syndrome. J Infect Dis 1991;163:117784. [16] Igarashi M, Gilmartin RC, Gerald B, et al. Cerebral arteritis and bacterial meningitis. Arch Neurol 1984;41:5315. [17] Tureen JH, Dworkin RJ, Kennedy SL, et al. Loss of cerebrovascular autoregulation in experimental meningitis in rabbits. J Clin Invest 1990;85:57781. [18] Feigin RD, Pearlman E. Bacterial meningitis beyond the neonatal period. In: Feigin RD, Cherry JD, editors. Textbook of pediatric infectious diseases. 4th edition. Philadelphia: WB Saunders; 1998. p. 40029. [19] Dinarello CA, Mier JW. Lymphokines. N Engl J Med 1987;317:9405. [20] Niemoller UM, Tauber MG. Brain edema and increased intracranial pressure in the pathophysiology of bacterial meningitis. Eur J Clin Microbiol Infect Dis 1989;8:10917. [21] Quagliarello VJ, Long WJ, Scheld WM. Morphologic alterations of the blood brain barrier with experimental meningitis in the rat: temporal sequence and role of encapsulation. J Clin Invest 1986;77:108495. [22] Scheld WM, Dacey RG, Winn HR, et al. Cerebrospinal uid outow resistance in rabbits with experimental meningitis. Alterations with penicillin and methylprednisolone. J Clin Invest 1980;66:24353.
1033
[23] Ashwal S, Tomasi L, Schneider S, Perkin R, Thompson J. Bacterial meningitis in children: pathophysiology and treatment. Neurology 1992;42:73948. [24] Cabral DA, Flodmark O, Farrell K, Speert DP. Prospective study of computed tomography in acute bacterial meningitis. J Pediatr 1987;111:2015. [25] Goiten KJ, Tamir I. Cerebral perfusion pressure in central nervous system infections in infancy and childhood. J Pediatr 1983;103:403. [26] Tauber MG. Brain edema, intracranial pressure and cerebral blood ow in bacterial meningitis. Pediatr Infect Dis J 1989;8:9157. [27] Dodge PR, Swartz MN. Bacterial meningitis: a review of selected aspects. II. Special neurologic problems, post meningitis complications and clinicopathological correlations. N Engl J Med 1965;272:95460. [28] Gaussorgues P, Guerin C, Boyer F, et al. Intracranial hypertension in comatose bacterial meningitis. Presse Med 1987;16:14203. [29] Klein JO, Feigen RD, McCracken GH, Jr. Report of the task force on diagnosis and management of meningitis. Pediatrics 1986;78(Suppl):95982. [30] Sell SH. Long term sequelae of bacterial meningitis in children. Pediatr Infect Dis J 1983;2:903. [31] Kadurugamuwa JL, Hengstler B, Zak O. Eects of anti-inammatory drugs on arachidonic acid metabolites and cerebrospinal uid proteins during infectious pneumococcal meningitis in rabbits. Pediatr Infect Dis J 1987;6:11534. [32] Mertsolla J, Ramilo O, Mustafa MM, et al. Release of endotoxin after antibiotic treatment of gram negative bacterial meningitis. Pediatr Infect Dis J 1989;8:9046. [33] Syrogiannopoulos GA, Olsen K, Reisch JS, McCracken GH, Jr. Dexamethasone in the treatment of experimental Haemophilus inuenzae type B meningitis. J Infect Dis 1987; 155:2139. [34] Tauber MG, Khayam-Bashi H, Sande MA. Eects of ampicillin and corticosteroids on brain water content, cerebrospinal uid pressure and cerebrospinal uid lactate levels in experimental pneumococcal meningitis. J Infect Dis 1985;151:52834. [35] Lebel MH, Hoyt J, Waagner DC, et al. Magnetic resonance imaging and dexamethasone therapy for bacterial meningitis. Am J Dis Child 1989;143:3016. [36] McIntyre PB, Berkey CS, King SM, et al. Dexamethasone as adjunctive therapy in bacterial meningitis. A meta-analysis of randomized clinical trials since 1988. JAMA 1997;278:92531. [37] Mustafa MM, Ramilo O, Saez-Llorens X, et al. Cerebrospinal uid prostaglandins, interleukin 1 beta, and tumor necrosis factor in bacterial meningitis. Clinical and laboratory correlations in placebo-treated and dexamethasone-treated patients. Am J Dis Child 1990;144:8837. [38] Odio CM, Faingezicht I, Paris M, et al. The benecial eects of early dexamethasone administration in infants and children with bacterial meningitis. N Engl J Med 1991; 324:152531. [39] Lebel MH, Freji BJ, Syrogiannopoulous GA, et al. Dexamethasone therapy for bacterial meningitis: results of two double-blind, placebo controlled trials. N Engl J Med 1988; 319:96471. [40] Girgis NI, Farid Z, Mikhail IA, et al. Dexamethasone treatment for bacterial meningitis in children and adults. Pediatr Infect Dis J 1989;8:84851. [41] Kaplan SL. Dexamethasone for children with bacterial meningitis: should it be routine therapy? Am J Dis Child 1989;143:2902. [42] Cabellos C, Martinez-Lacasa J, Martos A, et al. Inuence of dexamethasone on ecacy of ceftriaxone and vancomycin therapy in experimental pneumococcal meningitis. Antimicrob Agents Chemother 1995;39:215860. [43] Pacheco TR, Cooper CK, Hardy DJ, et al. Failure of cefotaxime treatment in an adult with Streptococcus pneumoniae meningitis. Am J Med 1997;102:3035.
1034
[44] Paris MM, Hickey SM, Uscher MI, et al. Eect of dexamethasone on therapy of experimental penicillin- and cephalosporin-resistant pneumococcal meningitis. Antimicrob Agents Chemother 1994;38:13204. [45] American Academy of Pediatrics Committee on Infectious Diseases. Haemophilus inuenzae infections. In: Pickering LK, editor. 2000 Red book. 25th edition. Elk Grove Village, IL: American Academy of Pediatrics; 2000. p. 26272. [46] American Academy of Pediatrics Committee on Infectious Diseases. Pneumococcal infections. In: Pickering LK, editor. 2000 Red book. 25th edition. Elk Grove Village, IL: American Academy of Pediatrics; 2000. p. 45260. [47] Canadian Paediatric Society. Infectious Diseases and Immunization Committee: dexamethasone therapy for bacterial meningitis: a time for caution. CMAJ 1990;142:3078. [48] Finch RG, Mandragos C. Corticosteroids in bacterial meningitis: not yet justied for all patients. BMJ 1991;302:6078. [49] Grimwood K. Dexamethasone treatment for bacterial meningitis. J Paediatr Child Health 1990;26:3113. [50] American Academy of Pediatrics Committee on Infectious Diseases. Recommendations for the prevention of pneumococcal infections, including the use of pneumococcal conjugate vaccine (Prevnar), pneumococcal polysaccharide vaccine, and antibiotic prophylaxis. Pediatrics 2000;106:3626. [51] Mustafa MM, Ramilo O, Mertsola J, et al. Modulation of inammation and cachectin activity in relation to treatment of experimental Haemophilus inuenzae type b meningitis. J Infect Dis 1989;160:81825. [52] Cochi SL, Broome CV, Hightower AW. Immunization of US children with Haemophilus inuenzae type b polysaccharide vaccine: a cost-eectiveness model of strategy assessment. JAMA 1985;253:5219. [53] Wenger JD, Hightower AW, Facklam RR, et al. Bacterial meningitis in the United States, 1986: report of a multistate surveillance study. J Infect Dis 1990;162:131623. [54] Schuchat A, Robinson K, Wenger JD, et al. Bacterial meningitis in the United States in 1995. N Engl J Med 1997;337:9706. [55] Shattuck KE, Chonmaitree T. The changing spectrum of neonatal meningitis over a fteen year period. Clin Pediatr 1992;31:1306. [56] Smith AL, Haas J. Neonatal bacterial meningitis. In: Scheld WM, Whitley RJ, Durack DT, editors. Infections of the central nervous system. New York: Raven Press; 1991. p. 31333. [57] Friedland IR, McCracken GH, Jr. Management of infections caused by antibioticresistant Streptococcus pneumoniae. N Engl J Med 1994;331:37782. [58] Doern GV, Heilmann KP, Huynh HK, et al. Antimicrobial resistance among clinical isolates of Streptococcus pneumoniae in the United States during 19992000, including a comparison of resistance rates since 19941995. Antimicrob Agents Chemother 2001;45: 17219. [59] Caputo GM, Appelbaum PC, Liu HH. Infections due to penicillin-resistant pneumococci. Clinical, epidemiologic and microbiologic features. Arch Intern Med 1993;153: 130110. [60] Ferroni A, Berche P. Alterations to penicillin-binding proteins 1A, 2B and 2X amongst penicillin-resistant clinical isolates of Streptococcus pneumoniae serotype 23F from the nasopharyngeal ora of children. J Med Microbiol 2001;50:82832. [61] Reichler MR, Allphin AA, Breiman RF, et al. The spread of multiply resistant Streptococcus pneumoniae at a day care center in Ohio. J Infect Dis 1992;166:134653. [62] Gray BM, Dillon HC, Jr. Clinical and epidemiologic studies of pneumococcal infection in children. Pediatr Infect Dis J 1986;5:2017. [63] Friedland IR, Paris M, Ehrett S, et al. Evaluation of antimicrobial regimens for treatment of experimental penicillin- and cephalosporin-resistant pneumococcal meningitis. Antimicrob Agents Chemother 1993;37:16306.
1035
[64] Viladrich PF, Gudiol F, Linares J, et al. Evaluation of vancomycin for therapy of adult pneumococcal meningitis. Antimicrob Agents Chemother 1991;35:246772. [65] Bradley JS, Scheld WM. The challenge of penicillin-resistant Streptococcus pneumoniae meningitis: current antibiotic therapy in the 1990s. Clin Infect Dis 1997;24(Suppl 2): S21321. [66] Spach DH, Jackson LA. Bacterial meningitis. Neurol Clin North Am 1999;17:71135. [67] Butler JC, Hofmann J, Cetron MS, et al. The continued emergence of drug-resistant Streptococcus pneumoniae in the United States: an update from the Centers for Disease Control and Preventions Pneumococcal Sentinel Surveillance System. J Infect Dis 1996;174:98693. [68] Black S, Shineeld H, Fireman B, et al. Ecacy, safety and immunogenicity of heptavalent pneumococcal conjugate vaccine in children. Pediatr Infect Dis J 2000;19: 18795. [69] Centers for Disease Control and Prevention. Decreased availability of pneumococcal conjugate vaccine. MMWR 2001;50:7834. [70] Armstrong C, Lillie RD. Experimental lymphocytic choriomeningitis of monkeys and mice produced by a virus encountered in studies of the 1933 St. Louis encephalitis epidemic. Public Health Rep 1934;49:101922. [71] Buchmeier MJ, Zajac AJ. Lymphocytic choriomeningitis virus. In: Ahmed R, Chen I, editors. Persistent viral infections. New York: Wiley; 1999. p. 575605. [72] Ambrosio AM, Feuillade MR, Gamboa GS, et al. Prevalence of lymphocytic choriomeningitis virus infection in a human population of Argentina. Am J Trop Med 1994; 50:3816. [73] Childs JE, Glass GE, Korch GW, et al. Lymphocytic choriomeningitis virus infection and house mouse (Mus musculus) distribution in urban Baltimore. Am J Trop Med Hyg 1992; 47:2734. [74] Lehman-Grube F. Portraits of viruses: arenaviruses. Intervirology 1984;22:12145. [75] Stephensen CB, Blount SR, Lanford RE, et al. Prevalence of serum antibodies against lymphocytic choriomeningitis virus in selected populations from two U.S. cities. J Med Virol 1992;38:2731. [76] Komrower GM, Williams BL, Stones PB. Lymphocytic choriomeningitis in the newborn. Probable transplacental infection. Lancet 1955;1:6978. [77] Barton LL, Budd SC, Mortt WS, et al. Congenital lymphocytic choriomeningitis virus infection in twins. Pediatr Infect Dis J 1993;12:9426. [78] Barton LL, Mets MB. Congenital lymphocytic choriomeningitis virus infection: decade of rediscovery. Clin Infect Dis 2001;33:3704. [79] Enders G, Varho-Gobel M, Lohler J, et al. Congenital lymphocytic choriomeningitis virus infection: an underdiagnosed disease. Pediatr Infect Dis J 1999;18:6525. [80] Wright R, Johnson D, Neumann M, et al. Congenital lymphocytic choriomeningitis virus syndrome: a disease that mimics congenital toxoplasmosis or cytomegalovirus infection. Pediatrics 1997;100:e9 (Electronic publication). [81] Jahrling PB, Peters CJ. Lymphocytic choriomeningitis virus: a neglected pathogen of man. Arch Pathol Lab Med 1992;116:4868. [82] Meyer HM, Jr, Johnson RT, Crawford IP, et al. Central nervous system syndromes of viral etiology: a study of 713 cases. Am J Med 1960;29:33447. [83] Chesney PJ, Katcher ML, Nelson DB, et al. CSF eosinophilia and chronic lymphocytic choriomeningitis virus meningitis. J Pediatr 1979;94:7502. [84] Biggar R, Woodall J, Walter P, et al. Lymphocytic choriomeningitis outbreak associated with pet hamsters: fty seven cases from New York state. JAMA 1975;232:494500. [85] Barton LL, Peters CJ, Ksiazek TG. Lymphocytic choriomeningitis virus: an unrecognized teratogenic pathogen. Emerg Infect Dis 1995;1:1523. [86] Mets MB, Barton LL, Khan AS, et al. Lymphocytic choriomeningitis virus: an underdiagnosed cause of congenital chorioretinitis. Am J Ophthalmol 2000;130:20915.
1036
[87] Brezin AP, Thulliez P, Cisneros B, et al. Lymphocytic choriomeningitis virus chorioretinitis mimicking ocular toxoplasmosis in two otherwise normal children. Am J Ophthalmol 2000;130:2457. [88] Bonthius DJ, Mahoney J. LCM virus infection of the developing rat brain: acute and delayed-onset pathology [abstract]. Soc Neurosci Abstr 2001;27:658.8. [89] Bonthius DJ, Mahoney J, Buchmeier M, et al. Critical role for glial cells in the propagation and spread of lymphocytic choriomeningitis virus in the developing rat brain. J Virol 2002;76:661836. [90] del Cerro M, Grover D, Monjan AA, et al. Congenital retinitis in the rat following maternal exposure to lymphocytic choriomeningitis virus. Exp Eye Res 1984;38:31324. [91] Monjan AA, Gilden DH, Cole GA, et al. Cerebellar hypoplasia in neonatal rats caused by lymphocytic choriomeningitis virus. Science 1971;171:1946. [92] Pearce BD, Steensen SC, Paoletti AD, et al. Persistent dentate granule cell hyperexcitability after neonatal infection with lymphocytic choriomeningitis virus. J Neurosci 1996;16:2208. [93] Bonthius DJ, Mahoney J, Van Hoesen GW, Perlman S. Cerebellar granule cell death induced by LCM virus in neonatal rats is lobule-dependent and apoptotic [abstract]. Soc Neurosci Abstr 1999;25:486. [94] Monjan AA, Bohl LS, Hudgens GA. Neurobiology of LCM virus infection in rodents. Bull World Health Org 1975;52:48792. [95] Levison SW, Goldman JE. Both oligodendrocytes and astrocytes develop from progenitors in the subventricular zone of postnatal rat forebrain. Neuron 1993;10:20112. [96] Luskin MB. Restricted proliferation and migration of postnatally generated neurons derived from the forebrain subventricular zone. Neuron 1993;11:17389. [97] Monjan AA, Cole GA, Nathanson N. Pathogenesis of LCM disease in the rat. In: Lehmann-Grube F, editor. Lymphocytic choriomeningitis virus and other arena viruses. New York: Springer-Verlag; 1973. p. 195206. [98] Centers for Disease Control and Prevention. Outbreak of West Nile-like viral encephalitisNew York, 1999. Morb Mortal Wkly Rep 1999;48:8459. [99] Hayes CG. West Nile fever. In: Monath TP, editor. The arboviruses: epidemiology and ecology, vol. 5. Boca Raton, FL: CRC Press; 1988. p. 5988. [100] Lanciotti RS, Roehrig JT, Deubel V, et al. Origin of the West Nile Virus responsible for an outbreak of encephalitis in the northeastern United States. Science 1999;286:23337. [101] Nash D, Mostashari F, Fine A, et al. The outbreak of West Nile virus infection in the New York City area in 1999. N Engl J Med 2001;344:180714. [102] Deubel V, Gubler DJ, Layton M, Malkinson M. West Nile virus: a newly emergent epidemic disease. Emerg Infect Dis 2001;7:536. [103] Centers for Disease Control and Prevention. Update: surveillance for West Nile virus in overwintering mosquitoesNew York, 2000. MMWR CDC Surveill Summ 2000;49:1789. [104] Centers for Disease Control and Prevention. Serosurveys for West Nile virus infection New York and Connecticut counties. MMWR CDC Surveill Summ 2000;50:379. [105] Marburg K, Goldblum H, Sterk VV, et al. The natural history of West Nile fever: clinical observations during an epidemic in Israel. Am J Hyg 1956;64:25963. [106] Craven RB, Roehrig JT. West Nile virus. JAMA 2001;286:6513. [107] Tyler KL. West Nile virus encephalitis in America. N Engl J Med 2001;344:18589. [108] Beaty BJ, Calisher CH, Shope RE. Arboviruses. In: Schmidt NJ, Emmons RW, editors. Diagnostic procedures for viral, rickettsial and chlamydial infections. 6th edition. Washington, DC: American Public Health Association; 1989. p. 797855. [109] Tardei G, Ruta S, Chitu V, et al. Evaluation of immunoglobulin M (IgM) and IgG enzyme immunoassays in serologic diagnosis of West Nile virus infection. J Clin Microbiol 2000;38:22329. [110] Briese T, Glass WG, Lipkin WI. Detection of West Nile virus sequences in cerebrospinal uid. Lancet 2000;355:16145.
1037
[111] Fine A, Mostashari F, Nash D, Layton M. Testing for West Nile virus. Lancet 2000; 356:111011. [112] Jordan I, Briese T, Fischer N, et al. Ribavirin inhibits West Nile virus replication and cytopathic eect in neural cells. J Infect Dis 2000;182:12147. [113] Centers for Disease Control and Prevention . Guidelines for arbovirus surveillance in the United States. Fort Collins, CO: US Department of Health and Human Services, Centers for Disease Control and Prevention; 1993. [114] Mount GA, Biery TL, Haile DG. A review of ultralow-volume aerial sprays of insecticide for mosquito control. J Am Mosq Control Assoc 1996;12:60118. [115] Salahuddin SZ, Ablashi DV, Markham PD, et al. Isolation of a new virus, HBLV, in patients with lymphoproliferative disorders. Science 1986;234:596601. [116] Yamanishi K, Okuno T, Shiraki K, et al. Identication of human herpesvirus-6 as a causal agent for exanthem subitum. Lancet 1988;8594:10657. [117] Barone SR, Kaplan MH, Krilov LR. Human herpesvirus-6 infection in children with rst febrile seizures. J Pediatr 1995;127:957. [118] Kondo K, Nagafuji H, Hata A, et al. Association of human herpesvirus-6 infection of the central nervous system with recurrence of febrile convulsions. J Infect Dis 1993;167:1197200. [119] Suga S, Suzuki K, Ihira M, et al. Clinical characteristics of febrile convulsions during primary HHV-6 infection. Arch Dis Child 2000;82:626. [120] Luppi M, Barozzi P, Maiorana A, et al. Human herpesvirus-6 infection in normal human brain tissue. J Infect Dis 1994;169:9434. [121] Jee SH, Long CE, Schnabel KC, et al. Risk of recurrent seizures after a primary human herpesvirus 6-induced febrile seizure. Pediatr Infect Dis J 1998;17:438. [122] Ablashi DV, Lusso P, Hung CL, et al. Utilization of human hematopoietic cell lines for the propagation and characterization of HBLV (human herpesvirus 6). Int J Cancer 1988;42:78791. [123] Caserta MT, Hall CB, Schnabel K, et al. Neuroinvasion and persistence of human herpes virus 6 in children. J Infect Dis 1994;170:15869. [124] He J, McMarthy M, Zhou Y, et al. Infection of primary human fetal astrocytes by human herpesvirus 6. J Virol 1996;70:1296300. [125] Ueda T, Miyake Y, Imoto K, et al. Distribution of human herpesvirus 6 and varicellazoster virus in organs of a fatal case with exanthem subitum and varicella. Acta Paediatr Jpn 1996;38:5905. [126] Hukin J, Farrell K, MacWilliams LM, et al. Case control study of primary human herpesvirus 6 infection in children with febrile seizures. Pediatrics 1998;101:e3 (Electronic publication). [127] Teach SJ, Wallace HW, Evans MJ, et al. Human herpesviruses types 6 and 7 and febrile seizures. Pediatr Neurol 1999;21:699703. [128] Flamand L, Gosselin J, DAddario M, et al. Human herpesvirus 6 induces interleukin-1b and tumor necrosis factor alpha, but not interleukin-6, in peripheral blood mononuclear cell cultures. J Virol 1991;65:510510. [129] Jankowski JL, Patterson PH. The role of cytokines and growth factors in seizures and their sequelae. Prog Neurobiol 2001;63:12549. [130] Kikuta H, Nakane A, Lu H, et al. Interferon induction by human herpesvirus-6 in human mononuclear cells. J Infect Dis 1990;162:358. [131] Bertolani MF, Portolani M, Marotti F, et al. A study of childhood febrile convulsions with particular reference to HHV-6 infection: pathogenic considerations. Childs Nerv Syst 1996;12:5349. [132] Drobyski WR, Knox KK, Majewski D, Carrigan DR. Fatal encephalitis due to variant B human herpesvirus-6 infection in a bone marrow-transplant recipient. N Engl J Med 1994;330:135660. [133] Yoshikawa T, Asano Y. Central nervous system complications in human herpesvirus-6 infection. Brain Dev 2000;22:30714.
1038
[134] Jones CM, Dunn HG, Thomas EE, et al. Acute encephalopathy and status epilepticus associated with human herpes virus 6 infection. Dev Med Child Neurol 1994;36:64650. [135] Carrigan DR, Harrington D, Knox KK. Subacute leukoencephalitis caused by CNS infection with human herpesvirus-6 manifesting as acute multiple sclerosis. Neurology 1996;47:1458. [136] Asano Y, Yoshikawa T, Kajita Y, et al. Fatal encephalitis/encephalopathy in human herpesvirus-6 infection. Arch Dis Child 1992;67:14845. [137] Hill AE, Hicks EM, Coyle PV. Human herpesvirus-6 and central nervous system complications. Dev Med Child Neurol 1994;36:6512. [138] Oki J, Yoshida H, Tokumitsu A, et al. Serial neuroimages of acute necrotizing encephalopathy associated with human herpesvirus 6 infection. Brain Dev 1995;17:3569. [139] Yanagihara K, Tanaka-Taya K, Itagaki Y, et al. Human herpesvirus 6 meningoencephalitis with sequelae. Pediatr Infect Dis J 1995;14:2402. [140] American Academy of Pediatrics Committee on Infectious Diseases. Dexamethasone therapy for bacterial meningitis in infants and children. Pediatrics 1990;86:1303. [141] Larsen PD, Wright R. Early clinical manifestations and long-term outcome in symptomatic congenital lymphocytic choriomeningitis virus infection [abstract]. Neurology 2001;56:A3940.
Neurol Clin N Am 20 (2002) 10391060
Congenital infections
James F. Bale, Jr, MD*
Division of Pediatric Neurology, The University of Utah School of Medicine and Primary Childrens Medical Center, Salt Lake City, UT, USA
Women encounter many infectious pathogens during their pregnancies, but few of these pathogens possess the ability to infect the placenta and damage the developing fetus. Nonetheless, these pathogens, grouped traditionally as toxoplasmosis, others, rubella, cytomegalovirus (CMV), and herpes (TORCH) infections [1], remain important causes of deafness, vision loss, and behavioral or neurologic disorders among children throughout the world. A list of infectious agents potentially associated with intrauterine central nervous system (CNS) infections is displayed in the box. Although the introduction of the rubella vaccine in the late 1960s led to a dramatic reduction in the incidence of the congenital rubella syndrome (CRS), the past decade has seen only modest changes in the epidemiology of congenital infections. Immunization programs are likely to reduce the risk of congenital varicella syndrome, but there are currently no eective strategies to prevent intrauterine CMV infection or congenital toxoplasmosis, which are important causes of permanent neurologic disability among children worldwide. This article describes the major pathogens causing intrauterine infections and summarizes the current status of treatment and prevention. Cytomegalovirus Epidemiology In the United States, 0.4% to 2.5% of newborns shed CMV at birth, and most human beings become infected with the virus thereafter, usually without recognizable symptoms [24]. Acutely infected persons have viremia that lasts up to 3 months, and CMV is then shed in their urine, saliva, semen (if
* Correspondence. Division of Pediatric Neurology, Suite 2700, Primary Childrens Medical Center, Suite 2700, 100 North Medical Drive, Salt Lake City, UT 84113, USA. E-mail address: pcjbale@ihc.com (J.F. Bale). 0733-8619/02/$ - see front matter 2002, Elsevier Science (USA). All rights reserved. PII: S 0 7 3 3 - 8 6 1 9 ( 0 2 ) 0 0 0 1 4 - 2
1040
J.F. Bale / Neurol Clin N Am 20 (2002) 10391060
Infectious agents potentially associated with intrauterine central nervous system infections Viruses Rubella Cytomegalovirus Herpes simplex virus type 2 Varicella zoster virus Lymphocytic choriomeningitis virus Western equine encephalitis virus Protozoa Toxoplasma gondii Plasmodium sp. Trypanosoma cruzi Other Treponema pallidum
male), and cervical uids (if female) for a year or more. Seropositive breastfeeding women shed infectious virus in breast milk within 1 month of lactation as a consequence of reactivated CMV infection [2,4]. Congenitally infected infants shed CMV in urine for several years [2]. The incidence of acquired CMV infection among children and adults, including women of childbearing age, averages 1% to 2% annually. By adulthood, 40% to 100% of the population has been infected with CMV [24]. Infants acquire CMV from their breast-feeding mothers, and young children become infected through direct contact with CMV-excreting playmates. Group child care in the United States plays a major role in the transmission of CMV among young children, their parents, and their care providers [58]. Adults acquire CMV via blood transfusion, sexual contact, organ transplantation, or contact with young children. Adults with multiple sexual partners have increased rates of CMV acquisition [4]. Seronegative women have the greatest risk of transmitting CMV to their unborn infants. Approximately 40% of women who acquire primary CMV infection during pregnancy transmit CMV to their fetuses, and approximately 10% of the infected infants have CMV disease [9]. Reinfection with new CMV strains occurs, particularly among young children in child care environments [10]. CMV-seropositive women can also be reinfected with new CMV strains, and these new strains can cause disease in their ospring [11]. Clinical manifestations Most acquired CMV infections in healthy children or adults do not produce recognizable symptoms. Some infected persons experience a
1041
self-limited infectious mononucleosis syndrome that resembles infection with the Epstein-Barr virus, a closely related member of the herpesvirus family. By contrast, symptomatic and potentially life-threatening CMV infections can occur in children or adults who are immunocompromised by AIDS, chemotherapy, organ transplantation, or other disorders aecting cellmediated immunity. Potential complications of CMV infection in such persons include pneumonitis, hepatitis, gastroenteritis, retinitis, and meningoencephalitis [12]. Most CMV-infected neonates have no apparent signs of CMV infection at birth. Approximately 10% of the infected newborns have systemic signs (Table 1), which consist of jaundice, hepatomegaly, splenomegaly, petechial or purpuric rash, intrauterine growth retardation, or respiratory distress [13,14]. Neurologic signs of intrauterine CMV infection include microcephaly, chorioretinitis, seizures, sensorineural hearing loss, and hypo- or hypertonia. Diagnosis The diagnosis of intrauterine infection is established best by detecting the virus in urine or saliva in clinical samples obtained during the rst 3 weeks of life [24]. Because congenitally infected infants shed enormous quantities
Table 1 Neonatal clinical features in infants with proven intrauterine cytomegalovirus, Toxoplasma gondii, or rubella infections Approximate prevalence Feature Petechiae Microcephaly Hydrocephalus Intrauterine growth retardation Hepatomegaly Splenomegaly Jaundice at birth Sensorineural hearing loss Abnormal tone Chorioretinitis Seizures Death Pneumonitis Congenital heart disease Cytomegalovirusa 50% 50% 5% 50% 45% 45% 40% 40% 25% 10% 10% 5% 5% T. gondiib 20% 15% 40% 10% 50% 50% 65% 30% 75% 15% 5% 10% Rubellac 35% 25% 60% 35% 35% 15% 60% ND 10% ND ND ND 70%
Data from a National Congenital Cytomegalovirus Disease Registry [13]. b Koskiniemi et al [28], Couvreur and Desmonts [36], and Roizen et al [37]. c Schluter et al [85]. d ND no data.
1042
of CMV in their urine, culturing urine using the shell vial assay is the gold standard for the diagnosis of intrauterine CMV infection. The polymerase chain reaction (PCR) can detect CMV DNA in several body uids, including urine, but the sensitivity of this approach does not exceed that of the shell vial assay. Measuring CMV-specic IgG or IgM in the infants serum has much lower sensitivity than culturing the infants urine for CMV, and the CMV antigenemia assay has no role in the diagnosis of congenital CMV infection. Intrauterine CMV infection can be diagnosed prenatally by performing amniocentesis and fetal blood sampling [1517]. CMV can be detected in amniotic uid by cell culture, shell vial assay, or PCR, and CMV-specic IgM can be detected in the serum of infected fetuses. Ultrasonography can be used to identify CMV-induced organ damage, particularly of the developing fetal brain. A normal ultrasound scan early in gestation does not preclude organ damage during the subsequent stages of intrauterine development, however. Neuroimaging studies provide useful information regarding the eects of CMV infection on the developing brain and the likelihood of neurodevelopmental sequelae [18,19]. Approximately 50% of infants with intrauterine CMV infection have periventricular calcications (Fig. 1), a hallmark of intrauterine infection, and infants with calcications have higher rates of permanent neurodevelopmental disabilities [18,19]. Neuroimaging studies in infants with intrauterine CMV infection may also show periventricular leukomalacia, polymicrogyria, pachygyria, or lissencephaly. Because CT detects small calcications more eectively than MRI, CT is the preferable initial imaging study. After the age of 2 years, MRI provides accurate information regarding cortical dysplasia, polymicrogyria, and other disorders of neuronal migration associated with congenital CMV infection. Prevention, treatment, and prognosis Pregnant women can reduce their risk of CMV infection by avoiding direct contact with the urine and saliva of toddler-aged children [20]. Should inadvertent skin contact occur, hand washing with soap and water eliminates the virus from the skin. Because the CNS damage associated with CMV largely occurs in utero, postnatal antiviral therapy has only a modest eect on neurodevelopmental outcomes. Sequelae of intrauterine CMV infection include cerebral palsy, mental retardation, and seizures [21]. The available data suggest that postnatal therapy with ganciclovir may reduce the severity of sensorineural hearing loss and may treat CMV-induced pneumonia in congenitally infected infants [22,23]. In the National Institute of Allergy and Infectious Diseases Childrens Antiviral Study Groupsponsored treatment trial, infants received 6 mg/kg intravenously every 12 hours for 6 weeks [22,24]. CMV vaccines, including subunit vaccines based on glycoprotein B, a CMV protein that induces robust immunologic responses, and vaccines
1043
Fig. 1. An unenhanced head CT scan of an infant with congenital cytomegalovirus (CMV) infection shows dense periventricular calcications. The white matter appears hypodense, suggesting periventricular leukomalacia. The cortical and white matter features of congenital CMV infection are imaged better by MRI, but MRI can miss small calcications, which are the most common central nervous system features of intrauterine CMV infection.
using attenuated live CMV, have been investigated in phase 1 and 2 trials [25,26]. These vaccines seem safe, but no data exist regarding their ecacy to prevent maternal or fetal infection. Recent observations indicating that reinfection with new CMV strains can cause intrauterine infection in seropositive women suggest that prevention of CMV is likely to remain a complex problem even when safe and eective vaccines are developed [11]. Most infants with silent CMV infections escape neurodevelopmental sequelae. Approximately 15% of these children later manifest varying degrees of sensorineural hearing loss. Because of the high incidence of intrauterine CMV infection, this disorder represents the most common cause of nongenetic deafness in many regions of the world [27]. By contrast, infants with CMV disease have high rates (>90%) of neurodevelopmental sequelae, consisting of cerebral palsy, seizures, sensorineural hearing loss, mental retardation, behavioral disorders, and vision loss [2,3,18,19,21]. Approximately 9000 infants annually in the United States have audiologic or neurologic sequelae of intrauterine CMV infections [2].
1044
Toxoplasma gondii Epidemiology Toxoplasma gondii, an obligate intracellular protozoan, infects birds and many mammals, especially cats, worldwide [28]. Human beings acquire infection by ingesting undercooked meat that contains bradyzoites, the encysted form of T. gondii found in tissues, or by ingesting fruits, vegetables, and other foodstus that are contaminated by oocysts, the highly infectious form of the organism [29]. Infected cats, important contributors to human infection, excrete vast quantities of oocysts, and these oocysts can remain viable in warm moist soils for a year or more. Wide variations reecting dietary habits, feline exposure, contact with soil, and other variables exist in the seroprevalence rates against T. gondii infection among human populations. Recent epidemiologic data from the Centers for Disease Control and Prevention (CDC) indicate that 15% of young women have serologic evidence of prior T. gondii infection [30], conrming that most women in the United States remain susceptible to T. gondii. By contrast, studies in France, a highly endemic region, show that 50% to 80% of women of childbearing age possess antitoxoplasma antibodies [31]. Approximately 0.1% to 2% of the adult population, depending on the geographic region and risk variables, acquire T. gondii annually [3032]; like CMV, fetal infections complicate approximately 40% of the infections in pregnant women [33]. Rates of congenital infection vary from 0.8 per 10,000 live births to as high as 1 per 1000 live birth [3235]. These data suggest that as many as 3000 to 4000 infants with congenital toxoplasmosis are born in the United States each year. Clinical manifestations The risk of experiencing symptoms during acquired toxoplasmosis, like that of CMV infection, depends largely on the immune status of the infected host. Although persons with AIDS and other immunocompromising disorders have high rates of invasive infections, healthy pregnant women who become infected with T. gondii rarely have recognizable symptoms. Occasionally, an acute mononucleosis-like illness with prominent lymphadenopathy develops [28]. As the TORCH acronym implies, infants with congenital toxoplasmosis resemble those with intrauterine CMV infection [28,3638]. A substantial number infants lack identiable symptoms of T. gondii infection at birth, but unlike CMV, most of these infants later have ophthalmologic or neurologic sequelae of intrauterine infection with T. gondii [36]. Symptomatic neonates commonly exhibit jaundice, splenomegaly, hepatomegaly, fever, anemia, chorioretinitis, hydrocephalus or microcephaly, and petechiae secondary to thrombocytopenia (see Table 1). Infants with congenital toxoplas-
1045
mosis frequently have hydrocephalus and chorioretinitis, whereas intrauterine growth retardation and sensorineural hearing loss are more likely in CMV-infected infants. Intrauterine coinfections with T. gondii and HIV or CMV have been described elsewhere [39]. Diagnosis The diagnosis of intrauterine infection with T. gondii is usually established postnatally using serologic methods [40,41], although inoculation of fresh placental tissues into laboratory mice can conrm the diagnosis of congenital toxoplasmosis. The PCR can be used to detect T. gondii in clinical specimens [42]. Antitoxoplasma antibodies of several classes (IgM, IgG, IgA, and IgE) can be detected in the infants serum. Because as many as 25% of infected infants lack antitoxoplasma IgM and detection of antitoxoplasma IgG does not distinguish intrauterine infection from passive transmission of maternal antibody, serologic studies of infants and their mothers should be sent to a laboratory, such as the Palo Alto Laboratories (telephone number: 650-8534828), that performs a panel of serologic studies for T. gondii [40]. Imaging studies in infants with congenital toxoplasmosis often reveal intracranial calcications and ventriculomegaly caused by either hydrocephalus or passive ventricular enlargement secondary to loss of brain parenchyma [38]. In contrast to the calcications observed in infants with congenital CMV infection, calcications in congenital toxoplasmosis tend to be scattered throughout the brain parenchyma rather than being predominantly periventricular (Fig. 2). Hydrocephalus results from periventricular inammation of the aqueduct of Sylvius or foramina of Monro. Prevention, treatment, and prognosis There are currently no eective vaccine strategies to prevent congenital toxoplasmosis. Pregnant women can reduce their risk of exposure to T. gondii by avoiding cats and consumption of undercooked meats (Table 2). In endemic regions, such as France and Italy, treating infected mothers with spiramycin or pyrimethamine-sulfadiazine has been advocated to reduce the risk of congenital toxoplasmosis [4346]. Treatment within 4 weeks of maternal seroconversion seems to be more eective than delayed therapy [47]. Therapeutic abortion has also been used when the diagnosis of intrauterine toxoplasmosis has been conrmed by amniocentesis or cordocentesis and evidence of CNS disease has been detected by fetal ultrasonography. Studies performed during the early 1990s indicated that the prognosis of infants with congenital toxoplasmosis can be improved by early shunting of obstructive hydrocephalus and prolonged postnatal therapy with sulfadiazine and pyrimethamine [37,48,49]. Currently, infants with proven congenital toxoplasmosis should be treated with 100 mg/kg/d of sulfadiazine and 1 mg/kg/d of pyrimethamine for 6 months followed by the same dose three
1046
Fig. 2. An unenhanced head CT scan of an infant with congenital toxoplasmosis shows scattered and dense parenchymal calcications. The infant underwent ventriculoperitoneal shunt placement in the perinatal period as indicated by the valves present on the surface of the calvarium bilaterally.
times per week for an additional 6 months. Infants require 5 to 10 mg of folinic acid three times weekly. Previous studies of untreated infants with congenital toxoplasmosis indicated that 80% of surviving infants have epilepsy, 70% have cerebral palsy, 60% have visual impairment, and nearly 60% have IQs below 70 [36]. By contrast, infants receiving prolonged courses of antitoxoplasma therapy with sulfadiazine and pyrimethamine have much lower rates of these complications [48,49].
Herpes simplex virus type 2 Epidemiology Herpes simplex virus type 1 (HSV-1) and herpes simplex virus type 2 (HSV-2) have approximately 50% nucleotide homology but display considerable dierences in their epidemiology and clinical manifestations [50]. HSV-1, typically transmitted by oral routes, causes stomatitis, keratitis, and meningitis or encephalitis in children or adults. By contrast, HSV-2, typically transmitted sexually, causes genital infection, myelitis, aseptic
Table 2 Strategies for the prevention and treatment of intrauterine infections Therapy Ganciclovira Acyclovirb None None None Sulfadiazine and pyrimethamine
Infectious disorder
Preventive measures
Cytomegalovirus
Herpes simplex virus Varicella zoster virus
Rubella Lymphocytic choriomeningitis virus
Toxoplasma gondii
Syphilis
Avoiding contact with young children during pregnancy, practicing monogamy None established Avoiding contact with persons with chickenpox immunizationc Immunization in childhood Avoiding contact with mice and hamsters during pregnancy Avoiding contact with cats and avoiding consumption of poorly cooked meats, prenatal therapy with spiramycin or sulfadiazine-pyrimethamine Avoiding contact with infected persons, maternal penicillin therapy Penicillin
Experimental (see text). Acyclovir therapy diminishes viral shedding but does not aect neurologic outcome in congenitally infected infants. Presumptive benet of childhood immunization.
1047
1048
meningitis, and infections in newborn infants. Intrauterine HSV disease, a rare disorder, results from HSV-2 infection. The incidence of HSV-2 infection is highest among young women, of whom approximately 2% acquire the virus each year [51]. Primary infection usually occurs without recognizable symptoms. Epidemiologic studies from several regions of the world, including the United States, indicate that approximately 30% of 30-year-old women have serologic evidence of prior HSV-2 infection [5255]. The seroprevalence of HSV-2 infection among women in the United States has risen steadily during the past three decades [51]. The risk of HSV-2 seropositivity among women in the United States is associated with several sociodemographic factors, including AfricanAmerican race, lower income, greater number of lifetime sexual partners, earlier age of sexual intercourse, and drug use [54,56]. As many as 2% of pregnant women become infected with HSV during pregnancy, but most of these infections do not produce fetal or neonatal disease [57]. Neonatal disease is more likely when mothers have genital HSV infection near the time of delivery. The risk of intrauterine transmission of HSV-2 to the fetus seems to be low, and the factors that determine the pathogenesis of congenital HSV infection have not been determined. Fewer than 50 cases of intrauterine HSV infection have been described. Clinical manifestations Approximately 5% of the infants with neonatal HSV-2 infections acquire the virus in utero [58]. Within 24 to 48 hours of birth, these congenitally infected infants have signs of HSV disease consisting of skin vesicles or scarring, chorioretinitis, microcephaly, and micro-ophthalmia [5961]. Intrauterine growth retardation can also be present. Of 13 such infants identied by the National Institutes of HealthCollaborative Antiviral Study Group (NIHCASG) [59], 12 had skin lesions, 8 had chorioretinitis, 7 had microcephaly, and 5 had hydranencephaly, evidence of severe brain necrosis. Less severe congenital HSV-2 infections likely occur, albeit infrequently. Diagnosis The diagnosis of congenital HSV-2 infection can be established best by detecting the virus in skin lesions or body uids using either cell culture or PCR [62,63]. Samples for virus detection should include skin vesicle uid; cerebrospinal uid (CSF); and swabs of the conjunctiva, rectum, and oropharynx. Serologic studies may reveal evidence of HSV-2specic antibody, but serologic diagnosis can be confounded by passive transmission of maternal antibody. Tzank preparation has low sensitivity. Ophthalmologic examination may reveal chorioretinitis and micro-ophthalmia, and imaging studies usually detect evidence of severe brain destruction. Infants with intrauterine HSV infections frequently have dense calcications of the basal
1049
ganglia and thalami, diuse hemispheric cystic lesions, and a lissencephaliclike appearance of the cerebral cortex (Fig. 3) [64]. Prevention, treatment, and prognosis Preventing neonatal HSV-2 infection, including intrauterine infection, remains problematic. In the NIH-CASG study described previously, 4 of the 13 mothers reported primary HSV-2 infections and 1 had recurrent genital herpes, but the remainder (8 of 13) had no known history of genital HSV infection [59]. These data are similar to those of other epidemiologic studies indicating that most women with genital HSV are either unaware of their infection or have asymptomatic shedding of HSV-2 [52,57]. Although infants with neonatal HSV infections require prolonged highdose therapy with acyclovir (60 mg/kg/d for 21 days) [65], acyclovir has little or no benecial eect on the severe CNS damage that accompanies intrauterine HSV-2 infection. Infants with congenital HSV-2 infections often die in infancy, and most of the infants who survive have neurologic sequelae consisting of cerebral palsy, epilepsy, and mental retardation [5961].
Fig. 3. An unenhanced head CT scan of a neonate with intrauterine herpes simplex virus (HSV) infection shows dense calcications of the deep nuclei, diuse white matter hypodensity, and cortical dysplasia resembling lissencephaly.
1050
Varicella zoster virus Epidemiology The varicella zoster virus (VZV) causes chickenpox (varicella) and shingles (herpes zoster). Chickenpox, reecting primary VZV infection, typically aects children less than 10 years of age, with a peak incidence between 5 and 9 years of age among unimmunized children in temperate climates [66]. Chickenpox, common during the winter and spring months, results from respiratory droplet transmission. Children with chickenpox are contagious for approximately 4 days before and 5 days after the onset of the rash. Chickenpox develops in approximately 0.05% of pregnant women [67]. Women who acquire VZV before the twentieth week of gestation have a 2% risk of delivering an infant with the fetal varicella syndrome [68,69]. This risk is greatest in rst-trimester infections, whereas the fetal varicella syndrome is highly unlikely in women who have chickenpox during the second half of pregnancy. Maternal VZV reactivation (ie, shingles) poses little or no risk to the developing fetus. Clinical manifestations Clinical manifestations of the congenital varicella syndrome consist of limb hypoplasia, chorioretinitis, cataracts, micro-ophthalmia, cutaneous scaring (cicatrix), and neurologic abnormalities [61,6971]. The latter can include microcephaly, hydrocephalus, seizures, extremity paralysis, Horner syndrome, and cranial neuropathies. Intrauterine growth retardation can be an additional feature. In general, earlier maternal infections are associated with more severe intrauterine disease. Diagnosis Although the diagnosis can be suspected on the basis of the maternal history, virologic conrmation of intrauterine varicella infection can be dicult. Infection can be established by detecting the virus in fetal tissues with cell culture or PCR and by detecting VZV-specic IgM in fetal blood samples obtained by cordocentesis [72,73]. Ultrasonography frequently reveals fetal anomalies compatible with varicella infection. At birth, infants with the fetal varicella virus syndrome no longer shed VZV, and VZV-specic IgM can be undetectable [72]. Imaging studies may show intracranial calcications, cortical dysplasia, or hydranencephaly [61,64]. Prevention, treatment, and prognosis Susceptible women can reduce their risk of contracting chickenpox by avoiding children during their pregnancies. Although the ecacy of varicella zoster immune globulin (VZIG) in preventing the fetal varicella syndrome is
1051
unproven, VZIG should be considered when susceptible pregnant women have signicant exposures to persons with chickenpox. VZIG should be given within 96 hours of the exposure [74]. Postnatal therapy with acyclovir or other antiviral medications provides no benet for infants with the congenital varicella syndrome. A substantial number of infants with the fetal varicella syndrome die during infancy. Many of the surviving infants have permanent sequelae aecting their vision or neurologic development [61,71].
Lymphocytic choriomeningitis virus Epidemiology Lymphocytic choriomeningitis (LCM) virus, rst identied as a cause of human infection in the 1930s, naturally infects the feral house mouse (Mus musculus) but can also infect hamsters [75]. Infected mice and hamsters chronically excrete LCM virus in urine, feces, and respiratory secretions. Human infection results from inhalation of infected aerosols, direct contact with infected animals, or ingestion of material contaminated by infected excreta. Occasional outbreaks have been associated with the purchase of pet hamsters or occupational contact with infected laboratory mice [76,77]. Most recognized human infections develop during the winter months, although infections can be observed throughout the year. Seroprevalence studies suggest that the incidence of LCM virus infection is low among the general population. Approximately 5% of the adults attending a sexually transmitted disease clinic in Baltimore had serologic evidence of prior LCM virus infection [78]. Trapping of mice from urban sites in Baltimore disclosed considerable variation in LCM virus infection rates among mouse colonies [79]. A seroepidemiologic study performed in Birmingham, Alabama indicated that persons older than 30 years of age were more likely to have serologic evidence of prior LCM virus infection (5.4% seroprevalence among persons >30 years of age versus 0.3% among persons \30 years of age) [80]. Seroprevalence rates of 2.4% and 4% were observed among adults in Argentina and Nova Scotia [81,82]. Intrauterine LCM virus infection is presumed to be a rare condition, but no surveillance studies have been performed to establish the incidence of the disorder [75]. Clinical manifestations The clinical features of congenital LCM virus infection mimic those of intrauterine CMV disease and congenital toxoplasmosis [75,83,84]. Among 26 reported cases reviewed by Wright and colleagues [84] in 1997, 88% had chorioretinopathy, 43% had macrocephaly at birth, and 13% were microcephalic. Virtually all infants were born at term and had normal birth weights. Occasional infants have vesicular or bullous skin lesions.
1052
Approximately 50% of mothers who give birth to infants with congenital LCM virus infection recalled inuenza-like illness during pregnancy, and one fourth reported contact with rodents, usually mice, during their pregnancies. Diagnosis Currently, the diagnosis of congenital LCM virus infection is established by serologic studies of the mother and infant [75,84]. Antibodies against LCM virus can be detected in the infants serum or CSF. Because of the low seroprevalence of LCM virus infection in the general population, detection of LCM virusspecic antibodies, particularly LCM virusspecic IgM, strongly supports the diagnosis. The virus can be detected by PCR, but this method does not yet have diagnostic utility in human infections. Imaging studies in infants with congenital LCM virus infections may show hydrocephalus, intracranial calcications, and cortical dysplasia [84]. Prevention, treatment, and prognosis No vaccine exists to prevent infection with LCM virus, but pregnant women can reduce their risk of LCM virus infection by avoiding mice and pet hamsters during their pregnancies. Eective postnatal antiviral therapy of LCM virus infections has not been described. Approximately 35% of infants die from complications of congenital LCM virus infection [84]. Most surviving infants have neurodevelopmental sequelae consisting of cerebral palsy, epilepsy, vision loss, and mental retardation. Because the virus causes an intense inammatory reaction within CSF pathways, infants can experience progressive hydrocephalus and require shunt placement (see additional discussions of LCM virus infections elsewhere in this issue). Rubella Epidemiology The epidemiology of CRS, the consequence of maternal rubella (German measles) infection, has changed considerably since Greggs initial description of the syndrome [85]. In the prevaccine era, epidemics of rubella appeared at 6- to 9-year intervals, causing thousands of cases of CRS, but after licensure of the rubella vaccine in 1969 and aggressive immunization programs targeting adolescents and adults in the 1980s, the incidence of the CRS in the United States and other nations declined dramatically [85]. In the United States, only one to three cases of CRS were reported to the CDC annually during the mid- to late 1980s, an incidence of less than 0.1 case of CRS per 100,000 live births [86,87]. In recent years, cases of CRS in the United States have occurred primarily among immigrants from regions without compulsory immunization [85,88].
1053
Clinical manifestations In 1941, Gregg [89] linked maternal rubella virus infection to cataracts, hearing loss, and heart disease in young infants. Later reports, especially those describing infants born during the rubella pandemic of the early 1960s, expanded the clinical features of the disorder [9092]. Neonates with CRS have a spectrum of abnormalities that includes cataracts, retinopathy, micro-ophthalmia, microcephaly, and sensorineural hearing loss as well as meningoencephalitis, osteopathy, pneumonitis, hepatitis, hepatosplenomegaly, thrombocytopenia, and jaundice (see Table 1). The propensity of the rubella virus to infect the heart causes myocarditis, patent ductus arteriosus, valvular stenosis, or septal defects at the atrial or ventricular level [91]. The fetal consequences of rubella relate directly to the gestational timing of infection [90,91]. Maternal rubella during the initial 8 weeks of gestation produces cataracts and congenital heart lesions, whereas hearing loss correlates with infection during the rst 16 weeks. Stillbirth or spontaneous abortion can also occur during this time. By contrast, infections after the sixteenth week of gestation usually leave no sequelae, although silently infected infants can have sensorineural hearing loss. Diagnosis The diagnosis of CRS can be conrmed by detecting infectious virus in body uids (nasal secretions, urine, or CSF) or rubella virusspecic IgM in the infants serum [93]. Postnatal persistence of rubella-specic IgG supports the diagnosis of CRS. Imaging studies can show periventricular calcications, periventricular leukomalacia, or subependymal cystic lesions. Prevention, treatment, and prognosis Because CRS cannot be treated eectively by postnatal antiviral therapy, infection must be prevented by vaccination. Serologic screening of women can determine the risk of fetal infection and should be performed before pregnancy and in pregnant women exposed to rubella [94]. The presence of maternal antibody before conception or at the time of rubella exposure indicates that the fetus is not at risk. The Committee on Infectious Diseases of the American Academy of Pediatrics recommends that serologic studies in women who are exposed to rubella be repeated in 3 to 4 weeks and once again 6 weeks after exposure if the second sample was also negative [94]. Persistently negative samples indicate that maternal infection did not occur, whereas seroconversion indicates infection and potential risk to the fetus. Termination of pregnancy is considered in conrmed primary rubella infections early in gestation. When termination is not an option, immune globulin can be considered, although the ecacy of this approach is unproven [94].
1054
Progressive sensorineural hearing loss can develop in children who survive CRS, indicating the need for serial audiometry, and children with CRS also have an increased risk of growth failure or diabetes mellitus beginning in the second or third decade of life [95,96]. Neurologic sequelae include microcephaly, language delay, autistic features, and developmental or mental retardation.
Syphilis Epidemiology Although the overall incidence of congenital syphilis, the consequence of maternal infection with Treponema pallidum, is low in the United States, syphilis remains a threat in urban areas and the rural South [9799]. In virtually all other developed countries, syphilis is rare. The resurgence of congenital syphilis in the United States during the late 1980s and early 1990s was associated with maternal use of illicit drugs, especially cocaine [99]. Untreated maternal infection causes perinatal death, stillbirth, or miscarriage in approximately 40% of pregnancies, and rates of fetal infections during secondary-stage maternal infections range from 60% to 100%. Clinical manifestations Early signs of congenital syphilis can be evident at birth or appear during the rst 2 years of life [100,101]. These consist of intrauterine growth retardation, rash, hepatosplenomegaly, jaundice, lymphadenopathy, pseudoparalysis, and bony abnormalities. Hemolytic anemia, thrombocytopenia, leukocytosis, or leukopenia can be present, and radiographic studies during the perinatal period may show signs of osteochondritis [101]. Late signs of congenital syphilis, features evident after 2 years of age, include sensorineural deafness, dental abnormalities, saddle nose, saber shins, hydrocephalus, and developmental delay. Diagnosis Infants with suspected congenital syphilis require (1) quantitative treponemal and nontreponemal assays of serum, (2) VDRL and routine studies of CSF, (3) complete blood cell and platelet counts, and (4) long bone radiographs. Denitive diagnosis is made by demonstrating spirochetes in exudates or tissues using darkeld or direct immunouorescence microscopy [102]. Serologic tests consist of nontreponemal tests, such as the VDRL slide test or rapid plasmin reagin (RPR) test, and treponemal tests, such as the uorescent treponemal antibody absorption (FTA-ABS) test or microhemagglutination test for T. pallidum (MHA-TP). Infants likely have congenital syphilis when spirochetes are present in tissues or when they have
1055
reactive CSF VDRL test results or serum quantitative treponemal test results four times higher than those of the mother [102]. Prevention and treatment Women with acquired syphilis require penicillin treatment using two doses of benzathine penicillin (2.4 million U administered intramuscularly) given 1 week apart, regardless of the timing of pregnancy [102,103]. No eective alternatives exist for penicillin-allergic persons, so desensitization should be performed. Neonates with proven or highly suspected symptomatic congenital syphilis require 50,000 U/kg of aqueous crystalline penicillin G given intravenously every 12 hours during the rst week of life and every 8 hours thereafter for a total of 10 days [103]. Alternatively, procaine penicillin G can be given intramuscularly at a dose of 50,000 U/kg once a day for 10 days. Infectious disease experts should be consulted regarding current treatment strategies for infants whose mothers received inadequate treatment, infants with asymptomatic infections, or infants older than 4 weeks with possible syphilis and neurologic involvement. Should penicillin G not be available, alternative treatment recommendations can be found at www. cdc.gov/nchstp/dstd/penicillinG.htm/. Other agents Intrauterine transmission of Plasmodium sp produces congenital malaria, a neonatal disorder with fever, jaundice, hepatosplenomegaly, thrombocytopenia, and hemolytic anemia [104]. The diagnosis can be suspected on the basis of sociodemographic data and conrmed by detecting parasites in Wright- or Giemsa-stained blood smears. Treatment consists of chloroquine administered orally for sensitive plasmodia or quinine plus a second antimalarial agent for chloroquine-insensitive organisms [105]. Infants with congenital Chagas disease, which is caused by intrauterine infection with Trypanosoma cruzi, have hepatosplenomegaly, petechiae, jaundice, anemia, and seizures. The diagnosis can be established by detecting serologic responses or parasitemia [106]. T. cruzi remains endemic in most of Latin America. Intrauterine infection with Venezuelan equine encephalitis (VEE) virus, another pathogen endemic to Latin America, causes severe fetal disease with stillbirth and hydranencephaly [107]. Conditions that mimic congenital viral infections Several noninfectious disorders produce clinical features that can resemble those of intrauterine infections [108]. Aicardi syndrome, Warburg syndrome, incontinentia pigmenti, and neonatal Graves disease are among the noninfectious disorders that occasionally may be confused with an intrauterine infection.
1056
References
[1] Nahmias AJ. The TORCH complex. Hosp Pract 1974;May:6572. [2] Demmler GJ. Summary of a workshop on surveillance of congenital cytomegalovirus disease. Rev Infect Dis 1991;13:31529. [3] Bale JF, Jr, Jordan MC. Cytomegalovirus. In: Vinken PJ, Bruyn GW, Klawans HL, editors. Handbook of clinical neurology, vol. 12. Viral disease. Amsterdam: Elsevier; 1989. p. 26379. [4] Britt WJ, Alford CA. Cytomegalovirus. In: Fields BN, Knipe DM, Howley JM, Fields virology. 3rd edition. Philadelphia: Lippincott-Raven; 1996. p. 2493523. [5] Pass RF, August AM, Dworsky M, et al. Cytomegalovirus infection in a day care center. N Engl J Med 1982;307:4779. [6] Pass RF, Hutto C, Ricks R, et al. Increased rate of cytomegalovirus infection among parents of children attending day care centers. N Engl J Med 1986;314:36670. [7] Adler SP. Cytomegalovirus and child day care: evidence for an increased infection rate among day care workers. N Engl J Med 1989;321:12906. [8] Murph JR, Baron JC, Brown CK, et al. The occupational risk of cytomegalovirus among day care providers. JAMA 1991;265:6038. [9] Stagno S, Pass RF, Cloud G, et al. Primary cytomegalovirus infection in pregnancy: incidence, transmission to fetus, and clinical outcome. JAMA 1986;256:19048. [10] Bale JF, Jr, Petheram SJ, Souza IE, Murph JA. Cytomegalovirus reinfection in children. J Pediatr 1996;128:34752. [11] Boppana SB, Rivera LB, Fowler KB, et al. Intrauterine transmission of cytomegalovirus to infants of mothers with preconceptional immunity. N Engl J Med 2001;344:136671. [12] Drew WL. Cytomegalovirus infection in patients with AIDS. Clin Infect Dis 1992;14: 60815. [13] Istas AS, Demmler GJ, Dobbins JG, Stewart JA. Surveillance for congenital cytomegalovirus disease: a report from the National Congenital Cytomegalovirus Disease Registry. Clin Infect Dis 1995;20:66570. [14] Boppana SB, Pass RF, Britt WJ, et al. Symptomatic congenital cytomegalovirus infection: neonatal morbidity and mortality. Pediatr Infect Dis J 1992;11:939. [15] Lazzarotto T, Varani S, Gabrielli L, et al. New advances in the diagnosis of congenital cytomegalovirus infection. Intervirology 1999;42:3907. [16] Lazzarotto T, Varani S, Guerra B, et al. Prenatal indicators of congenital cytomegalovirus infection. J Pediatr 2000;137:905. [17] Enders G, Bader U, Lindemann L, et al. Prenatal diagnosis of congenital cytomegalovirus infection in 189 pregnancies with known outcome. Prenat Diagn 2001;21:36277. [18] Noyola DE, Demmler GJ, Nelson CT, et al. Early predictors of neurodevelopmental outcome in symptomatic congenital cytomegalovirus infection. J Pediatr 2001;138:32531. [19] Boppana SB, Fowler KB, Vaid Y, et al. Neuroradiographic ndings in the newborn period and long term outcome in children with symptomatic congenital cytomegalovirus infection. Pediatrics 1997;99:40914. [20] Adler SP, Finney JW, Manganello AM, Best AM. Prevention of child-to-mother transmission of cytomegalovirus by changing behaviors: a randomized controlled trial. Pediatr Infect Dis J 1996;15:2406. [21] Pass RF, Stagno S, Meyers GJ, et al. Outcome of symptomatic congenital cytomegalovirus infection: results of long-term, longitudinal follow-up. Pediatrics 1980;66:75862. [22] Kimberlin DW, Lin CY, Sanchez P, et al. Ganciclovir (GCV) treatment of symptomatic congenital cytomegalovirus (CMV) infection: results of a phase III randomized trial [abstract 1942]. Presented at the 40th Interscience Conference on Antimicrobial Agents and Chemotherapy. Toronto, Ontario, Canada, September, 1720, 2000. [23] Hocker J, Cook L, Adams G, et al. Ganciclovir therapy of congenital cytomegalovirus pneumonia. Pediatr Infect Dis J 1990;9:7435.
1057
[24] Whitley RJ, Cloud G, Gruber W, et al. Ganciclovir treatment of symptomatic congenital cytomegalovirus infection: results of a phase II study. J Infect Dis 1996;175:10806. [25] Pass RF, Duliege AM, Boppana S, et al. A subunit cytomegalovirus vaccine based on recombinant envelope glycoprotein B and a new adjuvant. J Infect Dis 1999;180: 9705. [26] Plotkin SA. Vaccination against cytomegalovirus. Arch Virol Suppl 2001;17:12134. [27] Dahle AJ, Fowler KB, Wright JD, et al. Longitudinal investigation of hearing disorders in children with congenital cytomegalovirus. J Am Acad Audiol 2000;11:28390. [28] Koskiniemi M, Lappalainen M, Hedman K. Toxoplasmosis needs evaluation: an overview and proposals. Am J Dis Child 1989;143:7248. [29] Baril L, Ancelle T, Goulet V, et al. Risk factors for Toxoplasma infection in pregnancy: a case control study in France. Scand J Infect Dis 1999;31:3058. [30] Jones JL, Kruszon-Moran D, Wilson D, et al. Toxoplasma gondii infection in the United States: seroprevalence and risk factors. Am J Epidemiol 2001;154:35265. [31] Jeannel D, Niel G, Costagliola D, et al. Epidemiology of toxoplasmosis among pregnant women in the Paris area. Int J Epidemiol 1988;17:595602. [32] Evengard B, Petersson K, Engman ML, et al. Low incidence of toxoplasma infection during pregnancy and in newborns in Sweden. Epidemiol Infect 2001;127:1217. [33] Couvreur J. Problems of congenital toxoplasmosis: evolution over four decades. Presse Med 1999;28:7537. [34] Ambroise-Thomas P, Schweitzer M, Pinon JM, et al. Prevention of congenital toxoplasmosis in France. Risk assessment. Results and perspectives of prenatal screening and newborn follow up. Bull Acad Natl Med 2001;185:66583. [35] Mara M, Hsu HW, Eaton RB, et al. Epidemiology of congenital toxoplasmosis identied by population-based newborn screening in Massachusetts. Pediatr Infect Dis J 2001;20: 11325. [36] Couvreur J, Desmonts G. Congenital and maternal toxoplasmosis. Dev Med Child Neurol 1962;4:51930. [37] Roizen N, Swisher C, Stein M, et al. Neurologic and developmental outcome in treated congenital toxoplasmosis. Pediatrics 1995;95:1120. [38] Swisher C, Boyer K, McLeod R. Congenital toxoplasmosis. Semin Pediatr Neurol 1994;1:425. [39] Mitchell CD, Erlich SS, Mastrucci MT, et al. Congenital toxoplasmosis occurring in infants perinatally infection with human immunodeciency virus 1. Pediatr Infect Dis J 1990;9:5128. [40] Boyer KM. Diagnostic testing for congenital toxoplasmosis. Pediatr Infect Dis J 2001; 20:5960. [41] Naessens A, Jenum PA, Pollak A, et al. Diagnosis of congenital toxoplasmosis in the neonatal period: a multicenter evaluation. J Pediatr 1999;135:7149. [42] Johnson J, Butcher P, Savva D, et al. Application of the polymerase chain reaction to the diagnosis of human toxoplasmosis. J Infect 1993;26:14758. [43] Hohlfeld P, Daos F, Thulliez P, et al. Fetal toxoplasmosis: outcome of pregnancy and infant follow-up after in utero treatment. J Pediatr 1989;115:7659. [44] Ghidini A, Sitori M, Spelta A, Vergani P. Results of a preventive program for congenital toxoplasmosis. J Reprod Med 1991;36:2703. [45] Pratlong F, Boulat P, Villena I, et al. Antenatal diagnosis of congenital toxoplasmosis: evaluation of the biological parameters in a cohort of 286 patients. Br J Obstet Gynaecol 1996;103:5527. [46] Wallon M, Liou C, Garner P, Peyron F. Congenital toxoplasmosis: systematic review of evidence of ecacy of treatment in pregnancy. BMJ 1999;318:15114. [47] Gilbert R, Gras L, Wallon M, et al. Eect of prenatal treatment on mother to child transmission of Toxoplasma gondii: retrospective cohort study of 554 mother-child pairs in Lyon, France. Int J Epidemiol 2001;30:13038.
1058
[48] Guerina N, Hsu H, Meissner H, et al. Neonatal serologic screening and early treatment for congenital Toxoplasma gondii infection. N Engl J Med 1994;330:185863. [49] McAuley J, Boyer K, Patel D, et al. Early and longitudinal evaluations of treated infants and children and untreated historical patients with congenital toxoplasmosis. The Chicago Collaborative Treatment Trial. Clin Infect Dis 1994;18:3872. [50] Corey L, Spear PG. Infections with herpes simplex viruses. N Engl J Med 1986;314:686 91,7495. [51] Armstrong GL, Schillinger J, Markowitz L, et al. Incidence of herpes simplex virus type 2 infection in the Unites States. Am J Epidemiol 2001;153:91220. [52] Mertz GJ. Epidemiology of genital herpes infections. Infect Dis Clin North Am 1993; 7:82539. [53] Sugligoi B, Tchamgmena O, Sarmati L, et al. Prevalence and risk factors for herpes simplex virus type 2 infection among adolescents and adults in northern Cameroon. Sex Transm Dis 2001;28:6903. [54] Buchacz K, McFarland W, Hernandez M, et al. Prevalence and correlates of herpes simplex virus type 2 infection in a population-based survey of young women in lowincome neighborhoods of Northern California. The Young Womens Survey Team. Sex Transm Dis 2000;27:393400. [55] Patrick DM, Dawar M, Cook DA, et al. Antenatal seroprevalence of herpes simplex virus type 2 (HSV-2) in Canadian women: HSV-2 prevalence increases throughout the reproductive years. Sex Transm Dis 2001;28:4248. [56] Sucato G, Celum C, Dithmer D, et al. Demographic rather than behavioral risk factors predict herpes simplex virus type 2 infection in sexually active adolescents. Pediatr Infect Dis J 2001;20:4226. [57] Brown ZA, Selke S, Zeh J, et al. The acquisition of herpes simplex virus during pregnancy. N Engl J Med 1997;337:50915. [58] Baldwin S, Whitley RJ. Intrauterine herpes simplex virus infections. Teratology 1989;39:10. [59] Hutto C, Arvin A, Jacobs R, et al. Intrauterine herpes simplex virus infections. J Pediatr 1987;110:97101. [60] Hoppen T, Eis-Hubinger AM, Schild RL, et al. Intrauterine herpes simplex virus infection. Klin Padiatr 2001;213:638. [61] Grose C. Congenital infections caused by varicella zoster virus and herpes simplex virus. Semin Pediatr Neurol 1994;1:439. [62] Jacobs RF. Neonatal herpes simplex virus infections. Sem Perinatol 1998;22:6471. [63] Kimberlin DW, Lakeman FD, Arvin AM, et al. Application of the polymerase chain reaction to the diagnosis and management of neonatal herpes simplex virus disease. National Institute of Allergy and Infectious Disease Collaborative Antiviral Study Group. J Infect Dis 1996;174:11627. [64] Bale JF, Murph JR. Infections of the central nervous system of the newborn. Clin Perinatol 1997;24:787806. [65] Kimberlin DW, Lin CY, Jacobs RF, et al. Safety and ecacy of high-dose intravenous acyclovir in the management of neonatal herpes simplex virus infections. Pediatrics 2000;108:2308. [66] Arvin A. Varicella-zoster virus. Clin Microbiol Rev 1996;9:36181. [67] Brunell P. Varicella in pregnancy, the fetus, and the newborn: problems in management. J Infect Dis 1992;166:427. [68] Pastuszak A, Levy M, Schick B, et al. Outcome after maternal varicella infection in the rst 20 weeks of pregnancy. N Engl J Med 1994;330:9015. [69] Jones KL, Johnson KA, Chambers CA. Ospring of women infected with varicella during pregnancy: a prospective study. Teratology 1994;49:2932. [70] Paryani SG, Arvin AM. Intrauterine infection with varicella-zoster virus after maternal varicella. N Engl J Med 1986;14:15426.
1059
[71] Alkalay AL, Pomerance JJ, Rimoin DL. Fetal varicella syndrome. J Pediatr 1987; 111:3203. [72] Cuthbertson G, Weiner C, Giller R, et al. Clinical and laboratory observations: prenatal diagnosis of second-trimester congenital varicella syndrome by virus-specic immunoglobulin M. J Pediatr 1987;111:5925. [73] Isada NB, Paar DP, Johnson MP, et al. In utero diagnosis of congenital varicella zoster virus infection by chorionic villus sampling and polymerase chain reaction. Am J Obstet Gynecol 1991;165:172730. [74] American Academy of Pediatrics. Varicella zoster virus. In: Pickering LK, editor. 2000 Red book: report of the Committee on Infectious Diseases. 25th edition. Elk Grove Village, IL: American Academy of Pediatrics; 2000. p. 6301. [75] Barton LL, Mets MB. Congenital lymphocytic choriomeningitis virus infection: decade of rediscovery. Clin Infect Dis 2001;33:3704. [76] Deibel R, Woodall JP, Becher WJ, Schryver GD. Lymphocytic choriomeningitis virus in man. Serologic evidence of association with pet hamsters. JAMA 1975;232:5014. [77] Dykewicz CA, Dato VM, Fisher-Hoch SP, et al. Lymphocytic choriomeningitis outbreak associated with nude mice in a research institute. JAMA 1992;267:134953. [78] Childs JE, Glass GE, Ksiazek TG, et al. Human-rodent contact and infection with lymphocytic choriomeningitis and Seoul viruses in an inner-city population. Am J Trop Med Hyg 1991;44:11721. [79] Childs JE, Glass GE, Korch GW, et al. Lymphocytic choriomeningitis virus and house mouse (Mus musculus) distribution in urban Baltimore. Am J Trop Med Hyg 1992;47: 2734. [80] Park JY, Peters CJ, Rollin PE, et al. Age distribution of lymphocytic choriomeningitis virus serum antibody in Birmingham, Alabama: evidence of a decreased risk of infection. Am J Trop Med Hyg 1997;57:3741. [81] Marie TJ, Saron MF. Seroprevalence of lymphocytic choriomeningitis virus in Nova Scotia. Am J Trop Med Hyg 1998;58:479. [82] Ambrosio AM, Feuillade MR, Gamboa GS, Maiztegui JL. Prevalence of lymphocytic choriomeningitis virus infection in a human population in Argentina. Am J Trop Med Hyg 1994;50:3816. [83] Barton LL, Budd SC, Mortt WS, et al. Congenital lymphocytic choriomeningitis virus infection in twins. Pediatr Infect Dis J 1993;12:9426. [84] Wright R, Johnson D, Neumann M, et al. Congenital lymphocytic choriomeningitis virus syndrome: a disease that mimics congenital toxoplasmosis or cytomegalovirus infection. Pediatrics 1997;100:E9. [85] Schluter WW, Reer SE, Redd SC, Dykewicz CA. Changing epidemiology of congenital rubella syndrome in the United States. J Infect Dis 1998;178:63641. [86] Cochi SL, Edmonds LE, Dyer K, et al. Congenital rubella syndrome in the United States, 19701985. Am J Epidemiol 1989;129:34961. [87] Reef SE, Plotkin S, Cordero JF, et al. Preparing for elimination of congenital rubella syndrome (CRS). Summary of a workshop on CRS elimination in the United States. Clin Infect Dis 2000;31:8595. [88] Centers for Disease Control and Prevention. Rubella outbreak: Arkansas, 1999. Morb Mortal Wkly Rep 2001;50:113739. [89] Gregg NM. Congenital cataract following German measles in the mother. Transactions of the Ophthalmologic Society of Australia 1941;3:3541. [90] Dudgeon J. Congenital rubella. J Pediatr 1975;87:107886. [91] Ueda K, Nishida Y, Oshmia K, Shepard TH. Congenital rubella syndrome: correlation of gestational age at time of maternal rubella with type of defect. J Pediatr 1979;94: 7635. [92] Givens KT, Lee DA, Jones T, Ilstrup DM. Congenital rubella syndrome: ophthalmic manifestations and associated systemic disorders. Br J Ophthalmol 1993;77:35863.
1060
[93] Souza I, Bale JF, Jr. The diagnosis of congenital infections: contemporary strategies. J Child Neurol 1995;10:27182. [94] American Academy of Pediatrics. Rubella. In: Pickering LK, editor. 2000 Red book: report of the Committee on Infectious Diseases. 25th edition. Elk Grove Village, IL: American Academy of Pediatrics; 2000. p. 497. [95] Desmond MM, Fisher ES, Vorderman AL, et al. The longitudinal course of congenital rubella syndrome in nonretarded children. J Pediatr 1979;93:58491. [96] Desmond MM, Wilson CW, Vorderman AL. The health and educational status of adolescents with the congenital rubella syndrome. Dev Med Child Neurol 1985;27:7219. [97] Nakashima AK, Rolfs RT, Flock ML, Kilmarx P, Greenspan JR. Epidemiology of syphilis in the United States, 19411993. Sex Transm Dis 1996;23:1623. [98] Centers for Disease Control and Prevention. Congenital syphilisUnited States, 1998. Morb Mortal Wkly Rep 1999;48:75761. [99] Sison CG, Ostrea EM, Reyes MP, Salari V. The resurgence of congenital syphilis: a cocaine-related problem. J Pediatr 1997;130:28992. [100] Hollier LM, Cox SM. Syphilis. Semin Perinatol 1998;22:32331. [101] Murph JR. Rubella and syphilis: continuing causes of congenital infection in the 1990s. Semin Pediatr Neurol 1994;1:2635. [102] Alexander JM, Sheeld JS, Sanchez PJ, Mayeld J, Wendel GD. Ecacy of treatment of syphilis in pregnancy. Obstet Gynecol 1999;93:58. [103] American Academy of Pediatrics. Syphilis. In: Pickering LK, editor. 2000 Red book: report of the Committee on Infectious Diseases. 25th edition. Elk Grove Village, IL: American Academy of Pediatrics; 2000. p. 54759. [104] Hulbert T. Congenital malaria in the United States: report of a case and review. Clin Infect Dis 1992;14:9226. [105] American Academy of Pediatrics. Syphilis. In: Pickering LK, editor. 2000 Red book: report of the Committee on Infectious Diseases. 25th edition. Elk Grove Village, IL: American Academy of Pediatrics; 2000. p. 70510. [106] Bittencourt AL. Congenital Chagas disease. Am J Dis Child 1975;130:97103. [107] Wenger F. Venezuelan equine encephalitis. Teratology 1977;16:35862. [108] Bale JF, Jr. Conditions mimicking congenital infections. Semin Pediatr Neurol 1994;1: 6370.
Neurol Clin N Am 20 (2002) 10611077
Cerebral venous thrombosis and venous malformations in children

