Neurogenic Language Disorders in Children (Fabbro)

Neurogenic Language Disorders
in Children
Related Books
Title: The Sciences of Aphasia: From Therapy to Theory

Edited by: Was Papathanaslou & Ria De Bleser (eds)
ISBN: 0080440738
Title: Speech-Language Pathology

Author: Carl R. Schneiderman & Robert E. Potter
ISBN: 0126278741
Title: Manifestations of Aphasia Symptoms in Different Languages

Author: Michel Paradis (ed)
ISBN: 0080436625
Title: Foundations of Stuttering

Author: Marcel E. Wingate
ISBN: 0127594515
Title: Clinical Phonetics & Phonology: Making Sense of Disordered Speech

Author: Sara Howard & Barry Heselwood
ISBN: 0123569184
Title: Concise Encyclopedia of Language Pathology

Author: Franco Fabbro (ed)
ISBN: 0080431518
Related Journals
Journal of Fluency Disorders

Editor: Richard Curlee
Journal of Communication Disorders

Editor: Theodore Glattke
Journal of Neurolinguistics
Editors: John Marshall & Michel Paradis
Brain & Language

Editor: Harry A. Whitaker
For further information on these titles, visit www.elsevier.com
For free abstracts and sample copies of our journals, visit www.ScienceDirect.com
Neurogenic Language Disorders
in Children
EDITED BY
Franco Fabbro
"E. Medea"Scientific Institute, University of I]dine
INTERNATIONAL ASSOCIATION OF
LOGOPEDICS AND PHONIATRICS
2004
ELSEVIER
Amsterdam - Boston - Heidelberg - London - New York - Oxford - Paris

San Diego - San Francisco - Singapore - Sydney - Tokyo
ELSEVIERB.V. ELSEVIER Inc. ELSEVIER Ltd ELSEVIER Ltd
Sara Burgerhartstraat 25 525 B Street, Suite 1900 The Boulevard, Langford Lane 84 Theobalds Road
P.O. Box 211, 1000 AE San Diego, CA 92101-4495 Kidlington, Oxford OX5 1GB London WC1X 8RR
Amsterdam, The Netherlands USA UK UK
© 2004 Elsevier Ltd. All rights reserved.
This work is protected under copyright by Elsevier Ltd., and the following terms and conditions apply to its use:
Photocopying
Single photocopies of single chapters may be made for personal use as allowed by national copyright laws. Permission of the
Publisher and payment of a fee is required for all other photocopying, including multiple or systematic copying, copying
for advertising or promotional purposes, resale, and all forms of document delivery. Special rates are available for educational
institutions that wish to make photocopies for non-profit educational classroom use.
Permissions may be sought directly from Elsevier's Rights Department in Oxford, UK: phone (+44) 1865 843830,
fax (+44) 1865 853333, e-mail: permissions@elsevier.com. Requests may also be completed on-line via the Elsevier homepage
(http://www.elsevier.com/locate/permissions).
In the USA, users may clear permissions and make payments through the Copyright Clearance Center, Inc., 222 Rosewood Drive,
Danvers, MA 01923, USA; phone: (+1) (978) 7508400, fax: (+1) (978) 7504744, and in the UK through the Copyright Licensing
Agency Rapid Clearance Service (CLARCS), 90 Tottenham Court Road, London W1P 0LP, UK; phone: (+44) 20 7631 5555;
fax: (+44) 20 7631 5500. Other countries may have a local reprographic rights agency for payments.
Derivative Works
Tables of contents may be reproduced for internal circulation, but permission of the Publisher is required for external resale or
distribution of such material. Permission of the Publisher is required for all other derivative works, including compilations and
translations.
Electronic Storage or Usage

Permission of the Publisher is required to store or use electronically any material contained in this work, including any chapter
or part of a chapter.
Except as outlined above, no part of this work may be reproduced, stored in a retrieval system or transmitted in any form or by any
means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission of the Publisher.
Address permissions requests to: Elsevier's Rights Department, at the fax and e-mail addresses noted above.
Notice
No responsibility is assumed by the Publisher for any injury and/or damage to persons or property as a matter of products liability,
negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material
herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages
should be made.
First edition 2004
Library of Congress Cataloging in Publication Data

A catalog record is available from the Library of Congress.
British Library Cataloguing in Publication Data

A catalogue record is available from the British Library.
ISBN: 0 08 044549 7
@ The paper used in this publication meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper).
Printed in The Netherlands.
V
ACKNOWLEDGEMENTS
This book collects the papers presented at the International Symposium of the IALP Aphasia
Committee on Neurogenic Language Disorders in Children, held in Cividale del Friuli
(Udine, Italy), on 9-10 May 2003. The Symposium was organized by Scientific Institute
"E.Medea" of Association "la Nostra Famiglia", the Faculty of Education of the University of
Udine, and IALP. Financial support was granted by Fondazione CRUP. The success of the
meeting, in terms of scientific innovation and public participation, was also due to the work of
Dr. Alessandro Tavano, Scientific Secretary to the Symposium, in dealing with both scientific
and organizational issues. The administrative board and organizational secretariat of IRCCS
"E.Medea" of San Vito al Tagliamento (Pordenone. Italy) and Pasian di Prato (Udine, Italy)
provided invaluable assistance and deserve the utmost gratitude. Editorial assistance to this
volume was provided by Dr. Alessandro Tavano and Dr. Barbara Alberti.
This page is intentionally left blank
vii
Contents
1 Neurogenic Language Disorders in Children: An Introduction. 1-7
Franco Fabbro
2 Pathophysiological Basis of Aphasia and Verbal Outome in Landau- 9-23

Kleffner Syndrome.
Marie-Noelle Metz-Lutz and Steve Majerus
3 Acquired Language Disorders and Epilepsy: From Landau -Kleffner 25-35

Syndrome to Autistic Regression.
Roberto Tuchman
4 Persistent Subtle Language and Learning Deficits in a Child with 37-48

Acquired Epileptiform Opercular Syndrome.
Paola Cipriani, Anna M. Chilosi, Claudia Casalini, Lucia Pfanner,
Annarita Ferrari, Daniela Brizzolara and Renzo Guerrini
5 Cerebral Language Lateralization and Early Linguistic In Children 49-63

With Focal Brain Lesions.
Anna M. Chilosi, Chiara Pecini, Paola Cipriani, Daniela Brizzolara,
Paola Brovedani, Giovanni Ferretti, Lucia Pfanner and Giovanni
Cioni
6 Language Disorders Associated With Paroxysmal Abnormalities 65-85

During NREM Sleep After Very Early Brain Lesions.
Franco Fabbro, Alessandro Tavano, Guido Cristofori
and Renato Borgatti
7 Language and Phonological Awareness Abilities of Children treated 87-126

for Posterior Fossa Tumor.
Bruce E. Murdoch, Kimberley M. Docking, and Elizabeth C. Ward
8 Language Development in Children with Cerebellar Malformations.

Renato Borgatti, Alessandro Tavano, Guido Cristofori 127-145
and Franco Fabbro
viii Contents
9 Crossed Aphasia in Children.

Peter Marien, Philippe Paquier, Sebastiaan Engelborghs 147-180
and Peter P. De Deyn
10 Recognizable Spontaneous Language Characteristics in a Young 181 -197

Adult Twelve Years After She Became Aphasic as a Child.
Philippe F. Paquier, Valerie R. van Maldeghem,
Hugo R. van Dongen and Wouter L. Creten
11 Recovery From Aphasia After Polytrauma in a Czech Child: 199-229

What Is Lost and What Is Left.
Helena Leheckovd
12 Persistent Acquired Childhood Aphasia. 231-251

Isabel Pavao Martins
Introduction 1
NEUROGENIC LANGUAGE DISORDERS IN

CHILDREN: AN INTRODUCTION
Franco Fabbro
"E. Medea" Scientific Institute, Polo del Friuli Venezia Giulia, Italy
University ofUdine, Italy
INTRODUCTION
Language disorders in children are one of the most frequent causes of difficulties in
communication, social interaction, learning and academic achievement. It has been estimated
that over 5% of children present with some kind of language disorders (Fabbro, 1999).
Language disorders in children are distinguished into: 1) acquired language disorders; 2)
language disorders due to pre-perinatal lesions; 3) developmental language disorders; 4)
language disorders in genetic syndromes with mental retardation (Fabbro, 2000a).
ACQUIRED LANGUAGE DISORDERS IN CHILDREN
Acquired language disorders in children can be distinguished into three main groups: acquired
childhood aphasia, language disorders following posterior cranial fossa lesions (PCF) and
acquired epileptiform aphasias.
Acquired childhood aphasia
Acquired childhood aphasia refers to language deficits following brain lesions after the age of
acquisition of the first sentences, generally after the age of 2. The most common etiological
causes include vascular lesions, trauma, tumors and infections involving the language
dominant hemisphere {see Marien et al., this volume, for a review of the cases of "crossed
2 Neurogenic Language Disorders in Children
aphasia in children" or acquired childhood aphasia after right hemisphere lesions in right-
handed children). Before the late 1970s it was believed that aphasia in children was
characterized by "negative symptoms" such as mutism, dysarthria, reduction in sentence
length and telegraphic speech. More recently, neurolinguistic tests, including systematic
analyses of spontaneous and descriptive speech, and comprehensive language batteries have
allowed researchers to evaluate the different aspects of language (for example, auditory,
semantic and syntactic comprehension; syllable, word and sentence repetition; naming and
sentence production). In children with acquired aphasia, these methods highlight positive
symptoms such as logorrhea, paraphasia, perseverations and neologisms (see Leheckovd,
Paquier et al., this volume). Moreover, these symptoms correlate with clinical aphasic
profiles and lesion localizations similar to those of adults. Indeed, at least in their acute phase
and lesional phase aphasic syndromes in children have been found similar to most of the
aphasic syndromes in adults (cf. Van Hout, 2000). Recovery of acquired childhood aphasia
remains one of the most debated issue still today. Before the early 1970s language recovery in
childhood aphasia was believed to be rapid and complete (Lenneberg, 1967). Later studies
have shown that, even if language seems to return to normal, non-linguistic abilities (such as
working memory) are affected too. Therefore, irrespective of age and lesion etiology, children
encounter educational difficulties. These findings stress the need for rehabilitation and
educational/professional support in the chronic stages of childhood aphasia (see Pavao
Martins, this volume).
Language disorders following PCF lesions
Almost 50% of brain tumors in children involve the cerebellum (medulloblastomas, cerebellar
astrocytomas, ependymomas; cf. Becker and Jay, 1990), a structure localized in the posterior
cranial fossa (PCF). In 10% of the children surgical removal of the tumor can cause a
syndrome characterized by complete but transient loss of speech (transient cerebellar
mutism), followed by dysarthria. This syndrome is frequent in patients aged 2-16 years. There
have been reports of patients who became mute within 12 to 48 hours of surgery and the
period of mutism lasted from 1.5 to 12 week after onset. Transient cerebellar mutism seems to
be due to a diaschisis on the nervous structures of the brain stem which are responsible for
verbal expression (cf. Pollak, 1997; Esposito et al., 1999). Recently, the cerebellum was also
attributed an important role in the regulation of linguistic, cognitive and affective functions
(cf. Fabbro, 2000b). Resection of cerebellar tumors both in childhood and adult age may
cause a "Cognitive - Affective Syndrome" (Levisohn et al, 2000; Riva and Giorgi, 2000)
characterized by expressive language deficits, verbal memory deficits (mainly after right
hemisphere cerebellar lesions), deficits in the visuo-spatial functions (after left hemisphere
cerebellar lesions) and deficits in affect regulation (after vermis damage) (see Murdoch et al.,
this volume). Deficits in the development of linguistic and cognitive functions were also
described in patients with malformation lesions localized to the cerebellum (see Borgatti et
al., this volume).
Introduction 3
Acquired Epileptiform Aphasias
The main clinical syndrome of acquired epileptiform aphasias is acquired aphasia with
convulsive disorder. It manifests in children, generally at 3-8 years of age. It was first
described by Landau and Kleffner (1957); hence, its definition as Landau-Kleffner syndrome
(LKS) (cf. Lebrun and Fabbro, 2002). It affects children with normal psychomotor and
linguistic development. Language disruption may occur before or after seizures. At onset, the
most frequent symptom is breakdown in language comprehension, whereas hearing and
interpretation of non-linguistic sounds remain intact (word deafness). Following the
breakdown in comprehension, expressive language and vocabulary decay progressively, too.
Unlike other forms of acquired aphasia, in LKS verbal expression may be fluent with many
semantic and verbal paraphasias, neologisms and jargon. There may be an almost full
recovery after the first episode, even after a short period of time (days or weeks). In other
cases, recovery may be very slow (months or years). Relapse is frequent and often the
development of aphasia is fluctuating, with several aphasic episodes. The clinical picture
stabilizes before the end of adolescence. Language recovery is sometimes very good, at other
times it is defective, in which case subjects present with aphasic disorders throughout their
lives (see Metz-Lutz and Majerus, this volume). Epileptic manifestations are heterogeneous
(partial motor seizures, atypical absence, generalized tonic-clonic seizures) and their
frequency is quite variable. Children with LKS often show a rather limited number of seizures
which can be treated with anti-epileptic drugs. Seizures may disappear completely before the
age of 15. In a review of the first cases described, Mantovani and Landau (1981) reported
that, after more than 20 years, none of their patients still had seizures. On waking EEG
paroxysmal abnormalities in children with LKS may vary: bilateral temporal or temporo-
parietal spikes, bilateral 1-3 Hz slow wave activity over the temporal regions, generalized
sharp waves or slow wave discharges, and multifocal or unilateral spikes. The background
rhythm of the EEG in waking state is normal. Generally, the epileptic focus is never stable,
even if statistically it tends to be over the left temporal regions. An extremely significant
feature which seems to be common to all children with LKS is an EEG with bilateral
(focalized) paroxysmal abnormalities during non-REM sleep which is constantly associated
with regression of language functions (cf. Fabbro and Zucca, 2000). Recently, many other
clinical syndromes were associated to Landau-Kleffner Syndrome, and they include Acquired
Epileptiform Opercular Syndrome (AEOS) (see Cipriani et al., this volume), the Continuous
Spike-Waves during Slow Sleep (CSWS), the Benign Childhood Epilepsy with
Centrotemporal Spikes (BECTS), and the Autistic Regression with an Epileptiform EEG. All
of these clinical syndromes are associated with different degrees of cognitive disorders or
mental retardation and language disorders (see Tuchman, this volume).
LANGUAGE DISORDERS DUE TO PRE-PERINATAL LESIONS
The most frequent cause of prenatal lesion is cerebral palsy. The incidence of cerebral palsy
of pre-perinatal origin is approximately 2 in 1000 births. Of these cerebral palsy cases
approximately one-third have hemiplegia. The underlying unilateral hemispheric brain injury
is supposed be due to a thrombotic, vasospasmic or embolic episode in the middle cerebral
and/or the internal carotid artery. Between 30% and 40% of such hemiplegic cases develop a
cerebral seizure disorder. As documented by many studies, the neuropsychological sequelae
of early brain damage are relatively mild if compared with those of adults. Besides, contrary
to the case of adults, not all cases of early brain damage show a clear-cut correlation of
symptoms with characteristics of the lesion. The degree of preservation or recovery of
language and other cognitive functions may be dependent on several factors such as time of
injury, side of lesion, size of lesion; presence of epilepsy and role of anticonvulsant therapy.
Numerous studies have been carried out to verify the effects of early left hemisphere damage
(LHD) versus right hemisphere damage (RHD) on cognitive development, with particular
attention to language development. Some studies show that, regardless of the affected
hemisphere, linguistic functions tend to be preserved, while other studies reveal greater
difficulties in children with left hemisphere damage in the acquisition of vocabulary and
grammar (see Chilosi et al, this volume). Numerous reports demonstrate that seizures
accompanying early brain injury are associated with poorer language and cognitive outcome
(cf. Vargha-Khadem et al., 1992). The pathophysiological mechanism determining a deficit in
cognitive and linguistic development in children with early hemisphere damage and epilepsy
has not been clearly identified yet. Some authors have suggested that antiepileptic therapy
may play an unfavorable role, while other authors have drawn the attention to some variables
such as the presence of epileptiform abnormalities in NREM sleep, which associate these
clinical pictures to Landau-Kleffner Syndrome (see Fabbro et al., this volume).
DEVELOPMENTAL LANGUAGE DISORDERS
Developmental language disorders (DLDs) — also known as Specific Language Impairment

(SLI) or Developmental Dysphasia — are language acquisition disorders manifesting in
children with normal nonverbal intellectual development in the absence of hearing loss, frank
neurological deficits, severe emotional disorders and environmental differences and
deprivation. Children with specific language impairment show a very slow and laborious
language development. They speak late as compared to peers and show comprehension and
production disorders affecting more than one linguistic level (phonological, morphological,
lexical, syntactic and semantic). Developmental language disorders are to be distinguished
from acquired language disorders (acquired aphasia) affecting children with normal language
development until their occurrence. The frequency of DLDs in children aged 5-6 years is
surprisingly high, around 4% (Fletcher and Hall, 1993). Many classifications for
Introduction 5
developmental language disorders in children have been proposed, some of which are based
on statistical criteria (e.g., the International Classification of Diseases, 10th Edition, ICD -10th,
1992), others on clinical data (Rapin, 1996). In the ICD-10th classification developmental
language disorders are described in section F80 encompassing three syndromes: 1) Specific
speech articulation disorders (F 80.0): children use sounds that are below the norm as
compared to their mental age, whereas all other linguistic tasks are in the normal range; 2)
Expressive language disorder (F80.1): the expressive spoken language ability of the children
is markedly below the appropriate level for their mental age, and there may be an alteration in
speech articulation; 3) Receptive language disorder (F80.2): the receptive ability of the
children to understand spoken language is markedly below the appropriate level for their
mental age and expressive language will also be markedly affected. Many studies have
investigated the causes of DLDs (cf. Bishop, 1997; Leonard, 1999). At the pathophysiological
level, some children are unable to discriminate language sounds at the normal speech speed
(Merzenich et al, 1996), in others expressive language disorders are associated to
developmental verbal dyspraxia (Shriberg et al., 1997) or to language acquisition deficits with
a genetic origin (Gopnik, 2000). Several studies have shown that many children with DLDs
present with paroxysmal abnormalities in non-REM sleep similar to those found in Landau-
Kleffner Syndrome (Duvelleroy-Hommet et al, 1995). In particular, more than 50% of
children with receptive language disorder (F80.2) have epileptiform abnormalities in NREM
sleep (Picard et al, 1998; Fabbro et al, 2000). According to preliminary studies, an
association of pharmacological treatment with valproic acid to speech therapy markedly
improves language development in these children. Recently, Guerreiro et al. (2002) have
made a relevant progress in understanding the causes of DLDs. They systematically studied
the neuroimaging findings (MRI 2.0 T scanner) in a group of children with DLDs. All the
children presented with pictures of polymicrogyria that were related to the degree of severity
of dysphasia. Their findings suggest that developmental language disorders are associated to
malformations of cortical development.
LANGUAGE DISORDERS IN GENETIC SYNDROMES WITH MENTAL

RETARDATION
An important line of research in modern clinical neurolinguistics is the study of language

deficits in genetic syndromes with mental retardation. Recent studies have shown that some
genetic syndromes such as Down Syndrome, Williams Syndrome, Fragile X Syndrome,
Turner Syndrome, Prader-Willi Syndrome, etc., have deficits that are specific to each of them.
For example, in Down Syndrome, the phonetic-phonological and the morphosyntactic levels
are particularly affected, while in Williams Syndrome these levels are sufficiently spared,
while the pragmatic level is impaired (cf. Rondal and Edwards, 1997). Language deficits
peculiar to genetic syndromes are correlated with specific neurofunctional alterations that are
typical of each of these pictures (cf. Tager-Flusberg, 1999).
CONCLUSION
The different neurogenic language disorders described here and discussed in detail by the
authors who contributed to this volume suggest that language disorders in children, both
acquired and developmental, should be managed by an interdisciplinary approach. Each
single disorder should also be studied keeping in mind the whole range of childhood language
disorders.
REFERENCES
Becker, L.E. and V. Jay (1990). Tumors of the central nervous system in children. In:
Management of Childhood Brain Tumors (Deutsch M., ed.), pp. 5-51. Kluwer, Boston.
Bishop, D. V. (1997). Uncommon Understanding. Development and Disorders of Language
Comprehension in Children. Psychology Press, Hove.
Duvelleroy-Hommet, C, C. Billard, B. Lucas, P. Gillet, M. A. Barthez, J. J. Santini, E.
Degiovanni, F. Henry, B. De Toffol and A. Autret (1995). Sleep EEG and
developmental dysphasia: Lack of consistent relationsip with paroxysmal EEG activity
during sleep. Neuropediatrics, 26, 14-18.
Esposito, A., G. Demeurisse, B. Alberti and F. Fabbro (1999). Complete mutism after
midbrain periaqueductal gray lesion. NeuroReport, 10, 681-685.
Fabbro, F. (1999). Concise Encyclopedia of Language Pathology. Pergamon, Oxford.
Fabbro, F. (2000a). Languages Disorders in Children: An Introduction. Saggi, 26, 9-12.
Fabbro, F. (2000b). Introduction to language and cerebellum. JNeuroling, 13, 83-94.
Fabbro, F. and C. Zucca (2000). Acquired neuropsychological disorders in children with
epilepsy. Saggi, 26, 23-29.
Fabbro, F., C. Zucca, M. Molteni and R. Borgatti (2000). EEG abnormalities during slow
sleep in children with developmental language disorders. Saggi, 25, 41-48.
Fletcher, P. and D. Hall (1993). Specific Speech and Language Disorders in Children.
Singular, San Diego.
Gopnik, M. (2000). The investigation of genetic dysphasia. Saggi, 26, 31-40.
Guerreiro, M. M., S. R. Hage, C. A. Guimaraes, D. V. Abramides, W. Fernandes, P. S.
Pacheco, A. M. Piovesana, M. A. Montenegrol and F. Cendes (2002). Developmental
language disorders associated with polymicrogyria. Neurology, 59, 245-250.
International Statistical Classification of Diseases and Related Health Problems. Tenth
Revision (1992). World Health Organization, Geneva.
Landau, W. M. and F. R. Kleffner (1957). Syndrome of acquired aphasia with convulsive
disorder in children. Neurology, 7, 523-530.
Lebrun, Y. and F. Fabbro (2002). Language and Epilepsy. Whurr, London.
Lenneberg, E. H. (1967). Biological Foundations of Language. John Wiley & Sons, New
York.
Introduction 1
Leonard, L. B. (1998). Children with Specific Language Impairment. MIT Press, Cambridge.
Levishon, L., A. Cronin-Golomb and J. D. Schmahmann (2000). Neuropsychological
consequences of cerebellar tumor resection in children. Brain, 123, 1041-1050.
Mantovani, J. F. and W. M. Landau (1980). Acquired aphasia with convulsive disorder:
Course and prognosis. Neurology, 30, 524-529.
Merzenich, M. M., W. M. Jenkins, P. Johnston, et al. (1996) Temporal processing deficits of
language-learning impaired children ameliorating by training. Science, 271, 77-81.
Picard, A., F. Cheliout Heraut, M. Bouskraoui, M. Lemoine, P. Lacert and J. Delattre (1998).
Sleep EEG and developmental dysphasia. Dev Med Child Neurol, 40, 595-599.
Pollack, I. F. (1997). Posterior fossa syndrome. Int RevNeurobiol, 41, 411-432.
Rapin, I. (1996). Preschool Children with Inadequate Communication. Mac Keit Press,
London.
Riva, D. and C. Giorgi (2000). The Cerebellum contributes to higher functions during
development. Brain, 123, 1051-1061.
Rondal, J. A. and S. Edwards (1997). Language in Mental Retardation. Whurr, London.
Shriberg, L. D., D. M. Aram and J. Kwiatkowski (1997). Developmental apraxia of speech. J
Speech Hear Res, 40, 273-285.
Tager-Flusberg, H. (1999). Neurodevelopmental Disorders. The MIT Press, Cambridge.
Van Hout, A. (2000). An outline of acquired aphasia in children. Saggi, 26, 13-21.
Vargha-Khadem, F., E. Isaacs, S. Van der Werf, S. Robb, J. Wilson (1992). Development of
intelligence and memory in children with hemiplegic cerebral palsy. Brain, 115, 315-
329.
Pathophysiology of Landau-Kleffner Syndrome 9
PATHOPHYSIOLOGICAL BASIS OF APHASIA AND

VERBAL OUTCOME IN LANDAU-KLEFFNER
SYNDROME
Marie-Noelle Metz-Lutz
Louis Pasteur University, France
Steve Majerus
University of Liege and Fonds National de la Recherche Scientifique, Belgium
Abstract Acquired childhood aphasia with epilepsy described by Landau and Kleffncr (LKS) in
1957, differs, by its clinical features and prognosis, from other types of aphasia in children due to
brain structural lesions. Several critical features appear to influence consistently the prognosis of
language disorder: the age at onset and the duration of language disorder associated with EEG
abnormalities during wakefulness and sleep and the location of the epileptic focus. In order to
elucidate the pathophysiological basis of epileptic aphasia and particularly of its poor outcome, we
reexamine neuropsychological, electrophysiological and neuro-imaging data from the long-term
follow-up study of several cases of Landau and Kleffner Syndrome (LKS). Our neuropsychological
findings show that although the outcome of language abilities is variable, the common residual verbal
disorders associate an impaired phonological short-term memory with a permanent one-ear extinction
on dichotic listening tests contraleral to the temporal cortex previously affected by the epileptic focus.
To check the hypothesis of a permanent dysfunction in the auditory system suggested by these
findings, we studied auditory event-related potentials studies in order to localize the level of the
dysfunction in auditory language processing and along the auditory pathways. Using H215O labeled
positron emission tomography (PET), we compared the brain activation for immediate serial recall of
lists of 4 words, contratsted to single word repetition in three recovered LKS patients and 14 healthy
controls. Both ERP's and PET's findings suggest that the residual verbal impairments at the late
outcome of LKS might indeed be related to persistent dysfunction in the temporal regions that were
involved in the epileptic focus during the active phase.
Keywords: childhood epilepsy, Landau-Kleffner Syndrome, language, phonological short-

term memory, pathophysiology, brain imaging
INTRODUCTION
In 1957, Landau and Kleffner reported the cases of six children, instructed in a deaf institute
of Saint Louis-Missouri, who developed aphasia in relation to convulsive disorder. They
defined the syndrome as acquired childhood aphasia with convulsive disorder, called several
years later Landau-Kleffner Syndrome (LKS). Following this seminal report, no new cases of
acquired epileptic aphasia were described until 1968. From this date, an increasing number of
reports were published providing deeper analysis of the aphasic symptoms, the
electrophysiological features of epileptic discharges and the outcome of LKS. In 1992,
William Landau's editorial in Annals of Neurology defined the label "Landau-Kleffner
Syndrome" "an eponymic badge of ignorance" (Landau, 1992).
Indeed, the still prevailing use of the generic label seems to imply that our knowledge
about the aetiology, pathological physiology and therapy of the disease has made no
significant progress during the 35-year history of this clinical phenomenon. Have we really
made no advancements in the understanding of the pathophysiological basis of epileptic
childhood aphasia or was Landau overly pessimistic in his appraisal of the scientific
literature? In the present paper, we intend to provide a more optimistic review of the different
studies aimed at understanding the relationship between the two main symptoms of this
disease, i.e. epilepsy and aphasia. Firstly, we will summarize the main clinical and
electrophysiological features of the syndrome and discuss their relationships with other age-
related epileptic conditions. From a set of recent metabolic and electrophysiological findings
obtained both during the active phase and after recovery of LKS, we will set out a provisional
pathophysiological account of epileptic aphasia and its residual verbal impairment.
CLINICAL FEATURES OF ACQUIRED EPILEPTIC APHASIA
Beaumanoir (1992) reviewed about 200 cases described in 57 papers published between 1968
and 1990. In the following decade, no less than 145 papers were devoted to LKS adding 118
new clinical descriptions. This set of case reports provides a rather sound outline of the
common clinical features of acquired epileptic aphasia.
Epileptic disorders
Despite its rarity - in a twenty-year cohort study of 440 epileptic children, Landau-Kleffner
Syndrome represents less than 0.5 % (Kramer et al, 1998) - the syndrome comprises very
typical features. It is characterised by a sudden onset between the age of 3 and 8 in children
with otherwise usually normal neurocognitive development. In 70% of the cases, the onset is
before age 6 and rarely occurs after the age of 8 (7%). Less than 13% of the cases may have
had an impaired language development. Overt epileptic seizures are not present in all LKS
children, only 72% of them present at least one seizure. One third experiences only one
seizure, usually at the beginning of the disease. When present, seizures are of various types,
but generalised or partial motor seizures are the most frequent. Whatever seizures are present,
the EEG is always abnormal with repeated large-amplitude spikes followed by a large slow
wave. During wake, these spike-wave discharges (SWDs) occurring on an almost normal
background activity have a focal organisation (Figure 1). Usually the epileptic focus is
unilateral. The observation of bilateral or multiple foci is not rare and raises the issue of
secondary epileptic foci. Initially focal, the SWDs progressively spread to the whole
hemisphere, but remain predominant over the temporal derivations.
Figure 1 — A Landau-Kleffner Syndrome typical EEG recorded during wakefulness showing

paroxysmal spike-wave discharges (SWDs). The higher amplitude of SWDs over the left temporal
electrodes (C3/T3; T3/O1) is indicative of the temporal focus of SWDs.
Figure 2 — LKS EEG recorded during sleep showing an aspect of continuous spike wave discharges
during slow sleep (CSWSS).
During sleep, SWDs increase in frequency and spread to the contralateral hemisphere, leading
to an aspect of continuous spike wave discharges during slow sleep (CSWSS) (Figure 2),
which covers at least 80% of the sleep time. This sleep EEG profile, in the presence of non-
lesional acquired childhood aphasia, may be considered as a hallmark for the diagnosis of
LKS.
Typical neuropsychological impairments
Aphasia is the first symptom in 54% of LKS case reports. Verbal impairment begins with
comprehension disorder and auditory agnosia in 85% of cases. Expressive disorders appear
progressively and are characterised by vocabulary loss and speech reduction leading
progressively to muteness. The severe deficits in auditory comprehension are the most
common feature of LKS. Often LKS children are thought, at first, to be deaf, although
audiometric investigations show normal hearing. Auditory disorders often appear like an
auditory agnosia involving both verbal and non-verbal discrimination (Koeda & Kohno,
1992; Maquet et al, 1995). However, an in-depth evaluation of auditory discrimination
evidences a dissociation between the discrimination of environmental sounds and
phonological auditory discrimination, suggesting that the primary deficit of the receptive
aphasia is an impairment of auditory phonological discrimination rather than a generalised
auditory agnosia (Korkman et al, 1998; Metz-Lutz et al, 1999b).
In most cases reported in the literature, expressive language impairments follow the onset
of auditory comprehension deficits. Expressive disorders typically begin either with a
progressive loss of vocabulary or phonological disturbances. In one recent case report,
stuttering was the first symptom (Tutuncuoglu et al, 2002). Some authors have seen the
expressive disorders of LKS as secondary to the impairment of phonological decoding.
Expressive language disorders gradually increase and children become mute within several
weeks. However, various patterns of expressive disorders have been described with
agrammatism, echolalia, anomia or phonetic distortions (Paquier et al, 1992). In contrast to
childhood aphasia due to focal lesions, paraphasia and neologisms are frequent aphasic
symptoms observed before the complete loss of expressive language that may last several
months, even years, if antiepileptic treatment is ineffective.
Behavioural impairments with hyperactivity are mentioned in 78% of reported LKS cases.
Landau and Kleffner already mentioned attention disorders and hyperactivity, which they
viewed as a psychological reaction to impaired auditory comprehension. Preservation of non-
verbal abilities is another common feature of LKS. This allows the use of gestures or lip-
reading as an alternative mode of communication. Figure 3 illustrates the typical performance
profile on the different subtests of the Wechsler Intelligence Scale for Children Revised
(WISC-R) showing very poor performances on the verbal subtests, particularly the digit span
and vocabulary tests. The only disturbed nonverbal subtest is typically the coding test that
requires attention capacities.
Figure 3 — Typical performance profile on the Wechsler Intelligence Scale for Children Revised
observed in LKS. The stars indicate the subtests which are the more systematically affected.
CLINICAL EVOLUTION
Along the course of LKS, the degree of expressive impairments fluctuates. Periods of
transient recovery are often observed after the introduction or change of anti-epileptic
medication that successfully normalises the EEG, for several weeks or months ( Mantovani &
Landau, 1980; Dugas et al, 1982; Marescaux et al, 1990). Several group studies emphasised
the almost parallel fluctuation of aphasic disorders and EEG abnormalities and discussed the
pharmacological sensitivity of both symptoms (Marescaux et al, 1990; Paquier et al, 1992;
Robinson, Baird, Robinson, & Simonoff, 2001). They showed that medication active on the
GABAergic receptors like Benzodiazepine (BDZ) favourably influences epileptic and aphasic
symptoms, but most often only transiently. Conversely, medication typically active in focal
epilepsy, particularly in temporal lobe epilepsy, like carbamazepine, usually worsens
epilepsy, aphasia and behavioural disorders (Beaumanoir, 1992). Although their mode of
action remains unknown, the corticosteroids have been shown to be useful after relapse of
epilepsy and aphasia following an initial treatment by BDZ (Marescaux et al, 1990).
According to the specific pattern of aphasic disorders involving primarily auditory
comprehension, the use of a manual language has been proposed to prevent continuing
problems with communication (Baynes et al, 1998; Perez et al, 2001). In this way,
rehabilitation strategies involving signed language or cued speech have been shown to be
adequate to bypass language deprivation during the active period of LKS. Indeed, the active
period of epileptic aphasia ranging from 3 to 10 years overlaps the most critical period for the
development of phonological and syntactic skills as well as for verbal learning.
LATE OUTCOME OF LKS
While epileptic seizures and EEG abnormalities completely disappear at the age of 12 or 13
(Dugas et al, 1982; Mantovani & Landau, 1980; Paquier et al, 1992), recovery of receptive
and productive language is variable. The verbal outcome seems to be poorer in LKS than in
acquired childhood aphasia related to head injury. Moreover, in LKS outcome appears to be
better for late-onset aphasic disorders, in contrast to childhood aphasia consecutive to an
acquired brain lesion (Bishop, 1985; Beaumanoir, 1992; Kaga, 1999). Finally, poorer
outcome appears to be related to the duration of epileptic aphasia (Metz-Lutz et al, 1999b;
Robinson et al, 2001). Some children recover normal or near to normal written and oral
language (Mantovani & Landau, 1980; Soprano et al, 1994; Robinson et al, 2001).
However, most LKS patients show difficulties, affecting phonological aspects of language
processing most severely (Zardini et al, 1995; Metz-Lutz et al, 1996; Metz-Lutz et al,
1999a). Chronic auditory agnosia has been described in the presence of the most
unfavourable verbal outcome (Baynes et al, 1998; Sieratzki et al, 2001). Residual verbal
impairments have been reported to involve both phonological and metaphonological
judgments (Zardini et al, 1995; Notoya et al, 1991; Ege & Mouridsen, 1998; Vance et al,
1999), phonological verbal fluency (Metz-Lutz et al, 1999b) and articulatory aspects of
speech production (Soprano et al, 1994). Several long-term follow-up studies showed a
better outcome for naming and syntactic skills ( Zardini et al, 1995; Metz-Lutz et al, 1999b).
One very consistent finding is a deficit in phonological short-term memory (STM)
performance, even in patients showing relatively good language recovery (Soprano et al,
1994; Grote et al, 1999; Metz-Lutz et al, 1999b; Plaza et al, 2001; Robinson et al, 2001;
Majerus et al, 2004). For example, Soprano et al. (1994) observed that the performances of
LKS patients on auditory-verbal STM tasks were more severely impaired than their
performances on other language processing tasks, several months after the onset of recovery
from LKS. Similarly, Grote et al. (1999) and Robinson et al. (2001) showed that LKS
patients with complete or nearly complete language recovery still presented deficits on tasks
involving STM processing, such as verbatim sentence recall and serial recall of letter and
digit sequences.
More precisely, in a recent case study of 3 young adult patients who recovered from LKS
7-10 years ago, Majerus et al. (2004) observed selective deficits on tasks requiring short-term
storage of phonological information. However, performance for short-term retention of
lexico-semantic information was preserved. For example, the patients' scores for immediate
serial recall of nonword sequences and for a rhyme probe1 task were severely impaired in two
patients, and more mildy impaired for the third patient, while performance on a category
probe task2 was completely preserved in all three patients. Although two of the patients also
presented with residual phonological processing deficits (as evidenced by impaired
performance in speeded single nonword repetition and phonological awareness tasks), the
severity of the phonological processing deficit was not related to the severity of their
phonological STM impairment.
PATHOPHYSIOLOGICAL BASIS OF EPILEPTIC APHASIA
A focal epileptic dysfunction in the temporal lobe
As early as 1957, when the first LKS case report was published, a direct relationship between
epilepsy and acquired aphasia was suggested. Indeed, on the basis of the EEG findings,
Landau and Kleffner assumed "that persistent convulsive discharges in brain tissue largely
concerned with linguistic communication result in the functional ablation of these areas for
normal linguistic behavior". The almost parallel fluctuation of verbal impairments and
epilepsy, more particularly with the EEG abnormalities observed in the follow-up studies,
supported this assumption (Dugas et ah, 1982; Hirsch et ah, 1990; Paquier et ah, 1992; Lanzi
et ah, 1994; Soprano et ah, 1994).
As structural neuroimaging with computed tomography (CT) and magnetic resonance
imaging (MRI) usually do not disclose structural brain lesions, several studies have looked at
possible metabolic abnormalities common to LKS.
Evidence from metabolic studies. Several studies using single photon emission computerised
tomography (SPECT) have shown abnormal perfusion in the temporal lobe which correlates
with the localisation of the epileptic focus (Mouridsen et ah, 1993; Harbord et ah, 1999).
Similarly, positron emission tomography (PET) studies using 2-deoxy-2-[18F]fluoro-D-
glucose (FDG) and carried out in approximately 29 LKS children consistently reported
metabolic abnormalities in the temporal lobes; these abnormalities were either a unilateral or
bilateral increase or decrease of glucose uptake (Maquet et ah, 1995; da Silva et ah, 1997). In
one study, glucose consumption was investigated with 18FDG-PET at different times during
the evolution of LKS in the same patients. This study showed that the metabolic changes
appeared to be correlated with the epileptic activity, characterised by an hypermetabolic focus
during the active phase of epileptic aphasia and a hypometabolic focus when EEG activity
In a rhyme probe task, word lists of increasing length are presented auditorily. At the end of each list, a new
word is presented and the participants are asked to determine whether this probe word rhymes with one of the
words in the list.
2
The category probe task is very similar to the rhyme probe task but probes specifically STM retention for
semantic information. Here, the patients are asked to determine whether the probe word presented at the end of
the list belongs to the same semantic category as one of the words in the list.
went back to normal (Maquet et al, 1995). Several improvements of PET data analyses,
including the use of statistical parametric mapping (SPM), the comparison of each patient to a
large control group and the co-registration of PET activity with a high resolution 3D-MRI
scan of individual brains, have allowed a better definition of the spatial extent of the
metabolic abnormalities. This has permitted to establish that, in the active period of epilepsy,
the focus of hypermetabolism coincides with the topography of the predominant spike wave
discharges recorded on waking and sleep EEG. Interestingly, these studies showed that the
significantly higher glucose uptake only involved the associative but not the primary auditory
cortex (Maquet et al., 1999).
Neurophysiological evidences. Although the EEG recordings show both focal discharges and
bilateral generalized spike-and-wave discharges, electrophysiological studies based on dipole
mapping of spike discharge and magnetic source imaging provided arguments in favour of a
focal source of epileptic discharges localised in the superior temporal gyms (Morrell, 1995).
Using magneto-encephalographic recording (MEG) of SWD, Paetau et al. (1991)
demonstrated that all spikes originated close to the auditory cortex, contaminating the
ipsilateral auditory cortex and suppressing the contralateral auditory evoked potentials.
Another study using auditory evoked magnetic field recordings (Paetau, 1994) in six children
with LKS, suggested that the epileptiform activity may be produced by sound-responsive
neurons in the non-primary auditory cortex within the middle and posterior perisylvian
cortex.
This set of findings is congruent with the suggestion that the apparently bilateral epileptic
discharges of CSWSS are "driven" by a focal and unilateral source of primary epileptogenic
activity in the superior temporal cortex. It also suggests that the bilateralisation of epileptic
discharges and their generalisation during sleep makes the homotopic temporal cortex in the
opposite hemisphere unavailable for auditory processing and more specifically for verbal
processing. This might account for the main features of epileptic aphasia, i.e. the aphasic
impairments that do not depend on the side of the temporal epileptic focus and the auditory
agnosia observed in the active phase for about all LKS cases reported in the literature.
An age-related focal epilepsy. The EEG pattern of SWDs and CSWSS is specific to age-
related idiopathic focal childhood epilepsy (IFCE). Indeed, SWDs are common to a wide
range of so-called benign partial childhood epilepsies including rolandic epilepsy, the most
frequent childhood epilepsy. SWDs and CSWSS are encountered in children aged 3-10 years
and disappear in early adolescence, whatever the effectiveness of anti-epileptic treatment.
Several studies showed that in IFCE, the epileptic foci are mainly located in the associative
cortex whose maturation continues throughout childhood and adolescence. In LKS, the age of
onset (70% before the age of 6) corresponds to the period of brain maturation that follows the
peak of synaptogenesis in the associative temporal cortex.
Considering the typical aspect of the epileptic discharges in LKS, it has been suggested
that following a brief spike, the slow-wave component might be the manifestation of an
inhibitory mechanism (Engel, 1995; 1996). Such a mechanism would not only prevent the
occurrence of seizures dependent on the spike component but also inhibit the normal
functioning of the cortical area involved in the generation of epileptic discharges. As the
inhibitory component of the epileptic activity is predominant over the other EEG components,
the epileptic aphasia would be the expression of an excess of inhibitory activity limited to a
part of the temporal cortex.
To account for this particular pattern of epileptic activity, Maquet et al. (1999) have
hypothesised a focal alteration of the maturational processes leading to an imperfect neuronal
wiring that induces an imbalance between inhibitory and excitatory drives generating
epileptic discharges. According to this hypothesis, the large slow-wave component indicating
a predominant inhibitory mechanism would result from a local excess of activity of the
inhibitory interneurones within the epileptic focus. At the same time, the high density of
synapses characteristic of this period of cortical maturation would facilitate the diffusion of
SWDs leading to the aspect of CSWSS. The benefit of multiple subpial transection in LKS
reinforced this hypothesis. Indeed, this particular surgical procedure consisting in the
selective disruption of the intracortical horizontal fibers was used to mechanically interrupt
the deleterious intracortical circuitry (Morrell et al, 1995). This method permitting the
resumption of more normal synaptisation suppresses the focal epileptic activity and allows
the recovery of the normal functioning of the temporal associative cortex.
PATHOPHYSIOLOGICAL BASIS OF VERBAL OUTCOME IN LKS
The residual verbal impairments following LKS have been explained in different ways.
Mantovani and Landau (1980) suggested that, like in the case of focal brain damage, the long-
term effect of the epileptiform discharges on brain cells of a cortical area results in a
functional hemispheric reorganisation with the shifting of language area in a cortical area
spared by the epileptic activity. Bishop (1985) assumed that the loss of auditory verbal
comprehension, during the active period of epileptic aphasia, deprives the child from the
communicative verbal experience crucial for language development. Finally, Baynes et al.
(1998) suggested that the nature of dysfunction and the outcome depend on the stage of
language development at which LKS children experienced the disruption of auditory input.
As regards the verbal outcome of epileptic aphasia, two points are at issue. Do the
residual verbal disorders following LKS relate to some specific impairment in verbal
processing? Indeed, a very consistent finding is a deficit in phonological short-term memory
(STM) performance, even in patients showing relatively good language recovery as will be
shown below ( Soprano et al, 1994; Grote et al, 1999; Metz-Lutz et al., 1999a, b; Plaza et
al, 2001; Robinson et al, 2001; Majerus et al, 2003, 2004). The second point deals with the
pathophysiological mechanisms underlying the residual verbal deficits. The data of PET
studies performed at the recovery period of epileptic aphasia showing a focal hypometabolism
in the temporal area initially involved in the epileptic focus suggest a persistent functional
impairment of this cortical area. This persistent functional cortical impairment could also
underlie the residual verbal deficits.
Evidence for persistent dysfunction in the superior temporal cortex and its relationship to
impaired phonological STM. One clinical finding consistently reported in all LKS follow-up
studies including dichotic listening tasks is the one-ear dichotic listening extinction involving
the ear contralateral to the temporal cortex affected by the epileptic discharges during the
active phase of epilepsy (Metz-Lutz et al, 1997; Plaza et al, 2001). A similar one-ear
dichotic extinction was described in patients who suffered structural lesion involving the
temporal or parieto-temporal cortex and the geniculo-cortical pathway (Kimura, 1961;
Damasio & Damasio, 1979). In LKS, this finding might be the expression of a permanent
dysfunction in the temporal cortex. This assumption is supported by the findings of PET scan
studies performed in the late recovery period in LKS patients, which disclosed a focal
hypometabolism in the superior temporal region contralateral to the dichotic extinction
(Maquet et al., 1999).
In order to test this hypothesis, we used auditory evoked potentials (AEP) enabling us to
check the whole auditory system along the auditory pathways to the temporal associative
cortex in five children who had recovered from LKS and showed a right or left dichotic
listening extinction. The five patients were compared to five control subjects matched for age
and gender. This study showed normal and symmetrical early and middle latency auditory
evoked potentials but significant alteration in the late Nib and N250 components with
reduced amplitude over the temporal sites contralateral to the dichotic listening extinction
(Wioland et al, 2001). The Nib and N250 components are known to arise from the temporal
associative cortex.
In a second study, we explored more specifically the integrity of temporal associative
cortex and its relationship to the residual phonological STM deficits that characterise the late
outcome of LKS, using PET imaging (Majerus et al, 2003). The three adult patients for
which we had identified relatively specific phonological STM impairments (see above;
Majerus et al, 2004) also participated in this second study. They were asked to repeat
auditorily presented 4-word sequences (STM condition) or single words (control condition)
while they underwent H2I5O-PET imaging. During the STM condition, we observed
decreased activation in the bilateral posterior superior temporal gyms and adjacent
perisylvian cortex in the two patients who presented the most severe phonological STM
impairments. In the third patient whose phonological STM impairment was much milder, we
observed increased activation in the right posterior superior temporal gyms during the STM
condition. These data show that the posterior temporal cortex remains dysfunctional in the
late outcome of LKS, relative to healthy controls, and is related to the persistent phonological
STM impairment.
Furthermore, we also observed a relationship between the regions that displayed abnormal
glucose metabolism in the three patients during the active phase of the LKS (Maquet et al.,
1995) and the recent brain activation pattern during the STM task. The first of the two
patients with the most important phonological STM deficits had shown decreased glucose
uptake in the left superior and middle temporal regions during the active phase of LKS as
well as reduced glucose uptake bilaterally in superior temporal regions several months later.
In the recent brain activation study, reduced activity in the right posterior superior temporal
gyrus was observed. During the active phase the second patient had shown increased glucose
metabolism in the right middle and superior temporal regions and decreased glucose
metabolism in the left perisylvian cortex. In the recent study, diminished activity bilaterally in
the posterior superior temporal cortex was observed. The third patient with milder
phonological STM deficits had shown a very focal increase in glucose metabolism at the level
of the right STG, which became hypometabolic several months later. In the recent study,
increased activation in the right posterior as well as slightly reduced activity in the anterior
part of the right midtemporal gyrus (although only at uncorrected P-levels) was observed. In
sum, the results suggest that late outcome of LKS may indeed be characterised by a long-
lasting dysfunction of mainly posterior and superior temporal gyri and adjacent perisylvian
areas which had also been dysfunctional during the active phase of epilepsy. Furthermore,
this persistent dysfunction of superior temporal gyri appears to be related to the residual
phonological STM impairments.
DISCUSSION
In this paper, we attempted to provide a provisional pathophysiological account of aphasic

disorders and the late verbal outcome of LKS, a particular condition where acquired aphasia
appears as the main clinical symptom of a focal epileptic activity. Most importantly, this
epileptic activity has to be explained by a predominant inhibitory rather than by an excitatory
mechanism. Indeed, the main symptoms of LKS as opposed to other childhood focal
epilepsies are not sudden and brief episodes of behavioural changes in the form of seizures,
but the rapidly progressive disappearance of initially normally developing language
processing, appearing like an interictal disorder. Indeed, aphasic disorders are not transitory
as would be the clinical expression of seizure. Furthermore, the propensity of SWDs to
diffuse to the contralateral homologous region, probably due to the stage of maturation of the
cortical tissue at which SWDs occur, prevents any possible compensation as long as the
epilepsy is active. Thus, we think that the local excess of inhibitory mechanism can be viewed
as a plausible pathophysiological account for both the "functional ablation" of verbal cortex
and the inability to develop compensatory mechanisms during the active phase of epilepsy.
This account suggests a more or less direct relationship between the (inhibitory) epileptic
activity itself and the verbal impairment during the active phase. At outcome, a similar
relationship is observed between residual phonological STM deficits and the persistent
dysfunction of the superior temporal cortex that was involved in the epileptic focus during the
active phase. However, this association between aphasia and epilepsy does not necessarily
imply that aphasia and phonological STM deficits are exclusively and directly caused by the
epileptic activity. Instead, we cannot exclude that the aphasic and epileptic symptoms during
the active phase, as well as the residual verbal and STM impairments at later outcome, are
both the result of a third variable. This third variable could represent the focal alteration of
maturational processes of the cortex leading, in the case of LKS, to abnormal neuronal wiring
in the superior temporal cortex, as suggested by Maquet et al. (1999). This abnormal wiring,
in turn, will lead to epileptic activity within this area as well as to depressed language
processing. The epileptic activity will then further aggravate the language impairments by
further inhibiting functioning within the temporal cortex, via the inhibitory mechanisms
discussed above. In addition, the absence of environmental verbal stimulation resulting from
the severe receptive auditory-verbal impairments will also contribute to language decline.
Although epileptic activity will disappear in adolescence, the wiring of superior temporal
cortex may remain abnormal and thus language and STM processing will still be processed in
a less efficient way. If the hypothesis of a focal alteration of the maturational processes is
correct, it raises the issue of what determines this focal alteration and its more frequent
localisation in the associative temporal cortex. This will be the challenge for future work on
LKS.
Whatever the pathophysiological considerations, one should keep in mind that during the
active period of LKS, brain maturation is still in process. This maturation depends on both an
internal genetically determined programme and external stimulation, notably perceptual-
motor experience. Regarding verbal development, the experience of verbal input and inter-
individual verbal communication is especially determinant for further language development.
This implies that the development of effective rehabilitation programmes, aiming at
preventing and reducing the effect of deprivation of normal auditory feedback on
phonological, syntactic, and lexical development, remains of utmost importance.
ACKNOWLEDGEMENTS
Steve Majerus is a Postdoctoral Researcher at the Fonds National de la Recherche

Scientifique, Belgium.
REFERENCES
Baynes, K., J. A. Kegl, D. Brentari, C. Kussmaul and H. Poizner (1998). Chronic auditory
agnosia following Landau-Kleffner Syndrome: A 23 year outcome study. Brain Lang,
63(3), 381-425.
Beaumanoir, A. (1992). The Landau-Kleffner Syndrome. In: Epileptic Syndromes in infancy,
childhood and adolescence (2nd ed) (J. Roger, M. Bureau, C. Dravet, F. E. Dreifuss,
A. Perret and P. Wolf, editors), pp.231-244 John Libbey and Company Ltd, London.
Bishop, D. V. M. (1985). Age of onset and outcome in "Acquired aphasia with convulsive
disorder" (Landau-Kleffner Syndrome). Dev Med Child Neurol, 27, 705-712.
da Silva, E. A., D. C. Chugani, O. Muzik and H. T. Chugani (1997). Landau-Kleffner
Syndrome: metabolic abnormalities in temporal lobe are a common feature. J Child
Neurol, 12, 489-495.