Karen S. Carvalho, MD*, Bhuwan P. Garg, MBBS
James Whitcomb Riley Hospital for Children, Section of Pediatric Neurology, Indiana University Medical Center, 702 Barnhill Drive, Room #1757, Indianapolis, IN 46202-5200, USA
Cerebral venous thrombosis (CVT) usually involves the cerebral venous sinuses such as the sagittal and transverse sinuses, and may extend to involve the cerebral veins. The earliest description of the neuropathology of CVT in children is by Bailey and Hass in 1931 [1]. CVT has been considered a rare disorder with poor prognosis. It is now clear that CVT is more common than once believed and there is increasing awareness of the variety of clinical presentations [2]. It is estimated that CVT constitutes 25% of ischemic cerebrovascular disease in children with an annual incidence of 0.29 per 100,000 [3,4]. Outlook for CVT has also changed dramatically, owing to earlier diagnosis and eective treatment; however, with variable and subtle clinical presentation, it remains a diagnostic challenge.
Venous anatomy The venous drainage of the cerebral hemispheres can be subdivided into a supercial and a deep one. The intra-axial veins collect capillary blood and drain into the supercial and deep venous systems that empty into the dural sinuses, themselves drained mostly by internal jugular veins [5,6]. This network of vessels runs independent of the arteries, and anastomosis freely, and is unresponsive to changes in the systemic blood pressure. The supercial venous system is composed mainly of the superior and inferior supercial veins that drain into the superior sagittal sinus (SSS) and cavernous sinus, respectively. The middle cerebral veins are connected to the SSS by the great anastomotic vein of Trolard and to the transverse sinus by
* Corresponding author. E-mail address: kcarvalh@iupui.edu (K.S. Carvalho). 0733-8619/02/$ - see front matter 2002, Elsevier Science (USA). All rights reserved. PII: S 0 7 3 3 - 8 6 1 9 ( 0 2 ) 0 0 0 1 1 - 7
1062
K.S. Carvalho, B.P. Garg / Neurol Clin N Am 20 (2002) 10611077
` the vein of Labbe. The SSS empties into the right lateral sinus (transverse and sigmoid sinuses) and the right jugular veins in most cases. The deep venous system consists of the galenic system and variable basal veins. The galenic system is formed by the vein of Galen and its major tributaries, the basal vein of Rosenthal, and the internal cerebral veins. This system empties into the straight sinuses and ultimately into the left lateral sinus and the left jugular vein in most cases. Anteriorly, the paired cavernous sinuses drain blood from the orbits through the ophthalmic veins and from the anterior part of the base of the brain by the sphenoparietal sinuses and the middle cerebral veins. This system communicates with the jugular system by way of the superior and inferior petrosal sinuses [5]. The cerebellum is drained by the superior, inferior, and lateral sets of veins. The superior cerebellar veins drain into the straight sinus and the vein of Galen. The inferior cerebellar veins drain into the lateral, superior petrosal, and occipital sinuses. Pathophysiology Thrombosis of the venous system can occur because of venous stasis, prothrombotic states, involvement of the vessel wall, and less frequently, embolization [5]. The slower blood ow favors the formation and propagation of thrombus in the venous system. There is also a relative absence of thrombomodulin in the lining of cerebral sinuses that may further increase the tendency to thrombosis [7]. Impairment of venous drainage results in congestion and swelling of the underlying white matter with subsequent ischemia. Platelets play a minimal role in venous thrombosis. In neonates, the venous sinuses are located along the suture line and are prone to mechanical distortion during calvarial molding in the birth process that can predispose to cerebral venous thrombosis [8]. Superior sagittal sinus thrombosis usually results in bilateral cortical infarcts, often hemorrhagic. Deep venous thrombosis may produce thalamic or cerebellar infarcts. Recanalization of the occluded sinus may occur over several days and weeks. The site of the thrombosis and the area of infarction do not clearly correlate with the neurologic decit when compared with arterial strokes [9].
Clinical presentation The presenting clinical features of venous sinus thrombosis in children are age-dependent and can vary from minimal and nonspecic symptoms such as decreased oral intake and irritability to more ominous signs such as lethargy and coma. Neonates and younger children seem to be at greater risk for CVT. In our series of 31 patients with CVT, 19 patients (61.2%) were
1063
neonates with a median age of 14 days [2]. In another series 41% of the patients were neonates [10]. Seizures, fever, lethargy, or irritability and respiratory distress are common signs of CVT in neonates [2,11]. Older children commonly present with fever and lethargy often associated with the classic signs of intracranial hypertension such as vomiting, headache, papilledema, and abducens nerve palsy [2,1013]. Diagnosis of CVT in children is challenging because of the variable and nonspecic presentation and therefore a high index of suspicion is required [12].
Radiographic features Widespread availability of computerized tomography (CT), magnetic resonance imaging (MRI), and magnetic resonance venography (MRV), has resulted in increased and earlier diagnosis of CVT in children and neonates. Head CT (HCT) with or without contrast may be suggestive of CVT but can miss the diagnosis in 10%40% of patients and may underestimate the extent of the thrombosis [5,11]. Three abnormalities on HCT are considered direct signs of CVT: the cord sign, the dense triangle sign, and the empty delta sign (Box 1) [1416]. HCT may be false positive, particularly in neonates; the higher hematocrit and slower venous blood ow in this age group produce a high-density triangle in the torcula, mimicking sinus thrombosis [15]. MRI and MRV show details of the area of infarct and visualize the absence of ow and the presence of the thrombus and clot progression [17]. MRV shows decreased venous ow. The clot appears as increased signal on T1-weighted and T2-weighted images in a venous sinus or cerebral vein (Figs. 1 and 2). Four-vessel angiography (conventional or digital) shows in detail the partial or complete lack of lling of the cerebral veins or venous sinuses, enlarged collateral veins, delayed venous emptying, and reversal of normal venous ow direction. At present MRI and MRV that
Box 1. Direct signs of CVT visualized in CT of the head (HCT) 1. Cord sign (HCT without contrast): a linear area of increased density related to the thrombus in veins or sinuses. This sign is rare and its diagnostic value is debated [3]. 2. Dense triangle sign (HCT without contrast): opacication of torcula by freshly thrombosed blood [16]. 3. Empty delta sign (HCT with contrast): the delta sign is the area of hypodensity in the torcula correlating with the thrombus in the sinus surrounded with contrast. It is the most common sign, present in approximately 30 percent of the published cases [16].
1064
Author name / Nurs Clin N Am 20 (2002) 10611077
Fig. 1. Magnetic resonance imaging (MRI) scan of a 17-day-old boy showing occlusion of the superior sagittal sinus with a right occipital hemorrhagic stroke. FLAIR TR 10002, TE 148.
are more sensitive than CT are the methods of choice for diagnosis and follow-up of CVT, with angiography required only in dicult or doubtful cases [5].
Risk factors for cerebral venous thrombosis Cerebral venous thrombosis may be associated with a variety of local or systemic conditions. In approximately 25% of patients, no risk factor can be identied despite extensive search. The investigators found a predisposing risk factor in 24 of 31 (77%) patients [2]. In one prospective populationbased study, 65% of children with CVT had multiple risk factors and 40% had more than three [10]. Caution must be exercised when diagnosing idiopathic CVT, as continued surveillance may uncover causes that require treatment. Common risk factors for venous sinus thrombosis in children and adolescents are shown in Box 2. CVT is associated with states of altered cerebral hemodynamics such as dehydration, shock, and congestive heart failure [1,2,18]. Systemic sepsis and local infection such as mastoiditis remain important causes of CVT. The
1065
Fig. 2. Venogram of a 14-day-old boy showing partial sagittal sinus thrombosis (arrow).
investigators found mastoiditis in 7 of 12 children older than 1 month of age with CVT [2]. Other paranasal sinus infections and perioral and periorbital infections may also be associated with CVT. A hypercoagulable state, whether acquired or inherited, is important in the pathogenesis of CVT in children [2]. Inherited hypercoagulable disorders were present in 31% of children in one study [10]. Common causes of prothrombotic states are listed in Box 3. Patients with malignant tumor may have a variety of coagulation abnormalities leading to a prothrombotic state (Trosseau syndrome: malignancy-associated hypercoagulable state) [19]. Chemotherapy, particularly L-asparaginase, may also be associated with CVT. CVT has been described in children with non-Hodgkin lymphoma, leukemia, and neuroblastoma [20]. Connective tissue diseases including systemic lupus erythematosus and rheumatoid arthritis have been reported as rare causes of CVT in older children [21,22]. CVT may be seen in patients with nephrotic syndrome and is commonly attributed to renal loss of coagulation factors. Certain drugs including oral contraceptives, increasingly used by adolescent girls, predispose to CVT [23]. The prevalence of hereditary thrombophilia is estimated at 1:2500 to 1:5000 [24]. Newborns and infants less than 1 year of age are at the greatest risk for thromboembolic complications, and the incidence decreases signicantly following the rst year of life [24,25]. Protein C, protein S, and antithrombin III deciency, congenital or acquired because of renal loss,
1066
Box 2. Risk factors for cerebral venous thrombosis Septic Thrombosis Otitis Media and mastoiditis Paranasal sinusitis, scalp and face infections Purulent meningitis Sepsis in neonates Aseptic Thrombosis Dehydration Trauma Congenital heart disease, Congestive heart failure Leukemia and myeloproliferative disease Hemoglobinopathies Inherited prothrombotic disorders Iron deciency anemia Drugs: L-asparaginase, oral contraceptives Renal disease: nephrotic syndrome, Liver disease Disseminated intravascular coagulation Autoimmune disease: SLE, ulcerative colitis, Behcets disease Malignancy Surgery Pregnancy See text for details
hepatic failure, or disseminated intravascular coagulation have been associated with CVT [26]. Factor V Leiden gene mutation resulting in resistance to activated protein C (APCR) is an important cause of venous thrombosis [2,27,28]. Factor V Leiden gene mutation occurs in 5%12% of the general pediatric population [24] and accounts for 90%95% of all activated protein C resistance [29]. Heterozygous carriers have a sevenfold increased risk for thrombosis, whereas the risk for homozygous individuals is increased 80-fold [29]. The impact of Factor V Leiden mutation in children with CVT seems to be highest in newborns and young infants, reecting the dierent physiology of hemostasis, with proportionately lower levels of protein C and protein S frequently observed in this age group [24]. Inherited abnormalities of brinolysis including plasminogen deciency, plasminogen activator deciency, dysbrinogenemia, and Factor XII deciency have been linked to venous strokes in young adults [30,31]. Presence of antiphospholipid antibodies (aPL) has been related to venous thrombosis in children [32]. In the absence of connective tissue diseases, the
1067
Box 3. Prothrombotic abnormalities associated with CVT in children Protein C deciency Protein S deciency Antithrombin III deciency Activated protein C resistance Factor V Leiden gene mutation Prothrombin G20210A gene mutation Hereditary thrombophilia Antiphospholipid antibody syndrome Homocystinuria
association of aPL and stroke is known as primary antiphospholipid antibody syndrome, characterized by chronic headaches, livedo reticularis, pulmonary hypertension, recurrent episodes of deep venous thrombosis, and arterial and venous strokes [33]. In a prospective blood bank survey, antiphospholipid antibodies were detected in 6.5% of normal subjects with prevalence increasing with age [34]. The presence of aPL conveys a stroke risk 10 times that seen in patients without aPL [35]. Children with antiphospholipid antibodies have a signicantly increased risk for arterial and venous thrombosis, 50% of these occur in the central nervous system [32]. Arterial and venous thrombosis is a frequent complication in homocystinuria; however, cerebral venous thrombosis is rare [36]. Patients with this disorder have marfanoid body type, ectopic lentis, and mental retardation, and one third of these patients have thromboembolic events. The proposed mechanism is caused by abnormal collagen cross-links with consequent abnormalities in vessel walls and increased platelet adhesion [35]. The high incidence of inherited hypercoagulable disorders in children with venous sinus thrombosis suggests that a thorough hematologic evaluation should be done regardless of the presence of other risk factors or lack of family history of thrombotic events [2]. Plasma concentration of many coagulation proteins reaches adult ranges by several months of age and age-corrected values should be used [37]. Box 4 lists suggested investigations for hypercoagulable disorders in children with venous sinus thrombosis. Treatment Conservative general medical care and neurologic supportive care is the mainstay of treatment. Adequate hydration is paramount. Aggressive antibiotic therapy and antiepileptic medications should be used when appropriate. Surgical debridement is recommended for the treatment of mastoiditis [38]. The role of internal jugular vein ligation in patients with mastoiditis
1068
Box 4. Prothrombotic disorders laboratory studies for patients with CVT Complete Blood Cell Count Platelets count PT, INR, and aPTT Protein C activity, protein C Antigen Protein S clottable, protein S total Antigen Antithrombin III functional, antithrombin III Antigen Activated protein C resistance (APCR) Factor V Leiden mutation Prothrombin G20210A mutation Homocysteine level MTHFR polymorphism Lipid prole Lipoprotein (a) Antiphospholipid antibody panel dRVVT/Lupus anticoagulant assay Fibrinogen activity Fibrinogen Antigen Factor VIII assay
PT prothrombin time, INR international normalized ratio, aPTT activated partial thromboplastin time, dRVVT dilute Russell viper venom time, MTHFR Methylenetetrahydrofolate reductase.
is controversial and should be reserved for those cases in which thrombophlebitic or embolic spread beyond the sinus is suspected [39]. Some investigators have been more enthusiastic about systemic brinolysis and anticoagulation therapy. There is limited data regarding the ecacy and safety of systemic anticoagulation in a pediatric population. Low molecular weight heparin (LMWH) was found to be safe when used in children with CVT. This study did not evaluate the ecacy of anticoagulation in mortality and long-term neurologic morbidity [40]. Isolated case reports have shown improvement in patients with extensive thrombosis and low incidence of complications [41,42]. The role of thrombolytic therapy in children remains controversial. Outcome Some investigators have suggested that the outcome of cerebral venous thrombosis is age-related, with neonates and young children having unfavorable outcomes [2]. Barron et al reported 10 neonates with cerebral venous thrombosis with only 2 being entirely normal at subsequent exami-
1069
nations [11]. Other investigators have reported similar unfavorable results [43,44]. In our study of 19 neonates with CVT, 11 had developmental delays and only 5 had no neurologic decit. Fifty percent of the older children were neurologically intact [2]. Others have reported complete resolution of neurologic decits in 95% of cases [11]. Shevell et al reported normal outcomes in 88% of neonates with dural venous thrombosis [13]. The reason for the discrepancy in results is not clear and might represent selection bias. Outcome may also be related to localization of the thrombus and the presence of infarct or hemorrhage. Barron et al found that infants and children with infarction caused by deep venous thrombosis had persistent neurologic disability [11]. The investigators experience suggests that multiple sinus thrombosis is associated with higher morbidity [2]. In a large case series of 77 adult patients by Preter et al, 9 patients (11%) had recurrent sinus thrombosis, all in the rst year [45]. Recurrent thrombosis has not been reported in pediatric series [2,1113,44].
Vascular malformations The vascular malformations of the central nervous system are an important cause of hemorrhagic strokes in children. Arteriovenous malformations are characterized by direct anastomosis of arterial and venous channels without an intervening capillary bed [46]. These are discussed elsewhere. Here we discuss common venous malformations of childhood. Vein of Galen aneurysmal malformation The vein of Galen aneurysmal malformation (VGAM) accounts for approximately 10% of the vascular malformation in childhood [47]. VGAM corresponds to 60% of the lesions that present before 6 months of age, however [48]. Clinical features The clinical presentation of VGAM is age-dependent. Forty to fty percent of all VGAM presents in the neonatal period, usually with high output congestive heart failure [49]. Persistent pulmonary hypertension and heart murmur may be present [49]. In late infancy, patients usually present with hydrocephalus and seizures. Spontaneous intracranial hemorrhage is especially common. Dilated scalp veins and a loud bruit over the head are important diagnostic clues. Older children and adolescents usually present with seizures and intracranial hemorrhage. Signs of brainstem and cerebellar involvement are frequent. Once these lesions become symptomatic, the mortality is approximately 50%. Stroke in venous distribution can occur secondary to shunting of blood through the malformation and away from the
1070
brain parenchyma or steal phenomenon [50]. There is an interesting association between congenital cardiac defects, especially aortic coarctation and sinus venosus atrial septal defect, and VGAM [51]. Echocardiogram with attention to the aortic arch and pulmonary veins is imperative in the evaluation of children with VGAM [51]. Diagnostic studies Doppler ultrasound is critical for the prenatal diagnosis of VGAM, and shows an intracerebral hypoechoic cyst [52]. Intrauterine MRI and MRA usually conrm the diagnosis [53]. Three-dimensional power Doppler ultrasonography (3D-CPA) may assist in the diagnosis and provides a threedimensional visualization of the vasculature [54]. MRI is mandatory before treatment to assess the condition of the brain parenchyma [55]. Conventional angiography is essential for pretreatment evaluation. Treatment and outcome A few cases have been reported with spontaneous thromboses of VGAM after diagnostic angiography [56,57]. The approach is inuenced greatly by age, clinical symptoms, and the angiographic architecture of the malformation. Therapeutic options are primarily based on whether a true arteriovenous malformation (AVM) is present or if the malformation represents an arteriovenous stula involving the vein of Galen. Arterial endovascular approaches, microneurosurgery, or radiosurgery are preferred for the management of the former, whereas the transvenous endovascular approach has become the cornerstone for the treatment of the latter [58]. Endovascular embolization remains the treatment of choice in most cases, oering a high rate of cure with low morbidity [59], with multiple stage embolizations sometimes being necessary [60]. Some of the complications of endovascular treatment include perforation of the thin-walled aneurysm or the feeding vessels, cerebral infarction, or other neurologic sequelae [61]. Gamma knife surgery is a viable option in clinically stable patients, especially after failure of multiple embolizations [62]. Before the era of sophisticated imaging technologies and endovascular treatment, vein of Galen malformations were fatal in 90% of symptomatic patients less than 1 month of age and in half of those patients between 1 month and 1 year of age [63]. Treatment of VGAM has been revolutionized by transcatheter embolization techniques, with 70%80% survival among neonates and young infants, and cure rates of approximately 50% [64]. Neonates with VGAM presenting with early cardiac failure continue to have high morbidity and mortality, and outcomes are signicantly better in those presenting in later childhood [60]. Neurologic outcomes in survivors are generally poor. In one study of 22 children, there was a 37% incidence of mental retardation, and 82% had seizure disorder [65].
1071
Cavernous malformations Cavernous malformations (CM) are congenital vascular malformations characterized by irregular sinusoidal vessels lined by thin walls surrounded by normal brain tissue with no obvious feeding arteries or venous drainage. Estimated prevalence of CM ranges from 0.470.9% [6668]. Multiple CMs are common and they often coexist with other vascular anomalies [69]. CM may be sporadic or familial. They are inherited as an autosomal dominant trait. The familial cases have a higher incidence of multiple cavernous malformations and are more likely to be symptomatic [70]. Seventy-ve percent of patients who have multiple lesions and who present as sporadic cases in fact have a hereditary form of the disorder [70]. It has been estimated that the familial cases make up approximately half of the reported cases [69]. Dubovsky et al in 1995 mapped a gene (CCM1) associated with familial CMs to chromosome 7q11-q22 [71]. Additional genes have been found on chromosomes 7q, 7p, and 3q (KRIT1, CCM2, and CCM3) [72,73]. Clinical features Most patients with CMs are asymptomatic. CMs are more commonly identied in adults than in children. They may cause hemorrhage and focal neurologic decits [74]. Seizures are the presenting symptom in two thirds of the cases [75,76]; 40% of these may have refractory epilepsy [77]. CMs associated with venous angiomas are more likely to present with symptomatic hemorrhage [78]. Familial cases may also have retinal lesions and present with progressive visual loss [79]. Diagnostic studies CT of the head shows calcication in 50% of the patients. A low-density lesion with faint nonhomogenous enhancement is visualized on contrasted CT. MRI is more sensitive and shows a well-dened lesion with a central focus of mixed signal intensity surrounded by a rim of signal void caused by the paramagnetic eect of haemosiderin (Fig. 3) [76,80,81]. Angiography is normal in most of the cases or reveals an avascular mass [82]. Treatment and outcome The signicant risk for bleeding from CM is approximately 0.25% per person-year and many physicians reserve surgery for patients with intractable seizures or bleeding [66,83]. Management strategy for CM is complex and controversial. The general consensus is that microsurgical resection of the lesion is the best treatment in patients with intractable epilepsy. The success rate following surgical excision in patients with intractable seizures varies from 20%80% [77,8487]. The management of asymptomatic CM
1072
Fig. 3. T1-weighted magnetic resonance imaging (MRI) scan of a 5-year-old boy showing a right parietal cavernous angioma.
is conservative because of the low risk for bleeding. Radiosurgery may be an option for the treatment of epilepsy when the CM is located in an eloquent area. Venous angiomas Venous angiomas are neither true malformations nor tumors. They should properly be called developmental venous anomalies and represent anomalous but normal venous drainage [88]. They represent the most common type of intracranial vascular malformations documented by brain imaging and by autopsy, with prevalence as high as 3% [89]. Clinical features Venous angiomas are rarely symptomatic unless associated with another vascular malformation [78]. Rare cases presenting with seizures, headache, and intracerebral hemorrhage have been reported [90]. These patients often have mixed lesions with features of CM, venous angiomas, and AVM [66,90,91].
1073
Imaging studies Venous angiomas are often missed on CT without contrast. Contrastenhanced CT shows a hyperdense linear lesion perpendicular to the cerebral cortex. A hypodense lesion on T1-weighted sequences that enhances with gadolinium is seen on MRI [92]. The characteristic caput medusae pattern, corresponding to a large venous trunk fed by tributary veins is visualized in the late venous phase of conventional angiography [93]. Treatment and outcome Unless there is an associated CM or AVM, observation is indicated [66,92]. In cases of hemorrhage caused by mixed CM/venous angiomas, removal of the CM with preservation of the venous angioma may be necessary [88]. The venous angioma is a functional venous channel that drains normal parenchyma and the risk for venous infarction after surgery or radiosurgery is high. Radiosurgery has no role in the treatment of venous angiomas [88]. Summary Venous strokes are not as common as arterial strokes in the pediatric population, but may be associated with signicant mortality and morbidity. Cerebral vein thrombosis and venous sinus thrombosis are responsible for most venous strokes. Vein of Galen malformation is a rare but important cause of mortality in neonates and infants. Awareness of these potential causes of stroke in the pediatric population, early diagnosis, and appropriate therapeutic strategies are paramount to reduce mortality and improve neurologic outcome. References
[1] Bailey OT, Hass GM. Dural sinus thrombosis in early life: the clinical manifestations and extent of brain injury in acute sinus thrombosis. J Pediatr 1931;11:75571. [2] Carvalho KS, Bodensteiner JB, Connolly PJ, et al. Cerebral venous thrombosis in children. J Child Neurol 2001;16:57480. [3] Broderick J, Talbot GT, Prenger E, et al. Stroke in children within a major metropolitan area: the surprising importance of intracerebral hemorrhage. J Chil Neurol 1993;8:2505. [4] deVeber G, Adams M, Andrew M, et al. Canadian pediatric ischemic stroke registry (analysis III). Can J Neurol Sci 1995;22(Suppl):S214. [5] Ameri A, Bousser M. Cerebral venous thrombosis. Neurol Clin 1992;10:87. [6] DeMyer W. Blood supply of the central nervous system. In: Neuroanatomy. Baltimore: Williams & Wilkins; 1998. p. 395416. [7] Lin JH, McLean K, Morser J, et al. Modulation of glycosaminoglycan addition in naturally expressed and recombinant human thrombomodulin. J Biol Chem 1994;269:2502130. [8] Newton TH, Gooding CA. Compression or superior sagittal sinus by neonatal calvarial molding. Neuroradiol 1975;115:635. [9] Bergui M, Bradac GB, Daniele D. Brain lesions due to cerebral venous thrombosis do not correlate with sinus involvement. Neuroradiol 1999;41:41924.
1074
[10] deVeber G, Andrew M, Adams M, et al. Treatment of pediatric sinovenous thrombosis with low molecular weight heparin. Ann Neurol 1995;38(Suppl):S32. [11] Barron TF, Gusnard DA, Zimmerman RA, et al. Cerebral venous thrombosis in neonates and children. Pediatr Neurol 1992;8:1126. [12] Imai WK, Everhart FR, Sanders JM. Cerebral venous sinus thrombosis: report of a case and review of the literature. Pediatr 1982;70:96570. [13] Shevell MI, Silver K, OGorman AM, et al. Neonatal dural sinus thrombosis. Pediatr Neurol 1989;5:161. [14] Buonanno FS, Moody DM, Ball RM. CT scan ndings in cerebral sinovenous occlusion. Neurol 1982;12:28892. [15] Kriss VM. Hyperdense posterior falx in the neonate. Pediatr Radiol 1998;28:8179. [16] Virapongse C, Cazenave C, Quisling R, et al. The empty delta sign: frequency and signicance in 76 cases of dural sinus thrombosis. Radiol 1987;162:77985. [17] Medlock MD, Olivero WC, Hanigan WC, et al. Children with cerebral venous thrombosis diagnosed with magnetic resonance imaging and magnetic resonance angiography. Neurosurg 1992;31(5):8706. [18] Tyler HR, Clark DB. Cerebrovascular accidents in patients with congenital heart disease. Arch Neurol 1957;77:4839. [19] Graus F, Rogers LR, Posner JB. Cerebrovascular complications in patients with cancer. Medicine (Baltimore) 1985;64:1635. [20] Reddingius RE, Patte C, Couanet D, et al. Dural sinus thrombosis in children with cancer. Med Ped Oncol 1997;29(4):296302. [21] Uziel Y, Laxer RM, Blaser S, et al. Cerebral vein thrombosis in childhood systemic lupus erythematosus. J Ped 1995;126(5 Pt 1):7227. [22] Wiznitzer M, Masaryk TJ. Cerebrovascular abnormalities in pediatric stroke: assessment using parenchymal and angiographic magnetic resonance imaging. Ann Neurol 1991; 29:5859. [23] Buchanan DS, Brazinsky JH. Dural sinus and cerebral venous thrombosis. Incidence in young women receiving oral contraceptives. Arch Neurol 1970;22:4404. [24] Aschka I, Aumann V, Bergmann F, et al. Prevalence of factor V Leiden in children with thrombo-embolism. Eur J Ped 1996;155(12):100914. [25] Andrew M. Developmental haemostasis. Revelance to thrombo-embolic complications in paediatric patients. Thromb Haemostas 1995;74:41525. [26] Gouault-Heilmann M, Quetin P, Dreyfus M, et al. Massive thrombosis of venous cerebral sinuses in a 2-year-old boy with a combined inherited deciency of antithrombin III and protein C. Thromb Haemost 1994;72(5):7823. [27] Nowak-Gottl U, Koch HG, Aschka I, et al. Resistance to activated protein C (APCR) in children with venous or arterial thromboembolism. Br J Haematol 1996;92(4):9928. [28] Svesson PJ, Dahlback B. Resistance to activated protein C as a basis for venous thrombosis. N Engl J Med 1994;330:51722. [29] Pohl M, Zimmerhackl LB, Heinen F, et al. Bilateral renal vein thrombosis and venous sinus thrombosis in a neonate with factor V mutation (FV Leiden). J Pediatr 1998; 132(1):15961. [30] Hach-Wunderle V, Scharrer I, Lottenberg R. Congenital deciency of plasminogen and its relationship to venous thrombosis. Thromb Haemost 1988;59:27780. [31] Mammen EF. Congenital coagulation disorders. Dysbrinogenemia. Semin Thromb Hemost 1983;9:49. [32] deVeber G, Monagle P, Chan A, et al. Prothrombotic disorders in infants and children with cerebral thromboembolism. Arch Neurol 1998;55:153943. [33] Boggild MD, Sedhev RV, Fraser D, et al. Cerebral venous sinus thrombosis and antiphospholipid antibodies. Postgrad Med J 1995;71(838):4879. [34] Harum KH, Hoon AH Jr, Kato GJ, et al. Homozygous factor-V mutation as a genetic cause of perinatal thrombosis and cerebral palsy. Dev Med Child Neurol 1999;41(11):77780.
1075
[35] Levine SR, Brey RL. Neurological aspects of antiphospholipid antibody syndrome. Lupus 1996;5:34753. [36] Kerrin D, Murdoch Eaton D, Livingston J, et al. Homocystinuria presenting with sagittal sinus thrombosis in infancy. J Child Neurol 1996;11(1):701. [37] Andrew M, Vegh P, Johnston M, et al. Maturation of the hemostatic system during childhood. Blood 1992;80(8):19982005. [38] Kolenda J, Carr MM, Lemckert RJ, et al. Intracranial sinus thrombosis secondary to ear disease in an adolescent. J Otolaryngol 1997;26(3):2036. [39] Oyarzabal MF, Patel KS, Tolley NS. Bilateral acute mastoiditis complicated by lateral sinus thrombosis. J Laryng Otol 1992;106:5357. [40] deVeber G, Chan A, Monagle P, et al. Anticoagulation therapy in pediatric patient with sinovenous thrombosis: a cohort study. Arch Neurol 1998;55:15337. [41] Griesemer DA, Theodorou AA, Berg RA, et al. Local brinolysis in cerebral venous thrombosis. Pediatr Neurol 1994;10(1):7880. [42] Higashida RT, Helmer E, Halbach VV, et al. Direct thrombolytic therapy for superior sagittal sinus thrombosis. Am J Neuroradiol 1989;10(Suppl 5):S46. [43] Hanigan WC, Tracy PT, Tadros WS, et al. Neonatal cerebral venous thrombosis. Pediatr Neurosci 1988;14:17783. [44] Konishi Y, Kuriyama M, Sudo M, et al. Superior sagittal sinus thrombosis in neonates. Pediatr Neurol 1987;3(4):2225. [45] Preter M, Tzourio C, Ameri A, et al. Long-term prognosis in cerebral venous thrombosis: follow-up of 77 patients. Stroke 1996;27:2436. [46] Stein BM, Wolpert SM. Arteriovenous malformations of the brain I: current concepts and treatment. Arch Neurol 1980;37:15. [47] Golden GS. Stroke syndromes in childhood. Neurol Clin 1985;3(1):5975. [48] Knudson RP, Alden ER. Symptomatic arteriovenous malformation in infants less than 6 months of age. Pediatr 1979;64:23841. [49] Hendson L, Emery DJ, Phillipos EZ, et al. Persistent pulmonary hypertension of the newborn presenting as the primary manifestation of intracranial arteriovenous malformation of the Vein of Galen. Am J Perinatol 2000;17(8):40510. [50] Brunelle F. Arteriovenous malformation of the vein of Galen in children. Pediatr Radiol 1997;27(6):50113. [51] McElhinney DB, Halbach VV, Silverman NH, et al. Congenital cardiac anomalies with vein of Galen malformations in infants. Arch Dis Child 1998;78(6):54851. [52] Sepulveda W, Platt CC, Fisk NM. Prenatal diagnosis of cerebral arteriovenous malformation using color Doppler ultrasonography: case report and review of the litera ture. Ultrasound Obstet Gynecol 1995;6(4):2826. [53] Campi A, Scotti G, Filippi, et al. Antenatal diagnosis of vein of Galen aneurysmal malformation: MR study of fetal brain and postnatal follow-up. Neuroradiol 1996; 38(1):8790. [54] Lee TH, Shih JC, Peng SS, et al. Prenatal depiction of angioarchitecture of an aneurysm of the vein of Galen with three-dimensional color power angiography. Ultrasound Obstet Gynecol 2000;15(4):33740. [55] Sener RN. MR angiography of the vein of Galen malformation. Clin Imaging 1996; 20(4):2436. [56] Konus OL, Ilgit ET, Ozdemir A, et al. Spontaneous thrombosis of a vein of Galen aneurysmal malformation: possible eects of contrast media. Eur Radiol 2000;10(9):14568. [57] Nikas DC, Proctor MR, Scott RM, et al. Spontaneous thrombosis of vein of Galen aneurysmal malformation. Pediatr Neurosurg 1999;31(1):339. [58] Horowitz MB, Jungreis CA, Quisling RG, et al. Vein of Galen aneurysms: a review and current perspective. Am J Neuroradiol 1994;15(8):148696. [59] Halbach VV, Dowd CF, Higashida RT, et al. Endovascular treatment of mural-type vein of Galen malformations. J Neurosurg 1998;89(1):7480.
1076
[60] Mitchell PJ, Rosenfeld JV, Dargaville P, et al. Endovascular management of vein of Galen aneurysmal malformations presenting in the neonatal period. Am J Neuroradiol 2001; 22(7):14039. [61] Dowd CF, Halbach VV, Barnwell SL, et al. Transfemoral venous embolization of vein of Galen malformations. Am J Neuroradiol 1990;11(4):6438. [62] Payne BR, Prasad D, Steiner M, et al. Gamma surgery for vein of Galen malformations. J Neurosurg 2000;93(2):22936. [63] Johnston IH, Whittle IR, Besser M, et al. Vein of Galen malformation: diagnosis and management. Neurosurg 1987;20(5):74758. [64] Wiso JH, Friedman DM, Verma R, et al. Recent improvement in outcome using transcatheter embolization techniques for neonatal aneurysmal malformations of the vein of Galen. Pediatrics 1993;91(3):5836. [65] Friedman DM, Madrid M, Berenstein A, et al. Neonatal vein of Galen malformations: experience in developing a multidisciplinary approach using an embolization treatment protocol. Clin Pediatr 1991;30(11):6219. [66] Del Curling O Jr, Kelly DL Jr, Elster AD, et al. An analysis of the natural history of cavernous angiomas. J Neurosurg 1991;75(5):7028. [67] Robinson JR, Award LA, Little JR. Natural history of the cavernous angioma. J Neurosurg 1991;75:70914. [68] Sage MR, Brophy BP, Sweeney C, et al. Cavernous haemangiomas (angiomas) of the brain: clinically signicant lesions. Australas Radiol 1993;37(2):14755. [69] Rigamonti D, Hadley MN, Drayer BP, et al. Cerebral cavernous malformations. Incidence and familial occurrence. N Engl J Med 1988;319(6):3437. [70] Labauge P, Laberge S, Brunereau L, et al. Hereditary cerebral cavernous angiomas: clinical and genetic features in 57 French families. Societe Francaise de Neurochirurgie. Lancet 1998;352(9144):18927. [71] Dubovsky J, Zabramski JM, Kurth J, et al. A gene responsible for cavernous malformations of the brain maps to chromosome 7q. Hum Mol Gen 1995;4(3):4538. [72] Craig HD, Gunel M, Cepeda O, et al. Multilocus linkage identies two new loci for a Mendelian form of stroke, cerebral cavernous malformation, at 7p1513 and 3q25.227. Hum Mol Gen 1998;7(12):18518. [73] Laberge-le Couteulx S, Jung HH, Labauge P, et al. Truncating mutations in CCM1, encoding KRIT1, cause hereditary cavernous angiomas. Nat Genet 1999;23(2):18993. [74] Maraire JN, Awad IA. Intracranial cavernous malformations: lesion behavior and management strategies. Neurosurg 1995;37(4):591605. [75] Kattapong VJ, Hart BL, Davis LE, et al. Familial cerebral cavernous angiomas: clinical and radiologic studies. Neurol 1995;45(3 Pt 1):4927. [76] Simard JM, Garcia-Bengochea F, Ballinger WE, et al. Cavernous angioma: a review of 126 collected and 12 new clinical cases. Neurosurg 1986;18(2):16272. [77] Casazza M, Broggi G, Franzini A, et al. Supratentorial cavernous angiomas and epileptic seizures: preoperative course and postoperative outcome. Neurosurg 1996;39(1):2632. [78] Abdulrauf SI, Kaynar MY, Awad IA. A comparison of the clinical prole of cavernous malformations with and without associated venous malformations. Neurosurg 1999; 44(1):416, discussion 467. [79] Dobyns WB, Michels VV, Groover RV, et al. Familial cavernous malformations of the central nervous system and retina. Ann Neurol 1987;21(6):57883. [80] Imakita S, Nishimura T, Yamada N, et al. Cerebral vascular malformations: applications of magnetic resonance imaging to dierential diagnosis. Neuroradiol 1989;31(4): 3205. [81] Rigamonti D, Drayer BP, Johnson PC, et al. The MRI appearance of cavernous malformations (angiomas). J Neurosurg 1987;67(4):51824. [82] Becker DH, Townsend JJ, Kramer RA, et al. Occult cerebrovascular malformations. A series of 18 histologically veried cases with negative angiography. Brain 1979;102(2):24987.
1077
[83] Churchyard A, Khangure M, Grainger K, et al. Cerebral cavernous angioma: a potentially benign condition? Successful treatment in 16 cases. JNNP 1992;55(11):10405. [84] Acciarri N, Giulioni M, Padovani R, et al. Surgical management of cerebral cavernous angiomas causing epilepsy. J Neurosurg Sci 1995;39(1):1320. [85] Cohen DS, Zubay GP, Goodman RR. Seizure outcome after lesionectomy for cavernous malformations. J Neurosurg 1995;83(2):23742. [86] Dodick DW, Cascino GD, Meyer FB. Vascular malformations and intractable epilepsy: outcome after surgical treatment. Mayo Clin Proc 1994;69(8):7415. [87] Giulioni M, Acciarri N, Padovani R, et al. Results of surgery in children with cerebral cavernous angiomas causing epilepsy. Br J Neurosurg 1995;9(2):13541. [88] McLaughlin MR, Kondziolka D, Flickinger JC, et al. The prospective natural history of cerebral venous malformations. Neurosurg 1998;43(2):195200. [89] Sarwar M, McCormick WF. Intracerebral venous angioma. Case report and review. Arch Neurol 1978;35(5):3235. [90] Awad IA, Robinson JR Jr, Mohanty S, et al. Mixed vascular malformations of the brain: clinical and pathogenetic considerations. Neurosurg 1993;33(2):17988. [91] Ostertun B, Solymosi L. Magnetic resonance angiography of cerebral developmental venous anomalies: its role in dierential diagnosis. Neuroradiol 1993;35(2):97104. [92] Rigamonti D, Spetzler RF, Medina M, et al. Cerebral venous malformations. J Neurosurg 1990;73(4):5604. [93] Fierstein SB, Pribram HW, Hieshima G. Angiography and computed tomography in the evaluation of cerebral venous malformations. Neuroradiol 1979;17(3):13748.
Neurol Clin N Am 20 (2002) 10791100
Arterial strokes in children