Damasio, H. and A. Damasio (1979). "Paradoxic" ear extinction in dichotic listening:
Possible anatomic significance. Neurology, 29, 644-653.
Dugas, M., M. Masson, M. F. Le Heuzey and N. Regnier (1982). Childhood acquired aphasia
with epilepsy (Landau-Kleffner Syndrome). 12 personal cases. Rev Neurol, 138, 755-
780.
Ege, B. and S. E. Mouridsen (1998). Linguistic development in a boy with the Landau-
Kleffner Syndrome: A five year follow-up study. Journal ofNeurolinguistics, 11, 321-
328.
Engel, J. J. (1995). Inhibitory mecanisms of epileptic seizure generation. In: Negative motor
phenomenom (S. Fahn, editor), Vol. 12, pp. 157-171. Lippincott Raven Publishers,
Philadelphia.
Engel, J. J. (1996). Excitation and inhibition in epilepsy. Can J Neurol Sci, 23, 167-174.
Grote, C. L., P. Van Slyke and J. A. Hoeppner (1999). Language outcome following multiple
subpial transection for Landau- Kleffner Syndrome. Brain, 122, 561-566.
Harbord, M. G., R. Singh and S. Morony (1999). SPECT abnormalities in Landau-Kleffner
Syndrome. J Clin Neurosci, 6, 9-16.
Hirsch, E., C. Marescaux, P. Maquet, M. Metz-Lutz, M. Kiesmann, E. Salmon, G. Franckand
D. Kurtz (1990). Landau-Kleffner Syndrome: A clinical and EEG study of five cases.
Epilepsia, 31, 756-767.
Kaga, M. (1999). Language disorders in Landau-Kleffner Syndrome. J Child Neurol, 14, 118-
122.
Kimura, D. (1961). Some effects of temporal lobe damage on auditory perception. Can J
Psychol, 15, 156-165.
Koeda, T. and Y. Kohno (1992). Non-verbal auditory agnosia with EEG abnormalities and
epilepsy; an unusual case of Landau-Kleffner Syndrome. No To Hattatsu, 24, 262-
267.
Korkman, M., M. L. Granstrom, K. Appelqvist and E. Liukkonen (1998). Neuropsychological
characteristics of five children with the Landau-Kleffner Syndrome: dissociation of
auditory and phonological discrimination. JInt Neuropsychol Soc, 4, 566-575.
Kramer, U., Y. Nevo, M. Y. Neufeld, A. Fatal, Y. Leitner and S. Harel (1998). Epidemiology
of epilepsy in childhood: a cohort of 440 consecutive patients. Pediatr Neurol, 18, 46-
50.
Landau, W. M. (1992). Landau-Kleffner Syndrome. An eponymic badge of ignorance. Arch
Neurol, 49, 353.
Lanzi, G., P. Veggiotti, S. Conte, E. Partesana and C. Resi (1994). A correlated fluctuation of
language and EEG abnormalities in a case of the Landau-Kleffner Syndrome. Brain
Dev, 16, 329-334.
Majerus, S., S. Laureys, F. Collette, G. Del Fiore, C. Degueldre, A. Luxen, M. Van der
Linden, P. Maquet and M. N. Metz-Lutz (2003a). Phonological short-term memory
networks following recovery from Landau and Kleffner Syndrome. Hum Brain Mapp,
19, 133-144.
Majerus, S., M. Van der Linden, M. Poncelet and M. N. Metz-Lutz (2004). Can phonological
and semantic short-term memory be dissociated? Further evidence from Landau-
Kleffner Syndrome. Cogn Neuropsychol, (in press).
course and prognosis. Neurology, 30, 524-529.
Maquet, P., E. Hirsch, M. N. Metz-Lutz, C. Marescaux and G., Franck. (1999). PET studies
of Landau-Kleffner Syndromes and related disorders. In: Childhood Epilepsy (A.
Nehlig, J. Motte, S. Moshe and P. Plouin, editors), pp. 135-141. John Libbey and
Company Ltd., London.
Maquet, P., E. Hirsch, M. N. Metz-Lutz, J. Motte, D. Dive, C. Marescaux and G. Franck
(1995). Regional cerebral glucose metabolism in children with deterioration of one or
more cognitive functions and continuous spike-and-wave discharges during sleep.
Brain, 118, 1492-1520.
Marescaux, C, E. Hirsch, S. Finck, P. Maquet, F. Schlumberger, F. Sellal, M. N. Metz-Lutz,
Y. Alembik, E. Salmon, G. Franck and D. Kurtz (1990). Landau-Kleffner Syndrome:
A Pharmacologic Study of five cases. Epilepsia, 31, 768-777.
Metz-Lutz, M. N., A. de Saint Martin, E. Hirsch, P. Maquet and C. Marescaux (1996).
Auditory verbal processing following Landau and Kleffner Syndrome. Brain Lang,
55, 147-150.
Metz-Lutz, M. N., A. de Saint Martin, E. Hirsch, P. Maquet and C. Marescaux (1999a).
Impairments in auditory verbal processing and dichotic listening after recovery of
epilepsy in Landau and kleffner Syndrome. Brain Cogn, 40, 193-197.
Metz-Lutz, M. N., E. Hirsch, P. Maquet, A. de Saint Martin, G. Rudolf, N. Wioland and C.
Marescaux (1997). Dichotic listening performances in the follow-up of Landau and
Kleffner Syndrome. Child Neuropsychol, 3, 47-60.
Metz-Lutz, M. N., C. Seegmuller, C. Kleitz, A. de Saint Martin, E. Hirsch and C. Marescaux
(1999b). Landau-Kleffner Syndrome: a rare childhood epileptic aphasia. Journal of
Neurolinguistics, 12, 167-179.
Morrell, F. (1995). Electrophysiology of CSWS in Landau-Kleffner Syndrome. In:
Continuous spikes and waves during slow sleep (A. Beaumanoir, M. Bureau, T.
Deonna, L. Mira and C. A. Tassinari, editors), pp. 77-90. John Libbey & Company
Ltd., London.
Morrell, F., W. W. Whisler, M. C. Smith, T. J. Hoeppner, L. de Toledo-Morrell, S. J. Pierre-
Louis, A. M. Kanner, J. M. Buelow, R. Ristanovic, D. Bergen, et al. (1995). Landau-
Kleffner Syndrome. Treatment with subpial intracortical transection. Brain, 118,
1529-1546.
Mouridsen, S. E., C. Videbaek, H. Sogaard and A. R. Andersen (1993). Regional cerebral
blood-flow measured by HMPAO and SPECT in a 5-year-old boy with Landau-
Kleffner Syndrome. Neuropediatrics, 24, 47-50.
Notoya, M., S. Suzuki, M. Furukawa and H. Enokido (1991). A case of pure word deafness
associated with Landau-Kleffner Syndrome: a long-term study of auditory

disturbance. Auris Nasus Larynx, 18, 297-305.
Paetau, R. (1994). Sounds triggers spikes in the Landau-Kleffner Syndrome. J Clin
Neurophysiol, 11, 231 -241.
Paetau, R., M. Kajola, M. Korkman, M. Hamalainen, M. L. Granstrom and H. Hari (1991).
Landau-Kleffner Syndrome: epileptic activity in the auditory cortex. Neuroreport, 2,
201-204.
Paquier, P. F., H. R. Van Dongen and C. B. Loonen (1992). The Landau-Kleffner Syndrome
or "acquired aphasia with convulsive disorder". Long-term follow up of six children
and a review of recent literature. Arch Neurol, 49, 354-359.
Perez, E. R., V. Davidoff, A. C. Prelaz, B. Morel, F. Rickli, M. N. Metz-Lutz, P. Boyes
Braem and T. Deonna (2001). Sign language in childhood epileptic aphasia (Landau-
Kleffner Syndrome). Dev Med Child Neurol, 43, 739-744.
Plaza, M., M. T. Rigoard, C. Chevriemuller, H. Cohen and A. Picard (2001). Short-term
memory impairment and unilateral dichotic listening extinction in a child with
Landau-Kleffner Syndrome: Auditory or phonological disorder? Brain Cogn, 46, 235-
240.
Robinson, R. O., G. Baird, G. Robinson and E. Simonoff (2001). Landau-Kleffner Syndrome:
course and correlates with outcome. Dev Med Child Neurol, 43, 243-247.
Sieratzki, J. S., G. A. Calvert, M. Brammer, A. David and B. Woll (2001). Accessibility of
spoken, written, and sign language in Landau-Kleffner Syndrome: a linguistic and
functional MRI study. Epileptic Disord, 3, 79-89.
Soprano, A. M., E. F. Garcia, R. Caraballo and N. Fejerman (1994). Acquired epileptic
aphasia: neuropsychologic follow-up of 12 patients. Pediatr Neurol, 11, 230-235.
Tutuncuoglu, S., G. Serdaroglu and B. Kadioglu (2002). Landau-Kleffner Syndrome
beginning with stuttering: case report. J Child Neurol, 17, 785-788.
Vance, M., S. Dry and S. Rosen (1999). Auditory processing deficits in a teenager with
Landau-Kleffner Syndrome. Neurocase, 5, 545-554.
Wioland, N., G. Rudolf and M. N. Metz-Lutz (2001). Electrophysiological evidence of
persisting unilateral auditory cortex dysfunction in the late outcome of Landau and
Kleffner Syndrome. Clin Neurophysiol, 112, 319-323.
Zardini, G., B. Molterri, N. Nardocci, D. Sarti, G. Avanzini and T. Granata (1995). Linguistic
development in a patient with Landau-kleffner Syndrome: A nine year follow-up.
Neuropediatrics, 26, 19-25.
Acquired Epileptiform Aphasia 25
ACQUIRED LANGUAGE DISORDERS AND

EPILEPSY: FROM LANDAU-KLEFFNER
SYNDROME TO AUTISTIC REGRESSION
Roberto Tuchman
Dan Marino Center and Miami Children's Hospital, Florida, USA
Abstract — The relationship of epilepsy, either clinical or subclinical, to the acquired aphasia or the
loss of communicative intent that occurs in some children, most of whom are on the autistic spectrum,
remains controversial and not understood. The report of loss of language in any child should raise
concern and there are several clinical syndromes to consider when discussing the Acquired
Epileptiform Aphasias (AEA). These include: Acquired aphasia with convulsive disorder or Landau-
Kleffner Syndrome (LKS); Continuous Spike-Waves during Slow Sleep (CSWS) associated with
Electrical Status Epilepticus during Slow-Wave Sleep (ESES); Benign Childhood Epilepsy with
Centrotemporal Spikes (BECTS), and Autistic Regression with an Epileptiform EEG (AREE). The
lack of strict criteria and the overlap of disorders with language loss associated with epilepsy or an
epileptiform EEG have made classifying children who undergo a regression of their language in the
context of either clinical or subclinical seizures a confusing and controversial undertaking. An
important first step in our quest to understand the pathophysiology of the acquired epileptiform
aphasias and to develop rational interventions is to rigorously define these syndromes at a clinical
level. The purpose of this discussion is to review the nosology of acquired language disorders
associated with epilepsy.
Key words: language disorders, regression, epilepsy, autism, acquired aphasias, Landau-
Kleffner Syndrome (LKS); Continuous Spike-Waves during Slow Sleep (CSWS), Electrical
Status Epilepticus during Slow-Wave Sleep
INTRODUCTION
The classification of acquired language disorders in children and their relationship to epilepsy
is a complex clinical problem. The lack of strict criteria and the overlap of disorders with
language loss associated with epilepsy or an epileptiform EEG have made classifying children
who undergo a regression of their language in the context of either clinical or subclinical
seizures a confusing and controversial undertaking. The term I prefer to use for the above
disorders is Acquired Epileptiform Aphasias, as opposed to Acquired Epileptic Aphasias,
since epilepsy, defined as more than one unprovoked seizure, is not always present in this
group of encephalopathies.
The report of loss of language in any child should raise concern. Language regression was
considered rare until recent reports have indicated that up to one-third of all children with
autism regress in language or in communicative intent (Tuchman & Rapin, 1997). Although it
is most often parental awareness of regression or stagnation of expressive language in their
toddler or preschooler that brings the child to professional attention, it may be that there are
premonitory signs. Some of these are the inability to coordinate attention and share the
enjoyment of an event with a social partner. One important hallmark of social communication
in early development is joint attention. Joint attention skills refer to the capacity of individuals
to coordinate attention with a social partner vis-a-vis some object or event. This capacity,
which is an essential precursor to verbal communication, begins to emerge by 6 months of age
and takes on several different forms, each of which may be measured reliably in infants and
young children. The social-communication disturbance of autism is exemplified by a striking
failure to develop adequate joint attention skills or by the loss of these skills once developed.
Research indicates that joint attention impairment is characteristic of autism. There is also
evidence to suggest that joint attention may predict language, cognitive and social outcomes
in children on the autistic spectrum, and that it may be an index of the neurodevelopmental
components of the disorder. Consequently, joint attention has become an important dimension
to consider in early diagnosis and treatment research in autism and related disorders.
There are several clinical syndromes to consider when discussing the Acquired
Epileptiform Aphasias (AEA). These include: Acquired aphasia with convulsive disorder or
Landau-Kleffner Syndrome (LKS); Continuous Spike-Waves during Slow Sleep (CSWS)
associated with Electrical Status Epilepticus during Slow-Wave Sleep (ESES); Benign
Childhood Epilepsy with Centrotemporal Spikes (BECTS), and Autistic Regression with an
Epileptiform EEG (AREE). All of these clinical syndromes are associated with different
degrees of cognitive function or mental retardation (MR) and the language disorder most
likely to be present in the majority of the acquired epileptiform aphasias is that of Verbal
Auditory Agnosia (VAA). The purpose of this paper is to define, discuss and contrast this
group of encephalopathies.
ACQUIRED APHASIA WITH CONVULSIVE DISORDER OR LANDAU-

KLEFFNER SYNDROME (LKS)
Acquired epileptic aphasias have been used as an important model for expanding the concept
of childhood epilepsy to include prolonged language, cognitive and behavioral disorders as
primary epileptic manifestations (Deonna, 1991). The prototype of AEA is Landau-Kleffner
syndrome. Landau-Kleffner syndrome is defined as an acquired aphasia in association with
abnormal EEG demonstrating spikes, sharp waves or spike and wave discharges which are
usually bilateral and occur predominantly over the temporal and parietal regions (Rapin,
1995a). The definition of LKS has been widely expanded and behavioral problems such as
hyperkinesis, rage outbursts, aggressiveness, stereotypies and poor social communication
skills in children with language regression and associated epilepsy or with an epileptiform
EEG have been included under this eponym. Central to the original description of LKS in
1957 is language regression in association with an epileptiform EEG and either seizures or
acquired aphasia are equally likely to be the first presenting complaint in this disorder
(Landau & Kleffner, 1957). In LKS it is not the clinical epilepsy that is important in
producing the language manifestations of LKS, but the "subclinical seizures" as indexed by
epileptiform activity on the EEG as up to 25% of these patients may not have clinical seizures
(Tuchman & Rapin, 2002). A rigorous definition of LKS should include only those children
with primarily a regression in language and in whom the associated behavioral problems that
occur are secondary to the language disorder and not due to a primary behavioral or cognitive
regression (Tuchman, 1997).
The primary language disorder in the majority of children with LKS is a severe receptive
disorder amounting to verbal auditory agnosia (VAA) (Korkman et al., 1998; Klein et al.
2000). VAA is a receptive aphasia or dysphasia for acoustically, but not visually, presented
language and arises from inadequate auditory or phonologic processing that engages activity
in the primary or secondary auditory cortices and affecting output (expressive) language as
well (Majerus et al, 2003). A central theme of any discussion of acquired epileptiform
aphasia is the relationship of the abnormal electrical activity to the language disorder. The
cortical areas responsible for VAA are the same regions where the centrotemporal
epileptiform EEG activity characteristic of LKS and other AEA discussed below is found.
Furthermore, VAA is associated with the highest rate of epilepsy, both among children with
autistic spectrum disorders and those with developmental language disorders (Tuchman et al.,
1991a).
Age of symptoms may be important in differentiation of LKS from autistic regression
with an epileptiform EGG (AREE) and possibly from other acquired epileptiform aphasias.
The mean age of onset of the language regression in autistic spectrum disorders is 21 months
and over 90% of children with autism who undergo a regression do so before age 3 years
(Tuchman & Rapin, 1997). This is different from LKS where it is reported that mean age of
onset is after age 4 years (Robinson et al, 2001). The prognosis in LKS for recovery of
language and cognitive function is variable and in general not as good as it is for the seizures
(Bishop, 1985; Mouridsen, 1995).
ELECTRICAL STATUS EPILEPTICUS DURING SLEEP (ESES): CONTINUOUS

SPIKE-WAVES DURING SLOW SLEEP (CSWS) AND BENIGN CHILDHOOD
EPILEPSY WITH VENTROTEMPORAL SPIKES (BECTS)
Electrical status epilepticus during slow wave sleep (ESES) is considered an EEG-defined
syndrome characterized by nearly continuous (>85%) spike-and-slow-wave-complexes during
non-REM sleep, but not during REM sleep or the awake state. The clinical syndrome that has
been mostly associated with ESES is continuous spike-waves during slow sleep (CSWS)
(Veggiotti et al, 1999). The majority of children with CSWS have normal development prior
to the onset of ESES, but almost all deteriorate cognitively and behaviorally (increased
aggressiveness, poor social contact, decreased attention span, and hyperactivity) during ESES
(Roulet-Perez, 1993; Nieto-Barrera et al, 1997; Veggiotti et al, 2001). Language
deterioration out of proportion to other abilities occurs in some cases. At a behavioral level
the major difference between reported cases of LKS and CSWS is that children with CSWS
show more diffuse cognitive and behavioral deterioration with dementia and behaviors
consistent with autism than do children with LKS (Galanopoulou et al, 2000). Seizures are
common in CSWS, but as in LKS not always present. LKS and CSWS syndromes may be on
a continuum and as such some investigators have suggested that Landau-Kleffner may be too
narrowly defined (Hirsch et al, 2000). Nevertheless, I would suggest that strictly defining
these syndromes will be more productive in determining the relationship of acquired language
regression to clinical and subclinical epilepsy.
The electroencephalographic findings in LKS resemble the EEG findings in benign
childhood epilepsy with centrotemporal spikes (BECTS) both in morphology and distribution
and in both of these disorders the spikes are activated by sleep (Saint-Martin et al, 2001). At
a clinical level these disorders are very different and in general children with BECTS do not
have the associated clinically significant language or cognitive dysfunction found in LKS or
in CSWS. However, recent work suggests that there is a subgroup of children with BECTS
where the evolution of the spikes takes on a more atypical nature and the clinical evolution of
these children is more similar to that of children with LKS or CSWS (Fejerman et al, 2000).
There is also data suggesting that differences in morphology, topography, organization, and
abundance of interictal abnormalities during sleep can differentiate BECTS from LKS early
on and prior to the loss of language occurring in LKS (Massa et al, 2000). The outcome of
some children with BECTS with atypical evolution of their spikes on EEG and with
regression in language and behavior may be similar to those with LKS and CSWS (Yung et
al, 2000). Recent work using magnetoencephalography has suggested that the location and
possibly orientation of the spike might account for differences in clinical presentation of
disorders such as BECTS, LKS and AREE (Lewine et al, 1999; Sobel et al, 2000; Otsubo &
Snead, 2001).
AUTISTIC REGRESSION WITH AN EPILEPTIFORM EEG (AREE)
Autism is a life-long disorder with clinical symptoms that change with age and often improve
with early intervention. Approximately one-third of parents report a regression of language,
usually the loss of their toddler's first few words between 18 and 24 months, together with the
appearance of autistic behaviors. Autistic regression, including regression in language may
fluctuate for many months or even years and then improve, although rarely to complete
recovery (Wilson et al, 2003). It is most often parental awareness of regression or stagnation
of expressive language in their toddler or preschooler that brings the child to professional
attention. Early on in development children with autism display a syndrome-specific inability
to coordinate attention and share the enjoyment of an event with a social partner. Impairments
in this domain may be assessed with measures of joint attention skills (Mundy et al, 1990).
Since autistic regression occurs prior to age 2 years and as such may be associated with a the
loss of only a few single words identifying loss of joint-attention skills may be a better early
indicator of regression in children with autism.
Autistic regression with or without an epileptiform EEG should be differentiated from
children classified as having disintegrative disorder (DD) in whom language and behavioral
regression is delayed and may occur as late as age 10 years (Rapin, 1995b). Children with
disintegrative disorder, sometimes referred to as Heller's syndrome, are more likely to have
epilepsy than children with autism (Burd et al, 1989). This group of children differs from
children with LKS in the severity of their cognitive and behavioral manifestations and from
the much more numerous children with typical autistic regression in two main ways: (1) the
regression occurs later, usually between 3 and 6 years, prior to which development was
entirely normal, in contrast to children with autism in whom the mean age at regression is 18-
24 months and earlier development already worrisome in some cases; and (2) the regression is
even more profound and may leave these children frankly and permanently demented
(Malhotra & Gupta, 1999; Dawson, 2000). The difference between disintegrative disorder and
autism with regression is not crisp and what role epilepsy may play in the genesis of either or
both is disputed.
The relationship of epilepsy, either clinical or subclinical, to the acquired aphasia or the
loss of communicative intent that occurs in autistic regression remains controversial and not
understood. In a study of 314 children with autism and 237 children with language disorders
examined by one child neurologist, parents of 32% of the autistic and 4.6% of the language
disordered groups reported a regression (Tuchman et al, 1991a). Epilepsy (at least 2
unprovoked seizures) was correlated with motor and cognitive indices of the severity of the
underlying brain dysfunction (Tuchman et al, 1991b). In an independent sample of 585
consecutive children with autistic spectrum disorders seen by another child neurologist,
regression was stated to have occurred before age 2 years in 64% of cases and by age 3 years
in 95% of cases (Tuchman & Rapin, 1997). Epilepsy was no more frequent (12%) in children
who had regressed than in those without a history of regression (11%). An interesting but
unexplained observation was that regression was significantly associated with an epileptiform
EEG in the non-epileptic group, in that 14% of 155 non-epileptic children who had undergone
a regression had an epileptiform EEG during sleep, as opposed to 6% of 364 children with
neither regression nor epilepsy. There was no difference in the proportion of children with
epilepsy or epileptiform EEGs who had regressed before or after 2 years of age. On the other
hand, a recent multi-institutional study of 177 children with a history of language regression
found that children with a history of regression prior to age 36 months were more likely to
have autism than those who regressed after age 36 months, and that children with regression
after age 36 months were more likely to have epilepsy than those with an earlier regression
(Shinnar et at, 2001). Although all three studies found a mean age at regression between 21
to 22.8 months, mean age at referral to the specialist was uniformly over age 36 months.
The studies just reviewed suggest that there are two groups of children who experience
regression: somewhat older children who are more likely to have epilepsy but rarely
experience a serious behavioral-autistic regression (the few who do are likely to suffer from
the catastrophic disintegrative disorder phenotype), and a younger much larger group in
whom epilepsy is less frequent but in whom language regression has a high probability of
being associated with a serious behavioral/cognitive deterioration leading to autism. These
studies also suggest that there may be neurophysiological markers of regression such as
epileptiform discharges; however, their contribution to the autistic regression in this younger
group is not known for two reasons: (1) a minority of children without clinical seizures
undergo EEG recordings, let alone all night monitoring, and (2) the children are rarely seen at
the time of the regression (the mean interval is measured in years, not months). The main
cause for this long delay is that very early regression is regularly passed off because it is
insidious and occurs so early in the course of language development. This early regression is
not given the same importance as language regression in a fully verbal older child. This lack
of early recognition may prevent early intensive intervention. Future studies will have to
address this important concern.
MEDICAL AND SURGICAL MANAGEMENT OF CHILDREN WITH ACQUIRED

EPILEPTIFORM APHASIAS
Data regarding response to medication in well-defined subgroups of children with acquired

epileptiform aphasias are very limited. The treatment strategies that have been reported for
this group of children are those used for the management of children with LKS. Therapy in
LKS has been the subject of numerous case reports or short series and the lack of well-
controlled clinical trials has been frustrating to the clinician faced with a child with an
acquired aphasia thought to be secondary to the clinical or subclinical epilepsy.
Anticonvulsants, especially valproate, ethosuximide and the benzodiazepines have been

reported to improve the language of a limited numbers of children with LKS (Marescaux et
al, 1990; Lerman et al, 1991). The use of ACTH, steroids, or immunoglobulins has also been
the subject of several clinical reports which suggest improvements in language and behavior
in children with LKS treated with these medications (Prasad et al, 1996; Lagae et al., 1998;
Mikati et al, 1998; Tsuru et al, 2000). Several clinical reports of the use of Valproate in
children with autism with or without clinical seizures but with epileptiform abnormalities on
the EEG suggest an improvement in language and behavior in this group of children with
AREE (Nass & Petrucha, 1990; Plioplys, 1994; Childs & Blair, 1997; Hollander et al, 2001;
Holmes & Riviello, 2001).
In a selected group of children with LKS surgical transection of epileptogenic fronto-
temporal cortex has been performed and reported to produce short-term improvement in
language and behavior in perhaps half of the children (Morrell et al, 1995; Sawhney et al,
1995). Two studies of children with autistic regression and clinical seizures state that
aggressive treatment of the seizures with epilepsy surgery was associated with positive
outcomes (Neville et al, 1997; Nass et al., 1999). One study suggested that, in children with
autism and intractable seizures, surgery may improve the seizures but not the autism (Szabo et
al, 1999). The children in these case reports had intractable epilepsy and the epilepsy surgery
was being done for the treatment of the seizures and not for the behavior or language
dysfunction. A controversial study suggested that multiple subpial transections in children
with autism spectrum disorders, a history of language regression, multifocal epileptiform
EEGs, and possible minor clinical seizures (i.e. staring episodes, rapid eye blinking) without
overt clinical seizures improved in language and behavior after surgery (Lewine et al, 1999).
It is important to state that the role of surgery in children with LKS and especially in those
with AREE is only recommended for the treatment of the seizures and not for the concomitant
language or behavioral deficits.
CONCLUSION
From a clinical perspective all children with stagnation or regression in communication skills
should be promptly referred to a specialist that can evaluate them from a neurological
perspective. It is extremely critical that speech and language intervention is promptly begun in
all such children and that all forms of verbal and non-verbal communication systems are
taught. An EEG with adequate sleep should be part of the work-up of any child with a history
of language regression. Careful monitoring of the language progression should be carried out
and in those children who do not progress after communication intervention is begun an
overnight EEG with good amount of sleep recording should be obtained. If the EEG
demonstrates epileptiform activity consideration should be given to the use of antiepileptic
medications.
An important first step in our quest to understand the pathophysiology of the acquired
epileptiform aphasias and to develop rational interventions is to rigorously define these
syndromes at a clinical level. Progress has been made in our understanding of the clinical
differences and overlaps between Landau-Kleffner Syndrome and Autistic Regression with an
Epileptiform EEG. In the process we have also gained a deeper understanding of the role of
epilepsy, both clinical and subclinical, in all epileptic disorders in children with associated
acquired aphasia. The use of newer imaging and EEG techniques such as
magnetoencephalography is enhancing our ability to determine the role not only of the
quantity of EEG discharges but also of the importance of understanding how the topography
of the EEG may determine specific symptoms such as language regression. Clinical studies
suggest that timing of the seizures or the development of EEG discharges may also help in the
differentiation of the acquired epileptiform aphasias.
Acquired epileptic aphasias associated with epilepsy or epileptiform abnormalities are not
specific entities. The studies reviewed here suggest that they represent part of a spectrum
disorder with a common pathophysiology with varying severity of clinical manifestations
dependent on the location and quantity of the epileptiform activity. The differences in clinical
symptoms and in their relationship to epilepsy or EEG changes in these disorders may, in
selected cases, be due only to the time period in development when the seizures occur or to
the site or the amount of cortical or subcortical epileptogenic dysfunction. However, the
seizures and the EEG findings do not always correlate with the clinical picture and as such the
EEG and the seizures may be only epiphenomena that provides for the identification of a
diverse group of language-EEG-epileptic encephalopathies with diverse etiologies.
REFERENCES
Bishop, D. V. (1985). Age of onset and outcome in 'acquired aphasia with convulsive
disorder' (Landau-Kleffner syndrome). Dev Med ChildNeurol, 27,705-12.
Burd, L., W. Fisher and J. Kerbeshian (1989). Pervasive disintegrative disorder: Are Rett
syndrome and Heller dementia infantilis subtypes? Dev Med Child Neurol, 31, 609-
616.
Childs, J. A. and J. L. Blair (1997). Valproic acid treatment of epilepsy in autistic twins. J
Neurosci Nurs, 29, 244-8.
Dawson, G. (2000). What is Childhood Disintegrative Disorder, how is it different from
autism, and what is believed to be its cause? J Autism Dev Disord, 30,177.
Deonna, T. W. (1991). Acquired epileptiform aphasia in children (Landau-Kleffner
syndrome). J Clin Neurophysiol, 8, 288-98.
Fejerman, N., R. Caraballo and S. N. Tenembaum (2000). Atypical evolutions of benign
localization-related epilepsies in children: are they predictable? Epilepsia, 41, 380-90.
Galanopoulou, A. S., A. Bojko, F. Lado and S. L. Moshe (2000). The spectrum of

neuropsychiatric abnormalities associated with electrical status epilepticus in sleep.
Brain Dev, 22, 279-95.
Hirsch, E., C. Marescaux, P. Maquet, M. N. Metz-Lutz, M. Kiesmann, E. Salmon, et al.
(1990). Landau-Kleffner syndrome: a clinical and EEG study of five cases. Epilepsia,
31, 756-67.
Hollander, E., R. Dolgoff-Kaspar, C. Cartwright, R. Rawitt, S. Novotny (2001). An open trial
of divalproex sodium in autism spectrum disorders. J Clin Psychiatry, 62, 530-4.
Holmes, G. L. and J. J. Riviello (2001). Treatment of childhood idiopathic language
deterioration with valproate. Epilepsy Behav, 2, 272-276.
Klein, S. K., R. F. Tuchman and I. Rapin (2000). The influence of premorbid language skills
and behavior on language recovery in children with verbal auditory agnosia. J Child
Neurol, 15, 36-43.
Korkman, M, M. L. Granstrom, K. Appelqvist and E. Liukkonen (1998). Neuropsychological
characteristics of five children with the Landau-Kleffner syndrome: dissociation of
auditory and phonological discrimination. JInt Neuropsychol Soc, 4 566-75.
Lagae, L. G., J. Silberstein, P. L. Gillis and P. J. Casaer (1998). Successful use of intravenous
immunoglobulins in Landau-Kleffner syndrome. Pediatr Neurol, 18,165-8.
disorder in children. 1957. Neurology, 51, 1241-1249.
Lerman, P., T. Lerman-Sagie and S. Kivity (1991). Effect of early corticosteroid therapy for
Landau-Kleffner syndrome. Dev Med Child Neurol, 33, 257-60.
Lewine, J. D., R. Andrews, M. Chez, A. A. Patil, O. Devinsky, M. Smith, et al. (1999).
Magnetoencephalographic patterns of epileptiform activity in children with regressive
autism spectrum disorders. Pediatrics, 104, 405-18.
Majerus, S., S. Laureys, F. Collette, G. Del Fiore, C. Degueldre, A. Luxen, et al. (2003).
Phonological short-term memory networks following recovery from Landau and
Kleffner syndrome. Hum Brain Mapp, 19, 133-44.
Malhotra, S. and N. Gupta (1999). Childhood disintegrative disorder. J Autism Dev Disord,
29,491-8.
Marescaux, C, E. Hirsch, S. Finck, P. Maquet, E. Schlumberger, F. Sellal, et al (1990).
Landau-Kleffner syndrome: a pharmacologic study of five cases. Epilepsia, 31, 768-
77.
Massa, R., A. de Saint-Martin, E. Hirsch, C. Marescaux, J. Motte, C. Seegmuller, et al.
(2000). Landau-Kleffner syndrome: sleep EEG characteristics at onset. Clin
Neurophysiol, 111, S87-93.
Mikati, M., M. Fayad and R. Choueri (1998). IVIG in Landau-Kleffner syndrome. Pediatr
Neurol, 19, 399-400.
Morrell, F., W. W. Whisler, M. C. Smith, T. J. Hoeppner, L. de Toledo-Morrell, S. J. Pierre-
Louis, et al. (1995). Landau-Kleffner syndrome. Treatment with subpial intracortical
transection. Brain, 118, 1529-46.
Mouridsen, S. E. (1995). The Landau-Kleffner Syndrome: a review. Eur Child Adolesc

Psychiatry, 4, 223-8.
Mundy, P., M. Sigman and C. Kasari (1990). A longitudinal study of joint attention and
language development in autistic children. J Autism Dev Disord, 20, 115-28.
Nass, R. and D. Petrucha (1990). Acquired aphasia with convulsive disorder: a pervasive
developmental disorder variant. J ChildNeurol, 5, 327-8.
Nass, R., A. Gross, J. Wisoff and O. Devinsky (1999). Outcome of multiple subpial
transections for autistic epileptiform regression. Pediatr Neurol, 21, 464-70.
Neville, B. G., W. F. Harkness, J. H. Cross, H. C. Cass, V. C. Burch, J. A. Lees, et al. (1997).
Surgical treatment of severe autistic regression in childhood epilepsy. Pediatr Neurol,
16, 137-40.
Nieto-Barrera, M., F. Aguilar-Quero, E. Montes, R. Candau and P. Prieto (1997). Epileptic
syndromes which show continuous spike and wake complexes during slow wave
sleep. Rev Neurol, 25, 1045-51.
Otsubo, H., Snead OC, 3rd. Magnetoencephalography and magnetic source imaging in
children. J Child Neurol 2001;16(4):227-35.
Plioplys, A. V. (1994). Autism: electroencephalogram abnormalities and clinical
improvement with valproic acid. Arch Pediatr Adolesc Med, 148, 220-2.
Prasad, A. N., C. F. Stafstrom and G. L. Holmes (1996). Alternative epilepsy therapies: the
ketogenic diet, immunoglobulins, and steroids. Epilepsia , 37, S81-95.
Rapin, I. (1995a). Acquired aphasia in children. J Child Neurol, 10, 267-70.
Rapin, I. (1995b). Autistic regression and disintegrative disorder: how important the role of
epilepsy? Semin Pediatr Neurol, 2, 278-85.
Robinson, R. O., G. Baird, G. Robinson and E. Simonoff (2001). Landau-Kleffner syndrome:
course and correlates with outcome. Dev Med Child Neurol, 43, 243-7.
Roulet-Perez E., V. Davidoff, P. A. Despland and T. Deonna (1993). Mental and behavioural
deterioration of children with epilepsy and CSWS: acquired epileptic frontal
syndrome. Dev Med Child Neurol, 35, 661-74.
Saint-Martin, A. D., R. Carcangiu, A. Arzimanoglou, R. Massa, P. Thomas, J. Motte, et al.
(2001). Semiology of typical and atypical Rolandic epilepsy: a video-EEG analysis.
Epileptic Disord, 3, 173-82.
Sawhney, I. M., I. J. Robertson, C. E. Polkey, C. D. Binnie and R. D. (1995). Multiple subpial
transection: a review of 21 cases. J Neurol Neurosurg Psychiatry, 58, 344-9.
Shinnar, S., I. Rapin, S. Arnold, R. F. Tuchman, L. Shulman, K. Ballaban-Gil, et al. (2001).
Language regression in childhood. Pediatr Neurol, 24, 183-9.
Sobel, D. F., M. Aung, H. Otsubo and M. C. Smith (2000). Magnetoencephalography in
children with Landau-Kleffner syndrome and acquired epileptic aphasia. AJNR Am J
Neuroradiol, 21, 301-7.
Szabo, C. A., E. Wyllie, M. Dolske, L. D. Stanford, P. Kotagal and Y. G. Comair (1999).
Epilepsy surgery in children with pervasive developmental disorder. Pediatr Neurol,
20, 349-53.
Tsuru, T., M. Mori, M. Mizuguchi and M. Y. Momoi (2000). Effects of high-dose intravenous
corticosteroid therapy in Landau-Kleffner syndrome. Pediatr Neurol, 22, 145-7.
Tuchman, R. F. (1997). Acquired epileptiform aphasia. Semin Pediatr Neurol, 4, 93-101.
Tuchman, R. F. and I. Rapin (1997). Regression in pervasive developmental disorders:
seizures and epileptiform electroencephalogram correlates. Pediatrics, 99, 560-6.
Tuchman, R. F. and I. Rapin (2002). Epilepsy in autism. Lancet Neurol, 1, 352-8.
Tuchman, R. F., I. Rapin and S. Shinnar (1991a). Autistic and dysphasic children. I: Clinical
characteristics. Pediatrics, 88, 1211-8.
Tuchman, R. F., I. Rapin and S. Shinnar (1991b). Autistic and dysphasic children. II:
Epilepsy. Pediatrics, 88,1219-25.
Veggiotti, P., F. Beccaria, R. Guerrini, G. Capovilla and G. Lanzi (1999). Continuous spike-
and-wave activity during slow-wave sleep: syndrome or EEG pattern? Epilepsia, 40,
1593-601.
Veggiotti, P., S. Bova, E. Granocchio, G. Papalia, C. Termine and G. Lanzi (2001). Acquired
epileptic frontal syndrome as long-term outcome in two children with CSWS.
Neurophysiol Clin, 31, 387-97.
Wilson, S., A. Djukic, S. Shinnar, C. Dharmani and I. Rapin (2003). Clinical characteristics of
language regression in children. Dev Med Child Neurol, 45, 508-14.
Yung, A. W., Y. D. Park, M. J. Cohen and T. N. Garrison (2000). Cognitive and behavioral
problems in children with centrotemporal spikes. Pediatr Neurol, 23, 391-5.
Persistent Language and Learning Deficits in AEOS 37
PERSISTENT SUBTLE LANGUAGE AND

LEARNING DEFICITS IN A CHILD WITH
ACQUIRED EPILEPTIFORM OPERCULAR
SYNDROME (AEOS)
Paola Cipriani, Anna M. Chilosi,Claudia Casalini, Lucia Pfanner, Annarita Ferrari, Daniela
Brizzolara and Renzo Guerrini
"Stella Maris " Scientific Institute, Pisa, Italy

University of Pisa, Italy
Abstract — Persistent linguistic deficits are sometimes observed after prolonged anarthric status
epilepticus, even in the absence of structural abnormalities of the perisylvian cortex, leading some
authors to interpret Acquired Epileptiform Opercular Syndrome (AEOS) as an expressive variant of
Landau-Kleffner syndrome. The long-term outcome of children with AEOS is rarely reported, and no
systematic neurolinguistic studies have been carried out with the aim of analysing the effects of
abnormal articulatory experience on phonological coding in the course of language development. We
report on a child who was followed at our department from age 5 years to age 8 years for a fluctuating
clinical syndrome consisting of recurrent episodes of severe oral motor dysfunction, dysarthria and
drooling paralleled with focal EEG abnormalities that fluctuated in phase with the clinical disorder. At
follow-up, persistent subtle language and learning difficulties due to impaired phonological processing
skills were observed, in spite of a good response to antiepileptic drugs, improved EEG, normal MRI
and adequate nonverbal cognitive abilities.
Keywords: Operculum syndrome, epilepsy, acquired speech and language dysfunction,

phonological processing deficits.
INTRODUCTION
Opercular Syndrome (OS), also called Foix-Chavanny Marie Syndrome (FCMS) (Foix et al.,
1926) results from a structural or functional abnormality in the opercular or perisylvian areas
and is clinically manifested by anarthria/severe dysarthria and loss of voluntary muscular
functions of the face and tongue as well as impaired mastication and swallowing. These
symptoms result from a central disturbance of the volitional control of the facio-linguo-
glosso-pharingeal-masticatory muscles, with preserved automatic, involuntary and emotional
innervation (automatic-voluntary dissociation).
It has a variable etiology, which may be due to congenital or acquired bilateral damage of
the perisylvian cortex (Christen et al, 2000; Salas-Puig et al, 2000; Gordon, 2002). Rarely
does OS have an epileptic origin (Fejerman & Di Blasi, 1987; Roulet et al, 1989; Colamaria
et al, 1991; Deonna et al, 1993; Shafrir & Prensky, 1995; De Saint-Martin et al, 1999;
Galanopoulou et al, 2000; Shuper et al, 2000; Kramer et al, 2001; Tachikawa et al, 2001).
In these cases, the underlying mechanisms are not fully understood as there is no clear link
between the epileptic activity and the clinical manifestations.
The term "Acquired Epileptiform Opercular Syndrome" (AEOS) was first used by Shafrir
and Prensky in 1995 to describe a 5-year-old girl who developed recurrent prolonged episodes
of suprabulbar palsy in association with continuous spike-and-wave activity during slow
sleep. These authors interpreted the AEOS as an expressive variant of Landau-Kleffner
syndrome (LKS). AEOS and LKS would represent neurological disorders, sharing similar
pathophysiological mechanisms, in which long-standing electrical dysfunction of perisylvian
neurons translates into bilateral neurological dysfunction.
AEOS manifests with fluctuating signs of suprabulbar palsy resulting from a disruption of
the connections between the cortical motor areas and the brainstem nuclei. The symptoms
(identical to those attributed to bilateral structural abnormalities of the perisylvian cortex)
show an intermittent course, with widely variable onset, duration and relapse. Persistent
linguistic deficits are sometimes described (Deonna et al, 1993; De Saint-Martin et al, 1999;
Kramer et al, 2001) after prolonged anarthric status epilepticus, but the long-term outcome of
children with AEOS is rarely reported, and no systematic neurolinguistic studies have been
carried out. We describe a longitudinal study of a child with functional, epilepsy-related oral-
motor dysfunction who developed long-lasting oral and written language deficits.
CASE REPORT
This Italian right-handed male child patient was born after an uneventful pregnancy, delivery
and neonatal period and had a history of normal motor, cognitive and language development.
At the age of 5 years and 3 months he had his first seizure while falling asleep with twitching
of the right eyelid, corner of the mouth and right arm, lasting one minute, followed by an
inability to speak for about 10 minutes.The following day he was admitted to our Department:
He was alert and fully conscious; the main clinical features included pharyngeal, lingual and
masticatory motor deficits, drooling of saliva, and severely impaired speech initiation.
A video EEG showed almost continuous, high-voltage, bilateral centro-temporal
synchronous and asynchronous spikes and sharp and slow waves, more prominent on the left
(Figure 1). The child was severely dysarthric but responsive and able to follow commands.
Figure 1 — EEG at admission: almost continuous, high-voltage, bilateral centro-temporal synchronous

and asynchronous spikes and sharp and slow waves, more prominent on the left.
Figure 2 — EEG performed 3 minutes after an intravenous administration of diazepam (5mg): marked
improvement in the EEG; a slow subcontinuous activity, 2-3 Hz, persisted on the left centro-parietal
areas.
Language testing, performed during the EEG recording, revealed a severe difficulty in
naming familiar objects and pictures and an inability to reproduce simple oral gestures on
imitation.
Morphosyntactic comprehension as tested by TCGB, a multiple-choice test of receptive
grammar, (Chilosi & Cipriani, 1995), was also impaired. Intravenous diazepam abated EEG
discharges; a slow subcontinuous activity (2-3 Hz) persisted on the centro-parietal areas with
concomitant clinical improvement (Figure 2).
The child began to name objects and pictures and made some volitional orolingual
movements on request (sticking out the tongue, clicking the tongue to imitate the sound of a
horse galloping). Magnetic Resonance Imaging (MRI) was normal. Treatment with sodium
valproate was started and no other acute episodes of oral motor dysfunction and speech arrest
occurred. The child was followed at our clinic from the post-acute phase up to the age of 8
years and 7 months. Follow-up EEGs showed normal background activity and frequent sharp-
wave complexes, synchronous and asynchronous on both centro-temporo-parietal regions,
dramatically enhanced during sleep. In spite of these severe abnormalities, no overt clinical
symptoms or changes in speech fluency were demonstrated during the EEG recording.
At 5 years and 9 months and 6 years and 4 months he suffered two right-sided focal motor
seizures upon falling asleep. Various combinations of drugs (sodium valproate, clonazepam,
hydrocortisone, ethosuximide) only produced transient improvement on EEG paralleled by
improvement of language performance. Three years after the initial symptoms, the child was
seizure-free under sodium valproate, but the EEG was persistently abnormal.
LANGUAGE AND NEUROPSYCHOLOGICAL FOLLOW-UP
The child's neurological status, oromotor functions, language, speech and cognitive abilities
were repeatedly evaluated from 5.3 up to 8.7 years of age. The first full language and
cognitive assessment was performed a few days after the acute episode. Expressive language
was grammatically correct but simplified, slow and nonfluent; on a picture naming test
(Brizzolara et ai, 1994), global performance was within normal limits, but it was
characterised by an excessively high latency of response and by an abnormally high number
of anomias, as a consequence of word retrieval deficits.
The child's language comprehension was within the norms for his age (Chilosi & Cipriani,
1995). Cognitive assessment revealed a mild deficit of nonverbal (Leiter International
Performance Scale, LIPS) (Leiter, 1979) and visuo-motor integration abilities (VMI, Beery,
1997) and a severe impairment of verbal short-erm memory (Digit span) (Orsini et al., 1987)
and phonological working memory (recall of lists of words in the auditory-visual modality).
The child's subsequent course was relatively benign, though marked by some
inconsistencies in cognitive and language functioning, parallel with improvements and
relapses of the electroclinical conditions.