Karen S. Carvalho, MD*, Bhuwan P. Garg, MBBS
James Whitcomb Riley Hospital for Children, Section of Pediatric Neurology, Indiana University Medical Center, 702 Barnhill Drive, Room #1757, Indianapolis, IN 46202-5200, USA
Stroke in children is more common than once suspected. Stroke is a catastrophic event in any individuals life and is especially tragic in children, with potential long-term disability and burden for the victims, families, and the community. The loss of full earning potential is incalculable for the individual and cumulatively enormous for society. The World Health Organizations MONICA Project denes stroke as rapidly developing clinical signs of focal (or global) disturbance of cerebral function, with symptoms lasting 24 hours or longer or leading to death with no apparent cause other than of vascular origin [1]. The denition includes ischemic and hemorrhagic infarction and intracerebral and subarachnoid hemorrhage. In adults, 80%85% of strokes are ischemic, whereas the remaining 15%20% are hemorrhagic. In children, 55% are believed to be ischemic, and the remainder hemorrhagic [2,3]. Strokes in adults are often caused by atherosclerosis; the etiology of stroke in children is diverse and challenging. Although many dierent causes and potential risk factors for childhood stroke have been described, ischemic stroke in children frequently results from cardiac embolism, nonatherosclerotic vasculopathies, or prothrombotic states. The main causes of nontraumatic hemorrhagic strokes are ruptured vascular malformations, bleeding diathesis, sympathomimetic drug abuse, or intracranial tumoral bleeding [4]. Epidemiology There is limited information regarding stroke epidemiology in children. Retrospective studies have generally reported stroke incidence of approximately 2.5 to 3.1 cases per 100,000 children per year [5,6]. The reported incidence of 0.2 strokes per 100,000 children per year among Japanese children
* Corresponding author. E-mail address: kcarvalh@iupui.edu (K.S. Carvalho). 0733-8619/02/$ - see front matter 2002, Elsevier Science (USA). All rights reserved. PII: S 0 7 3 3 - 8 6 1 9 ( 0 2 ) 0 0 0 1 2 - 9
1080
was low not only because of the low rates of congenital heart disease but also because strokes caused by Moyamoya disease, a leading cause of stroke in Japan, were excluded [7]. The higher rate of approximately 13 cases of stroke per 100,000 children per year, in the prospective French study, probably reected the general availability of CT scan [8]. Hemorrhagic strokes account for approximately half of all strokes in the pediatric population (Table 1). Recurrence is estimated to occur in 20% of the children [9].
Ischemic cerebral infarction Several factors increase the risk for ischemic stroke in children. Congenital heart disease remains an important risk factor. Other risk factors include atherosclerotic and nonatherosclerotic vasculopathies, hemoglobinopathies such as sickle cell disease, cerebral, and meningeal infections, prothrombotic states, and trauma (Box 1). A variety of genetic disorders have been implicated in stroke in children [10]. Some inherited diseases, such as the hereditary dyslipoproteinemias, predispose to accelerated atherosclerosis. Deciencies of protein C and S, antithrombin-III deciency, prothrombin G20210A mutant genotype, and activated protein C resistance are examples of inherited hematologic abnormalities associated with ischemic stroke [11,12]. Other inherited metabolic disorders such as the syndrome of mitochondrial encephalomyopathy, lactic acidosis, and stroke-like episodes (MELAS), Fabry disease, Menkes disease, Tangier disease, and homocystinuria may be associated with strokes. Some children may have multiple risk factors.
Congenital heart disease Cardioembolic stroke presumably accounts for nearly a fth to a third of all ischemic strokes with congenital heart disease being an important risk factor in children (Fig. 1) [13]. Congenital heart disease is reported to occur
Table 1 Age-specic incidence rates of cerebral infarctions in children per 100,000 per year Population Rochester, MNa Linkoping, Sweden Tohoku, Japanb Cincinnati, OHc Dijon, Franced
a b
Years 196574 197079 197489 198889 198593
Age group 014 yrs <15 yrs <16 yrs <15 yrs <16 yrs
Rate 2.5 2.1 NA 2.7 13.0
Ischemic 0.63 NA 0.2 1.2 7.9
Hemorrhagic 1.89 NA NA 1.5 5.1
Stroke related to birth, intracranial infection, or trauma excluded. Ischemic cerebral infarctions only. Excluded cases of Moyamoya. c Traumatic brain hemorrhage and germinal matrix hemorrhage excluded. d Prospective study. NA, Not available or not reported.
1081
Box 1. Risk factors for ischemic strokes in children Cardioembolic Congenital cardiac diseases Cardiac surgery Myocarditis Infection Central nervous system infection Sepsis/septic embolism Hemoglobinopathies Sickle cell disease Vascular abnormalities Moyamoya disease Fibromuscular dysplasia Vasculitis Isolated CNS angiitis SLE vasculatilis Arterial dissection Spontaneous Trauma Antiphospholipid antibodies Antithrombin III deciency Prothrombin gene 20210A mutation Dysbrinogenemia Factor V Leiden mutation Protein S deciency Protein C deciency Hyperhomocysteinemia/MTHFR gene mutation Elevated lipoprotein A Polycythemia Hyperlipidemia Extrinsic arterial compression Mitochondrial diseases Drugs Cocaine Methylphenidate Phenylpropanolamine
in approximately 510 per 1000 live births [14]. Heart disease may be implicated in cerebral infarction in one of four ways: there may be (1) a rightto-left shunt, (2) valvular disease, (3) endocardial disease, or (4) thrombus or tumor in the left heart. Cardiac embolization also may occur during cardiac surgery using cardiopulmonary bypass. Finally, presumed cardioembolic
1082
Fig. 1. A 3-year-old girl with history of congenital heart disease (transposition of great vessel and ventricular septal defect) presented with a right-sided hemiparesis, seizures, and decreased mental status. Top, weighted-diusion MR scan shows left anterior cerebral artery (arrows) and, bottom, right posterior cerebral artery (arrowheads) distribution ischemic infarcts.
1083
stroke in fact may be caused by pulmonary arteriovenous stula. Children with congenital heart disease and anemia are at high risk for cerebral infarction whereas sinovenous thrombosis is more likely in children who have polycythemia and high hematocrit. A right-to-left shunt may occur across the ventricles as in ventricular septal defect or Tetralogy of Fallot or across the atria as in atrial septal defect (ASD) or patent foramen ovale (PFO). Under normal circumstances the left heart pressures are greater than the right heart pressures and shunting of blood if any is from left to right. In some situations, however, there may be a reversal of this gradient as may occur during a Valsalva maneuver or when there is obstruction to the right ventricular outow, as may be the case in Tetralogy of Fallot caused by pulmonary stenosis and in some patients with pulmonary hypertension. A thrombus may therefore pass from the venous system to the arterial system by way of this defect in the heart, resulting in a cardioembolic stroke. ASD/PFO is an important consideration in cerebral infarcts and is often uncovered during investigations following a stroke. PFO may be a common occurrence; it has been found in 35% of people between the ages of 1 and 29 years [15]. PFO was present in 40% of patients with stroke compared with 10% in the control group [16]. Closure of PFO has been advocated in such patients. The PFO may close spontaneously in the rst few years of life and hence it may be appropriate to watch children with early onset stroke when a small PFO is presumed to be the only risk factor. Valvular disease may be congenital, as in a bicuspid aortic valve, or it may be acquired. Acquired valvular heart disease may be rheumatic, prosthetic, infective, inammatory, marantic, myxomatous, degenerative, or traumatic. Rheumatic heart disease is the most common cause of acquired valvular heart disease and mitral valve is the most commonly aected valve, followed by aortic and pulmonary valves. Presence of atrial brillation increases the risk for thromboembolism [101]. Infective endocarditis is an important cause of thromboembolism in patients with congenital and acquired cardiac valvular disease. Strokes can result from septic embolization, mycotic aneurysm formation, and vasculitis. Infective endocarditis involving the left side and larger vegetations are associated with a higher risk for embolism. Intravenous drug users are at an increased risk for developing infective endocarditis. Development of mycotic aneurysm is an important complication. Arteriography may be necessary to exclude a mycotic aneurysm in patients who develop focal signs. Patients with prosthetic heart valves have an increased risk for thromboembolism. Prosthetic heart valves may be mechanical or biologic. Mechanical valves have a higher risk for thromboembolism compared with biologic valves and they require lifetime anticoagulation. The risk for systemic thromboembolism in patients receiving anticoagulation is 4% with mitral and 2% with aortic position mechanical heart valves [17]. Infective endocarditis should be a strong consideration in any febrile stroke patient. Endocarditis such as Libman-Sacks endocarditis
1084
may be associated with strokes. Patients with antiphospholipid antibody syndrome have been reported with ndings resembling verrucous endocarditis of the mitral and aortic valves [18]. Cardiomyopathies secondary to valvular, ischemic, hypertensive, or inammatory heart disease are uncommon in children. Primary cardiomyopathies may be dilated, restrictive, hypertrophic, or obliterative. Cardiomyopathies may also be seen in a variety of metabolic and genetic childhood neurologic diseases such as Duchenne and Becker muscular dystrophy, myotonic dystrophy, Friedreich ataxia, infantile acid maltase deciency (Pompe disease), carnitine-palmitoyl transferase II deciency, Refsum disease, and neonatal and infantile mitochondrial disorders. Complete heart block is frequently seen in patients with Kearns-Sayre syndrome [19]. Thrombus in the left atrium or the left ventricle may result in cardioembolic stroke. Patients with rheumatic heart disease with mitral valve involvement and dilated left atrium are at risk for developing a thrombus in the left atrium and appendage. Risk is increased when this is complicated by the presence of atrial brillation. Thrombus in the left ventricle may follow myocardial infarction causing a dyskinetic-hypokinetic ventricular musculature. Kawasaki disease, polyarteritis nodosa, rheumatic carditis with coronary artery involvement, and congenital anomalies of the origin of the coronary arteries may place some children at risk for myocardial infarction. Risk for cerebral embolism is 1%3% in acute myocardial infarction [19]. Atrial myxoma is the most common primary cardiac tumor and may be familial. Rhabdomyomas are most often associated with tuberous sclerosis and are rarely associated with strokes. The Fontan operation is one of the most common cardiac operations for children with congenital heart disease after the rst year of age and essentially consists of an anastomosis of the right atrium to the pulmonary artery. In a series of 645 patients who underwent the Fontan procedure over a 15year period, 17 patients (2.6%) suered a stroke following surgery. The risk period for stroke extended from the rst postoperative day to 32 months following the Fontan procedure [20]. In another review of 68 patients who underwent the Fontan procedure, 6 of the 64 surviving patients had stroke. They all had normal hematologic and coagulation parameters at the onset of the neurologic symptoms. Two patients were on platelet antiaggregants, and one patient was on warfarin [21]. Congenital abnormalities of the aortic arch and aortic valve may be associated with cerebrovascular disease such as Moyamoya disease, arterial dissection, and aneurysms [22,23]. Strokes have been reported in patients with Williams syndrome [24]. Anticoagulant therapy is recommended for stroke patients with acute myocardial infarction, prosthetic and rheumatic valvular heart disease, and for those patients who have stroke from other noninfective cardiac sources of emboli. Prolonged parenteral antibiotic therapy is the mainstay of treatment for infective endocarditis. Selective surgical or transcatheter repair
1085
may be indicated for ASD. Optimal treatment of paradoxic embolism associated with PFO or ASA is currently unknown. Prompt surgical resection is indicated for patients with atrial myxomas [2528]. Cervicocephalic arterial dissection Cervicocephalic arterial dissection is an important risk factor for stroke in children. Dissection of the extracranial and intracranial portions of the carotid and vertebrobasilar arteries may occur spontaneously or may be secondary to trauma or other underlying vascular risk factors. Among 263 consecutive patients with spontaneous cervicocephalic arterial dissections evaluated at the Mayo Clinic, 18 (6.8%) were 18 years of age or younger [29]. In another study of 59 children who underwent conventional arteriography, 12 patients (20%) were found to have dissection of cervicocephalic arteries [30]. Initial symptoms may be nonspecic, with patients complaining of headache or neck pain followed by focal neurologic signs. There seems to be a male predominance; dissections aecting the carotid circulation have been reported more often than those involving the vertebrobasilar system [31]. History of trauma often is not forthcoming or trauma may be trivial such as a fall [32]. Advances in neuroimaging, particularly in noninvasive techniques such as MR angiography (MRA), have improved our ability to diagnose dissections of the cerebral arteries. MRA may miss the vascular abnormality, particularly in dissections involving the posterior circulation. Particular attention should be paid to the C1-C2 region, as that is the most common site of vertebral artery dissection [33]. Four-vessel conventional angiography remains the gold standard [34]. The risk for recurrent dissection in a study spanning 10 years in children and adults was 12% and seems to be particularly high in the few months immediately following presentation [29]. Anticoagulation may be helpful in appropriate cases though data are too limited to make rm recommendations. Mortality seems to be higher among patients with anterior circulation dissection, particularly intracranial dissections [35]. Moyamoya disease Moyamoya disease (MMD) is a nonatherosclerotic, noninammatory, vasculopathy characterized by chronic progressive stenosis or occlusion of the terminal internal carotid arteries and the proximal portions of the anterior cerebral arteries and middle cerebral arteries. MMD, more common among the Japanese, Korean, and Chinese populations, occurs in patients of all races and ethnic groups. It may be familial [36]. Some investigators have suggested that MMD is most likely inherited in a polygenic mode or in an autosomal dominant fashion with a low penetrance [37]. Recent studies have mapped the locus for familial MMD to chromosomes 3, 6, and 17 [3739]. MMD has been reported in association with a variety of conditions
1086
such as neonatal anoxia, trauma, basilar meningitis, tuberculous meningitis, leptospirosis, cranial irradiation therapy for optic pathway gliomas, neurobromatosis, tuberous sclerosis, brain tumors, bromuscular dysplasia, polyarteritis nodosa, Marfan syndrome, pseudoxanthoma elasticum, hypomelanosis of Ito, Williams syndrome, cerebral dissecting and saccular aneurysms, sickle cell anemia, b thalassemia, Fanconi anemia, Apert syndrome, factor XII deciency, type I glycogenosis, NADH-coenzyme Q reductase deciency, renal artery stenosis, Down syndrome, and coarctation of the aorta [40]. Aneurysms and arteriovenous malformations have been detected in up to 11% of patients [41]. Presenting symptoms of MMD include headaches, progressive cognitive decline, seizures, and strokes. Nearly half of the patients present before 10 years of age. MRI shows infarcts in multiple arterial distributions. MRI with contrast shows the collateral developed secondary to stenosis of the major arterial branches (Fig. 2). The angiographic features of MMD include (1) bilateral stenosis or occlusion of the supraclinoid portion of the internal carotid
Fig. 2. (A) A 10-year-old girl with history of chronic headaches and developmental delay presented with sudden onset left-sided weakness. FLAIR MRI scan demonstrates right middle cerebral artery (large arrows) and right anterior cerebral artery (open arrows) distribution ischemic infarcts. (B) Axial contrast enhanced T1-weighted MR scan shows enlarged collateral vessels (arrowheads).
1087
Fig. 2 (continued )
artery (ICA) that extends to the proximal portions of the anterior cerebral artery (ACA) and the middle cerebral artery (MCA), and (2) the presence of parenchymal collateral vessels, or Moyamoya vessels (MMVs), from the suprasellar cistern to the cerebral base [42]. MR angiography is an important diagnostic modality for the screening and longitudinal follow-up of MMD [43]. Four-vessel conventional angiography remains the gold standard for the diagnosis of Moyamoya disease. Positron emission tomography and SPECT are useful in the evaluation of perfusion reserve and help guide therapeutic approach [44]. Antiplatelet agents such as aspirin and ticlopidine and calcium channel antagonists (nimodipine and nicardipine) have been used for the medical treatment of MMD [45]. Multiple surgical procedures including encephalomyoarteriosynangiosis (EMAS), encephaloduro-arteriosynangiosis (EDAS), and encephalomyosynangiosis (EMS) have been advocated. Strict maintenance of normotension, euvolemia, and normocapnia, and possibly the use of nimodipine perioperatively, are important measures [40,44,46]. Early
1088
surgical intervention may be benecial, delaying a decline in cognitive performance and activities of daily living score [47,48]. Cerebral vasculitis Cerebral vasculitis is an uncommon cause of stroke in children. The authors found that 4% of strokes were attributable to vasculitis in children [49]. Schoenberg et al did not nd any cause of vasculitis in children younger than 14 years of age with cerebrovascular disease [6]. Cerebral vasculitis may be infectious or noninfectious. Varicella, bacterial and tuberculous meningitis, and human immunodeciency virus (HIV) are associated with stroke (Fig. 3) [35,5054]. Noninfectious cerebral vasculitis may be idiopathic or secondary to a variety of systemic illnesses such as Behcet disease, sarcoidosis, Sjogren syndrome, ulcerative colitis, Kawasaki disease, and Henoch-Schonlein Purpura (HSP), among others [55,56].
Fig. 3. An 8-month-old girl with bacterial meningitis and sudden-onset left-sided hemiparesis and seizures. Axial head CT without contrast shows a large right middle cerebral artery territory ischemic infarct (arrows).
1089
Cerebral vasculitis should be considered when the stroke is associated with systemic manifestations such as persistent fever, weight loss, myalgias, arthralgias, renal disease, and skin lesions. It may present with encephalopathy, meningeal signs, and multiple focal neurologic decits. Stroke and other neurologic symptoms might be the initial manifestation of vasculitis or complicate the course of a previously diagnosed systemic illness. Primary angiitis of the central nervous system (PACNS) is an idiopathic vasculitis conned to the CNS. PACNS most commonly involves small vessels, although involvement of medium and large cerebral vessels has been reported [57]. Clinical course is gradual and variable, with headaches, focal seizures, multifocal neurologic decits, and neurobehavioral impairment, reecting multifocal or diuse brain damage. Neuroimaging reveals multiple cerebral ischemic lesions. Cerebral spinal uid might show elevated protein and mild pleocytosis but is nonspecic. Cerebral angiogram may show multiple vessel involvement or may be normal. Meningeal and brain biopsy may be diagnostic in some cases. Corticosteroids and cyclophosphomide are mainstays of treatment. Outcome is variable [55]. Radiation vasculopathy Radiation-induced vasculopathy may involve large or small vessels and is conned to the area of irradiation. Delayed capillary endothelial damage leads to breakdown of the bloodbrain barrier, vasogenic edema, endothelial hyperplasia, and brinoid necrosis of penetrating arterioles in the white matter of the CNS [58]. Large vessels show vessel wall brosis and evidence of accelerated atherosclerosis. Foulardi et al found that 25 of 421 consecutive children with CNS tumors treated with radiation therapy developed clinically silent lacunar strokes [59]. A Moyamoya-like picture following radiation therapy of optic pathway gliomas has been reported [60]. Drugs Illicit drugs are a rare but increasingly recognized cause of stroke in teenagers and young adults. In one case-control study of patients aged 1544 years, after controlling for identiable risk factors such as cardiac disease, diabetes mellitus, hypertension, smoking, alcohol use, and pregnancy, and excluding cases caused by endocarditis, the estimated overall relative risk for stroke among young adult illicit drug users was 6.5. The relative risk was even higher at 11.2 in patients younger than 35 years [61]. Cerebral infarcts and hemorrhage have been reported in patients abusing drugs such as amphetamines, ecstasy, cocaine, phencyclidine (PCP), and glue sning, among others. Suggested mechanisms include transient cerebral vasoconstriction, unmasking of a pre-existing cardiovascular disease, toxic vasculitis, and prothrombotic tendencies [62].
1090
Sickle cell disease Sickle cell disease (SCD) is common in the African American population with an estimated incidence of 1 in 400. Strokes occur in nearly 10% of patients with SCD and almost half of these will have a second stroke [63]. There is a high incidence of rst-time stroke among younger children, with most strokes occurring before 10 years of age [64]. Cerebral infarction is caused primarily by an occlusive arteriopathy involving the distal intracranial segments of the internal carotid artery, and proximal anterior and middle cerebral arteries. Transcranial Doppler (TCD) is valuable in detecting arterial stenosis in patients with SCD with velocities of 200 cm/sec or higher associated with a high risk for cerebral infarction [65,66]. There is a drastic reduction in the risk for a rst stroke in children with SCD who had abnormal results on TCD and are placed in a transfusion program [67]. Long-term transfusion therapy was associated with a reduced risk for stroke recurrence to as low as 10% and has become routine after stroke in children with SCD [68]. It is unclear how long transfusion should be continued as a means of preventing stroke in children with sickle cell anemia. Allogenic bone marrow transplantation (BMT) has been used in children with SCD [69]. There are substantial barriers to BMT, however, that include transplant-related toxicity and lack of a suitable human leukocyte antigen (HLA)-matched family donor that preclude its widespread application to patients with SCD [70].
Genetic and metabolic causes A variety of genetic and metabolic disorders may be associated with strokes in children. Conditions associated with cardioembolic strokes include mitral valve prolapse, Marfan syndrome, tuberous sclerosis complex when cardiac rhabdomyoma is present, hereditary cardiac conduction defects, and hereditary cardiomyopathies [71]. A variety of hereditary coagulation disorders may also predispose children to strokes, as discussed later. Occipital lobes are the most common site of stroke in MELAS. The strokes in MELAS often do not conform to a vascular territory. Several other metabolic disorders including Leigh disease, organic and amino-acidemias, Sneddon syndrome, carbohydrate-deciency glycoprotein syndrome, and carnitine deciency have been reported to have stroke-like symptoms [7274]. Homocystinuria is an autosomal-recessive disorder characterized by multiple abnormalities including mental retardation, lens dislocation, and predisposition to cerebrovascular disease. Classic homocystinuria results from deciency of cystathionine b-synthase enzyme activity. Combined homocystinuria and methylmalonic acidemia and a variant form caused by deciency of methylenetetrahydrofolate reductase enzyme also are believed to increase the risk for strokes [75]. Hyperhomocysteinemia may also result from nutritional deciency of folic acid or vitamin B12.
1091
Prothrombotic disorders Prothrombotic disorders are an important risk factor for stroke in children. A prothrombotic state is an impairment of the normal hemostatic system in which the balance has shifted toward thrombosis. It is most commonly caused by an abnormality or impairment of the vascular endothelium, the coagulation cascade, the brinolytic system, or the platelets. Cerebral infarction has been reported in children who have deciency of protein C, protein S, antithrombin III, and plasminogen, or the presence of activated protein C resistance (APCR). Presence of mutations such as Factor V Leiden mutation, prothrombin G20210A mutation, homozygous methylenetetrahydrofolate dehydrogenase (MTHFR) polymorphism and antiphospholipid antibodies, hyperhomocysteinemia, and elevated lipoprotein (a) are other risk factors [76,77]. Protein C, protein S, and antithrombin III deciency may be inherited or acquired. Infection, medications such Lasparaginase, or hepatic and renal disease are the most common causes of acquired deciency. Factor V Leiden mutation occurs in 5%12% of the general pediatric population. Heterozygous carriers have a sevenfold increased risk for thrombosis, whereas the risk for homozygous individuals is increased 80fold [78]. In a prospective blood bank survey, antiphospholipid antibodies were detected in approximately 6.5% of normal subjects [79]. Children with antiphospholipid antibodies have a signicantly increased risk for arterial and venous thrombosis, of which 50% occur in the central nervous system [80]. Lipoprotein (a) is an LDL-like lipoprotein that is usually associated with hypercholesterolemia and is an independent risk factor [81]. Lipoprotein (a) levels are increased in children with arterial strokes when compared with control subjects [82]. The overall incidence of prothrombotic states in children with ischemic strokes is reported to be 10%50% [80,82,83]. All children with cerebral infarction should be evaluated for the presence of a prothrombotic state. A suggested workup is outlined in Box 2.
Atherosclerotic cerebrovascular disease Atherosclerotic cerebrovascular disease is rare in children. Dyslipidemic states may hasten this process. In a large retrospective study, lipid abnormalities including elevated triglycerides and low-density lipoprotein cholesterol, and a depressed high-density lipoprotein cholesterol were seen in one third of 42 children with ischemic stroke [84]. More studies are needed to investigate the signicance of increased levels of lipoprotein (a) in childhood stroke. Other risk factors such as hypertension, hyperinsulinemia or insulin resistance, and obesity often coexist [85]. Hypertension is an important and independent risk factor [86]. Diabetes mellitus is another important risk
1092
Box 2. Coagulation work up in children with stroke Complete blood count including platelets count Prothrombin time and activated partial thromboplastin time Plasminogen level Fibrinogen level Protein C activity Protein S activity Activated protein C resistance Antithrombin III activity Antiphospholipid antibodies panel Homocysteine level, folate level, vitamin B12 level as indicated Factor V Leiden gene mutation Prothrombin 20210A gene mutation Methylene tetrahydrofolate reductase (MTHFR) polymorphism Lipoprotein a level Fasting lipid prole
factor in adults but is rarely associated with strokes in children. There has not been a systematic study of the eect of glucose-tolerance abnormalities in childhood stroke [87].
Treatment and outcome in children with arterial ischemic stroke Determination of the etiology of cerebral infarction in childhood is important because it strongly inuences outcome, recurrence risk, and treatment of choice [30]. There are no multicenter controlled clinical trials addressing the ecacy and safety of primary and secondary prevention of stroke in children except those associated with sickle cell disease. This paucity of clinical studies is attributable to a variety of factors: (1) cerebrovascular disorders in children are rare, (2) the diversity and overlapping of risk factors create a heterogeneous patient population, and (3) children tend to present to medical attention with stroke later than adults, making dicult the use of acute treatment such as thrombolytics [88]. Intervention in children with stroke usually follows extrapolation from randomized trial in adults, but the ecacy and safety of applying adult population data in children is not established [89]. When feasible, surgical correction of a cardiac defect is important to prevent recurrence. Anticoagulation is necessary in patients with a demonstrable thrombus either in the heart or in the vessels, as may be seen in a cervicocephalic dissection, because of the high risk for embolic stroke. Some children with a prothrombotic state may also require anticoagulation. Heparin, low
1093
molecular weight heparins, and warfarin are the most commonly used drugs for anticoagulation. Systemic anticoagulation has been controversial and ecacy has not been established in prospective pediatric studies. Intravascular thrombolysis is controversial and ecacy has not been established in pediatric patients. Thrombolytic agents such as urokinase and streptokinase promote clot lysis by converting plasminogen to plasmin, but also stimulate plasmin-mediated proteolysis of blood coagulation factors [90]. The use of these agents results in a high risk for catastrophic hemorrhage. The tissue plasminogen activator (t-PA) is clot-selective, has a shorter half-life, and produces lower levels of brinogen degradation products, but no trial has demonstrated acceptable safety in the pediatric stroke population. The delay in diagnosis that so often occurs in children with ischemic stroke reduces the likelihood that a child with an ischemic stroke will be seen early enough to benet from thrombolytic agents [91]. Repeated blood transfusions to keep the sickle cell level low are important in the primary and secondary prevention of stroke in children with SCD [67]. Moyamoya disease is best treated surgically, though others and the authors have also used calcium channel blocking drugs in some patients. Corticosteroids are often the mainstay of treatment in noninfectious vasculitis. Supplementation with folic acid and vitamin B12 and the use of cholesterol lowering agents may be appropriate in some children. Aspirin has been used in children despite lack of data supporting its ecacy in preventing stroke in the pediatric population.
Hemorrhagic strokes Germinal matrix hemorrhage is the most common intracranial hemorrhage in premature infants. Nontraumatic intracranial hemorrhage and that caused by coagulation factor defects has been described in term infants [92,93]. Trauma is the most common cause of intracranial hemorrhage in children. Thrombocytopenia also is an important risk factor for intracranial hemorrhage [94]. Structural vascular lesions are an important risk factor in children with spontaneous intraparenchymal hemorrhage [5,6]. Arteriovenous malformations are the most common cause of nontraumatic intracranial hemorrhage and should always be suspected in children with unexplained parenchymal hematoma. Aneurysms are a common cause of hemorrhage in adults but are rare in children. Other risk factors include hereditary or acquired hematologic disorders, drugs, and neoplastic disease.
Arteriovenous malformations Arteriovenous malformations (AVMs) of the brain are congenital highow vascular lesions. AVMs are characterized by a direct stulous connection between the arteries and veins without an intervening capillary bed.
1094
An AVM was found to be the cause of hemorrhage in one third of 68 children studied by Al-Jarallah et al [94]. Patients most commonly present with hemorrhage or seizures (Fig. 4); less often headache, mass eect, or focal neurologic decits are the presenting symptoms. Natural history studies suggest that the average risk for bleeding is 2%4% per year. The risk for death with an initial rupture of the brain AVM is 10% and the risk increases with each subsequent hemorrhage. Complete surgical resection is the ideal treatment and eliminates the risk for recurrent hemorrhage. This is not always possible, however, because of the size or location of the AVM. Patients may be treated by endovascular and stereotactic radiosurgery that sometimes may be followed by surgical extirpation [95]. Hereditary and acquired coagulation disorders Factor VII and factor VIII deciency have been associated with intracranial hemorrhage in children [92,96]. Intracranial hemorrhage may aect 2.6%13.8% of hemophiliacs and is one of the major causes of death among hemophiliacs. Diagnosis of hemophilia is often overlooked in children presenting with intracranial hemorrhage, particularly if there is a history of head trauma [92].
Fig. 4. (A) An 8-year-old boy presented with an acute left hemiparesis, severe headache, and nuchal rigidity. Axial head CT without contrast shows a right frontoparietal hemorrhage. (B) Right internal carotid angiogram, midarterial phase, lateral view, disclosed an arteriovenous malformation (arrow).
1095
Fig. 4 (continued )
Congenital abrinogenemia is a rare autosomal recessive disorder with an estimated incidence of two per one million births. The condition is characterized by a complete absence of brinogen (coagulation factor I), thus impairing brin clot formation. The incidence of spontaneous intracerebral bleeding in patients with abrinogenemia is unknown. Recurrent hemorrhage in patients with abrinogenemia may be fatal. Fibrinogen level must be checked in children presenting with spontaneous or traumatic intracranial hemorrhage [97]. Intracranial hemorrhage also has been reported with a variety of other hereditary and acquired coagulopathies including Factor V Leiden mutation, Vitamin K deciency, and renal or liver failure resulting in secondary coagulation factor deciency [98100]. Miscellaneous Hypertension is a common cause of intracranial hemorrhage in adults but not in children. In a series of 68 patients, intracranial hemorrhage could not be attributed to systemic arterial hypertension in any patient [94]. Intracranial hemorrhage may complicate an intracranial tumor. Intracranial hemorrhage may also occur during treatment of childhood malignancies. Intracranial hemorrhage may be the presentation of Moyamoya disease especially in young adults [40]. Hemorrhagic transformation of an ischemic stroke may occur, and broadens the dierential diagnosis of brain hemorrhage to include all of the risk factors for ischemic infarction [94].
1096
Diagnostic approach in children with arterial hemorrhagic stroke A complete medical history including detailed family history is important to identify inherited risk factors for intracranial hemorrhages. A comprehensive coagulation workup should be performed, particularly in the presence of family history of recurrent thrombosis or hemorrhage. CT of the head is the best imaging study to identify acute hemorrhage. Workup also should include MRI, magnetic resonance angiography, and conventional four-vessel angiography in selected cases, especially when a clear risk factor for hemorrhage cannot be established or for patients in whom there is a reason to suspect multiple risk factors. The best time to obtain conventional angiography is unclear, because it might be normal in the acute phase owing to the tamponade eect with the hematoma occluding the blood vessels. In this case angiography should be repeated in a few weeks. Medical treatment is mainly supportive with adequate oxygenation, ventilation, and hydration. Hypertension should be aggressively treated and secondary hypoxia avoided. Mannitol and steroids are used to decrease intracranial pressure. Surgical hematoma removal might be required, particularly if signicant mass eect is present. AVMs usually require surgical intervention. Outcome is variable depending on the cause, localization, and size of the hemorrhage.
References
[1] Thorvaldsen P, Asplund K, Kuulusmaa K, et al. Stroke incidence, case fatality, and mortality in the WHO MONICA Project. World Health Organization Monitoring Trends and Determinants in Cardiovascular Disease. Stroke 1995;26:36167. [2] Daniels SR, Bates S, Lukin RR, et al. Cerebrovascular arteriopathy (arteriosclerosis) and ischemic childhood stroke. Stroke 1982;13:36065. [3] Nencini P, Inzitari D, Baru MC, et al. Incidence of stroke in young adults in Florence, Italy. Stroke 1988;19:97781. [4] Garg BP, Durocher A, Biller J. Strokes in children and young adults. In: Ginsberg MD, Bogousslavsky J, editors. Cerebrovascular disease: pathophysiology, diagnosis, and management. Malden, Massachusetts: Blackwell Science; 1998. [5] Broderick J, Talbot T, Prenger E, et al. Stroke in children within a major metropolitan area: the surprising importance of intracerebral hemorrhage. J Child Neurol 1993;8:2505. [6] Schoenberg BS, Mellingier JF, Shoenberg DG. Cerebrovascular disease in infants and children: a study of incidence, clinical features and survival. Neurol 1978;28:7638. [7] Satoh S, Shirane R, Yoshimoto T. Clinical survey of ischemic cerebrovascular disease in children in a district of Japan. Stroke 1991;22:58689. [8] Giroud M, Lemesle M, Gouyon JB, et al. Cerebrovascular disease in children under 16 years of age in the city of Dijon, France: a study of incidence and clinical features from 1985 to 1993. J Clin Epidemiol 1995;48:134348. [9] Isler W. Stroke in childhood and adolescence. Eur Neurol 1984;23(6):4214. [10] Pullicino P, Greenberg S, Trevisan M. Genetic stroke risk factors. Curr Opin Neurol 1997; 10:5863. [11] Chaturvedi S, Dzieczkowski J. Multiple hemostatic abnormalities in young adults with activated protein C resistance and cerebral ischemia. J Neurol Sci 1998;159:20912.
1097
[12] deStefano V, Chiusolo P, Paciaroni K, et al. Prothrombin G20210A mutant genotype is a risk factor for cerebrovascular disease in young patients. Blood 1998;91:356265. [13] Riela A, Roach E. Etiology of stroke in children. J Child Neurol 1993;8:20120. [14] Mitchell SC, Korones SB, Berendes HW. Congenital heart disease in 56,109 births. Incidence and natural history. Circulation 1971;43:32332. [15] Moss AJ, Adams FH Jr. Atrial septal defects. In: Emmanouilides GC, editor. Heart disease in infants, children, and adolescents, including the fetus and young adults. 5th edition. Baltimore: Williams and Wilkins; 1995. p. 687703. [16] Lechat P, Mas JL, Lascault G, et al. Prevalence of patent foramen ovale in patients with stroke. N Engl J Med 1988;318:114852. [17] Vongptanasin W, Hillis LD, Lange RA. Medical progress: prosthetic heart valves. N Engl J Med 1996;345:40716. [18] Biller J, Challa VR, Toole JF, et al. Nonbacterial thrombotic endocarditis. A neurologic perspective of clinicopathologic correlations of 99 patients. Arch Neurol 1982;39(2):958. [19] Biller J, Garg BP, Rasura M. Epidemiology and etiology of stroke in the young: new therapeutic perspectives for prevention. In: Fieschi C, Fisher M, editors. Prevention of ischemic stroke. London: Martin Dunitz; 2000. p. 79103. [20] du Plessis AJ, Chang AC, Wessel DL, et al. Cerebrovascular accidents following the Fontan operation. Pediatr Neurol 1995;12:2306. [21] Day RW, Boyer RS, Tait VF, et al. Factors associated with stroke following the Fontan procedure. Pediatr Cardiol 1995;16:2705. [22] Lutterman J, Scott M, Nass R, et al. Moyamoya syndrome associated with congenital heart disease. Pediatr 1998;101:5760. [23] Schievink WI, Mokri B, Piepgras DG, et al. Intracranial aneurysms and cervicocephalic arterial dissections associated with congenital heart disease. Neurosurg 1996;39:68590. [24] Wollack JB, Kaifer M, LaMonte MP, et al. Stroke in Williams syndrome. Stroke 1996;27: 1436. [25] Caplan LR. Prevention of cardioembolic stroke. Heart Dis Stroke 1994;3(6):297303. [26] Ende DJ, Chopra RS, Rao PS. Transcatheter closure of atrial septal defect or patent foramen ovale with the buttoned device for prevention of recurrence of paradoxic embolism. Am J Cardiol 1996;78:23336. [27] Gu M, Bogousslavsky J, Jeanrenaud X, et al. Surgical prophylaxis of recurrent stroke in patients with patent foramen ovale: a pilot study. J Thorac Cardiovasc Surg 1996;112: 26063. [28] Wilson DG, Wiseheart JD, Stuart AG. Systemic thromboembolism leading to myocardial infarction and stroke after fenestrated total cavopulmonary connection. Br Heart J 1995; 73:48385. [29] Schievink WI, Mokri B, Piepgras DG. Spontaneous dissections of the cervicocephalic arteries in childhood and adolescence. Neurol 1994;44:160712. [30] Cabrier S, Husson B, Lasjaunias P, et al. Stroke in childhood: outcome and recurrence risk by mechanism in 59 patients. J Child Neurol 2000;15:2904. [31] Fullerton HJ, Johnston SC, Smith WS. Arterial dissection and stroke in children. Neurol 2001;57:115560. [32] Garg BP, Edwards-Brown MK. Vertebral artery compression due to head rotation in thalamic stroke. Pediatr Neurol 1995;12(2):1624. [33] Garg BP, Ottinger CJ, Smith RR, et al. Strokes in children due to vertebral artery trauma. Neurol 1993;43(12):25558. [34] Ganesan V, Savvy L, Chong WK, et al. Conventional cerebral angiography in children with ischemic stroke. Ped Neurol 1999;20:3842. [35] Fugate S, Roos KL. Stroke associated with infections including immunizations. Semin Cerebrovasc Dis Stroke 2001;1:2408. [36] Hamada JI, Yoshioka S, Nakahara T, et al. Clinical features of Moyamoya disease in sibling relations under 15 years of age. Acta Neurochir (Wien) 1998;140:45558.
1098
[37] Yamauchi T, Mitsuhiro T, Kiyohiro H, et al. Linkage of familial Moyamoya disease (spontaneous occlusion of the circle of Willis) to chromosome 17q25. Stroke 2000;31:9305. [38] Ikeda H, Sasaki T, Yoshimoto T, et al. Mapping of a familial Moyamoya disease gene to chromosome 3p24.2-p26. Am J Hum Gen 1999;64:5337. [39] Inoue TK, Ikezaki K, Sasazuki T, et al. Linkage analysis of Moyamoya disease on chromosome 6. J Child Neurol 2000;15:17982. [40] Garg BP, Bruno A, Biller J. Moyamoya disease and cerebral ischemia. In: Batjer HH, editor. Cerebrovascular disease. Philadelphia: Lippincott-Raven Publishers; 1997. p. 48999. [41] Chiu D, Shedden P, Bratina P, et al. Clinical features of Moyamoya disease in the United States. Stroke 1998;29:134751. [42] Takahashi M. Magnication angiography in Moyamoya disease. New observations on collateral vessels. Radiol 1980;136:37986. [43] Yamada I, Nakagawa T, Matsushima Y, et al. High-resolution turbo magnetic resonance angiography for diagnosis of Moyamoya disease. Stroke 2001;32:182531. [44] Ikezaki K. Rational approach to treatment of Moyamoya disease in childhood. J Child Neurol 2000;15:3506. [45] Hosain SA, Hughes JT, Forem SL, et al. Use of a calcium channel blocker (nicardipine HCl) in the treatment of childhood Moyamoya disease. J Child Neurol 1994;9:37880. [46] Smith JL, Scott RM. Treatment of Moyamoya syndrome in children. Sem Cerebrovasc Dis Stroke 2001;1:22539. [47] Matsushima Y, Aoyagi M, Masaoka H, et al. Mental outcome following encephalodurosynangiosis in children with Moyamoya disease with the onset earlier than 5 years of age. Childs Nerv Syst 1990;6:4403. [48] Matsushima Y, Aoyagi M, Nariai T, et al. Long-term intelligence outcome of postencephalo-duro-arterio-synangiosis in childhood Moyamoya patients. Clin Neurol Neurosurg 1997;99(Suppl):S14750. [49] Williams LS, Garg BP, Cohen M, et al. Subtypes of ischemic stroke in children and young adults. Neurol 1997;49(6):15415. [50] Berkefeld J, Enzensberger W, Lanfermann H. MRI in human immunodeciency virusassociated cerebral vasculitis. Neuroradiol 2000;42:5268. [51] Ganesan V, Kirkham FJ. Mechanisms of ischaemic stroke after chickenpox. Arch Dis Child 1997;76:5225. [52] Hayman M, Hendson G, Poskitt KJ, et al. Postvaricella angiopathy: report of a case with pathologic correlation. Pediatr Neurol 2001;24:3879. [53] Park YD, Belman AL, Kim TS, et al. Stroke in pediatric acquired immunodeciency syndrome. Ann Neurol 1990;28:30311. [54] Snyder RD, Stovring J, Cushing AH, et al. Cerebral infarction in childhood bacterial meningitis. JNNP 1981;44:5815. [55] Lopez-Yunez AM, Garg BP. Non-infectious cerebral vasculitis in children. Sem Cerebrovasc Dis Stroke 2001;1:24963. [56] Sokol DK, McIntyre JA, Short RA, et al. Henoch-Schonlein purpura and stroke: antiphosphatidylethanolamine antibody in CSF and serum. Neurol 2000;55(9):137981. [57] Lanthier S, Lortie A, Michaud J, et al. Isolated angiitis of the CNS in children. Neurol 2001;56:83742. [58] Baker DG, Krochak RJ. The response of the microvascular system to radiation: a review. Cancer Invest 1989;7:28794. [59] Fouladi M, Langston J, Mulhern R, et al. Silent lacunar lesions detected by magnetic resonance imaging of children with brain tumors: a late sequelae of therapy. J Clin Oncol 2000;18:82431. [60] Serdaroglu A, Simsek F, Gucuyener K, et al. Moyamoya syndrome after radiation therapy for optic pathway glioma: case report. J Child Neurol 2000;15:7657. [61] Kaku DA, Lowenstein DH. Emergence of recreational drug abuse as a major risk factor for stroke in young adults. Ann Int Med 1990;113:8217.
1099
[62] Sloan MA, Kittner SJ, Feeser BR, et al. Illicit drug-associated ischemic stroke in the Baltimore-Washington Young Stroke Study. Neurol 1998;50:168893. [63] Balkaran B, Char G, Morris JS, et al. Stroke in a cohort of patients with homozygous sickle cell disease. J Pediatr 1992;120(3):3606. [64] Ohene-Frempony K. Stroke in sickle cell disease. Demographic, clinical, and therapeutic considerations. Sem Hematol 1991;28:2139. [65] Adams RJ, McKie VC, Carl EM, et al. Long-term stroke risk in children with sickle cell disease screened with transcranial Doppler. Ann Neurol 1997;42:699704. [66] Verlhac S, Bernaudin F, Tortrat D, et al. Detection of cerebrovascular disease in patients with sickle cell disease using transcranial Doppler sonography: correlation with MRI, MRA and conventional angiography. Pediatr Radiol 1995;25(Suppl 1):S149. [67] Adams RJ, McKie VC, Hsu L, et al. Prevention of a rst stroke by transfusions in children with sickle cell anemia and abnormal results on transcranial Doppler ultrasonography. N Engl J Med 1998;339:511. [68] Adams RJ. Stroke prevention and treatment in sickle cell disease. Arch Neurol 2001;58: 5658. [69] Walters MC, Storb R, Patience M, et al. Impact of bone marrow transplantation for symptomatic sickle cell disease: an interim report. Multicenter investigation of bone marrow transplantation for sickle cell disease. Blood 2000;95:191824. [70] Hoppe CC, Walters MC. Bone marrow transplantation in sickle cell anemia. Curr Op Oncol 2001;13:8590. [71] Pavlakis SG, Kingsley PB, Bialer MG. Stroke in children: genetic and metabolic issues. J Child Neurol 2000;15:30815. [72] Thompson JE, Smith M, Castillo M, et al. MR in children with L-carnitine deciency. Am J Neuroradiol 1996;17:15858. [73] Wheeler PG, Medina S, Dusick A, et al. Livedo reticularis, developmental delay and stroke-like episode in a 7-year-old male. Clin Dysmorphol 1998;7:6974. [74] Young G, Driscoll MC. Coagulation abnormalities in the carbohydrate-decient glycoprotein syndrome: case report and review of the literature. Am J Hematol 1999;60:669. [75] Topic E, Timundic AM, Ttefanovic M, et al. Polymorphism of apoprotein E (APOE), methylenetetrahydrofolate reductase (MTHFR) and paraoxonase (PON1) genes in patients with cerebrovascular disease. Clin Chem Lab Med 2001;39:34650. [76] Chan AKC, deVeber G. Prothrombotic disorders and ischemic stroke in children. Sem Pediatr Neurol 2000;7:3018. [77] Walsh LE, Garg BP. Ischemic strokes in children. Indian J Pediatr 1997;64:61323. [78] Pohl M, Zimmerbackl LB, Heinen F, et al. Bilateral renal vein thrombosis and venous sinus thrombosis in a neonate with factor V mutation (FV Leiden). J Pediatr 1998;132:15961. [79] Levine SR, Brey RL. Neurological aspects of antiphospholipid antibody syndrome. Lupus 1996;5:34753. [80] deVeber G, Monagle P, Chan A, et al. Prothrombotic disorders in infants and children with cerebral thromboembolism. Arch Neurol 1998;55:153943. [81] Scanu AM. Atherothrombogenicity of lipoprotein (a): the debate. Am J Cardiol 1998; 82:Q2633. [82] Nowak-Gottl U, Strater R, Heinecke A, et al. Lipoprotein (a) and genetic polymorphisms of clotting factor V, prothrombin, and methylenetetrahydrofolate reductase are risk factors of spontaneous ischemic stroke in childhood. Blood 1999;94:367882. [83] Ganesan V, McShane MA, Liesner R, et al. Inherited prothrombotic states and ischaemic stroke in childhood. JNNP 1998;65:50811. [84] Abram HS, Knepper LE, Warty VS, et al. Natural history, prognosis, and lipid abnormalities of idiopathic ischemic childhood stroke. J Child Neurol 1996;11:27682. [85] Berenson GS, Srinivasan SR, Wattigney WA. Association between multiple cardiovascular risk factors and atherosclerosis in children and young adults. N Engl J Med 1998;338:16506.
1100
[86] MacMahon S, Peto R, Cutler J, et al. Blood pressure, stroke, and coronary heart disease, 1: prolonged dierences in blood pressure: prospective observational studies corrected for the regression dilution bias. Lancet 1990;335:76574. [87] Kirkham FJ, Prengler M, Hewes DKM, et al. Risk factors for arterial ischemic stroke in children. J Child Neurol 2000;15:299307. [88] Roach ES, Biller J. Treatment of ischemic stroke in children. Sem Cerebrovasc Dis Stroke 2001;1:26470. [89] Andrew M, Marzinotto V, Brooker LA, et al. Oral anticoagulation therapy in pediatric patients: a prospective study. Thromb Haem 1994;71:2659. [90] Griesemer DA, Theodorou AA, Berg RA, et al. Local brinolysis in cerebral venous thrombosis. Pediatr Neurol 1994;10:7880. [91] deVeber G, Roach ES, Riela AR, et al. Stroke in children: recognition, treatment, and future directions. Sem Pediatr Neurology 2000;7:30917. [92] Bray GL, Luban NL. Hemophilia presenting with intracranial hemorrhage: an approach to the infant with intracranial bleeding and coagulopathy. Am J Dis Child 1987;141: 12157. [93] Cartwright GW, Culbertson K, Schreiner RL, et al. Changes in clinical presentation of term infants with intracranial hemorrhage. Dev Med Child Neurol 1979;21:7307. [94] Al-Jarallah A, Al-Rifai MT, Riela AR, et al. Nontraumatic brain hemorrhage in children: etiology and presentation. J Child Neurol 2000;15:2849. [95] Smith JL, Garg BP. Treatment of arteriovenous malformations of the brain. Curr Neurol Neurosc Rep 2002;2:449. [96] Ries M, Wolfel D, Maier-Brandt B. Severe intracranial hemorrhage in a newborn infant with transplacental transfer of an acquired factor VII: C inhibitor. J Pediatr 1995;127: 64950. [97] Henselmans JM, Meijer K, Haaxma R, et al. Recurrent spontaneous intracerebral hemorrhage in a congenitally abrinogenemic patient: diagnostic pitfalls and therapeutic options. Stroke 1999;30:247982. [98] Citak A, Emre S, Sairin A, et al. Hemostatic problems and thromboembolic complications in nephrotic children. Pediatr Nephrol 2000;14:13842. [99] Melegh B, Stankovics J, Kis A, et al. Increased prevalence of factor V Leiden mutation in neonatal intracranial haemorrhage. Eur J Pediatr 1998;157:261. [100] Suzuki K, Fukushima T, Meguro K, et al. Intracranial hemorrhage in an infant owing to vitamin K deciency despite prophylaxis. Childs Nerv Syst 1999;15:2924. [101] Anonymous. Cardiogenic brain embolism. The second report of the Cerebral Embolism Task Force. Arch Neurol 1989;46(7):72743.
Neurol Clin N Am 20 (2002) 11011124
Movement disorders in children and adolescents

William M. McMahon, MDa,*, Francis M. Filloux, MDb, James C. Ashworth, MDc, Jenise Jensen, MSd
a
Departments of Psychiatry and Pediatrics, University of Utah; Salt Lake City, UT, USA b Division of Pediatric Neurology, Departments of Pediatrics and Neurology, University of Utah and Primary Childrens Medical Center; Salt Lake City, UT, USA c Department of Psychiatry, University of Utah and Primary Childrens Medical Center, Salt Lake City, UT, USA d Department of Psychiatry, University of Utah, Salt Lake City, UT, USA
Tourette syndrome (TS) and Sydenham chorea (SC) are prototypical movement disorders that have onset in children. Once considered distinct in etiology and clinical presentation, TS and SC may have overlapping pathophysiology involving common neural, genetic, infectious, and immune factors. Obsessions and compulsions co-occur with both, adding cognitive symptoms to these two disorders of movement. Moreover, the boundary between the two conditions has become blurred by the description of a putative new clinical entity known as pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections (PANDAS). All three of these conditions seem to involve the circuitry of the basal ganglia, and all may be associated with abnormal regulation of dopamine and/or serotonin. A fourth type of movement disorder, drug-induced dyskinesia, is rare in children but may become more germane as dopamine agonists and antagonists are increasingly prescribed for children. This article reviews current knowledge regarding TS, SC, PANDAS, and drug-induced dyskinesia in children. Children and adolescents may experience other movement disorders, such as neuroacanthocytosis, Wilson disease, Huntington disease, benign familial chorea, myoclonus, and a number of disorders associated with
This work was supported in part by grant MH58858 to Dr McMahon. * Corresponding author. University of Utah Clinics, 546 Chipeta Way, Suite 441, Salt Lake City, UT 84108, USA. E-mail address: william.mcmahon@hsc.utah.edu (W.M. McMahon). 0733-8619/02/$ - see front matter 2002, Elsevier Science (USA). All rights reserved. PII: S 0 7 3 3 - 8 6 1 9 ( 0 2 ) 0 0 0 1 5 - 4
1102
W.M. McMahon et al / Neurol Clin N Am 20 (2002) 11011124
tremor. These have been subjects of recent reports [1,2] and are outside the scope of this review.
Tics and Tourette syndrome Tics are sudden, stereotyped, coordinated, and involuntary movements or sounds. They may be simple, such as a blink or sni. Alternatively, tics may be complex, such as gesticulating or uttering words, phrases, or sentences. Although more commonly rapid and brief, tics may also present as sustained expressions, movements, or postures that may be referred to as dystonic tics. Tics may be preceded by premonitory sensations of tension that build until the tic occurs, with a feeling of relief accompanying the occurrence of the tic. Tics may be suppressible for brief periods of time, leading parents, teachers, or others to erroneously believe that the tics are willful or that a child can eliminate tics by trying hard. Tics may sometimes cause self-injury by hitting, biting, poking, or other complex actions. They may also result in pain or injury to muscles or joints because of repetitive or violent movements. Once thought common, coprolalia (the involuntary utterance of forbidden or insulting words, phrases, or sentences) is rare and is sometimes dicult to distinguish from a psychogenic or pseudo tic. The major problems caused specically by tics are usually psychologic and social, resulting from social intolerance and an erosion of the childs sense of self. Tics resulting in injury, pain, or disruption of normal activities deserve pharmacotherapy. Treatment of the child or adolescent with TS must often target symptoms secondary to other neuropsychiatric disorders, including obsessive-compulsive disorder (OCD) or attention-decit/hyperactivity disorder (ADHD). Tourette syndrome and the Tourette spectrum disorders In 1885, a French neurologist published a description of nine cases with a syndrome characterized by multiple motor and vocal tics, echolalia, and coprolalia [3]. The neurologist was Gilles de la Tourette, a student of Charcot at the Salpetriere Hospital. Tourette emphasized the childhood onset, waxing and waning course, and hereditary nature of TS. He also considered the syndrome to be rare and lifelong in duration, concepts that have recently been shown to be incorrect. Two overlapping systems of classication of tic disorders have recently been published: the American Psychiatric Associations Diagnostic and Statistical Manual of Mental Disorders, 4th edition, Text Revision (DSMIV-TR) [4] and the Tourette Syndrome Classication Study Groups system [5]. Both systems treat tic disorders as spectrum disorders with TS at the most severe end of the spectrum. The specic classes of tic disorders in both systems dier from each other on clinical grounds (eg, presence of motor
1103
or vocal tics, duration and course of tics). The Tourette Syndrome Study Group makes a further distinction between tic disorders based only on reported history versus tic disorders based on combined history and observation of tics by an examiner. Evidence supporting distinctions between classes of tic disorders is sparse, but the arbitrary distinctions seem to have utility for clinical and research purposes. In the DSM-IV-TR, the transient tic of childhood is dened by a duration of less than 12 months, duration of at least 4 weeks, and inclusion of either single or multiple motor and/or vocal tics. Chronic motor or vocal tics may also be single or multiple but are conned to one class (motor or vocal). Duration must be at least 1 year, with no tic-free period of more than 3 consecutive months. The DSM-IV-TR [4] denition of Tourette disorder requires the combination of motor and vocal tics, duration of at least 1 year with no tic-free interval longer that 4 months, onset before the age of 18 years, and exclusion of other causes (direct physiologic eects of a substance [eg, stimulant drugs] or a general medical condition [eg, Huntington disease or postviral encephalitis]). In contrast to the DSM-IV [6], however, the newer denition does not require marked distress or signicant impairment. Tic disorder not otherwise specied is the category for cases not fullling the required duration or age at onset criteria. Epidemiology of tics and tic disorders Gilles de la Tourette estimated the population prevalence of TS as one case per million. Modern studies all agree that the rate is much higher, but dierences in methodology give a wide range in prevalence estimates. Scahill et al [7] recently summarized 10 published studies of TS prevalence, estimating the true rate to be 10 to 30 cases per 10,000 children. In general, higher rates have been found in younger age groups, supporting the notion that many cases remit partially or completely by adulthood. Chronic tic disorders are more common than TS, occurring in 2% to 5% of school-aged children [7]. Transient tics seem to be at least as common as chronic tics and may be underestimated because of their transient and perhaps more subtle nature [7]. TS is more common in males than females, with male/female ratio ranging from a high of 10:1 [8] to a low of 2:1 [9,10]. TS also seems to be more common in children than in adults. A study of children and adults with TS in North Dakota resulted in estimates of 9.3 per 10,000 for boys and 0.77 per 10,000 for men, giving boys a prevalence 10 times greater than that of men [11]. Similarly, the rate for North Dakota girls, 1 per 10,000, was higher than the rate for women, 0.22 per 10,000. Tic onset is more common before 7 years of age, and tic severity usually peaks between the ages of 7 and 15 years [12,13]. The temporal clustering of tics within bursts seems to be a fractal pattern that is apparent at dierent time scales (eg, seconds, hours, days, months) [14].
1104
Causes of tics and Tourette syndrome Jankovic [15] recently reviewed the causes of tics, subgrouping them into primary and secondary tic disorders. Although the transient tics of childhood and chronic single tics may occur sporadically, they also seem to be alternative expressions of genes conferring risk for the Tourette spectrum. Multiple genes on dierent chromosomes operating singly or in combination with other genes and/or environmental factors likely contribute to dierences in the onset and severity of TS spectrum disorders. Molecular linkage studies suggest that loci on chromosomes 4, 8, and 11 are involved [16,17]. More recently, a reanalysis of the Tourette Syndrome Association (TSA) sib pair sample using OCD symptom factors and a recursive partitioning analytic technique found evidence for hoarding symptoms on chromosomes 4, 5, and 17 [18]. Other idiopathic or hereditary disorders may manifest tics, including Huntington disease, primary dystonia, neuroacanthocytosis, HallervordenSpatz disease, Wilson disease, and tuberous sclerosis. Tics may also be secondary to a long list of infections, drugs, toxins, or developmental disorders. SC and drug-induced tics, discussed in detail below, are examples of secondary disorders, although genetic vulnerability still seems to be an important factor. Perturbation of corticostriatothalamocortical (CSTC) circuits, basal ganglia, and dopaminergic overactivity seem to be common neuropathogenic factors that result in tics [19]. Both pathologic tics and adaptive habits that link sensory cues with motor action apparently arise from the complex parallel CSTC circuits [20]. Autopsy studies have implicated an array of neurochemicals, including decreased 5-hydroxytryptamine, glutamate, cyclic AMP, dopamine reuptake sites, and dynorphin-like immunoreactivity [2124]. Overall, neuroimaging studies of adults consistently implicate the basal ganglia. Much of the current literature may not be applicable to children, however, and may reect eects of drug treatment, development, and neuropsychiatric comorbidity [25]. A recent volumetric MRI study of children and adults with TS implicates larger volumes in dorsal prefrontal and parieto-occipital regions and smaller inferior occipital volumes compared with controls [26]. Peterson and colleagues [26] hypothesize that tics result from inhibitory decits in a broadly distributed action-attentional system involving these brain regions. Smaller volumes in the orbitofrontal and parieto-occipital regions were associated with worse symptoms, possibly resulting in insucient ability to inhibit tics. In contrast, the larger prefrontal volumes were hypothesized to be caused by an activity-dependent structural plasticity reecting active tic suppression. Fredericksen and colleagues [27] identied an increase in right frontal white matter, which may suggest a failure of axonal pruning. Functional imaging studies (positron emission tomography [PET], single photon emission computed tomography [SPECT], and functional MRI [fMRI]) have detected decreased blood ow in the basal ganglia, especially
1105
in the ventral striatum [2834]. Decreased blood ow of the frontostriatal pathway has been associated with symptom severity [27]. The most consistent dierences between groups have been found in the temporolimbic system, which includes the anterior cingulate, parahippocampal, and possibly insular cortices and could be involved with such behavioral functions as impulse control, reward contingency, and executive functions [33,35]. Brain regions involved in attentional functioning (frontal, temporal, and superior temporal association cortices) have also demonstrated signicant dierences between TS and normal control samples [35]. Radioligand studies have provided modest, although less conclusive, evidence for abnormal dopaminergic activity in the basal ganglia [3639]. Most ndings are sex specic to males and suggestive that females have a dierent phenotypic morphology [40,41]. This may reect neurodevelopmental dierences between the sexes [4247]. Treatment of Tourette syndrome There are three basic principles when considering treatment of TS or another tic disorder. First, the tics require treatment. Parental education, reassurance, and advocacy may be all that is required with less severe tic disorders, which often disappear by young adulthood. For mild to moderate cases of TS, education of the child and family about tics should be the rst treatment goal. Increasing tolerance of tics by the child, parents, teachers, coaches, and peers may be the only treatment needed. Tic-suppressing medication (Table 1) is indicated when tics interfere with activities, lead to ostracism or avoidance of social settings, or cause selfinjury. Tics that require pharmacologic intervention should be suciently severe that they overwhelm the ability to cope and aect the patients social or emotional functioning. The decision to begin tic-suppressing medication requires a thorough assessment of the childs tics, including the family and social context. A discussion of the target symptoms is helpful so that the child and parents realize what symptoms are likely to improve. Equally important is an explicit explanation of the limitations and risks for the medication chosen. For example, haloperidol or pimozide, when given alone, is unlikely to improve OCD or ADHD. Furthermore, the patient and the parents should understand that the goal of treating tics with tic-suppressing medication is to make tics tolerable and not to eliminate them. Therefore, improved coping is always a treatment goal. Second, use the least amount of medication that works. Although this is a good axiom to apply to all patients, it is particularly important for the young TS patient who may require potent medications with serious side eects. The corollary of this approach is to use enough medication to make a dierence. Insucient amounts of a good medication may be worse than no medication at all. The patient receives all the risks without compensatory benets. Thus, drug treatment requires repeated assessment, particularly as the severity of the disorder waxes and wanes.
1106
Table 1 Medical Management of Tics Medication Haloperidol Primary indication Suppression of tics Dose 0.25 mg to 35 mg/day, qd to tid Common side eects
Pimozide
Suppression of tics
Risperidone
Suppression of tics, calming of behavior Suppression of tics, calming of behavior Same as clonidine
Clonidine Clonidine patch Guanfacine
Imipramine Clonazepam
Calming of behavior; possibly suppression of tics Attentional diculties in person with tics Suppression of tics
Weight gain, lethargy, anhedonia, depression, EPS 0.5 mg to 68 mg/day, Weight gain, lethargy, qd to tid anhedonia, depression, EPS, prolonged QT 0.25 mg to 6 mg/day, Sedation, fatigue, qhs, bid to tid weight gain, tremor, milder EPS 0.025 mg to 0.1 mg tid Sedation, postural hypotension rare TTS-1 patch weekly; Same as clonidine increase to TTS-2 or sedation usually less TTS-3 patch as needed with patch 0.5 mg to 1 mg tid Same as clonidine
Tetrabenazinea Suppression of tics
10 to 75 mg/day qhs or bid to tid 0.25 mg qd to tid to 1 mg tid 12.5 mg to 75 mg/day
Pergolide
Suppression of tics
25 mcg qhs, increase to 50 mcg tid as needed
Sedation, tremor, fatigue, irritability Sedation, irritability, behavior change Depression, sedation, anxiety, insomnia, akathisia Minimal dizzyness, sedation, nausea
qd, each day; tid, three times daily; qhs, at bedtime; bid, twice daily; TTS, Transdermal Therapeutic System; EPS, extrapyramidal movement syndromes. Local botulinum toxin injection has been utilized as an alternative treatment for motor tics. a Not marketed in the US.
Third, comorbid disorders may seriously interfere in the patients life, although the tics themselves have little impact. A parent may enter the physicians oce with the goal of eradicating the newly discovered tics in a child, although the associated OCD, ADHD, or behavioral problems are the major causes of dysfunction. Physicians must assess the whole patient to recognize or exclude other disorders or psychosocial factors that contribute to distress and dysfunction. ADHD frequently co-occurs with TS, as does OCD [48,49]. Anxiety disorders, mood disorders, pervasive developmental disorders, mental retardation, and specic learning disorders have all been reported as comorbid, although such co-occurrence could represent an ascertainment bias rather than true comorbidity. Substance abuse (eg, cocaine, amphetamines) or the use of anabolic steroids can exacerbate tics. Haloperidol, rst reported to be an eective tic-suppressing medicine in the 1960s, is still a mainstay of therapy [5053]. Shapiro et al [54] argued
1107
that the possibility of eective treatment using haloperidol gave rise to increasing awareness of the disorder among patients and physicians. Pimozide, a dopamine (D2)-receptor antagonist, is currently the only US Food and Drug Administration (FDA)approved drug for the treatment of TS in children and adolescents. Both haloperidol and pimozide are widely used. Studies comparing haloperidol and pimozide have provided conicting results [55,56]. Fluphenazine has also been used with reported benet [57]. This medication can cause dystonias and dyskinesias. Newer antipsychotics have been tried with some reported success, but controlled studies are rare. A controlled study of clozapine (Clozaril) resulted in an exacerbation of tics at low doses and a lack of ecacy in reducing tics [58]. A comparative, doubleblind, parallel-group study of risperidone and pimozide has recently shown ecacy for both drugs in treating patients with TS [59]. Patients receiving risperidone had fewer extrapyramidal side eects. Ziprasidone was eective and tolerable in a randomized controlled trial of children and adolescents with TS [55]. Clonidine (Catapress) has been used frequently in children with TS. Mixed results indicate that this a-adrenergic receptor agonist has less ticsuppressing potency than the neuroleptics. It may be helpful for improving impulse control in comorbid ADHD, but the frequent sedation and short half-life can be problematic for the pill form, and contact dermatitis often limits use of the skin patch form of clonidine. Guanfacine, an a2-adrenergic cousin to clonidine that oers a longer half-life with fewer side eects, produced improvement in one controlled trial [60]. Pergolide is an ergot alkaloid derivative used in the management of Parkinson disease. Pergolide is a potent dopaminergic agonist at both D1 and D2 dopamine receptors. In a recent randomized, double-blind, crossover study, pergolide was shown to be an eective treatment for TS in children [61]. The medication was well tolerated with no serious side eects. Botulinum toxin (Botox) has been used to treat motor and vocal tics, and a recent double-blind, controlled, clinical trial demonstrated ecacy [62]. Both the frequency of the tic and its associated urge were decreased. Remarkably, patients did not report an overall benet from the treatment in this study, and the authors urged careful consideration when deciding whether a tic should be treated with Botox. In an earlier clinical trial, a baclofen/botulinum toxin type A preparation was used and found to be eective in treating TS [63]. Baclofen alone may benet children with TS, as suggested by the results of a recent double-blind, placebo-controlled, crossover trial [64]. Clinical management of Tourette syndrome Because of the variety of cognitive, behavioral, and emotional diculties associated with TS, clinicians play an important role in assisting parents and
1108
school professionals to support the childs academic, social, and developmental functioning. In mild to moderately severe cases, support and patience from teachers and parents may be all that is necessary. Educating teachers about classroom accommodations, such as allowing the child to leave the classroom to relieve tics or to take tests privately so as to decrease anxiety can be helpful. Many children with TS have visuomotor diculties that could be helped by allowing the use of a computer for written assignments, providing photocopies of classroom notes and daily assignments, and permitting oral tests. Clinical suspicion of a possible specic learning disability should lead to referral to a school psychologist for further academic assessment. Furthermore, collaboration with school professionals is recommended to determine the childs baseline functioning and to monitor response to any intervention [65]. Finally, the Tourette Syndrome Association has an excellent Internet site (http://www.tsa-usa.org) that provides parents, teachers, and professionals with information on educational materials, support services, and related web pages.
Sydenham chorea Generally considered a monophasic illness in most aected individuals, SC is characterized by the subacute onset of a movement disorder that generally remits over a period of weeks to months [6671]. Generalized chorea of variable severity is typical. The face, tongue, and upper extremities are usually aected, more or less symmetrically in 80% to 90% of cases. Socalled paralytic chorea, prominent hypotonia, and motor impairment simulating muscle weakness occur rarely (<10% of cases), whereas predominant unilateral involvement is evident in up to 40% of the recent Utah cases. The clinical features of SC include the typical movement disorder, hypotonia, dysarthria, apparent muscle weakness, motor impersistence, and restlessness or hyperkinesis [6770]. There are often associated emotional and behavioral disturbances [70,72], and writing diculties in school are common. Onset of SC, as with rheumatic fever (RF) in general, typically occurs between 8 and 12 years of age. There is a female predominance [6770]. Specic manifestations of the movement disorder and the associated motor impersistence include the milk-maids grip (inability to sustain continuous even hand grip pressure on the ngers of the examiner) and a darting tongue (erratic retraction, protrusion, and deviation of the tongue caused by chorea of aected muscles). The movement disorder disappears in sleep and is often intensied on intention. Children often unconsciously attempt to control the movements of the aected limbs (by constraining their movement using body position or other limbs [eg, sitting on their hands]) or consciously disguise the choreiform movements in semivoluntary acts.
1109
The emotional disturbance accompanying SC has long been recognized and classically has been said to include emotional lability, distractibility, and inattention as well as nightmares and night terrors. Obsessive-compulsive symptoms or clinical OCD may accompany SC. Swedo and her colleagues [70,73] reported that 82% of patients with SC had concomitant onset of obsessions and compulsions and that one third met requirements for OCD. More recently, Brazilian children with either SC (n 20) or RF without SC (n 22) were found to have increased rates of OCD (13.6% and 10%, respectively) compared with control children (n 20 [0%]) [74]. This Brazilian study also found increased rates of ADHD and major depressive disorder (MDD) in both SC and RF cases compared with controls. Interestingly, the ADHD and MDD rates were even signicantly higher in SC compared with RF. Furthermore, the frequency of tics or TS was also increased in both RF and SC, with the rate in SC being even higher than in RF. A mail survey of 65 Utah SC cases and 35 non-SC RF cases found evidence for OCD in 25% of SC cases compared with 9% of non-SC RF cases (unpublished data). These reports support the hypothesis that similar or related basal ganglia circuitry may be involved in SC, OCD, and other childhood neuropsychiatric disorders. Virtually all cases of SC resolve spontaneously with time over weeks to months. Recurrences may occur in up to 20%, and more than one recurrence is possible in a given individual [75,76]. The overlap between SC and related immune-mediated causes of chorea (eg, the primary antiphospholipid antibody syndrome or systemic lupus erythematosus) may account for some of these unusual recurrences or relapsing and remitting cases in individuals properly treated with prophylactic antibiotics [77,78]. In addition, pregnancy and oral contraceptive use may result in the re-emergence of chorea in previously aected women (eg, chorea gravidarum). Alterations in cerebral dopamine and other neurotransmitter receptors by sex steroids may provide one mechanism whereby this transient reappearance of chorea may be mediated. Relation to rheumatic fever SC is one of the major manifestations of rheumatic fever according to the original and modied Jones criteria (along with carditis, erythema marginatum, arthritis, and subcutaneous nodules) [79,80]. Unlike these other major features, however, SC generally follows streptococcal pharyngitis by a prolonged interval of weeks to 9 months [67]. In addition, in recent outbreaks in the United States, SC has occurred as a solitary manifestation of RF in up to 40% of instances [8183]. The long interval between streptococcal infection and appearance of chorea also results in a lower rate of positive serologic test results using antistreptolysin O titer (ASO) titer. This makes identication of rheumatic chorea dicult in many cases. Reliance on the ASO titer alone for serologic conrmation of prior infection is suboptimal,
1110
because the anti-DNase B titer persists for a longer interval after group A streptococcal pharyngitis and thus provides more reliable serologic evidence of previous streptococcal infection [84]. In recent outbreaks in Utah, clinically evident carditis was identied in approximately 30% of individuals with SC [82,83]. Other major manifestations of RF were more rare. Prior streptococcal infection had often passed unrecognized and therefore had gone untreated. Veasy and his colleagues [82,83] identied what they considered to be signicant subclinical mitral regurgitation on echocardiography in most children with chorea believed to be otherwise consistent with SC (ie, with other proximate causes excluded). If this silent regurgitation is accepted as a reliable indicator of rheumatic mitral valve involvement [8590], most individuals with putative SC in the recent Utah outbreak have carditis. The addition of silent regurgitation as a criterion for RF has facilitated the diagnosis of SC in Utah children with serologic evidence of streptococcal disease and absence of other major manifestations of RF [91]. Laboratory evaluation The dierential diagnosis of SC includes systemic lupus erythematosus [9294], the primary antiphospholipid antibody syndrome [95100], Wilson disease, neuroacanthocytosis, Huntington disease, benign familial chorea, other paroxysmal movement disorders, encephalitis [101], acute disseminated encephalomyelitis (ADEM), drug intoxication, and focal basal ganglia lesions caused by stroke, tumor, or even mitochondrial cytopathies. The rst step in diagnosis is a careful history and physical examination seeking to establish the occurrence of other major manifestations of RF and a history of prior streptococcal infection (recent culture or current ASO/ anti-DNase B assays). Secondary evaluation in some children with chorea and suspected SC includes the following: serum electrolytes, calcium, glucose, thyroid function tests, complete blood cell count, sedimentation rate, serum copper and ceruloplasmin, antinuclear antibodies, and anticardiolipin antibodies (IgG and IgM). The latter are often elevated in the setting of SC [78] but may also be elevated in the so-called primary antiphospholipid antibody syndrome [96] or in systemic lupus erythematosus (which may present with chorea before the appearance of other systemic disease) [93]. MRI of the brain is generally indicated and is usually normal (see comments below). Identication of serum antineuronal antibodies has been used as a research tool (see comments below), but such assays have limited clinical roles currently. The same applies to detection of the D8/17 antigen (discussed below).
Immunopathogenesis of Sydenham chorea: new developments The last two decades have witnessed a burgeoning of interest in the immunopathogenesis of RF, SC, and potentially related conditions. Most
1111
new developments have arisen from studies of the putative immunopathogenetic mechanisms of disease combined with novel neuroimaging methods. Following the hypothesis that molecular mimicry [102] may account for the occurrence of tissue injury in RF, Husby and colleagues [103] rst identied antineuronal antibodies directed against human cerebral tissue in the serum taken from children with SC [104106]. Little subsequent progress was made, however, until the recent resurgence of RF and SC in the United States [81,83] and the emergence of the PANDAS hypothesis [107]. Several subsequent studies have identied autoantibodies directed against human basal ganglia in children with SC [70,108110], but neither the sensitivity nor the specicity of the assay seems to be particularly high in many of these studies. Furthermore, similar rates of antineuronal antibodies have been identied in a variety of clinical settings (subsumed under the acronym PANDAS), including children with OCD [111113], children with tic disorders [114,115], and particularly children with ADHD and tics [109]. Likewise, a recent study describes and even better characterizes autoantibodies directed against cerebral tissue antigens in children with ADEM believed to be secondary to acute streptococcal infections (so-called PSADEM) [101]. Thus, the presence of autoantibodies directed against cerebral tissue antigens (particularly against basal ganglia) seems not to be specic to SC. These ndings suggest that similar mechanisms of molecular mimicry may underlie a variety of immune-mediated poststreptococcal neurologic disorders [110,116]. The eect of putative immune-mediated targeting of basal ganglia structures has been recently investigated using MRI morphometry and functional imaging (SPECT). Several limited case series have suggested transient T2 signal abnormalities within the caudate and putamen of some children with SC [117122]. This seems to be a rare nding in our experience and one that might be dicult to replicate. Quantitative studies, however, have identied subtle but statistically signicant transient increases in the volume of the basal ganglia during symptomatic SC, followed by a decrement in volume after resolution of the clinical illness [123]. These ndings have been proposed to reect subtle inammation or edema of these structures, presumably secondary to immune-mediated injury. Similar ndings have been reported in OCD and the PANDAS subgroups [124,125] but not in TS [126]. Likewise, transient hypermetabolism in the contralateral basal ganglia has been clearly demonstrated in several children with rheumatic hemichorea, with resolution of the hypermetabolism after clinical remission [66,69,119,127129]. Thus, a neurophysiologic correlate to the observed movement disorder may be demonstrable using functional neuroimaging techniques, indicating dysfunction of the extrapyramidal system in SC. No volumetric changes were found in the total cerebral, prefrontal, midfrontal, or thalamic cortices.
1112
Treatment of Sydenham chorea No report on a controlled treatment trial has been published regarding the management of SC. Approaches to treatment can be either symptomatic or immunomodulatory [67,68,130]. The most important strategy is to prevent recurrence by prophylactic antibiotic management. Some controversy exists regarding the optimal interval for administration of intramuscular benzathine penicillin, with some investigators promoting a 3-week interval rather than the standard 4-week interval [131]. Consensus on this issue remains to be reached [75,80]. The reader is referred to the 2000 Red Book published by the American Academy of Pediatrics [132] for current recommendations regarding antibiotic prophylaxis. Because SC remits spontaneously in most patients, treatment should be reserved for more severely aected children. Children who fall frequently or cannot feed themselves are candidates for drug treatment. Symptomatic treatment includes most medications used for the treatment of chorea and other hyperkinetic movement disorders in other circumstances, including neuroleptics (primarily haloperidol), benzodiazepines (clonazepam and clorazepate), valproic acid, depleters of synaptic dopamine (eg, reserpine and tetrabenazine), and perhaps baclofen. This wide array of options underscores the often unsatisfactory clinical response to such agents. The risk of tardive dyskinesia from haloperidol, although present, is probably remote. Immunomodulatory therapy may include oral steroids, intravenous immunoglobulin, and plasmapheresis, although the latter two are expensive and are not endorsed for general clinical application even by the investigators reporting on their use [133]. Benet from plasmapheresis and intravenous immunoglobulin in SC has been reported [130], and this response lends credence to the immunopathogenic hypothesis of SC. Since the resurgence of SC in the intermountain western United States in 1985, pediatric neurologists at the University of Utah have considered prednisone to be the drug of rst choice in SC of sucient severity to warrant treatment [134]. The drug is given at a dose of 1 to 2 mg/kg/d for 7 to 10 days, followed by a gradual taper over a similar period. Most children respond well, although the natural history of the disease makes assessment of ecacy difcult without a controlled trial. Occasional patients have seemed to relapse during or after reduction in the prednisone dose but have improved again with a second course. No serious complications have been encountered.
Pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections The acronym PANDAS is based on cases of OCD and tics studied at the National Institute of Mental Health (NIMH) [73,135137]. The conceptual model postulates that antibodies against basal ganglia are triggered by
1113
streptococcal infection in a manner analogous to SC. The NIMH working denition requires the abrupt onset of tics or OCD after infection with group A b-hemolytic streptococcus (GAS) in prepubertal children, with motoric hyperactivity and adventitious movements but not frank chorea [136]. The explosive onset of symptoms is said to distinguish PANDAS from the typical case of TS or OCD, which usually begins gradually. By this denition, PANDAS diers from RF and represents a subset of the TS spectrum disorders (tics, TS, and OCD). The precise boundaries of PANDAS, SC, and the TS spectrum have not been established [138141]. It seems possible that there may be multiple shared risk factors across these conditions, including genetic vulnerabilities in families, pathogenic features of particular GAS strains, seasonal exposure modulated by crowding, and age-specic brain or immune vulnerabilities. Unfortunately, proof of a GAS-triggered autoimmune pathogenesis is likely to be challenging because of the hypothesized extended duration from exposure GAS and onset of PANDAS, the high incidence of GAS infections in children, and the absence of laboratory or neuroimaging tests specic for either SC or PANDAS. On the positive side, several preliminary studies oer exciting and provocative results that support extensive follow-up studies. These early results include D8/17 assays, family studies, neuroimaging, treatment, and immune studies. D8/17 is a monoclonal antibody that reacts with antigens on B-lymphocytes (reviewed by Murphy and Goodman [142]). Originally developed in the context of RF research, greater D8/17 reactivity occurs in RF cases and perhaps in individuals at greater genetic risk for RF [143145]. Thus, D8/17 may be a marker for genetic susceptibility to RF and a marker for PANDAS. In Swedo et als (1997) study [137], 23 of 27 (85%) PANDAS patients, 8 of 9 (89%) SC patients, and 4 of 24 (17%) healthy children were D8/17positive. The results of this study seem to be consistent with Murphy et als study of children with TS and OCD [146]. Both studies raise the exciting possibility that D8/17 reactivity may help to clarify the relation between SC, PANDAS, and the genetic subtypes of TS and OCD. The reliability, sensitivity, specicity, and methodology of D8/17 have not been adequately documented, however [142]. Studies in Utah and elsewhere are underway, but assay of D8/17 has not been established as a clinical test at this point. Family studies support the PANDAS concept. The relation of PANDAS in probands to TS and other psychiatric disorders in family members has been studied by the NIMH group [147]. First-degree relatives of PANDAS probands had high rates of tics and/or OCD. This suggests that genetic factors probably contribute to the risk of PANDAS, although it can also be argued that family members share environmental exposures, especially to GAS. Preliminary data from a mail survey of 100 Utah RF cases support the NIMH observations but raise questions about the boundary between PANDAS and SC. In 14 of 65 (22%) individuals with SC in Utah, at least one rst-degree relative had a history of tics, TS, or OCD. Curiously, only
1114
2 of 35 (6%) of those RF patients without SC had a rst-degree relative with tics and/or OCD. The dierence in reported TS spectrum disorders between SC and non-SC RF families is statistically signicant (z 2.08; P 0.02). This preliminary result suggests that SC and TS spectrum disorders may share genetic susceptibility factors. It also raises the question as to whether PANDAS and SC are truly separate entities. Immune studies Besides D8/17, immunologic support for the PANDAS model comes from other antibody studies of subjects with tics and OCD, from a provocative animal model, and from treatment studies (described below). Several groups have detected elevated antineuronal antibodies in patient sera [109,110,115,148150]. Morshed and colleagues [110] reported elevated antinuclear antibodies in TS compared with controls. Serum ASO and antiDNase B titers were not found to be associated with tics or OCD, but higher titers were associated with ADHD [151]. Furthermore, higher titers were also associated with an increased volume of basal ganglia on MRI [152]. Finally, Hallett and colleagues [153] have reported that serum taken from TS patients causes motor and phonic stereotypies reminiscent of tics when microinfused into rat striatum. If replicated, this animal model may provide an important methodologic advance. Treatment of pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections PANDAS cases have been treated at the NIMH with total plasma exchange, intravenous immunoglobulin, and antibiotic prophylaxis [139]. Plasma exchange and immunoglobulin produced signicant improvement, but prophylaxis did not. Tucker and colleagues [150] at Yale University treated a single case of PANDAS with total plasma exchange, resulting in clinical improvement and a concomitant decreased volume of basal ganglia on MRI. Although these research studies lend support to the PANDAS concept, further research is needed before a clinical standard of care can be formulated. Currently, clinicians are advised to assess abrupt onset or treatment-resistant cases of TS and/or OCD for evidence of GAS pharyngitis and to treat children suering from pharyngitis with penicillin or an appropriate alternative [139]. Total plasma exchange and intravenous immunoglobulin should not be administered outside of a research protocol.
Drug-induced movement disorders Numerous medications can induce movement disorders. In children, the most common relevant medication exposures include stimulants, neuroleptics, and antidopaminergic antihistamines. Stimulants, whether prescribed
1115
to treat comorbid ADHD or used as substances of abuse, have been reported to provoke tic onset or severity. Recent prospective studies of stimulant treatment of children with concurrent tics and ADHD found that such treatment was relatively safe, particularly if dosage changes are small and gradual [154,155]. Long-term risks for stimulant exposure on the natural history of tics are not known, however [156]. Movement disorders have been attributed to many other drugs, but the literature describing such associations comes largely from adult populations [157]. Dopamine antagonist activity provides a model mechanism for the prototypical neuroleptics and antihistamines. Other drugs reported to cause movement disorders in children include phenytoin, carbamazepine, chloroquine, and vincristine. Acute dystonia, parkinsonism, akathisia, treatment emergent dyskinesia, and withdrawal emergent dyskinesia are all potential complications of neuroleptic treatment. Unfortunately, the literature on these disorders in children is sparse [157]. The term tardive dyskinesia (TD) was originally applied to a specic pattern of continuous orobuccal dyskinesia emerging during the chronic treatment of schizophrenic adults with neuroleptic medications. With time, variations of this condition have been recognized and discussed, including tardive dystonia, withdrawal dyskinesias, and tardive TS. The pathophysiology of these conditions remains to be claried, but dopamine receptor upregulation and other pharmacologically induced alterations in neural circuitry have been proposed as potential causes. Recent investigations suggest that specic inherited genetic polymorphisms of the dopamine type 3 (DRD3) receptor (a ser-to-gly substitution in exon 1 of DRD3) and/or of the serotonin type 2C (5-HT2C) receptor (cys-to-ser substitution) may predispose to the emergence of TD [158160]. Given the widespread use of typical and atypical antipsychotic medications in children with schizophrenia, autism, aggressive behavior disorders, mood disorders like bipolar disease, and tic disorders [161164], druginduced tardive movement disorders are a potential concern. Although studies of TD in pediatric patients have been largely retrospective, they suggest that such treatment is not benign, with TD or related tardive or withdrawal emergent dyskinesias occurring in up 10% to 20% of patients [162167]. The risk seems to be increased in girls, in children with a strong family history of movement disorders, during longer duration of neuroleptic administration at higher doses, and possibly in association with on-o use of these medications [168]. Extrapolating from studies in adults, it is likely that TD occurs more frequently in the setting of schizophrenia than in other neuropsychiatric conditions [168,169]. In fact, there is considerable evidence that orobuccal dyskinesias may represent an inherent feature of schizophrenia rather than a consequence its treatment [168,169]. In this respect, it is interesting to note that the risk of TD in the treatment of TS and tic disorders may be low. Although several experts have reported cases of TD in their TS patients
1116
treated with neuroleptics [170,171], pediatric TS clinics in Utah have not identied a single case of persistent TD in the cumulative experience of nearly 40 years (>300 pediatric patients treated up to early adulthood). Likewise, no cases of TD occurred among 65 children treated with pimozide by Regeur and colleagues [172]. Therefore, the risk of TD in children treated for TS and tic disorders must be quite low during childhood and adolescence [173]. It is likely that the risk of TD is higher with the increased exposure required by decades of treatment in severe lifelong TS. Finally, it is hoped that the use of atypical neuroleptics reduces the occurrence of TD in at-risk populations [161,174,175]. The development of these agents has in large part been directed toward identifying agents eective for the negative symptoms of schizophrenia but with lower risks of extrapyramidal symptoms. Although cases of TD have apparently been identied after treatment with risperidone, for example [174,176,177], studies to date suggest that the risk of TD and other tardive movement disorders is likely to be much lower with these atypical neuroleptic agents than with conventional higher potency dopamine antagonists [161]. No single treatment is recognized to be superior for the treatment of tardive movement disorders [1,2]. In children, discontinuation of the oending agent is usually followed by gradual disappearance of the dyskinesia. Various interim treatments have been proposed, including tetrabenazine, baclofen, valproic acid, benzodiazepines, and atypical neuroleptics [1,2]. References
[1] Robertson MM, Eapen V, editors. Movement and allied disorders in childhood. Chichester: John Wiley & Sons Ltd; 1995. [2] Saunders-Pullman R, Braun I, Bressman S. Pediatric movement disorders. Child Adolesc Psychiatr Clin North Am 1999;8:74765. [3] Goetz CG, Klawans HL. Gilles de la Tourette on Tourette syndrome. Adv Neurol 1982;35:116. [4] American Psychiatric Association. Diagnostic and statistical manual of mental disorders. 4th edition. Text revision. Washington, DC: American Psychiatric Association; 2000. [5] The Tourette Syndrome Classication Study Group. Denitions and classication of tic disorders. Arch Neurol 1993;50:10136. [6] American Psychiatric Association. Diagnostic and statistical manual of mental disorders. 3rd edition. Washington, DC: American Psychiatric Association; 1994. [7] Scahill L, Tanner C, Dure L. The epidemiology of tics and Tourette syndrome in children and adolescents. Adv Neurol 2001;85:26171. [8] Comings DE, Himes JA, Comings BG. An epidemiologic study of Tourettes syndrome in a single school district. J Clin Psychiatry 1990;51:4639. [9] Apter A, Pauls DL, Bleich A, et al. An epidemiologic study of Gilles de la Tourettes syndrome in Israel. Arch Gen Psychiatry 1993;50:7348. [10] Costello EJ, Angold A, Burns BJ, et al. The Great Smoky Mountains study of youth, goals, design, and the prevalence of DSM-III-R disorders. Arch Gen Psychiatry 1996; 53:12936. [11] Burd L, Kerbeshian J, Wikenheiser M, et al. A prevalence study of Gilles de la Tourette syndrome in North Dakota school-age children. J Am Acad Child Psychiatry 1986;25:5523.
1117
[12] Goetz CG, Tanner CM, Stebbins GT, et al. Adult tics in Gilles de la Tourettes syndrome: description and risk factors. Neurology 1992;42:7848. [13] Leckman JF, Zhang H, Vitale A, et al. Course of tic severity in Tourette syndrome: the rst two decades. Pediatrics 1998;102:149. [14] Peterson BS, Leckman JF. The temporal dynamics of tics in Gilles de la Tourette syndrome. Biol Psychiatry 1998;44:133748. [15] Jankovic J. Tourettes syndrome. N Engl J Med 2001;345:118492. [16] Merette C, Brassard A, Potvin A, et al. Signicant linkage for Tourette syndrome in a large French Canadian family. Am J Hum Genet 2000;67:100813. [17] The Tourette Syndrome Association International Consortium for Genetics. A complete genome screen in sib pairs aected by Gilles de la Tourette Syndrome. Am J Hum Genet 1999;65:142836. [18] Zhang H, Leckman JF, Pauls DL, et al. Genome wide scan of hoarding in sibling pairs both diagnosed with Gilles de la Tourette syndrome. Am J Hum Genet 2002;70:896904. [19] Leckman JF, Peterson BS, Anderson GM, et al. Pathogenesis of Tourettes syndrome. J Child Psychol Psychiatry 1997;38:11942. [20] Leckman JF, Riddle MA. Tourettes syndrome: when habit-forming systems form habits of their own? Neuron 2000;28:34954. [21] Anderson GM, Pollak ES, Chatterjee D, et al. Postmortem analysis of subcortical monoamines and amino acids in Tourette syndrome. Adv Neurol 1992;58:12333. [22] Haber SN, Kowall NW, Vonsattel JP, et al. Gilles de la Tourettes syndrome. A postmortem neuropathological and immunohistochemical study. J Neurol Sci 1986;75: 22541. [23] Haber SN, Wolfer D. Basal ganglia peptidergic staining in Tourette syndrome. A followup study. Adv Neurol 1992;58:14550. [24] Singer HS. Neurochemical analysis of postmortem cortical and striatal brain tissue in patients with Tourette syndrome. Adv Neurol 1992;58:13544. [25] Peterson BS. Neuroimaging studies of Tourette syndrome: a decade of progress. Adv Neurol 2001;85:17996. [26] Peterson BS, Staib L, Scahill L, et al. Regional brain and ventricular volumes in Tourette syndrome. Arch Gen Psychiatry 2001;58:42740. [27] Fredericksen KA, Cutting LE, Kates WR, et al. Disproportionate increases of white matter in right frontal lobe in Tourette syndrome. Neurology 2002;58:859. [28] Biswal B, Ulmer JL, Krippendorf RL, et al. Abnormal cerebral activation associated with a motor task in Tourette syndrome. AJNR Am J Neuroradiol 1998;19:150912. [29] Braun AR, Stoetter B, Randolph C, et al. The functional neuroanatomy of Tourettes syndrome: an FDG-PET studyI. Regional changes in cerebral glucose metabolism dierentiating patients and controls. Neuropsychopharmacology 1993;9:27791. [30] Chase TN, Georey V, Gillespie M, et al. Structural and functional studies of Gilles de la Tourette syndrome. Rev Neurol (Paris) 1986;142:8515. [31] Hall M, Costa DC, Shields J, et al. Brain perfusion patterns with Tc-99m-HMPAO/SPET in patients with Gilles de la Tourette syndrome. Eur J Nucl Med 1990;16:564(WP18). [32] Moriarty J, Costa DC, Schmitz B, et al. Brain perfusion abnormalities in Gilles de la Tourettes syndrome. Br J Psychiatry 1995;167:24954. [33] Peterson BS, Skudlarski P, Anderson AW, et al. A functional magnetic resonance imaging study of tic suppression in Tourette syndrome. Arch Gen Psychiatry 1998;55:32633. [34] Riddle MA, Rasmusson AM, Woods SW, et al. SPECT imaging of cerebral blood ow in Tourette syndrome. Adv Neurol 1992;58:20711. [35] Peterson BS, Skudlarski P, Gatenby JC, et al. An fMRI study of Stroop word-color interference: evidence for cingulate subregions subserving multiple distributed attentional systems. Biol Psychiatry 1999;45:123758. [36] Ernst M, Zametkin AJ, Jons PH, et al. High presynaptic dopaminergic activity in children with Tourettes disorder. J Am Acad Child Adolesc Psychiatry 1999;38:8694.
1118
[37] Heinz A, Knable MB, Wolf SS, et al. Tourettes syndrome: [I-123]beta-CIT SPECT correlates of vocal tic severity. Neurology 1998;51:106974. [38] Malison RT, McDougle CJ, van Dyck CH, et al. [123I]beta-CIT SPECT imaging of striatal dopamine transporter binding in Tourettes disorder. Am J Psychiatry 1995;152: 135961. [39] Wolf SS, Jones DW, Knable MB, et al. Tourette syndrome: prediction of phenotypic variation in monozygotic twins by caudate nucleus D2 receptor binding. Science 1996;273:12257. [40] Mostofsky SH, Wendlandt J, Cutting L, et al. Corpus callosum measurements in girls with Tourette syndrome. Neurology 1999;53:13457. [41] Zimmerman AM, Abrams MT, Giuliano JD, et al. Subcortical volumes in girls with Tourette syndrome: support for a gender eect. Neurology 2000;54:22249. [42] Durston S, Hulsho Pol HE, Casey BJ, et al. Anatomical MRI of the developing human brain: what have we learned? J Am Acad Child Adolesc Psychiatry 2001;40:101220. [43] Giedd JN, Castellanos FX, Rajapakse JC, et al. Sexual dimorphism of the developing human brain. Prog Neuropsychopharmacol Biol Psychiatry 1997;21:1185201. [44] Giedd JN, Rumsey JM, Castellanos FX, et al. A quantitative MRI study of the corpus callosum in children and adolescents. Brain Res Dev Brain Res 1996;91:27480. [45] Giedd JN, Vaituzis AC, Hamburger SD, et al. Quantitative MRI of the temporal lobe, amygdala, and hippocampus in normal human development: ages 418 years. J Comp Neurol 1996;366:22330. [46] Lange N, Giedd JN, Castellanos FX, et al. Variability of human brain structure size: ages 420 years. Psychiatry Res 1997;74:112. [47] Reiss AL, Abrams MT, Singer HS, et al. Brain development, gender and IQ in children. A volumetric imaging study. Brain 1996;119:176374. [48] Sheppard DM, Bradshaw JL, Purcell R, et al. Tourettes and comorbid syndromes: obsessive compulsive and attention decit hyperactivity disorder. A common etiology? Clin Psychol Rev 1999;19:53152. [49] Walkup JT, Khan S, Schuerholz L, et al. Phenomenology and natural history of ticrelated ADHD and learning disabilities. In: Leckman JF, Cohen DJ, editors. Tourettes syndrometics, obsessions, compulsions. Developmental psychopathology and clinical care. New York: John Wiley & Sons Ltd; 1999. p. 6379. [50] Caprini G, Melotti V. Un grave sindrome ticcosa guarita con haloperidol. Rivista Sperimentale di Freniatria e Medicina Legalle Delle Alienzioni Mentali 1961;85:1916. [51] Challas G, Brauer W. Tourettes disease: relief of symptoms with R1625. Am J Psychiatry 1963;120:2834. [52] Seignot MJN. Un cas de maladie des tics de Gilles de la Tourette gueri par le R-1625. Annales Medico-Psychologiques 1961;119:5789. [53] Shapiro AK, Shapiro E. Treatment of Gilles de la Tourettes syndrome with haloperidol. Br J Psychiatry 1968;114:34550. [54] Shapiro AK, Shapiro ES, Bruun RD, et al. History of tics and Tourette syndrome. In: Gilles de la Tourette syndrome. New York: Raven Press; 1978. p. 1182. [55] Sallee FR, Nesbitt L, Jackson C, et al. Relative ecacy of haloperidol and pimozide in children and adolescents with Tourettes disorder. Am J Psychiatry 1997;154:105762. [56] Shapiro E, Shapiro AK, Fulop G, et al. Controlled study of haloperidol, pimozide and placebo for the treatment of Gilles de la Tourettes syndrome. Arch Gen Psychiatry 1989;46:72230. [57] Goetz CG, Tanner CM, Klawans HL. Fluphenazine and multifocal tic disorders. Arch Neurol 1984;41:2712. [58] Caine ED, Polinsky RJ, Kartzinel R, et al. The trial use of clozapine for abnormal involuntary movement disorders. Am J Psychiatry 1979;136:31720. [59] Rabinowitz J, Hornik T, Davidson M. Rapid onset of therapeutic eect of risperidone versus haloperidol in a double-blind randomized trial. J Clin Psychiatry 2001;62:3436.
1119
[60] Scahill L, Chappell PB, Kim YS, et al. A placebo-controlled study of guanfacine in the treatment of children with tic disorders and attention decit hyperactivity disorder. Am J Psychiatry 2001;158:106774. [61] Gilbert DL, Sethuraman G, Sine L, et al. Tourettes syndrome improvement with pergolide in a randomized, double-blind, crossover trial. Neurology 2000;54:13105. [62] Marras C, Andrews D, Sime E, et al. Botulinum toxin for simple motor tics: a randomized, double-blind, controlled clinical trial. Neurology 2001;56:60510. [63] Awaad Y. Tics in Tourette syndrome: new treatment options. J Child Neurol 1999; 14:3169. [64] Singer HS, Wendlandt J, Krieger M, et al. Baclofen treatment in Tourette syndrome: a double-blind, placebo-controlled, crossover trial. Neurology 2001;56:599604. [65] Walter AL, Carter AS. Gilles de la Tourettes syndrome in childhood: a guide for school professionals. School Psychology Rev 1997;26:2846. [66] Dilenge ME, Shevell MI, Dinh L. Restricted unilateral Sydenhams chorea: reversible contralateral striatal hypermetabolism demonstrated on single photon emission computed tomographic scanning. J Child Neurol 1999;14:50913. [67] Garvey MA, Swedo SE. Sydenhams chorea. Clinical and therapeutic update. Adv Exp Med Biol 1997;418:11520. [68] Goldenberg J, Ferraz MB, Fonseca AS, et al. Sydenham chorea: clinical and laboratory ndings. Analysis of 187 cases. Rev Paul Med 1992;110:1527. [69] Heye N, Jergas M, Hotzinger H, et al. Sydenham chorea: clinical, EEG, MRI and SPECT ndings in the early stage of the disease. J Neurol 1993;240:1213. [70] Swedo SE, Leonard HL, Schapiro MB, et al. Sydenhams chorea: physical and psychological symptoms of St. Vitus dance. Pediatrics 1993;91:70613. [71] Van der Merwe PL, Kalis NN. Sydenhams choreaanalysis of 27 patients and a review of the literature. S Afr Med J 1997;87(Suppl 3):C15760. [72] Wilcox JA, Nasrallah H. Sydenhams chorea and psychopathology. Neuropsychobiology 1988;19:68. [73] Swedo SE, Rapoport JL, Cheslow DL, et al. High prevalence of obsessive-compulsive symptoms in patients with Sydenhams chorea. Am J Psychiatry 1989;146:2469. [74] Mercadante MT, Busatto GF, Lombroso PJ, et al. The psychiatric symptoms of rheumatic fever. Am J Psychiatry 2000;157:20368. [75] Berrios X, Quesney F, Morales A, et al. Are all recurrences of pure Sydenham chorea true recurrences of acute rheumatic fever? J Pediatr 1985;107:86772. [76] Gibb WR, Lees AJ, Scadding JW. Persistent rheumatic chorea. Neurology 1985;35:1012. [77] Abraham S, OGorman M, Shulman ST. Anti-nuclear antibodies in Sydenhams chorea. Adv Exp Med Biol 1997;418:1536. [78] Figueroa F, Berrios X, Gutierrez M, et al. Anticardiolipin antibodies in acute rheumatic fever [see comments]. J Rheumatol 1992;19:117580. [79] Mirkinson L. The diagnosis of rheumatic fever. Pediatr Rev 1998;19:3101. [80] Persellin RH. Acute rheumatic fever: changing manifestations. Ann Intern Med 1978; 89:10023. [81] Bisno AL. The resurgence of acute rheumatic fever in the United States. Annu Rev Med 1990;41:31929. [82] Veasy LG, Tani LY, Hill HR. Persistence of acute rheumatic fever in the intermountain area of the United States. J Pediatr 1994;124:916. [83] Veasy LG, Wiedmeier SE, Orsmond GS, et al. Resurgence of acute rheumatic fever in the intermountain area of the United States. N Engl J Med 1987;316:4217. [84] Kaplan EL, Rothermel CD, Johnson DR. Antistreptolysin O and anti-deoxyribonuclease B titers: normal values for children ages 2 to 12 in the United States. Pediatrics 1998; 101:868. [85] Abernethy M, Bass N, Sharpe N, et al. Doppler echocardiography and the early diagnosis of carditis in acute rheumatic fever. Aust NZ J Med 1994;24:5305.
1120
[86] Davies C, Sahn D. Detection and signicance of subclinical mitral regurgitation by colour Doppler techniques [letter, comment]. Heart 2001;85:36970. [87] Elevli M, Celebi A, Tombul T, et al. Cardiac involvement in Sydenhams chorea: clinical and Doppler echocardiographic ndings. Acta Paediatr 1999;88:10747. [88] Figueroa FE, Fernandez MS, Valdes P, et al. Prospective comparison of clinical and echocardiographic diagnosis of rheumatic carditis: long term follow up of patients with subclinical disease [see comments]. Heart 2001;85:40710. [89] Folger GM, Hajar R, Robida A, et al. Occurrence of valvar heart disease in acute rheumatic fever without evident carditis: colour-ow Doppler identication. Br Heart J 1992;67:4348. [90] Ozkutlu S, Ayabakan C, Saraclar M. Can subclinical valvitis detected by echocardiography be accepted as evidence of carditis in the diagnosis of acute rheumatic fever? Cardiology in the young 2001;11:25560. [91] Minich LL, Tani LY, Pagotto LT, et al. Doppler echocardiography distinguishes between physiologic and pathologic silent mitral regurgitation in patients with rheumatic fever. Clin Cardiol 1997;20:9246. [92] al Jishi F, al Kawi MZ, el Ramahi K, et al. Hemichorea in systemic lupus erythematosus: signicance of MRI ndings. Lupus 1995;4:3213. [93] Besbas N, Damarguc I, Ozen S, et al. Association of antiphospholipid antibodies with systemic lupus erythematosus in a child presenting with chorea: a case report [see comments]. Eur J Pediatr 1994;153:8913. [94] Font J, Cervera R, Espinosa G, et al. Systemic lupus erythematosus (SLE) in childhood: analysis of clinical and immunological ndings in 34 patients and comparison with SLE characteristics in adults. Ann Rheum Dis 1998;57:4569. [95] Angelini L, Zibordi F, Zorzi G, et al. Neurological disorders, other than stroke, associated with antiphospholipid antibodies in childhood. Neuropediatrics 1996;27:14953. [96] Cervera R, Asherson RA, Font J, et al. Chorea in the antiphospholipid syndrome. Clinical, radiologic, and immunologic characteristics of 50 patients from our clinics and the recent literature. Medicine (Baltimore) 1997;76:20312. [97] Kiechl-Kohlendorfer U, Ellemunter H, Kiechl S. Chorea as the presenting clinical feature of primary antiphospholipid syndrome in childhood. Neuropediatrics 1999;30:968. [98] Levine SR, Brey RL. Neurological aspects of antiphospholipid antibody syndrome. Lupus 1996;5:34753. [99] Masala C, Morino S, Zangari P, et al. Chorea in primary antiphospholipid syndrome. Clin Neurol Neurosurg 1996;98:2478. [100] Okseter K, Sirnes K. Chorea and lupus anticoagulant: a case report. Acta Neurol Scand 1988;78:2069. [101] Dale RC, Church AJ, Cardoso F, et al. Poststreptococcal acute disseminated encephalomyelitis with basal ganglia involvement and auto-reactive antibasal ganglia antibodies. Ann Neurol 2001;50:58895. [102] Fujinami RS, Oldstone MB. Molecular mimicry as a mechanism for virus-induced autoimmunity. Immunol Res 1989;8:315. [103] Husby G, v de Rijn I, Zabriskie JB, et al. Anti-neuronal antibody in Sydenhams chorea. Lancet 1977;1:1208. [104] Dorling J, Kingston D, Webb JA. Anti-streptococcal antibodies reacting with brain tissue. II. Utrastructural studies. Br J Exp Pathol 1976;57:25565. [105] Husby G, Forre O, Williams RC Jr . IgG subclass, variable H-chain subgroup, and light chain-type composition of antineuronal antibody in Huntingtons disease and Sydenhams chorea. Clin Immunol Immunopathol 1979;14:3617. [106] Kingston D, Glynn LE. Anti-streptococcal antibodies reacting with brain tissue. I. Immunouorescent studies. Br J Exp Pathol 1976;57:11428. [107] Swedo SE. Sydenhams chorea. A model for childhood autoimmune neuropsychiatric disorders. JAMA 1994;272:178891.
1121
[108] Bronze MS, Dale JB. Epitopes of streptococcal M proteins that evoke antibodies that cross-react with human brain. J Immunol 1993;151:28208. [109] Kiessling LS, Marcotte AC, Culpepper L. Antineuronal antibodies in movement disorders. Pediatrics 1993;92:3943. [110] Morshed SA, Parveen S, Leckman JF, et al. Antibodies against neural, nuclear, cytoskeletal, and streptococcal epitopes in children and adults with Tourettes syndrome, Sydenhams chorea, and autoimmune disorders. Biol Psychiatry 2001;50:56677. [111] Arnold PD, Richter MA. Is obsessive-compulsive disorder an autoimmune disease? CMAJ 2001;165:13538. [112] Murphy TK, Petitto JM, Voeller KK, et al. Obsessive compulsive disorder: is there an association with childhood streptococcal infections and altered immune function? Semin Clin Neuropsychiatry 2001;6:26676. [113] Swedo SE, Leonard HL. Childhood movement disorders and obsessive compulsive disorder. J Clin Psychiatry 1994;55(Suppl):327. [114] Singer HS, Giuliano JD, Hansen BH, et al. Antibodies against a neuron-like (HTB10 neuroblastoma) cell in children with Tourette syndrome. Biol Psychiatry 1999;46: 77580. [115] Singer HS, Giuliano JD, Hansen BH, et al. Antibodies against human putamen in children with Tourette syndrome. Neurology 1998;50:161824. [116] Visvanathan K, Triletti RR, Altemus M, et al. Autoimmune mechanisms in movement disorders. Semin Pediatr Neurol 2000;7:1037. [117] Castillo M, Kwock L, Arbelaez A. Sydenhams chorea: MRI and proton spectroscopy. Neuroradiology 1999;41:9435. [118] Emery ES, Vieco PT. Sydenham chorea: magnetic resonance imaging reveals permanent basal ganglia injury. Neurology 1997;48:5313. [119] Hill A, Herkes GK, Roche P. SPECT and MRI ndings in Sydenhams chorea. J Neurol Neurosurg Psychiatry 1994;57:763. [120] Ikuta N, Hirata M, Sasabe F, et al. High-signal basal ganglia on T1-weighted images in a patient with Sydenhams chorea. Neuroradiology 1998;40:65961. [121] Ju TH, Kao KP, Chen CC. Sydenham chorea. AJNR Am J Neuroradiol 1993;14:1265. [122] Traill Z, Pike M, Byrne J. Sydenhams chorea: a case showing reversible striatal abnormalities on CT and, MRI. Dev Med Child Neurol 1995;37:2703. [123] Giedd JN, Rapoport JL, Kruesi MJ, et al. Sydenhams chorea: magnetic resonance imaging of the basal ganglia. Neurology 1995;45:2199202. [124] Giedd JN, Rapoport JL, Garvey MA, et al. MRI assessment of children with obsessivecompulsive disorder or tics associated with streptococcal infection. Am J Psychiatry 2000; 157:2813. [125] Giedd JN, Rapoport JL, Leonard HL, et al. Case study: acute basal ganglia enlargement and obsessive-compulsive symptoms in an adolescent boy. J Am Acad Child Adolesc Psychiatry 1996;35:9135. [126] Singer HS, Reiss AL, Brown JE, et al. Volumetric MRI changes in basal ganglia of children with Tourettes syndrome. Neurology 1993;43:9506. [127] Goldman S, Amrom D, Szliwowski HB, et al. Reversible striatal hypermetabolism in a case of Sydenhams chorea. Mov Disord 1993;8:3558. [128] Lee PH, Nam HS, Lee KY, et al. Serial brain SPECT images in a case of Sydenham chorea. Arch Neurol 1999;56:23740. [129] Weindl A, Kuwert T, Leenders KL, et al. Increased striatal glucose consumption in Sydenhams chorea. Mov Disord 1993;8:43744. [130] Garvey MA, Swedo SE, Shapiro MB, et al. Intravenous immunoglobulin and plasmapheresis treatments of Sydenhams chorea [abstract]. Neurology 1996;46:A147. [131] Lue HC, Wu MH, Hsieh KH, et al. Rheumatic fever recurrences: controlled study of 3-week versus 4-week benzathine penicillin prevention programs. J Pediatr 1986;108: 299304.
1122
[132] American Academy of Pediatrics. 2000 Red book: report of the Committee on Infectious Diseases. 25th edition. Elk Grove Village, IL: American Academy of Pediatrics; 2000. [133] Perlmutter SJ, Leitman SF, Garvey MA, et al. Therapeutic plasma exchange and intravenous immunoglobulin for obsessive-compulsive disorder and tic disorders in childhood. Lancet 1999;354:11538. [134] Thompson JA, Tani LY, Bale JF. Sydenham chorea: the Utah experience. Ann Neurol 1999;46:523 (Abstract). [135] Allen AJ, Leonard HL, Swedo SE. Case study: a new infection-triggered, autoimmune subtype of pediatric OCD and Tourettes syndrome. J Am Acad Child Adolesc Psychiatry 1995;34:30711. [136] Swedo SE, Leonard HL, Garvey M, et al. Pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections: clinical description of the rst 50 cases. Am J Psychiatry 1998;155:26471. [137] Swedo SE, Leonard HL, Mittleman BB, et al. Identication of children with pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections by a marker associated with rheumatic fever. Am J Psychiatry 1997;154:1102. [138] Bessen DE, Lombroso PJ. Group A streptococcal infections and their potential role in neuropsychiatric disease. Adv Neurol 2001;85:295305. [139] Hamilton CS, Garvey MA, Swedo SE. Therapeutic implications of immunology for tics and obsessive-compulsive disorder. Adv Neurol 2001;85:3118. [140] Kurlan R. Could Tourette syndrome be a neurologic manifestation of rheumatic fever? Adv Neurol 2001;85:30710. [141] Murphy TK, Goodman WK, Ayoub EM, et al. On dening Sydenhams chorea: where do we draw the line? Biol Psychiatry 2000;47:8517. [142] Murphy T, Goodman W. Genetics of childhood disorders: XXXIV. Autoimmune disorders, part 7: D8/17 reactivity as an immunological marker of susceptibility to neuropsychiatric disorders. J Am Acad Child Adolesc Psychiatry 2002;41:98100. [143] Chapman F, Visvanathan K, Carreno-Manjarrez R, et al. A ow cytometric assay for D8/17 B cell marker in patients with Tourettes syndrome and obsessive compulsive disorder. J Immunol Methods 1998;219:1816. [144] Khanna AK, Buskirk DR, Williams RC, Jr, et al. Presence of a non-HLA B cell antigen in rheumatic fever patients and their families as dened by a monoclonal antibody. J Clin Invest 1989;83:17106. [145] Regelmann WE, Talbot R, Cairns L, et al. Distribution of cells bearing rheumatic antigens in peripheral blood of patients with rheumatic fever/rheumatic heart disease. J Rheumatol 1989;16:9315. [146] Murphy TK, Goodman WK, Fudge MW, et al. B lymphocyte antigen D8/17: a peripheral marker for childhood-onset obsessive-compulsive disorder and Tourettes syndrome? Am J Psychiatry 1997;154:4027. [147] Lougee L, Perlmutter SJ, Nicolson R, et al. Psychiatric disorders in rst-degree relatives of children with pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections (PANDAS). J Am Acad Child Adolesc Psychiatry 2000; 39:11206. [148] Kiessling LS, Marcotte AC, Culpepper L. Antineuronal antibodies: tics and obsessivecompulsive symptoms. J Dev Behav Pediatr 1994;15:4215. [149] Muller N, Riedel M, Forderreuther S, et al. Tourettes syndrome and mycoplasma pneumoniae infection. Am J Psychiatry 2000;157:4812. [150] Tucker DM, Leckman JF, Scahill L, et al. A putative poststreptococcal case of OCD with chronic tic disorder, not otherwise specied. J Am Acad Child Adolesc Psychiatry 1996;35:168491. [151] Peterson BS, Leckman JF, Tucker D, et al. Preliminary ndings of antistreptococcal antibody titers and basal ganglia volumes in tic, obsessive-compulsive, and attention decit/hyperactivity disorders. Arch Gen Psychiatry 2000;57:36472.
1123
[152] Garvey MA, Giedd J, Swedo SE. PANDAS: the search for environmental triggers of pediatric neuropsychiatric disorders. Lessons from rheumatic fever. J Child Neurol 1998;13:41323. [153] Hallett JJ, Harling-Berg CJ, Knopf PM, et al. Anti-striatal antibodies in Tourette syndrome cause neuronal dysfunction. J Neuroimmunol 2000;111:195202. [154] Castellanos FX, Giedd JN, Elia J, et al. Controlled stimulant treatment of ADHD and comorbid Tourettes syndrome: eects of stimulant and dose. J Am Acad Child Adolesc Psychiatry 1997;36:58996. [155] Gadow KD, Sverd J, Sprafkin J, et al. Long-term methylphenidate therapy in children with comorbid attention-decit hyperactivity disorder and chronic multiple tic disorder. Arch Gen Psychiatry 1999;56:3306. [156] Peterson BS, Leckman JF, Lombroso P, et al. Environmental risk and protective factors. In: Leckman JF, Cohen DJ, editors. Tourettes syndrometics, obsessions, compulsions. Developmental psychopathology and clinical care. New York: John Wiley & Sons Ltd; 1999. p. 21329. [157] Cunningham-Owens DG. Drug-related movement disorders. In: Robertson MM, Eapen V, editors. Movement and allied disorders in childhood. Chichester: John Wiley & Sons Ltd; 2001, p. 199236. [158] Basile VS, Masellis M, Badri F, et al. Association of the MscI polymorphism of the dopamine D3 receptor gene with tardive dyskinesia in schizophrenia. Neuropsychopharmacology 1999;21:1727. [159] Liao DL, Yeh YC, Chen HM, et al. Association between the Ser9Gly polymorphism of the dopamine D3 receptor gene and tardive dyskinesia in Chinese schizophrenic patients. Neuropsychobiology 2001;44:958. [160] Segman RH, Heresco-Levy U, Finkel B, et al. Association between the serotonin 2C receptor gene and tardive dyskinesia in chronic schizophrenia: additive contribution of 5HT2Cser and DRD3gly alleles to susceptibility. Psychopharmacologia 2000;152:40813. [161] Glick ID, Murray SR, Vasudevan P, et al. Treatment with atypical antipsychotics: new indications and new populations. J Psychiatr Res 2001;35:18791. [162] McDermid SA, Hood J, Bockus S, et al. Adolescents on neuroleptic medication: is this population at risk for tardive dyskinesia? Can J Psychiatry 1998;43:62931. [163] Richardson MA, Haugland G, Craig TJ. Neuroleptic use, parkinsonian symptoms, tardive dyskinesia, and associated factors in child and adolescent psychiatric patients. Am J Psychiatry 1991;148:13228. [164] Shay J, Sanchez LE, Cueva JE, et al. Neuroleptic-related dyskinesias and stereotypies in autistic children: videotaped ratings. Psychopharmacol Bull 1993;29:35963. [165] Campbell M, Armenteros JL, Malone RP, et al. Neuroleptic-related dyskinesias in autistic children: a prospective, longitudinal study. J Am Acad Child Adolesc Psychiatry 1997;36:83543. [166] Kumra S, Jacobsen LK, Lenane M, et al. Case series: spectrum of neuroleptic-induced movement disorders and extrapyramidal side eects in childhood-onset schizophrenia. J Am Acad Child Adolesc Psychiatry 1998;37:2217. [167] Silverstein FS, Johnston MV. Risks of neuroleptic drugs in children. J Child Neurol 1987;2:413. [168] Muscettola G, Barbato G, Pampallona S, et al. Extrapyramidal syndromes in neuroleptictreated patients: prevalence, risk factors, and association with tardive dyskinesia. J Clin Psychopharmacol 1999;19:2038. [169] Fenton WS, Blyler CR, Wyatt RJ, et al. Prevalence of spontaneous dyskinesia in schizophrenic and non-schizophrenic psychiatric patients. Br J Psychiatry 1997;171: 2658. [170] Golden GS. Tardive dyskinesia in Tourette syndrome. Pediatr Neurol 1985;1:1924. [171] Mizrahi EM, Holtzman D, Tharp B. Haloperidol-induced tardive dyskinesia in a child with Gilles de la Tourettes disease. Arch Neurol 1980;37:780.
1124
[172] Regeur L, Pakkenberg B, Fog R, et al. Clinical features and long-term treatment with pimozide in 65 patients with Gilles de la Tourettes syndrome. J Neurol Neurosurg Psychiatry 1986;49:7915. [173] Bruun RD. Subtle and underrecognized side eects of neuroleptic treatment in children with Tourettes disorder. Am J Psychiatry 1988;145:6214. [174] Fleischhacker WW, Lemmens P, van Baelen B. A qualitative assessment of the neurological safety of antipsychotic drugs; an analysis of a risperidone database. Pharmacopsychiatry 2001;34:10410. [175] Sallee FR, Kurlan R, Goetz CG, et al. Ziprasidone treatment of children and adolescents with Tourettes syndrome: a pilot study. J Am Acad Child Adolesc Psychiatry 2000; 39:2929. [176] Carroll NB, Boehm KE, Strickland RT. Chorea and tardive dyskinesia in a patient taking risperidone. J Clin Psychiatry 1999;60:4857. [177] Feeney DJ, Klykylo W. Risperidone and tardive dyskinesia. J Am Acad Child Adolesc Psychiatry 1996;35:14212.
Neurol Clin N Am 20 (2002) 11251141
Rett syndrome Current status and new vistas