The main clinical features, as summarised in Table 1 and graphically represented in
Figures 3-7 are as follows:
fluctuating levels of performance on nonverbal tasks (visuo-motor integration skills and
performance IQ)
improvements and relapses of grammatical comprehension
phonological processing deficits, manifested by an impaired performance on working
memory and phonemic fluency tasks
slow rate of articulation and mild oral dyspraxia
word-retrieval difficulties with a variable preponderance of anomias and paraphasic
speech manifesting as hesitations, circumlocutions, "conduites d'approche", and/or
phonemic and semantic substitutions.
Table 1 — Summary of clinical, neuropsychological and EEG data from 5.3 years to 6.4 years
5.3 (acute 5.3 (post 5.5 5.7 5.11 6.4

phase) acute phase)
EEG ++ +++
++++ ++ ++ +++
abnormalities ft
Partial motor
Y N N N Y Y
seizures
Oral motor +++ ++ + + + ++
dysfunction ft ft
SPEECH AND LANGUAGE
Dysarthria ++ +
+++ +++ + +
ft ft
Anomias ++
+++ +++ +++ +++ ++
ft
Grammatical normal impaired
impaired normal normal normal
comprehension ft
Phonemic fluency n.a. impaired impaired impaired impaired impaired
Semantic fluency n.a. impaired impaired borderline borderline borderline
Phonological
n.a. impaired impaired borderline borderline impaired
Working memory
COGNITIVE SKILLS
Nonverbal
n.a. 67 99 67 94 92
IQ
VMI
n.a. 72 85 77 79 83
(standard score)
Digit span
n.a. 2° 2° 22° 9° 9°
(percentile)
Note: U = better; 0 = worse; Y= yes; N = no; n.a. = not available

++++ = very severe; +++ = severe; ++ = moderate; += mild
Figure 3 — Neuropsychological follow-up: non verbal abilities: Performance IQ (PIQ) and Visuo-
motor integration skills (VMI Standard score).
Figure 4 — Neuropsychological follow-up: Grammatical Comprehension.
Figure 5 — Neuropsychological follow-up: Phonological Working Memory (auditory-visual

modality).
Figure 6 — Neuropsychological follow-up: Verbal fluency.
Figure 1 — Neuropsychological follow-up: Lexical production (types of errors).
LONG-TERM OUTCOME
When last seen, at the age of 8 years and 7 months (three years after the onset of the AEOS),
the child was attending the 3rd grade of primary school. The EEG showed frequent sharp-wave
complexes that were synchronous and asynchronous on both centro-temporo-parietal regions
which became almost continuous during sleep, but the child was seizure-free.
Upon examination, some residual signs of orofacial clumsiness and verbal dyspraxia were
observed with a persistently very slow rate of speech and some prosodic alteration. Nonverbal
cognitive abilities (PM47, Raven, 1956, 1984; Pruneti et al, 1996; VMI, Beery, 1997) and
receptive vocabulary (Peabody Picture Vocabulary Test, PPVT, Dunn & Dunn, 1981; Stella et
al, 2000) were within average range, whereas performance on morpho-syntactic
comprehension, naming and phonological processing tasks was below the norms (Table 2). A
formal assessment of literacy skills by means of standardized tests of reading and spelling
(Martini, 1995; Sartori et al, 1995) revealed a severe learning disability. Reading
performance was significantly impaired both for speed and accuracy. The child could read
and write only disyllabic words, making use of a prevalent letter by letter strategy and
manifested severe difficulties to find out the meaning of even single words. Writing to
dictation was also severely impaired (Table 2).
Table 2 — Summary of residual deficits
NON VERBAL ABILITIES

PM47 VMI
(percentile) (standard score)
30° 77
LANGUAGE
Receptive vocabulary Receptive grammar Expressive vocabulary
PPVT (standard score) TCGB (z-score) Naming test (z-score)
102 -2.5 - ,4
PHONOLOGICAL PROCESSINC
Phonological Working Memory Verbal fluency
(z-scores) (z-scores)
Short words Long words Phonemic Semantic
-2,2 -1,5 -1,6 -0,9
ACADEMIC SKILLS
Writing Reading (z-scores)
(single words) Speed Accuracy
(z-scores)
Short Long Letters Words Short words Long words
-1,95 > -5 > -5 > -5 > -5 > -5
DISCUSSION
This child sustained a severe functional, epilepsy-related disorder, with sudden onset of
suprabulbar palsy and focal seizures. The course of epilepsy was rather favourable (despite
persistent EEG abnormalities), but he was left with mild verbal dyspraxia, some expressive
language difficulties and a severe disorder of reading and writing. Some other rare cases of
AEOS with long-lasting high-level language deficits have occasionally been described in the
literature. Deonna et al.'s case 2 (1993) was reported to have difficulty with written language
(mainly severe dysorthography) and dysfluent speech. Patient 1 in the Kramer et al. 's series
(2001) showed word-retrieving difficulties on a formal language test, whereas patient 3 (in the
same series) had poor reading skills. In the authors' opinion, the above symptoms should not
be influenced by a sensorimotor deficit and represent a "higher function deficit" whose
significance is unclear. In a recent review of the opercular epilepsies with oromotor
dysfunction, Salas-Puig et al. (2000) pointed out that some children with perisylvian
developmental disorders (bilateral opercular malformations) may present with a quite
homogeneous language dysfunction that cannot be accounted for by a speech paretic disorder
and by oral dyspraxia (as in the classic form of FCMS), but should imply involvement of
language areas, though to a variable degree. At present, the pathophysiology of central
processing disorders in AEOS rests on speculative grounds (Fejerman et al, 2000). We may
hypothesise that the spreading of abnormal electrical brain activity to adjacent regions of the
frontal cortex subserving phonologically-based linguistic processes would interfere with
higher neurofunctional language mechanisms that sustain coding and rehearsal of
phonological and lexical information. This hypothesis would also explain reading and writing
difficulties, by assuming the presence of a central impairment in the acquisition of rapid and
automatized rules for recoding oral language knowledge into its written form.
From a neurolinguistic point of view, the relationship between AEOS and oral/verbal
dyspraxia, has not been systematically addressed in the literature. This may be due, at least in
part, to the following facts. First, in some papers on functional or structural Opercular
Syndrome (OS), the terms dyspraxia and dysarthria have been used interchangeably, although
the paretic origin of typical suprabulbar palsy is well recognised since Worster - Drought's
(1974) influential work. In our patient, both paretic and praxic defects of the facio-lingual-
glosso-pharingeal muscles were variably present during different stages of the illness. Second,
as described above, rare cases of AEOS with persisting subtle oral and written language
deficits have been reported, but few explanatory hypotheses have been proposed. Third, the
pathophysiology of acquired apraxia of speech in adults and of developmental apraxia of
speech (DAS) in children is very controversial. The debate concerns both the nature of the
disorder - in terms of a movement or a language disturbance (Robin, 1992) - and the specific
brain site responsible for it. Some suggestions for speculating on the localization of DAS arise
from different and sparse sources drawn from adult patients with acquired lesions and from
children with both congenital and acquired speech disorders (Habib & Demonet, 1996;
Vargha-Khadem et al, 1998; Van Mourik et al, 1997).
However, an involvement of perirolandic cortex in the adjacency of the inferior motor
strip (dedicated to the innervation of lips, tongue and glosso- pharyngeal muscles) and of the
'pars opercularis' of Broca's area has been advocated as the most likely candidate region by
several authors (Alexander et al, 1990; Foundas et al, 1998). In addition, based on data from
stroke patients with apraxia of speech, Dronkers (1996) identified the precentral gyrus of the
insula as the brain region co-ordinating speech articulation. A functional impairment of these
regions may result from epileptic activity involving them either primarily, or as a result of a
spreading effect from contiguous areas. However, confirmation of this hypothesis would need
sophisticated electrophysiological and neurofunctional studies, through two-dimensional
topographical EEG mapping and coherence/spectrum analysis, and fMRI exploration of brain
regions of interest.
In conclusion, the long-lasting, praxic and language deficits in AEOS might be related -
as suggested for some permanent sequelae in Landau-Kleffner Syndrome (LKS) - to a
disruption of normal synaptic connectivity in the perisylvian cortex during a sensitive stage of
its development. Similarly to LKS, the effects of epileptogenic discharges on neuronal
networks subserving language would 'activate and perpetuate synaptic arrangements that are
functionally inappropriate' (Morrell et al, 1995).
REFERENCES
Alexander, M. P., M. A. Naeser and C. Palumbo (1990), Broca's area aphasias: Aphasia after
lesions including the frontal operculum. Neurology, 40, 353-362.
Beery K.E. (1997) VMI. Developmental test of visual-motor integration, 4th Edition, Revised.
Toronto: Modern Curriculum Press.
Brizzolara, D., P. Cipriani, A. M. Chilosi and L. De Pasquale (1994). L'apprendimento del
linguaggio scritto nei bambini con difficolta di acquisizione del linguaggio orale:
continuita o discontinuity? In: Apprendimento e patologia neuropsichica nei primi
anni di scuola. Modelli interpretativi della clinica (G. Masi and A. Martini, editors),
pp. 124-135. Borla, Roma.
Chilosi, A. and P. Cipriani (1995). Test di comprensione grammaticale per bambini (TCGB).
Del Cerro, Tirrenia.
Christen, H. J., F. Hanefeld, E. Kruse, S. Imhauser, J. P. Ernst and M. Finkenstaedt (2000).
Foix-Chavany-Marie (anterior operculum) syndrome in childhood: a reappraisal of
Worster-Drought syndrome. Dev Med Child Neurol, 42, 122-32.
Colamaria, V., V. Sgro, R. Caraballo, M. Simeone, E. Zullini, E. Fontana, R. Zanetti, R.
Grimau-Merino and B. Dalla Bernardina (1991). Status epilepticus in benign rolandic
epilepsy manifesting as anterior operculum syndrome. Epilepsia, 32, 329-334.
De Saint-Martin, A., C. Petiau, R. Massa, R. Maquet, C. Marescaux, E. Hirsch and M. N.
Metz-Lutz (1999). Idiopathic rolandic epilepsy with "interictal" facial myoclonia and
oromotor deficit: A longitudinal EEG and PET study. Epilepsia, 40, 614-620.
Deonna, T. W., E. Roulet, D. Fontan and J. P. Marcoz (1993). Speech and oromotor deficits
of epileptic origin in benign partial epilepsy of childhood with rolandic spikes
(BPERS). Neuropediatrics, 24, 83-87.
Dronkers, N. G. (1996). A new brain region for coordinating speech articulation. Nature, 384,
159-61.
Dunn, L. and L. M. Dunn (1981). Peabody Picture Vocabulary Test - Revised. American
Guidance Service, Circle Pines, MN.
Fejerman, N. and M. Di Blasi (1987). Status epilepticus of benign partial epilepsies in
children: report of two cases. Epilepsia, 28, 351-355.
Fejerman, N., R. Caraballo, S. N. Tenembaum (2000). Atypical evolutions of benign

localization-related epilepsies in children: Are they predictable? Epilepsia, 41, 380-
390.
Foix, C, J. A. Chavany and J. Marie (1926). Diplegie facio-linguo-masticatrice d'origine
cortico-sous-corticale sans paralysies des membres. Rev Neurol, 33, 214-219.
Foundas, A. L., K. F. Eure, L. F. Luevano and D. R. Weinberger (1998). MRI asymmetries of
Broca's area: The pars triangularis and pars opercularis. Brain Lang, 64, 282-296.
Galanopoulou, A. S., A. Bojko, F. Lado and S. L. Moshe (2000). The spectrum of
neuropsychiatric abnormalities associated with electrical status epilepticus in sleep.
Brain Dev, 22, 279-295.
Gordon, N. (2002) Worster-Drought and congenital bilateral perisylvian syndromes. Dev Med
Child Neurol 44, 201-204.
Habib, M., J.-F. Demonet (1996). Cognitive neuroanatomy of language: the contribution of
functional neuroimaging, Aphasiology, 10, 217-234
Kramer, U., B. Ben-Zeev, S. Harel and S. Kivity (2001).Transient oromotor deficits in
children with benign childhood epilepsy with central temporal spikes. Epilepsia, 42,
616-620.
Leiter, R.G. (1979). Letter International Performance Scale. Stoelting Co, Chicago.
Martini, A. (1995). Le difficolta di apprendimento della lingua scritta. Criteri di diagnosi e
indirizzi di trattamento. Del Cerro, Tirrenia.
Morrell, F., W. W. Whisler, M. C. Smith, J. H. Thomas, L. de Toledo-Morrell, S. J. C. Pierre-
Louis, et al. (1995). Landau-Kleffner syndrome: Treatment with subpial intracortical
transection. Brain, 118, 1529-1546.
Orsini, A., D. Grossi, E. Capitani, M. Laiacona, C. Papagno and G. Vallar (1987). Verbal and
spatial immediate memory span: normative data from 1355 adults and 1112 children.
Ital J Neurol Sci, 8, 539-548.
Pruned, C, A. Fenu, G. Freschi, S. Rota, D. Cocci, M. Marchionni, S. Rossi and G.
Baracchini Muratorio (1996). Aggiornamento della standardizzazione italiana del test
della Matrici Progressive Colorate di Raven (CPM). Bollettino di Psicologia
Applicata, 217,51-57.
Raven, J. C. (1956) Guide to using the Coloured Progressive Matrices, sets A, AB and B.
London: H.K. Lewis. Italian version (1984): Progressivi matrici colore. Serie A, AB,
B. Firenze: Organizzazioni Speciali.
Robin, D. A. (1992). Developmental apraxia of speech. Am J Speech Lang Pathol, 1, 9-22.
Roulet, E., T. Deonna and P. A. Despland (1989). Prolongued intermittent drooling and
oromotor apraxia in benign chilhood epilepsy with centrotemporal spikes. Epilepsia,
30(5), 564-568.
Salas-Puig, J., A. Perez-Jimenez, P. Thomas, I. E. Scheffer, B. Dalla Bemardina and R.
Guerrini (2000). Opercular epilepsies with oromotor dysfunction. In: Epilepsy and
Movement Disorders (R. Guerrini, J. Aicardi, F. Andermann and M. Hallet, editors),
pp. 251-268. Cambridge University Press, Cambridge.
Sartori, G., R. Job and P.E. Tressoldi (1995). Batteria per la valutazione della dislessia e
delta disortografia evolutiva. Organizzazioni Speciali, Firenze.
Shafrir, Y. and A. L. Prensky (1995). Acquired epileptiform opercular syndrome: a second
case report, review of the literature, and comparison to the Landau-Kleffner syndrome.
Epilepsia,36, 1050-1057.
Shuper, A., B. Stahl and M. Mimouni (2000). Transient opercular syndrome: a manifestation
of uncontrolled epileptic activity. Ada Neurol Scand, 101, 335-338.
Stella, G., C. Pizzoli, P. E. Tressoldi (2000). Peabody. Test di Vocabolario Recettivo. Omega
Edizioni, Torino.
Tachikawa, E., H. Oguni, S. Shirakawa, M. Funatsuka, K. Hayashi and M. Osawa (2001).
Acquired epileptiform opercular syndrome: a case report and results of single photon
emission computed tomography and computer-assisted electroencephalographic
analysis. Brain Dev, 23, 246-250.
van Mourik, M., C. E. Catsman-Berrevoets, H. R. van Dongen and B.G.R. Neville (1997).
Complex orofacial movements and the disappearance of cerebellar mutism: Report of
five cases. Dev Med Child Neurol, 39, 686-690.
Vargha-Kadem, F., K. Watkins, C. J. Price, J. Ashburner, K. J. Alcock, A. Connelly, et al.
(1998) Neural basis of an inherited speech and language disorder. Proc Nat Acad Sci,
95, 12695-12700.
Worster-Drought, C. (1974). Soprabulbar Paresis. Congenital suprabulbar paresis and its
differential diagnosis, with special reference to acquired suprabulbar paresis. Dev Med
Child Neurol, 16, 1-33.
Brain Language Lateralization and Focal Lesions 49
CEREBRAL LANGUAGE LATERALIZATION AND

EARLY LINGUISTIC DEVELOPMENT IN
CHILDREN WITH FOCAL BRAIN LESIONS
Anna M. Chilosi, Chiara Pecini, Paola Cipriani, Daniela Brizzolara, Paola Brovedani,
Giovanni Ferretti, Lucia Pfanner and Giovanni Cioni
"Stella Maris " Scientific Institute, Pisa, Italy
Abstract — We conducted a longitudinal study of 20 children with unilateral focal brain lesions and
hemiplegia, 11 with left (LHD) and 9 with right hemisphere damage (RHD) to investigate the
relationship between lesion characteristics, early linguistic development and hemisphere lateralization
for language. Cerebral lateralization for language was measured by means of the Fused Dichotic
Words Test. Two comprehensive assessments of language comprehension and production were
performed at about 2 and 4 years of age. An early left side-specificity for language was revealed by
the presence of lexical and grammatical delay in the majority of LHD children. In 90% of LHD
children, plasticity and the potential for re-organization were documented by a shift in lateralization
for language to the right hemisphere on the dichotic listening test. There was an association,
irrespective of side, between the largest lesions, the most atypical hemispheric asymmetries (as
expressed by laterality coefficients) and delay in grammatical development. Lesion type seemed to
significantly affect hemispheric lateralisation and short-term language outcome, cortico-subcortical
lesions being significantly associated with a greater degree of lateralization and language delay in
comparison to periventricular white matter lesions. The presence of EEG abnormalities and/or seizures
negatively affected language outcome.
Keywords: language development, focal lesions, hemispheric lateralization, Fused Dichotic

Words Test.
INTRODUCTION
Several studies report that early brain lesions have relatively mild consequences on language
development in comparison to lesions acquired later in adulthood, and do not necessarily
show a clear-cut association to side, site or size of lesion (Vargha-Khadem el al, 1985;
Vargha-Khadem et al, 1992; Muter et al., 1997; Reilly et al., 1998). These data have been
interpreted in relation to the concepts of 'plasticity' and 'equipotentiality' of the immature
brain: while plasticity refers to the compensatory mechanisms underlying lesion-induced
neurofunctional and behavioural reorganisation, equipotentiality refers to the analogous
capacity of the two hemispheres to sub-serve language functions after unilateral brain damage
(Lenneberg, 1967). This theory has been challenged by several studies which support the
existence of an early specialisation of the left hemisphere for language acquisition, although
the differential effect of left and right lesions seems to depend on the specific stage of
language development. In fact, Bates et al. (1997) found that in the period from 10 to 17
months of age, children with a right lesion were at a greater risk for delays in word
comprehension and gesture production than children with left hemisphere damage (LHD), and
in the period from 10 to 44 months children with a lesion involving the left temporal lobe
showed significantly greater delay in expressive vocabulary and grammar (see also Thai et al.,
1991; Vicari et al, 2000; Chilosi et al, 2001). However, the effect of a left temporal damage
on grammar development was no longer detected past 5-6 years of age (Reilly et al, 1998).
These data on the linguistic development of LHD children suggest that 'equipotentiality' and
early 'left hemisphere specialisation' may represent the two poles of a continuum: the left
hemisphere may be innately predisposed to language learning and processing, but this
predisposition is sufficiently plastic for the non-dominant hemisphere to successfully acquire
and mediate language in conditions such as early focal brain damage (Vicari et al, 2000;
Chilosi etal, 2001; Satzefa/., 1990).
More recently, the plasticity of the immature brain and the concept of equipotentiality of
the two hemispheres was supported by new neuroimaging evidence which found a right
hemisphere specialization for language after early damage to the left language areas (Miiller
et al, 1999; Lazar et al, 2000). According to these studies, the reorganization for language in
the right hemisphere involves regions which are mostly homotopic to the language areas in
the left hemisphere of healthy right-handers, thus suggesting a 'near-equipotentiality' of the
two hemispheres also at a topological level (Staudt et al, 2002). However, it is still too early
to draw general conclusions from functional neuroimaging studies, as there are in fact two
main methodological limitations: a) normative data from healthy children are still scarce, and
b) most of the data are collected on patients with early-onset and severe epilepsy who are
candidates for neurosurgery.
The issue of how language reorganises after early focal damage has been traditionally
addressed by behavioural techniques, such as the dichotic listening paradigm. The use of a
behavioural paradigm allows for the study of larger samples of patients and is especially
appropriate for investigating language lateralisation in very young children in comparison to

neuroimaging methods. In the dichotic listening paradigm two competing verbal stimuli are
presented simultaneously to the two ears. According to Kimura's structural model (Kimura,
1961, 1967), in this condition, controlateral auditory pathways occlude the ipsilateral
pathways. This leads to a better processing of the stimuli presented to the right ear (REA,
right ear advantage) as they have more direct and faster access to the language-dominant left
hemisphere than the stimuli presented to the left ear, which are in fact assumed to access first
the right hemisphere and then the left through the corpus callosum (for a review, see Bryden,
1981). Dichotic techniques were employed to investigate whether language reorganization
after an early lesion occurs inter- or intrahemispherically in relation to the characteristics of
the lesion (Brizzolara et al., 2002).
The presence of epileptic activity is another factor that may in itself alter the pattern of
lateralization. Piccirilli et al. (1988), using a verbal-manual dual task, found that in children
with benign focal epilepsy with no documented lesion, a left unilateral epileptogenic focus
was associated with a bilateral representation of language processing. Riva et al. (1993)
compared the performance of epileptic children with unilateral foci and the presence or
absence of CT documented lesions on a verbal-spatial tachistoscopic task and found that
cerebral lateralisation was altered similarly in both groups. On the basis of these findings they
suggested that epilepsy alone can change the pattern of lateralisation for verbal information
processing.
More recently, Isaacs et al. (1996) addressed the issue of the effect of seizure disorder on
language lateralisation in hemiplegic children. On a dichotic listening task using digits, left
lesioned children with a history of clinical seizures displayed a stronger LEA than left
lesioned children without seizures, probably because of a more limited potential for language
processing in the hemisphere where the epileptic focus was active.
In a previous study conducted in our laboratory (Brizzolara et al., 2002), we investigated
cerebral lateralization for language by means of the Fused Dichotic Words Test (Wexler &
Halwes, 1985) in 26 hemiplegic children with congenital focal brain damage with the specific
aim to investigate the relation between lesion characteristics (side, size and localisation) on
MRI and the pattern of language lateralisation on the dichotic test. We found significant side
and site effects at group level: while children with right lesions showed the expected right ear
advantage (REA), in children with left hemisphere lesions there was a left ear advantage
(LEA). An analysis of individual data, however, revealed that type of lesion (e.g., cortico-
subcortical versus periventricular) occurring at term or pre-term respectively, may be the
primary factor responsible for inter- vs. intrahemispheric organisation of language after
congenital brain lesions. Only when the left lesions involved cortico-subcortical regions
encroaching the temporal lobe and occurred at term, language was reorganised in the right
hemisphere; when lesions (whether left or right) involved only the periventricular white
matter and occurred at pre-term, language was lateralised in the left hemisphere. However, the
relationship between lesion localisation, re-organisation of language and functional effect on
language development is still an open issue.
On the basis of these considerations, the aim of the present study was twofold:
a) to analyse the course of linguistic development in relation to different lesions
characteristics;
b) to investigate whether there is a relation between the degree of functional specialisation for
language, expressed by LEA and REA values on the dichotic test, and the timing and
trajectory of linguistic development.
We addressed these issues by examining language development and reorganisation from the
third to the fourth year of age in a sample of children with pre- or perinatal focal brain
damage.
SUBJECTS
Subjects were selected from a larger population of patients with congenital focal brain lesions
referred to the Division of Child Neurology and Psychiatry of the University of Pisa on the
basis of the following criteria:
unilateral focal brain lesions, occurring pre- or peri-natally, documented on the basis
of clinical records and MRI
absence of diffuse or progressive lesions or brain malformations
comprehensive longitudinal linguistic assessment with at least 2 evaluations between
the age of about 2 and 4 years, with an interval between the two consecutive
observations of at least 8 months
absence of mental retardation (Developmental Quotient > 80) at the time of the first
evaluation
absence of treatment- resistant epilepsy at the two evaluation time points
absence of auditory deficits and personality disorders.
MRI assessment
Brain MRI studies were performed under sedation using a 1.5 T system (GE, Signa
Advantage); images were obtained in the axial, coronal and sagittal planes with sections of 5
mm. When multiple scans were available, the most recent one was considered. MRI findings
were classified retrospectively by one of the authors (GC), blind to the results of linguistic
and cognitive assessment, according to the most recent indications in the literature about
neuroimaging findings in congenital hemiplegia (Tailairach & Tournoux, 1988; Barkovitch,
1995; Cioni et al, 1999). Results are reported in Table 2.
Lesions were classified according to three different criteria:
Side and site of lesion. Patients were assigned either to the "right" or "left" unilateral lesion
group. Moreover, cerebral lobes involved in the lesion and the encroachment of language
areas (frontal and temporal lobes) were recorded.
Brain Language Later•alization and Focal Lesions 53
Size of lesion. Lesion size was classified by a grading system adapted (with modifications)
from Vargha-Khadem et al. (1985). It consists of a six-point scale, ranging from 0 to 5,
providing an index of lateral ventricular dilatation and of the extension of encephaloclastic
cysts; score 5 indicates the most severe abnormality. This scale is presented in papers
previously published by the authors (Vicari et al, 2000; Chilosi et al, 2001).
Type of lesion. Both the pathophysiological mechanisms of the lesion and its probable timing
were taken into account. Two types of lesions were observed:
1) Periventricular white matter lesions (PV), likely due to parenchymal haemorrhages
(results of venous infarction or of haemorrhagic periventricular leukomalacia) or to
periventricular leukomalacia. These lesions usually occur either intra or extra utero (in
case of preterm birth) in the last trimester of gestation. Unilateral encephaloclastic cysts,
merged into a dilated ventricle, are often observed on MRI at older ages. In the other
cases, lesions consist of symmetrical or asymmetrical periventricular gliosis.
2) Cortico-subcortical (C-SC-PV) lesions, due to an infarction of a main cerebral artery
(generally main branch or cortical branch of middle artery), involving the cortex, the
white matter below the cortex and often the periventricular white matter. Sometimes,
the lesion concerns the deepest branches (exclusively or in association with other
lesions) involving diencephalic structures. These lesions usually occur at around term
age.
Electrophysiological assessment
EEGs were obtained for all patients, the majority of whom receiving repeated recordings.
EEG tracings closest to the time of linguistic and psychological observations were analysed
and results were classified according to the nature of the abnormality (diffuse, focal,
paroxysmal), its frequency and the condition during which it occurred (wakefulness, sleep,
other types of activation). Findings were scored as normal, mildly abnormal or severely
abnormal. The occurrence of clinical seizures with onset beyond the neonatal period and
excluding febrile convulsions were documented.
Linguistic assessment
Language evaluation was performed through a combination of indirect procedures (parental

interview to collect information on productive vocabulary size) and direct observations to
evaluate the level of expressive and receptive grammar. The latter were based on free-speech
samples and on a test of Early Verbal Comprehension (Chilosi et al, 2003).
Expressive vocabulary was tested by means of the Infant's and Toddler's MacArthur
Communication Developmental Inventories (Italian version: // Primo Vocabolario del
Bambino - PVB) (1995), for which normative data are available in the 8-30 months age range.
Because many of the children in our focal lesion sample were delayed in language
development, assignment of the Infant's or Toddler's form was based on language level rather
than on chronological age. The analyses presented here will focus on productive vocabulary
from both forms and will be expressed by total number words and by a lexical quotient (LQ)
corresponding to the ratio between lexical age (i.e., the age at which a particular score
corresponds to the median in the normative sample) and chronological age.
Expressive grammar was evaluated on the basis of language samples collected in our
laboratory during a standardised play situation involving the child and his/her parents. Speech
was transcribed independently by one of the authors (L.P.) and by a trained research assistant
(inter-observer agreement reached 90%). The utterances were then coded according to the
Child Language Exchange System (CHILDES) (Mac Whinney & Snow, 1985). For each child
the level of grammatical development was scored on a six-level rating system (see Table 1)
developed by Cipriani et al. (1993), ranging from Level 0 (pre-linguistic stage) to Level 5
(complex grammar).
Verbal comprehension was investigated by an acting-out task that has been standardised
on a sample of Italian children. It includes 20 simple verbal commands of increasing
complexity that the child is required to act out with a set of toys or familiar objects. For each
child, a z-score was obtained on the basis of normative data from 6 groups of normal children
aged 16-36 months (Chilosi et al, 2003).
Table 1 Levels of grammatical development
Level Expressive language
Level 0 Babble, sounds, gestures and sporadic single word

utterances (SW)
Level 1 Single-word utterances (SWU) start to be consistently
produced
Level 2 Emergence of combinatorial speech, but SWU prevail
Level 3 Ungrammatical or telegraphic multiword utterances
(MWU)
Level 4 Fully grammatical simple sentences
Level 5 Fully grammatical complex sentences
Dichotic Listening paradigm
The Fused Dichotic Words Listening Test (Wexler and Halwes, 1985) (consisting of pairs of
rhyming words) was used since it has been shown to be a more valid measure of cerebral
lateralisation than non-fused versions (Zatorre, 1989). Fifty-five high frequency two-syllable
words were used (Marconi et al., 1994), 28 CVCV and 27 CVCCV. The stimuli were
recorded in Digital Audio Tape mode in a noise-protected room and sampled by a digital
SoundBlaster Hard-Disk-Recording for PC. Sampling cared that words forming the dichotic
pair were synchronised for the beginning of the first consonant, for length and for some
internal features (especially where the accent fell). Stimuli were presented by a specific
program running on a PC. The experiment consisted in the dichotic presentation of 30 pairs of
fused words: twenty-five pairs differed for the first consonant (e.g., cane-pane) and 5 pairs
differed for the first vowel (e.g., luna-lana). Word pairs were presented twice to all subjects;
in the second session the assignment of stimuli to ears was reversed. The order of presentation
of the word pairs was fixed across subjects and varied across sessions. Overall, each subject
heard 60 stimuli in each ear through headphones (Sony Professional MDR-V50). Children
were instructed to repeat all the words they heard, after each presentation. The raw data were
then converted to a laterality coefficient (Lambda) according to the procedure proposed by
Bryden and Sprott (1981). It consists of the natural logarithm of the number of correct
responses for the words heard by the right ear plus 1, divided by the number of correct
responses for the words heard by the left ear plus 1 (Ln[(Right+l)/(Left +1)]). A positive
value is indicative of REA, reflecting a left hemisphere superiority for language processing.
Conversely, a negative value indicates a LEA, reflecting a right hemisphere superiority. Mean
laterality coefficient values, obtained on a large sample of normal children aged 4-10 years,
ranged from 0.32 to 0.54 and were stable across ages (Brizzolara et ah, 2000). In the present
study the dichotic test was administered as soon as children were able to cooperate reliably,
with mean chronological age of 5 years and 4 months.
RESULTS
Clinical and MRI characteristics
According to the selection criteria, a total of 20 children (13 males and 7 females) participated
in the study, 11 with LHD and 9 with RHD. Mean chronological age at the first evaluation
time point (77) was 23.5 months (SD 3.0, range 16-29) for LHD children and 21.8 months
(SD 5.1, range 17-32) for RHD children. At 77, mean age for LHD children was 38.3 months
(SD 1.6, range 36-42) and 39.2 months (SD 3.6, range 35-44) for RHD children. The mean
interval between 77 and T2 was 14.8 months for LHD children (range: 9-23) and 17.4 months
for RHD children (range: 8-26). Neither age at the two time-points nor the T1-T2 interval
significantly differed between the two groups (t-test). Table 2 shows the sample
characteristics, time of brain injury (hypothesized on the basis of medical history and
neuroradiological data), MRI findings, presence or absence of epilepsy, EEG findings and age
at the two evaluation time points. Nineteen children had a documented hemiplegia of
differing degree of severity. Four had epilepsy that was well controlled by mono- or
polytherapy at the time of the study. Fourteen children had C-SC-PV lesions, 5 had PV
lesions and only one child had a SC-PV lesion. The mean size of the lesions (grade) was 3.8,
with a range between 1 and 5. Sixteen children had EEG abnormalities that were mildly
abnormal in 10 and severely abnormal in 6. Four of the latter had clinical seizures at the time
of the study, whereas two cases (cases 6 and 14) presented with epilepsy within two years
from T2.
Table 2 - Clinical and neuroradiological characteristics of the sample
Case/ GA Time of MRI findings Age Age Epilepsy EEG

Gender (Wk) lesion atTl atT2 findings
Site (mo) (mo)
Side Extention Lobes Size
1/M 42 term L C,SC, FTP 4 23 37 N MA
PV
2/F 39 term L c, sc, FTP 4 23 38 Y SA
PV
3/M 38 term L c, sc, PTF 4 24 37 N MA
PV
4/F 39 pre-term L c, sc, PO 4 25 38 Y SA
PV
5/M 40 term L c, sc, FTPO 5 24 39 Y SA
PV
6/M 38 term L c, sc, TPO 4 16 39 Y SA
PV
7/M 41 term L c, sc, PTF 4 23 36 N MA
PV
8/F 40 pre-term L PV FP 3 24 38 N N
9/M 39 term L c, sc, TPF 5 24 42 N MA
PV
10/M 42 term L c, sc, FTP 5 29 38 N MA
PV
11/M 41 pre-term L PV PF 5 24 40 N N
12/M 38 pre-term R SC, ]PV FPT 3 20 35 N SA
13/F 34 pre-term R C sc, FPT 4 24 40 Y MA
PV
14/M 39 term R c, sc, PFT 4 19 44 Y SA
PV
15/M 40 term R c, sc, FTPO 5 17 32 N MA
PV
16/F 40 term R c, sc, FTP 4 26 40 N MA
PV
17/M 40 pre-term R PV FTP 1 22 42 N N
18/F 40 pre-term R PV PT 3 32 40 N MA
19/M 39 term R c, sc, T 4 21 39 N MA
PV
20/F 34 pre-term R PV P 1 15 41 N N
Note: GA = gestational age; Wk = weeks; Mo = months; R = right, L = left, C = Cortical; SC =

subcortical; PV = Periventricular; F = Frontal; P = Parietal; T = Temporal; O = Occipital; N = Normal;
MA = Mild abnormalities; SA = Severe abnormalities.
A preliminary analysis was conducted to verify whether LHD and RHD groups differed
for clinical and MRI characteristics:
Lesion type: there was a slightly higher incidence of cortico- subcortical lesions in LHD than
in RHD children but the difference was not significant (chi-square);
Lesion size: LHD children showed a tendency to have larger lesions than RHD children,
however this difference was not significant (chi-square);
Lesion site: 18 children had a lesion encroaching temporal and/or frontal lobes, without
significant differences between left and right lesions.
EEG findings: EEG abnormalities were associated with lesion type, with a higher incidence of
mild and severe abnormalities in C-SC-PV than in PV lesions (chi square = 11.7, p = 0.005);
no significant associations were found with lesion side and size.
Language development
The mean number of words produced by the whole sample at Tl was 66.6 (SD = 102.7, range
= 2-403). This corresponds to a mean age of 16.5 months and to a lexical quotient (LQ) of
78.7. At T2 the mean vocabulary raw score was 413 (SD = 218.9, range = 26 - 660; mean
increase from Tl to T2 = 346.5 words) corresponding to a mean age of 29.8 months and an
LQ of 77.9. The mean z-score for language comprehension at Tl was in the low average
range (z = -0.9), but 70% of LHD and 37.5% of RHD children scored more than one standard
deviation below the norm, showing a varying degree of delay. Table 3 shows the performance
of LHD and RHD children on linguistic tests.
Table 3 — Language performances of left (LHD) and right (RHD) damage children at Tl and T2
(means and standard deviations)
Language assessment
Vocabulary Expressive Language
Grammar Comprehension
N words Lexical Quotient Level * Z score
Mean sd Mean sd Mean sd
LHD Tl 27.9 (26.7) 63 (25.7) 1 -1.4 (0.8)
T2 360.7 (192.6) 70.9 (15.8) 4 -0.9 (0.9)
RHD Tl 109.8 (137) 89.1 (19.1) 1.7 -0.2 (1.1)
T2 485.2 (244) 86.5 (17.5) 4.5 0.09 (0.5)
Note: * median value

LHD = Left hemisphere damage; RHD = Right hemisphere damage; N words = number of words.
At T2, language comprehension improved in both groups (mean z score = -0.36) with only
2 LHD children and 1 RHD child showing a persistent delay. The anova analysis revealed a
significant effect of lesion side on LQ (F(l, 18) = 7.94, p < 0.01) and on verbal
comprehension (F(l,12) = 18.87, p < 0.001), with a lower performance in the LHD group
versus the RHD group; the absence of a significant interaction between lesion side and time of
evaluation indicated that this difference was stable from Tl to T2.
The level of expressive grammar of the whole sample varied widely at Tl, ranging from a
pre-linguistic to an early grammatical phase of development. However, while all LHD
children were delayed and did not produce any word combination (level 0 or 1), only two
children in the RHD group had not yet reached the level of combinatorial speech (chi
square(l) = 8.14, p < 0.005). The disadvantage of LHD children was still present at T2, as 9
out of 11 showed a persistent delay in grammatical development in comparison to the 2 RHD
children who maintained their delay (chi square (1) = 4.8, p < 0.05).
To further analyze the effects of different variables on lexical and grammatical
development, short-term language outcome at T2 was rated according to the following
criteria: age-appropriate language outcome (LQ > 80; expressive grammar level >4) and
delayed language outcome (LQ < 80, expressive grammar level <4). Significantly more LHD
than RHD children showed a delay both in vocabulary (chi square(l) = 5.15, p < 0.05) and
grammar (chi square(l) = 4.8, p < 0.05). Grammatical, but not lexical outcome, was also
affected by lesion type and size: in children with PV lesions grammatical outcome was more
frequently age-appropriate than in children with C-SC-PV lesions (chi square(l) = 6.7, p <
0.01); in children with lesion size greater than 3, grammatical outcome was more frequently
delayed in comparison to children with smaller lesions (chi square(l) = 4.4, p < 0.05).
A non-parametric correlation analysis (Spearman) between EEG findings (normal, mild
and severe abnormalities) and linguistic measures produced a negative correlation with
language comprehension (rho = -0.43, p < 0.05) and expressive grammar level (rho = -0.38, p
<0.05)atT2butnotatTl.
Hemispheric lateralization
A t-test on Lambda values at the dichotic test revealed a significant difference between LHD
and RHD groups (t(18) = -9.28, p < 0.001), with the former showing a left ear advantage and
the latter a right ear advantage (Figure 1).
A correlational analysis (Pearson) on the whole sample revealed that Lambda values
significantly correlated with Tl and T2 LQs (Tl, r = 0.43, p < 0.05; T2, r = 0.43, p < 0.05)
and language comprehension ( Tl, r = 0.44, p < 0.05; T2, r = 0.44, p < 0.05). Subjects were
divided into two groups on the basis of Lambda coefficient values: normal and atypical
(falling more than ± 2 SDs from the mean of the normative sample; Brizzolara et al., 2000).
Atypical values were significantly associated with delayed grammatical outcome (chi
square(l)= 7.5, p < 0.01). The Lambda absolute value, which reflects the magnitude of
hemispheric lateralisation, varied according to lesion type, with C-SC-PV lesions showing a
higher degree of lateralisation than PV lesions (t(18) = 2.3, p < 0.05). This result was
confirmed by the analysis of the individual Lambda values: regardless of lesion side, 93% of
the 14 children with C-SC-PV lesions had an atypical lateralisation, while only 34% of the
children with a lesion confined to the periventricular white matter were in the atypical range.
The Lambda absolute value also significantly correlated with lesion size (rho = 0.47, p <
0.05,) with larger lesions being associated with greater Lambda values. No significant
correlations were found between EEG findings and Lambda values.
Figure 1—Mean Lambda values in the LHD and RHD groups and in normal children
DISCUSSION
In the present longitudinal study, we investigated the relationship between unilateral focal
lesion characteristics, early linguistic development and hemispheric lateralisation for
language. The main results consisted in an atypical degree of hemispheric lateralization of
language in both left and right damaged children that was significantly associated with lexical
and grammatical delay in the short-run. We also found an early left side-specificity for
language, although a functional shift and reorganization of language in the right hemisphere
occurred in most LHD children, due to the plasticity of the developing brain.
Side-specificity was revealed by the presence of lexical and grammatical delay in the
majority of LHD children. On the second evaluation, 8 LHD children were still lagging
behind in the acquisition of vocabulary and 9 of them showed a delay in grammar, while only
2 RHD children presented with a delay in both vocabulary and grammar. Left hemisphere
damage was also associated with an initial delay in verbal comprehension, while right injuries
mostly appeared to be associated with a sparing of receptive skills, at least within the time-
widow considered in the present study.
Plasticity and the potential for reorganisation and recovery were documented by the
pattern of lateralisation for language on the dichotic listening test. As expected, 90% of LHD
children presented a shift for language processing to the right hemisphere, while RHD
children showed a left hemisphere lateralisation. Apart from lesion side, the rate and mode of
language development appeared to be influenced by other factors (alone or in combination)
such as lesion size and type, and the presence of EEG abnormalities and/or seizures.
Lesion size was significantly correlated with grammatical outcome and with the degree of
lateralisation, although there were no statistically significant differences between left and
right-sided lesions. Moreover, there was an association, irrespective of side, between the
largest lesions and the most atypical hemispheric asymmetries, confirming previous evidence
obtained by our group (Brizzolara et al, 2002). Interestingly, atypical Lambda values were
significantly associated with a delay in grammatical development. This association was
present in all left damaged children who had an atypical Lambda, but only in 1 out of 5 RHD
children with Lambda values in the atypical range. This may suggest that the right
hemisphere, which takes over language after left cortico-subcortical lesions, cannot
completely accomplish the task of learning grammar, at least at the stage when transition from
primitive combinatorial speech to grammaticisation occurs in normal development.
Lesion type was another variable that significantly affected hemisphere lateralization: C-
SC-PV lesions showed a greater degree of lateralisation than PV lesions, independently of the
side of damage: 13 out of 14 patients with C-SC-PV lesions showed an atypical Lambda in
comparison with only 1 out of 6 children with periventricular white matter lesions. Type of
lesion also seemed to influence language outcome as the only two LHD children with an age-
appropriate language development had a periventricular lesion.
In two of our focal brain damaged children later onset of epilepsy was preceded by severe
EEG abnormalities earlier in life. As outlined by several authors (Sussova et al, 1990; Claeys
et al, 1983; Vargha-Khadem et al, 1992), electrical abnormalities in areas surrounding the
lesion may limit the potential for neurofunctional substitution on the part of undamaged areas.
In conclusion, our results suggest that, in spite of the side-specificity of brain circuitries
dedicated to language, compensatory processes are at work from an early age, probably
because of the activation of substitution mechanisms on the part of uncommitted areas of the
right hemisphere, as revealed by the atypical lateralisation coefficients. The linguistic profile
and the hemisphere lateralization for language in the LHD group could be interpreted as the
consequence of re-allocation of language to alternative regions of the brain under the
influence of different gradients of vulnerability and the asymmetrical maturation timing of the
two hemispheres. According to the so-called 'maturation gradient hypothesis' proposed by
Corballis and Morgan (1978), the more rapid rate of development of the left hemisphere
would force a shift of verbal functions to homologous areas of the right hemisphere.
However, our longitudinal study also demonstrates that neural reorganisation has some costs.
In fact, the slow language development of our left-lesioned children suggests that re-
allocation of language functions in alternative regions of the brain and/or the emergence of
substitution strategies for language is an extended process that goes on well beyond the
normal age expectation.
From a clinical perspective, it must be pointed out that the high degree of variability of
developmental profiles in focal brain damaged children may make it impossible to predict the
long-term outcome. Our data suggest that the risk of a negative short-term language outcome
is represented, on the one hand, by the presence of a left cortico-subcortical hemispheric
lesion, which is strongly associated with atypical cerebral lateralization, and on the other, by
severe EEG abnormalities and/or seizures which worsen language outcome both in left and
right focal brain damage.
REFERENCES
Barkovitch, A. J. (1995). Pediatric neuroimaging, 2nd Edition. Raven Press, New York.
Bates, E., D. Thai, D. Trauner, J. Fenson, D. Aram, J. Eisele and R. Nass (1997). From first
words to grammar in children with focal brain injury. Dev Neuropsychol, 13, 275-343.
Brizzolara, D., P. Brovedani, G. Cioni, P. Cipriani, G. Ferretti and C. Pecini (2000). A
comparison of two different dichotic tests for the assessment of language lateralization
in children. Proceedings ofXXVll International Congress of Psychology. Stockholm,
23-28 July 2000, p.422.
Brizzolara, D., C. Pecini, P. Brovedani, G. Ferretti, P. Cipriani and G. Cioni (2002). Timing
and type of congenital brain lesion determine different patterns of language
lateralization in hemiplegic children. Neuropsychologia, 40, 620-32.
Bryden, M. P. and D. A. Sprott (1981). Statistical determination of degree of laterality.
Neuropsychologia, 19, 571-81.
Caselli, M. C. and P. Casadio (1995). // Primo Vocabolario del Bambino. Franco Angeli,
Milano.
Chilosi, A., P. Cipriani, B. Bertuccelli, L. Pfanner and G. Cioni (2001). Early cognitive and
communication development in children with focal brain lesions. J Child Neurol, 16,
309-316.
Chilosi A.M., Cipriani P., Pfanner L., Villani S. (2003). La valutazione della comprensione
verbale nella prima infanzia: dati normativi e indici predittivi nei ritardi di
acquisizione del linguaggio. Progetto di Ricerca CNR002D39-004. Technical report.
Cioni, G., B. Sales, P. B. Paolicelli, E. Petacchi, M. F. Scusa and R. Canapicchi (1999). MRI
and clinical characteristics of children with hemiplegic cerebral palsy.
Neuropediatrics, 30, 249-255.
Cipriani, P., A. M. Chilosi, P. Bottari and L. Pfanner (1993). L'acquisizione della
morfosintassi in Italiano: Fasi e Processi. Unipress, Padova.
Claeys, V., T. Deonna and R. Chrzanowski (1983). Congenital hemiparesis: the spectrum of
lesions: A clinical and computerized tomographic study of 37 cases. Helv Paediatr
Ada, 38, 439-455.
Corballis, M. and M. Morgan (1978). On the biological basis of human laterality: Evidence
for a matuartional left-right gradient. Behav Brain Sci, 1, 261-268.
Isaacs, E., D. Christie, F. Vargha-Khadem and M. Mishkin (1996). Effects of hemispheric
side of injury, age at injury, and presence of seizure disorder on functional ear and
hand asymmetries in hemiplegic children. Neuropsychologia, 34, 127-37.
Kimura, D. (1961). Cerebral dominance and the perception of verbal stimuli. Can J Psychol,
15, 166-71.
Kimura, D. (1967). Functional asymmetry of the brain in dichotic listening. Cortex, 3, 163-78.
Lazar, R. M., R. S. Marshall, J. Pile-Spellman, H. C. Duong, J. P. Mohr, W. L. Young, R. L.
Solomon, G. M. Perera and R. L. DeLaPaz (2000). Interhemispheric transfer of
language in patients with left frontal cerebral arteriovenous malformation.
Lenneberg, E. H. (1967). Biological foundations of language. Wiley, New York.
Mac Whinney, B. and C. Snow (1985). The Child Language Data Exchange System. J Child
Lang, 12,271-296.
Marconi, L., M. Ott, E. Pesenti, D. Ratti and M. Tavella (1994). Lessico elementare: dati
statistici suH'italiano scritto e letto dai bambini delle elementari. Zanichelli, Bologna.
Miiller, R.-A., R. D. Rothermel, M. E. Behen, O. Muzik, P. K. Chakraborty and H. T.
Chugani (1999). Language organization in patients with early and late left-hemisphere
lesion: a PET study. Neuropsychologia, 37, 545-57.
Muter, V., S. Taylor and F. Vargha-Khadem (1997). A longitudinal study of early intellectual
development in hemiplegic children. Neuropsychologia, 35, 289-298.
Piccirilli, M., P. D'Alessandro, C. Tiacci and A. Ferroni (1988). Language lateralization in
children with benign partial epilepsy. Epilepsia, 29, 19-25.
Reilly, J. S., E. Bates and V. A. Marchman (1998). Narrative discourse in children with early
focal brain injury. Brain Lang, 61, 335-375.
Riva, D., C. Pantaleoni, N. Milani and C. Giorgi (1993). Hemispheric specialization in
children with unilateral epileptic focus, with and without computed tomography-
demonstrated lesion. Epilepsia, 34, 69-73.
Satz, P., E. Strauss and H. Whitaker (1990). The ontogeny of hemispheric specialisation:
Some old hypotheses revisited. Brain Lang, 38, 596-614.
Staudt, M., K. Lidzba, W. Grodd, D. Wildgruber, M. Erb and I. Krageloh-Mann (2002).
Right-hemispheric organization of language following early left-sided brain lesions:
functional MRI topography. Neuroimage, 16, 954-67.
Sussova, J., Z. Seidl and J. Faber (1990). Hemiparetic forms of cerebral palsy in relation to
epilepsy and mental retardation. Dev Med Child Neurol, 32, 792-795.
Tailairach, J. and P. Tournoux (1988). A stereotactic coplanar atlas of the human brain.
Thieme Med Pub, New York.
Thai, D., V. Marchman, J. Stiles, D. Aram, D. Trauner, R. Nass and E. Bates (1991). Early
lexical development in children with focal brain injury. Brain Lang, 40, 491-527.
Vargha-Khadem, F., E. Isaacs, S. van der Werf, S. Robb and J. Wilson (1992). Development
of intelligence and memory in children with hemiplegic cerebral palsy. Brain, 115,
315-329.
Vargha-Khadem, F., A. O'Gorman and G. Watters (1985). Aphasia and handedness in relation
to hemispheric side, age at injury and severity of cerebral lesion during childhood.
Brain, 108, 677-96.
Vicari, S., A. Albertoni, A. M. Chilosi, P. Cipriani, G. Cioni and E. Bates (2000). Plasticity
and reorganization during language development in children with early brain injury.
Cortex, 36, 31-36.
Wexler, B. E. and T. Halwes (1985). Dichotic listening tests in studying brain-behavior
relationships. Neuropsychologia, 23, 545-59.
Zatorre, R. J. (1989). Perceptual asymmetry on the dichotic fused words test and cerebral
speech lateralization determined by the carotid sodium amytal test. Neuropsychologia,
27, 1207-19.
Language Disorders and EEG Abnormalities During NREM Sleep 65
LANGUAGE DISORDERS ASSOCIATED WITH

PAROXYSMAL ABNORMALITIES DURING
NREM SLEEP AFTER VERY EARLY BRAIN
LESIONS
Franco Fabbro, Alessandro Tavano, Guido Cristofori and Renalo Borgalti
"E.Medea " Scientific Institute, Polo del Friuli Venezia Giulia and Bosisio Parini (LC), Italy
University of Udine, Italy
Abstract - We investigated the role of paroxysmal abnormalities in NREM sleep in children with
early brain lesions and language disorders. The presence of epilepsy has been often associated to an
unfavorable prognosis for cognitive and language development in children with early brain lesions.
We assessed language development in six children with pre- or perinatal brain lesions and marked
language disorders by focusing on the following variables: 1) site and side of lesion; 2) presence of
epilepsy; 3) presence of paroxysmal abnormalities in NREM sleep; 4) presence of a clinical picture of
epilepsy and language regression. Cases 1, 2 and 3 presented with many epileptic episodes. The
remaining three cases showed language disorders in the absence of seizures and antiepileptic therapy.
In all cases, we found a high percentage of epileptiform abnormalities in NREM sleep. Cases I, 2 and
3 present with a clinical picture of acquired aphasia associated to epilepsy in the presence of a
neurological lesion (Landau-Kleffner Syndrome-Like). Results confirm the unfavorable prognosis of
the association of early brain lesion, epilepsy and Landau-Kleffner Syndrome for linguistic and
cognitive development. The association between language disorders and/or language regression in
children with early brain lesions and paroxysmal abnormalities in NREM sleep is highlighted. Our
study points out the need to constantly monitor the sleep EEG of children with an early brain injury,
since the presence of a relevant paroxysmal activity in sleep - even in the absence of epilepsy - seems
to be correlated to poor language development.
Key words: early brain lesion, congenital aphasia, developmental dysphasia, NREM sleep,
paroxysmal abnormalities, Landau-Kleffner Syndrome.
INTRODUCTION
The incidence of cerebral palsy of perinatal origin is approximately two in 1000 births. Of
these cerebral palsy cases approximately one-third have hemiplegia. The hemispheric brain
injury responsible for hemiplegia is believed to be due to a thrombotic, vasospasmic or
embolic episode occurring in the middle cerebral and/or internal carotid artery territories. The
etiology of the circulatory event is not yet well understood. It is believed to occur at some
time between the end of the second trimester of pregnancy and the early postnatal period.
Between 30% and 40% of such hemiplegic cases develop a cerebral seizure disorder (Brett,
1992). As documented by many studies, the neuropsychological sequelae of early brain
damage are relatively mild if compared with those of adults. The degree of sparing or
recovery of language and other cognitive functions seems to depend on several factors such
as: 1) time of injury, 2) side of lesion, 3) size of lesion; 4) presence of epilepsy and/or role of
anticonvulsant therapy.
Time of injury. Often, in adults strokes in the left hemisphere do not only lead to hemiplegia
but also to impairments in different aspects of language and/or other cognitive functions.
Similarly, hemiplegia following right hemisphere strokes in adults is often accompanied by
impairments in attentional, visuo-spatial and pragmatic functions (Cytowic, 1996). Contrary
to adults, the language deficits observed in children with early left-hemisphere injury are far
less pronounced than aphasic syndromes in adults (Lenneberg, 1967; Van Hout, 2000).
Reports of childhood aphasia with favorable and expedient language recovery often mention
the presence of lasting deficits in academic performance (e.g., reading and writing). These
findings suggest the presence of residual deficits in language ability (Aram & Eisele, 1992).
Numerous studies on the effects of early brain lesions have evidenced that the worst time for
injury in the human brain is likely to be the third trimester of pregnancy.
Side of lesion. Numerous studies have been carried out to verify the effects of early left
hemisphere damage (LHD) versus right hemisphere damage (RHD) on language
development. Vargha-Khadem et al. (1992) explored the effects of early brain lesions on the
intellectual development of 82 children with hemiplegic cerebral palsy (42 with LHD and 40
with RHD) compared to a control group. If children with cerebral palsy and epilepsy (30
participants, 14 with LHD and 16 with RHD) were excluded, there were no differences in
intellectual development between control children and children with cerebral palsy. Within
this group, no significant differences were found in VIQ across control children (C-VIQ =
104.3), LHD children (L-VIQ= 101.3) and RHD children (R-VIQ= 103.4). In contrast,
children with cerebral palsy (both the LHD and RHD groups) showed a similar, yet
significantly lower, PIQ than control children (C-PIQ= 109.6; L-PIQ=98.6; R-PIQ= 97.9).
Bates et al. (1999a) discussed the performances of 76 children - presented in previously
published studies - with pre- or perinatal stroke (46 LHD and 30 RHD) tested between 3 and
14 years of age. Mean full-scale IQs were in the normal range. There were absolutely no
Language Disorders and EEG A bnormalities During NREM Sleep 67
differences between LHD and RHD children in full-scale, verbal or nonverbal IQ. The Italian
sample (18 children with LHD and 15 RHD) received several language tests (Peabody Picture
Vocabulary Test, Boston Naming Test, Token Test, Test of Receptive Grammar, Semantic
Category Fluency Test). The brain-injured children performed significantly worse than
normal controls on all language measures. However, no difference was found between LHD
and RHD on any measure. Furthermore, when mental age was controlled for in analyses of
covariance, the differences between brain-injured children and normal controls disappeared,
except for a measure of lexical access (Boston Naming Test). In another study, Bates et al.
(1999b) studied some characteristics of language production (word types, word tokens, mean
length of utterance, grammatical complexity and word errors) based on samples of free
speech, in a group of 38 brain-injured children (24 LHD and 14 RHD) aged 5-8 years,
compared to 38 age- and gender-matched normal controls. There were no differences between
LHD and RHD children on any measure. There were also very few differences between brain-
injured children (combining LHD and RHD) and controls. More recently, Chilosi et al. (2001)
systematically investigated cognitive development and language abilities in a group of 18
children younger than 4 years with focal brain lesions (9 with LHD and 9 with RHD). Their
findings differ from those of other recent studies (Bates et al. 1999a, b; Vicari et al, 2000) in
that they suggest that children with LHD may have greater difficulties in vocabulary and
grammar acquisition. However, the different results obtained by these studies may be
attributed to age-related differences across samples.
Size of lesion. According to Lashley's mass action effect (1929), large lesions are related to
more severe cognitive deficits. However, several studies did not find a strong relationship
between lesion size and language-cognitive outcome (Aram & Eisele 1992; Eisele & Aram
1994; Chilosi et al. 2001). Animal experiments suggest that small lesions have little effect
because they are small; mid-size lesions are large enough to cause permanent behavioral
impairments, but not large enough to enable the brain to re-organize. Compared to small or
medium lesions, large lesions result in a better outcome because participants make a "fresh
start" (Me, 1990).
Presence of epilepsy. There have been reports which demonstrate that seizures accompanying
early brain injury are associated with poorer language and cognitive outcome (Van Dongen &
Loonen, 1977). Vargha-Khadem et al. (1992) showed that a history of seizures is the most
important predictor of cognitive and language impairments in brain-injured children,
regardless of side or size of injury or age of insult. In their study, children with cerebral palsy
(LHD and RHD) and epilepsy presented with a significantly lower cognitive development
than control children and children with cerebral palsy but without epilepsy. Cognitive
impairment in children with cerebral palsy and epilepsy affected both verbal and performance
skills. According to Chilosi et al. (2001) seizures are the single greatest risk factor for
language and cognitive outcome in children with unilateral damage. Chilosi et al. (2001)
suggested that the detrimental effect of epilepsy on the potential for recovery and
development can be probably interpreted as the consequence of a more diffuse neurological

dysfunction. In contrast, according to Vargha-Khadem et al. (1992) it is possible that
anticonvulsant drug therapy administered to all the children with epilepsy rather than seizures
per se may be responsible for these impairments.
In the present study we looked at language development in six children with pre- perinatal
brain lesions associated to important language deficits. We investigated the role of some
classical neurological variables in language development: 1) site and side of the lesion and 2)
presence of epilepsy. Further, we looked at the possible role of some factors whose relevance
has not been sufficiently explored in the literature, namely 3) the presence of a "Landau-
Kleffner Syndrome-Like" and 4) the presence of epileptiform abnormalities in non-REM
(NREM) sleep.
METHODS
Participants
We selected six participants (three boys and three girls) who suffered an early brain damage
before the end of their first year of life after a pregnancy without persistent problems. In all
the participants lesions were localized to specific brain structures on the basis of
neuroimaging procedures. Case 1 had suffered from bacterial meningitis with lesions
localized to the caudate nuclei bilaterally. Case 2 showed a single unilateral focal ischemic
lesion. Case 3 presented with a malformation syndrome affecting the cerebellum. The
remaining three patients (Cases 4-6) had a single unilateral focal ischemic lesion (cf. Table 1).
Neuropsychological assessment
Patients were administered the following standard neuropsychological tests: Standard