Alan K. Percy, MD*
Departments of Pediatrics, Neurology, and Neurobiology, University of Alabama at Birmingham School of Medicine, Birmingham, AL, USA
Imagine that you encounter a puzzling condition aecting young girls and, despite repeated attempts, are unable to convince other clinicians that they represent a unique and previously unrecognized syndrome. Then, fast forward 35 years and nd that not only has this syndrome been associated with mutations in a specic gene but that a myriad of other clinical presentations may result from mutations in this gene. Such is the history of Rett syndrome (RS). Unfortunately, Andreas Rett did not live to realize the fruits of these ndings. Nevertheless, his initial observations and his persistence in pursuing the underlying basis of RS have opened important new vistas in neurodevelopmental disorders and, more broadly, in neurobiology. In the early 1960s, Rett, a developmental pediatrician in Vienna, recognized this syndrome as a neurodevelopmental disorder predominantly aecting young girls [1]. His early accounts of RS were, however, brief and not widely circulated. Bengt Hagberg noted girls in Sweden with similar clinical features about the same time. Nevertheless, not until 1983 did Hagberg and colleagues [2] report their experiences with RS in Sweden, France, and Portugal in the rst widely circulated English language publication on the disorder. In short order, RS was identied in the United States, Japan, and throughout Western Europe [36]. Several epidemiologic studies (Table 1) were conducted, producing prevalence rates ranging from 1:10,000 in Sweden [7] to 1:22,000 in Texas [8]. As such, it is now clear that RS occurs throughout the world and is the leading cause of profound cognitive impairment in girls and women.
Supported in part by the International Rett Syndrome Association. * Correspondence. Departments of Pediatrics, Neurology, and Neurobiology, University of Alabama at Birmingham School of Medicine, 1600 7th Avenue South, Suite 516, Birmingham, AL 35233, USA. E-mail address: apercy@peds.uab.edu (A.K. Percy). 0733-8619/02/$ - see front matter 2002, Elsevier Science (USA). All rights reserved. PII: S 0 7 3 3 - 8 6 1 9 ( 0 2 ) 0 0 0 2 2 - 1
1126
A.K. Percy / Neurol Clin N Am 20 (2002) 11251141
Table 1 Prevalence of Rett syndrome Location Western Scotland Switzerland Western Sweden Japan Texas Australia Sweden Year 1982 1982 1982 1988 1990 1995 1996 Number 19 27 12 24 103 79 69 Prevalence 1:15,000 1:24,600 1:13,000 1:25,000 1:22,800 1:22,000 1:13,000
Clinical characteristics RS (Fig. 1) is characterized by profound cognitive impairment, communication dysfunction, stereotypic movements, and pervasive growth failure, which follow a period of apparently normal development for the rst 6 to 18 months of life [9]. All attempts to nd a biologic marker were unsuccessful until the recent identication of a gene for RS [10]. The diagnosis of RS, however, is based on clinical criteria (Table 2) [11,12], because not all girls with the RS phenotype have mutations in this gene (see below). These include normal pre- and perinatal periods and apparently normal development for the rst several months of life. Thereafter, purposeful hand skills are lost along with regression of psychomotor and communication functions. These changes may occur as early as 6 months of age or as late as 2.5 years. Some girls may have features of the autistic spectrum disorders in the sense that eye contact is poor and attempts at socialization and
Fig. 1. Six-year-old girl with typical features of Rett syndrome and mutation in MECP2.
A.K. Percy / Neurol Clin N Am 20 (2002) 11251141 Table 2 Diagnostic criteria for Rett syndrome Criteria Apparently normal prenatal and perinatal periods Psychomotor development may seem normal through 6 months or may be delayed from birth Postnatal deceleration of head growth rate in most children Decrease of purposeful hand skills Communication dysfunction and social withdrawal Severely impaired expressive and receptive language and psychomotor impairment Stereotypic movements Hand washing/wringing Hand clapping/patting Hand mouthing Impaired or absent locomotion Absence of other disease process Organomegaly Optic atrophy Retinal changes Intrauterine growth retardation Onset 68 months
1127
3 months4 years 6 months2.5 years 9 months2.5 years
13 years
14 years
communication are severely limited. Profound irritability without apparent explanation is often noted in these early years. Rett noted deceleration in the rate of head growth [1]. This nding may appear as early as 3 months of age and may represent the rst clinical sign of RS. It may also be profound, leading to microcephaly in some girls [13]. Stereotypic hand movements emerge between the ages of 1 and 3 years and are characterized by handwashing, hand-wringing, or hand-clapping/hand-patting movements. Hand stereotypies occur typically but not exclusively in the midline. Occasionally, hand mouthing or picking at the hair or clothes is the predominant stereotypy. Most girls with RS are able to walk; however, gait is apraxic, broad based, wandering, and purposeless in character. Walking is often initiated by rst stepping backward (retropulsion). Side-to-side rocking is common. The diagnosis of RS requires a careful history and neurologic evaluation to include growth and developmental parameters. Ancillary evaluations should include audiologic and ophthalmologic assessments, chromosome analysis with high-resolution banding, and the molecular probe for Angelman syndrome (AS). Many behavioral features associated with RS resemble clinical seizures. Thus, video-electroencephalography (EEG) monitoring, if available, is recommended. An extensive investigation for inherited metabolic disorders is not warranted. Despite intensive investigations of amino and organic acids, mitochondrial function, and urea cycle metabolism, no consistent metabolic abnormality has been identied. Denitive diagnosis is accomplished by mutation analysis on leukocyte DNA for the gene MECP2, which encodes the methylcytosine-guanosine (CpG)binding protein 2 (MeCP2).
1128
Clinical staging RS has been characterized in four clinical stages [14], providing a format for plotting clinical progression. The transition from one stage to the next is generally along a continuum rather than an abrupt change. The rst stage is the early-onset stagnation period, which occurs from the age of 6 to 18 months. Usually, this stage lasts weeks to months and consists of delay in developmental progress but no clear evidence of regression. The second stage is that of rapid developmental regression with an onset from the age of 1 to 3 or 4 years. This stage may be relatively brief, lasting days to weeks or as long as a year. Acquired skills in motor and communication function are lost and impairment of cognitive performance becomes apparent. At this stage, RS must be dierentiated from autistic spectrum disorders, infantile neuronal ceroid lipofuscinosis, AS, or an acute toxic or infectious encephalopathy. Infantile neuronal ceroid lipofuscinosis and AS feature deceleration in the rate of head growth, seizures, and stereotypic movements resembling RS. The natural history of these two disorders is quite dierent from that of RS, however, and both can now be diagnosed by appropriate molecular genetic or biochemical testing. The third or pseudostationary stage, which may last for several decades, is reserved for those girls with preserved ability to walk. During this stage, communication functions such as socialization and eye contact may improve remarkably. Conversely, motor function may slowly decline such that ambulation and stereotypic hand movements diminish in speed and frequency. Ambulation often continues into middle age. In stage 3, RS must be dierentiated from the ataxic static encephalopathies, spinocerebellar degeneration, AS, and neuronal ceroid lipofuscinosis as well as idiopathic psychomotor retardation. The fourth or late motor deterioration stage is achieved with the loss of ambulation, that is, when wheelchair dependency is complete. For those girls who never walk, staging moves directly from stage 2 to stage 4. Thus, stage 4 is now subdivided into girls who lose ambulation (4A) and those who never ambulate (4B). Girls in stage 4B are remarkably hypotonic and develop severe motor disability with muscle wasting and skeletal deformities. Despite transition to stage 4, eye contact and socialization may be quite intense even into adulthood.
Variant phenotypic expression Variant phenotypic expressions of RS are well described. The so-called forme fruste is the most common, consisting of delay in onset until the age of 8 to 10 years [15]. Other atypical forms include a preserved speech variant [16], a congenital form in which no period of developmental progress is noted, and an early-onset seizure form [17]. The relatively severe epileptic encephalopathy of the early-onset seizure form results in little or no normal early development.
1129
Hagberg and Skjeldal [18] developed criteria for delineating the variant phenotypes (Table 3). In a Swedish cohort, Hagberg [9] noted that of 130 girls, 82% of cases fullled the classic criteria, 12% were forme fruste, and the remainder included preserved speech or congenital forms.
Rett syndrome in boys and men Convincing descriptions of boys and men with RS are limited. RS has been described in boys with more than one X chromosome (Klinefelter syndrome [XXY]) [1922]. Alternatively, mutations in MECP2 have been identied in male patients with features quite dierent from those of RS. These include individuals with a rapidly progressive encephalopathy [19,2325], severe developmental delay [2628], spastic paraparesis [29], and nonspecic mental retardation [30,31].
Specic clinical issues Several specic clinical issues should be noted. These include longevity, cognitive impairment, seizures, breathing irregularities, scoliosis, ambulation, growth failure, gastrointestinal function, and self-abuse. In general, survival is quite prolonged, and growth failure is pervasive. Longevity Survival into middle age is typical. In the only systematic study, survival in RS followed that of all girls up to the age of 10 years. Survival to the age of 35 years was approximately 70% compared with 98% for all women and 27% for women with profound cognitive psychomotor impairments. These
Table 3 Diagnostic criteria for Rett syndrome variants Main criteria Loss of nger skills Loss of babble/speech Loss of communication skills Deceleration of head growth Hand stereotypies RS disease prole Supportive criteria Irregular breathing Teeth grinding Scoliosis/kyphosis Lower limb amyotrophy Cold purplish feet Laughing/screaming spells Bloating Gait dyspraxia RS electroencephalographic pattern RS eye pointing Pain indierence
RS, Rett syndrome. Inclusion criteria: patient must meet at least 3 of 6 main criteria and 5 of 11 supportive criteria.
1130
data have important implications for counseling parents about long-term care and future planning, because these women may well outlive their parents. Cognitive impairment Cognitive function is dicult to assess in RS. Without eective ne motor and communication skills, available standardized tests are problematic. Test results indicate a mental age at the 8- to 10-month level and gross motor function at the 12- to 18-month level. Assessments, which depend only on visual response, also yield cognitive levels in the severely impaired range. Adaptive skills (feeding, dressing, and toileting) are never acquired eectively, again presenting important implications for long-term care. To maximize their functional level, girls with RS should receive appropriate educational and habilitation service in physical, occupational, and speech therapy, including augmentative communication [32]. Seizures Reports of seizure frequency yield variable rates ranging from 30% to 80% [3336]. After the age of 2 years, EEG ndings are invariably abnormal, featuring slowing in background activity, reduction or loss of posterior dominant rhythm, and recurrent spike and slow spike and wave activity. Clinical seizures may be infrequent or absent in most girls. Dierentiating behavioral patterns from seizures may be dicult without video-EEG monitoring. Seizure control with carbamazepine or sodium valproate is typically not dicult. Lamotrigine has also proved eective [37,38]. Although generally well tolerated, carbamazepine may produce agitation or self-abusive behavior. Caution is recommended regarding the use of sodium valproate because it has been shown recently to inhibit the activity of histone deacetylase (HDAC), a key component in transcription regulation [39]. As such, inhibition of HDAC could exacerbate the eect of a mutation in MECP2 (see below). Breathing irregularities Irregular breathing during wakefulness, consisting of hyperventilation or breath holding, was rst recognized by Rett [4042]. In some instances, both may occur. Breath holding may be prolonged and alarming, occasionally exceeding 1 minute. In other girls, it may be quite subtle and hence underreported. Air swallowing (aerophagia) is common and may produce striking abdominal distention. Distention subsides during sleep. Typically, irregular breathing has its onset in early childhood (35 years) and is maximal during early school age (510 years). It may dominate much of the waking activities. Thereafter, breathing irregularities diminish in
1131
frequency and intensity. Eorts to modify breath holding or hyperventilation have generally been unsuccessful. The opiate antagonist naltrexone may provide benet, but this response is not uniform and may simply reect its sedating properties [43]. Irregular breathing during sleep is not typical of RS and should be assessed for causes of obstructive apnea. Scoliosis The incidence and severity of scoliosis increases with age, being present in approximately 8% of aected girls before school age and in more than 80% over 16 years of age [4448]. The overall incidence is approximately 50%. Onset is typically noted at the age of 8 years or somewhat before. Progression, which is much more common in those girls who do not walk, may become clinically signicant and require medical or surgical attention. Bracing is considered with a 25 curvature, but evidence of eectiveness is lacking. Surgery is recommended strongly when curvature exceeds 40. In general, this procedure is tolerated well. Ambulation Ambulation occurs in 80% of girls with RS, but one quarter of these girls lose their ability to walk during or after the period of regression. Overall, approximately 60% of girls with RS remain ambulatory. Ambulation is encouraged as long as possible; failing this, weight bearing, including the use of standing frames, is encouraged. Bones tend to be undermineralized, and weight bearing may aid this problem. Growth failure Growth failure is pervasive, with the rst evidence being deceleration in the rate of head growth as early as 3 months of age [13,49]. Median head circumference values fall to the second percentile for the normal population by the age of 4 to 5 years. Weight begins to decline from normal near the end of the rst year of life, with the median value falling below the fth percentile for the normal population by the age of approximately 7 years. Height or length falls o around 15 months of age. The median values again reach the fth percentile for the normal population around the age of 7 years. Hand and foot growth is also involved, with the feet affected more than the hands. The reduction in the rate of foot growth parallels that of height [50]. Gastrointestinal function Nutrition can be a major problem in RS. Girls with RS seem to have increased protein requirements [5153]. Gastrostomy feeding may be
1132
necessary in some girls to preserve growth as such. The guidance of a nutritionist with regard to dietary supplementation may be required. Gastroesophageal reux and esophagitis are common and help to account for recurrent periods of irritability or apparent distress. Constipation is another signicant problem. Various strategies, including high-ber foods, enemas, mineral oil, and milk of magnesia, have been variably successful. The frequent use of enemas may lead to dependency on this mode of treatment, and prolonged use of mineral oil may interfere with proper absorption of the fat-soluble vitamins. Despite the availability of avored milk of magnesia, many girls resist taking it. Most recently, polyethylene glycol (Miralax) has proved particularly eective. It is tasteless and odorless and may be dissolved in juice, making it more palatable and much better tolerated. Gallbladder dysfunction has also been recognized as a signicant problem. Systematic study has not been conducted, but the occurrence of this problem, especially in younger children, seems to exceed that in the general population.
Self-abuse Self-abusive behavior, such as hair pulling; biting of the ngers, hands, or other parts of the upper extremities; and hitting themselves about the face, is seen occasionally. Aggressive behavior toward others (hitting, biting, or hair pulling) may also occur. Before considering pharmacologic intervention, care should be taken to exclude other medical problems, particularly gastrointestinal dysfunction (gastroesophageal (G-E) reux, constipation, gallbladder dysfunction) as noted previously or as a side eect of medications already in use. After excluding such problems, low-dose risperidone (0.5 mg twice daily) may be eective.
Other associated features Bruxism (teeth grinding), interrupted sleep patterns, and vasomotor disturbances comprising cold feet and hands are common. Bruxism tends to be most prominent during early childhood and is resistant to medical management. Sleep is often fragmented [40], in some instances, for many nights in succession. This may result in disrupted sleep for the parents. Girls with RS may awaken in the night and be found playing quietly or laughing for no apparent reason. As such, chloral hydrate, diphenhydramine, or hydroxyzine may be required. Vasomotor disturbances seem to represent autonomic nervous system dysfunction. Sympathectomy occurring during surgery for scoliosis has been noted to reverse these ndings on the operated side. No eective treatment is available.
1133
Long-term management Presently, long-term management of girls with RS involves appropriate physical and occupational therapy, speech therapy, nutritional support, orthopedic intervention, and seizure control [9,32]. As noted previously, the potential longevity of women with RS requires long-term planning, because the parents may not be able to manage these needs as they age. Neuropathology The characteristic neuropathologic ndings in RS consist of reduced brain weight, reduced volume of frontal cortex and deep gray nuclei, reduced melanin deposition in the substantia nigra, smaller neurons, reduced dendritic arborizations, and the absence of any recognizable disease process, particularly a progressive neurodegenerative disorder [5458]. As such, the fundamental neurobiologic mechanism seems to be an arrest in normal neuronal maturation, giving RS the prole of a neurodevelopmental and not a neurodegenerative disorder. The brain has a normal appearance but is approximately 60% to 70% of expected weight for age. Brain weights are uniformly low across the age spectrum with no pattern of progressive reduction with increasing age. Neuronal migration is normal, but the neurons are too small, too close together, and have too few processes. Dendritic arbors are markedly shortened and relatively primitive, resulting in increased cell packing density. These ndings suggest a failure in the proper development or maintenance of synaptic connections. Several other neurodevelopmental disorders have similar neuropathologic features. In Down syndrome, dendritic branches are already decient by 4 months of age, remaining so into adulthood [59]; in AS and fragile X syndrome, dendritic arborizations and spines are reduced [60 62]; and in autistic spectrum disorder, increased cell packing density and decreased cell size are typical [63]. The involvement of multiple neurotransmitter systems and the abnormalities in dendrite formation suggest that the impact of cellular dysfunction begins in the last third of gestation or early in infancy. The facts that neuronal proliferation and migration proceed appropriately and that deceleration in the rate of head growth is already present by 3 months of life suggest this temporal boundary. Genetic basis of Rett syndrome RS is a genetic disorder predominantly aecting girls and women and is transmitted in an X-linked dominant fashion with presumed lethality in boys and men. RS is sporadic in more than 99% of aected girls and women; that is, recurrence within families is less than 1%. Thus, for the most part, RS represents new mutations.
1134
Despite the small number of familial cases, linkage studies were able to focus attention to Xq28, a gene-rich region [64,65]. As a result, mutations were identied in the gene MECP2, which encodes MeCP2 [10]. MeCP2 is expressed throughout the body but is highly expressed in brain. An array of MECP2 mutations has been dened in girls or women with RS, but about two thirds of those with a molecular diagnosis are represented by eight specic mutations [10,19,25,6682]. At present, mutations in MECP2 have been identied in 80% to 85% of girls and women with classic RS, a gure which is likely to increase as the gene is sequenced more completely. In the case of variant forms of RS, the proportion with mutations in MECP2 is less than 50%. Girls or women with RS demonstrate a broad range of functional capabilities presumably related to dierences in the specic MECP2 mutation. In terms of phenotypegenotype correlations in girls with classic RS, the most important determinant of clinical severity seems to be variability in X chromosome inactivation (XCI) [83], although the position and type of mutation are also important [19,72,74,83]. Family studies have demonstrated an unexpected spectrum of phenotypic abnormalities. In one family, these ranged from a mild learning disability in the mother and RS in her sister and daughter, to a severe and fatal encephalopathy in her son [23,25]. A similar pattern was noted subsequently in another family [24] and in a boy whose mother was asymptomatic [19]. The transmitting woman in each family had skewed XCI such that most cells expressed the X chromosome with normal MECP2. The range of disorders associated with MECP2 mutations now involves girls and women and well as boys and men (Table 4). Among girls and women, mutations have been identied most often in association with RS and its variants, including those with preserved speech [31,8486], but also in the Angelman phenotype [87], in autistic spectrum disorder [19,30], and in normal women as well as those with a mild learning disability. Other than Klinefelter syndrome [19,20,88], MECP2 mutations have been found in karyotypically normal boys with various forms of mental retardation as well as rapidly progressive encephalopathy as noted previously. MECP2 mutations
Table 4 Phenotypes related to MECP2 mutations Girls/women Rett syndrome Forme fruste Preserved speech variant Delayed-onset variant Angelman syndrome Autistic spectrum disorder Mild learning disability Normal carriers Boys/men Fatal encephalopathy Rett/Klinefelter syndrome X-linked mental retardation/progressive spasticity X-linked mental retardation Somatic mosaicism/neurodevelopmental delay
1135
were found in two 46, XY boys with severe cognitive impairment, macrocephaly, chronic diarrhea, and progressive spasticity [29]. Both mothers and the two boys share a common mutation in MECP2. One mother was normal, but the other had borderline intelligence. In contrast, a more Rett-like phenotype in two boys is explained by somatic mosaicism for the MECP2 gene; that is, some cells express a mutation in MECP2, although other cells have a normal copy of this gene [26,27]. To add further complexity, MECP2 mutations were detected in 4 of 185 boys and men with nonspecic mental retardation [89], a frequency similar to that of fragile X syndrome as the basis for mental retardation in male patients. Karyotypically normal boys and men carrying MECP2 mutations lack all the clinical features of RS, and regression in skills is not noted generally, but one common feature has been the absence of language.
MeCP2 function in developmental neurobiology MeCP2 plays an important role in transcriptional silencing, yet its precise relation to developmental neurobiology is largely unknown [90]. The association of the gene encoding MeCP2 with RS has provided new impetus for such studies. MeCP2 functions in transcriptional silencing by binding to methylated CpG dinucleotides in gene promoter regions. In the mammalian genome, 60% to 90% of the cytosine residues in CpG dinucleotides are methylated, a process established during early embryonic dierentiation [91]. Gene silencing via methylation is operative in such cellular processes as XCI, imprinting, silencing of endogenous retroviruses, and regulation of transcription. Transcriptional silencing in postmitotic cells, such as neurons, may promote ecient cellular function by preventing excess transcriptional noise [90,92]. The linkage between MeCP2 and transcriptional repression involves a corepressor protein, Sin3A, and HDAC. In this model (Fig. 2), MeCP2 mediates gene silencing by attracting HDAC to methylated DNA [93]. The target genes of MeCP2-mediated silencing and the specic role of this process in developmental neurobiology are unknown at present. Mutations in MECP2 such as those seen in RS do interfere with MeCP2 binding to methylated DNA [94]. The common MeCP2 missense mutations R106W and R133C and a third missense mutation result in a 100-fold reduction in anity for methylated DNA, whereas another common missense mutation (T158M) has only a twofold reduction in binding anity. Based on the neuropathology of RS, MECP2 seems to aect neuronal maturation and synaptic development [54,55]. Fragile X syndrome provides an interesting comparison with respect to dendritic arborizations [60,61] and transcriptional silencing [90]. The fragile X syndrome gene, FMR1, has an expansion of the CGG trinucleotide in the 5-untranslated region. This would be predicted to result in excessive methylation of FMR1 and
1136
Fig. 2. Interplay between the methylcytosine-guanosinebinding protein 2 (MeCP2), the corepressor protein Sin3A, and histone deacetylase (HDAC) in transcriptional regulation through the compaction of chromatin. In this model, MeCP2 would bind to methylcytosine. This would mobilize the corepressor protein Sin3A and HDAC. HDAC would then deacetylate histone acetate groups, leading to compaction of chromatin, which would, in turn, downregulate or inhibit transcription of as yet undened genes responsible for RS.
repression of its transcription through histone deacetylation as described previously [9597]. The FMR1 protein (Fmrp) is abundant in neurons and seems to be critical for protein synthesis [98,99]. Thus, in fragile X syndrome, excessive transcriptional silencing leads to a loss of function. In RS, the lack of transcriptional silencing seems to lead to a gain of function in that the downstream genes typically repressed by MeCP2 are no longer regulated properly. The enhanced interest in MECP2 arising from RS opens new vistas regarding specic neurobiologic events and the downstream genes underlying RS on the one hand and normal neurodevelopment on the other. The unfolding panorama of ongoing investigations in RS, including the elaboration of mouse models, should certainly provide important new insights for both [100,101].
1137
Summary RS, the most common cause of profound cognitive impairment in girls and women, is composed of characteristic clinical features, including communication dysfunction, stereotypic movements, and pervasive growth failure. Neuropathologic ndings indicate a failure of neuronal maturation with too small neurons and too few dendritic arbors and no evidence of a progressive neurodegenerative process. The combination of clinical and neuropathologic characteristics presents the prole of a neurodevelopmental disorder. Mutations in the gene MECP2, which encodes MeCP2, have been identied in 80% to 85% of girls and women with RS. Furthermore, the panorama of phenotypes with MECP2 mutations now extends far beyond RS to include normal girls and women, mild learning disability, autistic spectrum disorders, and X-linked mental retardation. These rapid advances in our understanding of RS over the past three decades have opened new avenues of study in developmental neurobiology.
References
[1] Rett A. Uber ein eigenartiges hirnatrophisches Syndrom bei Hyperammonamie im Kindesalter. Wien Med Wochenschr 1966;116:7236. [2] Hagberg B, Aicardi J, Dias K, et al. A progressive syndrome of autism, dementia, ataxia, and loss of purposeful hand use in girls: Retts syndrome: report of 35 cases. Ann Neurol 1983;14:4719. [3] Holm VA. Retts syndrome: a progressive developmental disability in girls. Dev Behav Pediatr 1985;6:326. [4] Kerr AM, Stephenson JB. A study of the natural history of Rett syndrome in 23 girls. Am J Med Genet 1986;24(Suppl 1):7783. [5] Nomura Y, Segawa M, Hasegawa M. Rett syndromeclinical studies and pathophysiological consideration. Brain Dev 1984;6:47586. [6] Percy AK, Zoghbi H, Riccardi VM. Rett syndrome: initial experience with an emerging clinical entity. Brain Dev 1985;7:3004. [7] Hagberg BA. Retts syndrome: prevalence and impact on progressive severe mental retardation in girls. Acta Paediatr 1985;74:4058. [8] Kozinetz CA, Skender ML, MacNaughton N, et al. Epidemiology of Rett syndrome: a population-based registry. Pediatrics 1993;91:44550. [9] Hagberg B. Rett syndromeclinical and biological aspects. London: MacKeith Press; 1993. p. 1120. [10] Amir R, Van den Veyver I, Wan M, et al. Rett syndrome is caused by mutations in X-linked MECP2, encoding methyl-CpG-binding protein 2. Nat Genet 1999;23:1858. [11] Rett Syndrome Diagnostic Criteria Work Group. Diagnostic criteria for Rett syndrome. Ann Neurol 1988;23:4258. [12] Hagberg B, Goutieres F, Hanefeld F, et al. Rett syndrome: criteria for inclusion and exclusion. Brain Dev 1985;7:3723. [13] Schultz RJ, Glaze DG, Motil KJ, et al. The pattern of growth failure in Rett syndrome. Am J Dis Child 1993;147:6337. [14] Hagberg B, Witt-Engerstrom I. Rett syndrome: a suggested staging system for describing impairment prole with increasing age towards adolescence. Am J Med Genet 1986; 24(Suppl 1):4759.
1138
[15] Hagberg B, Rasmussen P. Forme fruste of Rett syndromea case report. Am J Med Genet 1986;24(Suppl 1):17581. [16] Zappella M. The Rett girls with preserved speech. Brain Dev 1992;14:98101. [17] Aicardi J. Rett syndrome. Int Pediatr 1988;3:1659. [18] Hagberg BA, Skjeldal OH. Rett variants: a suggested model for inclusion criteria. Pediatr Neurol 1994;11:511. [19] Hobuhr K, Devaney JM, LaFleur B, et al. MeCP2 mutations in children with and without the phenotype of Rett syndrome. Neurology 2001;56:148695. [20] Leonard H, Silberstein J, Falk R, et al. Occurrence of Rett syndrome in boys. J Child Neurol 2001;16:3338. [21] Schwartzman JS, Zatz M, dos Reis Vasquez L, et al. Rett syndrome in a boy with a 47,XXY karyotype. Am J Hum Genet 1999;64:17815. [22] Vorsanova SG, Demidova IA, Ulas VY, et al. Cytogenic and molecular-cytogenic investigation of Rett syndrome: analysis of 31 cases. Neuroreport 1996;8:1879. [23] Schanen NC, Kurczynski TW, Brunelle D, et al. Neonatal encephalopathy in two boys in families with recurrent Rett syndrome. J Child Neurol 1998;13:22931. [24] Villard L, Cardoso AK, Chelly PJ, et al. Two aected boys in a Rett syndrome family: clinical and molecular ndings. Neurology 2000;55:118893. [25] Wan M, Lee SS, Zhang X, et al. Rett syndrome and beyond: recurrent spontaneous and familial MECP2 mutations at CpG hotspots. Am J Hum Genet 1999;65:15209. [26] Armstrong J, Pineda M, Aibar E, et al. Classic Rett syndrome in a boy as a result of somatic mosaicism for a MECP2 mutation. Ann Neurol 2001;50:692. [27] Clayton-Smith J, Watson P, Ramsden S, et al. Somatic mutation in MECP2 as a non-fatal neurodevelopmental disorder in males. Lancet 2000;356:8302. [28] Imessaoudene B, Bonnefont JP, Royer G, et al. MECP2 mutation in non-fatal, nonprogressive encephalopathy in a male. J Med Genet 2001;38:1714. [29] Meloni I, Bruttini M, Longo I, et al. A mutation in the Rett syndrome gene, MECP2, causes X-linked mental retardation and progressive spasticity in males. Am J Hum Genet 2000;67:9825. [30] Orrico A, Lam C, Galli L, et al. MECP2 mutation in male patients with non-specic X-linked mental retardation. FEBS Lett 2000;481:2858. [31] Vacca M, Filippini F, Budillon A, et al. Mutation analysis of the MECP2 gene in British and Italian Rett syndrome females. J Mol Med 2001;78:64855. [32] Budden SS. Rett syndrome: habilitation and management reviewed. Eur Child Adolesc Psychiatry 1997;6:1037. [33] Glaze D, Frost J, Zoghbi H, et al. Retts syndrome: correlation of electroencephalographic characteristics with clinical staging. Arch Neurol 1987;44:10536. [34] Glaze D, Schultz R, Frost J. Rett syndrome: characterization of seizures and non-seizures. Electroencephalogr Clin Neurophysiol 1998;106:7983. [35] Hagne I, Witt-Engerstrom I, Hagberg B. EEG development in Rett syndrome. A study of 30 cases. Electroencephalogr Clin Neurophysiol 1989;72:16. [36] Steenburg U, Hagberg G, Hagberg B. Epilepsy in a representative series of Rett syndrome. Acta Paediatr 2001;90:349. [37] Stenbom Y, Tonnby B, Hagberg B. Lamotrigine in Rett syndrome: treatment experience from a pilot study. Eur Child Adolesc Psychiatry 1998;7:4952. [38] Uldall P, Hansen FJ, Tonnby B. Lamotrigine in Rett syndrome. Neuropediatrics 1993; 24:33940. [39] Phiel CJ, Zhang F, Huang EY, et al. Histone deacetylase is a direct target of valproic acid, a potent anticonvulsant, mood stabilizer, and teratogen. J Biol Chem 2001;276:3673441. [40] Glaze D, Frost J, Zoghbi H, et al. Retts syndrome: characterization of respiratory patterns and sleep. Ann Neurol 1987;21:37782. [41] Lugaresi E, Cirignotta F, Montagna P. Abnormal breathing in the Rett syndrome. Brain Dev 1985;7:32933.
1139
[42] Southall DP, Kerr AM, Tirosh E, et al. Hyperventilation in the awake state: potentially treatable component of Rett syndrome. Arch Dis Child 1988;63:103948. [43] Percy AK, Glaze DG, Schultz RJ, et al. Rett syndrome: controlled study of an oral opiate antagonist, naltrexone. Ann Neurol 1994;35:46470. [44] Bassett GS, Tolo VT. The incidence and natural history of scoliosis in Rett syndrome. Dev Med Child Neurol 1990;32:9636. [45] Guidera KJ, Borrelli J, Jr, Raney E, et al. Orthopaedic manifestations of Rett syndrome. J Pediatr Orthop 1991;11:2048. [46] Harrison DJ, Webb PJ. Scoliosis in the Rett syndrome: natural history and treatment. Brain Dev 1990;12:1546. [47] Holm VA, King HA. Scoliosis in the Rett syndrome. Brain Dev 1990;12:1513. [48] Lidstrom J, Stokland E, Hagberg B. Scoliosis in Rett syndrome. Clinical and biological aspects. Spine 1994;19:16325. [49] Stenbom Y, Engerstrom IW, Hagberg G. Gross motor disability and head growth in Rett syndromea preliminary report. Neuropediatrics 1995;26:856. [50] Schultz R, Glaze D, Motil K, et al. Hand and foot growth failure in Rett syndrome. J Child Neurol 1998;13:714. [51] Motil K, Shultz R, Wong W, et al. Increased energy expenditure associated with repetitive involuntary movement does not contribute to growth failure in girls with Rett syndrome. J Pediatr 1998;132:22833. [52] Motil KJ, Schultz R, Brown B, et al. Altered energy balance may account for growth failure in Rett syndrome. J Child Neurol 1994;9:3159. [53] Motil KJ, Schultz RJ, Browning K, et al. Oropharyngeal dysfunction and gastroesophageal dysmotility are present in girls and women with Rett syndrome. J Pediatr Gastroenterol Nutr 1999;29:317. [54] Armstrong D, Dunn J, Antaly B, et al. Selective dendritic alterations in the cortex of Rett syndrome. J Neuropathol Exp Neurol 1995;54:195201. [55] Armstrong D, Dunn K, Antaly B. Decreased dendritic branching in frontal, motor and limbic cortex in Rett syndrome compared with Trisomy 21. J Neuropathol Exp Neurol 1998;57:10137. [56] Armstrong D, Dunn K, Schultz R, et al. Organ growth in Rett syndrome: a postmortem examination analysis. Pediatr Neurol 1999;20:1259. [57] Bauman ML, Kemper TL, Arin DM. Microscopic observations of the brain in Rett syndrome. Neuropediatrics 1995;26:1058. [58] Jellinger K, Armstrong D, Zoghbi HY, et al. Neuropathology of Rett syndrome. Acta Neuropathol (Berl) 1988;76:14258. [59] Takashima S, Ieshima A, Nakamura H, et al. Dendrites, dementia and the Down syndrome. Brain Dev 1989;11:1313. [60] Comery TA, Harris JB, Willems PJ, et al. Abnormal dendritic spines in fragile X knockout mice: maturation and pruning decits. Proc Natl Acad Sci USA 1997;94:54014. [61] Hinton VJ, Brown WT, Wisniewski K, et al. Analysis of neocortex in three males with the fragile X syndrome. Am J Med Genet 1991;41:28994. [62] Jay V, Becker LE, Chan FW, et al. Puppet-like syndrome of Angelman: a pathologic and neurochemical study. Neurology 1991;41:41622. [63] Kemper TL, Bauman M. Neuropathology of infantile autism. J Neuropathol Exp Neurol 1998;57:64552. [64] Schanen C, Dahle E, Capozzoli F, et al. A new Rett syndrome family consistent with X-linked inheritance expands the X chromosome exclusion map. Am J Hum Genet 1997; 61:63441. [65] Schanen NC. Molecular approaches to the Rett syndrome gene. J Child Neurol 1999; 14:80614. [66] Amano K, Nomura Y, Segawa M, et al. Mutational analysis of the MECP2 gene in Japanese patients with Rett syndrome. J Hum Genet 2000;45:2316.
1140
[67] Amir R, Zoghbi H. Rett syndrome: methyl-CpG-binding protein 2 mutations and phenotype-genotype correlations. Am J Med Genet 2000;97:14752. [68] Auranen M, Vanhala R, Vosman M, et al. MECP2 gene analysis in classical Rett syndrome and in patients with Rett-like features. Neurology 2001;56:6117. [69] Bienvenu T, Carrie A, de Roux N, et al. MECP2 mutations account for most cases of typical forms of Rett syndrome. Hum Mol Genet 2000;9:137784. [70] Bourdon V, Philippe C, Labrune O, et al. A detailed analysis of the MECP2 gene: prevalence of recurrent mutations and gross DNA rearrangements in Rett syndrome patients. Hum Genet 2001;108:4350. [71] Buyse IM, Fang P, Hoon KT, et al. Diagnostic testing for Rett syndrome by DHPLC and direct sequencing analysis of the MECP2 gene: identication of several novel mutations and polymorphisms. Am J Hum Genet 2000;67:142836. [72] Cheadle J, Gill H, Fleming N, et al. Long-read sequence analysis of the MECP2 gene in Rett syndrome patients: correlation of disease severity with mutation type and location. Hum Mol Genet 2000;9:111929. [73] Dragich J, Houwink-Manville I, Schanen C. Rett syndrome: a surprising result of mutation in MECP2. Hum Mol Genet 2000;9:236575. [74] Hampson K, Woods CG, Latif F, et al. Mutations in the MECP2 gene in a cohort of girls with Rett syndrome. J Med Genet 2000;37:6102. [75] Huppke P, Laccone F, Kramer N, et al. Rett syndrome: analysis of MECP2 and clinical characterization of 31 patients. Hum Mol Genet 2000;9:136975. [76] Kim SJ, Cook EH, Jr. Novel de novo nonsense mutation of MECP2 in a patient with Rett syndrome. Hum Mutat 2000;15:3823. [77] Laccone F, Huppke P, Hanefeld F, et al. Mutation spectrum in patients with Rett syndrome in the German population: evidence of hot spot regions. Hum Mutat 2001; 17:18390. [78] Lebo R, Ikuta T, Milunsky J, et al. Rett syndrome from quintuple and triple deletions within the MECP2 deletion hotspot region. Clin Genet 2001;59:40617. [79] Nielsen JB, Henriksen KF, Hansen C, et al. MECP2 mutations in Danish patients with Rett syndrome: high frequency of mutations but no consistent correlations with clinical severity or with the X chromosome inactivation pattern. Eur J Hum Genet 2001; 9:17884. [80] Obata K, Matsuishi T, Yamashita Y, et al. Mutation analysis of the methyl-CpG binding protein 2 gene (MECP2) in patients with Rett syndrome [letter]. J Med Genet 2000;37: 60810. [81] Van den Veyver IB, Zoghbi HY. Methyl-CpG-binding protein 2 mutations in Rett syndrome. Curr Opin Genet Dev 2000;10:2759. [82] Xiang F, Buervenich S, Nicolao P, et al. Mutation screening in Rett syndrome patients. J Med Genet 2000;37:2505. [83] Amir R, Van den Veyver I, Schultz R, et al. Inuence of mutation type and X chromosome inactivation on Rett syndrome phenotypes. Ann Neurol 2000;47:6709. [84] De Bona C, Zappella M, Hayek G, et al. Preserved speech variant is allelic of classic Rett syndrome. Eur J Hum Genet 2000;8:32530. [85] Inui K, Akagi M, Ono J, et al. Mutational analysis of MECP2 in Japanese patients with atypical Rett syndrome. Brain Dev 2001;23:2125. [86] Zappella M, Meloni I, Longo I, et al. Preserved speech variants of the Rett syndrome: molecular and clinical analysis. Am J Med Genet 2001;104:1422. [87] Watson P, Black G, Ramsden S, et al. Angelman syndrome phenotype associated with mutations in MECP2, a gene encoding a methyl CpG binding protein. J Med Genet 2001;38:2248. [88] Schwartzman JS, Bernardino A, Nishimura A, et al. Rett syndrome in a boy with a 47,XXY karyotype conrmed by a rare mutation in the MECP2 gene. Neuropediatrics 2001;32:1624.
1141
[89] Couvert P, Bienvenu T, Aquaviva C, et al. MECP2 is highly mutated in X-linked mental retardation. Hum Mol Genet 2001;10:9416. [90] Robertson KD, Wole AP. DNA methylation in health and disease. Nat Rev Genet 2000; 1:119. [91] Kafri T, Ariel M, Brandeis M, et al. Developmental pattern of gene-specic DNA methylation in the mouse embryo and germ line. Genes Dev 1992;6:70514. [92] Bird AP. Gene number, noise reduction and biological complexity. Trends Genet 1995; 11:94100. [93] Jones PL, Veenstra GJ, Wade PA, et al. Methylated DNA and MeCP2 recruit histone deacetylase to repress transcription. Nat Genet 1998;19:18791. [94] Ballestar E, Yusufzai TM, Wole AP. Eects of Rett syndrome mutations of the methylCpG binding domain of the transcriptional repressor MeCP2 on selectivity for association with methylated DNA. Biochemistry 2000;39:71006. [95] Coee B, Zhang F, Warren ST, et al. Acetylated histones are associated with FMR1 in normal but not fragile X-syndrome cells [erratum appears in Nat Genet 1999;22:209]. Nat Genet 1999;22:98101. [96] Kremer EJ, Pritchard M, Lynch M, et al. Mapping of DNA instability at the fragile X to a trinucleotide repeat sequence p(CCG)n. Science 1991;252:17114. [97] Oberle I, Rousseau F, Heitz D, et al. Instability of a 550-base pair DNA segment and abnormal methylation in fragile X syndrome. Science 1991;252:1097102. [98] Feng Y, Absher D, Eberhart DE, et al. FMRP associates with polyribosomes as an mRNP, and the I304N mutation of severe fragile X syndrome abolishes this association. Mol Cells 1997;1:10918. [99] Warren ST, Nelson DL. Trinucleotide repeat expansions in neurological disease. Curr Opin Neurobiol 1993;3:7529. [100] Chen RZ, Akbarian S, Tudor M, et al. Deciency of methyl-CpG binding protein-2 in CNS neurons results in a Rett-like phenotype in mice. Nat Genet 2001;27:32731. [101] Guy J, Hendrich B, Holmes M, et al. A mouse Mecp2-null mutation causes neurological symptoms that mimic Rett syndrome. Nat Genet 2001;27:3226.
Neurol Clin N Am 20 (2002) 11431161
Diagnosis and treatment of neurotransmitter-related disorders

Kathryn J. Swoboda, MDa,*, Keith Hyland, PhDb
University of Utah School of Medicine, Primary Childrens Medical Center, Division of Pediatric Neurology, 100 North Medical Drive, Suite 2700, Salt Lake City, UT 84113, USA b Department of Neurochemistry, Institute of Metabolic Disease, Baylor University Medical Center, 3812 Elm Street, Dallas, TX 75226, USA
a
The term neurotransmitter disorders constitutes a broad spectrum of neurologic conditions that share certain clinical features depending on the severity and pattern of neurotransmitter deciency or excess. Isolated exercise-induced dystonia, a gait disorder, or parkinsonian symptoms including oculogyric crises, hypokinesia, hypophonia, and tremor are more common in disorders associated with dopamine deciency. If more than one neurotransmitter is involved, or if an associated tetrahydrobiopterin (BH4) deciency is present, neurologic symptoms are typically more severe and include limb hypertonia or dystonia, axial and truncal hypotonia, abnormal eye movements, global developmental impairment, and autonomic symptoms. Less well recognized phenotypes include those characterized predominantly by encephalopathy, ataxia, and seizures. Screening for this group of disorders has been underused. This is unfortunate in that many patients can benet from directed treatment of their specic neurotransmitter deciency states or associated BH4 deciency, even before reaching a specic diagnosis. Cerebrospinal uid (CSF) neurotransmitter metabolite analysis is necessary to conrm the diagnosis in many of these disorders, but some can be identied by characteristic clinical features, results of phenylalanine loading studies, urine pterin and catecholamine studies, plasma catecholamine studies, or specic enzyme testing in blood cells or skin broblasts to conrm the diagnosis. For ease of classication, these disorders can be divided into four groups:
* Corresponding author. E-mail address: swoboda@genetics.utah.edu (K.J. Swoboda). 0733-8619/02/$ - see front matter 2002, Elsevier Science (USA). All rights reserved. PII: S 0 7 3 3 - 8 6 1 9 ( 0 2 ) 0 0 0 1 8 - X
1144
K.J. Swoboda, K. Hyland / Neurol Clin N Am 20 (2002) 11431161
(1) neurotransmitter deciency states with hyperphenylalaninemia, (2) neurotransmitter deciency states without hyperphenylalaninemia, (3) secondary neurotransmitter deciency states, and (4) disorders of neurotransmitter metabolism (Table 1).
Table 1 Primary neurotransmitter deciency disorders Typical phenotype Elevated plasma phenylalanine PTS deciency DHPR deciency GTPCH deciency Encephalopathy, dystonia, spasticity, axial hypotonia, autonomic symptoms, oculogyric crises, seizures Primapterinuria Benign hyperphenylalaninemia Normal plasma phenylalanine GTPCH deciency ALAAD deciency Exercise-induced dystonia, gait disorder Dystonia, spasticity, axial hypotonia, autonomic symptoms, psychomotor retardation, oculogyric crises Parkinsonian symptoms, psychomotor retardation, behavioral disturbances Gait disturbance, infantile parkinsonism Ataxia, speech delay, hypotonia, psychomotor retardation Orthostatic hypotension, lethargy, ptosis Mild mental retardation, tendency for violent or aggressive behavior Psychomotor retardation, ataxia, seizures 11q22.3-23.3 4p15.31 14q22.1-22.2 AR AR AR Locus Inheritance
10q22
AR
14q22.1-22.2 7p11
AD AR
SPR deciency
2p14-p12
AR
TH deciency TPH deciency
11p15.5 11p15.3-p14
AR, ? AD AR
DBH deciency MAOA deciency
9q34 Xp11.23 XR
SSADH deciency
6p22
AR
PTS, 6-Pyruvoyltetrahydrobiopterin; DHPR, dihydropteridine reductase; GTPCH, GTP cyclohydrolase; ALAAD, aromatic L-amino acid decarboxylase; SPR, sepiapterin reductase; TH, tyrosine hydroxylase; TPH, tryptophan hydroxylase; DBH, dopamine beta-hydroxylase; MAOA, monoamine oxidase A; SSADH, succinic semialdehyde dehydrogenase.
1145
Neurotransmitter deciency states with hyperphenylalaninemia The neurotransmitter deciency and hyperphenylalaninemia in infants in this group arise as a result of defects in BH4 metabolism (Fig. 1). Patients are usually identied by way of elevated phenylalanine levels on newborn screening, as BH4 is required for phenylalanine hydroxylation in the liver. The neurotransmitter deciency results from the lack of BH4, an obligatory cofactor required for the synthesis of the catecholamines and serotonin. Although most academic center-based biochemical genetics clinics that follow children with phenylketonuria (PKU) systematically perform the additional studies required to diagnose this group of disorders, occasionally children are not identied until they have progressive neurologic symptoms or clear evidence of developmental delay despite a phenylalanine-restricted diet. In the past, these patients were referred to as atypical phenylketonurics. In some cases they were missed because they were discharged before an adequate interval to detect elevated phenylalanine levels. Approximately 1%3% of patients with hyperphenylalaninemia have an associated BH4 deciency state; thus it is imperative to identify such children so that BH4 and neurotransmitter precursors can be supplemented as early as possible. The two most commonly identied disorders in these children are 6-pyruvoyltetrahydropterin synthase or PTS deciency (that results in inadequate BH4 synthesis) and dihydropteridine reductase (DHPR) deciency (that results in decreased regeneration of BH4 from dihydrobiopterin). Both are autosomal recessive disorders in which hyperphenylalaninemia results from a deciency of BH4. Because of the involvement of BH4 in catecholamine and serotonin synthesis, such infants also have a manifest deciency of neurotransmitter metabolites in addition to hyperphenylalaninemia. Other conditions in this category include autosomal recessive GTP cyclohydrolase deciency and primapterinuria. 6-Pyruvoyltetrahydropterin synthase (6-PTS or 6-PTPS) deciency 6-pyruvoyl tetrahydropterin synthase (6-PTPS, 6-PTS, or PTS) catalyzes the elimination of inorganic triphosphate from dihydroneopterin triphosphate to form 6-pyruvoyltetrahydropterin. Thus, patients have elevated neopterin to biopterin ratios in urine and plasma. Reduced PTS activity can be documented in red blood cells. In the classic form of the disorder, patients have reduced catecholamine and serotonin metabolites, and an increased neopterin to biopterin level in CSF. These patients are usually picked up on newborn screening as phenylketonurics, and show progressive signs of neurologic involvement in the rst few months of life, including extrapyramidal signs, axial and truncal hypotonia, hypokinesia, feeding diculties, choreoathetotic or dystonic limb movements, and autonomic symptoms. Many of these patients, despite early diagnosis and supplementation with
1146 K.J. Swoboda, K. Hyland / Neurol Clin N Am 20 (2002) 11431161
Fig. 1. Synthesis and catabolism of catecholamine and indoleamine neurotransmitters. GTPCH-GTP cyclohydrolase, PTS-6-pyruvoyl-tetrahydropterin synthase, SPR-sepiapterin reductase, DHPR-dihydropteridine reductase, PCD-pterin 4a-carbinolamine dehydratase, PAH-phenylalanine hydroxylase, THtyrosine hydroxylase, TPH-tryptophan hydroxylase, NOS-nitric oxide synthase, AADC-aromatic L-amino acid decarboxylase, B6-pyridoxine COMTcatechol-o-methyltransferase, MAO-monoamine oxidase, DBH-dopamine hydroxylase.
1147
BH4 and neurotransmitter precursors, continue to manifest delays in development [1]. A peripheral form of the disorder is characterized by normal central neurotransmitter levels, and less signicant or transient hyperphenylalaninemia [2]. Patients with the peripheral form have an excellent prognosis for normal neurologic development, provided the hyperphenylalaninemia is corrected by diet or BH4 administration. Dihydropteridine reductase deciency Dihydropteridine reductase (DHPR) deciency manifests in a variety of phenotypes, all with hyperphenylalaninemia. The clinical presentation is similar to that observed with PTS deciency. Without folinic acid to restore methyltetrahydrofolate status in the CNS, these patients can have progressive calcication of the basal ganglia and subcortical regions, despite treatment with BH4 and neurotransmitter precursors [3]. A juvenile variant has been reported in which siblings were developmentally normal until 6 years of age, at which time they developed progressive encephalopathy, epilepsy, and pyramidal, cerebellar, and extrapyramidal features on clinical examination [4]. Diagnosis can be conrmed by the pattern of urine pterins, and documentation of abnormal DHPR activity in skin broblasts [5]. Phenylalanine loading tests are abnormal, and phenylalanine status improves or normalizes with BH4 supplementation. CSF neurotransmitter and pterin analysis reveals reduced HVA, 5-HIAA, decreased or normal BH4, and elevated dihydrobiopterin levels. Autosomal recessive guanosine triphosphate cyclohydrolase deciency Although most mutations in guanosine triphosphate (GTP) cyclohydrolase to date have been documented in association with autosomal dominant dopa-responsive dystonia, or Segawas disease, these patients do not have hyperphenylalaninemia on routine plasma screening studies. Patients with the autosomal recessive form of GTP cyclohydrolase deciency, however, present in a similar fashion to patients with DHPR and PTS deciency. Such patients have severe global developmental impairment, marked hypotonia of the trunk and axial muscles, limb hypertonia, convulsions, and autonomic symptoms including temperature dysregulation, excessive diaphoresis, and blood pressure lability caused by the associated catecholamine deciency. They typically have absent GTP cyclohydrolase activity in blood cells, liver, and skin broblasts. By contrast, patients with autosomal dominant doparesponsive dystonia are believed to have preservation of some GTP cyclohydrolase activity in liver because of their heterozygous state, enough to maintain normal phenylalanine levels under usual circumstances. CSF neurotransmitter analysis reveals low levels of HVA and 5-HIAA, and low neopterin and biopterin levels.
1148
Pterin-4-alpha-carbinolamine dehydratase deciency (primapterinuria) Pterin-4-alpha-carbinolamine dehydratase deciency, or primapterinuria, is a cause of mild hyperphenylalaninemia [6,7]. These infants are usually identied on newborn screening, but generally have a benign course with normal development. Phenylalanine hydroxylase catalyzes the conversion of phenylalanine to tyrosine during which BH4 is converted to the unstable carbinolamine, 4a-hydroxytetrahydrobiopterin [8]. Carbinolamine dehydratase catalyzes the dehydration of carbinolamine to quinonoid dihydropterin qBH2. The decreased rate of dehydration is responsible for the production of 7-biopterin in some mildly hyperphenylalaninemic individuals [9]. Urine studies reveal an excess of 7-substituted pterins and reduced biopterin levels, and an increased neopterin to biopterin ratio. Role of BH4 in the central nervous system Because BH4 is required for the hydroxylation of aromatic amino acids, its importance to the central nervous system (CNS) becomes immediately apparent. Tyrosine and tryptophan are required for the synthesis of catecholamines and serotonin. A BH4-dependent process can be strongly suspected when normalization of plasma phenylalanine levels occurs following BH4 supplementation. A dose of 5 mg/kg BH4 is the usual recommended dose for correcting peripheral hyperphenylalaninemia. Because BH4 crosses the bloodbrain barrier poorly, however, lifelong supplementation with the neurotransmitter precursors L-dopa and 5-HTP, along with carbidopa to enhance CNS delivery, is necessary in most of the disorders mentioned earlier. A much higher dose of BH4, approximately 20 mg/kg, can normalize CSF BH4 levels, but is prohibitively expensive, and no studies exist as to the possible additional benet of such a regimen. The higher requirement of tyrosine hydroxylase for BH4 in comparison with tryptophan hydroxylase may explain the more severe impairment in the catecholaminergic system compared with the serotonergic system. The enzyme nitric oxide synthase also has an absolute requirement for BH4 for the oxidation of arginine to nitric oxide. The inability to replete normal levels of BH4 in the CNS with oral administration in its currently used doses may be one reason many children with BH4 decient disorders develop lifelong cognitive and developmental impairments despite other treatments. Nitric oxide is believed to play a critical role of neuroprotection in the CNS, and reduced eciency of this enzyme in the CNS may result in additional ongoing neuronal injury or cell death and vascular dysregulation and injury. Neurotransmitter deciency states without hyperphenylalaninemia Neurotransmitter deciency disorders not associated with hyperphenylalaninemia possess an increasingly complex spectrum of clinical phenotypes,
1149
ranging from ataxia and mental retardation to spastic diplegia to exerciseinduced dystonia. The lack of ascertainment by way of newborn screening and wide variance in phenotypes makes these disorders as a group much more challenging to recognize. Other than autosomal dominant doparesponsive dystonia, the remaining disorders in this category are all inherited in an autosomal recessive fashion with the exception of MAO-A deciency, a rare X-linked recessive disorder. Generally, heterozygous carriers for mutations in this latter group do not have a discernable phenotype, with rare exceptions in tyrosine hydroxylase deciency and combined MAO-A and -B deciency. Segawas disease, or autosomal dominant dopa-responsive dystonia The most well described and widely identied entity among this group of disorders is autosomal dominant dopa-responsive dystonia caused by GTP cyclohydrolase deciency, or Segawas disease [10]. Identication and treatment of this disorder can be extremely rewarding because patients often benet greatly with low dose L-dopa/carbidopa or even an anticholinergic agent [11; personal observations]. Patients with the classic presentation of exercise-induced dystonia are not dicult to recognize. This diagnosis should also be considered in patients with spastic diplegia, however, especially when signicant uctuation in gait or worsening gait at the end of the day is noted, and in patients with more atypical presentations, including writers cramp or asymmetric limb dystonia. In patients with classic presentations, the clinician frequently makes the diagnosis on a presumptive basis, after observing remission of symptoms with a trial of L-dopa/carbidopa. Although inheritance is autosomal dominant, penetrance is incomplete, and many reports exist about variable expressivity in dierent members of the same family [12]. In a family we have had the opportunity to follow for several years, the mother had spastic diplegia, one son had torticollis, and another son manifested a more typical exercise-induced dystonia phenotype. The female-to-male ratio in sporadic cases is 4:1, and investigators have conrmed increased penetrance of GTPCH I mutations in females [13,14]. Recently, we diagnosed a 55-year-old woman with a life-long history of a gait disorder and excessive fatigue with exertion. Her gait was suggestive of a hysterical gait disorder, strongly supported by her depressed, anxious aect. CSF analysis revealed, however, markedly decreased HVA, 5-HIAA, and BH4 levels; her phenylalanine loading test was consistent with heterozygote status for GTP cyclohydrolase activity, and skin broblast analysis revealed no enzyme activity. Her son has a similar history of gait disorder and depression and has attempted suicide on several occasions. There may be an increased burden of attentional diculties, dysphoria, or depression in some families, although this has not been systemically evaluated. CSF neurotransmitter metabolite and pterin studies are helpful in conrming the diagnosis in these patients, although they are not mandatory
1150
because GTPCH activity can be measured directly from patient broblasts. If one intends to perform CSF analysis, it should be obtained before institution of a treatment trial of L-dopa/carbidopa, because treatment results in increased levels of HVA and 3-O-methyldopa. It is most important to perform these studies in patients with cognitive impairment or atypical presentation, because they may well have an alternative neurotransmitter deciency disorder. The typical pattern in CSF in dopa-responsive dystonia is a low HVA level, normal or low 5-HIAA level, and reduced BH4 levels in CSF. Patients who are heterozygous for a GTPCH mutation, despite their normal blood phenylalanine levels on routine screening, can be shown to have abnormal phenylalanine metabolism if stressed by administration of an oral phenylalanine load (100 mg/kg). Therefore, if cytokine stimulated broblast enzyme analysis of biopterin metabolism is not feasible or patients decline a CSF examination and have an otherwise typical presentation, an oral phenylalanine load, with serial serum phenylalanine levels over the following 4 to 6 hours, can help conrm the clinical diagnosis. Aromatic L-amino acid decarboxylase, or dopa-decarboxylase deciency Aromatic L-amino acid decarboxylase is a pyridoxine dependent enzyme that decarboxylates L-dopa and 5-HTP to make dopamine and serotonin respectively. Patients with this disorder typically present in the rst few months of life with dystonia or intermittent limb spasticity, axial and truncal hypotonia, oculogyric crises, autonomic symptoms, ptosis, and psychomotor retardation [15]. As they get older, ataxia and expressive speech impairment are prominent features. CSF neurotransmitter metabolites show a characteristic pattern with low HVA and 5-HIAA, markedly elevated 3-Omethyldopa, 5-hydroxytryptophan and L-dopa, and normal biopterin and neopterin levels. Plasma L-dopa is markedly elevated. Urine catecholamines may be reduced or elevated [16,17]. These patients respond variably to treatment, with reported benet in some patients with monoamine oxidase inhibitors, dopamine receptor agonists, L-dopa, and pyridoxine [18]. All patients reported to date have had some degree of cognitive impairment, however. Most children with this disorder are not identied until at least 4 to 6 months of age. Sepiapterin reductase (SPR) deciency Sepiapterin reductase catalyzes the NADPH-dependent reduction of carbonyl derivatives, including pteridines, and plays an important role in BH4 biosynthesis. Somewhat surprisingly, the rst identied cases had normal plasma phenylalanine and urine pterin levels [19]. Blau et al have hypothesized that peripheral tissues can use alternative carbonyl, aldose, and dihydrofolate reductases to perform the last two steps in BH4 biosynthesis.
1151
Therefore, BH4 levels in the liver are likely to be normal, probably explaining the absence of hyperphenylalaninemia in these patients. In addition, it is likely that low dihydrofolate reductase activity in the brain allows accumulation of dihydrobiopterin that inhibits tyrosine and tryptophan hydroxylases, and uncouples nitric oxide synthase (nNOS), leading to neurotransmitter deciency and neuronal cell death. Thus, identication of low CSF neurotransmitter levels and the presence of elevated CSF dihyrobiopterin is crucial for making the diagnosis in these patients. Few patients have been reported to date [20,21]. Dystonic posturing, oculogyric crises, spasticity, tremor, and ataxia with recurrent falls were reported in one 9-year-old boy. He also had a depressed aect, aggressive behavior, and psychomotor retardation. Another child had psychomotor retardation, microcephaly, growth deciency, spasticity, and dystonia. Blau et al recently conrmed this diagnosis in a 22-year-old woman with cerebral palsy and lifelong cognitive impairment and a gait disorder, who had been having increasing diculties with head control, excessive fatigue and dystonia, with associated parkinsonian features including gait instability, hypophonia, and pallilalia. She had signicant diurnal variation of symptoms, with symptoms much worse in the evenings. She beneted greatly with supplemental 5-hydroxytryptophan (HTP) and L-dopa/carbidopa, but later discontinued the 5-HTP because of side eects. CSF neurotransmitter metabolite and pterin analysis reveals low levels of HVA and 5-HIAA, and high levels of biopterin and dihydrobiopterin. Diagnosis can be conrmed by documenting low SPR activity in skin broblast cultures. Tyrosine hydroxylase deciency, infantile parkinsonism, or autosomal recessive dopa-responsive dystonia Tyrosine hydroxylase deciency, sometimes referred to as autosomal recessive Segawas disease, displays a diverse phenotype, ranging from exercise-induced dystonia to progressive gait disturbance and tremor in childhood to severe infantile parkinsonism [2224]. An aberrant mRNA splice form is also over represented in progressive supranuclear palsy patients [22]. Low tyrosine hydroxylase activity results in signicant CSF catecholamine deciency as demonstrated by low HVA concentrations; CSF concentrations of S-HIAA, neopterin, and biopterin are normal. The children reported to date seem to have a paucity of autonomic features, suggesting a compensatory peripheral mechanism. In four patients whom the authors have tested, including one patient with the severe infantile variant of parkinsonism, peripheral plasma catecholamine levels have been normal, although reduced urine HVA levels have been noted. Patients variably respond to L-dopa/carbidopa, and some have complete reversal of symptoms. The exception to this is the patient with the severe infantile parkinsonism form. These patients sometimes tolerate L-dopa poorly, with excessive dyskinesia, irritability, and reux [25,26] personal
1152
observations. Slow institution of small doses of L-dopa/carbidopa, along with selegiline (MAO-B inhibitor) and an anticholinergic agent such as trihexyphenidyl, may be more benecial than L-dopa/carbidopa alone [27; personal observations]. When diagnosis occurs late in such patients, motor development must be recapitulated, and continued slow improvement over months is to be expected. One such patient whom we diagnosed at approximately 2 years of age had rigidity, tremor, bradykinesia, oculogyric crises, and severe psychomotor delay. Treatment with L-dopa/carbidopa alone caused severe dyskinesias with marked on-o eects. Addition of selegiline and then trihexyphenidyl provided signicant benet and led to attainment of motor skills and ability to ambulate independently over the next 3 years. Patients with a mild form of the disorder, such as an isolated gait disorder or exercise-induced dystonia, respond well to monotherapy with L-dopa/ carbide and rarely develop dyskinesia. Although inheritance to date in most families seems to be recessive, at least one family has been described in which the father of the aected proband had mild exercise-induced dystonia responsive to therapy, raising the possibility that tyrosine hydroxylase deciency could present in an autosomal dominant fashion in some families with a milder phenotype [28]. Tryptophan hydroxylase deciency Tryptophan hydroxylase (TPH) catalyzes the BH4-dependent hydroxylation of tryptophan to 5-HTP, which is then decarboxylated to form serotonin. TPH expression is limited to certain cells in the CNS and periphery, including raphe neurons, pinealocytes, mast cells, mononuclear leukocytes, beta cells of the islets of Langerhans, and enterochroman cells of the gut. Patients with presumed TPH deciency have recently been reported, although it is not yet certain that their symptoms result from TPH deciency [29]. Clinical features, consisting of ataxia, speech delay, mild psychomotor retardation, and hypotonia, are nonspecic. CSF neurotransmitter metabolite and pterin studies demonstrate the expected low 5-HIAA with normal HVA, neopterin, and biopterin levels. Mutations in the TPH gene have not yet been identied. Dopamine beta-hydroxylase deciency Dopamine beta-hydroxylase (DBH) is the enzyme that converts dopamine to norepinephrine. Presenting symptoms of this disorder have been largely attributed to the importance of this enzyme in postganglionic sympathetic neurons [30]. Patients with severe deciency of this enzyme, however, cannot synthesize norepinephrine, epinephrine, and octopamine in either the CNS or the peripheral autonomic neurons. Dopamine acts as a false neurotransmitter for noradrenergic neurons. Neonates with DBH deciency can have episodic hypothermia, hypoglycemia, and hypotension, but survivors then
1153
do fairly well until late childhood when overwhelming orthostatic hypotension profoundly limits their activities. The hypotension can be so severe as to lead to convulsive syncope with recurrent clonic seizures [31]. Most patients reported to date have been identied as young adults. Observation of severe orthostatic hypotension in a patient whose plasma norepinephrine/dopamine ratio is much less than one supports the diagnosis. Orthostatic hypotension, particularly after exercise, and ptosis are constant features. General lethargy and lassitude improve dramatically and blood pressure normalizes with treatment with d-l-threo-dihydroxyphenylserine, a synthetic amino acid that is converted to norepinephrine by aromatic L-amino acid decarboxylase. Whether or not these patients suer from attention problems or other subtle cognitive decits has not been adequately studied. Patients may undergo personality change, becoming more aggressive with treatment. Monoamine oxidase deciency Monoamine oxidase is a mitochondrial enzyme involved in the catabolism of biogenic amines. Monoamine oxidase A (MAO-A), the primary type in broblasts, preferentially degrades serotonin and norepinephrine. Monoamine oxidase B (MAO-B), the primary type in platelets and in the brain, preferentially degrades phenylethylamine and benzylamine. These enzymes are critical in the neuronal metabolism of catecholamine and indoleamine neurotransmitters. The genes are closely linked on the X-chromosome, near the Norrie disease locus, and only aected boys have been identied to date [32]. Comparisons of the neurochemical characteristics of previously described patients with combined MAO-A and MAO-B deciency and selective MAO-A deciency have led to an improved understanding of the roles of MAO-A and MAO-B in the metabolic degradation of catecholamines and other biogenic amines, including serotonin and the trace amines. Monoamine oxidase A deciency Brunner reported a family with an X-linked nondysmorphic mild mental retardation and a tendency to aggressive or violent behavior including arson, attempted rape, exhibitionism, and attempted suicide [33]. Urine studies revealed marked disturbance of monoamine metabolism. Normal platelet MAO-B activity suggested that the unusual behavior pattern in this family might be caused by isolated MAO-A deciency, which was later conrmed by identication in all aected males of a point mutation leading to premature termination of the protein [34]. Measurement of MHPG (3-methoxy, 4-hydroxyphenolglycol, a metabolite of norepinephrine) in plasma is the most sensitive index of MAO-A activity in humans, and can be used to screen potential cases. MAO-A enzyme activity can be measured directly from broblasts, however. The
1154
inability to identify additional patients despite screening in at-risk males with a mental retardation or behavioral phenotype makes it likely this disorder is rare [10]. Interestingly, a high activity MAO-A promoter allele has been found with increased frequency in women with panic disorder [35]. A possible association of decreased enzyme activity in women with bipolar disorder has also been reported [36]. Monoamine oxidase B deciency Isolated MAO-B deciency has not yet been reported in a patient. Two brothers with a microdeletion including the Norrie locus and MAO-B, however, had features consistent with Norrie disease alone, with congenital blindness and progressive hearing loss caused by cochlear degeneration in adolescence. These patients had neither abnormal behavior nor mental retardation, leading the authors to conclude that MAO-A plays a more signicant role than MAO-B in the metabolism of biogenic amines, and MAOB deciency alone may have a primarily neurochemical phenotype: that of increased phenylethylamine in urine [37]. Monoamine oxidase A and B deciency Conclusions regarding the phenotype of individuals with combined deciency of MAO-A and MAO-B come primarily from a study of a single boy with a microdeletion at Xp14 involving the Norrie locus and documented severe deciency of MAO-A and MAO-B activity [38]. He was severely mentally retarded, blind, and had other neurologic features including myoclonus and tendency for motor stereotypies. Because Norrie disease is an X-linked recessive disorder, obligate carriers would not be expected to have symptoms. In this family, two obligate carriers had normal IQ testing. The probands mother had psychiatric symptoms characterized by chronic hypomania and schizotypal features, however, and both carriers had low MAO activity. Succinic semialdehyde dehydrogenase deciency Succinic semialdehyde dehydrogenase (SSADH) deciency is an autosomal recessive inborn error of metabolism associated with a defect in the metabolism of 4-gamma-aminobutyric acid or GABA [39]. Phenotypic features range from nonspecic global developmental delay and hypotonia to ataxia, severe mental retardation, visual impairment, and seizures. Urine organic acid screening to detect elevated 4-hydroxybutyric acid is the most easily available screening strategy, but GABA levels in CSF and urine are also elevated. Recently, improvement of seizures in a mouse model of the disorder was demonstrated with treatment with vigabatrin or a GABA B
1155
receptor antagonist [40]. The relevance of these ndings to treatment of patients, if any, is not yet known. Secondary neurotransmitter deciency states Menkes disease is an X-linked recessive disorder in which aected males have progressive encephalopathy, spasticity, seizures, and sparse, brittle hair. The primary defect is reduced or absent function of the copper transporting ATPase, ATP7A. Multiple copper-dependent enzymes can be secondarily aected, including DBH, leading to secondary autonomic involvement and norepinephrine deciency. Hyperekplexia, or startle disease is a heterogeneous disorder caused by defects in the alpha-1 subunit of the glycine receptor [41]. The disorder occurs in autosomal dominant and recessive forms, and is characterized predominantly by stimulus-sensitive myoclonus. Transient hypertonia and hypokinesia in infancy in some families with the disorder has led to the designation sti baby syndrome. Dubowitz and colleagues reported an infant with classic startle disease in whom the CSF concentrations of gamma-aminobutyric acid (GABA) were substantially lower than normal during the rst weeks of life [42]. Infants with hyperekplexia have higher than expected rates of sudden infant death syndrome. Later in life patients develop involuntary myoclonus, markedly hyperactive brain-stem reexes, and a momentary generalized jerking on falling asleep. An exaggerated startle response persists throughout life; sudden, unexpected acoustic or tactile stimuli can precipitate a brief attack of intense rigidity with falling. Congenital dislocation of the hip and inguinal and abdominal hernias, presumably caused by increased intra-abdominal pressure, are more frequent in aected families. Dramatic improvement of symptoms occurs in most patients with clonazepam. Neurodegenerative disorders associated with ongoing cell loss are sometimes associated with reductions in neurotransmitter metabolites. We have seen such abnormalities in patients with leukodystrophy and progressive encephalopathy phenotypes in which we have been unable to identify a primary defect in neurotransmitter or pterin metabolism. Certain patients may still benet from directed treatment of the underlying neurotransmitter deciency. For example, in one young patient with an otherwise undened leukodystrophy, supplementation with L-dopa/carbidopa markedly ameliorated his lower limb spasticity. Nonetheless, he continued to show progressive neurologic involvement with time, as expected. Periods of hypoxia or ischemia can lead to secondary deciencies of serotonin and dopamine as demonstrated by low levels of HVA and 5HIAA in CSF. In addition, BH4 levels are low and neopterin levels can be elevated. This presents a confusing pattern that mimics the metabolite prole observed in 6PTPS deciency. The absence of hyperphenylalaninemia and the presence of signal abnormalities in the basal ganglia, thalamus,
1156
and cortex consistent with hypoxic-ischemic encephalopathy allows dierentiation. Undened neurotransmitter deciency states Several patients with documented neurotransmitter deciency states do not t easily into any of the above diagnostic categories, and the nature of their underlying defects remains unknown. One such example was a 21-year-old woman with mild global encephalopathy and stimulus-sensitive myoclonus, with low CSF HVA and 5-HIAA levels, but normal biopterin, neopterin, L-dopa, 5-HTP, and 3-O-methyldopa levels. She had a normal phenylalanine loading study. Another example is a 45-year-old man with apparent congenital cerebellar hypoplasia and stimulus-sensitive myoclonus. CSF studies revealed low HVA levels and biopterin levels, but he had a normal phenylalanine loading study. He has responded clinically to treatment with L-dopa/carbidopa. Additional studies are needed to determine the precise defects aecting neurotransmitter levels and to ascertain whether they are primary or secondary. In addition to neurotransmitter deciency, we have also seen excess levels of neurotransmitter metabolites in some cases. The etiology in these patients remains uncertain at this time, but may imply an underlying receptor defect, with secondary up-regulation. Weve identied one young boy with microcephaly, mental retardation, and progressive spastic paraparesis. CSF neurotransmitter metabolites and urine catecholamines were elevated, but the exact etiology of his disorder is unknown. A careful history of medications or herbal supplements is important, because some agents, such as serotonin reuptake inhibitors, could theoretically increase serotonin metabolite levels. Overall, CSF neurotransmitter metabolite and pterin assays provide powerful tools to help better characterize patients with otherwise undened or poorly dened neurologic disorders. Approach to treatment in patients with neurotransmitter deciency states Because patients with neurotransmitter deciency disorders caused by tyrosine hydroxylase or BH4 deciency have been decient for prolonged periods before treatment, they can be extremely sensitive to initiation of neurotransmitter precursors. Starting with extremely conservative dosages, increasing the dosage slowly over weeks or months, and ensuring that peripheral aromatic L-amino acid decarboxylase is fully blocked by providing ample carbidopa can make the transition to treatment much easier. The rate or degree to which children respond depends on a variety of factors including age of diagnosis, specic disorder and mutation, presence or absence of associated hyperphenylalaninemia, and presence or absence of central BH4 deciency. In general, optimism regarding improvement is warranted.
1157
Institution of neurotransmitter precursor treatment may lead to new problems, such as intermittent dyskinesia related to a peak dose eect, changes in appetite, gastroesophageal reux, diarrhea, or constipation. These problems, greatest in the rst few weeks of institution of treatment, tend to improve with time. With regard to replacement of L-dopa, for instance, we have found that many children respond best to use of a slow release form of the medication, but such formulations were not created for use in children, rather for use in adults with Parkinsons disease. Support for parents and children during this often dicult period of transition from initiation of treatment to adjustment of medications is critical, because these patients will likely require neurotransmitter precursor replacement throughout their lifetimes. In a disorder such as ALAAD, in which direct receptor agonists may be indicated, only adult formulations of these often potent medications are available, making the use of compounding necessary. Thus, giving more frequent and lower doses throughout the day may be necessary in some children. Although patients with primary neurotransmitter deciency states are more likely to respond optimally to treatment, patients with secondary neurotransmitter deciency may have some symptomatic benet from directed treatment of their underlying neurotransmitter deciency state. Neurologic disorders characterized by excess neurotransmitter levels: nonketotic hyperglycinemia and leukoencephalopathy with vanishing white matter Nonketotic hyperglycinemia (NKH) or glycine encephalopathy is a heterogeneous disorder associated with insucient activity of various components of the mitochondrial glycine cleavage system. The enzyme system for cleavage of glycine is composed of four protein components: P protein, a pyridoxal phosphate-dependent glycine decarboxylase, H protein, a lipoic acid-containing protein, T protein, a tetrahydrofolate-requiring enzyme; and L protein, a lipoamide dehydrogenase. NKH may be caused by a defect in any one of these enzymes. It is an autosomal recessive disorder with several reported phenotypes, including the classic severe neonatal form, an infantile variant, a mild-episodic childhood variant, a late-onset form, and a benign reversible form [43]. Most patients described to date have the neonatal and most severe phenotype, likely because it is the most distinctive phenotype. These patients present shortly after birth with lethargy, encephalopathy, hypotonia, myoclonic jerks, and apnea. EEG generally shows a burst suppression pattern. Those who survive the neonatal period generally develop intractable seizures and profound mental retardation. Patients with the infantile form have seizures and variable cognitive impairment following a short period of apparently normal development. In the mild-episodic form, patients typically present sometime after infancy with mild psychomotor retardation and may
1158
manifest episodes of delirium, chorea, and vertical gaze palsy during febrile illness. In the late-onset form, children present with progressive spastic diplegia and optic atrophy. They generally do not have seizures, and intellectual function is preserved. Diagnosis is best made by documenting an increased CSF to plasma glycine ratio [44]. In the neonatal form of NKH, CSF glycine can be 30 times normal levels. Plasma glycine is also typically high, but can be in the normal range. A CSF/plasma ratio of >0.08 is usually considered diagnostic, but mildly aected cases can have ratios of 0.04 to 0.1 [44]. (In rare cases, CSF to plasma ratios are normal and only elevations in plasma glycine occur [51].) Conrmation of diagnosis requires enzyme analysis in liver or transformed lymphoblasts [45]. Treatment with dextromethorphan and sodium benzoate treatment have led to variable improvement in improving seizure control and behavioral problems in some patients [46,47]. Leukoencephalopathy with vanishing white matter (also known as childhood ataxia with central white matter hypomyelinization) is a recently identied leukodystrophy [48]. This is a heterogeneous autosomal recessive disorder characterized by progressive ataxia and motor impairment and encephalopathy in which episodic deterioration is associated with infection or minor head trauma. A wide range of onset has been reported in the dozen patients reported to date. CSF glycine is elevated, and can be helpful in conrming the diagnosis [49]. At least two genes have been recently noted to have mutations in patients with this disorder, EIF2B5 and EIF2B2, the rst translation initiation factors implicated in human disease [50].
References
[1] Dudesek A, Roschinger W, Muntau AC, Seidel J, Leupold D, Thony B, et al. Molecular analysis and long-term follow-up of patients with dierent forms of 6-pyruvoyl-tetrahydropterin synthase deciency. Europ J Pediat 2001;160:26776. [2] Niederwieser A, Shintaku H, Leimbacher W, Curtius H-C, Hyanek J, Zeman J, et al. Peripheral tetrahydrobiopterin deciency with hyperphenylalaninaemia due to incomplete 6-pyruvoyl tetrahydropterin synthase deciency or heterozygosity. Europ J Pediat 1987; 146:22832. [3] Woody RC, Brewster MA, Glasier C. Progressive intracranial calcication in dihydropteridine reductase deciency prior to folinic acid therapy. Neurology 1989;39:6735. [4] Larnaout A, Belal S, Miladi N, Kaabachi N, Mebazza A, Dhondt JL, et al. Juvenile form of dihydropteridine reductase deciency in 2 Tunisian patients. Neuropediatrics 1998; 29:3223. [5] Milstien S, Kaufman S, Summer GK. Hyperphenylalaninemia due to dihydropteridine reductase deciency: diagnosed by measurement of oxidized and reduced pterins in urine. Pediatrics 1980;65:80610. [6] Adler C, Ghisla S, Rebrin I, Heizmann CW, Blau N, Curtius H-C. Suspected pterin-4acarbinolamine dehydratase deciency: hyperphenylalaninaemia due to inhibition of phenylalanine hydroxylase by tetrahydro-7-biopterin. J Inherit Metab Dis 1992;15: 4058.
1159
[7] Blau N, Dhondt JL, Guibaud P, Kuster T, Curtius H-C. New variant of hyperphenylalaninaemia with excretion of 7-substituted pterins [letter]. Europ J Pediat 1988; 148:176. [8] Citron BA, Kaufman S, Milstien S, Naylor EW, Greene CL, Davis MD. Mutation in the 4a-carbinolamine dehydratase gene leads to mild hyperphenylalaninemia with defective cofactor metabolism. Am J Hum Genet 1993;53:76874. [9] Thony B, Neuheiser F, Kierat L, Blaskovics M, Arn PH, Ferreira P, et al. Hyperphenylalaninemia with high levels of 7-biopterin is associated with mutations in the PCBD gene encoding the bifunctional protein pterin-4a-carbinolamine dehydratase and transcriptional coactivator (DCoH). Am J Hum Genet 1998;62:130211. [10] Segawa M, Hosaka A, Miyagawa F, Nomura Y, Imai H. Hereditary progressive dystonia with marked diurnal uctuation. Adv Neurol 1976;14:21533. [11] Harwood G, Hierons R, Fletcher NA, Marsden CD. Lessons from a remarkable family with dopa-responsive dystonia. J Neurol Neurosurg Psychiat 1994;57:4603. [12] Steinberger D, Weber Y, Korinthenberg R, Deuschl G, Benecke R, Martinius J, et al. High penetrance and pronounced variation in expressivity of GCH1 mutations in ve families with dopa-responsive dystonia. Ann Neurol 1998;43:6349. [13] Furukawa Y, Lang AE, Trugman JM, Bird TD, Hunter A, Sadeh M, et al. Gender-related penetrance and de novo GTP-cyclohydrolase I gene mutations in dopa-responsive dystonia. Neurology 1998;50:101520. [14] Ichinose H, Ohye T, Takahashi E, Seki N, Hori T, Segawa M, et al. Hereditary progressive dystonia with marked diurnal uctuation caused by mutations in the GTP cyclohydrolase I gene. Nat Genet 1994;8:23642. [15] Hyland K, Surtees RA, Rodeck C, Clayton PT. Aromatic L-amino acid decarboxylase deciency: clinical features, diagnosis, and treatment of a new inborn error of neurotransmitter amine synthesis. Neurology 1992;42:19808. [16] Abeling NG, Brautigam C, Homann GF, Barth PG, Wevers RA, Jaeken J, et al. Pathobiochemical implications of hyperdopaminuria in patients with aromatic L-amino acid decarboxylase deciency. J Inherit Metab Dis 2000;23:3258. [17] Abeling NG, van Gennip AH, Barth PG, van Cruchten A, Westra M, Wijburg FA. Aromatic L-amino acid decarboxylase deciency: a new case with a mild clinical presentation and unexpected laboratory ndings. J Inherit Metab Dis 1998;21:2402. [18] Swoboda KJ, Hyland K, Goldstein DS, Kuban KC, Arnold LA, Holmes CS, et al. Clinical and therapeutic observations in aromatic L-amino acid decarboxylase deciency. Neurology 1999;53:120511. [19] Bonafe L, Blau N, Burlina AP, Romstad A, Guttler F, Burlina AB. Treatable neurotransmitter deciency in mild phenylketonuria. Neurology 2001;57:90811. [20] Blau N, Bonafe L, Thony B. Tetrahydrobiopterin deciencies without hyperphenylalaninemia: diagnosis and genetics of dopa-responsive dystonia and sepiapterin reductase deciency. Mol Genet Metab 2001;74:17285. [21] Bonafe L, Thony B, Penzien JM, Czarnecki B, Blau N. Mutations in the sepiapterin reductase gene cause a novel tetrahydrobiopterin-dependent monoamine-neurotransmitter deciency without hyperphenylalaninemia. Am J Hum Genet 2001;69:26977. [22] Bodeau-Pean S, Ravassard P, Neuner-Jehle M, Faucheux B, Mallet J, Dumas S. A human tyrosine hydroxylase isoform associated with progressive supranuclear palsy shows altered enzymatic activity. J Biol Chem 1999;274:346975. [23] De Lonlay P, Nassogne MC, van Gennip AH, van Cruchten AC, Billatte de Villemeur T, Cretz M, et al. Tyrosine hydroxylase deciency unresponsive to L-dopa treatment with unusual clinical and biochemical presentation. J Inherit Metab Dis 2000;23: 81925. [24] de Rijk-Van Andel JF, Gabreels FJ, Geurtz B, Steenbergen-Spanjers GC, van Den Heuvel LP, Smeitink JA, et al. L-dopa-responsive infantile hypokinetic rigid parkinsonism due to tyrosine hydroxylase deciency. Neurology 2000;55:19268.
1160
[25] Brautigam C, Steenbergen-Spanjers GC, Homann GF, Dionisi-Vici C, van den Heuvel LP, Smeitink JA, et al. Biochemical and molecular genetic characteristics of the severe form of tyrosine hydroxylase deciency. Clin Chem 1999;45:20738. [26] Dionisi-Vici C, Homann GF, Leuzzi V, Hoken H, Brautigam C, Rizzo C, et al. Tyrosine hydroxylase deciency with severe clinical course: clinical and biochemical investigations and optimization of therapy. J Pediatr 2000;136:5602. [27] Haussler M, Homann GF, Wevers RA. L-dopa and selegiline for tyrosine hydroxylase deciency. J Pediatr 2001;138:451. [28] Furukawa Y, Graf WD, Wong H, Shimadzu M, Kish SJ. Dopa-responsive dystonia simulating spastic paraplegia due to tyrosine hydroxylase (TH) gene mutations. Neurology 2001;56:2603. [29] Raemaekers VT, Senderek J, Hausler M, et al. A novel neurodevelopmental syndrome responsive to 5-hydroxytryptophan and carbidopa. Mol Genet Metab 2001;73: 17987. [30] Robertson D, Goldberg MR, Onrot J, Hollister AS, Wiley R, Thompson Jr JG, et al. Isolated failure of autonomic noradrenergic neurotransmission: evidence for impaired betahydroxylation of dopamine. New Engl J Med 1986;314:14947. [31] Robertson D, Haile V, Perry SE, Robertson RM, Phillips III JA, Biaggioni I. Dopamine beta-hydroxylase deciency. A genetic disorder of cardiovascular regulation. Hypertension 1991;18:18. [32] Schuback DE, Mulligan EL, Sims KB, Tivol EA, Greenberg BD, Chang SF, et al. Screen for MAOA mutations in target human groups. Am J Med Genet 1999;88:258. [33] Brunner HG, Nelen MR, van Zandvoort P, Abeling NGGM, van Gennip AH, Wolters EC, et al. X-linked borderline mental retardation with prominent behavioral disturbance: phenotype, genetic localization, and evidence for disturbed monoamine metabolism. Am J Hum Genet 1993;52:10329. [34] Brunner HG, Nelen M, Breakeeld XO, Ropers HH, van Oost BA. Abnormal behavior associated with a point mutation in the structural gene for monoamine oxidase A. Science 1993;262:57880. [35] Deckert J, Catalano M, Syagailo YV, Bosi M, Okladnova O, Di Bella D, et al. Excess of high activity monoamine oxidase A gene promoter alleles in female patients with panic disorder. Hum Mol Genet 1999;8:6214. [36] Preisig M, Bellivier F, Fenton BT, et al. An association between bipolar disorder and monoamine A gene polymorphisms: results of a multicenter study. Am J Psychiatry 2000;157: 94855. [37] Lenders JWM, Eisenhofer G, Abeling NGGM, Berger W, Murphy DL, Konings CH, et al. Specic genetic deciencies of the A and B isoenzymes of monoamine oxidase are characterized by distinct neurochemical and clinical phenotypes. J Clin Invest 1996;97:10109. [38] Collins FA, Murphy DL, Reiss AL, Sims KB, Lewis JG, Freund L, et al. Clinical, biochemical, and neuropsychiatric evaluation of a patient with a contiguous gene syndrome due to a microdeletion Xp11.3 including the Norrie disease locus and monoamine oxidase (MAOA and MAOB) genes. Am J Med Genet 1992;42:12734. [39] Gibson KM, Homann GF, Hodson AK, Bottiglieri T, Jakobs C. 4-Hydroxybutyric acid and the clinical phenotype of succinic semialdehyde dehydrogenase deciency, an inborn error of GABA metabolism. Neuropediatrics 1998;29:1422. [40] Hogema BM, Gupta M, Senephansiri H, Burlingame TG, Taylor M, Jakobs C, et al. Pharmacologic rescue of lethal seizures in mice decient in succinate semialdehyde dehydrogenase. Nat Genet 2001;29:2126. [41] Shiang R, Ryan SG, Zhu Y-Z, Hahn AF, OConnell P, Wasmuth JJ. Point mutations in the gene encoding the alpha-1 subunit of the inhibitory glycine receptor cause the dominant neurologic disorder, hyperekplexia. Nat Genet 1993;5:3517. [42] Dubowitz LM, Bouza H, Hird MF, Jaeken J. Low cerebrospinal uid concentration of free gamma-aminobutyric acid in startle disease. Lancet 1992;340:801.
1161
[43] Hamosh A, Johnston MV. Nonketotic hyperglycinemia. In: Scriver CR, Beaudet AL, Sly WS, Valle D, editors. The metabolic and molecular bases of inherited disease. Vol. II. 8th edition. New York: McGraw-Hill; 2001. p. 206578. [44] Applegarth DA, Toone JR. Nonketotic hyperglycinemia (glycine encephalopathy): laboratory diagnosis. Mol Genet Metab 2001;74:13946. [45] Christodoulou J, Kure S, Hayasaka K, Clarke JT. Atypical nonketotic hyperglycinemia conrmed by assay of the glycine in lymphoblasts. J Pediatr 1993;123:1002. [46] Boneh A, Degani Y, Harari M. Prognostic clues and outcome of early treatment of nonketotic hyperglycinemia. Pediatr Neurol 1996;15:13741. [47] Hamosh A, Maher JF, Bellus GA, Rasmussen SA, Johnston MV. Long-term use of highdose benzoate and dextromethorphan for the treatment of nonketotic hyperglycinemia. J Pediatr 1998;132:70913. [48] van der Knaap MS, Barth PG, Gabreels FJM, Franzoni E, Begeer JH, Stroink H, et al. A new leukoencephalopathy with vanishing white matter. Neurology 1997;48:84555. [49] van der Knaap MS, Wevers RA, Kure S, Gabreels FJ, Verhoeven NM, van Raaij-Selten B, et al. Increased cerebrospinal uid glycine: a biochemical marker for a leukoencephalopathy with vanishing white matter. J Child Neurol 1999;14:72831. [50] Leegwater PAJ, Vermeulen G, Konst AAM, Naidu S, Mulders J, Visser A, et al. Subunits of the translation initiation factor eiF2B are mutant in leukoencephalopathy with vanishing white matter. Nat Genet 2001;29:3838. [51] Jackson AH, Applegarth DA, Toone JR, Kure S, Levy HL. Atypical nonketotic hyperglycinemia with normal cerebrospinal uid to plasma glycine ratio. J Child Neurol 1999;14:4647.
Neurol Clin N Am 20 (2002) 11631182
New antiepileptic drug therapies