Progressive Matrices (Raven, 1954), WPPSI (Wechsler, 1973), WISC-R (Wechsler, 1986),
Stanford- Binet, L-M (1968), Leitner-R (Roid & Miller, 2002), NEPSY battery (Korkman et
al. 1998). Praxic abilities were evaluated by a standardized test for the evaluation of oro-facial
praxis in children (Bearzotti & Fabbro, in press).
Language assessment
Parents of children aged 1-3 years received the Mac Arthur Questionnaire (Caselli & Casadio,
1990). Children aged 2-3 years were administered the Test del Primo Linguaggio battery
(TPL, First Language Test) (Axia, 1995). Children aged 3-6 years received the Test di
Valutazione del Linguaggio battery (TVL, Language Assessment Test) (Cianchetti &
Fancello, 1996) or the Esame del Linguaggio dai 4 ai 12 anni battery (4-12, Language
examination for children aged 4-12) (Fabbro, 1999). Children older than 12 years were
administered the Italian adaptation of the Bilingual Aphasia Test (BAT, Paradis, 1987). A
qualitative analysis of verbal expression was made according to Fabbro & Frau (2001).
Table 1 — Participants' characteristics and clinical data.
Case G HP A Lesion Site VIQ PIQ FIQ Convulsive Therapy

episodes*
1,MD F LH 10 Left caudate nucleus 79 76 75 3rd day Valproate
bilaterally 8 years sodium
2, GM F RH 10 Left upper parietal 70 91 77 7 years Lamotrigine;
region 9 years valproate
sodium;
ACTH one
cycle
3, RP M RH 24 Right cerebellar 54 81 66 7-12 years Carbamazepine
hemisphere
4, LS F LH 7 Left fronto-temporal 92 89 90 6 years Valproate
parietal sodium
5, LP M RH 4 Thalamus and deep 106 -
white matter
bilaterally
6, GM M RH 3 Left frontal cortico- 100 2nd day No
subcortical region
Note: G= gender; HP =hand preference; A= age at most recent assessment; VIQ=verbal intelligence
quotient; PIQ=performance intelligence quotient; FIQ= full-scale intelligence quotient; *= age at
onset.
CASE HISTORIES
Case 1 (MD)
MD is a left-handed girl aged 10 years and four months. Pregnancy was uneventful and she
was born at term (birth weight: 3.4 kg). On her third day of life she had meningitis owing to
group B beta-hemolytic streptococcus complicated by a convulsive seizure (upper limb and
mimic muscle spasms). She was under phenobarbital for a month. At 1 month of life, a brain
CT scan revealed a linear ischemic lesion lateral to the frontal horn and the head of the left
caudate nucleus. She achieved independent walking at 10 months of age and uttered her first
words when she was 12 months old. At 5 years of age she started attending our residential
kindergarten owing to emotional problems, attention difficulties and phonological deficits. On
that occasion, a mild left hemiparesis was noted, mainly affecting the lower limb. A recent
MRI scan evidenced the presence of lesions affecting the head of the caudate nucleus
bilaterally (cf. Figure 1).
Figure 1 — MRI. Lesions affecting the head of the caudate nucleus bilaterally (gross areas of signal
hyperintensity in T2). Possible outcome of inflammatory pathology.
At 6 years of age she underwent a systematic neurological and neuropsychological

evaluation. The EEG in wakefulness and NREM sleep showed a 8-9 Hz fundamental reactive
rhythm with normal morphology. In light sleep conditions spike/slow wave sequences were
found, recorded by bicentral derivations (cf. Figure 2).
Figure 2 — EEG. Diffusion of intraclinical paroxysmal abnormalities to both hemispheres in sleep

stage 2 (ca. 50%). Paroxysmal abnormalities diffuse from the right hemisphere to the contralateral
anterior regions.
On several occasions, other left parietal paroxysmal abnormalities were found. Cognitive
development was in the low normal range (VIQ= 78; PIQ= 73; FIQ= 73). On the 4 - 1 2
language battery (Fabbro, 1999) she performed 2 SDs below the norm on syntactic and
grammatical comprehension tasks, naming, sentence repetition and semantic fluency. Speech
was fragmentary and little contextual. On the neuropsychological evaluation with the NEPSY
battery (Korkman el al. 1998), which allows an in-depth analysis of executive, sensorimotor,
visuoperceptual and mnestic functions, she performed 1 SD below the mean on tests targeting
attention and executive functions, and 2 SDs below the mean on tests targeting sensorimotor,
visuo-spatial and memory functions. Neuropsychological rehabilitation and speech therapy
were started. As her response to treatment was insufficient and paroxysmal anomalies in sleep
persisted until she was 7 years old, she was started on valproate sodium (400 mg/day). Five
months thereafter her EEG tracing had considerably improved. However, right focal temporal
paroxysmal abnormalities were found. After a few months of therapy, her language abilities
had improved. At the age of 8 years, when she received a new language assessment, her
performance on grammatical comprehension and sentence repetition was still 2 SDs below the
norm. Her expression was fragmented and characterized by pragmatic deficits. Her
intellectual development was in the low normal range (VIQ = 79; PIQ = 76; FIQ = 75) (cf.
Table 1).
Table 2 — EEG abnormalities and type of language deficits.
Wa E
Case NREM sleep EEG anguage Language deficits
EEG Regression
Paroxysmal activity > One episode Receptive and expressive

1, MD Normal
50% NREM sleep at 8 years of age morphosyntax.
Spike - Diffuse paroxysmal Two episodes at 7 Expressive morphosyntax,

2, GM
waves activity and 9 years of age pragmatics.
Diffuse paroxysmal Between 7 and 12 Transcortical motor
3, RP Spike-waves . . , \
activity before 14 years years aphasia.
Subcontinuous
Receptive and expressive
4, LS Spike-waves paroxysmal activity > No episodes
morphosyntax. Anomias.
5, LP Normal Paroxysmal activity No episodes Verbal dyspraxia

Receptive and expressive
6, GM Normal Paroxysmal activity No episodes
morphosyntax.
At 8 years and 10 months she had a critical episode with loss of consciousness and right
eye lateroversion, followed by an aggravation of language deficits. As her clinical picture
which was typical of Landau-Kleffner Syndrome - with aphasic regression, seizures,
paroxysmal EEG during NREM sleep - was associated to a documented neurological lesion,
we defined it Landau-Kleffner Syndrome-Like. Ethosuximide was introduced to keep seizures
under control. The sleep EEG still revealed focal paroxysmal abnormalities in the right frontal
region, with mainly homolateral diffusion. Administration of the NEPSY battery showed a
clear aggravation of memory skills compared to the previous evaluation (from 67% correct to
54%). On the language examination her language deficits had worsened. She performed 2
SDs below the norm on the following tasks: verbal discrimination, syntactic comprehension,
grammatical comprehension, sentence repetition and semantic fluency (cf. Table 2).
Case 2 (GM)
GM is a right-handed girl aged 10 years and 8 months. Pregnancy was uneventful with normal
delivery (low birth weight: 2.4 kg). At 10 days of life she suffered an episode of low platelet
count, with bruises on her face, neck and trunk. At 1 year of life a left hemisyndrome was
observed. This neurological picture was attributed to perinatal brain suffering. The child
received kinesitherapy and speech therapy to improve her motor skills and pragmatic
problems - poor relationships with adults and peers - as well as phonological and lexical
deficits. At 4 years of age her IQ was in the normal range (WIPPSI: VIQ= 87; PIQ= 95; FIQ=
90). Her speech was little intelligible and she still had relational difficulties with adults. At 6
years and a half her intellectual development had improved (WISC-R: VIQ= 96; PIQ= 100;
FIQ= 97). She showed semantic and grammatical comprehension deficits, phonological
deficits, reading and writing disabilities.
First episode ofLKS. At 7 years and a half she suffered two generalized tonic-clonic seizures.
The EEG in wakefulness evidenced spike-wave paroxysmal activity in the fronto-temporal
region bilaterally, mainly prevalent on the left. She received carbamazepine (300 mg/day).
Progressively the child started to show language regression: her pragmatic problems
aggravated and she had difficulties in establishing relationships with others, and avoided eye
contact. Her speech was unintelligible. She uttered single words, with many anomias and
perseverations. To communicate, she resorted to pointing or drawing. She read very slowly.
On the WISC-R she showed increased difficulties (VIQ= 87; PIQ= 83; FIQ= 83). Her clinical
picture, which was typical of Landau-Kleffner Syndrome - aphasic regression, seizures,
paroxysmal EEG during NREM sleep - was associated to a documented neurological lesion.
Thus, we defined it Landau-Kleffner Syndrome-Like. Slowly she showed a progressive
language recovery. She was started on Lamotrigine (125 mg/day). At the age of 8.06 years
she underwent another neurological examination. MRI imaging and CT scan evidenced an
egg-like formation, 2.5 cm in diameter, in the left parietal region near the vertex and the
midline (possible outcome of prenatal lesion) (see Figure 3).
Language Disorders and EEG Abnormalities During NREMSleep 73
Figure 3 — MRI. Egg-like lesion due to an arachnoidal cyst possibly localized in the left upper
parietal region, with likely areas of cortical dysplasia (coronal section, T2).
An EEG in wakefulness evidenced a symmetric and reactive 10 Hz background rhythm. On

the frontal areas, isolated spikes and waves were noted, mainly prevalent on the left. Focal
paroxysmal abnormalities were found (triphasic spikes followed by slow waves) in the right
centro-temporal region throughout all sleep stages. Epileptiform abnormalities were repetitive
and more numerous in NREM sleep (see Figure 4).
Figure 4 — EEG: Left centro-parietal paroxysmal slow focus in wakefulness and drowsiness, with
immediate diffusion and generalization of paroxysms upon falling asleep.
Second episode of LKS. At the age of 9 years and a half seizures during sleep appeared again
(eye opening, mouth clonic spasms, scialorrhea, clonic spasms of the lower limbs) with abrupt
language regression. Her verbal comprehension was minimal and spontaneous speech was
absent. Only upon request did she produce single words. An important regression in IQ was
noted (WISC-R: VIQ= 70; PIQ= 91; FIQ= 77) (cf. Table 1). On the EEG in wakefulness the
spike-slow wave complexes were more marked in the left anterior areas. This activity was not
modified in sleep. ACTH therapy was started (Synacthen Depot 1, 1 via IM) with
administration of 6 doses. Valproate sodium (600 mg/day) was introduced in the AE therapy.
Progressively she showed language recovery. On the 4-12 language battery (Fabbro, 1999),
she showed full recovery of sentence repetition and fluency. Verbal fluency, MLU and
Type/Token Ratio of descriptive speech were age-appropriate. Omissions of free grammatical
morphemes and phonemic paraphasias were still present (cf. The Bird Nest Story Picture
Description Task; Paradis, 1987) (cf. Table 2).
Case 3 (RP)
RP is a right-handed young adult male aged 24 years. Pregnancy and labor were uneventful
(birth weight: 3 kg). At 2 years he was examined by a neurologist owing to severe motor and
linguistic deficits - independent walking was reached only at age 2 years when the child
produced only few words. The neurological examination revealed cerebellar problems and the
child was referred for outpatient rehabilitation. A CT scan evidenced a small poroencephalic
area in the right cerebellar lobe, with signs of olivo-ponto-cerebellar atrophy.
At 7 years he had seizures during sleep. Ictal episodes lasting about 1 minute were
followed by a transient disorder of language expression. In the following years he continued
to suffer from abrupt regression of language skills, receptive and expressive alike. Sometimes,
these episodes were associated to seizures. As his clinical picture which was typical of
Landau-Kleffner Syndrome - aphasic regression, seizures, paroxysmal EEG during NREM
sleep - was associated to a documented neurological lesion, we defined it Landau-Kleffner
Syndrome-Like. His IQ was in the normal range (Stanford - Binet Scale: IQ=93). The
neurological examination revealed phonological and morphosyntactic deficits. His expression
was characterized by many phonemic paraphasias and telegraphic style. The EEG in
wakefulness showed paroxysmal activity characterized by bi- and triphasic spikes, followed
by slow waves, in the left central and temporal regions. He was started on Carbamazepine (1 +
l
A cp/day).
At 8 years the child started to attend primary school at our Rehabilitation Center. He was
followed for epilepsy-related neurological problems and received speech therapy and
neuropsychological therapy for 12 years. At the age of 10 years his IQ was in the low normal
range (WISC-R: VIQ = 54; PIQ = 81; FIQ = 66) (cf. Table 1). Language was characterized by
deficits that are typical of acquired aphasia in children, with many phonological, lexical
access and morphosyntactic deficits. He showed clumsiness when performing fine distal
movements, intentional tremor and asynergy bilaterally. At the age of 12 years he suffered
from seizures in wakefulness, followed by transient aphasia. He also had many seizures in
sleep. At the age of 14 years the AE therapy (Carbamazepine: 700 mg/day) was discontinued.
He no longer had seizures. Until the age of 14 years he suffered from paroxysmal
abnormalities mainly in the right centro-parietal region, which became accentuated in NREM
sleep. They were also present in the contralateral hemisphere. Afterwards, the EEG
normalized. Despite the fact that he no longer suffered from seizures and paroxysmal
abnormalities, as shown by the EEG in sleep and wakefulness, his language was still
pathological, with symptoms that are typical of transcortical motor aphasia (cf. Table 2).
At the age of 24 years he received a detailed neuropsychological assessment and a
language evaluation. On Raven's Standard Progressive Matrices his IQ was in the low normal
range (Raven: IQ= 80). On the NEPSY his visuospatial skills were sufficiently developed,
while he showed sensorimotor deficits (54% correct), attentional and executive deficits (79%)
and memory deficits (88%). Language skills were assessed by the Italian version of the BAT
(Paradis, 1987). The BAT comes with no normative values. However, to assess the extent of
language deficits, we made reference to the values obtained from a control group, described in
Fabbro et al. (2004). The morphological and syntactic language levels were most impaired
(morphology = 39% of correct answers, <2SD below the mean; lexicon = 65%, <2SD), while
the most severely compromised language tasks were propositional skills (48%, <2SD), lexical
access (74%, <2SD) and comprehension (75%, <2SD).
Case 4 (LS)
LS is a left-handed girl aged 7 years and 3 months. A risk of miscarriage at the fourth month
was evidenced. The child was born at term (birth weight: 2.85 kg). As she tended to always
keep her head turned to the left, her mother consulted a neurologist.
Figure 5 — MRI. Large poroencephalic lesion affecting the temporal, parietal and frontal lobes, the
thalamus and the basal ganglia (caudate nucleus, globus pallidus) of the left hemisphere (T2). The
lesion affects over 40% of the left hemisphere, which is distinctly hypotrophic. Possible outcome of
prenatal ischemia.
A diagnosis of right hemiparesis, with more marked hypertonia of the lower limb (age: 5
months) was made. At the age of 6 months she started motor rehabilitation. The EEG in
wakefulness at 18 months showed diphasic spikes - either isolated or in short sequences - over
the left parieto-occipital derivations that tended neither to diffusion nor to generalization. At 6
years of age, an MRI evidenced an extended lesion to the left fronto-temporo-parietal lobe
due to prenatal brain ischemia (cf. Figure 5). EEGs in wakefulness and Holter sleep were
performed. The EEG in wakefulness was characterized by 9 Hz background activity.
Paroxysmal discharges in the form of spikes and waves at 3 c/s were more diffuse in the
anterior regions. Upon falling asleep and during sleep, subcontinuous paroxysmal activity at
2.5-3 c/s in the form of diffuse spike-wave complexes was evident (cf. Figure 6).
Figure 6— EEG: Intraclinical continuous paroxysmal activity (>75%) with 1-2 Hz spike-slow wave
complexes in sleep stage 3 (CSWS).
Her IQ was in the normal range (WPPSI at age 5.06: VIQ= 92; PIQ= 89; FIQ= 90) (cf.
Table 1). At the age of 6 she received a systematic neuropsychological and neurolinguistic
assessment. On the neuropsychological evaluation by NEPSY (Korkman et al, 1998) she
performed 1 or 2 SDs below the mean on all domains (attention, sensorimotor tasks, memory
and learning, visuo-spatial tasks). The neurolinguistic assessment was made at 5 years and 11
months by the 4 -12 language battery (Fabbro, 1999). The child made significant errors on
grammatical comprehension, word repetition and lexical fluency. Word comprehension was
between 1 and 2 SDs below the norm. Her verbal expression was characterized by word-
finding difficulties, many morphosyntactic errors (substitution and addition of free
inflectional morphemes and substitution of bound morphemes) and phonemic paraphasias.
Once her neuropsychological and neurolinguistic deficits had been established, she
received neuropsychological rehabilitation and speech therapy. In the evenings she also
received valproate sodium which reduced her paroxysmal activity during sleep to a very small
extent. She was also administered clobazam. After a few days the child showed short partial
epileptic fits, which was unprecedented. The therapy was changed to control seizures
(valproate sodium and ethosuximide). A language assessment was repeated at age 6 years and
8 months. A mild improvement in semantic comprehension, sentence repetition and semantic
fluency was observed. A recent language assessment (7 years and 3 months) showed an
improvement in verb naming. The patient also mildly improved in grammatical
comprehension (up to 2 SDs below the mean). Semantic fluency was still between 1 and 2
SDs as in the previous assessment. In an analysis of descriptive speech, verbal fluency (words
per minute) was between 1 and 2 SDs, while MLU and Type/Token Ratio were normal.
Anomias were present. While these results may be indicative of some grammatical recovery,
lexical difficulties still persist (cf. Table 2).
Case 5 (LP)
LP is a right-handed boy aged 4 years and 9 months. He was born at term after uneventful
pregnancy (birth weight: 3.4 kg). At 4 months of age, he was diagnosed as having insufficient
head control, axial hypotonia and rigidity of the upper and lower limbs. He started
rehabilitation at 6 months of age. He reached independent sitting at 13 months of age,
standing at 18 months, and walking at 21 months of age. At 4 years of age he underwent
clinical and neuropsychological evaluations. The neurological examination revealed a
dyskinetic tetraparesis, with greater impairment of the left side. Manipulative functions were
limited because of dystonias. An MRI evidenced areas of signal alteration in the thalamus and
the deep white matter bilaterally in the central, posterior and rolandic cortical-subcortical
regions (cf. Figure 7), with lesions probably due to prenatal brain ischemia.
Figure 7— MRI. Lesions affecting the thalamus bilaterally and the periventricular white matter (area
of signal hyperintensity in T2) probably due to a prenatal ischemic event.
EEG scans in wakefulness and sleep were run at the age of 3 and 4 years. The EEG in
wakefulness was characterized by unstable background activity. During sleep sharp waves
were followed by slow medium-voltage waves over the left centro-temporal regions and
asynchronous waves in the left central regions. During protracted sleep - up to NREM sleep
stages 2 - a marked diffusion of epileptiform abnormalities was noted (cf. Figure 8).
Paroxysmal abnormalities disappeared upon awakening. His IQ was in the normal range
(Leiter-R, IQ= 106) (cf. Table 1). The child presented with scialorrhea and dyspraxia
affecting the mouth, tongue and face (Bearzotti & Fabbro, in press). His verbal expression
was dysarthric. Lexicon was limited to 10 words and CV+ CV strings. He could produce all
vowels and semivowels, but only a few consonants (Iml, Ibl, Id/, Ik/). Word and sentence
comprehension were normal (Cianchetti & Fancello, 1996). Naming was 1 SD below the
norm, while sentence repetition was 2 SDs below the norm (cf. Table 2).
Figure 8 — EEG: Left focal spike activity. Spikes are more diffuse over the right hemisphere. A
tendency was found to stage 2 paroxysmal synchronization.
Case 6 (GM)
GM is a right-handed boy aged 3 years and 4 months. Pregnancy was uneventful and he was
born at term (birth weight: 3.5 kg). As fetal and neonatal suffering was observed, he was
admitted to Intensive Care Unit. Twenty hours after birth he was taken to Nursery again. At
43 hours he suffered from clonic spasms affecting the right upper limb. The EEG tracing
showed a marked depression in the left hemisphere, with convulsive activity starting from the
left frontal lobe. He received phenobarbital (5 mg twice a day) for a month. AE therapy was
thus discontinued and since then he no longer had seizures. Brain CT scan on his third day of
life evidenced a cortico-subcortical vascular ischemic lesion in the left lateral frontal region.
Another CT scan at age 1 month confirmed the extent of the ischemic lesion (cf. Figure 9).
Motor and intellectual development was normal. He reached independent walking at 10
months of age. When he was 3 years old, he was referred to our Center for a suspected
expressive language disorder. The neurological examination was normal. His nonverbal IQ
was in the normal range as assessed by Stanford - Binet Scale (cf. Table 1). A marked
attention deficit was present, with hyperactive behavior. EEGs in wakefulness and sleep were
performed at age 2 years and 10 months. The EEG in wakefulness evidenced a reactive 8.5
Hz fundamental rhythm. Paroxysmal discharges with large sharp spike-waves and atypical 4
Hz spikes-waves with maximal bicentral expression were present in the first sleep stages
(stages 1 and 2). Reactivity was normal on awakening (Figure 10).
Figure 9 — CT Scan: cortico-subcortical lesion in the left rolandic and frontal area (4-6 cm in size)
possibly due to an ischemic event.
Figure JO — EEG: Generalized paroxysmal discharges with 3.5-4 Hz spike-wave activity correlated
to distal myoclonias in sleep stage 2 (<10%).
Language examination showed a marked impairment in word comprehension (TPL Test,

Axia, 1995) which was 2 SDs below the norm, and on word production (28 words, normal
range > 300 words) on the Mac Arthur Scale (Caselli et al. 1990). He produced all the vowels
but only some consonants (/p/, Pol, III, Id/, Ikl, Iml, In/, Its/) (cf. Table 2).
DISCUSSION
All the patients showed an early brain lesion localized in different brain structures (mainly
cortical areas in cases 2, 4, 6; subcortical in cases 1 and 5 and cerebellar in case 3) and a
severe language disorder. Many studies have shown that early brain lesions do not affect
intellectual development and language acquisition in the absence of epilepsy (Vargha-
Khadem et al, 1992; Ballantyne & Trauner, 1999; Bates et al, 1999a,b). In our study three
cases had seizures (cases 1-3), while the remaining cases showed defective language
development without seizures. In all of them we found the presence of a high proportion of
epileptiform abnormalities in NREM sleep.
The first three cases did not only have epilepsy but showed a Landau-Kleffner Syndrome-
Like (LKS-L), that is a clinical picture of acquired aphasia associated to epilepsy in the
presence of a neurological lesion. Acquired aphasia with convulsive disorder is an acquired
language disorder associated with convulsive disorders and EEG abnormalities in children
(Landau & Kleffner, 1957). Language disruption may occur before or after seizures. At onset
the most frequent symptom is breakdown in language comprehension, while hearing and
interpretation of non-linguistic sounds remain intact (word deafness). Following the
breakdown in comprehension, expressive language and vocabulary decay progressively, too
(Dugas et al, 1995). There may be (almost) full recovery following the first episode even
after a short period of time (days or weeks). In other cases, recovery may be very slow
(months or years) (cf. cases 1 and 2). Relapse is frequent and often the development of
aphasia is fluctuating, with several aphasic episodes (cases 2 and 3). The clinical picture
stabilizes before the end of adolescence with complete disappearance of seizures before the
age of 15. Language recovery is sometimes very good, at other times defective. When
defective, participants present with aphasic disorders for the rest of their lives (cf. case 3)
(Mantovani & Landau, 1981). Despite the many studies and investigations carried out on LKS
its cause is still unknown (Landau, 1992). Neuroradiological investigations have revealed that
children with "classical" LKS do not exhibit quiescent or evolving cerebral lesions.
An extremely significant feature which seems to be present in all children with LKS and
which is revealed by the sleep EEG is a subcontinuous paroxysmal activity during NREM
Sleep (ESES) (Tassinari et al, 1985a, Fabbro & Zucca, 2000). This bilateral (focalized)
paroxysmal dysfunction during slow sleep is constantly associated with regression of
language functions (Hirsh et al, 1990). Ravnik (1984) maintains there is an interdependence
relationship between EEG abnormalities in sleep and verbal behavior. He documented
recovery of verbal expression in a child with global aphasia and LKS immediately after
intravenous injection of diazepam. The most frequent treatment is drug therapy.
Phenobarbital, carbamazepine and phenytoin were either ineffective or even aggravative (cf.
cases 1 and 3). Treatment with valproate sodium, ethosuximide, clobazam appeared to be
partially or transiently effective, whereas the most encouraging results were obtained with
corticosteroids (sleep-modifying drugs) (Marescaux et al, 1990). It has been observed that
some cycles with cortisone (hydrocortisone) tend to reduce the frequency of seizures with
paroxysmal abnormalities disappearing from tracings of both wake and sleep EEG. Often, an
improvement in language functions (cf. case 2) is evident, which seems to be associated with
a reduction or disappearance of paroxysmal abnormalities in sleep instead of seizures. Case 3
developed a language disorder, epilepsy and LKS-Like in the presence of a right cerebellar
malformation lesion. The association between posterior fossa malformation and focal epilepsy
has rarely been reported in the literature (Ducan el al, 1990; Parmeggiani et al, 1999), while
the association between LKS and right cerebellar lesion has never been reported, at least to
our knowledge. In contrast, the association between right cerebellar lesions and acquired or
developmental language disorders has already been established (cf. Fabbro, 2000; Fabbro et
al, 2000a). Cases 4, 5 and 6 presented a highly defective language development in the
absence of epilepsy, though case 4 suffered from seizures from age 6 years onwards. In these
cases too, paroxysmal abnormalities were present in NREM sleep. This pathological hallmark
was also found in many children with severe Developmental Language Disorders (DLDs) —
also known as Specific Language Impairment (SLI) or Developmental Dysphasia (cf. Bishop,
1997; Leonard, 1998). As early as the late 1960s it was reported that a large number of
children with DLDs presented epileptiform EEG abnormalities without seizures (Forrest et
al., 1967; Eisenson, 1968). These EEG abnormalities tended to be activated by drowsiness
and sleep, at times occurring before the age of 3 and never after the age of 16 (Blom et al.,
1972). Fois et al. (1968) discussed the issue of neuropsychological disorders following focal
epileptic discharges in children without seizures. In some cases, they proposed an anti-
epileptic therapy. Often, this therapy reduced hyperactivity, improved the attention span and
school performance, and eliminated vegetative and sleep disturbances.
Maccario et al. (1972) described 7 children with a severe picture of DLD associated with
epileptiform EEG abnormalities without convulsive seizures. The paroxysmal EEG
abnormalities comprised sharp waves and spike-wave discharges in focal fashion, or bilateral
synchronous bursts that were present in wakefulness and tended to intensify in slow wave
sleep. The authors stressed the differences between this clinical picture and LKS. In their
patients: 1) language impairment had been present from the start, and 2) no patient presented
with clinical epilepsy. Echenne and co-workers made two EEG studies of children with DLD.
In the former (Echenne et al, 1992) they studied 19 children with developmental dysphasia
with no previous history of epileptic seizures. Each child received a diurnal prolonged EEG
and a continuous overnight polygraph recording. Four children out of 19 presented
paroxysmal EEG abnormalities (PA) in diurnal EEG and 17 children out of 19 presented PA
in nocturnal EEG. These abnormalities clearly increased during NREM sleep. In a subsequent
study, Echenne (1996) studied 76 children with developmental dysphasia without convulsive
seizures. EEG paroxysmal abnormalities were found in 20% of the children with expressive
disorders and in 63.5% of children with comprehension difficulties.

Picard et al. (1998) studied the diurnal and nocturnal EEGs of 52 children with DLD (40
boys and 12 girls), aged 4 to 11 years (mean age = 9 years). All children received a
continuous overnight polygraph recording. In 46 out of 52 children with DLD the standard
EEG in wakefulness was normal. In the control group, all children showed a standard EEG in
the norm. In the nocturnal EEG paroxysmal activity during slow sleep was found in 50% of
the participants with DLD. Recently, we followed 50 children with DLD classified according
to ICD-10 criteria (Fabbro et al. 2000b). These children were administered: A) a standard
EEG, and B) an EEG in afternoon sleep. Four children out of 50 (8%) presented with
epileptiform abnormalities in the wake EEG. Twenty-eight children out of 50 (56%)
presented with paroxysmal abnormalities in the EEG during afternoon sleep after partial
deprivation of nocturnal sleep. According to Picard et al. (1998), paroxysmal abnormalities
are responsible for developmental language disorders as interictal paroxysms occurring in
epilepsy with rolandic paroxysms are responsible for transient cognitive problems, as
suggested by Aarts et al. (1984) and Binnie (1993). Another explanation is provided by recent
studies on memory and sleep. Slow sleep could be involved in structural modifications
underlying long-term memory processes (such as protein synthesis, cf. Jones, 1998), which
would be hampered by paroxysmal abnormalities. The presence of epileptiform abnormalities
in NREM sleep is therefore a key clinical predictive factor of cognitive and/or linguistic delay
in children with an early brain lesion. For this reason, it should be continuously monitored in
these children. Paroxysmal abnormalities in NREM sleep in children with early brain lesions,
children with LKS and children with developmental language disorders suggest the presence
of common physiopathological mechanisms across the three clinical conditions. Recent
studies with advanced neuroimaging techniques have shown that the majority of children with
severe DLDs have malformative lesions of the cerebral cortex (polymicrogyria, PMG) and the
parietal and perisylvian regions (Guerreiro et al., 2002). Furthermore, it is known that
perisylvian PMG is frequently associated to epilepsy (Guerrini, 1999).
REFERENCES
Aarts, J. H., C. D. Binnie, A. M. Smith and A. J. Wilkins (1984). Selective cognitive

impairment during focal and generalized epileptiform EEG activity. Brain, 107, 293-
308.
Aram, D. M. and J. A. Eisele. (1992). Plasticity and recovery of higher cognitive functions
following early brain injury. In: Handbook of Neuropsychology (F. Boiler and J.
Grafman, eds.), Vol. 6, pp. 73-92. Elsevier, Amsterdam.
Axia, G. (1995). Test delprimo linguaggio. Organizzazioni Speciali, Firenze.
Bates, E. S., S. Vicari and D. Trauner (1999a). Neural mediation of language development:
Perspectives from lesion studies of infants and children. In: Neurodevelopmental
Disorders (H. Tager-Flusberg, ed.), pp. 533-81. The MIT Press: Cambridge (MA).
Bates, E. S., B. Wulfeck, M. Opie, J. Fenson, S. Kriz, J. Reilly, N. Dronkers. L. Miller, R.
Language Disorders andEEG Abnormalities During NREMSleep 83
Jeffries and K. Herbst (1999b). Comparing free speech in children and adults with left
vs. right-hemisphere injury. Brain Lang, 69, 377-379.
Bearzotti, F. and F. Fabbro. II test per la valutazione delle prassie orofacciali nel bambino.
Giornale Ltaliano di Neuropsichiatria, in press.
Binnie, C. D. (1993). Significance and management of transitory cognitive impairment due to
subclinical EEG discharges in children. Brain Dev, 15, 39.
Bishop, D. V. M. (1997). Uncommon Understanding. Development and Disorders of
Language Comprehension in Children. Psychology Press, Hove.
Blom, S., J. Heijbel and P. G. Bergfors (1972). Benign epilepsy of children with centro-
temporal EEG-foci. Prevalence and follow-up study of 40 patients. Epilepsia, 13, 609-
619.
Brett, E. M. (1983). Pediatric Neurology. Churchill Livingstone, Edinburgh.
Caselli M. C. and P. Casadio (1995). // primo vocabolario del bambino. Franco Angeli,
Milano.
Chilosi, A.M., P. Cipriani, B. Bertuccelli, L. Pfanner and G. Cioni (2001). Early Cognitive
and Comunicative Development in Children with Focal Brain Lesions. J Child Neurol,
16,309-316.
Cianchetti, C. and S. Fancello (1996). Test di valutazione del linguaggio. Erikson, Trento.
Cytowic, R. E. (1996). The Neurological Side of Neuropsychology. The MIT Press,
Cambridge (MA).
Ducan, R., J. Patterson, D. M. Hadley and I. Bone (1990). Unilateral cerebellar damage in
focal epilepsy. JNeurol, Neurosurg, Psychiatr, 53, 436-442.
Dugas, M., S. Franc, Gerard C. Loi'c and M. Lecendreux (1995). Evolution of acquired
epileptic aphasia with or without continuous spikes and waves during slow sleep. In:
Continuous Spikes and Waves during Slow Sleep (A. Beaumanoir, M. Bureau, T.
Deonna, L. Mira and C. A. Tassinari, eds.), pp. 47-55. John Libbey, London.
Echenne, B., R. Cheminal, F. Rivier, C. Negre, J. Touchon, M. Billiard (1992). Epileptic
electroencephalographic abnormalities and developmental disphasias: a study of 32
patients. Brain Dev, 14, 216-235.
Echenne, B. (1997). Physiopathologies des troubles specifiques du developpement du langage
parle. Approche Neuropsychologique des Apprentissages chez I 'Enfant, 42, 52-55.
Eisele, J. and D. Aram (1995). Lexical and grammatical development in children with early
hemisphere damage: A cross-sectional view from birth to adolescence. In: The
Handbook of Child Language (P. Fletcher and B. MacWhinney, eds), pp. 664-689.
Blackwell, Oxford.
Eisenson, J. (1968). Developmental aphasia: a postulation of a unitary concept of the disorder.
Cortex, 4, 184-200.
Fabbro, F. (1999). Neurolinguistica e neuropsicologia dei disturbi specifici del linguaggio nel
bambino: proposta di un esame del linguaggio. Saggi, 24, 11-23.
Fabbro, F. (2000). Introduction to language and cerebellum. Journal of Neurolinguistics, 13,
83-94.
Fabbro, F. and G. Frau (2001). Manifestations of aphasia in Friulian. Journal of

Fabbro, F., and C. Zucca (2000). Acquired neuropsychological disorders in children with
epilepsy. Saggi, 25, 23-29.
Fabbro, F., R. Moretti and A. Bava (2000a). Language impairments in patients with cerebellar
lesions. Journal of Neurolinguistics, 13, 173-88.
Fabbro, F., C. Zucca, M. Molteni and R. Borgatti (2000b). EEG abnormalities during slow
sleep in children with developmental language disorders. Saggi, 25, 41-8.
Fabbro, F., A. Tavano, S. Corti, N. Bresolin, P. De Fabritiis and R. Borgatti (2004). Long-
term neuropsychological deficits after cerebellar infarctions in two young adult twins.
Neuropsychologia,42: 536-545.
Fois, A., S. Borgheresi and E. Luti (1968). Clinical correlates of focal epileptic discharges in
children without seizures. A study of 110 cases. Helvetica Pediatrica Acta, 23, 257-
265.
Forrest, T., J. Eisenson and J. Stark (1967). EEG findings in 113 non verbal children.
Electroencephalogr Clin Neurophysiol, 22, 291.
Guerreiro, M. M., S. R. Hage, C. A. Guimaraes, D. V. Abramides, W. Fernandes, P. S.
Pacheco, A. M. Piovesana, M. A. Montenegrol and F. Cendes (2002). Developmental
language disorders associated with polymicrogyria. Neurology, 59, 245-250.
Guerrini, R. (1999). Polymicrogyria and epilepsy. In: Abnormal cortical development and
epilepsy (R. Spreafico, G. Avanzini and F. Andermann, eds.), pp. 191-201. J. Libbey
& C, London.
Hirsch, E., C. Marescaux, P. Maquet, M. N. Metz-Lutz, M. Kiesmann, E. Salmon, G. Franck
and D. Kurtz (1990). Landau-Kleffner Syndrome: a clinical and EEG study of five
cases. Epilepsia, 31, 756-767'.
Jones, B. E. (1998). The neural basis of consciousness across the sleep-waking cycle. Adv
Neurol, 77, 75-94.
Me, E. (1990). An analysis of the correlation of lesion size, localization and behavioral effects
in 283 published studies of cortical and subcortical lesions in old-world monkeys.
Brain Research Review, 15, 181-213.
Korkman, M, U. Kirk and S. Kemp (1998). NEPSY. A developmental Neuropsychological
Assessment. The Psychological Corporation, San Antonio.
Landau, W. M. (1992). Landau-Kleffner Syndrome: an eponymic badge of ignorance. Arch
Neurol, 49,353.
Landau, W.M. and F. R. Kleffner (1957). Syndrome of acquired aphasia with convulsive
Lashley, K. S. (1929). Brain Mechanisms and Intelligence. University of Chicago Press,
Chicago.
Lenneberg, E. H. (1967). Biological Foundations of Language. Wiley, New York.
Leonard, L. B. (1998). Children with Specific Language Impairment. The MIT Press,
Cambridge (MA).
Maccario, M, S. J. Hefferen, S. J. Keblusek and K. A Lipinski KA (1982). Developmental

dysphasia and electroencephalographic abnormalities. Dev Med Child NeuroL 24, 141-
155.
Course and prognosis. Neurology, 30, 524-529.
Marescaux, C, E. Hirsch, S. Finck, P. Maquet, E. Schlumberger, F. Sellal, M. N. Metz-Lutz,
Y. Alembik, E. Salmon and G. Franck (1990). Landau-Kleffner Syndrome: a
pharmacological study of five cases. Epilepsia, 31. 768-777.
Paradis, M. (1987). The Assessment of Bilingual Aphasia. LEA, Hillsdale (NJ).
Parmeggiani, A.. A. Posar, M. C. Scaduto, S. Chiodo. M. Santucci and P. Giovanardi Rossi
(1999). Posterior fossa malformations and epilepsy. J Child Neurol, 14, 113-117.
Picard, A., F. Cheliout Heraut, M. Bouskraoui, M. Lemoine, P. Lacert and J. Delattre (1998).
Sleep EEG and developmental dysphasia. Dev Med Child Neurol, 40, 595-599.
Raven, J. C. (1954). Standard Progressive Matrices. The Psychological Corporation, New
York.
Ravnik, I. (1984). Un cas de syndrome de Landau-Kleffner: effet du Diazepam intraveineux.
In: Les syndromes epiletiques de I 'enfant et de I'adolescent (J. Roger, C. Dravet, M.
Bureau, F. E. Dreifuss, P. Wolf, eds.), pp. 196-7. John Libbey : London.
Roid, G. H. and L. J. Miller (2002). Leitner International Performance Scale ~ R.
Organizzazioni Speciali, Firenze.
Tassinari, C. A., M. Bureau, C. Dravet, B. Dalla Bernardina and J. Roger (1985). Epilepsy
with continous spikes and waves during slow sleep, otherwise described as ESES. In:
Epileptic Syndromes in Infancy, Chilhood and Adolescence (J. Roger, C. Draver, M.
Bureau, F. E. Dreifuss and P. Wolf P, eds.), pp. 194-204. John Libbey, London.
Terman, L. M. and M. A. Merrill (1968). Scala di Intelligenza Stanford - Binet, forma L-M.
Tuchman, R. and I. Rapin (2002). Epilepsy in autism. Lancet. Neurology, 1, 352-358.
Van Dongen, H. R. and M. C. B. Loonen (1977). Factors related to prognosis of acquired
aphasia in children. Cortex; 13, 131-136.
Van Hout, A. (2000). An Outline of Acquired Aphasia in Children. Saggi, 26, 13-21.
Vargha-Khadem, F., E. Isaacs, S. Van der Werf, S. Robb and J. Wilson (1992). Development
of intelligence and memory in children with hemiplegic cerebral palsy. Brain, 115,
315-329.
Vicari, S., A. Albertoni, A. M. Chilosi, P. Cipriani, G. Cioni and E. Bates (2000). Plasticity
and reorganization during language development in children with early brain injury.
Cortex, 36, 31-46.
Wechsler, D. (1973). WPPSI. Scala Wechsler a livello prescolare e di scuola elementare.
Wechsler, D. (1986). WISC-R. Scala di Intelligenza Wechsler per Bambini Riveduta.
Language in Children Treated for Posterior Fossa Tumor 87
LANGUAGE AND PHONOLOGICAL AWARENESS

ABILITIES OF CHILDREN TREATED FOR
POSTERIOR FOSSA TUMOR
Bruce, E. Murdoch, Kimberley M. Docking, and Elizabeth C. Ward

The University of Queensland, Australia
Abstract — The general and high-level language and phonological awareness abilities of twelve
children treated for posterior fossa tumor were examined in the context of recent theories implicating
the cerebellum in language function, and treatment effects. At a group level, twelve children exhibited
reduced performance in comparison to their individually matched peers on receptive and expressive
general language measures. Across high-level language parameters, a group of ten of the twelve
children with posterior fossa tumor demonstrated comparatively poor skills. At an individual case
analysis level, three of twelve children demonstrated specific general language deficits, particularly in
the areas of expressive language and syntax. However, three often children individually exhibited
global impairments in high-level language, with an additional three demonstrating specific
disturbances to problems solving, or lexical knowledge. Key contributing factors discussed included
the impact of direct involvement of cerebellar structures, fourth ventricle location and associated
hydrocephalus, tumor type, treatment effects, and young age at diagnosis. While all children with
identified general language deficits also demonstrated severe global high-level impairments, findings
indicated that high-level language disturbances may also occur in the presence of intact general
language skills.
Key words: posterior fossa, cerebellum, brain, tumor, children, language.

INTRODUCTION
The posterior fossa accounts for just one-tenth of intracranial volume and yet it is this location
that is responsible for over half of the brain tumors occurring in children (Cohen et al, 1982).
Specific signs and symptoms relating to posterior fossa tumor often include signs of increased
intracranial pressure due to obstruction of the cerebrospinal fluid pathways, or cerebellar
signs which can include gait disturbances, nystagmus, dysarthria, dysmetria, and hypotonia
(Heideman et al, 1993). Commonly occurring posterior fossa tumors include
medulloblastomas, cerebellar astrocytomas, ependymomas, and choroid plexus papillomas
(Becker & Jay, 1990).
Studies by Hudson et al. (Hudson et al, 1989; Hudson & Murdoch, 1992a, b; Murdoch &
Hudson-Tennent, 1994; Murdoch & Hudson, 1999a, c) constitute the most detailed reports of
language disorders following posterior fossa tumor in children to date. Language abilities in
the population of children with posterior fossa tumor were reported to range from above
average language abilities to significant global language impairment, with specific areas of
deficit noted to include expressive semantic and syntactic language, receptive language, and
word-finding (Murdoch & Hudson-Tennent, 1994). Overall, this group of researchers
reported that while not inevitable, language disorders do occur in children in the chronic stage
following treatment for posterior fossa tumor, noting that no particular pattern of language
impairment was considered characteristic, as all aspects of language were reported to be
impacted in the population examined (Murdoch & Hudson, 1999a).
In highlighting the heterogeneity of children with posterior fossa tumor examined across
the studies reported due to the varying profiles, tumor types, and differing treatment
approaches, several factors were proposed by Hudson et al. (1989) to have contributed to the
resulting language disorders reported in this population. These factors included obstructive
hydrocephalus, with the mass effect of the tumor impeding the flow of cerebrospinal fluid,
dilating the ventricular system, compressing the cerebral cortex, and potentially contributing
to tissue destruction and cortical compression (Murdoch & Hudson, 1999a). Other potential
contributing factors noted by these authors included young age at diagnosis and/or treatment,
tumor type, duration of pre-operative symptoms, extent of surgical excision, and insertion of
ventriculoperitoneal shunts. The primary factor considered by these researchers, however, to
have contributed to the presence of language deficits in children treated for posterior fossa
tumor was treatment.
One factor that was not considered by Hudson et al. (1989) was the potential for language
deficits to occur as a direct result of cerebellar damage associated with posterior fossa tumor.
While the significance of treatment effects on language function in children with posterior
fossa is acknowledged, in light of the recently identified role of the cerebellum in cognitive
and language functions, it is equally important to consider the direct effects of cerebellar
damage. As these functions are only now beginning to be recognized, such recent
contributions in the study of the cerebellum were unavailable for consideration in earlier
studies of language function in children treated for posterior fossa tumor.
Language in Children Treatedfor Posterior Fossa Tumor 89
Recent clinical and functional neuroimaging studies have provided evidence of the role of
the cerebellum in modulating cognitive and language functions via its reciprocal connections
with the cerebral cortex (Fabbro, 2000; Fabbro et al. 2000; Marien et al, 2000; Silveri &
Misciagna, 2000; Marien et al, 2001; Silveri et al. 2001). Damage to the cerebellum, in
particular the right cerebellar hemisphere and areas of the vermis, have been documented to
impact nonmotor language function and result in impairments such as reduced verbal fluency,
telegraphic speech, agrammatism, and word-finding difficulties (Riva, 1998, 2000; Schatz et
al., 1998; Steinlin et al, 1999; Levisohn et al, 2000; Riva & Giorgi, 2000a, b; Scott et al,
2001). Considering the early language findings by Hudson et al. (Hudson et al, 1989; Hudson
& Murdoch, 1992a, b; Murdoch & Hudson-Tennent, 1994; Murdoch & Hudson, 1999a) in
light of this new understanding of the role of the cerebellum in language, the patterns of
deficit reported may also have reflected direct damage to cerebellar function in addition to
treatment effects.
Disturbances to high-level language and cognition has also been observed in recent
clinical studies relating to cerebellar damage (Riva, 2000: Riva & Giorgi. 2000a, b). As
evidence outlining the cerebellar contribution to both language and cognition is now
emerging, the direct effects on these areas of function are considered to be plausible, in
addition to later developing processes which include high-level language (Levisohn et al,
2000). The potential for high-level language to be impacted following cerebellar involvement
is supported by several authors. Riva and Giorgi (2000b) reported deficits such as reduced
thinking flexibility and problem solving in children with cerebellar tumors. Other authors,
however, noted disturbances to the processing of verbal intelligence, the ability to produce
complex language structures, complex language tasks, and executive function and higher-
order behaviour (Riva, 1998; Schatz et al, 1998; Levisohn et al, 2000; Riva & Giorgi,
2000a).
Investigations into the presence of literacy and foundation pre-literacy difficulties in
children treated for posterior fossa tumor to date have also been lacking and somewhat
unclear. The reading and writing abilities of children with posterior fossa in a study by
Hudson and Murdoch (1992a) were reported to be comparable to the control group
performance as well as falling within the normal range on standardized assessment. Yet, at an
individual level, two of the six participants experienced significant difficulty in these areas
(Murdoch & Hudson-Tennent, 1994). Unfortunately, the specific underlying deficits and
areas of difficulty experienced by these two participants were not outlined. Sands et al. (1998)
included measures assessing the basic reading and spelling abilities of six children in their
investigation of the neuropsychological functioning of children treated for brain tumor. These
authors reported that mean performances were within the normal range on both tasks.
Specifically, five of the six children demonstrated abilities within the high average to average
range on the academic measure of reading, with the remaining child performing in the low
average range. Reading and writing difficulties have also been documented as a component of
acquired language disorders in children of varying aetiologies including brain tumor
(Alajouanine & Lhermitte, 1965; Hecaen, 1976; Cooper & Flowers, 1987).
The noted presence of reading and writing deficits in children with acquired language
deficits, combined with the young age at which treatment occurs for brain tumor, necessitates
a focus on the impact of brain trauma (including to the posterior cranial fossa) on the
development of pre-literacy skills that provide the foundation for literacy mastery. However,
no study to date has investigated possible disturbances to the pre-literacy precursors that may
result in reading and writing difficulties in children treated for posterior fossa tumor. This
area of development, namely phonological awareness, clearly requires further attention.
As relatively few studies to date have addressed language function in children treated for
posterior fossa tumor (Hudson & Murdoch, 1992a, b; Murdoch & Hudson-Tennent, 1994)
and the only contributing factors considered to impact were treatment effects, there is a need
to extend this area of research in light of current theories. The importance of investigating
high-level language abilities has been highlighted not only by recent evidence supporting the
potential for high-level language disturbances to exist, but also the discrepancy between
performance on general language tasks and high-level language tasks in populations of
children treated for posterior fossa tumor (Murdoch & Hudson-Tennent, 1994). Additionally,
while some reports of reading and writing difficulties have been previously documented in
populations of children treated for posterior fossa tumor and studies of children with acquired
language disorders resulting from varying aetiologies, the phonological awareness skills of
this population, to date, have not been investigated. Therefore, the intent of the present paper
is to examine general language, high-level language and pre-literacy skills in children
following treatment for posterior fossa tumor, in the context of tumor types and their
associated treatment effects, while also considering the impact of direct damage to the
cerebellum on language functions.
METHODOLOGY
Participants
Twelve participants, nine male and three female, were included in the present study ranging in
age from three years nine months to thirteen years six months (mean age = 9.76 years,
standard deviation = 3.30 years), who had been diagnosed with a tumor in the posterior fossa
and had completed treatment at least six months prior to involvement in the study.
Biographical details of these twelve participants are summarized in Table 1. Twelve control
participants individually matched for age and gender (mean age = 9.40 years, standard
deviation = 3.33 years) were also included in the study. All control participants had no history
of cancer, acquired brain injury, epileptic activity or seizures, or had a known history of
speech/language difficulties. All twenty-four participants spoke English as their only
language.
Table 1 — Biographical data of participants treated for posterior fossa tumor
Case/ AaA AaD TPT» Tumor Tumor TM Surgery TRD CHEMO

Gender type location drugs
1/M 13.6 12.3 1.1 MDB 4thV R — 36Gy —
cranial
54Gy to PF
36Gy to
spine
2/F 10.9 9.11 0.6 EPD RCH s, T 54Gy —
R
3/M 13.3 10.1 2.0 MDB Posterior S, N-T 36Gy —
to 41" V R craniospinal
54Gy to PF
4/M 10.11 2.6 8.5 CPP Roof of S N-T — —
4thV
5/M 7.9 7.3 0.6 JP AA Inferior s N-T 54Gy —
vermis,
LCH
6/M 7.4 1.8 3.5 EPD 4"'V s, NA — vincristine,
c CP, Mesna,
G-CSF, HD
methotrexate,
cisplatin,
carboplatin,
etoposide
7/F 6.11 2.11 3.9 EPD 4"'V s, NA 45Gy to PF
5.4 1.7 AA 4"'V R NA

8/M 13.0 11.1 1.2 JP AA Cerebellar s T
vermis s
9/M 13.0 11.1 1.2 EPD 4th V S-T 50Gy +
s, 5.4Gy to
R
residual
tumor
10/M 11.0 6.6 4.1 JP AA Cerebellar s T
midline,
LH
Il/F 3.9 2.5 0.8 EPD 4th V s, T
c
3.1 RCH, #
vermis,
RC
peduncle
12/M 5.11 5.4 0.7 AA LC s S-T
vermis,
LCH
Note: "Age and time presented in years, months; LG = low-grade; 4th V = fourth ventricle; L = left; S
= surgery, R = radiotherapy; C = Chemotherapy; NA = not available; - not applicable; Gy = grays; # =
no treatment for recurrence at time of testing; AaA = Age at assessment; AaD = Age at diagnosis; TPT
a
= Time post-treatment; TM = Treatment; TRD = Total radiation dosage; T = Total; N-T = Near Total;
S-T = Subtotal; RCH = Right cerebellar hemisphere; LCH = Left cerebellar hemisphere; LH = Left
hemisphere; CP = cyclophosphamide; MDB = Medulloblastoma; EPD = Ependymoma; AA =
Astrocytoma; CPP = Choroid plexus papilloma.
Procedure
All 24 participants underwent language testing in a quiet environment with limited

distractions. In all cases, testing was conducted over a number of sessions to reduce the
influences of fatigue.
General Language Assessment Battery. The general language abilities of each of the twelve
cases and their individually matched peers were assessed using three tests which reflect a
measure of general language skills: either the Clinical Evaluation of Language Fundamentals
- Third Edition (CELF-3) (Semel el al, 1995) or the Clinical Evaluation of Language
Fundamentals - Preschool (CELF-Preschool) (Wiig el al, 1992), the Peabody Picture
Vocabulary Test - Third Edition (PPVT-III) (Dunn & Dunn, 1997), and the Hundred Pictures
Naming Test (HPNT) (Fisher & Glenister, 1992).
High-level Language and Phonological Awareness Assessment Battery. The high-level

language and phonological awareness abilities of a group of ten of the twelve participants
treated for posterior fossa described above (excluding Cases 11 and 12) (mean age = 10.74
years, standard deviation = 2.57 years), and their matched peers (mean age = 10.36, standard
deviation = 2.65) were administered a battery of standardized assessments designed to
examine higher-level language function and pre-literacy ability. As Cases 11 and 12 did not
meet the minimum age requirements of the high-level language and literacy assessments,
these cases were not included in the present analysis. The assessment battery included: the
Test of Problem Solving - Elementary (Revised) (TOPS-Elementary) (Zachman et al, 1994)
or the Test of Problem Solving - Adolescent (TOPS-Adolescent) (Zachman et al., 1991), the
Test of Word Knowledge (TOWK) (Wiig & Secord. 1992). the Test of Language
Competence - Expanded (TLC-E) (Wiig & Secord, 1989), and the Queensland University
Inventory of Literacy (QUIL) (Dodd el al, 1996). Again, the age of each participant treated
for posterior fossa tumor and their individually matched peer determined the version or age
group level that was completed for each of the high-level language assessments.
RESULTS
Group Analysis
Non-parametric tests were adopted in the present analysis, due to an incongruity across some
parameters relating to significance on a measure of homogeneity of variance (Levene's Test
for Equality of Variance). The Mann-Whitney U was employed to determine the presence of
statistically significant discrepancies across all parameters. These results are summarized
respectively in Tables 2 and 3. Due to the multiplicity of subtests comprising the QUIL, a
stringent alpha level of p<0.01 was applied for this assessment only (Shearer, 1982). For all
other assessments, an alpha level of p<0.05 was adopted. The group of twelve children with
Language in Children '/'recitedfor Posterior Fossa Tumor 93
posterior fossa tumor performed significantly below an individually matched control group on
the Receptive (p < 0.01), Expressive (p < 0.01) and Total Language (p< 0.01) component of
both versions of the CELF assessments, and the Peabody Picture Vocabulary Test - Third
Edition (p < 0.05). These results are summarized in Table 2.
Table 2 — Posterior fossa tumor and control group analysis: Means (MX standard deviations (SD),
and Mann Whitney U comparisons for the Clinical Evaluation of Language Fundamentals - Third
Edition / Preschool (CELF-3/Prcschool), Peabody Picture Vocabulary Test - Third Edition (PPVT-III),
and the Hundred Pictures Naming Test (HPNT)
Parameter Posterior fossa Control group Mann-Whitney Asymp. Sig.