Ann M. Bergin, MB, MRCP (UK)a,*, Mary Connolly, MB, FRCPCb
Division of Epilepsy and Clinical Neurophysiology, Childrens Hospital, 300 Longwood Avenue, HU2, Boston, MA 02115, USA b Division of Neurology, British Columbia Childrens Hospital, 4480 Oak Street, Vancouver, British Columbia, V6H 3V4 Canada
a
The goal of seizure management is control of seizures with no or minimal adverse eects. This is achieved less frequently than one would hope, despite that many new antiepileptic medications have been licensed in the past decade. In a study of recently diagnosed patients with epilepsy, 47% were seizure-free with the rst antiepileptic drug, 13% after the second drug, and 1% after the third drug [1]. Only 3% were seizure-free with two concomitant antiepileptic drugs and none with three antiepileptic drugs. The most frequently prescribed medications in this study were carbamazepine, valproate, and lamotrigine. Other studies suggest that less than 50% of patients respond to the rst antiepileptic drug [24]. Thus, approximately 60% of patients become seizure-free on monotherapy, usually with the rst or second antiepileptic drug. Thirty to forty percent of patients have seizures that are dicult to control. The role of the newer antiepileptic drugs in this group of patients seems modest to date. In this article we review the new antiepileptic drugs available and discuss what is known about their mechanisms of action, indications, adverse eects, interactions, and other relevant data. Topiramate Topiramate (Topamax), a novel antiepileptic medication derived from a naturally occurring monosaccharide, is approved in more than 65 countries. It has several mechanisms of action, including modulation of voltagedependent sodium conductance, enhancement of gamma aminobutyric acid (GABA) activity at GABA-A receptors, reduction in glutamate release, and
* Corresponding author. E-mail address: ann.bergin@tch.harvard.edu (A.M. Bergin). 0733-8619/02/$ - see front matter 2002, Elsevier Science (USA). All rights reserved. PII: S 0 7 3 3 - 8 6 1 9 ( 0 2 ) 0 0 0 1 7 - 8
1164
A.M. Bergin, M. Connolly / Neurol Clin N Am 20 (2002) 11631182
carbonic anhydrase inhibition [57]. It is well absorbed following oral ingestion with a time to maximum concentration of 24 hours. The rate of absorption may be slowed by food, and it is 15% bound to plasma protein. It has linear pharmacokinetics and is excreted unchanged in the urine with an elimination half-life of 1923 hours. It should be administered twice daily. Concomitant carbamazepine and phenytoin reduce plasma concentrations and half-life. Topiramate has no eect on the plasma concentration of carbamazepine [8]. Ecacy has been demonstrated for the treatment of all seizure types [9 15]. In a recent randomized, multicenter, double-blind study of 252 adults and children with newly or recently diagnosed partial epilepsy, low-dose (2550 mg/day) topiramate was compared with high-dose 200 or 500 mg/day [13]. Time to rst seizure was longer in the high- than in the low-dose group (median 317 versus 108 days). Fifty-four percent in the high-dose group and 39% in the low-dose group became seizure-free. It is also eective in patients with the Lennox-Gastaut syndrome and infantile spasms [1618]. Evidence of topiramate tolerability is based on data from more than 800,000 patients. Topiramate shows a good safety record with no evidence to date of serious or potentially life-threatening adverse eects, or idiosyncratic organ toxicity. The dose should be reduced by 50% in the presence of impaired renal function. Blood levels of phenytoin may be increased in combination with topiramate, and the ecacy of oral contraceptives is reduced. The most frequent adverse eect is decreased appetite and weight loss. Renal stones occur in 1.5% of patients. The risk for renal stones is increased with concomitant acetazolamide or the ketogenic diet. Cognitive slowing and behavioral changes are well described and are the most common reason for discontinuing the medication [19]. There are reports of ecacy of topiramate in bipolar disorder, migraine, neuropathic pain, and essential tremor, but this is not discussed here. Glaucoma has recently been described in association with topiramate [20,21]. Topiramate is available as 25-, 100-, and 200-mg tablets and 15- and 25-mgs sprinkles. In adults, the recommended starting dose is 2550 mg daily, increasing by 25 mg weekly as tolerated. The maximum dose depends on adverse eects and seizure control; doses as high as 1600 mg daily have been used but are not recommended. If there is impaired renal function, the dose should be reduced. In children, the initial recommended dose is 13 mg/ kg/day, increasing weekly by 13 mg/kg/day to a maximum dose of 59 mg/ kg/day. In our experience, an initial dose of 1525 mg/day, increasing by 1525 mg weekly, is better tolerated than a more rapid titration in dose. Vigabatrin Since vigabatrin (Sabril), was initially licensed in 1989, it has been approved for use in 69 countries but not in the United States. Vigabatrin (gamma-vinyl GABA; MDL 71,754) was designed to achieve irreversible
1165
inhibition of the enzyme gamma aminobutyric acid alpha-oxo-glutarate transaminase (GABA-T), the enzyme responsible for the catabolism of the inhibitory amino acid GABA. Vigabatrin increases the concentration of GABA in the brain, thus increasing inhibitory neurotransmission. Vigabatrin diers from GABA structurally by the presence of a vinyl moiety on the amine-bearing carbon. Vigabatrin is rapidly absorbed, and peak plasma concentrations are reached within 2 hours. The plasma half-life is 511 hours and kinetics are linear. It is not protein bound, and excretion is primarily unchanged in the urine. The drug may lower serum levels of phenytoin, phenobarbital, and primidone. Vigabatrin is an eective add-on treatment for patients with refractory partial epilepsy [2226]. It is also indicated for the treatment of infantile spasms, and many neurologists regard it as the treatment of choice, especially in light of the potential for greater toxicity with other medications such as ACTH (Adrenocorticotropic hormone) or sodium valproate. It is an eective treatment for infantile spasms [2730]. In the treatment of infantile spasms it has been demonstrated to be as eective as ACTH, particularly in the setting of tuberous sclerosis [31,32]. In a study of infantile spasms, assessing the ecacy of vigabatrin as initial monotherapy, with ACTH or sodium valproate as a second drug in nonresponders, 11 (26%) of 46 infants responded to vigabatrin, 82% within the rst week [29]. Of the 31 infants who did not respond to vigabatrin, 21 were tried on ACTH and 11 of the 31 responded to ACTH. ACTH was associated with more adverse eects in this study. Six of the vigabatrin nonresponders were treated with sodium valproate and one responded to treatment. Vigabatrin is not eective in the treatment of myoclonic or absence seizures and may exacerbate such seizures [33]. The greatest concern with vigabatrin at present is the risk for persistent visual eld defects [34,35]. Many patients are asymptomatic and for this reason, this adverse defect did not emerge in early studies. In a study of 32 patients on long-term vigabatrin, 13 (40%) were found to have a visual eld constriction [36]. In a prospective study of patients evaluated for epilepsy surgery, 20 of 118 patients (17%) who had received vigabatrin had concentric visual eld defects, compared with 0 of 39 patients (0%) who never received vigabatrin [37]. Of 99 patients treated with vigabatrin for 2.76.7 years, 29 (29%) had visual eld defects [38]. Age, body weight, and daily or cumulative dose of vigabatrin were not associated with an increased risk for visual eld defect. The pattern of vigabatrin-induced visual eld defects is believed to be unique, typically aecting nasal aspects of the visual elds [38]. In infants and young children perimetric visual eld testing is impossible or extremely dicult, and this raises special concerns in using vigabatrin in children. It would seem prudent, in infantile spasms, to try to withdraw vigabatrin after a few months of seizure control and to discontinue treatment when it is ineective. Visual eld defects associated with vigabatrin are considered irreversible, but rare reports suggest that reversal may occur with early withdrawal [3941].
1166
In a review of double-blind placebo-controlled trials of add-on vigabatrin in 717 patients with refractory partial epilepsy, there was a signicantly higher incidence of depression and psychosis in the vigabatrin treated group versus placebo group [42], but no increase in aggressive behavior, agitation, manic behavior, anxiety, emotional lability, or suicide attempt was observed. Vigabatrin, supplied as 500-mg tablets and 500-mg powder sachet, is not available in the United States. The recommended starting dose in adults is 2 g/day administered twice daily. The daily dose may be increased in increments of 0.51 g, depending on response and tolerability, to a maximum of 4 g/day. In children, the recommended starting dose is 40 mg/day, increasing to 80100 mg/day. In infantile spasms, the initial dose is 4050 mg/kg/ day, increasing by 50 mg/kg/day until the spasms are controlled or to 150200 mg/kg/day. Thus, ecacy in infantile spasms may be determined in a few days. Oxcarbazepine Oxcarbazepine (Trileptal), rst introduced in Denmark in 1990, has been registered in more than 50 countries, including the United States, for use as monotherapy and as add-on treatment of partial epilepsy with or without secondary generalization. Oxcarbazepine and its active metabolite, the 10monohydroxy derivative (MHD), limit high frequency repetitive neuronal ring by blocking voltage-dependent sodium channels [43]. MHD reduces the frequency of epileptiform spike discharges induced in hippocampal slices by penicillin, produces a reversible decrease in high voltage activated calcium currents, and reduces transmission of glutamate. Oxcarbazepine and MHD are similar to carbamazepine in their spectrum of antiseizure activity in animal models. Oxcarbazepine is rapidly absorbed from the gastrointestinal tract and rapidly reduced to MHD by cytosolic ketoreductases. MHD is then glucuronidated by UDP-glucuronosyltransferase. The plasma half-life of oxcarbazepine is 12.7 hours and the plasma half-life of MHD is 810 hours [44]. Steady state is reached after 4 doses of oxcarbazepine, administered twice daily. Forty percent of MHD is bound to plasma proteins. There is little competitive interaction with other drugs for protein binding sites. Unlike carbamazepine there is no autoinduction of metabolism. Pharmacokinetics of oxcarbazepine or MHD are not aected by hepatic dysfunction, but impaired renal function may result in delayed clearance of MHD [45]. Neither oxcarbazepine nor MHD inhibit most human cytochrome P450 enzymes in vitro [46]. It is possible, however, that high doses of oxcarbazepine, in conjunction with other medications metabolized by CYP2C19, could aect the metabolism or serum concentrations of other medications such as phenytoin or lamotrigine. Oxcarbazepine induces a subgroup of
1167
cytochrome 450 3A isoenzymes that are responsible for the metabolism of oral contraceptives. In early double-blind studies, the ecacy of oxcarbazepine monotherapy was comparable with phenytoin, carbamazepine, and sodium valproate, in patients with newly diagnosed partial onset or generalized tonic clonic seizures [47]. Recent placebo and active-controlled studies conrm the ecacy of oxcarbazepine monotherapy in patients with partial epilepsy [48,49]. Studies of oxcarbazepine in children and adults with refractory partial epilepsy have demonstrated ecacy [50,51]. Skin reactions occur less frequently than they do with carbamazepine: 68 of 2436 oxcarbazepine treated patients (2.8%), versus 18 of 277 carbamazepine treated patients (6.5%), according to the manufacturers database. Seventy-ve percent of patients who experience skin reactions to carbamazepine do not react to oxcarbazepine. Nonetheless caution is recommended in using oxcarbazepine in patients who had severe reactions to carbamazepine. Hyponatraemia occurs more frequently with oxcarbazepine than with carbamazepine but is usually not clinically signicant [52]. No clinically signicant uctuations in white blood cell counts or elevations of liver enzymes occurred in clinical studies. There is limited information on the safety of oxcarbazepine and MHD on the developing fetus. Oxcarbazepine and MHD cross the placenta and the placenta may metabolize oxcarbazepine to MHD. Oxcarbazepine may be eective in trigeminal neuralgia [53,54]. Oxcarbazepine is supplied as 150-mg, 300-mg, and 600-mg tablets and a suspension of 300 mg/5ml. The starting dose is 300600 mg daily, increasing by 300 mg/day at 37 day intervals to a maximum of 2400 mg daily. If there is renal impairment, the initial dose should be lower and the rate of titration slower. In children, the initial dose is 810 mg/kg/day (maximum initial dose 600 mg daily), increasing by 810 mg/kg/day as tolerated at 37day intervals. Gabapentin Gabapentin (Neurontin) is approved in 34 countries for the treatment of partial epilepsy and neuropathic pain. Although structurally similar to GABA, the mechanism of gabapentins action is not understood. Gabapentin is absorbed from the gastrointestinal tract through an active transport system. There is no protein binding. It is eliminated unchanged by the renal system, indicating that the dose should be reduced in patients with impaired renal function. It does not interact with hepatic enzymes, and thus there is minimal potential to interact with other drugs [55]. In a study of patients with newly diagnosed partial epilepsy, gabapentin 300, 900, and 1800 mg daily was compared with carbamazepine 600 mg daily [56]. Gabapentin was eective, and doses of 1800 mg were well tolerated as monotherapy. Retention rates were similar to carbamazepine. Another double blind, multicenter study compared the ecacy and tolerability of gabapentin and lamotrigine in 309 patients with partial or generalized
1168
tonic-clonic seizures. The endpoint in this study was time to an exit event signaled by lack of ecacy, occurrence of status epilepticus, addition of another antiepileptic medication, or withdrawal because of a drug-related adverse event. Gabapentin (12003600 mg daily) was as eective as lamotrigine (100300 mg daily) in seizure control and tolerability. Of those patients who completed the study (69.6% of the gabapentin-treated patients and 66.2% of the lamotrigine-treated patients), 76% were seizure-free for the nal 12 weeks of treatment. Studies of gabapentin add-on treatment for partial epilepsy in children aged 312 years showed that gabapentin was eective and well tolerated in children [57]. In a systematic review, the ecacy of gabapentin as add-on treatment was short term. Doses of gabapentin used in regulatory trials in adults were modest (9001800 mg daily), but doses of 3600 mg daily or even up to 6000 mg daily have been used [58]. Gabapentin is eective in the management of neuropathic pain and social phobia, and it may have a role in movement disorders, migraine prophylaxis, and cocaine dependence [59]. Gabapentins ecacy and tolerability have been observed across 2 million exposures [60]. No life threatening or serious adverse events have been described. The absence of interactions with other medications is advantageous in patients with complex medical problems and those patients who receive other medications. Gabapentin is supplied as 100-, 300-, 400-, 600-, and 800-mg tablets. The initial dose is 300 mg daily, increasing by 300 mg per day or more slowly as tolerated. The maximum dose is not well established, but doses of 3.6 g daily may be used, administered three times daily. For children, doses of up to 50 mg/kg/day have been used safely, starting at 10 mg/kg/day. Levetiracetam Levetiracetam (Keppra) is a pyrrolidone, a group of compounds investigated for their cognition-enhancing eects. Piracetam was the rst member of this group to have clinical application. Levetiracetam, a novel anticonvulsant drug, has an unknown mechanism of action in epilepsy. A selective binding site for levetiracetam in the brain has been postulated [61], and there is evidence of selective inhibition of N-type calcium channels in hippocampus CA1 cells [62]. Levetiracetam has a unique prole of anticonvulsant eect in animal studies, lacking eect in maximal electroshock and pentylenetetrazole models, but eective in suppressing seizures and spike wave activity in rat models of genetic absence epilepsy, and possibly preventing kindling [63,64]. Levetiracetam is well absorbed with oral bioavailability of nearly 100%. Mean time to peak concentration is 1.3 hours, with an elimination half-life of 68 hours in healthy children and adults. Half-life is prolonged to 1011 hours in elderly patients because of reduction in renal function. There is no
1169
protein binding. There is no hepatic metabolism, and the drug does not interact with or inhibit the hepatic CYP450 enzyme system. Most of the drug is excreted unchanged in the urine, with 24% hydrolyzed to an inactive metabolite. Reduction of dosage is recommended in the setting of renal impairment. There are no known signicant drug interactions [65]. Levetiracetam was shown to be eective as add-on therapy in 324 patients with refractory partial epilepsy at doses of 1000 mg/day and 2000 mg/day [66]. A subsequent crossover trial comparing these two dosage levels with placebo indicated a signicantly greater responder rate at the higher levetiracetam dose [67]. In the United States, another add-on study of 294 patients with refractory partial epilepsy revealed a signicant dierence in responder rate (>50% reduction in seizures) between placebo group (7.4%), 1000 mg/day group (37.1%), and 3000 mg/day group (39.6%) [68]. A study of later monotherapy among 86 responders in a conventional add-on study (3000 mg/day) indicated ecacy and tolerability in a subgroup of 36 patients who completed the monotherapy evaluation period. Nine patients in this group were seizure-free throughout the monotherapy evaluation period [69]. There are few reports of experience with levetiracetam treatment in children. In an open-label study, 12 of 24 children aged 612 years treated with levetiracetam (40 mg/kg/day) as add-on therapy for refractory partial complex seizures had a >50% reduction in seizures. Two patients were seizurefree [70]. Another open-label study of 65 children with highly refractory partial or generalized epilepsy revealed ecacy and tolerability in focal onset and primarily generalized convulsive seizures [71]. The adverse eects occurring more commonly in the levetiracetam than in the placebo treated group were: somnolence (14.8% versus 8.4%), asthenia (14.7% versus 9.1%), infection (primarily common cold) (13.4% versus 7.5%), and dizziness (8.8% versus 4.1%). There was no relationship to dosage [72]. Coordination diculties were reported more commonly in levetiracetam-treated patients (3.4% versus 1.6%). Behavioral abnormalities occurred in 13.3% on active drug versus 6.2% of placebo-treated patients. Symptoms included agitation, hostility, anxiety, apathy, and depression. Psychotic symptoms (0.7% for levetiracetam versus 0.2% for placebo) and suicidal behavior (0.5% versus 0%) were also reported. These eects were reported early in therapy and led to treatment withdrawal. Other symptoms requiring treatment withdrawal were somnolence, dizziness, and asthenia. There were statistically but not clinically signicant reductions in red blood cell count, hemoglobin, and hematocrit, and leukocyte count values for levetiracetam compared with placebo. Occurrence of infection was unrelated to white blood cell count. None of these changes required alteration of dosage. Four children (ages 4, 13, 16, and 17 years) treated at one center experienced rapidly reversible psychotic symptoms on doses of 1533 mg/kg/day for 290 days. All had prior cognitive decits, and the three adolescents had
1170
a history of mild behavioral problems [73]. Little is known regarding developmental eects of levetiracetam in humans. In pregnant rats, treatment is associated with minor fetal skeletal abnormalities and growth retardation. Levetiracetam is supplied as 250-, 500-, and 750-mg tablets. Treatment initiation at 500 mg twice daily has been tolerated and this dose has been shown to have anticonvulsant eect. Dosage increments of 5001000 mg/ day may be introduced at 2 week intervals to the recommended maximum of 3000 mg/day. In elderly patients, initiation at 500 mg/day may prevent adverse eects. There are no recommended dosage guidelines for pediatric patients. Clinical trials in children have initiated treatment at 10 mg/kg/day, and increased to 4060 mg/kg/day. Zonisamide Zonisamide (Zonegran), a sulfonamide derivative, has anticonvulsant properties in animal studies similar to phenytoin and valproate. It blocks sustained ring of neurons mediated by voltage-sensitive Na channels and also reduces voltage-dependent T-type Ca current [74]. Although the drug has carbonic anhydrase activity, it is much less potent than acetazolamide and is not believed to exert signicant anticonvulsant eect by way of this mechanism. Bioavailability of zonisamide is estimated to be greater than 50% after oral administration. Time to peak plasma level is approximately 3 hours, with an elimination half-life of 5070 hours. It is approximately 50% protein bound [75]. Most zonisamide is excreted unchanged in urine, with most of the remaining drug excreted as a reductive metabolite of the benzisoxazole ring. CYP450-3A isoenzyme is responsible for metabolism. Enzymeinducing anticonvulsants increase the metabolism of zonisamide, lower plasma levels, and reduce its half-life (30 hours). Lamotrigine may inhibit clearance of zonisamide [76]. Zonisamide has no eect on hepatic microsomal enzymes, and apart from decreasing the metabolism of carbamazepine to its epoxide [77], it does not alter other anticonvulsant drug levels to a clinically signicant degree [78]. Zonisamide has been shown to be eective for control of refractory localization-related epilepsies in several placebo-controlled, blinded studies [79,80]. It has been at least equivalent to carbamazepine in that setting [81]. A recent multicenter, double-blinded, placebo-controlled trial in 203 patients older than 12 years of age with refractory partial-onset seizures revealed ecacy at all doses tested (100, 200, and 400 mg/day) with best response at 400 mg/day [82]. Zonisamide has been reported to be eective in generalized epilepsy but there are no robust clinical trials in this area. In an open-label study of zonisamide monotherapy for pediatric epilepsy, 5 out of 5 patients (100%) with idiopathic generalized epilepsy achieved control, and the seizures of 7 out of 8 patients (88%) with symptomatic generalized epilepsy were controlled [83]. Eleven of 54 infants treated for infantile
1171
spasms refractory to vitamin B6 achieved seizure control, but three subsequently relapsed [84]. Several reports suggest ecacy of zonisamide in progressive myoclonic epilepsies [78,85,86]. Adverse eects occurring signicantly more commonly in patients treated with zonisamide during early studies were dizziness, somnolence, anorexia, ataxia, confusion, and abnormal thinking. In the most recent placebocontrolled add-on study, a lower incidence of adverse eects was attributed to a slower titration rate. Only weight loss was signicantly more common in zonisamide-treated patients, although anorexia and ataxia were also reported more frequently in the zonisamide group. Cognitive eects are believed to be more likely to occur at zonisamide levels >30 mcg/ml [87]; reversible psychotic symptoms [88] and drug-induced behavior abnormalities in children [89] have been reported. Renal calculi developed in 2.6% of 505 patients treated in the United States and Europe, compared with 0.2% of 1008 patients in Japan. Most were small and did not require specic treatment [78]. A case of renal tubular acidosis in association with zonisamide treatment has been reported [90]. Serious skin reactions including Stevens-Johnson syndrome (SJS) and toxic epidermal necrolysis (TEN) have been observed in patients receiving zonisamide, and seven deaths occurred in the rst 11 years of marketing in Japan. Rash resulting in discontinuation of the drug occurred in 1.4% of patients during drug development in the United States and Europe. No cases of SJS or TEN were seen. Hyperpyrexia caused by oligohydrosis in children, particularly during hot weather, has also been reported, perhaps caused by inhibition of cholinergic innervation of sweat glands [91]. Leukopenia and agranulocytosis have been reported rarely. Elevation of hepatic transaminases may also occur. Zonisamide crosses the placenta and is present in breast milk at levels similar to those of plasma. Anencephaly was detected in one exposed pregnancy and atrial septal defect in another, in a review of 26 exposed ospring [92]. Both mothers were taking additional anticonvulsant drugs. Zonisamide is available as a 100-mg powder-containing capsule. Treatment may be initiated in adults with 100 mg/day, with 100-mg increments at 2-week intervals to usual maximum of 400600 mg/day, divided twice daily to minimize uctuations in plasma levels. Levels of 2030 mcg/ml have been associated with ecacy. In children, the recommended initial dose is 24 mg/kg/day divided twice daily, with increments at 2-week intervals to 68 mg/kg/day and a possible maximum of 12 mg/kg/day. Lamotrigine Lamotrigine (Lamictal), a phenyltriazine compound, inhibits voltagesensitive Na+ currents by interaction with the slow inactivated Na channel, a mechanism similar to that of phenytoin and carbamazepine [93]. Unlike these two drugs, however, lamotrigine is active against partial onset and
1172
generalized seizures. Other as yet unidentied molecular eects may be responsible for lamotrigines broader anticonvulsant spectrum. Oral bioavailability of lamotrigine is virtually 100%. Time to peak serum level is 13 hours [94]. Lamotrigine is 55% protein bound, a fraction that remains stable in the presence of other protein-bound anticonvulsants (eg, phenytoin, valproate). Lamotrigine does not alter the protein binding of other drugs. It is metabolized by glucuronosyl-transferases, and only 10% of the drug is excreted unchanged. None of the metabolites, all excreted in urine, have anticonvulsant properties. The glucuronidation step is rate limiting in the elimination of lamotrigine. The drugs half-life is approximately 30 hours in monotherapy, 14 hours in patients taking enzyme-inducing drugs, and up to 80 hours in patients taking valproate. Lamotrigine increases clearance of valproate by approximately 25%, but has no signicant eect on carbamazepine. It has little eect on oral contraceptive levels [95]. Lamotrigine has been shown to be eective as add-on therapy for refractory partial onset seizures in adults [9698] and children [99], and in refractory generalized epilepsy in adults [100] and children [101], including Lennox-Gastaut syndrome [102]. Monotherapy trials in newly diagnosed epilepsy have included a comparison with carbamazepine in which lamotrigine was shown to be as eective but better tolerated than carbamazepine [3], and a comparison with phenytoin in which there was equal ecacy between the drugs. Lamotrigine was better tolerated overall, with fewer central nervous system side eects, but had a higher incidence of rash [103]. A study comparing lamotrigine with valproate in patients previously treated with either phenytoin or carbamazepine monotherapy for partial seizures and weaned to monotherapy with one of the study drugs revealed signicantly better maintenance of monotherapy on lamotrigine [104]. A placebo-controlled trial of monotherapy in typical absence seizures in children revealed signicant benet over placebo [105]. Franz et al reported seizure freedom in 24 out of 57 patients (42%) with tuberous sclerosis and epilepsy, and a further 21 (37%) who had >50% reduction in seizures without becoming seizure-free [106]. Ecacy in a heterogeneous group with refractory infantile spasms has been reportedasymmetrical spasms and postnatal brain injury were associated with response [107]. There are several reports suggesting a benecial pharmacodynamic interaction between lamotrigine and depakote [108110]. A synergistic eect of lamotrigine and topiramate has also been reported [111]. Common adverse eects of lamotrigine in monotherapy include headache (22%), asthenia (16%), nausea (10%), dizziness (8%), and somnolence (8%). Lamotrigine compares favorably with many antiepileptic drugs with regard to sedative side eects. Several studies have reported fewer cognitive eects in lamotrigine therapy when compared with other drugs [112], and a mental activating eect has been postulated by Meadoretal in 1997. The occurrence of potentially life-threatening skin rash, however, is the greatest concern during lamotrigine therapy. The overall incidence of rash
1173
on lamotrigine therapy is 10%12%. Severe, potentially life-threatening rash such as Stevens-Johnson syndrome (SJS) or toxic epidermal necrolysis (TEN) occurs in an estimated 1:1000 adults, and 1:100200 children [113]. Although most cases occur within 8 weeks of treatment initiation, there have been occasional reports of SJS or TEN later in therapy. The incidence of serious rash, but not overall rash incidence, has declined following reduction in rate of initial dose escalation [114]. Rash may be slightly more common when lamotrigine is added to valproate therapy. It is recommended that treatment be discontinued with the onset of rash. In children, when a rash characteristic of a viral syndrome occurs in a clinical context suggestive of that syndrome, it may be possible to continue lamotrigine therapy. Other rare but potentially fatal adverse eects include hepatic failure [115], agranulocytosis, and aplastic anemia. Causal relationship with lamotrigine is unclear. Lamotrigine has also been reported to induce tic disorder [116] and to exacerbate some cases of juvenile myoclonic epilepsy [96]. Lamotrigine is available as 25-, 100-, 150-, and 200- tablets, and 5-mg and 25-mg chewable/dispersible tablets. Dosing regimen varies depending on whether the depakote is being used concomitantly. In adults not receiving valproate, treatment can be initiated with 50 mg/day for 2 weeks, followed by 50 mg twice daily for 2 weeks, thereafter increasing by 100 mg every 12 weeks to 300500 mg/day. In those patients receiving valproate, initial dose is 25 mg on alternate days for 2 weeks, then 25 mg/day for 2 weeks, thereafter increasing by 2550 mg every 12 weeks to 100400 mg/day. Children starting on Lamictal monotherapy may take 0.3 mg/kg/day for 2 weeks, then 0.6 mg/kg/day for 2 weeks, gradually increasing to 28 mg/kg/ day later. Children already on valproate should start on 0.15 mg/kg/day for 2 weeks, then 0.3 mg/kg/day for 2 weeks with further gradual increments to 15 mg/kg/day. Children already taking enzyme-inducing antiepileptic drugs may start at 0.6 mg/kg/day, increasing after 2 weeks to 1.2 mg/kg/day, with further increments at 12 weekly intervals to 515 mg/kg/day. When sodium valproate is added to a regimen already including lamotrigine, lamotrigine dosage should be reduced by 25%50%. Tiagabine Tiagabine (Gabitril), a molecule designed specically to block reuptake of GABA, inhibits the GAT 1-25 transportation system, thus increasing intrasynaptic GABA concentrations. Tiagabine has been eective in a variety of animal models for convulsive epilepsy, but in some models of absence epilepsy, the proportion of time that the EEG shows spike and wave discharges is increased [117,118]. Bioavailability of tiagabine approaches 100%. Absorption is slowed by administration with food. Time to peak activity is 3090 minutes. It is 96% protein bound. Metabolism is extensive, and only 2% of the drug is
1174
excreted unchanged in urine. Hepatic metabolism by the microsomal enzyme CYP450-3A is inducible. Elimination half-life is 79 hours in noninduced patients, and 24 hours in those patients taking inducing agents [119]. Tiagabine neither induces nor inhibits hepatic microsomal enzymes. The metabolism of tiagabine is slightly inhibited by valproate [94]. Tiagabine has been shown to be eective as add-on therapy for refractory partial seizures in adults [120]. This study and subsequent dose-frequency studies indicate that 32 mg/day is the minimum eective dose, and that either twice daily or four times a day dosing was eective [121]. Given that many of the study patients were taking enzyme-inducing antiepileptic agents, this suggests the therapeutic half-life exceeds the elimination half-life. When compared with phenytoin as add-on therapy to prior carbamazepine treatment, tiagabine ecacy was comparable to that of phenytoin, and tiagabine was better tolerated [122]. There have been some studies of monotherapy with tiagabine, usually in the setting of refractory partial epilepsy, that have been promising. Although pilot studies of tiagabine in pediatric patients have been mostly promising, no denitive study has been completed [123]. Adverse eects of tiagabine, usually mild or moderate and transient, most often involve the central nervous system. Dizziness, tremor, and diculty concentrating were among the common side eects in controlled studies. Adverse eects are more common in polytherapy. In long-term, open label studies the most commonly reported unwanted eects were dizziness, somnolence, accidental injury, asthenia, and headache. Fifteen percent of patients in long-term studies discontinued tiagabine because of adverse eects [123]. There have been several reports of nonconvulsive status in patients treated with tiagabine. Review of these reports suggests that tiagabine does not cause nonconvulsive status, but may be associated with confusional states [124]. There also may be worsening of spike-wave discharges in individual patients at certain doses that resolves with dose adjustment and may not prohibit continued use of tiagabine [123]. Unlike vigabatrin, tiagabine does not seem to cause peripheral visual eld loss [125]. Tiagabine is available as 4-, 12-, 16-, and 20-mg tablets. In adult patients, treatment should be initiated at 4 mg/day. The dose may be increased at weekly intervals by increments of 48 mg, to a maximum of 56 mg/day. The daily dose may be divided twice daily to four times daily. There are no dosing guidelines as yet for children younger than 12 years. Studies in younger children have used an initiating dose varying from 0.250.5 mg/kg/day, with increments of 0.250.5 mg at 24 week intervals, to maximum doses ranging from 12 mg/kg/day [126]. Felbamate Felbamate (Felbatol), an eective broad-spectrum anticonvulsant drug, was rst introduced in the United States in 1993 [127]. Initially believed to be free of serious toxicity, the drug was virtually withdrawn from the
1175
market 1 year later, because of severe bone marrow and liver toxicity in several cases, leading to fatality in some. The American Academy of Neurology issued a practice advisory providing recommendations for its use based on quality of evidence for benet, and understanding of attendant risks [128]. Situations considered to support use of felbamate are: (1) patients with Lennox-Gastaut syndrome older than age 4 years and unresponsive to primary antiepileptic drugs, (2) intractable partial seizures in patients older than age 18 years and unresponsive to standard antiepileptic agents (monotherapy with felbamate providing better risk/benet ratio than polytherapy), (3) patients already on felbamate for 18 months at time of withdrawal of the drug. It is believed that the idiosyncratic responses to felbamate may be because of a specic metabolite that certain individuals are unable to clear eectively [126,129]. A urine screening test has been developed but not yet validated that may allow determination of individuals at lower risk for such responses [130]. Milder adverse eects include anorexia, weight loss, insomnia, and somnolence. Oral bioavailability of felbamate is approximately 90%. Time to peak serum concentration is 14 hours. It is 20% 25% protein bound and has an elimination half-life of 20 hours. Felbamate is excreted unchanged in urine (50%), but also undergoes hepatic metabolism by a variety of enzymes, including the cytochrome P450 enzymes. Felbamate is a potent inhibitor of hepatic microsomal enzymes [94]. This complicates addition of felbamate to existing regimens. Doses of concurrently administered drugs subject to hepatic metabolism, particularly valproate, should be reduced by 20%30% when felbamate is initiated. A further reduction of up to one third the original dose is recommended as felbamate doses are escalated. Further adjustments may be needed [126]. Unlike phenytoin and valproate, levels of carbamazepine decrease slightly, but the proportion of the drug converted to the epoxide derivative increases. Felbamate is available as 400- and 600-mg tablets, and 600-mg/5ml suspension. Initial dose in adults in 400 mg three times daily, increasing at 2-week intervals by an increment of 600 mg/day, to a maximum of 3600 mg/day. In children, initial doses of 15 mg/kg/day, increasing at 2-week intervals to 45 mg/kg/day have been used. Higher doses may be tolerated in children. Fosphenytoin Fosphenytoin (Cerebyx), which has no anticonvulsant activity of its own, is a prodrug of phenytoin. It was developed to avoid local complications of parenteral administration of phenytoin. It is rapidly and completely converted to phenytoin. It has been studied in adults and children of all ages. It was well tolerated intravenously and intramuscularly. After intravenous fosphenytoin administration there was a 9% incidence of pain or burning at the infusion site, compared with 90% after phenytoin administration.
1176
Systemic adverse eects, such as cardiovascular complications (arrhythmia, hypotension) are infrequent. Fosphenytoin dosage is expressed as phenytoin sodium equivalents (PE). Dose is 1520 mg PE/kg for loading dose, usually administered at 0.52.0 mg PE/kg/min, with a maximum rate of 3 mg PE/kg/min [131]. Sulthiame Sulthiame (Ospolot), a sulfonamide derivative available in Europe for many years, is a carbonic anhydrase inhibitor that also blocks sodium channels [132]. It is eective in benign and atypical Rolandic epilepsy [133,134]. In a recent study of 66 children with benign partial epilepsy with central temporal spikes (BECTS), sulthiame 5 mg/kg/day was compared with placebo [133]. Eighty-one percent of the sulthiame treated group and 29% of the placebo treated group completed the study without any treatment failure events. Sulthiame was well tolerated and no patient stopped sulthiame because of adverse eects. Alterations of acidbase equilibrium, hyperventilation, and paresthesiae have been described with sulthiame. This drug is not available in the United States. Stiripentol Stiripentol, a new antiepileptic drug, has been studied in a limited way to date. It inhibits synaptosomal uptake of GABA in rodents and indirectly inhibits several P450 enzymes in humans. The drugs precise mechanism of action is poorly understood. If used in combination with other drugs metabolized by the cytochrome P450 system, it causes elevated levels of these medications, including carbamazepine and valproate. Stiripentol is eective in severe myoclonic epilepsy of infancy when used in combination with valproate and clobazam [135,136]. In a double-blind study comparing stiripentol with placebo in 41 children with severe myoclonic epilepsy of infancy, 71% of the stiripentol treated group had a greater than 50% reduction in frequency of generalized tonic-clonic or clonic seizures, and 9 of 21 in this group were seizure-free. Side eects such as drowsiness, loss of appetite, and hematologic abnormalities were common, and were often caused by elevations in the blood levels of coadministered drugs. Stiripentol is available as 250-mg tablets. In adults, the recommended starting dose is 250500 mg daily, increasing as tolerated to 1500 mg/day. In children, the starting dose is 50 mg/kg/day, increasing as tolerated to 100 mg/kg/day. This drug is not available in the United States. Summary The introduction of these new antiepileptic drugs, from felbamate to levetiracetam, raised hope of control of epilepsy with fewer adverse eects and
1177
improved quality of life. Unfortunately, many patients continue to experience refractory epilepsy despite the use of these new agents, and dose-related adverse eects and idiosyncratic reactions continue to be problematic. A recent report describes six new compounds in preclinical development, and ve in clinical trials [131]. As the number of available, eective, but imperfect antiepileptic drugs increases, many challenges remain. These include: choosing the drug appropriate for the epileptic syndrome, assessing accurately the range of a drugs adverse eects in an individual patient, and considering carefully the drugs interactions in combination drug therapy. In considering drug combinations, diering mechanisms of drug action and favorable pharmacodynamic interactions (an area requiring additional studies) are of importance. Clinicians caring for children who have epilepsy anticipate further advances in the pharmacogenetics and molecular pathophysiology of epilepsy, leading to individually tailored, eective, and safe therapy. References
[1] Kwan P, Brodie MJ. Eectiveness of rst antiepileptic drug. Epilepsia 2001;42(10): 125560. [2] Beydoun A. Monotherapy trials of new antiepileptic drugs. Epilepsia 1997;38(Suppl 9):S2131. [3] Brodie MJ, Richens A, Yuen AWC, The UK Lamotrigine/Carbamazepine Monotherapy Trial Group. Double-blind comparison of lamotrigine and carbamazepine in newly diagnosed epilepsy. Lancet 1995;345:4769. [4] Kwan P, Brodie MJ. Early identication of refractory epilepsy. N Engl J Med 2000;342:3149. [5] Brown SD, Wolf HH, Swinyard EA. The novel anticonvulsant topiramate reduces GABA-mediated chloride ux. Epilepsia 1993;34(Suppl 2):1223. [6] Coulter DA, Sombati S, DeLorenzo RJ. Selective eects of topiramate on sustained repetitive ring and spontaneous bursting in cultured hippocampal neurons. Epilepsia 1993;34(Suppl 2):123. [7] Herrero AI, Olmo N, Gonzalez-Escalada JR, et al. Two new actions of topiramate: inhibition of depolarizing GABA (A)-mediated responses and activation of a potassium conductance. Neuropharmacology 2002;42(2):21020. [8] Bourgeois BF. Pharmacokinetics and pharmacodynamics of Topiramate. J Child Neurol 2000;15(Suppl 1):S2730. [9] Tassinari CA, Michelucci R, Chauvel P, et al. Double-blind, placebo-controlled trial of topiramate (600 mg daily) for the treatment of refractory partial epilepsy. Epilepsia 1996;37(8):7638. [10] Ben-Menachem E. Clinical ecacy of topiramate as add-on therapy in refractory partial epilepsy: the European experience. Epilepsia 1997;38(Suppl 1):S2830. [11] Biton V, Montouris GD, Ritter F, et al. A randomized placebo-controlled study of topiramate in primary generalized tonic-clonic seizures. Neurology 1999;53(5):11307. [12] Elterman RD, Glauser TA, Wyllie E, et al. A double blind randomized trial of topiramate as adjunctive therapy for partial-onset seizures in children. Neurology 1999;52(7):133844. [13] Gilliam FG, Reife RA, Wu SC. Topiramate monotherapy: randomized controlled trial in patients with recently diagnosed localization-related epilepsy. Neurology 1999;52(Suppl 2):A248. [14] Mohamed K, Appleton R, Rosenbloom L. Ecacy and tolerability of topiramate in childhood and adolescent epilepsy: a clinical experience. Seizure 2000;9(2):13741.
1178
[15] Ritter F, Glauser TA, Elterman RD, et al. Eectiveness, tolerability and safety of topiramate in children with partial onset seizures. Epilepsia 2000;41(Suppl 1):S825. [16] Sachdeo RC, Glauser TA, Ritter F, et al. A double-blind, randomized trial of topiramate in Lennox-Gastaut syndrome. Neurology 1999;52:18827. [17] Glauser TA, Clark PO, Strawsburg R. A pilot study of topiramate in the treatment of infantile spasms. Epilepsia 1998;39(12):13248. [18] Glauser TA, Clarke PO, McGee K. Long-term response to Topiramate in patients with West syndrome. Epilepsia 2000;41(Suppl 1):S9194. [19] Tatum WO, French JA, Faught E, et al. The PADS Investigators. Postmarketing experience with topiramate and cognition. Epilepsia 2001;42(9):113440. [20] Rhee DJ, Goldberg MJ, Parrish PK. Bilateral angle-closure glaucoma and ciliary body swelling from topiramate. Arch Ophthalmol 2001;119(11):17213. [21] Sankar PS, Pasquale LR, Grosskreutz CL. Uveal eusion and secondary angle-closure glaucoma associated with topiramate use. Arch Ophthalmol 2001;119(8):12101. [22] Nabbout RC, Chiron C, Mumford J, Dumas C, Dulac O. Vigabatrin in partial seizures in children. J Child Neurol 1997;12(3):1727. [23] Brodie MJ, Mumford JP. Double-blind substitution of vigabatrin and valproate in carbamazepine-resistant partial epilepsy. Epilepsy Res 1999;34(2):199205. [24] Chadwick DW, for the Vigabatrin European Monotherapy Study Group. Safety and ecacy of vigabatrin and carbamazepine in newly diagnosed epilepsy: a multicenter randomized double-blind study. Lancet 1999;354:139. [25] Dean C, Moiser M, Penry K. Dose-response of vigabatrin as add-on therapy in patients with uncontrolled complex partial seizures. Epilepsia 1999;40:7482. [26] Gobbi G, Pini A, Bertani G, et al. Prospective study of vigabatrin monotherapy in childhood partial epilepsies. Epilepsy Res 1999;35(1):2937. [27] Villeneuve N, Souet C, Plouin P, et al. Treatment of infantile spasms with vigabatrin as rst-line therapy and in monotherapy. Arch Pediatr 1998;5(7):7318. [28] Appleton RE, Peters AC, Mumford JP, et al. Randomized, placebo-controlled study of vigabatrin as rst line treatment on infantile spasms. Epilepsia 1999;40(11):162733. [29] Granstrom ML, Gaily E, Liukkonen E. Treatment of infantile spasms: results of a population based study with vigabatrin as the rst drug for spasms. Epilepsia 1999;40(7):9507. [30] Elterman RD, Shields WD, Manseld KA, et al. Randomized trial of vigabatrin in patients with infantile spasms. Neurology 2001;57(8):141621. [31] Hancock E, Osborne JP. Vigabatrin in the treatment of infantile spasms in tuberous sclerosis: literature review. J Child Neurol 1999;14(2):714. [32] Curatolo P, Verdecchia M, Bombardieri R. Vigabatrin for tuberous sclerosis complex. Brain Dev 2001;23(7):64953. [33] Marciani MG, Maschio M, Spanedda F, et al. Development of myoclonus in patients with partial epilepsy during treatment with vigabatrin: an electroencephalographic study. Acta Neurol Scand 1995;91(1):15. [34] Krauss GL, Johnson MA, Miller NR. Vigabatrin associated retinal cone system dysfunction. Electroretinogram and ophthalmologic ndings. Neurology 1988;50:6148. [35] Sankar R, Wasterlain CG. Is the devil we know the lesser of two evils? Vigabatrin and visual elds. Neurology 1999;52:15378. [36] Kalviainen R, Nousiainen I, Mantyjarvi M, et al. Vigabatrin, a GABAergic antiepileptic drug, causes concentric visual eld defects. Neurology 1999;53(5):9226. [37] Hardus P, Verduin WM, Postma G, et al. Concentric contraction of the visual eld in patients with temporal lobe epilepsy and its association with the use of vigabatrin therapy. Epilepsia 2000;41(5):5817. [38] Wild JM, Martinez C, Reinshagen G, et al. Characteristics of a unique visual eld defect attributed to vigabatrin. Epilepsia 1999;40(12):178494. [39] Krakow K, Polizzi G, Riordan-Eva P, et al. Vigabatrin: reversibility of visual eld constriction following discontinuation of vigabatrin. Seizure 2000;9(4):28790.
1179
[40] Vanhatalo S, Paakkonen L. Visual eld constriction in children treated with vigabatrin. Neurology 1999;52:17134. [41] Vanhatalo S, Alen R, Riikoned R, et al. Reversed visual eld constriction in children after vigabatrin withdrawaltrue retinal recovery or improved performance only?. Seizure 2001;10(7):50811. [42] Levinson DF, Devinsky O. Psychiatric adverse events during vigabatrin therapy. Neurology 1999;53(7):15031511. [43] Schmutz M, Brugger F, Gentsch C, et al. Oxcarbazepine: preclinical anticonvulsant prole and putative mechanism of action. Epilepsia 1994;35(Suppl 5):S4750. [44] Lloyd P, Flesch G, Dieterle W. Clinical pharmacology and pharmacokinetics of oxcarbazepine. Epilepsia 1994;35(Suppl 3):S1013. [45] Rouan MC, Lecaillon JB, Godbillon J, et al. The eect of renal impairment on the pharmacokinetics of oxcarbazepine and its metabolites. Eur J Clin Pharmacol 1994;47:1617. [46] Baruzzi A, Albani F, Riva R. Oxcarbazepine: pharmacokinetics interactions and their clinical relevance. Epilepsia 1994;35(Suppl 3):S1419. [47] Sachdeo RC, Beydoun A, Schachter S, et al. Oxcarbazepine (Trileptal) as monotherapy in patients with partial seizures. Neurology 2000;57(5):86471. [48] Schachter SC, Vazquez B, Fisher RS, et al. Oxcarbazepine: double-blind, randomized, placebo-control monotherapy trial for partial seizures. Neurology 1999;52(4):7327. [49] Beydoun A, Sachdeo RC, Rosenfeld WE, et al. Oxcarbazepine monotherapy for partial onset seizures: a multicenter, double-blind, clinical trial. Neurology 2000;54(12):224551. [50] Barcs G, Walker EB, Elger CE, et al. Oxcarbazepine placebo-controlled dose ranging trial in refractory partial epilepsy. Epilepsia 2000;41(12):1597607. [51] Glauser TA, Nigro M, Sachdeo RC, et al. Adjunctive therapy with oxcarbazepine in children with partial seizures. Neurology 2000;54(12):223742. [52] Sachdeo R, Wassertein AG, DSouza J. Oxcarbazepine (Trileptal) eect on serum sodium. Epilepsia 1999;40(Suppl 7):103. [53] Grant SM, Faulds D. Oxcarbazepine: a review of its pharmacology and therapeutic potential in epilepsy, trigeminal neuralgia and aective disorders. Drugs 1992;43(6):8738. [54] Remillard G. Oxcarbazepine and intractable trigeminal neuralgia. Epilepsia 1994; 35(Suppl 3):S289. [55] Leiderman DB. Gabapentin as add-on therapy for refractory partial epilepsy: results of ve placebo controlled trials. Epilepsia 1994;35(Suppl 5):S746. [56] Chadwick DW, Anhut H, Greiner MJ, et al. The International Gabapentin Monotherapy Study Group 94577. A double-blind trial of gabapentin monotherapy for newly diagnosed partial seizures. Neurology 1998;51(5):12828. [57] Appleton R, Fichtner K, LaMoreaux L, et al. Gabapentin as add-on therapy in children with refractory partial seizures: a 12-week, multicenter, double-blind, placebo-controlled study. Epilepsia 1999;40(8):114754. [58] Wilson EA, Sills GJ, Forrest G, et al. High dose Gabapentin in refractory partial epilepsy: clinical observations in 50 patients. Epilepsy Res 1998;29(2):1616. [59] Magnus L. Non-epileptic uses of Gabapentin. Epilepsia 1999;40(Suppl 6):S6672. [60] Morris GL. Gabapentin. Epilepsia 1999;40(Suppl 5):S6370. [61] Noyer M, Gillard M, Matagne A, et al. The novel antiepileptic drug levetiracetam (ucb L059) appears to act via a specic binding site in CNS membranes. Eur J Pharmacol 1995;286: 13746. [62] Lukyanetz EA, Shkryl VM, Kostyuk PG. Selective blockade of N-type calcium channels by levetiracetam. Epilepsia 2002;43(1):918. [63] Klitgaard H, Matagne A, Gobert J, et al. Evidence for a unique prole of levetiracetam in rodent models of seizures and epilepsy. Eur J Pharmacol 1998;353:191206. [64] Gower AJ, Hirsch D, Boehrer A, et al. Eects of levetiracetam, a novel antiepileptic drug, on convulsant activity in two genetic rat models of epilepsy. Epilepsy Res 1995;22:20713. [65] Shorvon S. Pyrrolidone derivatives. Lancet 2001;358:188592.
1180
[66] Shorvon S, Lowenthal A, Janz D, et al. Multicenter double-blind, randomized, placebocontrolled trial of levetiracetam as add-on therapy in patients with refractory partial seizures. Epilepsia 2000;41:117986. [67] Boon P, Chauvel P, Pohlmann-Eden B, et al. Dose response of levetiracetam 1000 and 2000 mg/day in partial epilepsy. Epilepsy Res 2002;48:7789. [68] Cereghino JJ, Biton V, Abou-Khalil B, et al. Levetiracetam for partial seizures: results of a double-blind, randomized clinical trial. Neurology 2000;55:23642. [69] Ben-Menachem E, Falter U. Ecacy and tolerability of levetiracetam 3000mg/d in patients with refractory partial seizures: a multicenter, double-blind, responder-selected study evaluating monotherapy. Epilepsia 2000;41:127683. [70] Glauser T, Bebin M, Ritter F, et al. Open-label ecacy and safety of levetiracetam in pediatric patients with partial onset seizures [abstract]. Epilepsia 1999;40:161. [71] Bourgeois BFD, Holder DL, Valencia I, et al. Open label assessment of levetiracetam ecacy and adverse eects in a pediatric population. Epilepsia 2001;42(Suppl 7):S5354. [72] Harden C. Safety prole of levetiracetam. Epilepsia 2001;42(Suppl):S3639. [73] Kosso EH, Bergey GK, Freeman JM, et al. Levetiracetam psychosis in children with epilepsy. Epilepsia 2001;42:11613. [74] Suzuki S, Kawakami K, Nishimura S, et al. Zonisamide blocks T-type calcium channel in cultured neurons of rat cerebral cortex. Epilepsy Res 1992;12:217. [75] Perucca E. The clinical pharmacokinetics of the new antiepileptic drugs. Epilepsia 1999;40(Suppl 9):S713. [76] McJilton J, DeToledo J, DeCerce J, et al. Cotherapy of lamotrigine/zonisamide levels results in signicant elevation of zonisamide levels [abstract]. Epilepsia 1996;37(Suppl 5):S137. [77] Minami T, Ichiro I, Ohtsubo K, et al. Inuence of additional therapy with zonisamide on protein binding and metabolism of carbamazepine. Epilepsia 1994;35:10235. [78] Leppik IE. Zonisamide. Epilepsia 1999;40(Suppl 5):S239. [79] Leppik IE, Willmore LJ, Homan RW, et al. Ecacy and safety of zonisamide: results of a multicenter study. Epilepsy Research 1993;14:16573. [80] Schmidt D, Jacob R, Loiseau P, et al. Zonisamide for add-on treatment of refractory partial epilepsy: a European double blind trial. Epilepsy Res 1993;15:6773. [81] Seino M, Naruto S, Ito T, et al. Other antiepileptic drugs. In: Levy RH, Mattson RH, Meldrum BS, editors. Antiepileptic drugs. 4th edition. New York: Raven Press; 1995. p. 1011. [82] Faught E, Ayala R, Mountouris GG, et al. The Zonisamide 922 Trial Group. Randomized controlled trial of zonisamide for the treatment of refractory partial-onset seizures. Neurology 2001;57:17749. [83] Kumagai N, Seki T, Yamawaki H, et al. Monotherapy for childhood epilepsies with zonisamide. Jpn J Psychiatr Neurol 1991;45:3579. [84] Suzuki Y. Zonisamide in West syndrome. Brain Dev 2001;23:65861. [85] Henry T, Leppik IE, Gumnit RJ, et al. Progressive myoclonus epilepsy treated with zonisamide. Neurology 1998;38:92831. [86] Kyllerman M, Ben-Menachem E. Long-term treatment of progressive myoclonic epilepsy syndromes with zonisamide and N-acetylcysteine [abstract]. Epilepsia 1996;37(Suppl 5):172. [87] Berent S, Sackellares JC, Giordani B, et al. Zonisamide (CI-912) and cognition: results from preliminary study. Epilepsia 1987;28:617. [88] Charles CL, Stoesz L, Tollefson G. Zonisamide-induced mania. Psychosomatics 1990; 31:2147. [89] Kimura S. Zonisamide-induced behavior disorder in two children. Epilepsia 1995;35:4035. [90] Inoue T, Kira R, Kaku Y, et al. Renal tubular acidosis associated with zonisamide therapy. Epilepsia 2000;41:16424. [91] Shimizu T, Yamashita Y, Satoi M, et al. Heat stroke-like episode in a child caused by zonisamide. Brain Dev 1997;19:3668.
1181
[92] Kondo T, Kaneko S, Amano Y, et al. Preliminary report on teratogenic eects of zonisamide in the ospring of treated women with epilepsy. Epilepsia 1996;37:12424. [93] Coulter DA. Antiepileptic drug cellular mechanisms of action: where does lamotrigine t in? J Ch Neurol 1997;12(Suppl 1):S29. [94] Faught E. Pharmacokinetic considerations in prescribing antiepileptic drugs. Epilepsia 2001;42:1923. [95] Garnett WR. Lamotrigine: pharmocokinetics. J Ch Neurol 1997;12(Suppl 1):S105. [96] Binnie CD, Debets RM, Engelsman M, et al. Double-blind crossover trial of lamotrigine (Lamictal) as add-on therapy in intractable epilepsy. Epilepsy Res 1989;4:2229. [97] Matsuo F, Bergen D, Faught E, et al. The US Lamotrigine Protocol 0.5 Clinical Trial Group. Placebo-controlled study of the ecacy and safety of lamotrigine in patients with partial seizures. Neurology 1993;43:228491. [98] Messenheimer J, Ramsay RE, Willmore LJ, et al. Lamotrigine therapy for partial seizures: a multicenter placebo-controlled, double-blind, cross-over trial. Epilepsia 1994;935:11321. [99] Duchowny M, Pellock JM, Draf WD, et al. The Lamictal Pediatric Partial Seizure Study Group. A placebo-controlled trial of lamotrigine add-on therapy for partial seizures in children. Neurology 1999;53:172431. [100] Beran RG, Berkovic SF, Dunagan FM, et al. Double-blind, placebo-controlled, crossover study of lamotrigine in treatment-resistant generalised epilepsy. Epilepsia 1998;39: 132933. [101] Eriksson A-S, Nergardh A, Hoppu K. The ecacy of lamotrigine in children and adolescents with refractory generalized epilepsy: a randomized, double-blind, crossover study. Epilepsia 1998;39:495501. [102] Motte J, Trevathan E, Arvidsson JFV, et al. The Lamictal Lennox-Gastaut Study Group. Lamotrigine for generalized seizures associated with the Lennox-Gastaut syndrome. N Engl J Med 1997;337:180712. [103] Steiner RJ, Dellportas CI, Findley LJ, et al. Lamotrigine monotherapy in newly diagnosed untreated epilepsy: a double-blind comparison with phenytoin. Epilepsia 1999;40:6017. [104] Gilliam F, Vazquez JC, Chang GY, et al. An active-control trial of lamotrigine monotherapy for partial seizures. Neurology 1998;51:101825. [105] Frank LM, Enlow T, Holmes GL, et al. Lamictal (lamotrigine) monotherapy for typical absence seizures in children. Epilepsia 1999;40:9739. [106] Franz DN, Tudor C, Leonard J, et al. Lamotrigine therapy of epilepsy in tuberous sclerosis. Epilepsia 2001;42:93540. [107] Vegiotti P, Cieuta C, Rey E, et al. Lamotrigine in infantile spasms. Lancet 1994;244: 13756. [108] Panayiotopoulos CP, Ferrie CD, Knott C, et al. Interaction of lamotrigine with sodium valproate. Lancet 1993;341:445. [109] Brodie MJ, Yuen AW. Lamotrigine substitution study: evidence for synergism with sodium valproate? 105 Study Group. Epilepsy Res 1997;26:42332. [110] Pisani F, Oteri G, Russo MF, et al. The ecacy of valproate-lamotrigine comedication in refractory complex partial seizures: evidence for a pharmacodynamic interaction. Epilepsia 1999;40:11416. [111] Stephen LJ, Sills GJ, Brodie MJ. Lamotrigine and topiramate may be a useful combination. Lancet 1998;351:9589. [112] Meador KJ, Loring DW, Ray PG, et al. Dierential cognitive and behavioral eects of carbamazepine and lamotrigine. Neurology 2000;56:117782. [113] Pellock JM. Lamotrigine. In: Pellock JM, Dodson WE, Bourgeois BFD, editors. Pediatric epilepsy, diagnosis and therapy. 2nd edition. New York: Demos; 2001. p. 4616. [114] Wong ICK, Mawer GE, Sander JWAS. Factors inuencing the incidence of lamotriginerelated skin rash. Ann Pharmacother 1999;33:103742.
1182
[115] Arnon R, DeVivo D, Defelice AR, et al. Acute hepatic failure in a child treated with lamotrigine. Pediatr Neurol 1998;18:2512. [116] Sotero de Menezes MAS, Rho JM, Murphy P, et al. Lamotrigine-induced tic disorder: report of ve pediatric cases. Epilepsia 2000;41:8627. [117] Coenen AM, Blezer EH, Luijtelaar EL. Eects of the GABA-uptake inhibitor tiagabine on electroencephalogram, spike-wave discharges and behaviour of rats. Epilepsy Res 1995;21:8994. [118] Meldrum BS, Chapman AG. Basic mechanisms of Gabatril (tiagabine) and future potential developments. Epilepsia 1999;40(Suppl 9):S26. [119] Leach JP, Brodie MJ. Tiagabine. Lancet 1998;351:2037. [120] Uthman BM, Rowan J, Ahmann PA, et al. Tiagabine for complex partial seizures. Arch Neurol 1998;55:5662. [121] Sachdeo R, Leroy R, Krauss G, et al. Tiagabine therapy for complex partial seizures: a dose-frequency study. Arch Neurol 1997;54:595601. [122] Vasquez B, Sachdeo RC, Chang G, et al. Tiagabine or phenytoin as rst add-on therapy of complex partial seizures. Neurology 1998;50(Suppl 4):A199. [123] Shinnar S, Sommerville KW. Tiagabine. In: Pellock JM, Dodson WE, Bourgeois BFD, editors. Pediatric epilepsy diagnosis and therapy. 2nd edition. New York: Demos; 2001. p. 48998. [124] Shinnar S, Berg AT, Sommerville KW, et al. Evaluation of spike wave discharges and complex partial status epilepticus in association with tiagabine therapy. Neurology 1997;48:A39. [125] Kalviainen R, Nousiainen I, Mantyjarvi M. Absence of concentric visual eld defects in patients with initial tiagabine monotherapy. Epilepsia 1999;40(Suppl 2):S259. [126] Pellock JM. Tiagobin experience in children. Epilepsia 2001;42(Suppl 3):S4951. [127] Bourgeois BFD. Felbamate. In: Pellock JM, Dodson WE, Bourgeois BFD, editors. Pediatric epilepsy diagnosis and therapy. 2nd edition. New York: Demos; 2001. p. 46773. [128] French J, Smith M, Faught E, et al. Practice advisory: the use of felbamate in the treatment of patients with intractable epilepsy. Report of the quality standards subcommittee of the American Academy of Neurology and the American Epilepsy Society. Neurology 1999;52:15405. [129] Pellock JM. Epilepsia. 1999;40(Suppl 5):S5762. [130] Leppik IE. Issues in the treatment of epilepsy. Epilepsia 2001;42(Suppl 4):S16. [131] Bialer M, Johannsen SI, Kupferberg H, et al. Progress report on new antiepileptic drugs: a summary of the Fifth Eilat Conference (EILAT V). Epilepsy Res 2001;43:1158. [132] Madeja A, Wolf C, Speckmann EJ. Reduction of voltage operated sodium currents by the anticonvulsant drug sulthiame. Brain Res 2001;900(1):8894. [133] Rating D, Wolf C, Bast T, The Sulthiame Study Group. Sulthiame as monotherapy in children with benign childhood epilepsy with centrotemporal spikes: a 6-month randomized, double-blind, placebo controlled study. Epilepsia 2000;41(10):12848. [134] Rating D. Treatment in typical and atypical rolandic epilepsy. Epileptic Disord 2000;2(Suppl 1):S6972. [135] Chiron C, Marchand MC, Tran A, et al. Stiripentol in severe myoclonic epilepsy of infancy: a randomised placebo-controlled syndrome dedicated trial. Lancet 2000; 356:163842. [136] Meador KJ, Baker GA. Behavioral and cognitive eects of lamotrigine. J Child Neurol 1997;12(Suppl 1):S447.
Neurol Clin N Am 20 (2002) 11831194
Intractable pediatric epilepsy: Vagal nerve stimulation and the ketogenic diet
Raj D. Sheth, MD*, Carl E. Stafstrom, MD, PhD
Department of Neurology, University of Wisconsin at Madison, Madison, WI, USA
Of the approximately 25% of patients with intractable seizures, more than half are not candidates for resective surgery. In adults, intractable epilepsy is generally dened as the persistence of seizures despite at least two appropriately chosen antiepileptic drugs (AEDs) that have been progressively increased until toxicity has been experienced. As shown recently, intractability can often be identied early in the course of epilepsy [1]. The probability of achieving control of seizures after failure of the second AED is approximately 3% [1]. The denition of medical intractability should take into account the natural history of seizures in childhood and the tendency for seizures to become less frequent and severe with passing age [2]. Nevertheless, the approach to intractability in childhood seems to be similar to that in adults. Despite the availability of newer AEDs, the rates of intractability do not seem to have decreased [3]. The likelihood of controlling seizures after two appropriately chosen AEDs fail is low [4]. In most children, at least three AEDs should be tried before a child is said to have intractable epilepsy [5]. An additional factor that should be considered in children is the cognitive decline children experience when having frequent seizures. This latter factor is particularly apparent when there is reversal of cognitive decline after successful control of seizures with epilepsy surgery [6,7]. Surgery is not an option for more than half of children with intractable epilepsy, however. The consequences of epilepsy beyond the occurrence of seizures is considerable. Patients with uncontrolled seizures experience a poorer quality of life, including poor self-esteem, higher levels of depression, and other limitations [8].
* Corresponding author. Department of Neurology, University of Wisconsin at Madison, H6/574 CSC, 600 Highland Avenue, Madison, WI 537925132, USA. E-mail address: sheth@neurology.wisc.edu (R.D. Sheth). 0733-8619/02/$ - see front matter 2002, Elsevier Science (USA). All rights reserved. PII: S 0 7 3 3 - 8 6 1 9 ( 0 2 ) 0 0 0 2 1 - X
1184
R.D. Sheth, C.E. Stafstrom / Neurol Clin N Am 20 (2002) 11831194
The ketogenic diet and vagal nerve stimulation (VNS) oer adjunctive options to AED therapy for children with intractable epilepsy. Evaluation and patient selection for vagal nerve stimulation and the ketogenic diet Patients should be selected for either therapeutic option only after a comprehensive evaluation. The patients seizures should be carefully characterized, and nonepileptic events should be excluded. The latter is particularly important in children with developmental delay and behavioral problems. Many of these patients have movements and behaviors that are mischaracterized as epileptic seizures [911]. Video-electroencephalographic (EEG) monitoring often results in the diagnosis of events that are either pure psychogenic events in isolation or associated with epileptic seizures or, in the younger child, represent physiologic events that mimic seizures [1215]. A diagnosis of nonepileptic events should be considered in all children who do not seem to be responding to multiple medications or have variable and erratic response to therapy. Abnormal ndings on routine (interictal) EEG may actually confound the diagnosis, because many children with static encephalopathy have epileptiform discharges. Such patients should be considered for video-EEG recording, which can help to distinguish nonepileptic from epileptic seizures [16].
Vagal nerve stimulation In the mid-1980s, Zabara [17] rst proposed VNS as a method of desynchronization of electrocerebral activity, thereby relieving seizures. The rst patient with VNS who became seizure-free was reported in 1988 [18,19]. (The NeuroCybernetic Prosthesis (NCP) Houston, Texas) system was developed commercially and consists of (1) an encased pulse generator/battery, (2) bipolar stimulating leads, (3) a handheld programming wand, and (4) a magnet with which to activate the device externally. The pulse generator is typically implanted below the left clavicle in the subcutaneous tissue of the upper chest and is connected to the left vagus nerve in the neck via the bipolar lead. Placement of the pulse generator varies; in young children, the pulse generator may be implanted in the abdomen (for technical reasons), and it may be implanted in the lower anterior axilla beside the pectoral muscle in women (for cosmetic reasons). VNS is the only nonpharmacologic intervention that has been approved by the US Food and Drug Administration for the treatment of epilepsy in adults and adolescents older than the age of 12 years; since 1997, more than 10,000 patients have received implants. Although there is a well-developed literature on its use in adults and older adolescents, there is also an emerging literature on the role of VNS in childhood [20].
1185
Mechanism of action Partial and generalized seizures are both reduced by chronic intermittent stimulation of the left vagus nerve. The left vagus nerve projects on both nuclei of the tractus solitarius as well as on the brainstem reticular formation and has projection on other medullary nuclei. Postsynaptic projections from the medulla occur in the pons, thalamus, amygdala, and insula. The reticular activating system and the limbic system feature prominently in the activation of the vagus nerve. Despite widespread synaptic projections beyond the medulla, the exact mechanism by which VNS reduces seizures is unknown. VNS reduces penicillin-induced cortical interictal spiking rates by one third in rats. In animals, the stimulation eect may increase or decrease EEG synchronization depending on the rate of stimulation. This mechanism does not seem to be the primary factor underlying ecacy, however, and it has not been demonstrated in human patients. VNS does not seem to exert its eect by changes in the EEG. Changes in cerebrospinal uid (CSF) inhibitory amino acids have been noted with chronic VNS, although these changes were observed in both responders and nonresponders to VNS. An interesting clinical observation is the increase in antiseizure eect that occurs with time. This increased seizure control, which occurs 6 to 12 months after stimulation, is also unexplained. Ecacy, tolerability, and safety After the rst implantation in 1988, Penry and Dean [18] reported that 4 of 11 patients with intractable epilepsy became seizure-free after VNS. Early studies demonstrated that one third of patients experienced a greater than 50% reduction in seizure frequency. Randomized high versus low stimulation showed similar rates of seizure control, with this rate increasing to 52% at 18 months after VNS [2125]. Pediatric studies have demonstrated highly variable ecacy rates of control compared with the adult series discussed previously [26]. Of 309 children between the ages of 3 and 18 years who received implants at various centers across the United States, favorable response rates (>50% seizure reduction) were seen in 19% to 53% of patients followed between 3 and 24 months [20]. Surgical considerations and complications The midcervical portion of left vagus nerve chosen for electrical attachment is relatively free of branches and lies within the carotid sheath. It carries aerents from the cardiac atrioventricular node. The recurrent laryngeal nerve travels with the midcervical portion of the left vagus nerve. Stimulation of the midcervical portion therefore causes hoarseness, which occurs each time the device is stimulated. Typically, implantation takes 2 hours; patients may be observed overnight after implantation and are administered antibiotics for 24 hours.
1186
The overall infection rate is 3%. Many patients were successfully treated with antibiotics, although 1% needed the device removed. Vocal cord paralysis is the most common surgical complication, occurring in approximately 1% of patients in the E05 study. Vocal cord dysfunction is suspected if the patient presents with persistent hoarseness dysphagia. Vocal cord dysfunction can be minimized when care is taken to avoid nerve retraction and damage to the vascular supply to the nerve. Avoidance of high-stimulation intensities may help to reduce the degree of vocal cord dysfunction. Other less frequently occurring complications include Horner syndrome, facial weakness, breakage of the leads, and bradycardia [2325]. Initiation and maintenance parameters After implantation, the device is typically turned on in the operating room to deliver a stimulation of 0.25 mA. Thereafter, 0.25-mA increments are adjusted to seizure response, or intolerable side eects may appear. Concomitant AED therapy is kept stable for 12 weeks while the device is being ramped up before a reduction in the dosage or number of AEDs is considered. Current generator batteries are expected to last 10 years, although this is dependent on stimulation parameters [22]. Other considerations with vagal nerve stimulation The cost of the device and its implantation is a major disadvantage of this form of treatment. One third of patients experience no benet after VNS. Unfortunately, there is no way to identify which patients are most likely to benet from the device. Furthermore, only a few patients experience freedom from seizures. Advantages of VNS include absence of the cognitive side eects that are so common with AED treatment. In some patients, a reduction in the number or dosage of concomitant AEDs is possible. The availability of a relatively straightforward alternative therapy in patients who are not surgical candidates is an added advantage. The ketogenic diet The ketogenic diet (KD) is a high-fat, low-carbohydrate, adequate-protein regimen that has been used for more than 80 years for the treatment of intractable epilepsy. By now, the KD is familiar to most physicians caring for children with epilepsy. It is important for neurologists to understand some basic aspects of the KD and how it is used so as to select patients who are most likely to benet from it. The KD was originated in 1920 by Wilder [27]. It was devised to mimic fasting, which has been known for centuries to improve seizure control in persons with epilepsy. With the advent of multiple anticonvulsant medications in the middle and late twentieth century, the KD was relinquished
1187
to the back closet of most epilepsy clinics. In the 1990s, however, the diet witnessed a resurgence, and it is now a common adjunctive therapy to standard AEDs. Here, we briey review the scientic basis for the KD, describe recent studies of its ecacy, and discuss cautionary measures that need to be taken when embarking on KD therapy. Details about other aspects of the KD are available in recent publications, including its history [2830], possible mechanisms of action [3134], details of administration [21,30], and complications [29,35,36]. Ecacy of the ketogenic diet Numerous reports from the 1930s attest to the eectiveness of the KD in small uncontrolled series of patients [3739]. In most of these older reports, there is a paucity of clinical details, and the seizures are not classied according to our modern schema. Nevertheless, these anecdotal reports document the eectiveness of the diet. Although a meta-analysis of those reports is not possible, roughly one third to one half of treated children had a good or excellent response to the diet, variously dened as at least 50% fewer seizures. In the modern era, the eectiveness of the KD has been conrmed. Interestingly, despite advances in epilepsy diagnosis and other aspects of modern medical care, the clinical response rate to the KD has been similar over many decades [40]. A multicenter study of KD ecacy that involved seven comprehensive epilepsy centers and included 51 children showed that more than 40% of children had at least a 50% decrease in seizure frequency when evaluated after 1 year on the diet [41]. The study found no relation between KD ecacy and patient age at diet initiation, seizure type, or EEG ndings. Although this study was neither randomized nor blind, it showed that the KD could be successfully applied across a wide variety of clinical scenarios and in geographically diverse medical centers. A large prospective study of KD ecacy was performed by the Pediatric Epilepsy Group at John Hopkins Hospital, which has been a leader in KD use. They prospectively followed 150 children (mean age 5.3 years) with intractable epilepsy of dierent types [42]. These patients were extremely refractory, averaging 410 seizures per month and having been treated with an average of 6.2 medications before the KD. Overall, 75 of the 150 children had a persistent decrease in seizure frequency of greater than 50%. Thirty children had a greater than 90% seizure reduction, and an additional 7 children became seizure-free. Therefore, the KD has excellent ecacy across a wide variety of seizure types, ages, and etiologies. Children who remained on the diet longer tended to be those who responded well; those who discontinued the diet at less than 1 year tended to have a poorer response. The same population has now been followed for up to 6 years, and the results remain impressive [43]. Twenty children remain seizure-free; only
1188
1 is still on the KD, and the other 19 have been able to discontinue it and remain seizure-free. An additional 21 children had a greater than 90% reduction in seizure frequency (8 remain on the diet). Those who discontinued the diet had a lack of response, considered it too restrictive, or had complications at times of intercurrent illnesses. In addition to seizure control, many children were able to reduce or discontinue their standard anticonvulsants, with concomitant improvement in alertness and cognitive function and fewer anticonvulsant-related side eects [44]. Despite the theoretic and practical diculties of designing a blind crossover study of KD ecacy, such a study is underway. Freeman [45] has enrolled children with frequent atonic seizures (most have Lennox-Gastaut syndrome). After 24 hours of video-EEG to quantify baseline seizure frequency, the children are randomly assigned either to the KD or to an identical KD with glucose added (to negate the ketosis). In the KD treatment arm, placebo (articial sweetener) is added; this does not counteract the ketosis but makes the diet as sweet as that in the glucose-added arm. After 1 week on the respective diets, the groups are fasted briey and then crossed over to the other treatment. Video-EEG data are also collected before and after the 1-week crossover phase. The data from this study, which is still ongoing, should provide important information regarding the short-term role of the KD (and ketosis) in atonic seizure control. The benecial eects of the KD on seizure control, cognitive function, and neurodevelopment have also been documented in other studies both in the United States and abroad [4650]. It has been shown to be eective across the age spectrum from infants to adults [51,52], although it has been thought to work best in young and school-aged children for theoretic reasons related to eciency of ketone extraction by the brain [53]. Finally, all types of seizures and epilepsy may be amenable to KD treatment [30,54]. Mechanism of action Although it is reasonable to suspect that ketosis, a result of the ingestion of ketogenic foods, plays some role in anticonvulsant protection, it is not known exactly how the KD works. Overall, human and animal studies support the notion that ketosis is necessary but probably not sucient for KD eectiveness. Other mechanisms that have been considered in the past include acidosis, dehydration, elevated lipid levels, and electrolyte derangement [55]. None of these hypotheses has stood the test of time. At present, there is an intense laboratory eort to uncover the mechanism of action of the KD [31,5659]. Hopefully, such information will allow optimization of the administration, formulation, and composition of the KD. The basis of the KD lies in the brains use of fats rather than glucose as the primary cerebral energy source. Under conditions in which glucose is not available, such as fasting or the ketogenic diet, fatty acids are oxidized in the
1189
liver to ketones (b-hydroxybutyrate, acetoacetic acid, and acetone) (Fig. 1). The liver lacks the enzymes to degrade these ketone bodies, so they enter the bloodstream and circulate to tissues (eg, brain, muscle), where they may be broken down for energy. The brain can extract and break down ketone bodies, which are funneled into the tricarboxylic acid cycle and then to the electron transport chain, with resultant energy production. Ordinarily, the brain is an obligate user of glucose for its energy. There is a small arterial-venous gradient for ketones under ordinary dietary circumstances. During fasting or KD feeding, however, this gradient increases, and ketones cross the bloodbrain barrier via a monocarboxylic acid transport system. Therefore, during ketosis, the brain uses ketones as its energy source [60]. Exactly how this metabolic transformation from carbohydrates to ketones for energy leads to an antiepileptic eect is currently unknown. Possibilities include alteration of the brains energy charge (energy reserve) [61], a direct eect of ketones on the excitability and synaptic function of neurons [62], enhancement of gamma-aminobutyric acid (GABA) synthesis or function [59], or some other mechanism [33]. Clinical use of the ketogenic diet The rst question might be which patients are most likely to benet from the KD. To date, children have been placed on the diet when all else has failed. It might be possible to achieve seizure control at an earlier stage in the course of the childs epilepsy, however, before exhausting all standard AEDs. Although exact criteria for trying the KD have not been elucidated, any child with intractable epilepsy should be considered a candidate. There are a few conditions under which KD treatment is essential. DeVivo et al [63] described the syndrome of glucose transporter defect. This genetic disorder is characterized by a lack of the protein that transports glucose from the blood into the central nervous system. Therefore, aected children cannot use glucose properly for cerebral energy. The clinical presentation of this syndrome usually involves developmental delays and seizures. Aected children require the KD to provide sucient energy for cerebral function. There are certain conditions under which the KD could lead to neurologic deterioration, including pyruvate carboxylase deciency, fatty acid oxidation disorders, mitochondrial disorders, and carnitine deciency [36]. In all these disorders, switching to fats as the primary energy source could stress the bodys metabolic regulatory systems and lead to energy failure. Ketogenic diet formulation There are several alternative KD formulations, but the classic diet consists of a 4:1 ratio (by weight) of fat/[protein + carbohydrates]. Obviously, a dietitian familiar with the KD is essential to teach the family meal plans
1190
Fig. 1. Summary of ketogenesis and ketone body utilization. (A) Ketogenesis in the liver. FAs from the circulation enter hepatocytes. They cross into the inner mitochondrial membrane either by diusion (short- and medium-chain FAs) or via carnitine (long-chain FAs). Under conditions of fasting or the high-fat low-carbohydrate ketogenic diet, carbohydrate substrate is lacking; thus, FAs are metabolized. Because oxaloacetate (a TCA cycle intermediate) is being diverted to gluconeogenesis, the TCA is not actively involved in energy generation. Acetyl CoA molecules are therefore not funneled into the TCA but are used for ketone body synthesis (AcAc, BHB). The ketones are exported into the circulation, because the liver lacks the enzymes required to catabolize AcAc and BHB. (B) Ketone body oxidation and utilization in the brain. BHB and AcAc enter neurons via the monocarboxylic acid transport system. In mitochondria, BHB and AcAc are broken down (by enzymes 1 and 2) into acetyl-CoA molecules that can then enter the TCA cycle for energy production. FA, fatty acid; CoA, coenzyme A; TCA, tricarboxylic acid; AcAc, acetoacetate; BHB, b-hydroxybutyrate; ATP, adenosine triphosphate; MCT, monocarboxylic acid transporter; 1, BHB dehydrogenase; 2, succinylCoA transferase (3-ketoacid CoA transferase); 3, acetoacetyl CoA thiolase.
1191
and to calculate appropriate dietary needs for each child. Attention must be paid to total calories, sucient protein for growth, and appropriate vitamins and minerals in addition to maintaining the rigid ratio of dietary components. Detailed guidelines for KD formulation are available [30]. An alternative formulation involving medium chain triglycerides (MCTs), which produces an equivalent degree of ketosis with a less restrictive fat-to-carbohydrate ratio, was popular in the 1980s [64]. Although the MCT formulation allows a greater amount of carbohydrates and is thus more palatable, it often resulted in severe diarrhea and is used only occasionally today. The KD is most commonly initiated during a 3- to 5-day hospitalization. The diet begins with an initial fast, with modest uid restriction to approximately 75% of maintenance. Once urinary ketones reach 3 to 4 as measured by urine dipsticks (correlating to approximately 80160 mmol), the diet is started at a 4:1 ratio, with one third of the total calories as a ketogenic eggnog on day 1, two thirds on day 2, and, nally, the full-calorie diet on day 3. After that, ketogenic meals are started. Hospitalizing children has been considered important because of potential side eects during the period of fasting and initial KD administration, including dehydration, hypoglycemia, and other metabolic problems.
Complications of ketogenic diet use The KD is a form of medical therapy rather than a fad diet. Although relatively safe in experienced medical hands, there are many potential side eects that might occur. Renal stones develop in approximately in 10% of children on the KD [65]. Some children on the KD develop reduced bone mass, and growth must be monitored carefully. Recently reported adverse eects include bruising as a result of altered platelet function [66], pancreatitis [67], and cardiomyopathy [68]. Interestingly, one might expect that long-term atherogenic complications would occur in children treated with this high-fat diet, but this has not been reported. It must be stressed that with close monitoring, the KD is safe in most children [30]. Some serious adverse events have been reported in children on the KD. In one study, such events occurred in 5 of 52 children on the KD and included hypoproteinemia, lipemia, hemolytic anemia, renal tubular acidosis, and elevated liver transaminases [35]. Many of these side eects were associated with concurrent valproic acid use; this anticonvulsant must be used with caution with the KD. For optimal administration of the KD, an interdisciplinary program is suggested. This involves the coordinated care of a child by a team of health care professionals, including a neurologist, dietitian, nurse, and social worker [69]. Before initiating the KD in a child, it must be ensured that the parents are fully invested in this complicated approach.
1192
Summary The KD has been proven an eective alternative epilepsy treatment in children refractory to standard anticonvulsants. Children to be placed on the diet must be carefully selected, monitored, and followed. The diet is to be regarded as a strict medical regimen and requires a comprehensive medical team approach in concert with intensive parental involvement. With better understanding of the scientic principles underlying brain ketosis, we should be able to optimize the KD to achieve even better results. References
[1] Kwan P, Brodie MJ. Early identication of refractory epilepsy. N Engl J Med 2000; 342:3149. [2] Cameld C, Cameld P, Gordon K, Smith B, Dooley J. Outcome of childhood epilepsy: a population-based study with a simple predictive scoring system for those treated with medication. 1993;122:8618. [3] Pellock JM. Managing pediatric epilepsy syndromes with new antiepileptic drugs. Pediatrics 1999;104(Part 1):110616. [4] Cameld PR, Cameld CS, Gordon K, Dooley JM. If a rst antiepileptic drug fails to control a childs epilepsy, what are the chances of success with the next drug? J Pediatr 1997;131:8214. [5] Cameld PR, Cameld CS. Antiepileptic drug therapy: when is epilepsy truly intractable? Epilepsia 1996;37(Suppl 1):S605. [6] Shields WD, Shewmon DA, Chugani HT, Peacock WJ. Treatment of infantile spasms: medical or surgical? Epilepsia 1992;33(Suppl 4):S2631. [7] Chugani HT, Shewmon DA, Shields WD, Sankar R, Comair Y, Vinters HV, et al. Surgery for intractable infantile spasms: neuroimaging perspectives. Epilepsia 1993;34:76471. [8] Dunn DW, Austin JK, Huster GA. Symptoms of depression in adolescents with epilepsy. J Am Acad Child Adolesc Psychiatry 1999;38:11328. [9] Duchowny MS, Resnick TJ, Deray MJ, Alvarez LA. Video EEG diagnosis of repetitive behavior in early childhood and its relationship to seizures. Pediatric Review 1988;4:1624. [10] Neill JC. Pseudoseizures in impaired children. Neurology 1990;40:1146. [11] Carmant L, Kramer U, Holmes GL, Mikati MA, Riviello JJ, Helmers SL. Dierential diagnosis of staring spells in children: a video-EEG study. Pediatr Neurol 1996;14: 199202. [12] Sheth RD, Bodensteiner JB. Eective utilization of home-video recordings for the evaluation of paroxysmal events in pediatrics. Clin Pediatr (Phila) 1994;33:57882. [13] Wyllie E, Friedman D, Luders H, Morris H, Rothner D, Turnbull J. Outcome of psychogenic seizures in children and adolescents compared with adults. Neurology 1991;41:7424. [14] Olson DM. Success of ambulatory EEG in children. J Clin Neurophysiol 2001;18:15861. [15] Olson DM. Evaluation of children for epilepsy surgery. Pediatr Neurosurg 2001;34:15965. [16] Metrick ME, Ritter FJ, Gates JR, Jacobs MP, Skare SS, Loewenson RB. Nonepileptic events in childhood. Epilepsia 1991;32:3228. [17] Zabara J. Inhibition of experimental seizures in canines by repetitive vagal stimulation. Epilepsia 1992;33:100512. [18] Penry JK, Dean JC. Prevention of intractable partial seizures by intermittent vagal stimulation in humans: preliminary results. Epilepsia 1990;31(Suppl 2):S403. [19] Uthman BM, Wilder BJ, Penry JK, Dean C, Ramsay RE, Reid SA, et al. Treatment of epilepsy by stimulation of the vagus nerve. Neurology 1993;43:133845. [20] Crumrine PK. Vagal nerve stimulation in children. Semin Pediatr Neurol 2000;7:21623.
1193
[21] Wheless JW, Baumgartner J, Ghanbari C. Vagus nerve stimulation and the ketogenic diet. Neurol Clin 2001;19:371407. [22] Helmers SL, Wheless JW, Frost M, Gates J, Levisohn P, Tardo C, et al. Vagus nerve stimulation therapy in pediatric patients with refractory epilepsy: retrospective study. J Child Neurol 2001;16:8438. [23] Amar AP, DeGiorgio CM, Tarver WB, Apuzzo ML. Long-term multicenter experience with vagus nerve stimulation for intractable partial seizures: results of the XE5 trial. Stereotact Funct Neurosurg 1999;73:1048. [24] Amar AP, Heck CN, DeGiorgio CM, Apuzzo ML. Experience with vagus nerve stimulation for intractable epilepsy: some questions and answers. Neurol Med Chir (Tokyo) 1999;39:48995. [25] Amar AP, Heck CN, Levy ML, Smith T, DeGiorgio CM, Oviedo S, et al. An institutional experience with cervical vagus nerve trunk stimulation for medically refractory epilepsy: rationale, technique, and outcome. Neurosurgery 1998;43:126580. [26] Murphy JV, Wheless JW, Schmoll CM. Left vagal nerve stimulation in six patients with hypothalamic hamartomas. Pediatr Neurol 2000;23:1678. [27] Wilder RM. The eects of ketonemia on the course of epilepsy. Mayo Clin Proc 1921;2:3078. [28] Wheless JW. The ketogenic diet: fa(c)t or ction. J Child Neurol 1995;10:41923. [29] Swink TD, Vining EPG, Freeman JM. The ketogenic diet: 1997. Adv Pediatr 1997;44:297329. [30] Freeman JM, Freeman JB, Kelly MT. The ketogenic dieta treatment for epilepsy. 3rd edition. New York: Demos Publications; 2000. [31] Schwartzkroin P. Mechanisms underlying the anti-epileptic ecacy of the ketogenic diet. Epilepsy Res 1999;37:17180. [32] Stafstrom CE. Animal models of the ketogenic diet: what have we learned, what can we learn? Epilepsy Res 1999;37:24159. [33] Stafstrom CE. Eects of fatty acids and ketones on neuronal excitability: implications for epilepsy and its treatment. In: Mostofsky DI, Yehuda S, Salem N, editors. Fatty acidsphysiological and behavioral functions. Totowa, NJ: Humana Press; 2001. p. 27390. [34] Bough KJ, Stafstrom CE. The ketogenic diet: scientic principles underlying its use. In: Rho JM, Sankar R, Cavazos JE, editors. Epilepsy: scientic foundations of clinical management New York: Marcel Dekker; in press. [35] Ballaban-Gil K, Callahan C, ODell C, Pappo M, Moshe S, Shinnar S. Complications of the ketogenic diet. Epilepsia 1998;39:7448. [36] Wheless JW. The ketogenic diet: an eective medical therapy with side eects. J Child Neurol 2001;16:6335. [37] Peterman MG. The ketogenic diet in epilepsy. JAMA 1925;84:197983. [38] Helmholz HF, Keith HM. Eight years experience with the ketogenic diet in the treatment of epilepsy. JAMA 1930;95:7079. [39] Livingston S. Dietary treatment of epilepsy. In: Livingston S, editor. Comprehensive management of epilepsy in infancy, childhood and adolescence. Springeld, IL: Charles C. Thomas; 1972. p. 378405. [40] Kinsman SL, Vining EPG, Quaskey SA, Mellits D, Freeman JM. Ecacy of the ketogenic diet for intractable seizure disorders: review of 58 cases. Epilepsia 1992;33:11326. [41] Vining E, Freeman JM, Ballaban-Gil K, Cameld C, Cameld P, Holmes G, et al. A multicenter study of the ecacy of the ketogenic diet. Arch Neurol 1998;55:14337. [42] Freeman J, Vining E, Pillas D, Pryzik P, Casey J, Kelly M. The ecacy of the ketogenic diet 1998: a prospective evaluation of intervention in 150 children. Pediatrics 1998;102:135863. [43] Hemingway C, Freeman JM, Pillas DJ, Pyzik PL. The ketogenic diet: a 3- to 6-year followup of 150 children enrolled prospectively. Pediatrics 2001;108:898905. [44] Pulsifer M, Gordon J, Vining E, Freeman J. Eects of the ketogenic diet on development and behavior: preliminary report of a prospective study. Dev Med Child Neurol 2001;43:3016. [45] Freeman JM. The ketogenic diet and epilepsy. Nestle Nutrition Workshop Series Clinical and Performance Program 2001;5:30721.
1194
[46] LeFever F, Aronson N. Ketogenic diet for the treatment of refractory epilepsy in children: a systematic review of ecacy. Pediatrics 2000;105. Available at: www.pediatrics.org/cgi/ content/full/105/4/e46 (Electronic publication). [47] Kaytal NG, Koehler AN, McGhee B, Foley CM, Crumrine PK. The ketogenic diet in refractory epilepsy: the experience of Childrens Hospital of Pittsburgh. Clin Pediatr 2000;39:1539. [48] Hassan AM, Keene DL, Whiting SE, Jacob PJ, Champagne JR, Humphreys P. Ketogenic diet in the treatment of refractory epilepsy in childhood. Pediatr Neurol 1999;21:54852. [49] Magrath G, MacDonald A, Whitehouse W. Dietary practices and use of the ketogenic diet in the UK. Seizure 2000;9:12830. [50] Coppola G, Veggiotti P, Cusmai R, Bertoli S, Cardinali S, Dionisi-Vici C, et al. The ketogenic diet in children, adolescents and young adults with refractory epilepsy: an Italian multicentric experience. Epilepsy Res 2002;48:2217. [51] Nordli DR, Kuroda MM, Carroll J, Koenigsberger DY, Hirsch LJ, Bruner HJ, et al. Experience with the ketogenic diet in infants. Pediatrics 2001;108:12933. [52] Sirven J, Whedon B, Caplan D, Liporace J, Glosser D, ODwyer J, et al. The ketogenic diet for intractable epilepsy in adults: preliminary results. Epilepsia 1999;40:17216. [53] Kraus H, Schlenker S, Schwedesky D. Developmental changes of cerebral ketone body utilization in human infants. Hoppe-Seylers Z. Physiol Chem 1974;355:16470. [54] Maydell BV, Wyllie E, Akhtar N, Kotagal P, Powaski K, Cook K, et al. Ecacy of the ketogenic diet in focal versus generalized seizures. Pediatr Neurol 2001;25:20812. [55] Withrow CD. The ketogenic diet: mechanisms of anticonvulsant action. In: Glaser GH, Penry JK, Woodbury WM, editors. Antiepileptic drugs: mechanisms of action. New York: Raven Press; 1980. p. 63542. [56] Bough KJ, Eagles DA. A ketogenic diet increases the resistance to pentylenetetrazoleinduced seizures in the rat. Epilepsia 1999;40:13843. [57] Muller-Schwarze AB, Tandon P, Liu Z, Yang Y, Holmes GL, Stafstrom CE. Ketogenic diet reduces spontaneous seizures and mossy ber sprouting in the kainic acid model. Neuroreport 1999;10:151722. [58] Likhodii SS, Musa K, Mendonca A, Dell C, Burnham WM, Cunnane SC. Dietary fat, ketosis, and seizure resistance in rats on the ketogenic diet. Epilepsia 2000;41:140010. [59] Yudko M, Daikhin Y, Nissim I, Lazarow A, Nissim I. Ketogenic diet, amino acid metabolism, and seizure control. J Neurosci Res 2001;66:93140. [60] Sokolo L. Metabolism of ketone bodies by the brain. Annu Rev Med 1973;24:2719. [61] DeVivo DC, Leckie MP, Ferrendelli JS, McDougal DB. Chronic ketosis and cerebral metabolism. Ann Neurol 1978;3:3317. [62] Thio L, Wong M, Yamada K. Ketone bodies do not directly alter excitatory or inhibitory hippocampal transmission. Neurology 2000;54:32531. [63] DeVivo DC, Triletti RR, Jacobson RI, Ronen GM, Behmand RA, Harik SI. Defective glucose transport across the blood-brain barrier as a cause of persistent hypoglycorrhachia, seizures, and developmental delay. N Engl J Med 1991;325:7039. [64] Huttenlocher PR, Wilbourn AJ, Signore JM. Medium-chain triglycerides as a therapy for intractable childhood epilepsy. Neurology 1971;21:1097103. [65] Furth SL, Casey JC, Pyzik PL, Neu AM, Docimo SG, Vining EPG, et al. Risk factors for urolithiasis in children on the ketogenic diet. Pediatr Nephrol 2000;15:1258. [66] Berry-Kravis E, Booth G, Taylor A, Valentino LA. Bruising and the ketogenic diet: evidence for diet-induced changes in platelet function. Ann Neurol 2001;49:98103. [67] Stewart WA, Gordon K, Cameld P. Acute pancreatitis causing death in a child on the ketogenic diet. J Child Neurol 2001;16:682. [68] Best TH, Franz DN, Gilbert DL, Nelson DP, Epstein MR. Cardiac complications in pediatric patients on the ketogenic diet. Neurology 2000;54:232830. [69] MacCracken KA, Scalisi JC. Development and evaluation of a ketogenic diet program. J Am Diet Assoc 1999;99:15548.
Neurol Clin N Am 20 (2002) 11951215
Epilepsy surgery Presurgical evaluation