group (n = 12) (n = 12) V (2-tailed)
M SD M SD
CELF-3/Preschool
Receptive Language Score 107.17 12.66 121.25 11.43 27.5 0.010**
Expressive Language Score 98.42 17.23 117.08 12.39 24.5 0.006**
Total Language Score 102.67 15.26 119.75 11.92 24.0 0.006**
PPVT-III 102.25 12.05 114.17 9.33 32.0 0.021*
HPNT (Raw score /100) 91.58 12.31 97.91 2.66 40.5 0.066
Note: ** p < 0.01; *p < 0.05; Normative data for CELF-3/Preschool and PPVT = 100 ± 15; HPNT
scores reported in raw value out of 100.
Additionally, statistical measures applied to results of the high-level language abilities and
phonological awareness skills revealed significantly reduced performance by the group often
children treated for posterior fossa tumor on the TOPS, the Receptive (p<0.01), Expressive
(p<0.05) and Total Composites (p<0.05) of the TOWK, and the Interpreting Intents (p<0.05),
Expressing Intents (p<0.05), and Total component (p<0.05) of the TLC-E, when compared to
an individually matched control group (see Table 3). Results yielded by the QUIL revealed no
significant differences on any of the phonological awareness parameters (see Table 3).
Individual analysis
While group findings revealed significantly reduced performance across most general and all
high-level language parameters, individual variation is a prominent feature of populations of
children treated for posterior fossa tumor (Murdoch & Hudson-Tennent, 1994). Therefore,
individual areas of strengths and weakness in the context of varying tumor types, associated
symptomatology, and treatment programs, may be masked by group level analysis. An

examination of each case treated for posterior fossa tumor on an individual basis was,
therefore, carried out to determine the presence of specific signs of language impairment in
the context of tumor type, site, and treatments employed.
Table 3 — Posterior fossa tumor and control group analysis: Means (M), standard deviations (SD),
and Mann Whitney U comparisons for the Test of Problem Solving (TOPS), Test of Word Knowledge
(TOWK), Test of Language Competence - Expanded (TLC-E), and Queensland University Inventory
of Literacy (QUIL)
Posterior fossa Control group Mann- Asymp. Sig.

group (n = 10) (n = 10) Whitney (2-taiIed)
U
TOPS 88.80 15.35 109.70 5.25 12.0 0.004**
TOWK
Receptive Composite 98.60 14.65 110.50 8.86 23.5 0.044*

Expressive Composite 93.90 11.32 110.70 8.07 9.5 0.002**
Total Score 95.80 12.44 111.30 8.04 17.5 0.014*
TLC-E
Interpreting Intents 91.40 14.86 106.40 14.15 23.5 0.043*
Expressing Intents 95.50 18.56 110.60 11.53 21.0 0.027*
Total Score 93.40 16.83 109.60 13.90 23.5 0.045*
QUIL
Nonword spelling 12.70 3.40 14.30 1.95 36.5 0.300
Nonword reading 11.20 2.44 11.50 2.80 47.0 0.816
Syllable identification 11.00 1.25 10.80 1.87 48.5 0.905
Syllable segmentation 10.10 2.13 11.20 1.62 35.5 0.205
Spoken rhyme 9.30 3.27 8.80 3.94 48.5 0.906
Visual Rhyme 7.80 3.65 8.33 3.24 39.5 0.648
Spoonerisms 11.70 2.21 11.60 2.95 42.5 0.553
Phoneme detection 10.60 2.55 9.90 3.60 45.0 0.702
Phoneme segmentation 11.70 2.91 11.20 2.57 41.5 0.513
Phoneme manipulation 10.30 3.34 10.80 2.39 29.0 0.936
Note: ** = p<0.01; * = p<0.05; p significant at < 0.01 for QUIL subtests.
An individual analysis of the general and high-level language, and phonological awareness
abilities of each of the twelve participants treated for posterior fossa tumor will therefore
follow. Performance was analyzed by comparing each standard score to the normative data
for each assessment, where less than one standard deviation from the mean is considered to be
below the normal range. Seven of the twelve cases (Cases 2, 3, 5, 6, 7, 9, and 11) were noted
to exhibit evidence of language deficits, while the remaining cases (Cases 1, 4, 8, 10, and 12)
were found to function within normal limits. On the general language assessment battery,
varying degrees of expressive language deficits were observed in the performance of Cases 6,
7, and 11, however a pattern of reduced performance in the area of syntax was also noted in
the profile these three cases. Naming difficulties were only experienced by Case 11. However,
on measures of high-level language and phonological awareness, three of these cases (Cases
6, 7, and 9) exhibited global deficits, with more specific areas of deficit in three children
(Cases 2, 3, and 5). More specific phonological awareness difficulties were noted in five cases
(Cases 3, 4, 6, 7, and 8). Specific details of Cases 2 - 9 and 11 are discussed below. Cases 1,
10, and 12 did not demonstrate reduced performance across all assessments of general and
high-level language, as well as phonological awareness, and will therefore not be discussed in
detail in the case analysis of this paper.
Case 2. Following the diagnosis of a 2 cm low-grade ependymoma in the right cerebellum at

the age of nine years eleven months, Case 2 underwent a treatment regimen of total surgical
removal and an eight week course of radiotherapy (see Figure 1). At the age often years nine
months, Case 2 participated in the current study. The MRI conducted six weeks prior to
language testing revealed no evidence of recurrent tumor (see Figure 2). Case 2 demonstrated
intact general language abilities according to the normative information provided for each of
the assessments (see Table 4). In fact, performances on the Receptive, Expressive and Total
Language components of the CELF-3, including the subtests, Word Classes and Sentences
Assembly, were considered above the normal range. Receptive vocabulary skills measured by
the PPVT-III were also noted to be above the normal range.
Figure 1 — Case 2 at diagnosis: T2 weighted coronal MRI scan demonstrating a 2 cm mass in anterior
cerebellum, just to the right of the midline and adjacent to the right posterior aspect of the fourth
ventricle.
Figure 2 — Case 2 at language testing (six months post treatment): Tl weighted coronal MRI scan
revealing a small cystic areajust posterior to fourth ventricle in keeping with post-surgical change. No
evidence of recurrent tumor following radiotherapy treatment.
Table 4 — Individual general language assessment results (represented in standard scores) of Cases 1
- 5 treated for posterior fossa tumor on the Clinical Evaluation of Language Fundamentals - Third
Edition (CELF-3), Peabody Picture Vocabulary Test - Third Edition (PPVT-III), and the Hundred
Pictures Naming Test (HPNT)
Tests Case I Case 2 Case 3 Case 4 CaseS

CELF-3
Receptive Language 131 118 108 110 106
Concepts & Directions 15 12 16 13 10
Word Classes 14 14 8 12 12
Semantic Relationships1^ / ll#
15A 13A 10A 10A
Sentence Structure#
Expressive Language 125 120 92 104 96
Formulated Sentences 15 13 7 9 8
Recalling Sentences 12 12 7 13 9
Sentence AssemblyA / 11#
15A 15A 12A 10A
Word Structured
Total Language 129 120 100 107 101
107
PPVT-III 120 116 111 100
HPNT (Raw score /100) 100 93 93 96 95
Note: # = Level 1 subtest variation; A = Level 2 subtest variation; Standard scores in italics = normal
range 85 - 115; Subtest standard score normal range = 7-13.
Table 5 — Individual high-level language and phonological awareness assessment results (represented
in standard scores) of Cases 1 - 5 treated for posterior fossa tumor on the Test of Problem Solving
(TOPS), Test of Word Knowledge (TOWK), Test of Language Competence - Expanded (TLC-E), and
Queensland University Inventory- of Literacy (QUIL)
Tests Case 1 Case 2 Case 3 Case 4 CaseS

TOPS 112 78* 96 91 78*
TOWK
Receptive Composite 109 109 94 121 115

Synonyms A / Word Opposites # 12A 11 A 9A 14A 11#
Figurative Usage A / Receptive Vocabulary # 11 A 12A 9A 13A 14#
Expressive Composite 111 100 81* 97 100
Word Definitions 11 10 4* 9 9
Multiple Contexts A / Expressive 13A 10A 9A 10 11
Vocabulary #
Total 111 105 86 109 108
TLC-E
Interpreting Intents 135 91 100 94 109
Listening Comprehension: Making 15 8 12 8 13
Inferences
Figurative Language 17 9 8 10 10
Expressing Intents 112 97 76* 94 100
Ambiguous Sentences 15 12 9 1 8
Oral Expression: Recreating Sentences'" / 9 A
7A 3A* 11 A nu
Speech Acts#
Total 127 93 86 93 105
QUIL
Nonword spelling 15 16 11 12 13
Nonword reading 12 13 10 11 15
Syllable identification 11 12 11 13 10
Syllable segmentation 12 10 12 13 9
Spoken rhyme 11 12 4* 11 10
Visual Rhyme 12 11 3* 5* 11
Spoonerisms 13 13 13 12 13
Phoneme detection 12 14 8 14 9
Phoneme segmentation 13 14 12 14 14
Phoneme manipulation 13 11 13 11 11
A
Note: # = Level 1 subtest variation; = Level 2 subtest variation; * = below normal range (Standard
scores in italics: normal range 85 - 1 15; Subtest standard score normal range = 7-13);
The highest age range with respect to normative data on the HPNT comparable to Case
2's age often years nine months was seven years one month to nine years two months (with a
mean of seven years seven months). The mean accuracy score for this group of children
(n=260) was 84.11, with a standard deviation of 9.85 (Fisher & Glenister, 1992). Therefore,
while Case 2's raw accuracy score of 93 appears to be within the normal range for these lower
age limits, it is impossible to speculate whether performance represents intact naming skills
for this participant, or a mild deficit, at age ten years nine months.
High-level language and phonological awareness assessments revealed performance on
the TOPS that was considered below the normal range (see Table 5). All other performances
across the remaining measures were considered within the normal range.
Case 3. Case 3 was diagnosed at the age often years one month with a 3 cm medulloblastoma
located in the posterior fossa immediately posterior to the fourth ventricle (see Figure 3). A
six month history of increasing headaches associated with vomiting was also reported. A near
total surgical removal was undertaken, followed by a four week course of radiotherapy,
Craniospinal irradiation was administered to the whole brain and spinal axis at a dose of 36
Gy in twenty fractions, with the posterior boosted with the lateral oblique fields to a dose of
54 Gy in 30 fractions. No acute effects were noted.
Figure 3 — Case 3 at diagnosis: Sagittal MRI scan demonstrating a 3 cm medulloblastoma

immediately posterior to the fourth ventricle.
Two years following completion of treatment, at the age of thirteen years three months,
Case 3 participated in the present study. A MRI scan of the brain carried out two weeks
following language testing revealed stable appearance of the tumor with no evidence of tumor
recurrence or metastases (see Figure 4). Case 3 demonstrated intact general language abilities
across all assessments of general language including the CELF-3, PPVT-III and the HPNT
(see Table 4).
Reduced performance below the normal range, however, was noted on the Expressive
Composite of the TOWK, and the expressive subtest Word Definitions, as well as the
Expressing Intents section of the TLC-E and the expressive subtest Oral Expression:
Recreating Sentences (see Table 5). It is noted that a reduced subtest score within each of
these components would contribute to an overall reduction in the overall score for that
component. Performance below the normal range was demonstrated on both the Spoken and
Visual Rhyme subtests of the QUIL.
Case 4. Case 4 was diagnosed with a choroid plexus papilloma in the roof of the fourth
ventricle at the age of two years six months (see Figure 5). A history of increasing head tilting
to the right, ataxia, vomiting after breakfast, and a one year history of irritability (with greater
increase over the ten days prior to diagnosis) was also noted. Case 4 had associated chronic
hydrocephalus. Treatment involved subtotal removal of the tumor, followed by shunt
insertion ten days later. Cranial nerve dysfunction was observed subsequent to surgery, with
residual bulbar palsy and a failure of return of the gag reflex. Neither radiotherapy or
chemotherapy were administered.
Case 4 has a history of ongoing speech pathology intervention from the time of tumor
diagnosis until the present involvement in the current study. Management of severely
impaired swallowing function has been extensive, in addition to a significant oral motor
impairment and vocal fold impairment. General receptive and expressive language skills were
also monitored over this protracted period of time.
Figure 4 — Case 3 at language testing (two years post treatment): Tl weighted sagittal MRI scan post
surgery and radiotherapy treatment revealing stable appearances with no evidence of local tumor in the
posterior fossa, or metastases.
Figure 5 — Case 4 at diagnosis: Coronal CT scan revealing a 2.8 x 2.6 cm choroid plexus pailloma in
ihe midline of the posterior fossa. Surrounding oedema and associated hydrocephalus noted. Marked
dilatation of lateral and third ventricles. Fourth ventricle displaced anterosuperiorly.
Clinical history reported that Case 4 presented with a mild expressive language delay
initially post-surgery. Improvements in both comprehension and expressive language were
noted over time, with some word finding skills also noted at age four years six months. Some
delays in grammatical comprehension were also noted at this time. While at five years two
months, expressive semantic skills were reportedly within normal limits Case 4 demonstrated
mild deficits in expressive semantics and word retrieval skills during structured tasks at age
six years. Expressive semantic difficulties were again noted six months later, in addition to
significant difficulties in sentence formulation. A weakness in the area of sentence
formulation was again noted at nine years five months.
At the age often years eleven months, eight years five months following treatment, Case
4 underwent assessments for the present study. The most recent MRI available was performed
six years and two months prior to language testing and revealed no evidence of recurrent
tumor (see Figure 6). Performance across all subtests of the CELF-3, the PPVT-HI, and the
HPNT, indicated intact general language abilities in the areas of receptive and expressive
language, receptive vocabulary, and naming (see Table 4). High-level language and
phonological awareness testing for the present study revealed intact high-level language
abilities across all measures. A reduced score on the Visual Rhyme subtest of the QUIL,
however, was evident (see Table 5).
Figure 6 — Case 4 six years two months prior to language testing (most recent MRI prior to testing):
Post-surgical coronal CT scan indicating that the ventricular system remains dilated, despite no
progression or evidence of recurrent tumor in the fourth ventricle.
Case 5. Case 5 was diagnosed with a 4cm juvenile pilocytic astrocytoma arising from the
inferior vermis and involving the left hemisphere (see Figure 7), and obstructive
hydrocephalus, subsequent to a four month history of increasing intermittent headaches,
nausea, and vomiting.
Figure 7 — Case 5 at diagnosis: Tl weighted sagittal MRI scan demonstrating a large complex cystic
and solid mass lying within and expanding the fourth ventricle. There is cystic extension inferiorly
through the foramina of Magendie into the Cisterna Magna, extending through the foramen magnum
as far as the inferior margin of the Cl arch.
Figure 8 — Case 5 at language testing (six months post treatment): Post-surgical TI weighted coronal
MR1 scan indicating persistent non-progressive residual tumor on left side of fourth ventricle and
medial aspect of left cerebellar hemisphere.
A near total surgical resection was carried out leaving some residual tumor and associated
oedema. Post-operatively, hydrocephalus was considered present as well as slight ataxia and
left-sided weakness. A neuropsychological assessment revealed performance in the average
range, with no difficulties evident before or after surgery. Some sequencing and divided
attention difficulties were noted. Case 5 was also seen by the speech pathology department at
the hospital post surgical intervention for motor speech difficulties due to a mild dysarthria.
At this time, language abilities were considered intact. Case 5 was discharged from this
service three months prior to involvement in the current study. Six months following
completion of treatment, at the age of seven years nine months, Case 5 participated in the
present study. A MRI of the brain carried out five days prior to language testing indicated
unchanged residual tumor in the left side of the fourth ventricle and medial aspect of the left
cerebellar hemisphere (see Figure 8). Performance on the CELF-3, PPVT-III, and HPNT
indicated the presence of intact general language abilities according to the normative data for
each of the assessments (see Table 4). Performance on high level language and phonological
awareness tasks revealed abilities that were within the normal range across all assessments,
except for reduced abilities noted on the TOPS (see Table 5).
Case 6. Case 6 was diagnosed with a 4 x 4cm ependymoma located in the fourth ventricle at
the age of one year eight months (see Figure 9). A two month history of headaches, vomiting,
weight loss, irritability, tiredness and ataxia was reported. On examination Case 6
demonstrated ataxia with intention tremor, bilateral papilloedema and gross hydrocephalus.
Case 6 underwent surgical resection, and over two weeks later required a shunt insertion for
persisting hydrocephalus. Post-surgically, Case 6 was reportedly irritable and lethargic and
demonstrated bilateral sixth nerve palsy. The shunt was removed two months following
insertion due to infection, and did not require replacement.
Figure 9 — Case 6 at diagnosis: Axial view MRI scan demonstrating a 4 x 4 cm ependymoma located
within the fourth ventricle. Marked hydrocephalus also noted within the lateral ventricles.
Figure 10 — Case 6 at language testing (three years five months post treatment): Coronal view MRI
scan revealing no evidence of recurrent tumor in the fourth ventricle post surgery and chemotherapy
treatment.
Chemotherapy treatment was commenced one month following resection. A personal

protocol was administered over two years with each cycle fifty six days in length. The
chemotherapeutic drugs vincristine, cyclophosphamide, Mesna, G-CSF, high-does
methotrexate, cisplatin, carboplatin, and etoposide were administered. Acute effects included
progressive ototoxicity related to high-dose cisplatin.
At the age of seven years four months (three years five months following completion of
treatment), Case 6 participated in the present study. A MRI of the brain was carried out one
week following language testing (see Figure 10).
General language abilities were considered below the normal range on both the
Expressive Language and Total Language components of the CELF-3, all of the expressive
subtests (Word Structure, Formulated Sentences, and Recalling Sentences), and the receptive
subtest, Sentence Structure (see Table 6). The Word Structure subtest assesses the expressive
knowledge of word structures, whereas the Formulated Sentences subtest examines the
formulation of simple, compound, and complex sentences (Semel et al, 1995). Recalling
Sentences assesses recall and reproduction of sentence surface structure as a function of
syntactic complexity (Semel et ai, 1995). The receptive subtest, Sentence Structure, assesses
comprehension of structural rules at the sentence level. Performance on the Receptive
Language component of the CELF-3 and both the PPVT-III and the HPNT were considered
intact. High-level language and phonological awareness measures revealed reduced
performance that was considered below the normal range on the TOPS, and the Expressive
Composite and Total score of the TOWK (including the receptive subtest Receptive
Vocabulary and the Expressive subtest Word Definitions) (see Table 7). It was also noted that
performance on the Expressing Intents section (including both expressive subtests,
Ambiguous Sentences and Oral Expression) and the Total score of the TLC-E were below the
normal range. The QUIL subtests, Nonword Spelling, Nonword Reading, Visual Rhyme,
Spoonerisms, Phoneme Detection, and Phoneme Manipulation were also considered reduced.
Case 7. Case 7 was diagnosed with a 4 cm ependymoma in the fourth ventricle, which
extended posteriorly between the cerebellar hemispheres and through the foramen magnum to
the inferior border of the axis, including inferior extension along the brainstem. Obstructive
hydrocephalus was also noted. A four day history of vomiting and mallena was reported prior
to diagnosis at the age of two years eleven months. Treatment involved surgical resection as
well as a five week course of radiotherapy, delivering a dose of 45 Gy to the posterior fossa in
25 fractions. A recurrence in the fourth ventricle at the age of five years four months was
preceded by a two week history of headaches (see Figure 11). Treatment involved resection of
the tumor, which was histologically classified an astrocytoma.
A history of speech pathology intervention was noted over a period of several years
following diagnosis for Case 7, continuing up to time of involvement in the present study.
While therapy had focussed on Case 7's significant oral motor inefficiency, motor speech
disturbances (with a noted right facial paralysis), and impaired swallowing abilities, no formal
language assessment had been undertaken at the time of testing. Language testing for the
present study was undertaken at the age of six years eleven months (three years nine months
following the initial treatment program consisting of surgery and radiotherapy treatment, and
one year seven months following surgical treatment for tumor recurrence).
Figure 11 — Case 7 at diagnosis of recurrent tumor: Coronal MRI scan revealing recurring tumor in
fourth ventricle.
Table 6 — Individual general language assessment results (represented in standard scores) of Cases 6
- 10 treated for posterior fossa tumor on the Clinical Evaluation of Language Fundamentals - Third
Edition (CELF-3), Peabody Picture Vocabulary Test - Third Edition (PPVT-III). and the Hundred
Tests Case 6 Case 7 Case 8 Case 9 Case 10

CELF-3
Receptive Language 88 90 114 118 110
Concepts & Directions 10 8 12 11 10
Word Classes 8 7 10 9 13
Semantic Relationships'^/ 6#* 10# 15A
Sentence Structured
Expressive Language 61* 82* 102 106 108

Formulated Sentences 6* 8 12 10 13
Recalling Sentences 4* 6* 11 8 11
Sentence AssemblyA / Word Structure# 4#* 7# 10A 13A 10A
Total Language 73* 85 110 110 109

PPVT-III 89 84* 108 90 110
HPNT (Raw score / 100) 94 89 97 98 98
Note: # = Level 1 subtest variation; A = Level 2 subtest variation; Standard scores in italics = normal
range 85 - 115; Subtest standard score normal range = 7-13.
Table 7 — Individual high-level language and phonological awareness assessment results (represented
in standard scores) of Cases 6 - 1 0 treated for posterior fossa tumor on the Test of Problem Solving
(TOPS), Test of Word Knowledge (TOWK), Test of Language Competence - Expanded (TLC-E), and
Queensland University Inventory of Literacy (QUIL)
Tests Case 6 Case 7 Case 8 Case 9 Case 10

TOPS 79* 62* 94 88 100
TOWK
Receptive Composite 86 81* 91 83* 97
Synonyms A / Word Opposites # 10# 7# 10A 8A 12A
Figurative Usage A / Receptive Vocabulary # 5#* 6#* 7A 6A* 7A
Expressive Composite 81* 78* 97 88 106
Word Definitions 4* 4* 9 6* 11
Multiple Contexts A / Expressive 10A 8# 10A 10A 11A
Vocabulary #
Total 82* 78* 93 85 101
TLC-E
Interpreting Intents 94 76* 97 65* 94
Listening Comprehension: Making 10 7 1 3* 8
Inferences
Figurative Language 8 5* 12 7 10
Expressing Intents 73* 82* 100 76* 118
Ambiguous Sentences 6* 6* 8 6* 12
Oral Expression: Recreating Sentences'" / 5#* 8# 12A 6A* 14A
Speech Acts#
Total 81* 76* 98 69* 106
QUIL
Nonword spelling 5* 11 16 12 16
Nonword reading 6* 9 12 12 12
Syllable identification 1 1 9 11 9 11
Syllable segmentation 9 9 6* 9 12
Spoken rhyme 8 3* 11 12 11
Visual Rhyme 3* 5* 7 9 12
Spoonerisms 6* 10 12 12 13
Phoneme detection 6* 11 10 10 12
Phoneme segmentation 7 12 6* 13 11
Phoneme manipulation 4* 5* 13 9 13
A
Note: # = Level 1 subtest variation; = Level 2 subtest variation; * = below normal range (Standard
scores in italics: normal range 85 - 115; Subtest standard score normal range = 7-13)
A recent MRI of the brain was taken three months prior to language testing and indicated
no evidence of recurrence or metastases (see Figure 12). Performance on both the Expressive
Language component of the CELF-3, and the expressive subtest, Recalling Sentences, was
deemed below the normal range (see Table 6). The Recalling Sentences subtest of the CELF-
3 is responsible for assessing recall and reproduction of sentence surface structure as a
function of syntactic complexity (Semel et al., 1995). All remaining subtests of the CELF-3,
as well as the Receptive Language and Total components, were considered intact according to
the normative data. Receptive vocabulary abilities as measured by the PPVT-III, were also
below the normal range. Naming ability assessed by the HPNT, however, was considered
within the normal limits for a child aged six years eleven months.
Figure 12 — Case 7 at language testing (one year seven months post treatment for recurrence): Tl
weighted coronal MRI scan revealing evidence of right inferior vermian and cerebeilar tonsil resection
via right posterior craniotomy. No locally recurrent soft tissue mass present. No positive mass effect.
No tumor cyst in posterior fossa or at craniocervical junction. Also, no mass effect or enhancement
above tentorium. Cerebral ventricles are decompressed.
In an individual analysis of high-level language and phonological awareness tasks,

performance revealed widespread difficulties across the expressive, receptive, and total
language components of both the TOWK and the TLC-E, and most subtests (see Table 7).
Both the Spoken and Visual Rhyme subtests of the QUIL were also noted to be below the
normal range.
Case 8. Case 8 was diagnosed with a 6 x 4 x 4 cm juvenile pilocytic astrocytoma in the

cerebeilar vermis and hydrocephalus at the age of eleven years ten months (see Figure 13). A
six to eight week history of morning nausea and vomiting, headaches, increased clumsiness,
and blurred vision preceded diagnosis, with the most recent symptoms prior to diagnosis
including unsteadiness on feet and papilloedema. A total resection and third ventriculostomy
was carried out. Post-surgical complications included a right high-frequency hearing loss and
a significant deterioration in the vision of the left eye. Additionally, increased intracranial
pressure, headaches, and bitemporal hemianopia was noted post operatively, which was
reported to resolve. Persisting difficulties included truncal ataxia, decreased coordination,
right-sided hearing difficulties, left optic atrophy with secondary left visual diplopia.
Figure 13 — Case 8 at diagnosis: Sagittal view MRI scan demonstrating a 6 x 4 x 4 cm juvenile

pilocytic astrocytoma in cerebellar vermis.
Neuropsychological assessment was carried out both twelve weeks and eleven months
following surgery. In the initial assessment following surgical treatment, it was reported that
performance was considered in the average range of intellectual functional (verbal average to
above average, nonverbal/visual in average range). It was reported that Case 8 generally
performed within the normal range on all aspects of cognition and memory, although it was
reportedly unknown whether his average performances in these areas represented a relative
decline from previous functioning. No reported concerns in academic ability was noted prior
to illness, with academic performance considered to be well above the average range (with
particular strengths in English and Mathematics). Case 8 was reported to have not
experienced any difficulty returning to school. The second neuropsychological evaluation
eleven months post surgery also indicated performance in the average range of intellectual
functioning, with most areas of performance considered consistent with the previous
assessment, or slightly improved. However, certain areas of weakness were also reported with
some areas of relative decline from the previous assessment. These included metal arithmetic
reasoning abilities (although still in average range), and a relative weakness in speed of
information processing (still in average range). At this time Case 8's mother also reported
difficulties with concentration, motivation, organization, and task completion.
At the age of thirteen years, Case 8 participated in the present study. Testing was carried
out one year two months post treatment, and three months following the second
neuropsychological testing session. Case 8 had been seen by the Speech Pathology team at the
hospital one year prior to testing for the current study, when language abilities were
considered to be within normal limits. In the present study, general and high-level language,
and phonological awareness parameters were considered to be well within the normal range
(see Tables 6 and 7).
Case 9. At the age of six years six months. Case 9 was diagnosed with a large ependymoma
located in the fourth ventricle and extending out of the foramen of Magendie into the cisterna
magna and partly extending into the cervical canal. Possible invasion of the right cerebellar
hemisphere was also reported and obstructive hydrocephalus noted. An eleven month history
of headaches, neck stiffness, and photophobia was reported prior to diagnosis. A partial
surgical resection was carried out, followed by post-operative complications consisting of a
right sixth nerve palsy with a vertical diplopia, and mildly impaired coordination. A six week
course of radiotherapy was commenced one month following surgery. A dose of 50Gy was
delivered in 25 fractions over 36 days using right and left posterior oblique fields with a
smaller volume dose to the site of the residual tumor of a further 5.4 Gy in 3 fractions over 3
days. A ventriculoperitoneal shunt was inserted one month following completion of
radiotherapy treatment due to increased intracranial pressure and a three week history of
occasional headache, ataxia, cerebellar signs, nystagmus, and mild right facial nerve palsy.
Involvement in the present study was carried out at the age of eleven years, four years and
one month following treatment completion. Case 9 demonstrated intact general language
abilities across all measures of receptive and expressive language, receptive vocabulary,
naming (see Table 6). High-level language and phonological awareness assessments,
however, revealed performance below the normal range on the Receptive Composite of the
TOWK, the receptive subtest Figurative Usage (which may have contributed to an overall
reduction in the overall receptive component), and the expressive subtest Word Definitions
(see Table 7). More widespread difficulties were noted, however, across both the Expressing
Intents and Interpreting Intents components of the TLC-E assessment, and most subtests.
Performance on the QUIL, however, indicated abilities within the normal range.
Case 11. Case 11 was diagnosed with an ependymoma of the fourth ventricle at the age of
two years five months (see Figure 14). Surgical resection was carried and a shunt inserted.
Postoperatively a left sixth nerve palsy was reported, as well as residual ataxia and mild right-
sided hemiparesis. Two months later, chemotherapy treatment was commenced, with the
drugs vincristine, cyclophosphamide, cisplatin, and etoposide employed. A language
assessment carried out one month following commencement of chemotherapy noted delays in
receptive and expressive language skills and speech skills. Case 11 was reported to perform at
an eighteen to twenty-three month old level (at the age of two years eight months), and
demonstrated a failure to understand verbs in context, spatial concepts, pronouns, quantity
concepts, and recognize actions in pictures. Case 11 demonstrated a similar level of skill in
the expressive language component in the use of pronouns, and naming objects and pictures,
with an inability to use plural forms, verb + ing, and respond to what/where and yes/no
questions. A follow-up assessment session seven months later at the age of three years three
months, again reported delays in receptive and expressive language development. At this
time, Case 11 was reportedly performing at a 24 to 29 month receptive language level, and
failed to comprehend pronouns and descriptive concepts. At an expressive level, Case 11 was
experiencing difficulty using verb + ing or possessives, and was only using three to four word
sentences. It was also noted that Case 11 tended to substitute words within a related semantic
category.
Figure 14 — Case 11 at diagnosis: Axial view MRI scan demonstrating an ependymoma in the fourth
ventricle.
At the age of three years nine months, Case 11 underwent general language testing for the
present study. An MRI taken one month following language testing revealed evidence of
recurrence of an ependymoma in the right cerebellar hemisphere, vermis, and right middle
cerebellar peduncle (see Figure 15). While performance was considered to be within the
normal range across most components of the CELF-Preschool, performance on the expressive
subtest, Word Structure, was considered reduced at one standard deviation below the mean
(see Table 8). The Word Structure subtest evaluates a child's knowledge and use of early
acquired morphological rules and forms (Wiig & Secord, 1992). Although normative data is
not available for children below the age of four years six months on the HPNT, it is
considered that reduced naming abilities were evident for Case 11 with a score of 55, as the
mean accuracy score for the nearest age range, the 4;6-4;ll age category (n=2), was 74.50.
High-level language assessments were not conducted due to age requirements.
Language in Children Treatedfor Posterior Fossa Tumor Ill
Figure 15 — Case 11 at language testing (eight months post treatment): Tl weighted coronal MR]
scan revealing evidence of recurrence of ependymoma with a 2.2 cm enhancing mass in right
cerebellar hemisphere, vermis, and right middle cerebellar peduncle. No associated hydrocephalus.
Table 8 — General Language Assessments Results (represented in Standard Scores) of Case 11 & 12
with Posterior Fossa Tumor, on the Clinical Evaluation of Language Fundamentals - Preschool
(CELF-Preschool), Peabody Picture Vocabulary Test - Third Edition (PPVT-HI), and the Hundred
Tests Case 11 Case 12

CELF-Preschool
Receptive Language 95 98
Linguistic Concepts 8 12
Basic Concepts 8 9
Sentence Structure 10 8
Expressive Language 85 100

Recalling Sentences in Context 9 11
Formulating Labels 9 8
Word Structure 4* 11
Total Language 89 99
PPVT-III 87 107
HPNT (Raw score /100) 55* 89
Note: Standard scores in italics = normal range 85 - 115; Subtest standard score normal range = 7-13.
DISCUSSION
A group comparison between twelve children treated for posterior fossa tumor and their
individually matched peers revealed significantly reduced performance on measures of
general language in both receptive and expressive language, as well as at an overall language
level indicated by the Total Language score. Receptive vocabulary skills on the PPVT-III
were also noted to be significantly reduced compared to the control group. Naming abilities,
however, were not considered significantly different. Reports of reduced receptive language
(including receptive vocabulary) and expressive language in a group of children treated for
posterior fossa tumor have been previously noted by Hudson, Murdoch and colleagues
(Hudson & Murdoch, 1992a; Murdoch & Hudson-Tennent, 1994; Murdoch & Hudson,
1999a).
Results of a group comparison between ten children treated for posterior fossa tumor and
ten control participants matched for age and gender revealed significant differences across all
measures of higher-level language in the areas of problem solving (TOPS), word and concept
knowledge (TOWK), and abstract and complex linguistic structures (TLC-E). As mentioned
previously, Hudson and Murdoch (1992a) and Murdoch and Hudson-Tennent (1994) are the
only authors to date that have investigated the high-level language abilities of children treated
for posterior fossa tumor. These researchers noted deficits in the ability to understand and
manipulate complex and abstract language structures (Hudson & Murdoch, 1992a).
Research evidence of high-level language and cognitive deficits in children with
disturbances to the posterior fossa is supported by recent clinical and functional neuroimaging
evidence implicating the cerebellum in these domains (Muller et al, 1998; Riva, 1998, 2000;
Riva & Giorgi, 2000a, b). Results from the present study indicating reduced performance of
high-level language abilities across areas involving problem solving, high-level semantic
word and concept knowledge, and abstract and complex language structures in children
treated for posterior fossa tumor may therefore be considered in the context of direct impact
of a tumor in the cerebellum. Indeed, clinical findings relating to impaired higher language
abilities following cerebellar disturbances, including tumor, have been previously
documented.
Riva and Giorgi (2000a) reported reduced thinking flexibility and problem solving in
groups of children with left or right cerebellar involvement. Additionally, both Schatz et al.
(1998) and Riva and Giorgi (2000a) reported deficits in verbal processing and complex
language tasks in patients with cerebellar damage. A case study presented by Riva (1998)
indicated a severely impaired ability to produce complex language structures following a viral
inflammatory lesion in the cerebellum. Impairments in executive function and higher-order
behaviour were also reported by Levisohn et al. (2000), in an examination of nineteen
children who had undergone surgery for a cerebellar tumor in the absence of either
radiotherapy or chemotherapy.
Examination of twenty children treated for posterior fossa tumor at both group level and
individual case level revealed patterns of individual variability were evident in reports by
Hudson and colleagues (Hudson & Murdoch, 1992a; Murdoch & Hudson-Tennent, 1994).
Such previously reported variability was also noted in the present study. Despite an overall
reduction in performance as a group on both Receptive and Expressive Language components
of the CELF-3 (and the Total Language score) and receptive vocabulary abilities on the
PPVT-III, specific general language impairments were noted in just three of the twelve
children with posterior fossa (Cases 6, 7. and 11). Across high-level language parameters,
global reductions in group performance was demonstrated despite global high-level language
impairments noted in just three of the ten participants (Cases 6, 7, and 9). Specific weaknesses
in areas such as problem solving (Cases 2 and 5) and lexical knowledge (Case 3), however,
were demonstrated by an additional three cases. Phonological awareness deficits ranging from
mild to severe were noted in the profiles of five of the ten children examined (Cases 3, 4, 6, 7
and 8), despite being commensurate with their individually matched peers at a group level.
It was noted that in the profiles of all three participants with identified general language
performance (Cases 6, 7, and 11), that most of the demonstrated areas of difficulty were
represented by expressive language deficits. In particular, Case 6, who exhibited difficulty
across the greatest number of expressive tasks (all expressive language subtests of the CELF-
3, the Expressive Language Score, and Total Language Score), yielded the most reduced
scores, such that performance on the Expressive Language component was more than two
standard deviations below the mean. Murdoch and Hudson-Tennent (1994) also noted
reduced performance by six children with posterior fossa tumor in expressive semantic and
/or syntactic tasks. More recently, researchers investigating the presence of language disorders
associated with cerebellar damage have also noted particular dysfunction in expressive
language (Riva, 1998; Levisohn et al, 2000; Riva & Giorgi, 2000b).
Syntax was also noted to be a predominant area of weakness represented by performance
across subtests of the CELF-3/Preschool in all three cases with demonstrated general
language deficits (Cases 6, 7, and 11). Difficulties specifically in the area of syntax are
particularly prevalent across cerebellar clinical studies specifically investigating the role of
the cerebellum in language and cognitive function (Silveri et al, 1994; Marien et al, 1996;
Molinari et al, 1997; Schmahmann & Sherman, 1998; Gasparini et al, 1999; Fabbro et al,
2000; Marien et al, 2000; Riva & Giorgi 2000b; Silveri & Misciagna, 2000). In an
examination of four patients with varying tumor types located in the cerebellum (including
the right cerebellar hemisphere, the left cerebellar hemisphere, the vermis, and the superior
vermis), Fabbro et al. (2000) recognised syntax as the most compromised language feature in
all cases. In the context of an expressive aphasic syndrome, a patient with a history of
ischemic infarction in the vascular territory of the right arteria cerebellaris superior was also
noted by Marien et al. (2000; 1996) to have particularly impaired syntactic abilities at both a
comprehension and expressive level, together with reduced verbal output, perseverations,
word-finding difficulties, decreased speech rate, and lack of content words. Difficulties were
also exhibited by seven children of the twenty children with posterior fossa tumor (Murdoch
& Hudson-Tennent, 1994). Residual impaired syntactic abilities were also noted in the profile
of the only participant with a tumor (posterior fossa), in a group of brain injured children with
acquired language disorders examined by Cooper and Flowers (1987).
Although as a group, naming difficulties were not observed to be significantly reduced in
comparison to an individually matched control group, at an individual level severe naming
difficulties were observed in one participant (Case 11). Such a severe disturbance representing
abilities that were two standard deviations below the normal range existed in the presence of
generally intact skills, except for a specifically reduced subtest score in the area of expressive
syntax. As previously mentioned, both Hudson and Murdoch (1992a) and Hudson-Tennent
and Murdoch (1993) failed to find evidence of naming difficulties at a group level in their
studies of children with posterior fossa tumor. However, in their individual examination of
these same twenty children, Murdoch and Hudson-Tennent (1994) found that seven cases
performed poorly on assessments measuring naming ability.
Recent studies examining language disorders following cerebellar damage have also noted
naming difficulties in these populations (Akshoomoff et al, 1992; Marien et al., 1996;
Beldarrain et al., 1997; Schmahmann & Sherman, 1998; Gasparini et al., 1999; Marien et al.,
2000; Riva & Giorgi, 2000a). In a study investigating the language and cognitive abilities of
twenty-six children following removal of cerebellar tumor by Riva and Giorgi (2000b), a
slight worsening in naming was noted to be associated with cerebellar damage to the
hemispheres, regardless of laterality. The significant naming difficulties experienced by Case
11 in the present study may reflect the impact of a recurrent tumor at the time of testing on
right cerebellar hemispheric function.
Examination of the individual high-level language abilities of children treated for
posterior fossa revealed that three of the ten participants (Cases 6, 7 and 9) demonstrated
global deficits across most areas of high-level language. Of these participants, Cases 6 and 7
were noted to demonstrate difficulties in both general and high-level language. Furthermore, a
pattern demonstrating particular expressive language weaknesses on high-level tasks followed
a similar pattern in the general language profile of Case 6. While Case 7 demonstrated less
severity of general language deficit than Case 6, with scattered mild-moderate difficulties in
expressive syntax at a general level, Case 7 demonstrated the most severe global impairment
in high-level language ability. In contrast to Cases 6 and 7, however, Case 9 demonstrated no
general language deficits in general language ability despite quite widespread difficulties
evident across assessments of high-level language in the present analysis. Therefore, it is
significant that no evidence in the general language profile of this case indicated a need for
either further assessment or signalled the severity of difficulties in this area of high-level
language.
In addition to the three patients with high-level language global deficits, three additional
cases with no previous general language difficulties (Cases 2, 3, and 5) also demonstrated
some specific disturbances to this level of language. Case 3 performed below the normal
range on both the Expressive Composite of the TOWK and the Expressing Intents section of
the TLC-E, with one expressive subtest below the normal range on each assessment (Word
Definitions (TOWK) and Oral Expression: Recreating Speech Sentences (TLC-E)). It is
suggested that as the scores for each of these two subtests were severely reduced, the overall
expressive scores were adversely impacted by both subtests respectively. The Word
Definitions subtest of the TOWK assesses the ability to provide definitions that include
category membership and semantic features, whereas the Oral Expression subtest in Level 2
of the TLC-E addresses the ability to recreate intent-driven complex sentences that
incorporate lexical items and respond to situational contexts. Both of these assessments
involve the ability to retrieve and demonstrate the complexities of lexical knowledge and
present this in context in a sentence form. Therefore, it is evident that Case 3 experienced
specific difficulty in this area.
In the present analysis, Cases 2 and 5 were the only two cases to demonstrate intact
language in all areas except the specific area of problem solving. It was noted, however, that
both Cases 2 and 5 shared a similar tumor location within the cerebellar hemispheres (right
cerebellar hemisphere and the inferior vermis / left hemisphere respectively). As mentioned
previously, Riva and Giorgi (2000a, b) reported that children with right or left hemispheric
involvement exhibited reduced abilities in thinking flexibility and problem solving subsequent
to an examination of the specific functions of both cerebellar hemispheres and the vermis in
cognition, language, and executive functions of children who had undergone surgical removal
of a tumor in these areas. As both Case 2 and 5 underwent total and near total tumor removal
respectively in these critical areas, it is suggested that the observed disturbances to problem
solving may be related to damage to the cerebellar structures. Other cases with disturbances to
problem solving included Cases 6 and 7, who also experienced quite widespread impairment
across most high-level language tasks, with tumor locations in the fourth ventricle. It must
also be acknowledged that Case 8, who also had a tumor located in the vermis, did not
experience problem solving difficulties in the present study.
It was also noted that both Cases 2 and 5 participated in the present study at six months
post treatment. Therefore, as has been suggested previously, while an impairment was evident
in this specific area of problem solving only six months following treatment, the potential for
these cases to develop more widespread difficulties in high-level language may exist due to
evidence of neuropathological changes occurring from six months up to twenty years later,
including radionecrosis, white matter density, intracerebral calcifications, cerebral atrophy,
and ventricular and subarachnoid dilation (Arya et ai, 1986; Hoppe-Hirsch, 1993; Plowman,
1992).
Variable findings in the present study demonstrating reduced general receptive and
expressive language abilities as well as overall high-level language reductions at the group
level, in addition to both global high-level language impairments and specific areas in either
general or high-level language at an individual level, may be attributable to any one or a
combination of varied factors that are present in the current group of children with posterior
fossa tumor. These factors include tumor type, tumor location, treatment and/or combinations
of treatments employed, treatment complications/effects, age at diagnosis/treatment, the co-
occurrence of hydrocephalus, and other associated symptoms, and have been recognized to
significantly impact general language ability in children with posterior fossa tumor,
particularly in children undergoing radiotherapy treatment (Murdoch & Hudson, 1999a, b). In
fact, the variability in individual findings described is also not surprising given the
inconsistency of factors between each case (Murdoch & Hudson, 1999c). It is suggested that
all of the possible factors existing in the presentation and management of children with
posterior fossa may play a part in contributing to the presence of language deficits.
It is of interest that all five cases with a posterior fossa ependymoma (Cases 2, 6, 7, 9, and
11) presented with some form of language disturbance, accounting for all children with
general language deficits and all cases with global high-level language impairment.
Additionally, the ependymoma in four of the five children (Cases 6, 7, 9, and 11) had arisen
in the fourth ventricle, a commonly reported location for this tumor type (Cohen el al, 1982;
Gumbinas, 1983; van Eys, 1991; Plowman, 1992). Due to a characteristic presentation of the
ependymoma as a space-occupying mass, the most common clinical findings are those
resulting from increased intracranial pressure, including nausea, vomiting, headaches, and
obstructive hydrocephalus (Wilson, 1975; Gumbinas, 1983; van Eys, 1991; Plowman, 1992;
Heideman et al, 1993). Severe obstructive and persisting hydrocephalus was evident in all
three cases with global high-level language impairment, with symptoms evident in all five
children. Long-term and later developing effects resulting from hydrocephalus have been
reported by several authors and included effects on cognition (Wilson, 1975; Danoff et al,
1982; McWhirter & Masel, 1987; Silverman & Thomas, 1990; van Eys, 1991). A study by
Danoff el al. (1982) documented an increase in the incidence of intellectual impairment
occurring in children treated for brain tumor who had presented with hydrocephalus compared
to those who did not. Murdoch and Hudson-Tennent (1994) considered the presence of
hydrocephalus to be associated with language disturbances in their twenty cases with
posterior fossa tumor. The use of shunts to treat hydrocephalus in children with posterior
fossa tumor (as in Cases 6 and 11) was also examined by these authors, although the
influences of this treatment strategy on language function were considered unclear. Therefore,
it is possible that the severe and persisting long-term nature of the hydrocephalus in at least
three of the five cases due to an ependymoma in the fourth ventricle (Cases 6, 7. and 9) was a
factor contributing to the widespread and more severe reduction in performance across all
high-level language tasks observed, and the presence of increased intracranial pressure in the
remaining two instances related to specific language disturbances in general. However, recent
advances in the understanding of connections of the cerebellum to the language centres of the
cerebral cortex may indicate the potential of a more direct impact by a tumor located in the
posterior fossa.
The site of the recurring tumor in Case 11 was noted to involve the right cerebellar
hemisphere, the vermis, and the right cerebellar peduncle. As previously mentioned, it has
been reported that critical areas of the cerebellum recently noted to be involved in language
function include the right cerebellar hemisphere and the vermis (Oki et al, 1999; Riva, 2000;
Riva & Giorgi, 2000a, b; Scott et al, 2001), and may account for the severely reduced
naming difficulties in this case. As previously mentioned, naming and word finding
difficulties have been commonly reported to be associated with disruptions to the cerebellum,
Language in Children Treatedfor Posterior Fossa Tumor 1 17
particularly the right hemisphere and vermis (Akshoomoff et al. 1992; Marien et al, 1996;
Beldarrain el al, 1997; Schmahmann & Sherman, 1998; Gasparini et al, 1999; Marien et al,
2000).
Due to the potential for an ependymoma to be locally infiltrative with a tendency for
dissemination via the cerebrospinal fluid (Cohen et al, 1982; Gumbinas. 1983; Segall et al,
1985; van Eys, 1991; Plowman, 1992; Tait et al, 1992; Heideman et al, 1993), together with
its comparatively reduced prognostic favorability, treatment is often more aggressive for this
type of tumor (e.g. higher dosages of radiotherapy or chemotherapy, more combinations of
treatment techniques, more heroic attempts at surgical removal), as it is often recommended
that adjuvant treatment is necessary in combination with surgical intervention in cases of
ependymoma (Cohen et al, 1982). This is highlighted by the use of post-operative
radiotherapy or chemotherapy in all five cases with ependymoma and some degree of
language difficulties. While Cases 6 and 11 were treated with surgery followed by an
intensive chemotherapeutic program, Cases 2, 7 and 9 underwent radiotherapy subsequent to
surgical intervention. It is therefore also suggested that the intensive treatments employed
may have also contributed to the profiles noted, as the effects of chemotherapy and
radiotherapy on language and cognitive function have been well-documented (Cohen et al,
1982; Arya et al, 1986; Di Chiro et al, 1988; Hudson et al, 1989; Barrett & Donaldson,
1992; Heideman et al, 1993; Kun & Moulder, 1993; Murdoch et al, 1999) and explored in
the present study. Therefore, the necessity of careful long-term monitoring of the high-level
abilities of children treated with radiotherapy for posterior fossa tumor is highlighted.
Case 3, who demonstrated specific high-level deficits, also received radiotherapy
treatment following surgical intervention. Additionally, Case 3 was one of two cases in this
analysis to have presented with a medulloblastoma, with the site identified as directly
posterior to the fourth ventricle, and is only child with this tumor type to have been treated
with radiotherapy following surgery. The other case with a medulloblastoma, Case 1,
however, also received radiotherapy and did not experience any difficulties on both high-level
or general language tasks. The presence of increased intracranial pressure in cases of
medulloblastoma is considered a common finding due to compression of the fourth ventricle
(Segall et al, 1985; Neidhardt el al, 1986; van Eys, 1991; Plowman, 1992; Heideman et al,
1993). While signs of increased intracranial pressure are reportedly seen early, these
nonspecific findings may be intermittent and subtle. Therefore, the possibility of these
symptoms being overlooked and the duration of symptoms occurring three or more months
prior to diagnosis is within the realms of possibility (Heideman et al, 1993). It is suggested
that the long-term presence of increased intracranial pressure evident in Case 3 prior to
diagnosis may have been responsible for exerting prolonged effects on cerebellar function that
contributed to disturbances to high-level language. Findings of long-term effects associated
with prolonged increased intracranial pressure have been reported by several authors (Wilson,
1975; Danoff et al, 1982; McWhirter & Masel, 1987; Maria & Halpern, 1989; Silverman &
Thomas, 1990; van Eys, 1991).
It was noted in the present study that all three cases with demonstrated language
impairment (Cases 6, 7, and 11), two of who also exhibited global high-level language
impairment, were initially diagnosed and subsequently treated under the age of three years.
Although a recurrent tumor had been noted at the time of testing for Case 11, treatment had
not yet been carried out and is therefore exempt from the present consideration of treatment
effects. Therefore, it is suggested that the factor of very young age in these three cases may
have led to the occurrence of the general language deficits observed, due to reports that
effects of treatment, particularly radiotherapy, are more serious in very young children (van
Eys, 1991). Young age at tumor diagnosis has been noted by several authors to be directly
linked to impaired neuropsychological functioning and intellectual dysfunction (Alajouanine
& Lhermitte, 1965; Mulhern et al, 1992; Broadbent et al, 1981; Dennis el al., 1996). In a
retrospective study by Sue el al. (1990), examining twenty children who had been diagnosed
with a brain tumor under the age of three, eighteen cases were noted to have exhibited adverse
sequelae. Impaired cognitive function was reported in 85% of the children examined,
neurological sequelae in 65%, and central endocrinopathies in 70%.
All three cases diagnosed under the age of three years in the present study were also
treated with surgical intervention prior to radiotherapy or chemotherapy. While age at surgical
treatment was not considered a strong predictor of IQ performance in a study by Kao el al.
(1994), some associated declines in intellectual function were noted. Surgery was also noted
by Sue et al (1990) to be a possible cause of the late effects observed in their study of
children under three years of age. In contrast, however, Levisohn et al. (2000) reported that
younger children treated with surgery were less likely to exhibit neuropsychological deficits
(demonstrated in 33 % of children under age seven compared to 80% of children aged over
seven years). Therefore, it is possible other factors such as radiotherapy or chemotherapy
following surgery may have been a contributing factor in the deficits observed in these
participants than the surgery itself. Such a suggestion receives some support in the present
study by an absence of a general language impairment in Case 4, who was diagnosed and
subsequently treated under the age of three years for a posterior fossa choroid plexus
papilloma with surgery only. However, it must be noted that ongoing speech pathology
intervention from the time of diagnosis until involvement in the present study included
monitoring of general language abilities, with some mild difficulty in expressive semantics
and sentence formulation noted two years prior to involvement in the present study.
The most prevalent information regarding the adverse impact of young age on cognitive,
neuropsychological, and language function, however, is related to the use of radiotherapy
(Danoff et al, 1982; Duffner et al., 1983; Silverman et al, 1984; van Eys, 1991; Barrett &
Donaldson, 1992; Plowman, 1992; Tait et al, 1992; Heideman et al, 1993; Hoppe-Hirsch,
1993; Kun & Moulder, 1993;). Case 7 was treated by radiotherapy subsequent to surgery and
exhibited the most severe language deficits out of all twelve subjects examined in the present
study. It is acknowledged that prior to involvement in the present study Case 7 had been
treated for both an initial tumor and a subsequent recurrence. Initial treatment involved
surgery followed by radiotherapy, with surgical removal of the recurred tumor. Therefore, not
only did this case undergo radiotherapy for treatment of posterior fossa tumor with its known
associated language effects, Case 7 also underwent surgical intervention twice. While repeat
craniotomy was noted by Kao et al, 1994 to be strongly associated with declines in IQ scores,
and may explain similar findings related to language in Case 7, a case described by Murdoch
and Hudson-Tennent (1994) who also underwent two craniotomies (second to remove
residual tumor five weeks following initial surgery), did not display evidence of general
language difficulties.
Packer et al. (1986) indicated that the immature brain is more susceptible to the effects of
radiation due to axonal growth, dendriatic arborization and synaptogenesis occurring in the
early years of life. In fact according to Sue el al. (1990) irradiation was clearly responsible for
intellectual sequelae in seven of the twenty children treated for brain tumors under the age of
three in their study. These authors concluded that the high rate of intellectual impairment was
consistent with the concept that radiotherapy is very hazardous before the age of three.
Considering the disruption to higher functions of intelligence documented in children, it could
be reasonably expected that the adverse impact on the language abilities of the children in the
present study noted to be under age three years at the time of diagnosis were related to age. It
is noteworthy, however, that Cases 2, 3, and 9, also underwent radiotherapy as part of their
treatment program and while did not exhibit general language difficulties, all demonstrated
disturbances of varying degrees to high-level language abilities (Danoff et al, 1982; Duffner
et al, 1983; Silverman et al, 1984; van Eys, 1991; Barrett & Donaldson, 1992; Plowman,
1992; Tait et al, 1992; Heideman et al, 1993; Hoppe-Hirsch, 1993; Kun & Moulder, 1993).
Irrespective of age, however, both cases (Cases 6 and 11) treated with chemotherapy
following surgery exhibited language disturbances. While the effects across areas of language
varied from specific difficulties to widespread impairment, the degree of these disturbances
were noted to be consistently severe in both cases. Adverse effects on brain structure,
cognition, and language function have been reported in populations of children treated with
chemotherapy irrespective of age (Murdoch el al, 1999). Effects resulting from
chemotherapy treatment have been considered to be able to be examined more definitively
when examined in populations of children treated for acute lymphoblastic leukaemia. It is
suggested that as chemotherapy is often used in isolation, or with minimal radiation dosages,
findings may be more clearly attributed to chemotherapy than in populations of children with
brain tumor (Murdoch et al, 1999).
The phonological awareness abilities of children treated for posterior fossa tumor were
also explored in the present analysis. Despite previous reports indicating deficits in the
overlying literacy manifestations of reading and writing in populations of children with brain
trauma, inclusive of tumor (Hecaen, 1976; Cooper & Flowers, 1987; Van Dongen et al,
1995), no significant differences were evident between the present group of children treated
for posterior fossa tumor and their individually matched peers with respect to phonological
awareness skills. These results relate to the findings of Hudson and Murdoch (1992a), who
found no significant differences between a group of six children treated for posterior fossa
tumor and a control group, on measures of both reading and writing.
At an individual level, however, some deficits were noted in the children treated for
posterior fossa tumor. Similar to the presentation of widespread difficulties in high-level
language, Case 6 also demonstrated extensive difficulties in phonological awareness. Severe
deficits were demonstrated in both the Visual Rhyme and Phoneme Manipulation subtests.
Other areas of weakness included Nonword Spelling and Nonword Reading, Spoonerisms,
and Phoneme Detection. Cases 3, 4, and 7, however, presented with more specific areas of
weakness, as all three cases demonstrated difficulties specifically in the area of rhyme, as
assessed by the Spoken and Visual Rhyme subtests. In fact, reduced performance in the area
of rhyme was the only area of difficulty observed across all assessments inclusive of general
and high-level language and phonological awareness in the profile of Case 4. It was also
noted that while Case 6 did experience many difficulties in the area of literacy, the most
severe impairment observed was evident on the Visual Rhyme subtest. On the basis of this
preliminary pattern, it is suggested that rhyme may present as an area of weakness in the
phonological awareness skills of children treated for posterior fossa tumor, and warrants
further investigation in this population.
Particular difficulty in the Phoneme Manipulation subtest by two cases was also noted.
Case 6 and 7 were observed to perform poorly on this subtest Case 8, however, experienced
difficulty in the area of segmentation, with reduced performance on both the Syllable
Segmentation and Phoneme Segmentation subtests. It is possible that due to an absence of any
other difficulties across all measures of general and high-level language assessments and
other literacy subtests, Case 8 may have experienced difficulty understanding the
requirements of the tasks involving segmentation.
It is suggested that as phonological awareness skills are recognised to provide the
foundation for the development of literacy abilities such as reading, spelling, and writing (e.g.
Kahmi & Catts, 1991; MacDonald & Cornwall, 1995; Stackhouse & Wells, 1997), these
findings highlight the potential for reduced phonological awareness skills to lead to possible
impaired abilities in these areas. As few previous studies have specifically investigated the
contributing factors that may result in impaired pre-literacy or literacy skills in these
populations, it is suggested from the present study that location, as well as associated factors
such as hydrocephalus or post surgical management, may also impact on the development
phonological awareness skills. The impact of location is supported by an examination of fifty-
five children with dyslexia and their matched control participants by Fawcett el al. (1996), in
which cerebellar impairment was considered to be associated with dyslexia. However, it must
also be considered that periods of hospitalization in this population also occur and may
somewhat result from disturbances to schooling and therefore, the manifestation of these
types of disturbances. More indepth investigations, however, are required to further clarify
patterns of deficits and the contribution of specific tumor location on this skill.
CONCLUSION
The present study confirmed the presence of both general and high-level language deficits at a
group level in the performance of children treated for posterior fossa tumor. At an individual
level three children demonstrated evidence of specific general language deficit, largely in the
area of expressive language and syntax. Global high-level language deficits were noted in
three of ten participants, as well as more specific and subtle areas of high-level language
impairment (expressive semantic and lexical knowledge and problems solving) noted in three
additional cases.
Contributing factors noted included direct involvement of tumor location on cerebellar
connections involved in language function, fourth ventricle tumor location and the indirect
impact of hydrocephalus, tumor type, young age at diagnosis / treatment, and treatment
effects such as those associated with radiotherapy or chemotherapy. It was highlighted that
children with no initial observed general language difficulties in the acute phase post
management be monitored long-term. The present analysis also highlighted the importance of
investigating high-level language abilities in this population in the presence of intact general
language, due to observed high-level language abilities in children who did not present with
any evidence of general language deficits. Individual deficits identified in 50% of the group
children highlights the need for further analysis of both the pre-literacy awareness and later
literacy abilities of children treated for posterior fossa tumor.
REFERENCES
Akshoomoff, N. A., E. Courchesne, G. .A Press and V. Iragui (1992). Contribution of the

cerebellum to neuropsychological functioning: evidence from a case of cerebellar
degenerative disorder. Neuropsychologia, 30, 315-328.
Alajouanine, T. and F. Lhermitte (1965). Acquired aphasia in children. Brain, 88, 653-662.
Arya, S., E. F. Gilbert and T. A. Wiedrich (1986). Complications of treatment of neoplasia in
childhood, including second malignancies. In: Pathology of Neoplasia in Children and
Adolescents (M. Finegold, ed.), Vol. 18, pp. 433-464. W.B. Saunders Company,
Philadelphia.
Barrett, A. and S. S. Donaldson (1992). Radiation therapy. In: Cancer in Children: Clinical
Management (P. A. Voute, A. Barrett & J. Lemerle, eds.), Third Revised Ed., pp. 42-
50. Springer-Verlag, Berlin.
Becker, L. E. and V. Jay (1990). Tumors of the central nervous system in children. In:
Management of Childhood Brain Tumors (M. Deutsch. ed.), pp. 5-51. Kluwer
Academic Publishers, Boston.
Beldarrain, M. G., J. C. Garcia-Monco, J. M. Quintana, V. Llorens and E. Rodeno (1997).
Diaschisis and neuropsychological performance after cerebellar stroke. Eur J Neurol,
37, 82-89.
Broadbent, V. A., N. D., Barnes and T. K. Wheeler (1981). Medulloblastoma in childhood:

long-term results of treatment. Cancer, 48, 26-30.
Cohen, M. E., P. K. Duffner and C. K. Tebbi (1982). Brain tumors in children: diagnosis and
management. In: Major Topics in Pediatric and Adolescent Oncology (C. K. Tebbi,
ed.), pp. 240-289. G.K. Hall Medical Publishers, Boston.
Cooper, J. A. and C. R. Flowers (1987). Children with a history of acquired aphasia: residual
language and academic impairments. J Speech Hear Disord, 52, 251-262.
Danoff, B. F., S. Cowchock, C. Marquette. L. Mulgrew and S. Kramer (1982). Assessment of
the long-term effects of primary radiation therapy for brain tumors in children.
Cancer, 49, 1580-1586.
Dennis, M., B. J. Spiegler, C. R. Hetherington and M. L. Greenberg (1996).
Neuropsychological sequelae of the treatment of children with medulloblastoma. J
Neurooncol, 29, 91-101.
Di Chiro, G., E. Oldfield, D. C. Wright, D. De Michele, D. A. Katz, N. J. Patronas, et al.
(1988). Cerebral necrosis after radiotherapy and/or intraarterial chemotherapy for
brain tumors: PET and neuropathologic studies. AJR, 150, 189-197.
Dodd, B., A. Holm, M. Oerlemans and M. McCormick. (1996). Queensland University
Inventory of Literacy. St. Lucia: Department of Speech Pathology & Audiology, The
University of Queensland.
Duffner, P. K., M. E. Cohen and P. Thomas. (1983). Late effects of treatment on the
intelligence of children with posterior fossa tumors. Cancer, 51. 233-237.
Dunn, L. M. and L. M. Dunn (1997). Peahody Picture Vocabulary Test - Third Edition.
American Guidance Service, Circle Pines.
83-94.
Fabbro, F., R. Moretti and A. Bava (2000). Language impairments in patients with cerebellar
Fawcett, A. J., R. I. Nicolson and P. Dean (1996). Impaired performance of children with
dyslexia on a range of cerebellar tasks. Annals of Dyslexia, 46, 259-283.
Fisher, J. P. and J. M. Glenister (1992). The Hundred Pictures Naming Test The Australian
Council for Educational Research Ltd, Melbourne.
Gasparini, M., V. Di Piero, O. Ciccarelli. M. M. Cacioppo, P. Pantano and G. L. Lenzi
(1999). Linguistic impairment after right cerebellar stroke: a case report. Eur JNeurol,
6,353-356.
Gumbinas, M. (1983). Tumors of the central and peripheral nervous systems. In: Malignant
Diseases of Infancy, Childhood and Adolescence (A. J. Altman and A. D. Schwartz,
eds.), Second Ed., pp. 347-367. W.B. Saunders Company, Philadelphia.
Hecaen, H. (1976). Acquired aphasia in children and the otogenesis of hemispheric functional
specialization. Brain Lang, 3, 114-134.
Heideman, R. L., R. J. Packer, L. A. Albright, C. R. Freeman and L. B. Rorke (1993). Tumors
of the central nervous system. In: Principles and Practice of Pediatric Oncology (P.
A. Pizzo & D. G. Poplack, eds.), Second Ed., pp. 633-681. J.B.Lippincott Company,
Philadelphia.
Hoppe-Hirsch, E. (1993). Intellectual and psychosocial complications of posterior fossa tumor
surgery and supplemental (radiation therapy/chemotherapy) treatment. In: Posterior
Fossa Tumors (A. J. Raimondi, M. Choux & C. Di Rocco, eds.) pp. 194-200.
Springer-Verlag, New York.
Hudson, L. J. and B. E. Murdoch (1992a). Chronic language deficits in children treated for
posterior fossa tumor. Aphasiology, 6, 135-150.
Hudson, L. J. and B. E. Murdoch (1992b). Language recovery following surgery and CNS
prophylaxis for the treatment of childhood medulloblastoma: a prospective study of
three cases. Aphasiology. 6, 17-28.
Hudson, L. J., B. E. Murdoch and A. E. Ozanne (1989). Posterior fossa tumors in childhood:
associated speech and language disorders post-surgery. Aphasiology, 3, 1-18.
Hudson-Tennent, L. J. and B. E. Murdoch (1993). Retention of naming in children with an
acquired language disorder subsequent to treatment for posterior fossa tumor. J. Med.
Speech-Lang. PathoL, 1, 95-105.
Kahmi, A. and H. Catts (1991). Reading Disabilities: A Developmental Language
Perspective. Allyn and Bacon, Boston.
Kao, G. D., J. W. Goldwein, D. J. Schultz, J. Radcliffe, L. Sutton and B. Lange (1994). The
impact of perioperative factors on subsequent intelligence quotient deficits in children
treated for medulloblastoma/posterior fossa primitive neuroectodermal tumors.
Cancer, 74, 965-971.
Kun, L. E. and J. E. Moulder (1993). General principles of radiation therapy. In: Principles
and Practice of Pediatric Oncology (P. A. Pizzo & D. G. Poplack, eds.), Second Ed.,
pp. 273-302). J.B. Lippincott Company, Philadephia.
Levisohn, L., A. Cronin-Golomb and J. D. Schmahmann (2000). Neuropsychological
consequences of cerebellar tumor resection in children: cerebellar cognitive affective
syndrome in a paediatric population. Brain, 123, 1041-1050.
MacDonald, G. and A. Cornwall (1995). The relationship between phonological awareness
and reading and spelling achievement eleven years later. J Learn Disabil, 28, 523-527.
Maria, B. L. and J. Halpern (1989). Increased intracranial pressure. In: Handbook of Pediatric
Oncology (R. A. Gottlieb & D. Pinkel, eds.), pp. 165-166. Little, Brown and
Company, Boston.
Marien, P., S. Engelborghs and P. P. De Deyn (2001). Cerebellar neurocognition: a new
avenue. Ada Neurol Belg, 101, 96-109.
Marien, P., S. Engelborghs, B. A. Pickut. and P. P. De Deyn (2000). Aphasia following
cerebellar damage: fact or fallacy? Journal of Neurolinguistics, 13, 145-171.
Marien, P., J. Saerens, R. Nanhoe, E. Moens, G. Nagels, B. A. Pickut, et al. (1996).
Cerebellar induced aphasia: case report of cerebellar induced prefrontal language
phenomena supported by SPECT findings. J Neurol Sci, 144, 34-43.
McWhirter, W. R. and J. P. Masel (1987). Paediatric Oncology: An Illustrated Introduction.

Williams & Wilkins and Associates Pty Limited, Sydney.
Molinari, M , M. G. Leggio and M. C. Silveri (1997). Verbal fluency and agrammatism. Int
Rev Neurobiol, 41, 325-339.
Mulhern, R. K., J. Hancock, D. Fairclough, and L. Kun (1992). Neuropscyhological status of
children treated for brain tumors: a critical review and integrative analysis. Med
Pediatr Oncol, 20, 181 -191.
Muller, R., H. T. Chugani, O. Muzik, and T. J. Mangner (1998). Brain organization of motor
and language functions following hemispherectomy: a [ l5 O]-water positron emission
tomography study. Child Neurol, 13, 16-22.
Murdoch, B. E. and L. J. Hudson (1999a). Language disorders in children treated for brain
tumors. In: Communication Disorders in Childhood Cancer (B. E. Murdoch, ed.), pp.
55-75. Whurr Publishers, London.
Murdoch, B. E. and L. J. Hudson (1999b). Language recovery following treatment for
paediatric brain tumors. In: Communication Disorders in Childhood Cancer (B. E.
Murdoch, ed.), pp. 76-88. Whurr Publishers, London.
Murdoch, B. E. and L. J. Hudson (1999c). Variability in patterns of language impairment in
children following treatment for posterior fossa tumor. In: Communication Disorders
in Childhood Cancer (B. E. Murdoch, ed.), pp. 89-125. Whurr Publishers, London.
Murdoch, B. E. and L. J. Hudson (1994). Differential language outcomes in children
following treatment for posterior fossa tumors. Aphasiology, 8, 507-534.
Murdoch, B. E., L. J. Hudson and D. L. Boon (1999). Effects of treatment for paediatric
cancer on brain structure and function. In: Communication Disorders in Childhood
Cancer (B. E. Murdoch, ed.), pp. 21-54. Whurr Publishers, London.
Neidhardt, M. K., M. Bamberg and H. Riehm (1986). Medulloblastoma: strategies for
therapy. In: Malignant Neoplasms in Childhood and Adolescence (H. Riehm, ed.).
Vol. 18, pp. 296-315. Karger, Basel.
Oki, J., S. Takahashi, A. Miyamoto and Y. Tachibana (1999). Cerebellar hypoperfusion and
developmental dysphasia in a male. Pediatric Neurology, 21, 745-748.
Packer, R. J., R. A. Zimmerman, and L. T. Bilaniuk (1986). Magnetic resonance imaging in
the evaluation of treatment-related central nervous system damage. Cancer, 58, 635-
640.
Plowman, P. N. (1992). Tumors of the central nervous system. In: Paediatric Oncology:
Clinical Practice and Controversies (P. N. Plowman & C. R. Pinkerton, eds.), PP-
240-267. Chapman & Hall Medical, London.
Riva, D. (1998). The cerebellar contribution to language and sequential functions: evidence
from a child with cerebellitis. Cortex, 34. 279-287.
Riva, D. (2000). Cerebellar contribution to behaviour and cognition in children. Journal of
Riva, D. and C. Giorgi (2000a). The cerebellum contributes to higher functions during
development: evidence from a series of children surgically treated for posterior fossa
tumors. Brain, 123, 1051-1061.
Riva, D. and C. Giorgi (2000b). The contribution of the cerebellum to mental and social
functions in developmental age. Hum Physiol. 26, 21-25.
Sands, S. A., W. G. van Gorp, and J. L. Finlay (1998). Pilot neuropsychological findings from
a treatment regimen consisting of intensive chemotherapy and bone marrow rescue for
young children with newly diagnosed malignant brain tumors. Childs Nerv Syst, 14,
587-589.
Schatz, J., S. Hale and J. Myerson (1998). Cerebellar contribution to linguistic processing
efficiency revealed by focal damage. JInt Neuropsychol Soc, 4, 491-501.
Schmahmann, J. D. and J. C. Sherman (1998). The cerebellar cognitive affective syndrome.
Brain, 121,561-579.
Scott, R. B., C. J. Stoodley, P. Anslow, C. Paul, J. F. Stein, E. M. Sugden, et al. (2001).
Lateralized cognitive deficits in children following cerebellar lesions. Dev Med Child
Neurol, 43, 685-691.
Segall, H. D., S. Batnitsky, C. Zee. J. Ahmadi, R. Bird, and M. E. Cohen (1985). Computer
tomography in the diagnosis of intracranial neoplasms in children. Cancer, 56, 1748-
1755.
Semel, E., E. H. Wiig, and W. A. Secord (1995). Clinical Evaluation of Language
Fundamentals - Third Edition. The Psychological Corporation, San Antonio.
Shearer, W. M. (1982). Research Procedures in Speech, Language, and Hearing. Williams &
Wilkins, Baltimore.
Silveri, M. C. and S. Misciagna (2000). Language, memory, and the cerebellum. Journal of
Silveri, M. C , M. G. Leggio and M. Molinari (1994). The cerebellum contributes to linguistic
production: a case of agrammatic speech following a right cerebellar lesion.
Neurology, 44, 2047-2050.
Silveri, M. C , S. Misciagna, and G. Terrezza (2001). Right side neglect in right cerebellar
lesion. J Neurol Neurosurg Psychiatry, 71, 114-117.
Silverman, C. L. and P. R. M. Thomas (1990). Long-term neuropsychologic and intellectual
sequelae in brain tumor patients. In: Management of Childhood Brain Tumors (M.
Deutsch, ed.), pp. 493-500. Kluwer Academic Publishers, Boston.
Silverman, C. L., H. Palkes, B. Talent, E. Kovnar, J. W. Clouse and P. R. M. Thomas (1984).
Late effects of radiotherapy on patients with cerebellar medulloblastoma. Cancer, 54,
825-829.
Stackhouse, J. and B. Wells (1997). Children's Speech and Literacy Difficulties: A
Psycholinguistic Framework. Whurr Publishers, London.
Steinlin, M., M. Styger and E. Boltshauser (1999). Cognitive impairments in patients with
congenital nonprogressive cerebellar ataxia. Neurology. 53, 966-973.
Sue, E., C. Kalifa, R. Brauner, J. L. Habrand, M. J. Terrier-Lacombe, G. Vassal, et al. (1990).

Brain tumors under the age of three. The price of survival: a retrospective study of 20
long-term survivors. Ada Neurochir, 106, 93-98.
Tait, D. M., C. C. Bailey, and M. M. Cameron (1992). Tumors of the central nervous system.
In: Cancer in Children: Clinical Management (P. A. Voute, A. Barrett & J. Lemerle,
eds.), Third Revised Ed., pp. 184-206. Berlin: Springer-Verlag.
van Eys, J. (1991). Malignant tumors of the central nervous system. In: Clinical Pediatric
Oncology (D. J. Fernbach & T. J. Vietti, eds.), Fourth Ed., pp. 409-426. Mosby - Year
Book, Inc., St.Louis.
Van ongen, H. R., M. C. B. Loonen, and K. J. VanDongen (1985). Anatomical basis for
acquired fluent aphasia in children. Ann Neurol, 17, 306-309.
Wiig, E. H. and M. Secord (1989). Test of Language Competence - Expanded Edition. The
Psychological Corporation: San Antonio.
Wiig, E. H. and M. Secord (1992). Test of Word Knowledge. The Psychological Corporation,
San Antonio.
Wiig, E. H., W. A. Secord and E. Semel (1992). Clinical Evaluation of Language
Fundamentals - Preschool. The Psychological Corporation, San Antonio.
Wilson, C. B. (1975). Diagnosis and surgical treatment of childhood brain tumors. Cancer,
35, 950-956.
Zachman, L., M. Barrett, R. Huisingh, J. Orman and C. Blagden (1991). Test of Problem
Solving - Adolescent. LinguiSystems, East Moline.
Zachman, L., R. Huisingh, M. Barrett, J. Orman, and C. LoGiudice (1994). Test of Problem
Solving - Elementary, Revised. LinguiSystems, East Moline.
Language Development and Cerebellar Malformations 127
LANGUAGE DEVELOPMENT IN CHILDREN WITH

CEREBELLAR MALFORMATIONS
Renato Borgatti, Alessandro Tavano, Guido Cristofori and Franco Fabbro
"E.Medea " Scientific Institute, Polo del Friuli Venezia Giulia and Bosisio Parini (LC), Italy
University ofUdine, Italy
Abstract We investigated language development in three participants with cerebellar

malformations. The extent of the lesions varies from near total absence of the cerebellum to
hypoplasia of one cerebellar hemisphere. The effect of cerebellar malformations on language
development is not yet known. Acquired focal cerebellar lesions in adults may determine mild
expressive language deficits, which tend to regress rather rapidly over time. However, subtle deficits
may persist. We show that malformations localized to one or both cerebellar hemispheres and to the
vermis may result in permanent receptive and expressive language deficits. The degree of impairment
varies from subtle deficits (e.g., right cerebellar hypoplasia) to severe disorders of language and
communication (autistic-like behavior associated to vermal hypoplasia). However, we document that
near total absence of the cerebellum does not prevent general cognitive functioning and language
acquisition to a sufficient extent. Considerations about the plasticity of the human brain suggest that
cortical and subcortical mechanisms may compensate for the cerebellar contribution to language to a
considerable extent.
Key words: cerebellum, cerebellar malformations, cerebellar aplasia, language development

INTRODUCTION
Over the last 20 years, besides being attributed a major role in mediating motor functions
(Dow & Moruzzi, 1958), the cerebellum has been consistently involved in the modulation of
non-motor functions (Leiner et al, 1986; Leiner et al, 1991; Schmahmann, 1991; Fiez, 1996;
Fabbro, 2000). Recently, Schmahmann and Sherman (1998) identified a "Cerebellar
Cognitive Affective Syndrome" in a group of adult patients with acquired cerebellar lesions.
This syndrome is characterized by: 1) impairment of executive functions; 2) impaired spatial
cognition; 3) affect change; and 4) expressive linguistic deficits. Investigations involving
children with acquired cerebellar lesions showed a similar pattern of deficits (Levinsohn et al,
2000; Riva & Giorgi, 2000). Levinsohn et al. (2000) studied 19 children who underwent
excision of cerebellar tumor. Results showed that younger children were the least likely to
show deficits in language, cognition or affect. Similarly, Riva and Giorgi (2000) studied 26
children who underwent surgery for cerebellar hemisphere or vermal tumor. Vermal lesions
determined two different outcomes: 1) postsurgical mutism, which evolved into speech or
expressive language disorders similar to agrammatism; 2) disturbances of social and
communicative skills, ranging from irritability to autistic-like behavior. Excision of tumors
confined to the cerebellar hemispheres led to selective language deficits. Children with right
cerebellar tumors presented with disturbances of auditory sequential memory and language
disorders, particularly on the expressive side, as is reflected by low MLU scores (mean length
of utterance), an index of expressive grammar (cf. the Methods section). Children with left
cerebellar tumors showed a marked impairment in lexical access but on tasks tapping
comprehension and expressive skills. They also showed a less marked decrease in executive
functions than children with right cerebellar tumors.
In a review of studies on acquired cerebellar hemisphere lesion and subsequent language
disturbances in adults, Marien et al. (2001) argued for language lateralization in the right
cerebellar hemisphere pointing out an involvement of the right cerebellar hemisphere in: 1)
verbal working memory; 2) articulatory planning; 3) semantic and phonological word
retrieval; 4) syntactic processing. Some neuroimaging data seem to support this proposal.
Patterns of cerebellar and cortical (notably Broca's area and Supplementary Motor Area,
SMA) activation have been documented in healthy participants performing a word association
task, namely producing appropriate verbs for given nouns (Petersen et al, 1989; Petersen &
Fiez, 1993; Desmond & Fiez, 1998). At the neuroanatomical level, the cerebellum has
extensive input and output connections to the frontal and prefrontal cortices (so-called
'Frontal lobe system'). Input connections run from association areas in the frontal, temporal
and parietal lobes through the pontine nuclei and the red nucleus - inferior olive system
(Leiner et al, 1993). Output connections to the frontal and prefrontal cortices (Broca's area,
SMA) run through the thalamus, in particular the ventrolateral, ventral-anterior and centro-
median nuclei (Leiner et al, 1993). Cerebrocerebellar connections are organized in a crossed
pattern, whereby the right cerebellar hemisphere is connected to the left cerebral hemisphere,
and the left cerebellar hemisphere is connected to the right cerebral hemisphere.
Based on these considerations, some researchers have mentioned the phenomenon of
'crossed cerebello-cerebral diaschisis' to explain the development of expressive agrammatic
and expressive-receptive syndromes following right cerebellar hemisphere damage (Fabbro et
al, 2000a; Marien et al, 1996; Molinari et al. 1997; Silveri et al, 1994). An important issue
is that the extent of expressive agrammatic symptoms following acquired cerebellar vascular
lesions tends to regress rapidly over time (Marien et al., 2000; Silveri et al., 1998). The
phenomenon of 'crossed cerebello-cerebral diaschisis' is a temporary functional deactivation
of frontal lobe areas subsequent to damage to contralateral cerebellar hemisphere. As
neurofunctional activation becomes normal, language deficits rapidly regress in severity and
extent. Another important issue is that language deficits following cerebellar damage are
sometimes difficult to detect through administration of formal language tests. They can
become manifest at a qualitative analysis of spontaneous or descriptive speech (Fabbro et al.,
2004). Zettin et al. (1997) described the case of a 46-year-old highly educated right-handed
patient who developed severe expressive agrammatism (omission of function words) after a
hemorrhagic stroke involving the right cerebellar hemisphere. However, such deficit was
manifest only in spontaneous speech and the patient is reported to have fully recovered
language three months post-onset. Besides rapidity in regression over time and limitations in
deficit extent, this finding outlines a third important issue, that is the possible persistency of
residual linguistic difficulties. Unlike the patient studied by Zettin et al. (1997), the patients
studied by Fabbro et al. (2004) showed minor linguistic deficits that were still detectable
many years after insult. The patients were aware of such difficulties and their impairing effect
on daily activities and working abilities, hi conclusion, although subtle and transient, language
disorders following acquired focal or large cerebellar lesions point to some involvement of the
cerebellum in language processing (Fabbro, 2000).
The effects of cerebellar aplasia or hypoplasia on higher order cognitive functions are yet
to be investigated. As for language, two main issues may be raised. First, there are no
available data about the influence of cerebellar malformations on language acquisition.
Malformation-associated cerebellar lesions are likely to induce a consistent reorganization of
cerebello-cortical connections. Notably, the cerebellum seems to partly sustain the acquisition
and processing of procedural aspects of language such as phonological and morphosyntactic
rules (Molinari et al, 1997b; Ullman, 2001). Therefore, malfonnative lesions may delay or
disrupt normal language acquisition patterns. Second, it is not known whether malformations
affecting single cerebellar structures have differential effects on language acquisition. In the
case of near total absence of the cerebellum - indeed, a very rare event - the few case studies
reported in the literature (Stewart, 1956; Sener & Jinkins, 1993; Gardner et al., 2001)
documented that initially patients show a very marked delay in development but slowly and
steadily recover motor, affective and cognitive functions. However, these studies lack detailed
information on language development and language skills in adult age. Linguistic and
communicative outcomes of other types of cerebellar malformations show great variability
depending, for example, on the extent of vermian involvement. A more important

involvement of the vermis seems to be linked to severe pervasive impairments in social and
communication skills (autism or autistic-like behavior), although mainly neuropsychological
and linguistic deficits have been described (Gilberg & Coleman, 1992; Courchesne, 1997).
In this study we discuss the general development and, in particular, language development
in three patients with cerebellar malformations, ranging in localization and extent from near
total absence of the cerebellum to a large malformative lesion limited to the right cerebellar
hemisphere. Our aim is to provide some preliminary answers to the following three questions:
1) Is the cerebellum involved in language acquisition? 2) What is the effect of malformative
lesions on language acquisition and language development in adult age? 3) What are the
characteristics of language deficits associated to cerebellar malformations?
METHODS
Participants
Three participants were selected from a pool of 20 patients (15 adults, 5 children) who were
referred to the "E. Medea" Scientific Institute in Bosisio Parini (Lecco - Italy) over a 13-year
period (1991-2003). The pool of 20 patients was selected among a total number of 156
participants with cerebellar lesions of different etiology. Exclusion criteria were: 1) suspected
perinatal cerebellar lesion; 2) degenerative cerebellar disease; 3) diagnosis of Joubert
syndrome; 4) diagnosis of Dandy-Walker syndrome; 5) diagnosis of Chiari syndrome Type I
and II; 6) a clinical history of seizures. The pool of 20 participants then underwent extensive
clinical, neuroradiological and neuropsychological investigations as part of a current study
aiming at verifying whether a Cognitive Affective Syndrome is also associated to
malformative cerebellar lesions apart from acquired cerebellar lesions (see Schmahmann &
Sherman, 1998).
Three male patients were selected for the present study: Case 1, aged 32.01, right-handed,
presents with near total absence of the cerebellum; Case 2, aged 10.03, right-handed, presents
with mild vermian hypoplasia and mega cisterna magna; Case 3, aged 21.10, left-handed,
shows severe hypoplasia of the right hemisphere.
Clinical, radiological and neuropsychological assessments
Patients were administered a standard neurological examination based on the "International

Cooperative Ataxia Rating Scale" (Trouillas et al, 1996). They also received an EEG and
MRI scanning (0.5 or 1.5 Tesla, Tl, T2-weighted spin-echo sequences and protonic density,
along axial and coronal sections of 1.3 mm). Images were analyzed according to the Atlas by
Schmahmann et al. (1999) by one neuroradiologist blind to the patients' identity and clinical
history. In each case the malformation was classified into one of the following categories
according to Altman et al. (1992): Type I: Complete cerebellar agenesis (case 1); Type III:
Hypoplasia involving the vermis and both hemispheres (case 2); Type IV: Malformation
involving cerebellar hemispheres (one or both) (case 3).
Neuropsychological assessment included the following tests:
IQ: Wechsler Adult Intelligence Scale (WAIS, Weschler, 1974), Wechsler Adult
Intelligence Scale-Revised (WAIS-R) (Weschler, 1997); Wechsler Intelligence Scale for
Children-Revised (WISC-R) (Wechler 1986); Wechsler Preschool and Primary Scale of
Intelligence (WPPSI) (Weschler, 1973); Griffiths Scale (Griffths, 1986), Stanford-Binet
Intelligence Scale, L-M form (Terman & Merrill, 1968).
Visuospatial functions: Test of Visual-Perceptual Skills (non-motor) (Gardner, 1996);
Line Orientation Judgment (Benton & Benton, 1983); Temporal Orientation Test (Spinnler &
Tognoni, 1987); Rey Complex Figure "A" (Rey, 1967); Constructive Apraxia Test (Spinnler
& Tognoni, 1987); Developmental Test of Visual Motor Integration Test (VMI) (Beery &
Buktenica, 2000).
Executive functions: (Planning and Control): Wisconsin Card Sorting Test (Heaton et al,
2000); Tower of London Test (Shallice, 1982).
Memory: Wide Range Assessment of Memory and Learning (WRAML) (Adams &
Sheslow, 1990); Rey Complex Figure "B" (Rey, 1967); Corsi Span (Corsi, 1972); Digit span
(Spinnler & Tognoni, 1987); Word span, forward (Spinnler & Tognoni, 1987); Long-term
Memory Test (Spinnler & Tognoni, 1987).
Language assessment was organized as follows. Case 2 was administered the Esame del
Linguaggio dai 4 ai 12 anni, a standardized language battery for children aged 4-12 (Fabbro,
1999), which includes tests of comprehension, repetition, production and articulation.
Comprehension: Auditory and Verbal Discrimination tests (Bearzotti, Lilli & Fabbro, in
press); British Picture Vocabulary Scale (De Agostini et al., 1998); Token test (short-version,
21 items) (Vender et al, 1981; De Agostini et al, 1998); Test di Comprensione Grammaticale
per Bambini (TCGB), a test of grammatical comprehension in children (Chilosi & Cipriani,
1995); Repetition: Word and non-word repetition (De Agostini et al, 1998); Sentence
repetition (Ferrari et al, 1981; Vender et al, 1981); Production: Naming (De Agostini et al,
1998); Semantic Fluency (De Agostini et al, 1998); a picture description task, the Bird Nest
Story (Paradis, 1987). Case 1 and 3 were administered the Bilingual Aphasia Test, Part B
(Paradis 1987, 1999). The BAT, Part B consists of 32 subtests (427 items) and allows an in-
depth investigation of linguistic levels and linguistic skills. This test comes with no normative
values for the Italian population. We thus compared the performances of Cases 1 and 3 to the
values obtained for a control group of 19 participants (12 males, 7 females) with high
educational level (equivalent to MSc or A level degree), mean age 35.6 (age range 31-42),
described in detail in Fabbro et al. (2004). Reading: Prove di lettura MT (Cornoldi & Colpo
1981), age-appropriate reading tasks standardized for an Italian school-age sample.
CASE 1, NEAR TOTAL ABSENCE OF THE CEREBELLUM
Clinical data
Case 1 is a 32 year old right-handed male who, despite near total absence of the cerebellum,
presents with mild mental retardation and a harmonious profile as far as cognitive functions
are concerned. An MRl at the age of 31 evidenced near total absence of the cerebellum with
only a tiny cerebellar remnant and flat ventral pons (cf. Figures 1 and 2).
Figure 1 — MRl: Near total absence of the cerebellum with only a tiny cerebellar remnant positioned
superiorly within the posterior fossa. Loss of prominence of the ventral pons is also evident (coronal
section, Tl- weighted).
Figure 2 — MRl: Same lesion (sagittal section, Tl- weighted).

Standard neurological examination based on the "International Cooperative Ataxia Rating

Scale" (Trouillas et al, 1996) yielded a score of 29/100, evidencing cerebellar ataxia (Postural
and gait disturbances = 8/34; Limb ataxia = 16/52; Speech = 3/8; Oculomotor disorders =
2/6).
Development
Case 1 presents with no familiar history of neurologic or psychiatric diseases. He was born
from first-degree cousins, pregnancy and delivery were normal. Weight at birth was 2550gr.
The participant's motor and cognitive development soon appeared to be profoundly
compromised. However, he presented with slow but steady progress in all functional domains
up to almost adequate levels.
Motor skills were severely impaired during the first years of life due to a marked
hypotony. He did not achieve sitting until after the age of 2; as a child, he dragged himself
along and achieved crawling at 5 years of age. He achieved standing and walking if sustained
when he was 10. Independent walking was reached at the age of 22. His motor skills
developed further and at the time of testing a neurological examination detected only a
marked ataxia when walking, a mild dysmetria and intentional tremors (cf. neurological
assessment above).
He was first seen at our Institute when he was 4.06 years old. At that time he showed an
autistic-like behavior, he did not actively search for other people, spending time on his own
and being engaged in stereotyped and repetitive activities. If cared for, he did not reject
attention. Up to age of 8-10 years his personality was extremely fragile and emotionally
passive: socialization was hardly possible. During adolescence his social skills improved and
he started actively interacting with parents and other significant adults. At present, affect and
behavior are normal, without psychopathological signs.
Language development was also markedly impaired. He uttered his first words after age 2
years. Until age 7 years he produced only one-word sentences. He was able to correctly
pronounce all single phonemes of the Italian language but articulation of words was still
effortful and often incorrect. At age 12 years he was able to produce a simple sentence
(Subject - Predicate - Complement), short and contextually linked but sufficient for
communication with peers. He had learnt to read and write in capital letters with many
spelling errors. Language skills steadily progressed over time up to complete functionality.
Regarding quality of life, since he was 5 years of age he has been attending a special
school at our Rehabilitation Institute where he lived during the school year. Teachers reported
that his performances significantly improved if someone helped him to plan the task and
reinforced his attention steadily. He needed constant stimulation as he seemed unable to
initiate a requested or desired action on his own. When he was 17 he entered a family
residence of our Institute in the framework of a rehabilitation program centered on learning of
basic self-sufficiency skills, and he still lives there today. He is completely independent in his
personal life, being able to perform a simple work activity (assembling electronic
components), knows the value of money and is able to shop. He also has personal interests in
music and movies, and keeps himself informed by reading music and film magazines. He is
able to use a video recorder and a mobile phone to call and send text messages to his family
and friends. He is also able to travel independently by bus from the community where he lives
(in a valley in Lombardy, Northern Italy) to a near-by city (Bergamo, Northern Italy), where
he stops for lunch and shops for the friends he will be visiting.
Neuropsychological assessment at age 32.01
Intelligence Quotient (IQ). Administration of the WAIS (Weschler, 1974) evidenced a

mild mental retardation with a harmonious profile: VIQ = 72, PIQ= 67, FIQ= 68. The patient
showed more marked impairments on the following subtests: Arithmetics (standardized
score= 3), Coding (standardized score = 3), Similarities (standardized score = 4), and Picture
Arrangement (standardized score = 4).
Visuospatial functions. On the Rey Complex Figure "A" Test the participant scored
comparably to children with a mental age of 6.06 years (Copy) and 7.06 years (Memory),
showing a marked impairment in appreciating the organizing structure of the figure
(Schmahmann & Sherman, 1998). Corsi Test yielded a mildly deficient result (<1 SD; raw
score= 4; Mean = 5.11 for the 40-49 age group, SD= 1.1). No signs of constructive apraxia
were found (raw score = 1 3 ; Mean = 13.2 for the 40-49 age group, SD = 1,48).
Executive functions. The Tower of London Test score disclosed a marked impairment in
attention (<2 SD; raw score= 8.62; equivalent score = 0 for the age group of 40 years).
However, on the Wisconsin Card Sorting Test he did not show any impairment in executive
functions (Total errors = 18%, Mean and standard deviation = 24.32 ±15.11; Total categories
completed = 6, Mean and standard deviation = 5±1.6).
Memory. On the Verbal Span Test the patient scored within normal limits (raw score = 4;
Mean = 4.75 for the 40-49 age group, SD = 0.86); however, administration of the Long-term
Memory Test (Spinnler & Tognoni, 1987) highlighted a markedly defective performance (<1
SD; raw score= 8.5; Mean = 13.32 for the 40-49 age group, SD= 2.65). Procedural memory as
measured on the Star Test was unimpaired.
Language. Voice was hypophonic, scanned, nasalized (non-occluding velum). The
phonological and phonetic deficits and the irregular patterns of articulatory breakdown
suggest a diagnosis of ataxic dysarthria. Spontaneous speech is characterized by phonologic
and phonetic problems to a greater extent than descriptive speech (ciamo instead of siamo,
"we are"; /zi/ instead of the Italian affermative particle /si/).
The patient was administered the Italian adaptation of the Bilingual Aphasia Test, part B
(Paradis 1987, 1999). Results show a marked impairment of syntax (72% correct) and
semantics (85%), with subtler deficits of lexicon (93%), morphology (95%) and phonology
(96%). With regard to language skills, results show a marked impairment in writing (60%
correct), comprehension (78%), sentence construction (86%), reading (89%), with a somewhat
milder impairment of lexical access (92%). A closer investigation of his performance on the
BAT subtests shows marked deficits in Mental Arithmetic (13% correct), Sentence dictation
(20%), Auditory comprehension of a short story (40%), Reading Comprehension of sentences
(50%), Syntactic comprehension (70%), Antonyms (80%).
Reading. Administration of the MT reading tests evidenced a score average for a 5lh grade
as far as correctness, rapidity and comprehension are concerned. The analysis of a sample of
descriptive speech (The Bird Nest Story; Paradis, 1987) showed numerous morphosyntactic
errors (omissions of function words, substitution of bound grammatical morphemes) and
semantic paraphasias. In addition, the patient was not able to appreciate the "gist" of the story.
In conclusion, the patient seems to have acquired language to a considerable extent and is
able to use it for everyday activities. However, he presents with marked permanent
impairments of receptive and expressive language and of speech, with more accentuated
comprehension difficulties, probably due to syntactic processing deficits.
CASE 2, DIFFUSE VERMIAN HYPOPLASIA
Clinical data
Case 2 is a 10 year old right-handed boy. He presents with moderate mental retardation, with a
harmonious profile. An MRJ revealed Mega Cisterna Magna with diffuse vermian hypoplasia.
Cerebellar hemispheres are near to normal (cf. Figure 3). The patient was diagnosed as having
a Type III malformation, according to Altman et al. (1992).
Figure 3 — MRI: Mega Cisterna Magna with diffuse vermian hypoplasia. Cerebellar hemispheres are
nearly normal (coronal section, Tl- weighted).
Development
The patient was referred to our Institute when he was 9.01 for severe language delay and
reduced social and affective skills. Neuromotor development was slightly retarded. He
reached independent walking when he was 16 months of age. Development of affective,
cognitive and linguistic skills was markedly impaired. He had been living in an English-
speaking country until he was 4 years old and was thus exposed to both English and Italian
(parent's language) input. A marked communicative and linguistic delay was reported by
operators at the English-speaking kindergarten. He uttered his first words in Italian when he
was 5 years of age. Language production improved over time, but the patient soon showed
autistic-like behaviors.
Clinical and neuropsychological assessment at age 9
Clinical assessment. A neurological examination evidenced mild hypotonia, hypodiadococine-

sis, and mildly deficient fine and gross motor skills. Karyotype was normal and a diagnosis of
Fragile X syndrome was excluded. The EEG showed epileptiform abnormalities (SW) over
the central-posterior regions, bilaterally, mainly on the right side, and over the anterior regions
in NREM sleep (cf. Figure 4). On the psychomotor evaluation he showed a severe delay as he
could not complete the proposed tasks.
Figure 4 — Epileptiform abnormalities (SW) over the central-posterior regions, bilaterally, mainly on
the right side, and over the anterior regions during NREM sleep.
IQ. On the Stanford-Bmet Scale the patients obtained a full-scale IQ score of 50,
comparable to a mental age of 4.06.
Language. Articulation showed some phonological problems (phonemic simplifications).
His descriptive speech (Birds' Nest Story; Paradis, 1987) was characterized by numerous
morphosyntactic errors (omissions of free grammatical morphemes), anomic errors, phonemic
paraphasias. Fluency (words per minute) was markedly reduced (= 21.3, < 2 SD) as was MLU
= 3 (<2SD). The patient substituted the description of one story character for another,
demonstrating an impaired ability at understanding the story structure. Spontaneous speech
revealed inappropriate language switching episodes (from Italian to English in interaction with
a non English-speaking Italian operator), tangential answers and perseverations (Fabbro et al.,
2000b). Grammatical structure was more preserved, with few omissions of free grammatical
morphemes. However, answers were often in the form of a list, thereby limiting the extent of
possible morphosyntatic errors.
Clinical and neuropsychological assessment at age 10
At the age of 10 Case 2 received a complete clinical reassessment. The neurological

examination confirmed mild hypotonia, adiadococinesis, and mildly deficient fine and gross
motor skills. The psychomotor evaluation confirmed a severe delay as the child could not
complete the proposed, age-appropriate, tasks.
IQ. On the WISC-R (Weschler, 1986) he obtained the following scores: VIQ = 51; PIQ =
49; FIQ = 49, which confirm the moderate retardation evidenced on the first evaluation.
Language. Language scores on formal tests showed a change in profile, with word and
non-word repetition within normal limits and a decrease in semantic comprehension and
action naming. Scores were <2 SD on the following tests: syntactic comprehension (Token
Test), grammatical comprehension (TCGB), sentence repetition, object naming, action
naming, semantic fluency. Semantic comprehension was <1 SD. Language articulation
showed only mild inaccuracies. His descriptive speech (Birds' Nest Story) was characterized
by poor narrative skills and omission of bound grammatical morphemes. The patient delivered
an empty story, perseverating on just one aspect (whether one character falls or not) and
applying it to all figures inappropriately. Neologisms and inappropriate list-like object
descriptions (IT: "Un'auto ha tre targhe", ENG: "A car has three number plates") were present
in his spontaneous speech.
In conclusion, Case 2 presents with a moderate retardation, with language and
affect/relational skills being the most severely impaired functional areas. Remarkably, on
continuous and regular language therapy, the patient showed a regression in language
competence, while peripheral linguistic components such as articulation improved (as
evidenced by testing word and nonword repetition in the second assessment).
13 8 Neurogenic Language Disorders in Children
CASE 3, SEVERE HYPOPLASIA OF THE RIGHT CEREBELLAR HEMISPHERE
Clinical data
Case 3 is a highly educated 21 year-old left-handed male who, despite severe hypoplasia of
the right cerebellar hemisphere, presents with normal cognitive development. He was referred
to our Institute following a sudden onset of two hyperventilation crises associated with
postural tremors to the right forearm and diurnal drowsiness (onset of postural tremor to the
right forearm at 15 years of age). An MRi (Spin Echo) evidenced a severe hypoplasia of the
right cerebellar hemisphere (which is practically absent) (cf. Figure 5). Case 3 has been
classified as showing a Type IV category malformation, which includes all types of
malformations limited to one or both cerebellar hemispheres.
Figure 5 — Severe hypoplasia of the right cerebellar hemisphere. Axial MRI shows near total
absence of the right cerebellar hemisphere with an asymmetric IV ventricle which is more enlarged to
the right side (coronal section, T2- weighted).
Development
His parents reported some difficulties in acquiring reading and writing skills. One episode of
sudden and transient partial loss of consciousness at the age of 14 years was reported,
followed by headache. Language development was reported to be normal, with some initial
difficulties in acquisition of reading and writing skills. The patient is left-handed and tends
not to actively use the right hand also in tasks requiring bi-manual coordination. The
neurological examination revealed the following cerebellar signs: appearance of distal kinetic
tremor in the right upper limb under emotional stress conditions, tendency to move left when
walking with eyes closed, tremor and right dysmetria in finger to nose test,
hypodiadococinesis. At present Case 3 attends University classes with good results. He also
practices various types of sports.
Neuropsychological assessment
IQ. On the WAIS-R, the patient shows a harmonious profile (VIQ=107, PIQ=100,
FIQ=105), with a marked difficulty in performing the figure reconstruction task.
Visuospatial functions. On the Rey Complex Figure "A" Test the participant scored
comparably to a mental age of 6.00 years (Copy), 5.00 years (Memory immediate), 4.06 years
(Memory delayed), showing a marked impairment in appreciating the organizing structure of
the figure. On the Corsi Test he performed within the norm (Span = 6). No signs of
constructive apraxia were noted (raw score = 14; Mean = 13.2 for the 40-49 age group, SD =
1.48).
Executive functions. The Wisconsin Card Sorting Test did not evidence any impairment in
executive functions (Total errors = 19%, Mean and standard deviation = 24.32±15.11; Total
categories completed = 6, Mean and standard deviation = 5±1.6).
Memory. Mnestic functions were not impaired as assessed by the Verbal Span Test (raw
score = 5; Mean = 4.75 for the 40-49 age group, SD = 0.86); he performed in the normal range
on the Long-term memory test (raw score = 12.7; Mean = 13.32 for the 40-49 age group, SD =
2.65).
Language. The patient was administered the Italian adaptation of the Bilingual Aphasia
Test (Paradis 1987, 1999). He showed a moderate impairment in semantics (83% correct),
mainly pertaining to lexical knowledge (antonyms, synonyms). With regard to language skills,
he had subtle deficits in reading (95% correct) and comprehension (97% correct). A closer
look at his performance on the BAT shows that the patient had difficulties with reading
comprehension of a short story (50% correct), listening comprehension of a short story (80%),
synonyms (80%) and repetition of sentences (86%). The analysis of a sample of descriptive
speech (Bird Nest Story; Paradis, 1987) reveals fluent speech and grammatically well
articulated discourse (MLU = 14.2; subordinate clauses = 6), a higher than normal type/token
ratio (0.69), some morphosyntactic errors (2 omissions of function words) and 2 instances of
word-finding difficulty. In addition, the patient was not able to appreciate the "gist" of the
story. Voice was normal. Spontaneous speech manifested extremely rare morphosyntactic
difficulties (in a sample of 100 utterances, 2 substitutions of bound grammatical morphemes:
IT "Sono tornata (FEM) a Milano", ENG "I went back to Milan", instead of IT "Sono tomato
(MASC) a Milano".
In conclusion, the patient seems to have very subtle language difficulties, which become
more manifest in tasks tapping descriptive and spontaneous speech or lexical search and
decision (such as choice of a synonym). The higher than normal type/token ratio is indicative
of a conscious control over speech production.
DISCUSSION
In this study we set out to study language development in three subjects with different types
(localization and extent) of congenital cerebeliar malformations. We also presented
preliminary data on the characteristics of language disorders (e.g., receptive vs. expressive) in
congenital cerebeliar malformations as distinct from acquired cerebeliar lesions. Our findings
show that two out of three cases (Cases 1 and 2) have a marked language delay. Case 1
presented with a continuous although remarkably unusual language development, with
language acquisition phenomena occurring also at an adult age. In addition, Case 2 presented
with an episode of language regression in the presence of EEG epileptiform discharge. Case 3
presented with no language delay, although at present very subtle language deficits may be
detected. The different outcome between Cases 1 and 2 and Case 3 may be correlated to lesion
localization and extent. In Case 1 permanent deficits, more marked on the receptive side
(syntactic comprehension), are evident. The patient also presents with ataxic dysarthria,
phonological and phonetic deficits, and a scanned voice. Case 2 shows a marked language
delay, especially evident on syntactic and grammatical comprehension tasks. He presents with
language regression and EEG paroxysmal abnormalities during NREM sleep, in the presence
of a documented neurological lesion and thus shows remarkable similarities with the cases
described by Fabbro et al. (cf. this volume). In Case 3 only very subtle language deficits could
be detected, mainly on semantic comprehension tasks (e.g., synonym comprehension; cf.
Fabbro et al, 2004). Therefore, the study of language development in the three cases of
cerebellar malformations that we have presented suggests that the cerebellum is involved in
language acquisition.
However, the effect of malformative lesions on language acquisition and language
development in adult age seems to vary according to lesion extent and localization. Near total
absence of the cerebellum does not seem to prevent a considerable degree of language
development, up to full communicative functionality, although the acquisition of
morphosyntax remains incomplete. The event of near total absence of the cerebellum is very
rare. Recently, Gardner et al. (2001) described three new cases and reviewed the relevant
literature which included only five other cases (Sener & Jinkins, 1993; Sener et al, 1995; Van
Hoof & Wilmink, 1996; Velioglu et al, 1998). The authors identified a syndrome termed
"Near-total absence of cerebellum with flat ventral pons and relatively mild clinical
affection". The same diagnosis has been made for our Case 1. However, Case 1 is remarkable
in that the MRI shows a very limited cerebeliar residue (possibly of the vermis) compared to
the cases described by Gardner et al. (2001). Nonetheless, like in Gardner's cases, cognition
and behavior are rather mildly affected in contrast with the extent of the anatomical defect. A
general feature of the "Near-total absence of cerebellum" syndrome seems to be that initially
patients show a very marked delay in development but slowly and steadily recover motor,
affective and cognitive functions. Recovery may occur very late in life, compared to normal
developmental milestones. In Case 1, plasticity factors may act on the remaining normal
cerebellar tissue and on a process of "cerebellization" of supratentorial regions (Macchi &