Raj D. Sheth, MD
Comprehensive Epilepsy Program, Departments of Neurology and Pediatrics, University of Wisconsin, 600 Highland Avenue, H6/574 CSC, Madison, WI 53792-5132, USA
Approximately 5% of all children experience a seizure with a further 1% 2% having recurrent unprovoked seizures [1]. Major advances in the management and consequences of epilepsy have emerged in the last decade. Improved classication of seizures and epilepsy syndromes and the availability of new antiepileptic drugs (AEDs) account for a large portion of the advances. Despite these improvements, approximately 25% of children treated with appropriately chosen AED polytherapy have intractable epilepsy [1]. In many of these patients the intractability is established as early as following the rst 2 years of initiation of treatment, or if seizures persist after a trial with two AEDs [24]. Although in the long term many of these children may eventually become seizure-free as adults, they will have sustained substantial social and educational burdens. Patients whose epilepsy does not remit also have an increased risk for death [5]. Many of these patients may benet from epilepsy surgery performed early in the course of their epilepsy. Despite the possibility for becoming free from seizures, there is considerable delay in considering surgery for the treatment of seizures [6]. Several reasons seem to underlie this delay in consideration of epilepsy surgery. Older outcome gures for pediatric epilepsy surgery are in the range of a seizure-free rate of 20%30%, a rate similar to that obtained with a new AED trial [7]. These data, however, originate from series collected between 1950 and 1980, before the availability of modern structural and functional neuroimaging and EEG techniques. Extra-temporal epilepsy is much more common than temporal lobe epilepsy in childhood. Approximately 80% of adults considered for epilepsy surgery have temporal lobe epilepsy, often of mesial temporal onset. The natural history of temporal lobe epilepsy is now well dened and the results of surgery are largely predictable [8]. Adults
E-mail address: sheth@neurology.wisc.edu (R.D. Sheth). 0733-8619/02/$ - see front matter 2002, Elsevier Science (USA). All rights reserved. PII: S 0 7 3 3 - 8 6 1 9 ( 0 2 ) 0 0 0 1 3 - 0
1196
R.D. Sheth / Neurol Clin N Am 20 (2002) 11951215
undergoing an anterior temporal lobectomy for complex partial seizures have seizure-free rates of 60%, with a further 26% experiencing only rare seizures. The known natural history of temporal lobe epilepsy and the results of contemporary epilepsy surgery have recently prompted a multicenter randomized trial to examine the role of early surgery when a patient has failed their rst AED! Only 20% of intractable childhood epilepsy is of temporal lobe origin. Accordingly, pediatric neurologists have been reluctant to consider surgery as an option for their intractable patients. Familiarity with febrile seizures and the benign childhood epilepsies have resulted in a prevailing belief that seizures may not be of much consequence in children. Some of the benign partial epilepsy syndromes, however, have been increasingly associated with cognitive consequences [9]. Early observations of improvement in development following successful epilepsy surgery suggest that seizures may have serious consequence for the developing brain [10]. Contemporary clinical surgical series support these initial observations that behavior and cognition improves following successful surgery [11]. Furthermore, accumulating animal data suggests that intense neural activity, such as occurs in a seizure, may be of signicant consequence in the developing brain [1218]. Recently, pediatric surgical outcomes have become available for more than 200 children with intractable epilepsy. Seizure-free rates of 59%67% were across all pediatric age groups [19], with a further 11%20% experiencing only rare seizures. These outcomes match or closely approximate those reported for adults. Surprisingly, such results are seen across all age groups. Even in infancy, with careful patient selection, seizure-free rates of 50% 65% and rare seizures in a further 13%25% were reported in more than 50 infants undergoing epilepsy surgery. Medical intractability epilepsy Seizure frequency is an important component of the denition of intractability. Seizure severity and seizure persistence despite therapeutic compliance and optimal therapy and the behavioral/educational impact of ongoing seizures, however, are important considerations [1]. Patients should be considered for epilepsy surgery only after careful demonstration of medical intractability [6]. Intractable epilepsy can be dened as seizures that are not brought under control despite 2 years of treatment with appropriately chosen AEDs. It assumes that AEDs have been tried in a methodical fashion, and adequate serum AED concentrations have been documented or clinical signs of toxicity have been demonstrated [11,20]. Surgical outcomes The decision to consider surgery requires denition of an outcome measure. Freedom from seizures is the most favorable outcome, although in
1197
patients continuing to have seizures the reduction in severity and frequency may improve overall quality of life [21,22]. The latter factor should be considered on an individual basis. Epilepsy surgery in children who have been carefully chosen can result in either seizure freedom or a marked (>90%) reduction in seizures in approximately two thirds of children with intractable seizures [2,23,24]. These results are comparable to outcomes obtained in adults. Furthermore, the result are far superior to those obtained for pediatric epilepsy surgery performed in the pre-MRI era between 1940 and the late 1970s [7]. Eighty-ve percent of parents surveyed following epilepsy surgery viewed the surgical outcome favorably [20], suggesting that there was a perceived improvement in the quality of life [25]. Favorable factors include the presence of disabling epilepsy, a localized epileptogenic zone, and a low risk for postoperative decits [20]. Careful patient selection can reduce this rate in infants [19,26]. Mortality rates in infants undergoing surgery (particularly when associated with extensive resection), however, are approximately 2% and not surprisingly higher than in other age groups. Presurgical evaluation The primary determinant of successful outcome following epilepsy surgery is accurate identication of the site of seizure onset. The primary aim of the presurgical evaluation is the identication of the epileptogenic zone that is dened as the area necessary and sucient for initiating seizures and whose removal or disconnection is necessary for abolition of seizures. In practical terms this refers to the minimum amount of tissue that needs to be resected to ensure seizure freedom [27]. Seizure semiology The manner in which a seizure evolves (seizure semiology) oers important clues to location of seizure onset. Every attempt should be made to understand the number of seizure types that a patient experiences and the precise semiology of each type [28]. Often multiple partial seizure types are a manifestation of diering degrees of evolution and diering propagation pathways from a single focus. Precise semiology that is conrmed following successful epilepsy surgery oers the best conrmation of seizures that originate from a particular lobe. Such evidence is now available in adults, although it is still lacking in children. Seizure classication in the young child and infant are inherently problematic because alterations in mental status cannot be conrmed [2932]. In the older child, however, seizure semiology often has features resembling localization related seizures in adults [33]. The discussion that follows examines seizure semiology originating within a cerebral lobe as described in adults and the noninfant child.
1198
Frontal lobe onset Frontal lobe seizures are generally of short duration, typically unassociated with postictal confusion, and frequently nocturnal [34]. Dierentiating nocturnal seizures from nocturnal parasomnias can be made on clinical grounds, although video-EEG may be required [35]. The presence of paroxysmal arousals with motor behaviors usually indicates epileptic seizure rather than parasomnias [36]. Nocturnal frontal lobe epilepsy is more common in boys than girls, with a typical onset in childhood [37]. Tonic or postural features are common in frontal lobe seizures and are suggestive of onset in the contralateral hemisphere. Furthermore, complex partial seizures of frontal lobe onset are more likely to be associated with thrashing, pedaling, and kicking the lower extremities [35]. Not surprisingly, once seizures have propagated from the frontal lobe to the temporal lobe it is not possible to determine region of onset. Complex gestural automatisms are frequently present and help dierentiate frontal lobe seizures from those of temporal lobe onset in which simple orobucal or hand automatisms are common. Leg movements are prominent in frontal lobe seizures that do not propagate to other areas, whereas hand posturing is common in seizures that remain restricted to the temporal lobe [35,38]. When present, these features collectively support a nding of frontal lobe onset for seizures. Seizures originating from various regions within the frontal lobes often have unique dierentiating features that help in localization [39,40]. Supplementary motor seizures are associated with preservation of consciousness, tonic posturing of the extremities (often bilaterally), and are often mistaken for psychogenic seizures [40]. The diagnosis can almost always be veried with prolonged video-EEG recording [40]. Temporal lobe onset Temporal lobe seizures are often preceded by simple partial seizures that are reported as an epigastric rising sensation and may have autonomic accompaniments [41,42]. The semiology of temporal lobe seizures [43] is not clearly dened in childhood, although children older than age 6 years seem to demonstrate a semiology similar to adults. In younger children, it may be dicult to separate typical semiology of temporal lobe onset from seizures of frontal lobe onset [43]. Furthermore, automatisms tend to be simpler in younger children, typically limited to lip smacking and fumbling hand gestures [43]. Behavioral arrest, orofacial automatism, and convulsive activity are much more commonly seen in the young child [44]. These dierences may stem from dierences in seizure etiology. Younger children are unlikely to have mesial temporal sclerosis and are much more likely to have tumors or dysplasia underlying temporal lobe epilepsy [4547]. Unlike frontal lobe epilepsy, postictal confusion is prominent particularly in patients with atrophy restricted to the amygdala. Children, like adults with temporal lobe epilepsy, may have a history of prolonged febrile seizures [48].
1199
Occipital lobe onset Occipital lobe seizures are usually characterized by simple visual auras (sparks, ashes, scotoma, or amaurosis) followed by contraversion of eyes and head or forced eyelid closure [49]. Children often report a sensation of ocular oscillation [50,51]. In adult patients who had undergone successful epilepsy surgery for occipital lobe seizures, clinical features alone suggested an occipital onset in more than two thirds of the patients [52]. Visual auras, most commonly elementary hallucinations and ictal blindness, occurred in 73% of patients. Contralateral eye deviation, blinking, a sensation of eye movement, and nystagmoid eye movements are also seen [53,54]. A third of patients exhibited another seizure type, suggesting ictal spread, with 50% having temporal lobe automatism and 38% having focal motor seizures [52]. Similar ndings have been reported by other investigators [5557]; therefore, video-EEG monitoring is frequently required to identify region of seizure onset. Parietal lobe onset Parietal lobe epilepsy is less common than seizures originating from other lobes. Prominent sensory changes that spread in a Jacksonian march are often present [58]. Clinical manifestations of parietal lobe epilepsy conrmed by tumoral surgery in 34 patients [59] suggest that auras are present in 79% of patients. Somatosensory (62%), visual (12%), and aphasia were the most common auras [59]. Localized ictal pain is a rare phenomena that when present is a strong predictor of parietal lobe seizures [60]. Synthesis of seizure semiology Generating a hypothesis about region of seizure origin based on seizure semiology is the initial step in the strategy to localize seizures. It forms the basis for planning the remainder of the presurgical evaluation. There are several limitations with localization based on seizure semiology alone. As pointed out earlier, infants and young children often do not demonstrate localization-specic semiology [31]. Multiple seizure types that are suggestive of multiorigin and localization areas provide important information that may suggest that surgery is not a good option for the patient being evaluated. Propagated seizures may mislead as to site of origin and also require video-EEG. Alternatively, in patients with true multifocal seizures it may help delineate the most disabling target seizure to tackle.
Physical examination The physical examination provides important clues in the evaluation of seizures and is particularly important in the presurgical evaluation [61].
1200
Particular attention should be placed on the identication of skin lesions associated with tuberous sclerosis, neurobromatosis, Sturge-Weber syndrome, and linear sebaceous nevus syndrome [61,62]. Identication of a tuberous sclerosis would help identify a cerebral tuber that underlies the patients particular seizure [63]. The presence of hemiparesis and visual eld decits oers important clues to seizure localization and helps dene the approach for seizures originating from or close to eloquent cerebral cortex. Children with hemiparesis or limb deformity without evidence of an acquired lesion should prompt a search for focal cortical dysplasia [64], an important nding indicating a lesion that could be amenable to epilepsy surgery. Interictal electroencephalogram The area of cortex that generates epileptiform discharges is referred to as the irritative zone. In approximately 50% of patients, scalp interictal EEG determines the approximate site of seizure onset [65], although in many patients epileptiform discharges extend far beyond the region that underlies the epileptogenic zone. At best, the interictal EEG oers important information allowing planning for further noninvasive and invasive studies. The epileptogenic zone can also be approximated by the region of polymorphic focal slow activity with focal attenuation of background activity. Interictal EEG also allows identication of unsuspected generalized epilepsy. In temporal lobe epilepsy, the interictal EEG may also provide clues to the nature of the epileptogenic lesion. Similar to adults [66], anterior and inferiorly distributed temporal interictal sharp waves suggest mesial temporal sclerosis, whereas lateral or poorly localized epileptiform discharges are more often seen in children with low-grade temporal tumors [67]. Video-electroencephalogram monitoring Video-EEG monitoring gives the clinician an opportunity to carefully analyze the clinical seizure, corroborate the historical description of the seizure, and attempt localization based on semiology [6875]. It also allows the exclusion of nonepileptic seizures and the identication of all the seizure types that the patient experiences. The presence of multiple seizure types may indicate that surgery is not a good option or that in a particular clinical context one of the seizure types that is the most bothersome should be targeted. The ictal onset zone is the cerebral parenchyma that underlies the onset of seizures. The ictal onset zone can usually be identied with ictal videoEEG, although it suers from some of the same limitations as the interictal EEG described earlier [76]. Ictal onset in deep structures or regions not represented on the convexity often is seen only as slow activity occurring many seconds after the onset of clinical manifestations. Most patients in the presurgical evaluation require video-EEG study, although video-EEG evaluation may not be required in all patients undergoing epilepsy surgery [77].
1201
Video-electroencephalogram monitoring and antiepileptic drug withdrawal Video-EEG and intracranial EEG monitoring need to be performed in a timely and cost-ecient manner. To facilitate this, AEDs may need to be withdrawn either completely or partially so that seizures can be captured. Abrupt AED withdrawal leads to concern about provoking status epilepticus and on seizure localization and semiology. Furthermore, the order of individual AED withdrawal must also be addressed. Not surprisingly, AED withdrawal in general results in an increased seizure frequency. Seizure semiology and EEG characteristics following AED withdrawal, however, are similar to prewithdrawal characteristics [78]. These ndings have been shown for AED withdrawal in general and specifically for withdrawal of carbamazepine and phenytoin [79,80]. Rapid withdrawal of the barbiturates or benzodiazepines, however, may result in generalized seizures. The precipitation of previously not experienced seizures has been specically shown for withdrawal of clonazepam [79]. Furthermore, new types of partial seizures uncovered by discontinuation of these agents may be evidence of previously quiescent multifocal epilepsy [80]. In summary, currently available literature suggests that nonbenzodiazepine and nonbarbiturate AED withdrawal is not likely to result in false ictal localization or atypical seizure presentation. Furthermore, withdrawal of phenytoin, carbamazepine, or valproic acid may increase the quantity of interictal EEG epileptiform discharges [81,82]. The rate of AED withdrawal that increases seizure frequency varies by individual AED. For carbamazepine or phenytoin withdrawal, absent or subtherapeutic serum concentration rather than the rate of serum concentration decline was most likely to be associated with an increase in seizure frequency [79]. Furthermore, seizure rates do seem to be signicantly related to the rate of carbamazepine withdrawal, with secondarily generalized tonic-clonic seizures occurring more frequently when this agent is withdrawn over 4 days versus 10 days [83,84]. Theodore et al [85] observed that seizure frequency is related to declining serum concentrations, with the frequency of complex partial seizures increasing as phenobarbital levels decrease through the range of 1520 mcg/ml and generalized convulsions occurring as serum levels decline to less than 10 mcg/ml. Initial serum concentration and rate of concentration decline do not seem to be related to increased seizure frequency. Reduction of nonbarbiturate AED dosage should be done one drug at a time over 23 days. Recognizing that the desired increase in seizures may be most closely associated with absolute serum concentration, the extent of dosage reduction therefore depends on factors such as baseline drug concentration and baseline seizure frequency and severity. Eorts to reduce barbiturate or benzodiazepine medications should be done cautiously and grad ually and may begin well before hospital admission. If generalized tonicclonic seizures do occur during withdrawal, reintroduction of small subtherapeutic doses of the withdrawn AED is reasonable.
1202
Magnetoencephalography (MEG) MEG is a new epilepsy evaluation technique in which the electromagnetic waves of the brain are detected and localized. Because epileptic discharges are fundamentally electrical phenomena, their origin establishes the functional location of the epileptogenic zone. Standard electroencephalography cannot always accurately localize the original tissue source of the electrical disturbance because signal reaching the recording electrodes can be distorted by craniotomy defects or overlying tissues. In contrast, the magnetic counterpart of the electromagnetic wave passes through the calvarium unimpeded, allowing a more precise anatomic localization. To measure the tiny magnetic signals of the brain, a large array of specialized superconducting ampliers (SQUIDs) is placed over the head, and spontaneous electromagnetic activity is recorded. By coregistering the MEG data to the patients MRI, a composite image reecting electromagnetic function and associated structure can be seen. This combination of MEG with MRI has been termed magnetic source imaging (MSI). Early studies have shown the ability of MSI to localize epileptogenic zones in a group of medically refractory epilepsy patients. It seems to have particular usefulness in patients with neocortical sources, an area that is traditionally dicult to localize by EEG methods. In patients with multiple or widespread cortical lesions, the MEG localizations may allow direct visualization of the relation of the MEG seizure spike to the MRI lesion. In this same presurgical epilepsy population, MEG aords the opportunity to perform mapping of the auditory, visual, and somatosensory regions noninvasively before any planned resection. Neuroimaging: magnetic resonance imaging, magnetic resonance spectroscopy, functional magnetic resonance imaging, and positron emission tomography Magnetic resonance imaging (MRI) MRI is the imaging modality of choice [86], although the specic techniques need to be specically tailored to the patient [8789]. Cranial computed tomography may be complementary in some cases to detect calcication, which might help explain the presence of low signal intensity in MRI images or help conrm a diagnosis of cysticercosis. MRI angiography or conventional angiography may be indicated if MRI ndings suggest a vascular malformation. Intravenous contrast medium, although not required routinely in the evaluation of chronic seizures, may be useful in cases in which tumors or actively pathologic processes are suspected [90,91]. Improvement in imaging techniques has increased the percentage of cases in which lesions are detected. In the group of patients with complex partial seizures, mesial temporal sclerosis may be found in one third of patients. The use of surface coils and techniques with high spatial and contrast resolution have increased the accuracy of MRI for subtle cortical dysplasia [9294].
1203
The MRI examination of patients being evaluated for epilepsy surgery includes a sequence of T1-weighted images in the sagittal plane and images with T2 weighting. These may be obtained in axial or coronal plane with fast spine echo (FSE). These techniques include uid attenuating inversion recovery (FLAIR), which uses an inversion time that cancels the signal from CSF. This sequence has the advantage that the signal from CSF does not obscure lesions near the CSF. A second advantage is the greater dynamic range in the image for the normal and pathologic cerebral tissues. Disadvantages of the technique are artifacts related to CSF pulsations and artifacts near the border of CSF and brain. Another technique used to defeat the signal intensity from CSF is inversion recovery prepared fast spin echo. Although tailored protocols are critical to the detection of MTS [47,95], the pediatric magnetic resonance examination must also adequately cover the whole brain to assess general brain morphology and myelination status [96]. Evaluation of the cortex can be particularly challenging in the incompletely myelinated infant brain, because gray-white tissue contrast is suboptimal. In the rst 2 years of life, heavy T1 and T2 weighting is required (TR is typically increased to approximately 3000 msec for T2 series). Phakomatoses such as tuberous sclerosis may be clinically unsuspected and rst suggested by specic magnetic resonance ndings. The use of high-resolution volumetric techniques and phased-array MRI surface coils has improved the detection of subtle abnormalities of the cerebral cortex in patients with cortical dysplasia. The presence of a cortical dysplasia may be suspected because of a cortical dimple, an anomalous vessel, or abnormally small or large gyri. With additional images acquired by way of the surface coil or thin slice technique, the indistinctness of the graywhite junction characteristic of a cortical dysplasia can be detected. The proton-density images are often particularly useful in showing a funnel-shaped region of abnormal signal tapering downward from the cortex toward the ventricular margin. Longstanding dysplasias may have areas of gliosis that show hyperintense signal by T2 or FLAIR, making distinction from low-grade tumor dicult at times (by MRI and pathologic analysis). Dysplasia is favored over tumor when there is tissue loss, absence of mass eect, and presence of dysplastic cortical veins overlying the region. In the absence of imaging clues to the localization of a suspected cortical dysplasia, knowledge of the semiology or EEG localization may be useful to help direct surface coil placement for more detailed MR analysis. Diusionweighted imaging (DWI) has a role in the evaluation of some patients with chronic epilepsy [97]. In diusion-weighted proton MRI, regions with limited diusion or cytotoxic edema are characterized by higher signal intensity, whereas vasogenic edema shows low signal [98]. DWI is sensitive to a shift of water into the intracellular compartment that results from the loss of ATP. Although its more conventional use is in patients with stroke, DWI has been shown to reveal reversible changes in association with electrical brain activity in animals and in association with status epilepticus, either
1204
convulsive or nonconvulsive, in humans. Following a seizure, imaging with DWI may demonstrate a region of diminished diusion that appears as a region of increased signal intensity [95,99]. A region of abnormal diusion in DWI in early reports seems to localize to the epileptogenic focus, although it is not conned to the focus. Magnetic resonance spectroscopy Magnetic resonance spectroscopy (MRS) resolves the signal received from the region of interest into separate resonant frequency peaks for specic metabolites [100,101]. Proton MRS can be performed as a single voxel study in which the relative concentrations of choline, creatine, Nacetyl-aspartate (NAA), lactate, and other metabolites can be measured in a region of interest. It can be performed also as a two-dimensional or three-dimensional study in which the relative concentrations of these metabolites can be measure pixel by pixel in a large region of interest. It requires less than 15 minutes in conjunction with an imaging study. The concentration of NAA in the mesial temporal lobe is a sensitive measure of neuronal dropout and sclerosis. The addition of MRS to the imaging study can improve the detection of mesial temporal sclerosis, particularly in cases with questionable or borderline morphologic ndings. The concentration of NAA in the temporal lobe contralateral to the temporal lobe targeted for resection is a good predictor of outcome from tailored temporal lobectomy. Functional magnetic resonance imaging Functional magnetic resonance imaging (fMRI) may be used in selected patients before epilepsy surgery to estimate the risk for a postoperative neurologic decit [102,103]. With fMRI, the proximity of eloquent brain to the proposed resection can be determined noninvasively. fMRI allows for hemispheric dominance for language to be assessed accurately, obviating the invasive Wada test in some patients. The localization of memory functions by fMRI is under study. If fMRI localized memory accurately, it would likely replace Wada testing in patients with epilepsy. The ability of fMRI to show dynamic regional changes in cerebral blood ow and oxygenation can pinpoint anatomic sites of activation correlated to simultaneously acquired ictal or interictal EEG. Positron emission tomography Positron emission tomography (PET) and SPECT methods are usually reserved for the dicult epilepsy patient who is under consideration for seizure surgery [104]. Both techniques use injected radioactive tracers to produce functional brain images. The rationale behind their use is that epileptogenic zones tend to be hypermetabolic during a seizure, and may
1205
become hypometabolic in the interictal period. Most of the commonly used radioligands are distributed according to blood ow and metabolism, with an uptake curve time course on the order of minutes. Recent advances in SPECT and PET instrumentation have improved resolution such that high quality sublobar localization data can now be expected from these techniques. In SPECT studies, we generally use Tc-99labeled ECD or HMPAO; for PET studies, F-18 deoxyglucose is most commonly used. Advanced techniques that probe receptor kinetics or distinguish blood ow from metabolic changes also can be performed with other radio tracers. Ictal SPECT can be accomplished during inpatient hospitalizations for epilepsy monitoring and has a high sensitivity (80%90%) for localizing the epileptogenic zone. Hypometabolic areas shown by interictal PET correlate highly with ultimate clinical lateralization of mesial and neocortical epilepsies. Neuropsychologic evaluation The neuropsychologic evaluation in epilepsy consists of psychometric assessment of mental status, evaluation of behavioral and emotional function, and determination of laterality of language and memory by way of the intracarotid sodium amobarbital procedure (Wada test). Anterior temporal lobectomy for surgical treatment of complex partial epilepsy is the most widely studied neuropsychologic area of epilepsy surgery. The role of the neuropsychologic evaluation in the young child with extratemporal epilepsy is still evolving. The neuropsychologic assessment provides a baseline assessment of mental status against which the eects of epilepsy surgery can be assessed. It can also provide prognostic information regarding cognitive outcome and may help with localizing and lateralizing information [105110]. Rausch [111] found considerable diversity across epilepsy centers regarding the depth, sophistication, and characteristics of neuropsychologic assessments. The Wada test is a nonspecic moniker for a series of procedures that assess language dominance and memory patency in each cerebral hemisphere. The diversity of procedures used has been collectively called the Wada test. Although there have been several calls for standardization of the procedure, only recently has this been undertaken in a meaningful way [112]. The Loring protocol entails presentation of a standard set of eight to-beremembered objects to the patients intact visual eld for approximately 510 seconds each. Following recovery from the neurologic eects of the perfusion (typically approximately 10 minutes post-drug administration), return of the patients mental status to baseline condition, and cessation of unilateral EEG slowing, recognition memory testing is undertaken. The eight target stimuli are randomly intermixed with 16 foils and the subject is to indicate whether the presented item was or was not part of the initial target set. A net correct score for each hemisphere is derived. Left versus right hemisphere memory asymmetry scores have been shown to be associated with the presence or
1206
absence of hippocampal sclerosis in the to-be-resected mesial temporal lobe region and to be predictive of favorable surgical outcome [113]. In younger children, the Wada test is often ineective in helping determine the functional integrity of the ipsilateral medial temporal lobe and hippocampus. The ability to make condent predictions concerning postsurgical cognitive outcome in young children is weaker than in adolescents and adults. Little systematic research has been published on the eects of epilepsy surgery in postoperative memory functioning in children and on the rate of cognitive development in children undergoing surgery at a young age. Freedom from seizures is the primary predictor of future employment and social integration [114,115]. Hence, careful objective assessment of cognitive, behavioral, and functional status of surgical candidates gives a sense of the prognosis for change in these areas of function, helps in the derivation of reasonable expectations for what may be achieved by epilepsy surgery, and facilitates determination of what rehabilitation or assistance the patient may need postoperatively. Intracranial electroencephalograph In my opinion, factors favoring an intracranial electroencephalograph (EEG) pattern being ictal include the following: (1) a sudden change in ongoing background activity, that is (2) rhythmic rather than irregular, (3) focal rather than diuse, (4) of higher frequency rather than lower at onset, (5) sustained for tens of seconds, (6) evolving in frequency, amplitude, and spatial distribution, and (7) associated with behavioral change. As always there are exceptions to the above, such as the loss or attening of background activity often referred to as an electrodecremental event or seizure onset. The question here is whether the apparent absence of activity relates more to the recording characteristics of our electrodes and ampliers rather than to an actual secession of activity. Equally important is recognizing EEG changes that are less likely to represent a seizure onset and thus should not be used to localize the focus. Irregular slowing is thought by many, including myself, to be a change seen in cortex surrounding the focus rather than a marker of ictal activity. Similarly, regional changes in background rhythms without associated high-frequency discharges are not localizing. These include periodic sharp waves that may be recorded after a neocortical seizure onset. Jayakar (this issue) advises looking for abnormal background rhythms and not just ictal events. Sick cortex produces a dierent EEG background than does healthy cortex, and such areas are seizure prone. Unfortunately, the same sick cortex may generate poorly formed seizure activity, as well as background rhythms. The so-called subtle transformation from an abnormal background rhythm to seizure activity can be very dicult to appreciate, so it is prudent to pay particular attention to sick cortex when attempting to identify seizure onset.
1207
Nearly everyone agrees that the earliest EEG change is the most important for localization. A frequent problem is how to interpret this change when it appears to be widespread. Is the seizure unlocalized or simply unlocalizable by the electrodes implanted. Spatial undersampling may be the culprit, and it is a major concern when interpreting intracranial EEG. As has been known for some time, the recording properties of intracranial electrodes tend to make them very nearsighted. Local near-eld cortical activity is of signicantly higher amplitude than that from cortex only a few millimeters and certainly centimeters away. Successfully identifying the epileptogenic focus is made more likely if its general location has been determined by noninvasive procedures and if sucient numbers of electrodes are used to cover this area. ICEEG should not be considered an exploratory procedure. If you have no idea from where seizures may be coming, the limited spatial sampling of this technique is likely to lead to failure or worse, a wrong localization. This is particularly true of neocortical epilepsies which are more diverse in their location and extent than classic mesial temporal epilepsy. Given the expense, time, eort, and potential morbidity of ICEEG, it makes little sense to proceed without a reasonable expectation of obtaining adequate focus localization. Even if the intracranial EEG changes appear to be focal from our electrode array, can we be certain it represents the seizure origin and not ictal activity that has propagated from some other region where we do not have electrodes? There is no consensus on how this can be achieved reliably. We look for behavioral alterations before EEG change to warn us of this possibility, but neocortical seizures can spread with lightning speed, making this observation less likely. Some believe that low-voltage gamma activity or focal DC shifts are only observed at the initial focus. Less optimistic colleagues recount the old adage among intracranial electroencephalographers that seizures only begin where there are electrodes. Certainly further investigations are needed to help identify distinguishing criteria or analysis techniques. Advances in intracranial EEG recording, particularly the increase in electrode numbers, has brought with it problems as well. At many epilepsy centers, 100 or more electrode contacts may be used in a single patient. Appreciating the location of these electrodes relative to the underlying cortex and to each other is often very dicult. Accordingly, the spatial analysis of these intracranial EEG data is often quite confusing when looking simply at traces. The spatiotemporal character of intracranial voltage elds may provide new localizing information; however, to accomplish this, the locations of electrodes and cortex need to be dened and displayed in a two- or preferably three-dimensional representation. All of these data are currently available, but seldom are they combined and used to interpret ICEEG. In many respects ICEEG interpretation is at a stage similar to scalp EEG analysis 10 years ago, when quantitative and computer-assisted techniques were seldom used and localization was based principally on pattern recognition.
1208
Since that time there have been signicant strides in scalp EEG interpretation using a variety of digital techniques including voltage topography, source localization (such as dipole modeling), and spectral and coherence analyses. Many of these same techniques will probably nd useful applications with intracranial EEG. Under few circumstances except surgical evaluations do we get the opportunity to record directly from the human brain. These are our most valuable data, both by providing diagnostic and localizing information for the patient and by allowing us to advance our knowledge of the electrophysiology of the brain. With these data we can directly validate and thus more eciently improve our noninvasive techniques. It is time to look critically once more at our methods. Just because we have been interpreting ICEEG for 25 years does not necessarily mean that there are not better ways to do it. Although pressures for clinical productivity encourage us simply to localize the seizure focus and move on to the next case, it is worthwhile for all concerned to take the extra time to learn how to get more information from intracranial EEG, lest we nd ourselves in the near future constrained both by old ideas and new regulations that limit further opportunity to learn. Intracranial EEG is obtained when surgery is actively considered, not as an exploratory procedure [27]. It is obtained by placing strip or grid electrodes in the subdural space. Subdural electrodes measuring 24 mm in diameter made of platinum alloy are embedded in polyurethane and placed either by way of a craniotomy (for grids) or under uoroscopy through burr holes (for strips). EEG recording (electrocorticography, ECoG) and electrical stimulation (functional mapping) can be performed by way of the electrodes. The risk for infection and hemorrhage is less than 1% [116]. There are two approaches to obtaining intracranial EEG before resective surgery that are described here. One-stage operative strategy A one-stage operation uses ECoG in the operating room immediately before the resection. This strategy uses interictal activity to direct the resection. Generally, ECoG is used to ne tune a resection in which there is already relative clarity of the relationship of the lesion to the epileptogenic zone. Interictal ECoG features used to dene the epileptogenic zone include (1) consistent focality in interictal spiking, (2) rhythmic features, (3) trains of focal fast beta activity, and (4) focal attenuation of background [27,117 119]. ECoG that reveals almost continuous focal rhythmic discharges may not require ictal recordings. The one-stage strategy using ECoG is typically used to tailor resection of a tumor that manifests with seizures. It is also useful when there is congruence between the imaging lesion and the ictal onset zone. A major advantage of a one-stage procedure is the limitation of cost
1209
and risk for infection. Awake intraoperative ECoG to allow mapping of eloquent cortex, however, is particularly dicult in children [120]. Furthermore, ECoG is performed in the operating room while the child is under the inuence of general anesthesia that may suppress or occasionally paradoxically activate spikes. Two-stage operative strategy In a two-stage operation, subdural electrodes are placed during the initial operation and the patient is then moved to a monitoring unit [118,120124]. Seizures are captured and the ictal onset zone dened [125]. Functional mapping is performed if necessary and the relationship of eloquent cortex to the epileptogenic zone is dened. These data then allow a tailored resection at a second operation several days later [126,127]. This strategy is used when the epileptogenic zone is discrete, imaging lesions are not present, or the planned resection is adjacent to eloquent cortex [27]. Disadvantages with a two-stage strategy include the increased cost and risk for infection. Intracerebral depth electrodes seem to be of limited importance in children and are not discussed here. Resection of regions most active on ECoG seem to be important for successful outcomes, although removal of all discharging areas may not be necessary to achieve seizure control [128]. Surgical outcomes can be predicted by intracranial EEG [129133]. Intraoperative cortical stimulation elicited a habitual aura in 37% of 29 patients. Patients in whom stimulation elicits a habitual aura have a good prognosis for becoming seizure-free following resection. Lateralizing clinical features were seen in almost two thirds of patients: contralateral head deviation occurred in half, 59% had visual eld defects contralateral to the epileptogenic area, and 64% had abnormal imaging studies ipsilateral to the side of surgery [52]. Functional cortical mapping plays an important role in tailored neocortical resections [102,134]. Dening the relationship between eloquent cortex and the epileptogenic zone is important in planning the resection [117, 135,136]. Typically, a margin of at least 0.51 cm is required to spare function. Language cortex can be mapped in children older than 4 years [110], or more recently in children as young as age 2.5 years [135,136]. Summary The presurgical evaluation should result in a clear understanding of whether surgery can be undertaken and its associated risks and potential for benet. The results of surgery are best when there is congruence in the seizure semiology, the irritative zone on interictal EEG, and the ictal onset zone with the epileptogenic lesion as dened on MRI and PET, and when there is a clear understanding of the ictal onset zones relationship to eloquent cortex as dened by neuropsychologic evaluation, the intracarotid amobarbital test, and cortical functional mapping.
1210
References
[1] Hauser WA, Hesdorer DC. Epidemiology of intractable epilepsy. In: Luders HO, Y CG, editors. Epilepsy surgery. Philadelphia: Lippincott Williams & Wilkins; 2001. p. 559. [2] Cameld PR, Cameld CS, Gordon K, Dooley JM. If a rst antiepileptic drug fails to control a childs epilepsy, what are the chances of success with the next drug? J Pediatr 1997;131(6):8214. [3] Hauser E, Freilinger M, Seidl R, Groh C. Prognosis of childhood epilepsy in newly referred patients. J Child Neurol 1996;11(3):2014. [4] Berg AT, Levy SR, Novotny EJ, Shinnar S. Predictors of intractable epilepsy in childhood: a case- control study. Epilepsia 1996;37:2430. [5] Sillanpaa M, Jalava M, Kaleva O, Shinnar S. Long-term prognosis of seizures with onset in childhood. N Engl J Med 1998;338(24):171522. [6] Dlugos DJ. The early identication of candidates for epilepsy surgery. Arch Neurol 2001;58(10):15436. [7] Fish DR, Smith SJ, Quesney LF, Andermann F, Rasmussen T. Surgical treatment of children with medically intractable frontal or temporal lobe epilepsy: results and highlights of 40 years experience. Epilepsia 1993;34(2):2447. [8] Engel Jr J. Surgery for seizures. N Engl J Med 1996;334(10):64752. [9] Deonna TW, Roulet E, Fontan D, Marcoz JP. Speech and oromotor decits of epileptic origin in benign partial epilepsy of childhood with rolandic spikes (BPERS). Relationship to the acquired aphasia-epilepsy syndrome. Neuropediatrics 1993;24(2):837. [10] Vasconcellos E, Wyllie E, Sullivan S, Stanford L, Bulacio J, Kotagal P, et al. Mental retardation in pediatric candidates for epilepsy surgery: the role of early seizure onset. Epilepsia 2001;42(2):26874. [11] Gilliam F, Wyllie E, Kashden J, Faught E, Kotagal P, Bebin M, et al. Epilepsy surgery outcome: comprehensive assessment in children. Neurology 1997;48(5):136874. [12] Holmes GL. Do seizures cause brain damage? Epilepsia 1991;32(Suppl 5):S1428. [13] Holmes GL. Pathogenesis of epilepsy: the role of excitatory amino acids. Cleveland Clinic Journal Medicine 1995;62(4):2407. [14] Holmes GL. Epilepsy in the developing brain: lessons from the laboratory and clinic. Epilepsia 1997;38(1):1230. [15] Sutula TP, Pitkanen A. More evidence for seizure-induced neuron loss: is hippocampal sclerosis both cause and eect of epilepsy? Neurology 2001;57(2):16970. [16] Sutula T, Lauersdorf S, Lynch M, Jurgella C, Woodard A. Decits in radial arm maze performance in kindled rats: evidence for long-lasting memory dysfunction induced by repeated brief seizures. J Neurosci 1995;15(12):8295301. [17] Sutula TP. Secondary epileptogenesis, kindling, and intractable epilepsy: a reappraisal from the perspective of neural plasticity. Int Rev Neurobiol 2001;45:35586. [18] Stafstrom CE, Lynch M, Sutula TP. Consequences of epilepsy in the developing brain: implications for surgical management. Semin Pediatr Neurol 2000;7(3):14757. [19] Wyllie E, Comair YG, Kotagal P, Bulacio J, Bingaman W, Ruggieri P. Seizure outcome after epilepsy surgery in children and adolescents. Ann Neurol 1998;44(5):7408. [20] Gilman JT, Duchowny M, Jayakar P, Resnick TJ. Medical intractability in children evaluated for epilepsy surgery. Neurology 1994;44:13413. [21] Wiebe S, Eliasziw M, Matijevic S. Changes in quality of life in epilepsy: how large must they be to be real? Epilepsia 2001;42(1):1138. [22] Wiebe S, Blume WT, Girvin JP, Eliasziw M. A randomized, controlled trial of surgery for temporal-lobe epilepsy. N Engl J Med 2001;345(5):3118. [23] Pellock JM. Managing pediatric epilepsy syndromes with new antiepileptic drugs. Pediatrics 1999;104(5):110616. [24] Cameld PR, Cameld CS. Antiepileptic drug therapy: when is epilepsy truly intractable? Epilepsia 1996;37(Suppl 1):S6065.
1211
[25] Olson DM. Evaluation of children for epilepsy surgery. Pediatr Neurosurg 2001; 34(3):15965. [26] Wyllie E. Surgical treatment of epilepsy in children. Pediatr Neurol 1998;19(3):17988. [27] Jayakar P. Invasive EEG monitoring in children: when, where, and what? J Clin Neurophysiol 1999;16(5):40818. [28] Sheth RD, Bodensteiner JB. Eective utilization of home-video recordings for the evaluation of paroxysmal events in pediatrics. Clin Pediatr (Phila) 1994;33(10): 57882. [29] Blume WT. Clinical prole of partial seizures beginning at less than four years of age. Epilepsia 1989;30:8139. [30] Wyllie E, Comair YG, Kotagal P, Raja S, Ruggieri P. Epilepsy surgery in infants. Epilepsia 1996;37(7):62537. [31] Nordli Jr DR, Bazil CW, Scheuer ML, Pedley TA. Recognition and classication of seizures in infants. Epilepsia 1997;38(5):55360. [32] Duchowny M, Jayakar P, Resnick T, Harvey AS, Alvarez L, Dean P, et al. Epilepsy surgery in the rst three years of life. Epilepsia 1998;39(7):73743. [33] Blume WT. Uncontrolled epilepsy in children. Epilepsy Res 1992;5:1924. [34] Swartz BE, Delgado-Escueta AV, Walsh GO, Rich JR, Dwan PS, DeSalles AA, et al. Surgical outcomes in pure frontal lobe epilepsy and foci that mimic them. Epilepsy Res 1998;29(2):97108. [35] Kramer U, Riviello Jr JJ, Carmant L, Black PM, Madsen J, Holmes GL. Clinical characteristics of complex partial seizures: a temporal versus a frontal lobe onset. Seizure 1997;6(1):5761. [36] Provini F, Plazzi G, Tinuper P, Vandi S, Lugaresi E, Montagna P. Nocturnal frontal lobe epilepsy. A clinical and polygraphic overview of 100 consecutive cases. Brain 1999; 122(6):101731. [37] Oldani A, Zucconi M, Smirne S, Ferini-Strambi L. The neurophysiological evaluation of nocturnal frontal lobe epilepsy. Seizure 1998;7(4):31720. [38] Sheth RD, Gutierrez AR, Riggs JE. Rolandic epilepsy and cortical dysplasia: MRI correlation of epileptiform discharges. Pediatr Neurol 1997;17(2):1779. [39] So NK. Mesial frontal epilepsy. Epilepsia 1998;39(Suppl 4):S4961. [40] Baumgartner C, Flint R, Tuxhorn I, Van Ness PC, Kosalko J, Olbrich A, et al. Supplementary motor area seizures: propagation pathways as studied with invasive recordings. Neurology 1996;46(2):50814. [41] OBrien TJ, Kilpatrick C, Murrie V, Vogrin S, Morris K, Cook MJ. Temporal lobe epilepsy caused by mesial temporal sclerosis and temporal neocortical lesions. A clinical and electroencephalographic study of 46 pathologically proven cases. Brain 1996; 119(6):213341. [42] Meador KJ. Emergence of temporal lobe surgery for epilepsy. Arch Neurol 2001; 58(6):10112. [43] Duchowny M, Levin B, Jayakar P, Resnick T, Alvarez L, Morrison G, et al. Temporal lobectomy in early childhood. Epilepsia 1992;33(2):298303. [44] Goldstein R, Harvey AS, Duchowny M, Jayakar P, Altman N, Resnick T, et al. Preoperative clinical, EEG, and imaging ndings do not predict seizure outcome following temporal lobectomy in childhood. J Child Neurol 1996;11(6):44550. [45] Harvey AS, Berkovic SF, Wrennall JA, Hopkins IJ. Temporal lobe epilepsy in childhood: clinical, EEG, and neuroimaging ndings and syndrome classication in a cohort with new-onset seizures. Neurology 1997;49(4):9608. [46] Kuzniecky RI, Barkovich AJ. Pathogenesis and pathology of focal malformations of cortical development and epilepsy. J Clin Neurophysiol 1996;13(6):46880. [47] Mohamed A, Wyllie E, Ruggieri P, Kotagal P, Babb T, Hilbig A, et al. Temporal lobe epilepsy due to hippocampal sclerosis in pediatric candidates for epilepsy surgery. Neurology 2001;56(12):16439.
1212
[48] Shinnar S. Prolonged febrile seizures and mesial temporal sclerosis. Ann Neurol 1998;43(4):4112. [49] Maganti R, Sheth RD, Rutecki PA, Gidal BE, Hermann BP. Lesional occipital epilepsy of childhood onset. Epilepsia 2001;42(7):173. [50] Yalcin AD, Kaymaz A, Forta H. Childhood occipital epilepsy: seizure manifestations and electroencephalographic features. Brain Dev 1997;19(6):40813. [51] Sheth RD, Riggs JE. Persistent occipital electrographic status epilepticus. J Child Neurol 1999;14(5):3346. [52] Salanova V, Andermann F, Olivier A, Rasmussen T, Quesney LF. Occipital lobe epilepsy: electroclinical manifestations, electrocorticography, cortical stimulation and outcome in 42 patients treated between 1930 and 1991. Surgery of occipital lobe epilepsy. Brain 1992;115:165580. [53] Kuzniecky R, Gilliam F, Morawetz R, Faught E, Palmer C, Black L. Occipital lobe developmental malformations and epilepsy: clinical spectrum, treatment, and outcome. Epilepsia 1997;38(2):17581. [54] Kuzniecky R. Symptomatic occipital lobe epilepsy. Epilepsia 1998;39(Suppl 4):S2431. [55] Aykut-Bingol C, Bronen RA, Kim JH, Spencer DD, Spencer SS. Surgical outcome in occipital lobe epilepsy: implications for pathophysiology. Ann Neurol 1998;44(1):609. [56] Aykut-Bingol C, Spencer SS. Nontumoral occipitotemporal epilepsy: localizing ndings and surgical outcome. Ann Neurol 1999;46(6):894900. [57] Huott AD, Madison DS, Niedermeyer E. Occipital lobe epilepsy. A clinical and electroencephalographic study. Eur Neurol 1974;11(6):32539. [58] Quesney LF. Extratemporal epilepsy: clinical presentation, pre-operative EEG localization and surgical outcome. Act Neurol Scand 1992;140(Suppl):8194. [59] Salanova V, Andermann F, Rasmussen T, Olivier A, Quesney LF. Tumoural parietal lobe epilepsy. Clinical manifestations and outcome in 34 patients treated between 1934 and 1988. Brain 1995;118(5):1289304. [60] Williamson PD, Boon PA, Thadani VM, Darcey TM, Spencer DD, Spencer SS, et al. Parietal lobe epilepsy: diagnostic considerations and results of surgery. Ann Neurol 1992; 31(2):193201. [61] Kotagal P, Rothner AD. Epilepsy in the setting of neurocutaneous syndromes. Epilepsia 1993;34(Suppl 3):S718. [62] Sujansky E, Conradi S. Outcome of Sturge-Weber syndrome in 52 adults. Am J Med Genet 1995;57(1):3545. [63] Bebin EM, Kelly PJ, Gomez MR. Surgical treatment for epilepsy in cerebral tuberous sclerosis. Epilepsia 1993;34:6517. [64] Palmini A, Andermann F, Olivier A, Tampieri D, Robitaille Y. Focal neuronal migration disorders and intractable partial epilepsy: results of surgical treatment. Ann Neurol 1991;30(6):7507. [65] Blume WT, Young GB, Lemieux JF. EEG morphology of partial epileptic seizures. Electroencephalogr Clin Neurophysiol 1984;57(4):295302. [66] Hamer HM, Najm I, Mohamed A, Wyllie E. Interictal epileptiform discharges in temporal lobe epilepsy due to hippocampal sclerosis versus medial temporal lobe tumors. Epilepsia 1999;40(9):12618. [67] Wyllie E, Chee M, Granstrom ML, DelGiudice E, Estes M, Comair Y, et al. Temporal lobe epilepsy in early childhood. Epilepsia 1993;34:85968. [68] Acharya JN, Wyllie E, Luders HO, Kotagal P, Lancman M, Coelho M. Seizure symptomatology in infants with localization-related epilepsy. Neurology 1997;48(1):18996. [69] Chen LS, Mitchell WG, Horton EJ, Snead III OC. Clinical utility of video-EEG monitoring. Pediatr Neurol 1995;12(3):2204. [70] Duchowny MS, Resnick TJ, Deray MJ, Alvarez LA. Video EEG diagnosis of repetitive behavior in early childhood and its relationship to seizures. Pediatric Neurology 1988; 4:1624.
1213
[71] Munari C, Tassi L, Di Leo M, Kahane P, Homann D, Francione S, et al. Video-stereoelectroencephalographic investigation of orbitofrontal cortex. Ictal electroclinical patterns. Adv Neurol 1995;66:27395. [72] Sheth RD. Electroencephalogram in developmental delay: specic electroclinical syndromes. Semin Pediatr Neurol 1998;5(1):4551. [73] Kurle PJ, Sheth RD. Gelastic seizures of neocortical origin conrmed by resective surgery. J Child Neurol 2000;15(12):8358. [74] Sheth RD, Hermann BP, Rowley H, Gidal BE, Haughton VM, Bell BD, et al. Pediatric epilepsy surgery: neuroimaging, neuropsychology, and anticonvulsants. Semin Pediatr Neurol 2000;7(3):16677. [75] Sheth RD. Intractable pediatric epilepsy: presurgical evaluation. Semin Pediatr Neurol 2000;7(3):15865. [76] Mizrahi EM. Pediatric electroencephalographic video monitoring. J Clin Neurophysiol 1999;16(2):10010. [77] Blume WT, Ravindran J, Lowry NJ. Late lateralizing and localizing EEG features of scalp-recorded temporal lobe seizures. J Clin Neurophysiol 1998;15(6):51420. [78] So N, Gotman J. Changes in seizure activity following anticonvulsant drug withdrawal. Neurology 1990;40(3):40713. [79] Marks DA, Katz A, Scheyer R, Spencer SS. Clinical and electrographic eects of acute anticonvulsant withdrawal in epileptic patients. Neurology 1991;41(4):50812. [80] Spencer SS, Spencer DD, Williamson PD, Mattson RH. Ictal eects of anticonvulsant medication withdrawal in epileptic patients. Epilepsia 1981;22(3):297307. [81] Duncan JS, Smith SJ, Forster A, Shorvon SD, Trimble MR. Eects of the removal of phenytoin, carbamazepine, and valproate on the electroencephalogram. Epilepsia 1989; 30(5):5906. [82] Gotman J, Koer DJ. Interictal spiking increases after seizures but does not after decrease in medication. Electroencephalogr Clin Neurophysiol 1989;72(1):715. [83] Malow BA, Blaxton TA, Stertz B, Theodore WH. Carbamazepine withdrawal: eects of taper rate on seizure frequency. Neurology 1993;43(11):22804. [84] Malow BA, Lynch B, Blaxton TA, Mikati MA. Relationship of carbamazepine reduction rate to seizure frequency during inpatient telemetry. Epilepsia 1994;35(6):11604. [85] Theodore WH, Porter RJ, Raubertas RF. Seizures during barbiturate withdrawal: relation to blood level. Ann Neurol 1987;22(5):6447. [86] Jackson GD. New techniques in magnetic resonance and epilepsy. Epilepsia 1994; 35(Suppl 6):S213. [87] Barkovich AJ, Rowley HA, Andermann F. MR in partial epilepsy: value of highresolution volumetric techniques. Am J Neuroradiol 1995;16(2):33943. [88] Barkovich AJ, Rowley H, Bollen A. Correlation of prenatal events with the development of polymicrogyria. Am J Neuroradiol 1995;16(Suppl 4):S8227. [89] Shinnar S, ODell C, Mitnick R, Berg AT, Moshe SL. Neuroimaging abnormalities in children with an apparent rst unprovoked seizure. Epilepsy Res 2001;43(3): 2619. [90] Bronen RA, Vives KP, Kim JH, Fulbright RK, Spencer SS, Spencer DD. Focal cortical dysplasia of Taylor, balloon cell subtype: MR dierentiation from low-grade tumors. Am J Neuroradiol 1997;18(6):114151. [91] Patel H, Garg BP, Salanova V, Boaz JC, Luerssen TG, Kalsbeck JE. Tumor-related epilepsy in children. J Child Neurol 2001;16(2):1415. [92] Barkovich AJ, Peacock W. Sublobar dysplasia: a new malformation of cortical development. Neurology 1998;50(5):13837. [93] Barkovich AJ, Kuzniecky RI, Bollen AW, Grant PE. Focal transmantle dysplasia: a specic malformation of cortical development. Neurology 1997;49(4):114852. [94] Schaefer GB, Sheth RD, Bodensteiner JB. Cerebral dysgenesis. An overview. Neurol Clin 1994;12(4):77388.
1214
[95] Raymond AA, Fish DR, Sisodiya SM, Alsanjari N, Stevens JM, Shorvon SD. Abnormalities of gyration, heterotopias, tuberous sclerosis, focal cortical dysplasia, microdysgenesis, dysembryoplastic neuroepithelial tumour and dysgenesis of the archicortex in epilepsy. Clinical, EEG and neuroimaging features in 100 adult patients. Brain 1995;118(3):62960. [96] Zupanc ML. Neuroimaging in the evaluation of children and adolescents with intractable epilepsy. II. Neuroimaging and pediatric epilepsy surgery. Pediatr Neurol 1997;17(2): 11121. [97] Rowley HA, Grant PE, Roberts TP. Diusion MR imaging. Theory and applications. Neuroimaging Clin N Am 1999;9(2):34361. [98] Dillon WP, Rowley H. The reversible posterior cerebral edema syndrome. Am J Neuroradiol 1998;19(3):591. [99] Lansberg MG, OBrien MW, Norbash AM, Moseley ME, Morrell M, Albers GW. MRI abnormalities associated with partial status epilepticus. Neurology 1999;52(5): 10217. [100] Duncan JS. Imaging and epilepsy. Brain 1997;120(2):33977. [101] Duncan JS. Magnetic resonance spectroscopy. Epilepsia 1996;37(7):598605. [102] Sheth RD, Haughton VM, Iskandar B, Hermann BP, Gidal BE. fMRI identies functional striate cortex in patients undergoing epilepsy surgery for occipital lesions. Epilepsia 2000;41:189. [103] Detre JA, Sirven JI, Alsop DC, OConnor MJ, French JA. Localization of subclinical ictal activity by functional magnetic resonance imaging: correlation with invasive monitoring. Ann Neurol 1995;38(4):61824. [104] Decoo D, Destee A. PET studies in epilepsy. Acta Neurol Belg 1997;97(3):1969. [105] Wyler A. The surgical management of epilepsy. In: Wyler A, Hermann B, editors. Boston: Butterworth-Heinemann; 1994. p. 12938. [106] Spencer D, Inserni J. Temporal lobectomy. In: Luders H, editor. Epilepsy surgery. New York: Raven Press; 1991. p. 53345. [107] Dodrill C, Hermann B, Rausch R, Chelune G, Oxbury S. Use of neuropsychological tests for assessing prognosis following surgery for epilepsy. In: Engel J, editor. Surgical treatment of the epilepsies. New York: Raven Press; 1993. p. 26371. [108] Trenerry MR, Jack Jr CR, Ivnik RJ, Sharbrough FW, Cascino GD, Hirschorn KA, et al. MRI hippocampal volumes and memory function before and after temporal lobectomy. Neurology 1993;43(9):18005. [109] Chelune G. Hippocampal adequacy versus functional reserve: predicting memory functions following temporal lobectomy. Arch Clin Neuropsychol 1995;10:41332. [110] Ojemann GA. Mapping of neuropsychological language parameters at surgery. Int Anesthesiol Clin 1986;24(3):11531. [111] Rausch R. Psychological evaluation. In: Engel J, editor. Surgical treatment of the epilepsies. New York: Raven Press; 1987. p. 18195. [112] Loring DW, Hermann BP, Meador KJ, Lee GP, Gallagher BB, King DW, et al. Amnesia after unilateral temporal lobectomy: a case report. Epilepsia 1994;35(4): 75763. [113] Loring DW, Meador KJ, Lee GP, King DW, Nichols ME, Park YD, et al. Wada memory asymmetries predict verbal memory decline after anterior temporal lobectomy. Neurology 1995;45(7):132933. [114] Fraser R, Gumnit R, Thorbecke R, Dobkin B. Psychosocial rehabilitation: a pre- and postoperative perspective. New York: Raven Press; 1992. [115] Hermann BP, Wyler AR, Somes G. Preoperative psychological adjustment and surgical outcome are determinants of psychosocial status after anterior temporal lobectomy. J Neurol Neurosurg Psychiatry 1992;55(6):4916. [116] Zaccariotti VA, Pannek HW, Holthausen H, Oppel F. Evaluation with subdural plates in children and adolescents. Neurol Res 1999;21(5):46374.
1215
[117] Paolicchi JM, Jayakar P, Dean P, Yaylali I, Morrison G, Prats A, et al. Predictors of outcome in pediatric epilepsy surgery. Neurology 2000;54(3):6427. [118] Jayakar P, Duchowny M, Resnick TJ. Subdural monitoring in the evaluation of children for epilepsy surgery. J Child Neurol 1994;9(Suppl 2):616. [119] Jayakar P, Duchowny M, Resnick TJ, Alvarez LA. Localization of seizure foci: pitfalls and caveats. J Clin Neurophysiol 1991;8(4):41431. [120] Adelson PD, ORourke DK, Albright AL. Chronic invasive monitoring for identifying seizure foci in children. Neurosurg Clin N Am 1995;6(3):491504. [121] Adelson PD, Black PM, Madsen JR, Kramer U, Rocko MA, Riviello JJ, et al. Use of subdural grids and strip electrodes to identify a seizure focus in children. Pediatr Neurosurg 1995;22(4):17480. [122] Behrens E, Zentner J, van Roost D, Hufnagel A, Elger CE, Schramm J. Subdural and depth electrodes in the presurgical evaluation of epilepsy. Acta Neurochir 1994;128 (14):847. [123] De Salles AA, Swartz BE, Lee TT, Delgado-Escueta AV. Subdural recording and electrical stimulation for cortical mapping and induction of usual seizures. Stereotact Funct Neurosurg 1994;62(14):22631. [124] Luders H, Awad I, Burgess R, Wyllie E, Van Ness P. Subdural electrodes in the presurgical evaluation for surgery of epilepsy. Epilepsy Res Suppl 1992;5:14756. [125] Lee KH, Park YD, King DW, Meador KJ, Loring DW, Murro AM, et al. Prognostic implication of contralateral secondary electrographic seizures in temporal lobe epilepsy. Epilepsia 2000;41(11):14449. [126] Wennberg R, Quesney F, Olivier A, Rasmussen T. Electrocorticography and outcome in frontal lobe epilepsy. Electroencephalogr Clin Neurophysiol 1998;106(4):35768. [127] Quesney LF, Cendes F, Olivier A, Dubeau F, Andermann F. Intracranial electroencephalographic investigation in frontal lobe epilepsy. Adv Neurol 1995;66:24358. [128] Alarcon G, Garcia SJ, Binnie CD, Martin MM, Juler J, Polkey CE, et al. Origin and propagation of interictal discharges in the acute electrocorticogram. Implications for pathophysiology and surgical treatment of temporal lobe epilepsy. Brain 1997; 120(12):225982. [129] Armon C, Radtke RA, Friedman AH, Dawson DV. Predictors of outcome of epilepsy surgery: multivariate analysis with validation. Epilepsia 1996;37(9):81421. [130] Spencer SS, Guimaraes P, Katz A, Kim J, Spencer D. Morphological patterns of seizures recorded intracranially. Epilepsia 1992;33(3):53745. [131] Spencer SS, Marks D, Katz A, Kim J, Spencer DD. Anatomic correlates of interhippocampal seizure propagation time. Epilepsia 1992;33(5):86273. [132] Lieb JP, Engel Jr J, Babb TL. Interhemispheric propagation time of human hippocampal seizures. I. Relationship to surgical outcome. Epilepsia 1986;27(3):28693. [133] Jung WY, Pacia SV, Devinsky R. Neocortical temporal lobe epilepsy: intracranial EEG features and surgical outcome. J Clin Neurophysiol 1999;16(5):41925. [134] Spreer J, Martin P, Greenlee MW, Wohlfarth R, Hammen A, Arnold SM, et al. Functional MRI in patients with band heterotopia. Neuroimage 2001;14(2):35765. [135] Jayakar P, Duchowny M, Alvarez L, Resnick T. Intraictal activation in the neocortex: a marker of the epileptogenic region. Epilepsia 1994;35(3):48994. [136] Jayakar P, Alvarez LA, Duchowny MS, Resnick TJ. A safe and eective paradigm to functionally map the cortex in childhood. J Clin Neurophysiol 1992;9(2):28893.