Bentivoglio, 1987). Malformations involving the vermis do not seem to prevent language
acquisition but profoundly disrupt communication (up to autistic-like behavior) and therefore
markedly delay language acquisition. Malformations limited to the right cerebellar hemisphere
may not impair language acquisition. This may be due to two main reasons: 1) language may
not be lateralized in the cerebellum, and thus it can be sustained by either remaining cerebellar
hemisphere, 2) language may be lateralized to the right cerebellar hemisphere, as proposed by
Marien et al. (1996), but plasticity factors may induce transfer of language functions to the
contralateral cerebellar hemisphere.
Language deficits associated to cerebellar malformations show important similarities and
differences with language deficits associated to cerebellar lesions. Both congenital and
acquired cerebellar lesions can show a great improvement of language impairment over time,
although cerebellar malformations involving the vermis show a specific language acquisition
delay and require constant speech therapy, while linguistic symptoms following acquired
cerebellar lesions usually regress spontaneously and rapidly to a considerable extent (but see
Fabbro et al., 2004; for a discussion on the persistency of residual deficits). Both congenital
and acquired cerebellar lesions may induce deficits in language comprehension and
expression. However, congenital cerebellar malformations involving the vermis seem to
induce more marked and permanent language disorders, especially in language
comprehension. This can be explained if one considers that the cerebellum seems to be
involved in procedural language acquisition, in particular in the acquisition of phonology and
syntax (see Ullman, 2001). A malformative lesion may impair procedural learning of language
and require long-term, effortful explicit training for language to be acquired, as in Case 1. On
the contrary, acquired cerebellar lesions seem to induce mild language expression deficits. In
conclusion, though our data do not allow us to take a definite stance on the type
(neuromodulation vs. processing) of cerebellar contribution to language, it seems likely that
normal language acquisition is crucially supported by full functionality of the cerebellum.
REFERENCES
Adams, W. and D. Sheslow (1990). Wide Arrangement of Memory and Learning (WRAML).
Wide Range, Wilmington (DEL).
Altman, N. R., T. P. Naidich and B. H. Braffman (1992). Posterior fossa malformations.
AJNR, 13, 691-724.
Bearzotti, F., S. Lilli and F. Fabbro. Standardizzazione di due test per la valutazione della
discriminazione uditiva. Psichiatria dell 'Infanzia e dell 'Adolescenza, in press.
Beery, K. E. and N. A. Buktenica (2000) Developmental Test of Visual Motor Integration Test
(VMI). Organizzazioni Speciali, Firenze.
Benton, A. L., K. deS. Hamsher, N. R Varney and O. Spreen (1983). Judgment of Line
Orientation. In: A. L. Benton, K. deS. Hamsher, N. R Varney and O. Spreen,

Contributions to Neuropsychological Assessment, pp. 44-54. Oxford University Press,
New York.
Chilosi, A. M., P. Cipriani (1995). TCGB. Test di comprensione grammaticale per i bambini.
Edizioni del Cerro, Pisa.
Cornoldi, C and G. Colpo (1981). Prove di lettura MT per la scuola elementare.
Corsi, P. M. (1972). Human memory and the medial temporal region of the brain.
Unpublished Ph.D. Thesis, McGill University, Montreal.
Courchesne, E. (1997). Brainstem, cerebellar and limbic neuroanatomical abnormalities in
autism. Curr Opin Neurobiol, 7, 269-278.
De Agostini, M., M. N. Metz-Lutz, A. Van Hout, M. Chavance, G. Deloche, I. Pavao-Martins
and G. Dellatolas (1998). Batterie devaluation du language oral de l'enfant aphasique:
Standardisation francaise (4-12 ans). Revue de Neuropsychologie, 8, 319-367.
Desmond, J. E. and J. A. Fiez (1998). Neuroimagins studies of the cerebellum: Language,
learning and memory. Trends in Cognitive Sciences, 2, 355-361.
Dow, R. S. and G. Moruzzi (1958). The Physiology and Pathology of the Cerebellum. The
University of Minnesota Press, Minneapolis.
83-94.
Fabbro, F., R. Moretti and A. Bava (2000a). Language impairments in patients with cerebellar
Fabbro, F., M. Skrap and S. Aglioti (2000b). Pathological switching between languages after
frontal lesions in a bilingual patient. J. Neurol Neurosurg Psychiatry, 68, 650-652.
Fabbro, F., A. Tavano, S. Corti, N. Bresolin, P. De Fabritiis and R. Borgatti (2004). Long-
term neuropsychological deficits after cerebellar infarctions in two young adult twins.
Ferrari, E., E. De Renzi, P. Faglioni and E. Barbieri (1981). Standardizzazione di una batteria
per la valutazione dei disturbi del linguaggio in eta scolare. Neuropsichiatria Infantile,
235, 148-158.
Fiez, J. A. (1996). Cerebellar contribution to cognition. Neuron, 16, 13-15.
Gardner, M. F. (1996). Test of Visual Perceptive Skills (non motor). Psychological and
Educational Publication, Hydesville (CA).
Gardner, R. J. M, L. T. Coleman, L. A. Mitchell, L. J. Smith, A. S. Harvey, I. E. scheffer, E.
story, M. J. Nowotny, R. A. Sloane and L. Lubitz (2001). Near-total absence of the
cerebellum. Neuropediatrics, 32, 62-68.
Gilberg, C. and M. Coleman (1992). The biology of autistic syndromes. MacKeith Press,
London.
Griffiths, R. (1986). The Abilities of Babies. A Study in Mental Measurement. The Test
Agency Ltd, Henley-on-Thames (Oxon).
Heaton, R., G. Chelune, J. Talley, G. Kay and G. Curtiss (2000). Wisconsin Card Sorting Test.
Leiner, H. C, A. L. Leiner and R. S. Dow (1986). Does the cerebellum contribute to mental
skills? Behav Neurosci, 100, 443-454.
Leiner, H. C, A. L. Leiner, R. S: Dow (1991). The Human cerebro-cerebellar system: its
computing, cognitive and language skills. Behavioural Brain Research, 44, 113-128.
Leiner, H. C, A . L. Leiner and R. S. Dow (1993). Cognitive and language functions of the
human cerebellum. Trends Neurosci, 16, 444-447.
Levisohn, L., A. Cronin-Golomb and J. D. Schmahmann (2000). Neuropsychological
consequences of cerebellar tumor resection in children. Cerebellar cognitive affective
syndrome in apaediatric population. Brain, 123, 1041-1050.
Marien, P., S. Engelborghs, F. Fabbro and P. P. De Deyn (2001). The lateralized linguistic
cerebellum: A review and a new hypothesis. Brain and Language, 79, 580-600.
Marien, P., S. Engelborghs, B. A. Pickut and P. P. De Deyn (2000). Aphasia following
cerebellar damage: fact or fallacy? Journal of Neurollnguistics, 13, 145-171.
Marien P., J. Saerens, R. Nahoe, E. Moens, G. Nagels, B. A. Pickut, R. A. Dierckx and P. P.
De Deyn (1996). Cerebellar induced aphasia: case report of cerebellar induced
prefrontal aphasia language phenomena supported by SPECT findings. J Neurol Sci,
144, 34-43.
Molinari, M., M. G. Leggio and M. C. Silveri (1997a). Verbal fluency and agrammatism.
International Review ofNeurobiology, 41, 325-339.
Molinari, M., M. G. Leggio, A. Solida, R. Ciorra, M. Misciagna, M. C. silveri and L Petrosini
(1997b). Cerebellum and procedural learning: Evidence from focal cerebellar lesions.
Brain, 120, 1753-1762.
Paradis, M. (1987). The assessment of bilingual aphasia. Lawrence Erlbaum, Hillsdale.
Paradis, M. (1999). Valutazione dell'afasia bilingue. EMS, Bologna.
Petersen, S. E., P. T. Fox, M. I: Posner, M. Mintun and M. E. Raichle (1989). Positron
emission tomographic studies of the processing of single words. Journal of Cognitive
Neuroscience, 1, 153-170.
Petersen, S. E., and J. A. Fiez (1993). The processing of single words studied with positron
emission tomography. Annu. Rev. Neurosci., 16, 509.
Rey, A. (1967). REY. Reattivo della Figura Complessa. Organizzazioni speciali: Firenze
Riva, D. and C. Giorgi (2000). The cerebellum contributes to higher functions during
development. Evidence form a series of children surgically treated for posterior fossa
tumours. Brain, 123, 1051-1061.
Schmahmann, J. D. (1991). An emerging concepi: the cerebellar contribution to higher
function. Arch Neurol, 48, 1178-1187.
Schmahmann, J. D., J. Doyon, D. McDonald, C. Holmes, K. Lavoie, A. S. Hurwitz, et
al.(1999). Three-dimensional MRI atlas of the human cerebellum in proportional
stereotaxic space. Neuroimage, 10, 233-260.
Schmahmann J. D. & J. C. Sherman (1998). The cerebellar cognitive affective syndrome.

Brain, 121, 561-579.
Sener, R. N. and J. R. Jinkins (1993). Subtotal agenesis of the cerebellum in an adult. MRI
demonstration. Neuroradiology, 35, 286-287.
Sener, R. N. (1995). Cerebellar agenesis versus vanishing cerebellum in Chiari II
malformation. Comput Med Imaging Graph 19, 491-494.
Shallice, T. (1982). Specific impairments of planning. Philosophical Transactions of the
Royal Society of London, 298, 199-209.
Silveri, M. C, A. M. Di Betta, V. filippini, M. G. Leggio and M. Molinari (1998). Verbal
short-term store-reharsal system and the cerebellum. Evidence from a patients with a
right cerebellar lesion. Brain, 121, 2175-2187.
Silveri, M. C, M. G. Leggio and M. Molinari (1994). The cerebellum contributes to
linguistics production: a case of agrammatic speech following a right cerebellar lesion.
Neurology, 44, 2047-2050.
Spinnler, H. and G. Tognoni (1987). Standardizzazione e Taratura Italiana di test
Neuropsicologici. The Italian Journal of Neurological Sciences, 6, supplement 8.
Stewart, R.M. (1956). Cerebellar agenesis. J. Men Sci, 102, 67-77.
Terman, L. M. and M. A. Merrill (1968). Scala di Intelligenza Stanford - Binet, forma L-M.
Trouillas P., T. Takayanagi, M. Hallett, R. D. Currier, S. H. Subramony, K. Wessel, et al.
(1997) International Cooperative Ataxia Rating Scale for pharmacological assessment
of the cerebellar syndrome. The Ataxia Neuropharmacology Committee of the World
Federation of Neurology. JNeurol Sci, 145(2), 205-11.
Ullman, M. T. (2001). The neural basis of lexicon and grammar in first and second language:
the declarative/procedural model. Bilingualism: Language and Cognition, 4, 105-122.
Van Hoof, S. C. and J. T. Wilmink (1996). Cerebellar agenesis. J. Beige Radiol, 79, 282.
Velioglu, S. K., K. Kuzelyli and M Zzmenoglu (1998). Cerebellar agenesis: a case report with
clinical MR imaging finding and a review of the literature. Eur. J. Neurol, 5, 503-506.
Vender, C, D. A. Anghini, S. Cumer Bruno, R. Borgia and G. Zardini (1979). Contributo allo
studio del Token Test in eta evolutiva. Neuropsichiatria Infantile, 215, 436-439.
Vender, C, R. Borgia, S. Cumer Bruno, P. Freo and G. Zardini (1981). Un test di ripetizione
di frasi. Analisi delle perfomances di bambini normali. Neuropsichiatria infantile, 243,
819-831.
Zettin, M., S. F. Cappa, A. D'Amico, R. Rago, C. Perino, D. Perani and F. Fazio (1997).
Agrammatic speech production after a right cerebellar haemorrage. Neurocase, 3, 375-
380.
Weschler, D. (1974). WAIS. Scala di Intelligenza Weschler per Adulti. Organizzazioni
Speciali, Firenze
Weschler, D. (1973). WPPSI. Scala Weschler a livello prescolare e di scuola elementare.
Weschler, D. (1986). WISC-R. Scala di Intelligenza Weschler per Bambini - Riveduta.

Weschler, D. (1997). WAIS-R. Scala di Intelligenza Weschler per Adulti - Riveduta.
Organizzazioni Speciali, Firenze
Crossed Aphasia in Children 147
CROSSED APHASIA IN CHILDREN
Peter Marien
General Hospital Middelheim, University of Antwerp and Free University of Brussels,
Belgium
Philippe Paquier
University Hospital Erasme, University of Antwerp and Free University of Brussels, Belgium
Sebastiaan Engelborghs and Peter P. De Deyn

General Hospital Middelheim, University of Antwerp, Belgium
Abstract - In 1899, Byrom Bramwell (1899) introduced the concept of crossed aphasia (CA) as an
exception to the prevailing insight of an inherent association between cerebral dominance for
language and hand preference (so-called Broca's dogma). He defined as such the phenomenon of
aphasia caused by brain damage ipsilateral to the dominant hand (i.e. aphasia resulting from a lesion
to the left hemisphere in sinistrals and aphasia following a lesion to the right hemisphere in dextrals).
Rather striking in the development of this concept is the absence of any reference to the understanding
of acquired childhood aphasia (ACA) in which aphasia after a right hemisphere lesion was considered
a frequent phenomenon. However, following erosion of Bramwell's positions and a decay of the
feasibility of so-called Broca's dogma, a confluence of concepts was established between CA, aphasia
in sinistrals and ACA seven decades later. The first part of this chapter reviews the history of the
genesis, development and collusion of concepts in these atypical populations. The second part
addresses CA in children in more detail. In contrast to the estimated low incidence in dextral adults
(among 1%), CA in children is generally considered a common finding. Reviewing the literature from
1975 onwards, we only found five children (2.7%) in a corpus of 180 dextrals with aphasia following
a right hemisphere lesion (Marien et al, 2001c). A critical analysis rendered three of the reported
cases ambiguous for a CAD diagnosis and hence not suitable to draw conclusions. The
neurobehavioural characteristics of the two representative childhood CAD cases are discussed and
compared with adult CAD and ACA.
Keywords: Crossed aphasia, acquired childhood aphasia, language dominance

PART I: DEVELOPMENT AND CONFLUENCE OF CONCEPTS - FORMATION OF

CONCEPTS
The standard view on acquired childhood aphasia (ACA)
In 1868, Cotard for the first time systematically studied the phenomenon of ACA. In his
elaborated monograph entitled 'Etude sur l'atrophie cerebrale' he reported the post-mortem
findings of seven left hemisphere lesioned adults with right hemiplegia and normal
neurocognitive development since early childhood. Cotard (1868) stated that language in these
patients had developed in the right hemisphere, and that, in general, functions of the lesioned
hemisphere are taken over by the intact one, if damage is sustained at a young age. As a
consequence, he postulated that aphasic disorders do not occur in children. Clarus (1874),
however, disputed this position and concluded from an analysis of 50 published cases that: 1)
aphasia in children is not rare, 2) prognosis is not invariably benign but depends in part on
both etiology and extent of the lesion, and 3) the right hemisphere can take over language
functions when the left hemisphere is damaged.
Given the relatively large proportion of left hemiplegic aphasic children Sach and Peterson
(1890) advanced the view that during the first years of life both hemispheres are equally
equipped for the development of language and that during the expansion of language
functions, the role of the right hemisphere progressively decreases in favour of the left one.
This view was later to be designated by Basser (1962) and Lenneberg (1967) as the hypothesis
of 'hemispheric equipotentiality and progressive lateralization of language development'.
Freud (1897) seconded this position with the observation that aphasia in children, entirely
unlike aphasia in adults, occurs relatively frequently after right hemisphere lesions.
In spite of an ongoing controversy surrounding the theme, Bernhardt (1885) defined
acquired childhood aphasia (ACA) on five characteristics: 1) ACA is not rare, 2) ACA is
predominantly of the nonfluent (expressive, motor) type, 3) ACA entails a benign prognosis,
4) ACA lasts only a short time, and 5) ACA is entirely compensated for by the expansion of
language functions in the right hemisphere. Towards the end of the 19th century, this
definition became generally accepted, giving rise to a 'standard doctrine' which would last
more than 70 years.
Crossed aphasia: a new concept in adult aphasia
During the last decades of the 19th century, so-called Broca's doctrine assigned left
hemisphere dominance for language to dextrals and right hemisphere dominance for language
to sinistrals. This 'dogma' established a strict anatomo-functional connection -much stricter
indeed than Broca himself had implied in his seminal 1865 paper- between motor control of
the dominant hand and control over language functions. However, in the same period, a
number of observations casted doubt on such a clear-cut view. The counter-evidence brought
forth by the early, often anecdotical case-reports was not very strong. Hughlings Jackson
(1880), for instance, described aphasia with left hemiplegia in a sailor who, though left-
handed, was obliged to consistently use the right hand for mending the sails (a large share of
his work), and hence was, at least in part, a dextral. A few years later, Oppenheim (1889)
observed aphasia associated with anatomopathologically verified right hemisphere lesions (a
solitary tuberculoma and a subcortical sarcoma) in two right-handed patients, one of whom,
however, had become left-handed following injury to the right hand at the age of 17 years.
Before the turn of the century, more convincing evidence against the model of cerebral
dominance for language and handedness was provided by Bramwell (1899). He described a
36-year-old left-handed man who developed persistent nonfluent aphasia and a right
sensorimotor deficit following a left hemisphere stroke. Bramwell introduced the term crossed
aphasia (CA) to denote, in a broad sense, any aphasic syndrome resulting from a cerebral
lesion 'ipsilateral' to the dominant hand. He particularly emphasized the exceptional character
of this observation because of the persistent aphasia and noted that in most CA cases, the
language symptoms are temporary.
In agreement with his predecessors, Broca and Hughlings Jackson, Bramwell attributed
what he considered the inherently transient character of CA to the compensatory role of the
non-dominant hemisphere. In contrast to temporary manifestations of CA, which he assumed
to occur in both left- and right-handers, Bramwell postulated that CA characterized by
persistent language disturbances is extremely rare and limited to the sinistral population.
Although several reports of dextral aphasics with anatomopathologically verified lesions of
the right hemisphere had already been described (e.g. Farge, 1877; Oppenheim, 1890;
Preobrashenski, 1893; Moltschanow, 1897), Bramwell stated explicitly that he knew of no
pure cases of CA in dextrals (CAD).
Indeed, Bramwell believed that crossed aphasia was always a transient condition in
dextrals who: (1) had not suffered brain damage during early development, (2) were born to
dextral parents and (3) had learned to write with the right hand. To explain the exceptional
observations of 'pure' right-handers in whom acute damage to the anatomical language region
of the left hemisphere induced no aphasia (negative cases), he proposed that language
dominance of the lesioned left hemisphere was immediately taken over by the language
centres of the right hemisphere.
Notably absent in the formation of the concept of CA is any reference to the understanding
of ACA, which, at the turn of the 19th century, had arrived at a consensus within a standard
doctrine. This is striking given the fact that ACA was also considered: 1) to result frequently
from cerebral lesions ipsilateral to the dominant hand, and 2) to occur frequently as a transient
phenomenon due to the compensatory role of the non-lesioned hemisphere. The fact that,
within ACA, the right hemisphere was deemed central for both the genesis of language as well
as the recovery from aphasia makes its absence from the development of the CA concept all
the more conspicuous.
EVOLUTION OF CONCEPTS IN THE 2OTH CENTURY
Acquired Childhood Aphasia (ACA)1: 1900-1978
Neglected early evidence against semiological uniformity. During the first decades of the 20th
century, a few reports (e.g. Potzl, 1926; Wagner & Mayer, 1933; Brunner & Stengel, 1932; De
Girardier & Jeannin, 1939) documented a range of fluent (sensory) language deficits in
children, questioning the classical views on the predominantly motor character of ACA
(Bernhardt, 1885). Although these few scattered counter-examples constituted a striking
deviation from the homogeneous semiological picture portrayed by the standard doctrine, they
did not succeed in modifying the standard doctrine.
The aphasiogenic and compensatory role of the right hemisphere. Several authors, including
Taylor (1905) and Sachs & Hausmann (1926) subscribed to the notion that ACA frequently
occurs in association with lesions of the right hemisphere. By contrast, Smithies' review
(1907) of the literature did not confirm the purported role of the right hemisphere in ACA but
he unfortunately did not pursue this finding.
Marie (1922) again emphasized the absence of aphasia in children with right hemiplegia at
a young age and, in accord with the findings of Cotard (1868) and Jendrassik & Marie (1885),
focused attention on the transient character of ACA. Convinced that the anatomical centres are
bilaterally and symmetrically represented at birth, Marie (1922) posited that the language
centres are not innate, but that they develop self-sufficiently in each individual. Much like
Broca (1865) before him, Marie (1922) attributed left hemisphere superiority in this process to
the anatomical fact that the left hemisphere maturates faster than the right one. In a similar
sense, Mingazzini (1925) supported the views of Sachs & Peterson (1890) claiming that,
during development the role of the right hemisphere in language progressively decreases.
Recovery reconsidered. The prevalent optimism regarding favourable recovery from ACA
was not shared by Potzl (1926), Ford & Schaffer (1927), Minkowski (1930), and Brunner &
Stengel (1932), among others. They stated that prognosis for recovery is markedly worse for
sensory than for motor aphasia. Minkowski (1930), moreover, pointed out that residual
language symptoms undermine neurocognitive and social development.
Guttmann (1942) for the first time highlighted ACA as a multifaceted disorder. He
analyzed research data of 30 patients between two and 14 years of age with primarily
unilateral brain damage caused by different etiologies (traumas, tumors, abscesses, and
thromboses). Sixteen of the patients exhibited aphasia. With respect to incidence, Guttmann
(1942) found that aphasia following left hemisphere lesions is not at all infrequent in children
(14/16 cases); in fact, prevalence was comparable to that of acquired aphasia in adults.
Regarding semiology, he noted that irrespectively of the localization of the lesion, all children
under 10 years of age exhibited reduced spontaneous speech, ranging from mutism to
dysarthria and telegraphic style during recovery. In addition to expressive (nonfluent)
language symptoms, children with posterior lesions also presented sensory (fluent) deficits,
which, as in adults, could take the form of logorrhea. Although the data did not permit
rigorous deductions due to the variability of etiology and follow-up, Guttmann (1942)
concluded that: 1) a combination of expressive (nonfluent/motor) and receptive
(fluent/sensory) aphasia entails a less favourable prognosis than purely expressive aphasia, 2)
recovery from expressive aphasia occurs notably faster and 3) prognosis for recovery is
reserved if the aphasic symptoms are still present after four weeks. Following Guttmann
(1942), the view of AC A as a transient infliction slowly changed. The exclusively motor
character and the unequivocally successful recovery, however, remained indubitable even
during the 1950s (Branco-Lefevre, 1950; Critchley, 1950; De Ajuriaguerra, 1958; Subirana,
1960) and the influence of these notions was felt well into the 1980s (Denckla, 1979).
hi a study of 102 cases with acquired infantile hemiplegia, Basser (1962) discerned 30
patients whose cerebral lesion (15 in the left and 15 in the right hemisphere) occurred during
language development. A lesion in the left hemisphere gave rise to aphasia in 13 of the
children, while in seven of the children aphasia was due to a lesion in the right hemisphere.
According to Basser (1962), duration of aphasia did not depend on the severity of the
hemiplegia, but rather on the age of onset. A better outcome was attributed when the lesion
was sustained before the age of two years.
In 1965, Alajouanine & Lhermitte studied 32 children between 6 and 15 years of age with
an aphasiogenic lesion of traumatic, vascular or tumoral origin in the left hemisphere. They
found semiological differences between the children under and over 10 years of age. In the
group of children younger than 10 years the aphasia was predominantly characterized by: 1)
severely reduced verbal expression, 2) distorted articulation and phonetic disintegration, 3)
impaired auditory-verbal comprehension, 4) severe spelling impairments and 5) lack of
paraphasias. Aphasia in children older than 10 years consisted mostly of: 1) a markedly lower
frequency of articulation distortions, 2) an increased incidence of paraphasias, which even
took the form of jargon speech in some children, 3) impaired reading comprehension and 4)
disturbed spelling. With respect to recovery, Alajouanine & Lhermitte (1965) observed
complete recovery of language functions in 75% of the population sampled, although none of
the children made normal academic progress consequent to learning difficulties.
Lenneberg (1967), in turn, reformulated classical insights into AC A and posited that,
when the lesion is limited to one hemisphere and sustained before the age of 9 years, complete
remission of aphasia will ensue. One year later, Collignon et al. (1968) confirmed a negative
1
The Landau-Kieffher syndrome (Landau and Kleffner, 1957) is not reviewed here because of its
special position within acquired childhood aphasia. For critical reviews, we refer to the contributions
by Gordon (1990), Deonna (1991), Paquier, Van Dongen & Loonen (1992) and Marien et al. (1993).
prognosis for recovery after bilateral lesions, hi agreement with the findings of Alajouanine &
Lhermitte (1965) and Collignon et al. (1968), the contributions by Riese & Collison (1964) as
well as Aicardi et al. (1969) further weakened the case for widespread complete recovery
from ACA. In contrast to Lenneberg's conclusions (1967), which indicated that ACA after the
age of 10 years leaves undisputable traces, Hecaen (1976) found excellent recovery in three
patients who had acquired aphasia at the age of 14 years. In accord with Collignon et al.
(1968), Hecaen (1976) indicated that extent and bilateral cerebral damage are the most
important variables in the recovery process.
Van Dongen & Loonen (1977) investigated recovery of ACA with a follow-up of three
years in a group of 15 right-handed children between 4 and 14 years of age. Eight children,
seven of which with a traumatic etiology, recovered while the four children with vascular
aphasia barely recovered at all. In addition to the impact of etiology, the authors also found a
correlation between recovery and type of aphasia, and identified severity of impaired
comprehension as a crucial factor.
Brown & Hecaen (1976) summarized the predominant insights into ACA in the mid-
1970s as follows: 1) the initial phase of ACA is characterized by mutism or agrammatism and
is followed by anomia and nonfluent phonemic paraphasias, 2) a distorted articulation appears
in a context of a nonfluent or borderline fluent output, 3) fluent phonemic paraphasias
(conduction aphasia) are uncommon, 4) logorrhea and semantic or neologistic jargon do not
arise, 5) impaired comprehension is found in one-third of patients, 6) recovery is manifestly
superior compared to acquired aphasia in adults, and 7) at a young age, there is an
approximately equal chance to develop aphasia following damage to either hemisphere.
Crossed aphasia in sinistrals and dextrals: 1900-1970
In the course of seven decades, Bramwell's positions (1899) on CA were shown to be

untenable and the initial concept of CA completely eroded. Firstly, it became clear that
permanent aphasia following a hemispheric lesion ipsilateral to the dominant hand is not
limited to sinistrals, but is also occasionally observed in dextrals. In further contrast with
Bramwell's views (1899), consensus arose regarding the fact that CA in sinistrals is not the
exception, but the rule.
The conceptual turning point in classical thinking about CA was due to the insights
derived from systematic investigation of sinistral aphasics during the 1950s and 1960s rather
than to the increasing number of observations of crossed aphasia in dextrals (CAD), which
were often reduced to artifacts and impelled to accommodate the prevalent view of an intrinsic
connection between cerebral dominance for language and hand preference (so-called Broca's
doctrine). Explanations such as hidden sinistrality (e.g. Joffroy, 1903), absence of a
decussation of the pyramidal tract (e.g. Souques, 1910), familial left-handedness inducing a
genetic predisposition for right hemisphere dominance for language (e.g. Kennedy, 1916),
bilateral language representation linked to gradations in manual preference (e.g. Rothschild,
1931) or consequent to developmental complications (e.g. Stone, 1934), undetected damage of

the left hemisphere (e.g. Scollo, 1926) and the presumed linguistic role of phylogenetically
older subcortical structures in which language functions were assumed not to be lateralized
(e.g. Ardin-Delteil et al, 1923) did not put so-called Broca's doctrine fundamentally into
question.
The relation between hand preference and hemispheric dominance for language was
investigated by Conrad (1949) in a study of sinistral war victims inflicted with traumatic brain
injury. He concluded that: 1) aphasia occurs more frequently in sinistrals than in dextrals, 2)
the chance of recovery is greater in sinistrals than in dextrals, and 3) in contrast to the
accepted view, aphasia in the majority of cases is not caused by lesions of the right
hemisphere, but by lesions of the left hemisphere. Despite considerable differences in the
proportion and incidence of cases, these findings were corroborated by numerous studies
(Goodglas & Quadfasel, 1954; Bingley, 1958; Penfield & Roberts, 1959; Hecaen & De
Ajuriaguerra, 1963, Hecaen & Sauguet, 1971; Newcombe & Ratcliff, 1971; Satz, 1980;
Annett, 1985). Although only a very limited number of these studies systematically paid
attention to the semiological characteristics, they also gave rise to the long-standing view that,
irrespective of the lateralization of the lesion, aphasia in sinistrals significantly differs from
uncrossed aphasia in dextrals. In comparison to uncrossed aphasia in dextrals a more uniform
semiological syndrome was advanced for aphasia in sinistrals, typically characterized by: 1)
markedly less severe language disturbances, 2) a significantly lower incidence of
comprehension deficits, 3) a higher incidence of expressive language disturbances and 4) a
much better, often complete recovery (Luria, 1947; Goodglass & Quadfasel, 1954; Subirana,
1958, 1969; Hecaen & De Ajuriaguerra, 1963).
The implication that CA in sinistrals is the rule rather than the exception led to the disuse
of the term in the sinistral population and became synonym of crossed aphasia in dextrals
(CAD). In addition, the change of insights in the anatomoclinical configurations of aphasia in
sinistrals and the decay of the feasibility of the doctrine established the basis for the
development of other types of explanations for CAD in the 1970s. In accord with the view
that left hemisphere language dominance reflects the end-point of a successful maturation
process which ensues progressively from an initially bilateral representation of language
functions, the hypothesis was made that bilateral language representations in CAD reflect a
disrupted maturation. This view resulted from a confluence of concepts stemming from
acquired aphasia in three atypical populations: children, sinistrals and anomalous dextrals.
A CONFLUENCE OF CONCEPTS
In their contribution on the relation between cerebral dominance for language and hand
preference, Anastasopoulos & Kokkoni (1962) incorporated the ideas from AC A and made
them central to their discourse regarding the gradual differences of cerebral lateralization of
language and hand preference. In agreement with the hypotheses of hemispheric

equipotentiality and progressive maturation of language dominance, they posited that
complete unilateral representation of language and hand preference are the end product of
optimal development, which ensues progressively from an initially bilateral cerebral potential
for giving rise to manifold brain functions.
At the basis of the lateralization process they posited a dynamic mechanism of inhibiting
influences exerted on the contralateral homologous brain area by the hemisphere that is
becoming dominant. Based on this assumption, they subsequently claimed that language
capacity and hand preference will be represented within one and the same hemisphere only if
the process of cerebral lateralization develops fully into unilaterality. According to the
authors, variants of a separate development of both functions would arise as soon as complete
unilateral cerebral dominance becomes unattainable.
Anastasopoulos & Kokkoni (1962) explicitly related the conclusions from comparative
studies of left- and right-handed aphasics carried out in the 1950s to language representation
in children:
'Various authors (...) who studied aphasia in left-handers state that these
individuals, in contrast to right-handers, are more frequently made aphasic
irrespective of the side of the lesion, and that the occurring aphasia is on the
whole less severe and less persistent. These characteristics are attributed to a
significant participation of both hemispheres in language in left-handers, the
same as in children ' (p. 14)
They also attributed the parallels between the pattern of recovery from ACA and aphasia
in sinistrals to the adoption of the language functions by the non-lesioned, active language
centres of the contralateral hemisphere. Despite gradual differences in the extent of cerebral
dominance, Anastasopoulos & Kokkoni (1962) deemed unlikely the possibility of a complete
separation in the representation of language and hand preference.
Influenced by this view that variants of a dissociated lateralization of hand preference and
language reflect developmental irregularities, Brown & Hecaen (1976) and Brown & Wilson
(1973) achieved a quasi complete symbiosis between ACA, aphasia in sinistrals, and CAD.
They acknowledged two phases within ACA, namely 1) mutism, or agrammatism at an early
infantile age and 2) anomia with nonfluent phonemic paraphasias at a later infantile age, and
found these phases prominently represented as distinct phenomena in CAD and aphasia in
sinistrals, respectively. The temporal substrate of these mutual commonalities led them to
conclude that these three special forms of aphasia are situated in a continuum of gradual
development from bilateral to unilateral left hemisphere dominance for language. As such,
they proclaimed ACA, aphasia in left-handers, and CAD to be representative of an immature
lateralization process.
Concurring with the model of a disrupted maturation process from initial bilateral to
unilateral left hemisphere language dominance, they posited that CAD reflects various forms
of bilateral language representation.
Following these contributions, the semiological uniform type of CAD was systematically
reproduced in many additional observations. In the 1970s, the uniform pattern of these
observations rapidly led to a so-called standard doctrine for CAD. In accord with the classical
conception of ACA, nonfluent characteristics were ascribed to prototypical CAD irrespective
of the location of the lesion. Agrammatism, mutism at onset, relatively preserved word
retrieval, and variability in repetition and comprehension were propagated in numerous
publications, while the possibility of jargon was formally excluded. The swiftness and extent
of recovery were found to be superior compared to the classical types of uncrossed aphasia.
As in ACA, a comparable dissociation in CAD was frequently observed between superior oral
and inferior written language (Table 1).
Table 1 — Overview of characteristics of acquired aphasia in children, aphasia in smistrals and CAD
during the seventies
Acquired Aphasia in CAD

Childhood aphasia sinistrals
Nonfluent symptoms
Mutism + rare +
Agrammatism + - +
Distorted articulation + + +
Phonemic paraphasias + + +
Disturbed verbal fluency + + +
Fluent symptoms
Verbal paraphasias rare rare rare
Fluent phonemic paraphasias rare rare rare
Logorrhea - rare -
Semantic j agon - - -
Neologistic jargon - - -
Disturbed comprehension rare rare rare
Anomia + + variable
Disturbed repetition + + rare
Oral better than written + ? +
Recovery good good if expressive good
Lesion site bilateral/LH/RH bilateral/LH/RH RH
Legend: + = present; - = absent; LH = left hemisphere; RH = right hemisphere

EROSION OF STANDARD DOCTRINES
Despite several contestations of the classical ideas expressed in the standard doctrine on AC A,
a new era on the conceptual development only followed Woods & Teuber's (1978) landmark
review paper entitled 'On changing patterns of childhood aphasia'. In radical contrast to the
standard opinion, they demonstrated that: 1) crossed aphasia does not occur more frequently
in children than in adults, 2) the extent of recovery does not clearly correlate with age at onset
and 3) exceptions exist regarding the predominance of nonfluent aphasia characteristics. A
sizable number of reports following this study systematically demonstrated the heterogeneous
character of ACA (for a review, see Paquier & Van Dongen, 1996; 1998). Moreover, the
remarkable conformity with acquired aphasia in adults provided a direct impetus for a radical
break with the leading views on the neurobiological mechanisms of language acquisition such
as the hemispheric equipotentiality and progressive lateralization hypotheses, which had held
sway for more than a century. The untenability of the postulate that during the initial process
of normal language acquisition both hemispheres are equally actively involved was
furthermore supported by an increasing number of neuroanatomical, neurophysiological and
neuroradiological studies which contend an innate predisposition of the left hemisphere for
language acquisition. The view of aphasiologic uniformity and rapid recovery was also
contended in CAD under the cumulative evidence of reports illustrating a broad spectrum of
aphasic manifestations. In the first place, numerous CAD patients with receptive (fluent)
aphasia were described and neuroradiologically documented. These reports established that, as
a consequence of the methodological limitations inherent to the predominantly clinical
approach in the pre-neuroradiological period, fluent types of CAD have been heavily
underrepresented throughout the conceptual evolution of CAD. Carr et al. (1981) also
expanded the evidence against a homogeneous nonfluent CAD syndrome reporting similar
distribution figures of respectively 72% and 27% for the nonfluent-fluent dichotomy in both
standard forms of aphasia and CAD. Secondly, several CAD patients were described with
language profiles that did not at all or not significantly improve during follow-up (see
Coppens & Hungerford, 1998; Marien et al, 2004). Thirdly, in opposition to the frequently
reported dissociation between superior oral and inferior written language, reversed patterns in
which oral language is more prominently affected than written language have been reported as
well (see Marien et al, 2004).
PART 2: CROSSED APHASIA IN CHILDREN - SELECTION OF CASES
Irrespectively of etiology we selected from the literature all right-handed patients with aphasia
following a right hemisphere lesion. 1975 was chosen as a starting-point for this selection
since Faglia, Rottoli and Vignolo (1990) showed that all fully acceptable CAD cases were
published after this date. From this corpus of 170 cases, enriched with 10 personal
observations (Marien et al, 2001b; Paghera et al, 2003), all childhood cases were selected.
This resulted in a total number of five cases (2.7%). The first case was described by Assal &
Deonna (1977) with a close follow-up of almost four years. Twelve years later, Assal (1987)
reinvestigated this patient. The other patients were reported by Martins et al. (1987), Burd et
al, (1990), Martins et al. (1995) and Marien et al. (2001a). Schematic case presentations such
as the one in the sample of Hecaen (1976) and Woods & Teuber (1978) (case 18) were
considered to be too concise to allow any further analysis and were not withheld in the present
review. Demographic, clinical and neuroradiological data of these cases are summarized in
Appendices 1 and 2. To allow for a comparison between cases -and hence theoretical
deductions- all case descriptions were critically evaluated in terms of their reliability for a
CAD diagnosis according to the revised criteria developed by Marien et al. (2004). A
diagnosis of possible CAD was made whenever the three following criteria were met: 1) clear-
cut evidence of aphasia, 2) evidence of natural (i.e. not shifted) right-handedness as
documented by a formal test, and 3) evidence of lesions strictly confined to the right
hemisphere, leaving the left hemisphere structurally intact. Patients not complying with these
three criteria were considered unreliable CAD. Two additional criteria were required to justify
the diagnosis of reliable CAD: 4) absence of familial left-handedness or ambidexterity, and 5)
no history of early brain damage and/or seizures in childhood. In the reliable CAD cases,
anatomo-clinical correlations were judged to be reliable only if language was formally
assessed during the lesion phase. The following were not considered to be exclusion criteria
for CAD: 1) illiteracy and schooling, 2) bi- or multi-lingualism, 3) use of a tonal language,
and 4) use of an ideographic script. As shown in the algorithm for childhood CAD (Figure 1),
three cases did not meet the mandatory criteria: Assal & Deonna (1977), Assal (1987),
Martins et al. (1987) and Martins et al. (1995). Assal & Deonna (1977) and Assal (1987)
reported a five-year-old boy who sustained an extensive cortico-subcortical fronto-parieto-
temporal infarction secondary to a complete obstruction of the right internal carotid artery.
After a ten-day period of mutism and severe auditory-verbal comprehension deficits, the
aphasic syndrome evolved within three years to a nonfluent, adynamic and agrammatic output
disorder with marked sequelae in the subsequent acquisition of reading and writing. A
residual adynamic output syndrome associated with semantic naming defects and insufficient
reading and writing capacities persisted 12 years after the onset of neurological symptoms. A
two-year delay in schooling reflected learning disabilities which particularly consisted of
memory disturbances and calculation defects. Twelve years post-onset, the neurological
condition was characterized by a paralysis of the left arm. Unfortunately, neither the initial
description, nor the follow-up report after 12 years mention any formal measures to assess the
patient's hand preference. Despite thorough descriptions on the linguistic, neurocognitive and
neurological level, absence of a standardized test to assess handedness casts doubt on the
diagnosis of acquired CAD in this case. Both cases reported by Martins et al. (1987; 1995)
were also classified as unreliable cases (cf. Figure 1). Martins et al. (1987) reported a 15 year-
old-boy with an initially fluent aphasic syndrome that aggravated and evolved to a nonfluent
aphasia in association with the fatal course of a right posterior oligodendroglioma. Irrespective
of surgical treatment and irradiation, the tumor progressed, invaded more anterior brain
regions and caused the patient's death four months after onset of the neurological symptoms.
Along the progression of actual right hemisphere destruction, a considerable mass effect of the
infiltrative right hemisphere lesion on the left hemisphere was demonstrated on computerized
tomography (CT) images. Though in this patient a close correlation between the aphasic
symptoms and right hemisphere damage might be considered likely, distant intra- and
interhemispheric mass effects as well as functional compensation mechanisms inherent to this
type of pathology blurr the circumscription of the exact impact of the lesion on the clinical
syndrome. Because of a possible impact on the left hemisphere, a CAD diagnosis is generally
rejected when, as in this case, the underlying cause is an infiltrative or space-occupying brain
lesion. In the second case reported by Martins et al. (1995), a penetrating gunshot trauma
caused destruction of right frontal brain tissue in a 13-year-old right-handed boy. Only five
days after this event the boy developed repeated complex partial and focal motor seizures as
the first neurological symptoms. Analogous to Todd's paresis, a motor aphasia developed
along a left-sided weakness following the epileptic fits. On admission, a CT scan of the brain
disclosed a right frontal abscess and a midline shift with some mass effect on the left
hemisphere. Martins et al. (1995) claimed that the clinical course ruled out any dysfunction of
the left hemisphere arguing that the aphasia: 1) coincided with the development of the right
hemispheric abscess, 2) transiently aggravated in association with the left-sided motor
seizures indicating that the epileptic activity originated in the motor cortex of the right
hemisphere, and 3) followed the course of treatment of the abscess. However, viewing this
patient as an unambiguous CAD can be challenged for various reasons. Firstly, penetrating
brain injuries cause more diffuse brain damage especially when complicated by concomitant
brain tissue infections. In this respect, it should be noted that the highest CAD frequencies
have been reported in association with trauma - up to 18% in the Mohr et al.'s study (1980).
The extreme values reported in these populations (Ludwig, 1939; Mohr et al., 1980)
underscore the importance to control for etiology when a CAD diagnosis is considered. The
fact that no aphasic symptoms were encountered at the time of actual brain destruction but
only in association with the genesis and course of the frontal abscess makes it equally
plausible that the brain was more diffusely affected. Secondly, it is not appropriate to infer
definite conclusions with respect to cerebral language representation on the basis of epileptic
phenomena, since even in focal epilepsies bilateral cerebral dysfunction cannot be excluded
conclusively. Thirdly, the child presented with a history of developmental language
disturbances. Until the age of four the boy was said to produce only unintelligible sounds and
to communicate by means of pointing and gesturing. In marked contrast to his extremely
limited expressive language capacities, auditory-verbal comprehension was considered within
the normal ranges. Speech therapy, which was initiated from the age of three years onwards
and which was continued for four subsequent years, apparently yielded a favourable outcome.
At the age of seven the boy entered the first grade of a normal school but displayed learning
Crossed Aphasia in Ch ildren 15 9
disabilities. He had to repeat the second and fourth grade twice, and difficulties in learning to
read and write as well as difficuties with the acquisition of a foreign language were reported.
An increasing amount of evidence has corroborated the hypothesis that, in cases of
developmental language disorders, alterations in the typical pattern of cerebral dominance for
language may take place. Lou et al. (1984), for instance, showed that confrontational naming
tasks performed in children with developmental aphasia fail to disclose the expected blood
perfusion levels in the left perisylvian region, while Jernigan et al. (1991) as well as Plante et
al. (1991) documented absence of normal anatomical brain asymmetries on magnetic
resonance imaging (MRI). Given the assumption that in Martins et al.'s (1995) patient, the
aphasia solely originated from focal destruction of right frontal brain tissue, the history of a
developmental language disorder might be considered the reflection of an early pathological
condition in which the failure of the left hemisphere to acquire expressive language functions
might have led to a shift in cerebral dominance for language. The Martins et al.'s (1995) case
can therefore alternatively be considered an exponent of a subgroup of patients with
pathologically induced atypical cerebral dominance and anomalous intrahemispheric
specialization for language. The case highlights the importance to distinguish patients with
developmental disturbances from patients with a normal cognitive development in discussions
on functional brain organization.
Figure 1 - Reproduced from Marien et al. (2001a; 2004)

The cases reported by Burd et al. (1990) and Marien et al. (2001a) fulfilled the criteria for
definite CAD. Burd et al. (1990) assessed handedness in a four-year-old boy who had become
aphasic after an infarct in the right middle cerebral artery territory by parental report using the
Edinburgh Inventory (Oldfield, 1971). The integrity of the left hemisphere was confirmed on
CT, which disclosed a large cortico-subcortical fronto-parietal and posterior parietal hypodens
lesion of the right hemisphere. A vascular cause was clearly established and the aphasic as
well as neurocognitive symptoms were thoroughly documented during a three year follow-up
period. Though the child had no personal history of developmental disabilities, his familial
history was positive for dyslexia and seizures on the maternal side and for dyslexia and
dysorthographia on the paternal side. Besides these problems, attentional and concentration
disturbances were reported in the extended family. The patient's subsequent learning
impairments for reading, his further language development, his attention and concentration
problems might either constitute the sequelae of the acquired brain lesion - in which context
the family history may have been coincidental - or might form the expression of a genetic
predisposition for learning disabilities. Marien et al. (2001a) reported a 13-year-old strongly
right-handed girl who developed aphasic symptoms following a right cortico-subcortical
temporo-parietal haemorrhage. Repeat CT and MRI of the brain confirmed the structural
integrity of the left hemisphere. Handedness was assessed by means of the Edinburgh
Inventory (Oldfield, 1971) which yielded a laterality quotient of +100. Medical history,
growth and developmental milestones were normal. She was born at term after normal
gestation and labour and there had been no perinatal or postnatal problems. Her scholastic
achievements had always been above average and there was no familial history of
(neuro)developmental disorder or learning disability. No familial strain of left-handedness
was found after careful inquiry. During a longitudinal follow-up of ten years the aphasic
manifestations were described on a temporal basis in agreement with the three epoch time-
frames for models of aphasia (Mazzocchi & Vignolo, 1979; Basso et al., 1985; Alexander,
1989). To avoid remote functional effects of the hemorrhagic lesion, generally accounting for
the instability of aphasic profiles during the first two or three weeks post-onset (acute phase),
extensive language examinations were performed in the lesion phase. During this phase,
which might last up to four months, the purest and most robust anatomo-clinical correlations
are found (Alexander, 1989). To capture various effects of recovery and functional brain
reorganization, a third extensive neurolinguistic examination was performed during the late
phase, ten years after onset of neurological symptoms. Neurocognitive investigations using
standardized test batteries were performed on a similar temporal basis.
Acquired childhood CAD characteristics
Only the definite CAD cases reported by Burd et al. (1990) and Marien et al. (2001a)
constitute a sufficiently reliable source of information to allow comparison between cases and
theoretical deductions. Though this small number of patients does not allow general
conclusions, potentially relevant tendencies might nevertheless already be suggested with
respect to aphasic semiology and lesion-behavior relationships (Marien et al., 2001a).
Aphasia symptoms. As displayed in Table 2, aphasic symptoms were listed according to a

three epoch time-frame model encompassing acute, lesion and late phase findings (Mazzocchi
& Vignolo, 1979). To present the semiological characteristics in more detail, we added to this
model symptom descriptions 'at early onset' and at an 'end stage' after follow-up. At onset,
both cases invariably presented with a similar aphasic syndrome consisting of mutism and
marked auditory-verbal comprehension disturbances. Within the acute phase, they partly
recovered the ability to comprehend and they also started to speak again. Within a fortnight
the syndrome evolved to a severe adynamic output disorder combined with still prominent
auditory-verbal comprehension defects. Both patients subsequently developed a variety of
symptoms on distinct linguistic levels within the acute phase. The patient described by Burd et
al. (1990) showed severe anomia and prosodic disturbances after an eight-day period of
mutism. The restricted verbal output of the patient described by Marien et al. (2001a) was
contaminated by phonematic paraphasias and dysarthria. The cardinal feature of the dysarthria
was a hypertonic, effortful articulation. During the lesion phase, auditory-verbal
comprehension of daily conversations improved to functional levels for both patients. Apart
from adynamia, they showed prosodic deficits (bradylalia) and anomic disturbances. The
anomia in Burd et al.'s (1990) patient had improved but confrontational colour naming
remained moderately impaired. As part of an adynamic output syndrome, word-retrieval
disturbances in the patient reported by Marien et al. (2001a) were restricted to self-generated
speech. In this patient, repetition normalized during the lesion phase but reading and writing
remained deficient due to (morpho)syntactic errors. The patient described by Burd et al.
(1990) had not yet initiated the process of written language acquisition at the moment of brain
injury. During the late phase, commencing around three months post-onset, the most
important changes in the limited sample of childhood CAD patients seem to consist of a
remission of verbal-auditory comprehension defects. In the patients reported by Burd et al.
(1990) and Marien et al. (2001a), the adynamic features resolved within respectively nine and
four months. In Burd et al.'s (1990) patient, a mild dysarthria and resolution of prosody were
reported as late phase findings, nine months after onset. Dysarthric features were no longer
mentioned at the end of the follow-up, 27 months post-onset. As developmental outcome
cannot be anticipated after this restricted follow-up period, we interpreted the findings
reported in Burd et al. (1990) as late-phase phenomena. Defective comprehension of
grammatical morphemes, insufficient sentence repetition and insufficient oral vocabulary
characterized the language profile of this seven-year-old boy in this period. In contrast to this
observation, the longitudinal follow-up of 10 years of the patient reported by Marien et al.
(2001a) disclosed residual anomia in confrontational naming tasks.
Table 2 — Summary of speech and language symptoms in two childhood CAD cases represented in a
time-frame model
Acute Phase Lesion Phase Late Phase