Neurologic Clinics Pediatric Neurology Part I

Uploaded by

Copyright:

Available Formats

Neurologic Clinics Pediatric Neurology Part I

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Neurologic Clinics Pediatric Neurology Part I

Uploaded by

Copyright:

Available Formats

Neurol Clin N Am 20 (2002) xixii

Pediatric Neurology, Part I

John B. Bodensteiner, MD James F. Bale Jr., MD Guest Editors

J.B. Bodensteiner, J.F. Bale / Neurol Clin N Am 20 (2002) xixii

Neurol Clin N Am 20 (2002) 917939

Brain development and the genetics of brain development

G.D. Clark / Neurol Clin N Am 20 (2002) 917939

G.D. Clark / Neurol Clin N Am 20 (2002) 917939

G.D. Clark / Neurol Clin N Am 20 (2002) 917939

G.D. Clark / Neurol Clin N Am 20 (2002) 917939

G.D. Clark / Neurol Clin N Am 20 (2002) 917939

G.D. Clark / Neurol Clin N Am 20 (2002) 917939

G.D. Clark / Neurol Clin N Am 20 (2002) 917939

G.D. Clark / Neurol Clin N Am 20 (2002) 917939

G.D. Clark / Neurol Clin N Am 20 (2002) 917939

G.D. Clark / Neurol Clin N Am 20 (2002) 917939

G.D. Clark / Neurol Clin N Am 20 (2002) 917939

G.D. Clark / Neurol Clin N Am 20 (2002) 917939

G.D. Clark / Neurol Clin N Am 20 (2002) 917939

G.D. Clark / Neurol Clin N Am 20 (2002) 917939

G.D. Clark / Neurol Clin N Am 20 (2002) 917939

G.D. Clark / Neurol Clin N Am 20 (2002) 917939

G.D. Clark / Neurol Clin N Am 20 (2002) 917939

G.D. Clark / Neurol Clin N Am 20 (2002) 917939

G.D. Clark / Neurol Clin N Am 20 (2002) 917939

G.D. Clark / Neurol Clin N Am 20 (2002) 917939

G.D. Clark / Neurol Clin N Am 20 (2002) 917939

G.D. Clark / Neurol Clin N Am 20 (2002) 917939

Neurol Clin N Am 20 (2002) 941964

R.S. Kandt / Neurol Clin N Am 20 (2002) 941964

R.S. Kandt / Neurol Clin N Am 20 (2002) 941964

R.S. Kandt / Neurol Clin N Am 20 (2002) 941964

R.S. Kandt / Neurol Clin N Am 20 (2002) 941964

R.S. Kandt / Neurol Clin N Am 20 (2002) 941964

Fig. 2. A hypomelanotic macule of tuberous sclerosis complex.

R.S. Kandt / Neurol Clin N Am 20 (2002) 941964

R.S. Kandt / Neurol Clin N Am 20 (2002) 941964

Fig. 5. A esh-colored ungual broma with a central red core.

R.S. Kandt / Neurol Clin N Am 20 (2002) 941964

R.S. Kandt / Neurol Clin N Am 20 (2002) 941964

R.S. Kandt / Neurol Clin N Am 20 (2002) 941964

Fig. 8. Axial T1-weighted MRI demonstrating subependymal nodules.

R.S. Kandt / Neurol Clin N Am 20 (2002) 941964

R.S. Kandt / Neurol Clin N Am 20 (2002) 941964

R.S. Kandt / Neurol Clin N Am 20 (2002) 941964

R.S. Kandt / Neurol Clin N Am 20 (2002) 941964

Fig. 10. Cafe au lait spot.

R.S. Kandt / Neurol Clin N Am 20 (2002) 941964

R.S. Kandt / Neurol Clin N Am 20 (2002) 941964

R.S. Kandt / Neurol Clin N Am 20 (2002) 941964

R.S. Kandt / Neurol Clin N Am 20 (2002) 941964

R.S. Kandt / Neurol Clin N Am 20 (2002) 941964

R.S. Kandt / Neurol Clin N Am 20 (2002) 941964

R.S. Kandt / Neurol Clin N Am 20 (2002) 941964

R.S. Kandt / Neurol Clin N Am 20 (2002) 941964

R.S. Kandt / Neurol Clin N Am 20 (2002) 941964

Neurol Clin N Am 20 (2002) 965981

Neuroradiology of the central nervous system in childhood

G.L. Hedlund / Neurol Clin N Am 20 (2002) 965981

G.L. Hedlund / Neurol Clin N Am 20 (2002) 965981