Onset 21 days >21 days > 3 months 10 yrs
Authors B M B M B M B M M
Symptoms 9m 27m
Comprehesion defects + + + + 0 sem - gm 0
Mutism + + - - - - - -
Adynamia + + 0 + - - -
Agrammatism 0 0 0 + 0 0 0
Repetition defects 0 0 + - + + -
Naming defects + 0 + + + + 0 +
Phonemic Paraphasias 0 + 0 - 0 0 0 0
Reading disturbances 0 0 0 + 0 0 0
Writing disturbances 0 0 0 + 0 0 0
Prosodic disturbances + 0 + + - - 0
Dysarthria 0 + 0 + + 0 0
Legend: B = Burd; M = Marien; + = applicable; - = not applicable; 0 = no data available; y = years;

m = months; sem = semantics; gm = grammatical morphemes
Cognitive and behavioural symptoms. In the acute phase no thorough neurocognitive

assessments were performed in either of the two children under consideration (Table 3).
Neurocognitive assessments performed by Burd et al. (1990) during the lesion phase
disclosed a drawing apraxia, calculation problems and memory disturbances. During this
phase, the patient of Marien et al. (2001a) also displayed calculation deficiencies secondary to
a distorted visuo-spatial processing, right-left orientation problems and a triad of gnostic
disturbances consisting of an autotopagnosia, finger agnosia and astereognosis. In the late
phase, the gnostic and spatial disturbances persisted in this patient and were complicated by
concentration defects, constructional difficulties and minor learning disabilities for
mathematics and foreign language learning.
The patient reported by Burd et al. (1990) showed learning disabilities, concentration
difficulties and neglect phenomena during the late phase. A slight, though not significant
discrepancy between verbal (VIQ = 85) and performance (PIQ = 92) intelligence levels was
reported by Burd et al. (1990) after 27 months of follow-up.
Contrary to these findings, Marien et al. (2001a) reported a more pronounced dissociation
between the VIQ and PIQ in their patient three months post-onset. The WISC revealed a
normal global intelligence quotient of 100 with a discrepancy of 14 points between the verbal
(92) and performance intelligence level (106). After 10 years the IQ levels had homogenized.
Both patients under consideration also developed social and psychological problems. The
patient of Burd et al. (1990) displayed enuresis and elective mutism at school. The patient of
Marien et al. (2001a) withdrew from social contacts and developed feelings of inferiority and
depressive stigmata secondary to physical incapacities.
Table 3 — Summary of lesion and late phase cognitive symptoms of two childhood CAD cases
Lesion LatelPhase
Phase
>21 days > 3 months lOyrs
B M B M M
Authors
9m 27m
Cognitive Domains
Apraxia
bucco-labio-lingual - -
ideomotor - -
ideational - -
constructive + - + +
Agnosia
visual - -
tactile + - +
fingers + - +
autotopagnosia + - +
Neglect
visuo-spatial + + +
motor - +
R/L orientation + -
Calculation + + -
Learning + + +
Memory + + +
Concentration + + + +
Behavior + + +
Intelligence
Global IQ 97 93* 100 99
Verbal IQ 85 106 100
Performance IQ 109 92 98
Legend: B = Burd; M = Marien; + = applicable; - = not applicable; y = years; m = months; blank = no

data available; * personally derived intelligence levels from available scaled scores using the Flemish
and Dutch Wisc-R norms.
Anatomoclinical correlations. On the basis of the anatomoclinical characteristics both cases

were divided into mirror image and anomalous, according to Henderson's (1983) and
Alexander et al.'s criteria (1989). These authors denoted cases with lesion-aphasia
relationships comparable to those following an analogous lesion in the left hemisphere as
mirror-image and cases with an unexpected lesion-aphasia profile as anomalous.
In Burd et al.'s (1990) patient, large cortico-subcortical damage to fronto-parietal and
posterior parietal structures initially caused a severe aphasic disorder with equal involvement
of expressive and receptive abilities. Since the lesion-aphasia profile does not match standard
lesion-symptom configurations this patient represents the first example of the 'anomalous'
CAD in the ACA literature. In the light of the anatomical structures involved and the severity
of the aphasic syndrome at onset, the rapid evolution within 27 months to an only discrete
syndrome of insufficient linguistic performances on the level of grammatical morpheme
comprehension, oral vocabulary and sentence imitation represents a finding beyond the late
phase expectations for standard aphasia. In contrast to this case, the anatomoclinical findings
reported by Marien et al. (2001a) are not within plausible expectations. The posteriorly
located, high parietal lesion which extended subcortically to the crus posterior of the internal
capsule induced an adynamic aphasic syndrome complicated by agrammatism. Two distinct
dominant hemisphere areas are most typically discerned as the anatomical substrate for this
type of aphasia: the area situated anterior and/or superior to Broca's area and the territory
supplied by the anterior cerebral artery. As evidenced by repeated neuroimaging, neither of
these areas had sustained damage. Destruction of the posterior limb of the internal capsule
could also not be held responsible for the adynamic language phenomena (Marien et al.,
2001). Since the lesion-aphasia profile does not match standard lesion-symptom
configurations, within the restricted group of childhood cases this patient represents the first
example of the 'anomalous' CAD variant.
Acquired childhood CAD compared
Acquired childhood CAD versus classical ACA. In contrast to the statement in the updated
standard doctrine on ACA (Brown & Hecaen, 1976), age at onset in the small sample of
childhood CAD patients does not seem to be a significant parameter for aphasia typology. In
this respect, mutism not only occurred in the youngest CAD patient described by Burd et al.
(1990) but also in the patient reported by Marien et al. (2001a) who was 13 years of age at the
onset of neurological symptoms. Agrammatic features were not encountered in Burd et al.'s
(1990) four-year-old patient but were present in the patient reported by Marien et al. (2001a).
Articulatory defects were found in both cases in the context of predominantly nonfluent
aphasia. Though inherently fluent aphasic characteristics such as logorrhoea, semantic or
neologistic jargon did not occur in the limited corpus of childhood CAD patients,
comprehension disturbances might possibly represent a more prominent feature than the one-
third ratio mentioned in the doctrine for patients with standard ACA. In agreement with the
standard doctrine, and similar to adult forms of aphasia, no association between logorrhoea
and comprehension defects was found. In the restricted childhood CAD group, mutism was
found as an early acute phase phenomenon irrespective of lesion location. No such correlation
seems to hold currently for the comprehension disturbances which occurred within the
expectations of standard anatomoclinical correlations. Dysarthric speech characteristics,
prosodic abnormalities and difficulties in written language represent childhood CAD
phenomena that are not dealt with in the standard thinking about ACA.
Acquired childhood CAD versus adult CAD. Our limited sample of childhood CAD cases
seems to share a number of anatomoclinical similitudes with adult CAD. In the childhood
CAD group, the lesion-symptom profiles of the lesion phase follow the distinction made in
adult CAD between anomalous - mirror image cases (Marien et ai, 2004). In the light of the
conceptual evolution of both ACA and adult CAD, it further seems plausible to expect that the
heterogeneity of symptoms of adult CAD will also be reproduced in future descriptions of
childhood CAD. Apart from a variety of expressive and receptive aphasic manifestations, the
limited group of reported patients seems to indicate that prognosis for complete recovery is
not univocally favourable when poor academic performance and subsequent scholastic
difficulties are also taken into consideration.
Acquired childhood CAD and language lateralization
A wide variety of explanations has been proposed for CAD such as absence of decussation of
cortico-spinal tracts (Souques, 1910), a familial strain of sinistrality (Ardin-Delteil et ai,
1923), involvement of phylogenetically older subcortical structures (Habib et ah, 1983), a
simultaneously acquired invisible small left hemisphere lesion (Castro-Caldas et al., 1986),
inhibitory metabolic repercussions of the right hemisphere lesion on left hemisphere
functioning (Schweiger, 1987) and natural variation in functional brain organization due to the
presence or absence of a single gene (Alexander & Annett, 1996). To these, Brown and
colleagues (1973; 1976) added the hypothesis that CAD is the result of incomplete
lateralization of language functions in the left hemisphere during maturation. This hypothesis
implies that during normal maturation a gradual shift takes place from initially bilateral to
unilateral left hemisphere dominance for language. The concept of innate equipotentiality of
both cerebral hemispheres for language at birth and the subsequent process of progressive
lateralization of language functions was firmly established when Lenneberg (1967) claimed -
on the basis of a high incidence of CA in children2 and the transient nature of symptoms in
ACA- that there exists a critical age (nine to ten years of age) below which language functions
are not yet clearly lateralized. This view was called into question and was fundamentally
contested. Hecaen (1976) proposed to lower the critical age referring to Kxashen's (1973)
R i i c c p r ( 1 Q l ^ ^ nKtdinP'/^ nrt ini--ii-J*»ti <-»*=> f i n i i r p ' r\f A A"/,

observation that Basser's (1962) study contained no CAD patients with right hemisphere
damage sustained after the age of five. In their population of 65 brain-damaged children
Woods & Teuber (1978) found a marked disparity between the aphasiogenic power of left
(15/34) and right (4/31) hemisphere lesions. Following the fact that these findings sharply
contrasted with the prevailing insights of hemispheric equipotentiality for language, they
extensively reviewed the literature on AC A and noticed a dramatic change of the incidence of
childhood CAD over time. While one-third of the patients in the studies published before the
1930s were claimed to be CAD representatives, only a proportion of 5% was found in the
studies undertaken after 1940. Woods & Teuber (1978) held undetected bilateral cerebral
involvement before the introduction of antibiotics and mass immunization programs in the
earlier studies responsible for the higher incidence. Similarly, Satz & Bullard-Bates (1981)
concluded from a review based on ACA cases reported since 1940 that: 1) irrespective of age,
the risk of ACA is substantially greater after left than right hemisphere damage, 2) CAD in
children is rare after three to five years of age (or perhaps even earlier), and 3) CA is more
commonly observed in sinistral children regardless of age. Carter et al. (1982) critically
analyzed the data of five ACA studies (Basser, 1962; Shillito, 1964; Isler, 1971; Hecaen,
1977; Woods & Teuber, 1978) applying exclusionary rules and advanced statistical methods
for estimating the distribution of language organization. Data were moreover seperately
analyzed according to the critical age boundary of five years of age as postulated by Krashen
(1973) and Hecaen (1976). For the age group from six to 15 years consisting of 107 non-left-
handed children, Carter et al. (1982) obtained a figure of 97% left-hemispheric language
dominance with a standard deviation of six percent. When the Basser's (1962) study was
included in the analysis of the data of the 64 younger children, figures diminished to 69% and
31%, respectively for the proportion of children with left-sided and bilateral language
representation. The estimated proportion of bilateral language representation dropped from
31% to 16%, while the proportion of left-sided language representation increased from 69% to
84% when the Basser (1962) study was not taken into consideration. Carter et al. (1982)
additionally stated that, when all cases with trauma and undetermined handedness were
excluded from the four remaining studies, only four patients with a right-hemiphere lesion
remained. In none of these children the lesion caused aphasia. Considering the developmental
maturation hypothesis, Carter et al. (1982) concluded from their results that language
distribution for older right-handed children is nearly identical to that for right-handed adults.
They further stated that, even when the hypothesis is adjusted with the modification that right-
handed children below the age of five years have speech organization distributions similar to
those of sinistrals, the developmental maturation hypothesis remains incorrect since the
estimated figure of 3 1 % for bilateral language representation in this group is not compatible
with the 70% figure for left-handed adults. In a less stringent analysis retaining only the
children with undetermined hand preference, the authors obtained similar adjusted proportions
of aphasia following left and right hemisphere lesions to those reported for adults, resulting in
estimated figures for left and right hemisphere language dominance of 0.94 and 0.06,
respectively. On the basis of these findings, Carter et al. (1982) rejected the developmental
maturation hypothesis and adopted the 'developmental invariance position'. This hypothesis
advocates that 1) lateralized cerebral specialization for language is innate, 2) the cerebral
organization of language in children is similar to that of adults, and 3) it does not change with
time.
The second founding factor of the equipotentiality theory concerns the speed and extent of
recovery which were reported to be faster and better in children than in adults. Though there is
general agreement about the fact that functional recovery from comparable aphasic lesions
decreases with age, the variety of different interrelated variables determining the outcome acts
as a confounding factor. Among others, type of language disorder (expressive versus receptive
or combinations of both), etiology (e.g. trauma versus vascular lesions), extent of the lesion,
duration of post-traumatic coma, motor symptoms, and electroencephalographic
characteristics have all been granted important prognostic variables. Neuroanatomical,
neurophysiological and neurobehavioural studies have also discredited the developmental
maturation hypothesis with explicit evidence. Apart from right-left asymmetries on the gross
morphological level, interhemispheric histological differences of cytoarchitectonically defined
regions have been reported in the adult, neonatal and even fetal brain (e.g. Geschwind &
Levitsky, 1968; Teszner et al., 1972; Galaburda et al., 1978; Falzi et al., 1982; Eidelberg &
Galaburda 1984). These biological differences favour the view of a genetic predisposition of
particular brain regions to develop specific functions.
On a functional-anatomical level, interhemispheric differences in the processing of speech
and nonspeech signals, which were demonstrated in the neonate brain using dichotic listening
procedures (e.g. Bertoncini et al, 1989) and auditory event-related potentials (e.g. Molfese et
al. 1988), also contributed to the hypothesis that lateralized specialization for language is
already present at the onset of language acquisition. Our analysis of children with CAD also
suggests that the equipotential and progressive maturation hypothesis is not tenable. In the
CAD patient with brain damage sustained around the age of five years (Burd et al, 1990),
residual language disturbances contend the notion of a complete interhemispheric language
reorganization. Furthermore, Marien et al.'s, (2001a) findings of severe language symptoms
encountered in a 13-year-old patient are inconsistent with the notion that maturation of
cerebral dominance for language should be accomplished around puberty. On the other hand,
the low incidence of CAD in children indirectly corroborates the view that lateralized cerebral
dominance for language represents an innate neurobiological condition.
CONCLUSIONS
From a historical perspective, striking similarities have characterized the conceptual

development and evolution of ACA and CAD. On the common basis of a presumed
incomplete language lateralization process, overt application and even transposition of the
concepts of the standard doctrine on ACA to the clinically comparable CAD profiles led in the
1970s to a confluence of both aphasic conditions. Similar to the conceptual development of

AC A, a wide range of counterexamples provided strong evidence against the 'standard
doctrine on CAD' and modified the concept of a uniform syndrome that, irrespectively of
lesion location, would consist of nonfluent, agrammatic and often transient language
symptoms with a frequent dissociation between superior oral and inferior written language.
No symptom constellations have so far been reported that distinguish CAD from the rich
spectrum of standard aphasic syndromes following damage to the left hemisphere. From the
restricted case study of five children documented in the literature since 1975, only two (Burd
et al, 1990; Marien et al, 2001a) could be reliably assigned an unambiguous diagnosis of
childhood CAD. The other cases (Assal & Deonna, 1977; Assal, 1987; Martins et al, 1987;
1995), however, underscore the importance to stringently control for handedness, etiology and
neurodevelopmental factors when CAD is considered and its impact on functional brain
organization discussed. Supporting evidence exists in favour of developmental language
disturbances as an influencial variable in the functional organization of the brain.
We therefore proposed to form distinct groups of childhood CAD based on
developmental characteristics (Marien et al, 2001a). Patients in whom a partial or complete
right hemisphere dominance for language is suspected in the absence of prior developmental
language anomalies belong to a group of 'primary CAD cases'. Patients with developmental
language disturbances belong to a group of 'secondary CAD cases'. In this group, the
presence of developmental language disturbances might reflect a maturation dysfunction of
the language-dominant hemisphere that likely accounts for the supposed selective or complete
shift of cerebral dominance for language to the right hemisphere.
The common aphasic pattern of the limited number of convincing childhood CAD
representatives consists of an initially severe expressive and receptive language syndrome
characterized by speechlessness and severe auditory-verbal comprehension defects. In its
further course, this syndrome seems to evolve rapidly to a predominantly adynamic output
disorder that may be of long duration. A variety of residual semantic deficits might be
expected as (final) outcome variables after longitudinal follow-up. Agrammatic features,
repetition disturbances, phonematic paraphasias, (subsequent) reading and writing disorders,
prosodic disturbances and dysarthric features occurred. None of these symptoms persisted
after longitudinal follow-up. A variety of neurocognitive symptoms seems to characterize the
cognitive profiles of the representative cases. Visuo-constructive disturbances of apraxic or
perceptual origin occurred in both patients at different epochs. Calculation disturbances
represented relatively early reported lesion-phase phenomena and, along with learning
disabilities, memory disturbances and concentration defects, they constituted a major
(complicating) factor during recovery. Autotopagnostic disturbances occurred in one patient
as well. Although the distribution of the IQ profile might remain asymmetrical, intelligence
levels normalized. Psychogenic socio-behavioural problems were also encountered.
Alterations in social contact and secondary depressive symptoms constituted major reactive
phenomena.
On the anatomoclinical level, several tentative conclusions may be proposed. In the first
place, acquired CAD in the selected patients is irrespective of lesion localization characterized
at onset by a uniform nonfluent aphasic syndrome. The subsequent differential evolution of
predominantly nonfluent aphasic symptoms towards a variety of symptom complexes in the
late phase is a rather unexpected finding. Although this evolution emphasizes the importance
of temporal factors in the description of neurological language phenomena, it might
additionally reflect the anomalous patterns of inter- and intrahemispheric language
organization inherent to this group of exceptional patients.
From the limited number of observations, the childhood CAD characteristics so far
identified differ in a number of aspects from the concepts expressed in the standard doctrine
on ACA. Aphasic typology in childhood CAD seems to be unaffected by age at onset and
comprehension disturbances seem to represent at least initially a much more prominent feature
than the one to three ratio mentioned in the standard doctrine for patients with standard ACA.
The concept of a complete interhemispheric language re-organization after early cerebral
damage is also contended by the aphasic sequelae reported in the CAD children. Taken
together with the extremely low incidence of CAD in children, these considerations
corroborate the view that lateralized cerebral dominance for language represents an innate
neurobiological condition.
There is a need for carefully controlled clinical research on larger groups of cases with
childhood CAD to improve the insights in the pathophysiological mechanisms of anomalous
cerebral language organization in the child. At an elementary level, future clinical studies
might, for instance, be directed to the question whether the semiological heterogeneity in adult
CAD is also present in childhood CAD and whether the anatomoclinical correlations can be
further divided into distinct patterns such as the "mirror image - anomalous case' dichotomy.
APPENDIX I
Case summaries of acquired childhood aphasia associated with right hemisphere lesions:
Clinical and neurolinguistic data.
[Abbreviations: y: years; m: months; M: male; F: Female; AVM: arteriovenous malformation]
ASSAL & DEONNA (1977)
Age/Gender: 5y 9m / M
Etiology: Infarction
Aphasic symptoms: Type and course
During 10 days: mutism, minimal comprehension after heavy prompting; day 14: some words
while singing; during the following weeks: monophasic output; after lm: sporadic
comprehension errors in semi-complex instructions; after 2m: first comprehensible words; the
following months: agrammatic output, poor vocabulary; after 6.5m: unaltered comprehension,
spontaneous speech limited to interjections, phonemic distortions in responses, disturbed
repetition, long delay in naming; after 3v 10m; normal comprehension, reduced spontaneous
speech (<7 words), phonemic distortions, agrammatic errors, repetition disturbances for words
(>4 syllables), slow syllabic reading with phonemic errors, insufficient writing.
12v post-onset: sparse and fragmented spontaneous speech, semantic naming errors,
insufficient verbal fluency, slow reading, inverted spelling, bad orthography (normal verbal
comprehension, repetition, no word deformations).
Lesion site localization
Arteriography on day 2: complete obstruction of the right internal carotid artery
CT after 12 years: right fronto-parieto-temporal lesion involving the 3rd frontal, the 1st
temporal convolution & the supramarginal gyrus, the anterior limb of the internal capsule, the
anterior & middle parts of the caudate nucleus & the putamen, atypical frontal asymmetry (left
anterior region more developed than the right).
MARTINS ETAL. (1987)
Age/Gender: 15 / M
Etiology: Oligodendroglioma, partially resected, irradiation therapy 14 days post-surgery
lm before admission: word-finding difficulties; admission: fluent anomic speech with rare
literal and verbal paraphasias, impaired sentence repetition, impaired Token Test, limited
word reading, partial comprehension of simple texts, agraphia; 7 days post-surgery:
unchanged picture and agrammatism; 14 days post-surgery: worsened aphasia, change to
nonfluent output; 40 days post-surgery: further language deterioration, only sparse isolated
word production, severe anomia, paraphasias, mild dysarthria, agraphia, alexia.
CT on admission: right temporo-parieto-occipital tumor; CT 14 days post-surgery: no
evidence for a regrowth of the tumor or a hematoma; CT 40 days post-surgery: anterior
extension with tumor growth in the basal ganglia, the internal capsule and the frontal white
matter.
BVRDETAL. (1990)
Etiology: Infarction
For 8 days after admission: speechlessness; at day 8: single word utterances, severe anomia,
speech initiation defects; within a few days: correct completion of one-step commands, more
fluent output, repetition of 3 numbers and words, impaired naming and prosody; after 3m (age
5y1m): functional comprehension of conversations, fluent spontaneous speech, short
sentences, mild dysarthria, impaired colour naming; after 9m (age 5y7m): normal sentence
length, sentence repetition most impaired, selective naming difficulties, no prosodic defects;
after 27m (age 7ylm): insufficient comprehension of grammatical morphemes, oral
vocabulary and sentence imitation.
CT on admission: large right cortico-subcortical fronto-parietal and posterior parietal lesion.
Age/Gender: 13/M
Etiology: Gunshot injury & infection
Developmental language disturbances: unintelligible speech until age 4, gestural
communication, normal comprehension; between age 3-7 good recovery with speech therapy,
difficulties in foreign language learning and written language acquisition. Day 3 of admission:
impaired comprehension at simple levels, loss of verbal initiative, nonfluent speech, isolated
words, multiple pauses, word-finding difficulties, perseverations, verbal paraphasias, normal
word repetition. On day 10: mixed transcortical aphasia with impaired comprehension,
nonfluent low verbal output of isolated words or telegraphic sentences (normal articulation,
prosody, automatic speech, repetition of sounds and sentences), rare verbal paraphasias,
moderately impaired naming with semantic paraphasias and perseverations, disturbed reading
comprehension, spelling errors, disturbed written syntax, paragraphias, perseverations and
neologisms; at 3 weeks: speech deterioration; till 10 days post-surgery: total speech loss; at
2.5m: recovery of comprehension.
CT on admission: right frontal abscess.
MARIEN ETAL. (2001)
Age/Gender: 1 3 / F
Etiology: Hemorrhage, AVM resection on day 73
On admission: non-responsive; at day 2: global aphasia; during the following 3 weeks:
nonfluent output, adynamia, phonemic paraphasias, hypertonic dysarthria, disturbed
comprehension; at day 25: full-blown receptive and expressive agrammatism, disturbed
semantic knowledge, bradylalia, hypertonic articulation, agrammatic reading and writing
errors; at day 83 (10 days post-operatively): receptive agrammatism, bradylalic and
dysprosodic output, agrammatic written output; after 10 y: discrete residual anomia on the
BNT visual confrontation naming test.
CT on admission: right temporo-parieto-occipital lesion.
Arteriography: hemorrhagic dislocation of the middle cerebral artery branches, small parietal
AV malformation.
MRI after 10 years: extensive lesion extending from the crus posterior of the internal capsule
to high parietal and encroaching from the precentral sulcus upon the focally dilatated right
lateral ventricle with marked atrophy of the posterior part of the corpus callosum.
APPENDIX 2
Case summaries of acquired childhood aphasia associated with right hemisphere lesions:
Clinical, neurocognitive and neurological data.
[Abbreviations: y: years; m: months; M: male; F: Female; IQ: intelligence quotient; VIQ:
verbal IQ; PIQ: performal IQ]
ASSAL & DEONNA (1977)
Neurocognitive & neurobehavioural symptoms: Type & course
Onset: coma, left hemiparesis, eye deviation to the right; after lm: independent walking,
paralysis of left arm (normal sensation & visual fields); after 12y: unchanged neurological
tableau.
Age/Gender: 15 / M
Admission: visuo-spatial defects, left-sided neglect (normal bucco-facial, limb and drawing
praxis); 40 days post-surgery: mildly impaired constructional praxis (no visual neglect).
1 month before admission: progressive loss of vision, headaches, word-finding difficulties; I
week before admission: left-sided paresis; admission: bilateral papiloedema, left hemianopia,
left hemiplegia and hypoesthesia; died 3m later.
BURD ETAL. (1990)
After 3m (age 5ylm): difficulties drawing simple shapes, insufficient visual motor ntegration,
deficient preschool mathematics level, impaired attention and short-term memory, elective
mutism; after 9 months (age 5y7m): Stanford-Binet IQ = 97, concentration difficulties,

moderate problems on visual discrimination and visual closure tests (no gnostic or apractic
deficits); after 27m (age 7ylm): delay in schooling, special education for learning disabled
children, persisting concentration and memory problems, spatial distortions in clock drawing,
average IQ, motor neglect left hand.
Clinical and neurological findings
Sudden onset of left hemiplegia; on admission: left hemiparesis affecting the arm more than
the leg, confusion; after 9m (age 5y7m): minimal left hand weakness (no sensory loss or
hemianopia); after 27m (age 7ylm): residual left hemiparesis affecting the arm more than the
leg.
Age/Gender: 13/M
Developmental language problems: repeated grades 2 & 4 at school; day 10 of admission:
acquired acalculia, mild left neglect (no spatial dysgraphia or dyscalculia); at 2.5m: remission
of visual neglect, onset behavioural disturbances with attentional deficits, impulsive,
desinhibited behaviour: at 6 m: desinhibited, provocative behaviour made a return to school
impossible, WISC VIQ = 75, PIQ = 105, GIQ = 89.
Admission 5 days after a gunshot injury: complex partial seizures, focal motor seizures,
postictal aphasia, left hemiparesis; at 3 weeks: deterioration of speech and drowsiness
(surgical drainage of the abscess); at 2.5m: fully recovered hemiparesis.
MARIEN ETAL. (2001)
Age/Gender: 13/F
Day 27: unilateral astereognosis, fmgeragnosia, autotopagnosia, afferent dysgraphia and
spatial dyscalculia, right-left disorientation; day 85: amelioration of visuo-spatial
disturbances, unchanged gnostic defects, WISC-R VIQ = 106, PIQ = 92; at lOy normalized IQ
profile, poor concentration, poor visuo-constructional abilities
On admission: sub-coma, left hemiplegia; post-surgery (after 73d & 10y) improvement to
mild residual left-sided sensory-motor hemisyndrome; lOv post-onset: normal.
REFERENCES
Aicardi, J., J. Amsili and J. J. Chevrie, (1969). Acute hemiplegia in infancy and childhood.
DevMed Child Neurol, 11, 162-173.
Alexander, M. P. (1989). Clinical-anatomical correlations of aphasia following predominantly
subcortical lesions. In: Handbook of Neuropsychology (F. Boiler and J. Grafman,
editors), pp. 47-66. Elsevier Science Publishers, Amsterdam.
Alexander, M. P. and M. Annett (1996). Crossed aphasia and related anomalies of cerebral
organization: case reports and a genetic hypothesis. Brain Lang, 55, 213-239.
Alexander, M. P., M. R. Fischette and R. S. Fischer (1989). Crossed aphasias can be mirror
image or anomalous. Brain, 112, 953-973.
Anastasopoulos, G. and D. Kokkini (1962) Cerebral dominance and localisation of the
language functions. Psychiatria Neurologica, 143, 6-19.
Annett, M. 1985. Left, Right, Hand and Brain: The Right Shift Theory. Lawrance Erlbaum
Associates, London.
Ardin-Delteil, Levi-Valensi and Derrieu (1923). Deux cas d'aphasie: I. Aphasie de Broca par
lesion de l'hemisphere droit chez une droitiere, II. Aphasie avec hemiplegie droite chez
une gauchere. Rev Neurol, 1, 14-24.
Assal, G. (1987). Aphasie croisee chez un enfant. Rev Neurol, 143, 532-535.
Assal, G. and T. Deonna (1977). Aphasie par thrombose de la carotide interne droite, chez un
enfant droitier. Oto-Neuro-Ophtalmologia, 49, 321-326.
Basser, L. S. (1962) Hemiplegia of early onset and the faculty of speech with special reference
to the effects of hemipherectomy. Brain, 85, 427-460.
Basso, A., E. Capitani, M. Laiacona and M. E. Zanobio (1985). Crossed aphasia: One or more
syndromes? Cortex, 21, 25-45.
Benson, D. F. (1994). The neurology of thinking. Oxford University Press, New York.
Bernhardt, M. (1885). Ueber die spastische Cerebralparalyse im Kindesalter (Hemiplegia
spastica infantilis), nebst einem Excurse iiber "Aphasie bei Kindern". Virchow's
Archiv fur Pathologische Anatomie und Physiologie und fur Klinische Medicin, 102,
26-80.
Bertoncini, J., J. Morais, R. Bijeljac-Babic, S. McAdams, I. Peretzand J. Mehler (1989).
Dichotic perception and laterality in neonates. Brain Lang, 37, 591-605.
Bingley, T. (1958). Mental symptoms in temporal lobe epilepsy and temporal lobe gliomas.
Ada Psychiatrica et Neurologica, 33, 1-151.
Bramwell, B. (1899). On 'crossed' aphasia and the factors which go to determine whether the
'leading' or 'driving' speech-centres shall be located in the left or in the right
hemisphere of the brain, with notes on a case of 'crossed' aphasia (aphasia with right-
sided hemiplegia in a left-handed man). Lancet, I, 1473-1479.
Branco-Lefevre, A. (1950). Contribuicao para o estudos da psicopatologia da afasia em
criancas. Arq Neuropsiquiatr, 8, 345-393.
Broca, P. (1865). Du siege de la faculte du langage articule. Bulletin de la Societe
d'Anthropologie, 6, 377-95.
Brown, J. W. and F. R. Wilson (1973). Crossed aphasia in a dextral. Neurology, 23, 907-911.
Brown, J. W. and H. Hecaen (1976). Lateralization and language representation: observations
on aphasia in children, left-handers, and 'anomalous' dextrals. Neurology, 26, 183-189.
Brunner, H. and E. Stengel (1932). Zur Lehre von den Aphasien im Kindesalter
(Wortsummheit bei linksseitigem otogenem Schlafelappenabscess). Zentralblatt fur
die gesamte Neurologie und Psychiatrie, 142, 430-450.
Burd, L., G. Gascon, R. Swenson and R. Hankey (1990). Crossed aphasia in early childhood.
DevMed Child Neurol, 32, 528-546.
Carr, M. S., T. Jacobson and F. Boiler (1981). Crossed aphasia: Analysis of four cases. Brain
Lang, 14, 190-202.
Carter, R. L., M. K. Hohenegger and P. Satz (1982). Aphasia and speech organization in
children. Science, 218, 797-799.
Castro-Caldas, A., A. Confraria, T. Paiva and A. Trindale (1986). Contrecoup injury in the
misdiagnosis of crossed aphasis. J Clin Exp Neuropsychol, 8, 697-701.
Claras, A. (1874). Ueber Aphasie bei Kindern. Jahrbuchfur Kinderheilkunde, 7, 369-400.
Collignon, R., H. Hecaen and R. Angelergues (1968). A propos de 12 cas d'aphasie acquise de
l'enfant. Ada Neurol Belg, 68, 245-277.
Conrad, K. (1949). Uber aphasischen Sprachstorungen bei hirnverletzten Linkshander.
Nervenarzt,20, 148-154.
Coppens, P. and S. Hungerford (1998). Crossed aphasia. In: Aphasia in atypical populations
(P. Coppens, Y. Lebrun & A. Basso, editors), pp. 203-260. Lawrance Erlbaum
Associates, New Jersey.
Cotard, J. (1868). Etude sur I'atrophie cerebrate. Faculte de Medecine, Paris.
Critchley, M. (1950). Discussion on speech defects in children. Proc R Soc Med, 43, 4-6.
De Ajuriaguerra, J. (1958). Troubles de langage chez l'enfant au cours des lesions cerebrales
en foyer et des demences. JFr Otorhinolaryngol, 7, 225-240.
De Girardier, J. (1938). Un cas d'aphasie amnesique consecutif a une fracture du crane chez
une enfant de 7 ans. Bulletin de la Societe de Pediatrie, 36, 733-736.
De Girardier, J. and J. Jeannin (1939). Fracture du crane accompagnee d'aphasie apres
intervalle libre chez une enfant de 7 ans. Lyon Chir, 36, 183-189.
Denckla, M. B. (1979). Childhood learning disabilities. In: Clinical Neuropsychology (K.
Heilman and E. Valenstein, editors), pp. 535-573. Oxford University Press, New
York.
Deonna, T. W. (1991). Acquired epileptiform aphasia in children (Landau-Kleffner

syndrome). J Clin Neurophysiol, 8, 288-298.
Eidelberg, D. and A. M. Galaburda (1984). Inferior parietal lobule: divergent architectonic
asymmetries in the human brain. Arch Neurol, 41, 843-885.
Faglia, L., M. R. Rottoli and L. A. Vignolo (1990). Aphasia due to lesions confined to the
right hemisphere in right handed patients: a review of the literature including the
Italian cases. ItalJ Neurol Sci, 11, 131-144.
Falzi, G., P. Perrone and L. Vignolo (1982). Right-left asymmetry in anterior speech region.
Arch Neurol, 39, 239-240.
Farge, E. (1877). Hemiplegie gauche avec aphasie, observation suivie de quelques reflexions
sur la gaucherie cerebrale. Gazette Hebdomadaire de Medecine et de Chirurgie, 31,
488-490.
Ford, F. R. and A. J. Schaffer (1927). The etiology of infantile acquired hemiplegia. Archives
of Neurology and Psychiatry, 18, 323-347.
Freud, S. (1897). Die infantile Cerebrallamung. Alfred Holder, Wien.
Galaburda, A. M., F. Sanides and N. Geschwind (1978). Human brain: cytoarchitectonic left-
right asymmetries in the temporal speech region. Arch Neurol, 35, 812-817.
Geschwind, N. and W. Levitsky (1968). Human brain left right asymmetries in temporal
speech region. Science, 161, 186-187.
Gloning, I., G. Gloning, G. Haub and R. Quatember (1969). Comparison of verbal behavior in
right-handed and non-right-handed patients with anatomically verified lesion of one
hemisphere. Cortex, 5, 43-52.
Goodglass, H. and F. Quadfasel (1954). Language laterality in left-handed aphasics. Brain, 11,
521-548.
Gordon, N. (1990). Acquired aphasia in childhood: the Landau-Kleffner syndrome. Dev Med
Guttmann, E. (1942). Aphasia in children. Brain, 65, 205-219.
Habib, M., Y. Joanette, A. Ali-Cherif and M. Poncet (1983). Crossed aphasia in dextrals: a
case report with special reference to site of lesion. Neuropsychologia, 21, 413-418.
Hecaen, H. (1976). Acquired aphasia in children and the ontogenesis of hemispheric
functional specialization. Brain Lang, 3, 115-134.
Hecaen, H. (1977). Language representation and brain development. In: Brain fetal and
infant: current research on normal and abnormal development (S.R. Berenberg,
editor), pp. 112-123. Martinus Nijhoff Medical Division, The Hague.
Hecaen, H. and J. de Ajuriaguerra (1963). Les gauchers: prevalence manuelle et dominance
cerebrale. Presses Universitaires de France, Paris.
Hecaen, H. and J. Sauguet (1971). Cerebral dominance in left-handed subjects. Cortex, 7, 19-
48.
Henderson, V. W. (1983). Speech fluency in crossed aphasia. Brain, 106, 837-857.
Huglings Jackson, J. (1880). On aphasia with left hemiplegia. Lancet, 2, 637-638.
Humphrey, M. and O. Zangwill (1952). Dysphasia in left-handed patients with unilateral brain
lesions. JAWP, 15, 184-193.
Isler, W. (1971. Acute hemiplegias and hemisyndromes in childhood. Clinics in
developmental medicine. William Heinemann Medical Books, London.
Jendrassik, E. and P. Marie (1885). Contribution a l'etude de l'hemiatrophie cerebrale par
sclerose lobaire. Archives de Physiologie, 1, 51-105.
Jernigan, T., J. Hesselink, E. Sowell and P. Tallal (1991). Cerebral structure on magnetic
resonance imaging in language and learning impaired children. Arch Neurol, 48, 825-
829.
Joanette, Y., M. Puel, J. L. Nespoulous, A. Rascol and A. R. Lecours (1982). Aphasie croisee
chez les droitiers. Rev Neurol, 138, 575-586.
Joffroy, M. A. (1903). Sur un cas d'aphasie sensorielle avec lesion temporo-parietale droite.
Rev Neurol, 16, 112-115.
Kennedy, F. (1916). Stock-brainedness, the causative factor in the so-called "crossed
aphasias". Am JMedSci, 152, 849-859.
Krashen, S. D. (1973). Lateralization, language learning, and the critical period: some new
evidence. Language Learning, 23, 63-74.
Lenneberg, E. (1967). Biological foundations of language. John Wiley, New York.
Lou, H. C , L. Henriksen and P. Bruhn (1984). Focal cerebral hypoperfusion in children with
dysphasia and/or attention deficit disorders. Arch Neurol, 41, 825-829.
Ludwig, M. E. (1939). Beitrag zur Frage der Bedeutung der unterwertigen Hemisphare. Zeit
Ges Neurol Psychiatr, 164, 734-747.
Marien, P., B. Paghera, P. P. De Deyn and L. A. Vignolo (2004). Adult crossed aphasia in
dextrals revisited. Cortex, 40, 39-72.
Marien, P., J. Saerens, W. Verslegers, F. Borggreve and P. P. De Deyn (1993). Some
controversies about type and nature of aphasie symptomatology in Landau-Kleffner's
syndrome: a case study. Acta Neurol Belg, 93, 183-203.
Marien, P., Ph. Paquier, S. Cassenaer and P. P. De Deyn (2002). The history of crossed
aphasia: early development of concepts and hypotheses. Journal of Neurolinguistics,
15, 129-142.
Marien, P., Ph. Paquier, S. Cassenaer and P. P. De Deyn (2003). The history of crossed
aphasia: confluence of concepts. Journal of Neurolinguistics, 26, 1-12.
Marien, P., Ph. Paquier, S. Engelborghs and P. P. De Deyn (2001c) Acquired crossed aphasia
in dextral children revisited. Brain Lang, 79, 426-443.
Marien, P., S. Engelborghs, L. A. Vignolo and P. P. De Deyn (2001b). The many faces of
crossed aphasia in dextrals: report of nine cases and review of the literature. Eur J
Neurol, 8, 643-658.
Marien, P., S. Engelborghs, Ph. Paquier and P. P. De Deyn (2001a). Anomalous cerebral
language organization in a dextral child with acquired crossed aphasia. Brain Lang, 76,
145-157.
Martins, I. P., J. Ferro and A. Trindale (1987). Acquired crossed aphasia in a child. Dev Med
Martins, I. P., N. L. Antunes, A. Castro-Caldas and J. L. Antunes (1995). Atypical dominance
for language in developmental aphasia. Dev Med Child Neurol, 37, 85-90.
Mazzocchi, F. and L. A. Vignolo (1979). Localisation of lesions in aphasia: clinical-CT scan
correlations in stroke patients. Cortex, 15, 627-654.
Milner, B., G. Branch and T. Rasmussen (1964). Observations on cerebral dominance. In:
Disorders of Language (A. V. S. Reuck and M. O'Connor, editors), pp. 200-214.
Churchill, London.
Mingazzini, G. (1925). Uber den heutigen Stand der Aphasielehre. Klin Wochenschr, 4, 1289-
1294.
Minkowski, M. (1930). Gutachten uber einen Fall von kindlicher Aphasie nach Trauma.
Nervenarzt, 3, 411-416.
Mohr, J. P., G. H. Weiss, W. F. Caveness, J. D. Dillon, J. P. Kistler, A. M. Meirowskyand B.
L. Rish (1980). Language and motor disorders after penetrating head injury in Viet
Nam. Neurology, 30, 1273-1279.
Molfese, D. L. and J. C. Betz (1988). Electrophysiological indices of the early development of
lateralization for language and cognition, and their implications for predicting later
development. In: Brain lateralization in children - developmental implications (D. L.
Molfese and S. J. Segalowitz, editors), pp. 171-190. Guilford Press, New York.
Moltschanow (1897). Ein Fall linksseitiger Hemiplegie mit Aphasie. Medizinskoje Obosrenje,
XLVI, n° 24. - C.R. Cbttf. Nervenh., XXVIII, 206-207.
Newcombe, F. and G. G. Ratcliff (1971). Handedness, speech lateralization and ability.
Oldfield, R. (1971). The assessment and analysis of handedness: the Edinburgh inventory.
Oppenheim, H. (1890). Zur Pathologie der Grosshirngeschwiilste. Arch Psychiatr, 15, 560-
587.
Paghera, B., P. Marien and L. A. Vignolo (2003). Crossed aphasia with left spatial neglectt
and visual imperception. Neurol Sci, 23, 317-322.
Paquier, P. F. and H. R. Van Dongen (1996). Review of research on the clinical presentation
of acquired childhood aphasia. Ada Neurol Scand, 93, 428-436.
Paquier, P. F. and H. R. Van Dongen (1998). Is acquired childhood aphasia atypical? In:
Aphasia in atypical populations (P. Coppens, Y. Lebrun and A. Basso, editors), pp.
67-155. Lawrance Erlbaum Associates, New Jersey.
Paquier, P. F., H. R. Van Dongen and M. C. B. Loonen (1992). The Landau-Kleffner
syndrome or 'acquired aphasia with convulsive disorder'. Arch Neurol, 49, 354-359.
Penfield, W. and L. Roberts (1959). Speech and Brain Mechanisms. Princeton University
Press, Princeton.
Plante, E., L. Swisher and R. Vance (1991). MRI findings in boys with specific language
impairment. Brain Lang, 41, 52-66.
Potzl, O. (1926). Uber sensorische Aphasie im Kindesalter. Arch Ohren Nasen
Kehlkopfheilkd, 14, 190-216.
Preobrashenski, P. A. (1893). Zur Pathologie des Gehirns. Neurologisches Zentralblatt, 6,
759-760.
Riese, W. and J. Collison (1964). Aphasia in childhood reconsidered. J Nerv Ment Dis, 138,
293-295.
Sachs, B. and F. Peterson (1890). A study of cerebral palsies of early life, based upon analysis
of one hundred and forty cases. J Nerv Ment Dis, 17, 295-332.
Sachs, B. And L. Hausman (1926). Nervous and mental disorders from birth through
adolescence. Paul B. Hoeber, New York.
Satz, P. (1980). Incidence of aphasia in left-handers: A test of some hypothetical models of
cerebral speech organization. In: Neuropsychology of Left-Handedness (J. Herron,
editor), pp. 399-426. Academic Press, New York.
Satz, P. and C. Bullard-Bates (1981). Acquired aphasia in children. In: Acquired aphasia (M.
T. Sarno, editor), pp. 399-426. Academic Press, San Diego.
Schweiger, A. (1996). Anomaly in relations of hand, language and brain: crossed aphasia in
history cross-examined. In: Classic cases in neuropsycholog y(C. Code, C. Wallesch,
Y. Joanette and A. R. Lecours, editors), pp. 262-273. Psychological Press, New York.
Scollo, G. (1926). Sull'afasia motoria di Broca. A proposito di un interessante trauma del
capo con complicazioni di ascesso cerebrale. Policlinico [Pratica], 13, 437-41.
Shillito, J. (1964). Carotid arteritis: a cause of hemiplegia in childhood. JNeurosurg, 21, 540-
551.
Smithies, F. (1907). Hemiplegia as a complication in typhoid fever, with report of a case.
JAMA, 49, 389-395.
Souques, M. A. (1910). Aphasie avec hemiplegie gauche chez un droitier. Rev Neurol, 20,
547-549.
Stone, L. (1934). Paradoxical symptoms in right temporal lobe tumor. J Nerv Ment Dis, 79, 1-
13.
Subirana, A. (1958). The prognosis of aphasia in relation to the factor of cerebral dominance
and handedness. Brain, 81, 415-425.
Subirana, A. (1960). Vision panoramica de los problemas que comporta la afasia infantil. An
Med, 46, 141-155.
Subirana, A. (1969). Handedness and cerebral dominance. In: Handbook of Clinical
Neurology (P. J. Vinken and G. W. Bruyn, editors, Vol. 4, pp. 248-272. North-
Holland, Amsterdam.
Taylor, J. (1905). Paralysis and other diseases of the nervous system in childhood and early
life. J & A Churchill, London.
Teszner, D., A. Tzavaras, J. Gruner and H. Hecaen (1972). L'assymetrie droit-gauche du
planum temporale: a propos de l'etude anatomique de 100 cerveaux. Rev Neurol, 179,
444-448.
Van Dongen, H. R. and M. C. B. Loonen (1977). Factors related to prognosis of acquired
aphasia in children. Cortex, 13, 131-136.
Wada, J. and T. Rasmussen (1960). Intracarotid injection of sodium amytal for the
lateralization of cerebral speech dominance: Experimental and clinical observations.
Neurosurgery, 17, 266-282.
Wagner, W. and K. Mayer (1933). Psychologische Untersuchung an der Sprachstorung einer
Zwolfjahrigen (Psychopathologische Studie zur Frage der Grundfunktionsstorungen).
Monatsschr Psychiatr Neurol, 87, 108-55.
Wallenberg, A. (1886). Ein Beitrag zur Lehre von den cerebralen Kinderlahmungen. Arch
Kinderheilkd, 24, 384-439.
Woods, B. T. and L. Teuber (1978). Changing patterns of childhood aphasia. Ann Neurol, 3,
273-280.
Aphasic spontaneous speech at 12-year follow-up 181
10
RECOGNIZABLE SPONTANEOUS LANGUAGE

CHARACTERISTICS IN A YOUNG ADULT
TWELVE YEARS AFTER SHE BECAME APHASIC
AS A CHILD
Philippe F. Paquier
Department of Neurology, University Hospital Erasme ULB; Department of Linguistics, Free
University of Brussels (VUB-ULB); Department of ENT-surgery, School of Medicine,
University of Antwerp, Belgium.
Valerie R. van Maldeghem

Speech and Hearing Rehabilitation Center, University of Ghent, Belgium.
Hugo R. van Dongen

Department of Child Neurology, Children's University Hospital Sophia-Rotterdam, The
Netherlands.
and Wouter L. Creten

Department of Physics, Research Group on Biomedical Physics, University of Antwerp,
Belgium.
Abstract — We report on a patient who developed fluent aphasia when she was 9. Spontaneous
language characteristics were analyzed in the acute phase, and 1014 and 12 years post-onset to
determine why her aphasia was still instantly recognizable in adulthood by the way she talked. Video-
and audiotaped recordings were rated according to several psycholinguistic variables, among which
mean length of utterance (MLU). Pragmatic aspects of communication such as presupposition and
topic-maintenance were also assessed. Evaluations in the acute phase and at follow-up did not show
any significant difference between MLU measurements. The patient produced long utterances and
spoke fluently. A number of language characteristics were present to a varying degree both in the
acute phase and at follow-up: semantic verbal paraphasias (including superordinate substitutions),
passe-partout words, circumlocutions, conduites d'approche, dyssyntaxis, false starts, inappropriate
182 Neurogenic Language disorders in Children
use of anaphors, and empty speech. At follow-up, a quantitative improvement of the patient's verbal
output was observed. The follow-up did not reveal any new language features or anomalies. The
patient's verbal output was still recognizable at 12-year follow-up because of a combination of factors:
(a) persisting anaphoric difficulties which, along with lack of cohesion, explicit information and
accuracy, resulted in inadequate pragmatic interaction and disjointed and fragmented discourse; (b) the
presence of comparable paralinguistic aspects of communication (such as intelligibility, voice quality,
prosody, and rate of speaking) in the acute phase of aphasia and at follow-up; and (c) the presence of
mild but lasting word-finding difficulties with similar strategic adaptations at follow-up. Although
MLU measurements did not differ significantly between assessments in the acute phase and at follow-
up, they did not add much value to the characterization - hence, the recognition - of the patient's
verbal output.
Keywords: acquired aphasia, children, tumor, recovery
INTRODUCTION
Long-term follow-up studies of children with early (i.e., before 6 months of age) left-sided
focal brain injury are numerous and tend to demonstrate that, if they do not develop epilepsy,
these children usually acquire language skills that are within the normal range, that is, they
are not aphasic (e.g., Woods and Carey, 1979; Bates et al, 1997, 2001; Nass, 1997; Reilly et
al, 1998; Bates, 1999a, b). In contrast, children with acquired aphasia - i.e., a childhood
language disorder resulting from a cerebral lesion sustained after onset of language
acquisition - have a less favorable prognosis than children with early brain lesions (Paquier
and Van Dongen, 1996, 1998). Most longitudinal follow-up studies, which mainly focus on
school-aged children, emphasize the deleterious effects of acquired childhood aphasia (ACA)
on school achievement even in children who clinically recovered from aphasia (e.g.,
Alajouanine and Lhermitte, 1965; Hecaen, 1983; Cooper and Flowers, 1987; Cranberg et al,
1987; Klein et al, 1992; Martins and Ferro, 1992; Pitchford et al, 1997; Pitchford, 2000).
However, little is known about the long-term outcome of ACA in adult age, with the
likely exception of language outcome associated with Landau-Kleffner syndrome, i.e.,
acquired epileptic aphasia (Mantovani and Landau, 1980; Deonna et al, 1989; Van Dongen et
al, 1989). Woods and Teuber (1978) reported on a patient who at the age of 5 acquired jargon
aphasia as a result of stroke. The patient was seen again when he was 21. At follow-up he was
not clinically aphasic. Nevertheless, he displayed some difficulties in reading aloud, in
naming to category and from definition, in spelling, and in comprehending spelled words. He
performed normally on the Token Test. Watamori et al. (1990) described the long-term
outcome of linguistic and nonlinguistic functions in three adults with ACA. This study
revealed that, despite the early restoration of functional communication ability, language
recovery was not complete, and syntactic processing difficulties, limitation of lexical-
semantic abilities, as well as written language processing difficulties were still present in
adulthood. When compared to controls matched for lexical-semantic ability, all three patients
showed poorer performance on spontaneous speech measures such as production of complex
syntactic structures and efficiency of production. Their case histories suggested that the
influence of residual linguistic impairments was not restricted to academic achievement but
also had an effect on the subjects' personalities and social development. Watamori et al's
(1990) findings also revealed that child-onset aphasics used similar strategies for functional
compensation as seen in recovered adult aphasics.
We had the opportunity of assessing a previously reported girl with ACA (Paquier and
Van Dongen, 1991, case 2) both in the acute phase (at the age of 9 years 3 months) and at 12-
year follow-up. Remarkably, after all these years, the patient could still blindly be recognized
as aphasic just by the way she expressed herself, even by naive listeners who never met her
and who only had the opportunity of listening to recordings of her spontaneous language.
This study was thus geared to analyze the patient's spontaneous verbal output in the acute
phase and 10/4 and 12 years post-onset, to determine why she was still instantly recognizable
in adulthood. Several linguistic and nonlinguistic variables as well as pragmatic aspects of
communication were evaluated.
PATIENT AND METHODS
Patient
At the age of 9 years and 3 months, this right-handed, Dutch-speaking girl was admitted
because of lasting complaints of headache, nausea, and fatigue. Her psychomotor and
language development had been normal, and she was attending normal primary school.
Upon admission, the girl was conscious and well-oriented. No aphasic signs were present.
A clinical neurological examination revealed papilledema, a right homonymous hemianopia,
and a right-sided hypoesthesia. An electroencephalogram showed left temporal focal
abnormalities. A space-occupying neoplastic lesion in the left temporoparietal area was
revealed by CT scan and confirmed by cerebral angiography. The tumor - a ganglioneuroma
as revealed by the neuropathological examination - was partially resected fifteen days after
admission. After surgery the patient developed grand-mal seizures and, later on, right-sided
focal insults. She was aphasic and slightly hemiparetic. Her general conditions improved
steadily, and she was discharged one month after admission. However, she still had epilepsy,
and presented with a persisting right homonymous hemianopia.
At follow-up, two years post-onset the girl demonstrated mild aphasic disturbances which
increased during and after epileptic seizures, thus resulting in a transitory exacerbation of
aphasia, with different clinical manifestations, including speech arrests, depending on the
location of the epileptic focus. Nine years post-onset, the patient was started on chemotherapy
because of increasing vision problems. They were believed to be due to residual tumor
growth, as later revealed by MRI. One year later she was irradiated as vision further
deteriorated despite chemotherapy. Since then her sight has remained stable, and the recurring
epileptic insults which used to produce a transient aggravation of her language impairment
disappeared. The patient has been seizure-free ever since but she is still under anti-epileptic
therapy because she fears new fits.
Immediately after the post-operative time, the girl presented with fluent aphasia which in
the early phase was predominantly characterized by logorrheic utterances, mild auditory
comprehension and repetition difficulties, and by lasting word-finding difficulties with
conduites d'approche1. She was spontaneous and cheerful during the post-operative
assessments, as well as lO'A and 12 years post-onset. Her spontaneity was never affected by
1
For detailed information regarding the patient's neurological and aphasiological condition in the acute phase
and 2 years post-onset, see Paquier and Van Dongen (1991).
the language disorder. Nevertheless, she could not resume normal education after discharge
from hospital, and went to a school for visually disabled children. The teachers reported
learning and memory difficulties which eventually resulted in poor academic achievements.
Despite these difficulties she succeeded in completing secondary school, and she is currently
working in a day nursery. Twelve years post-onset, the patient complained of impaired vision,
memory difficulties, and attentional problems. She did not openly mention the presence of
language difficulties, although she was presenting with a residual anomic aphasia. Her
spontaneous language was fluent and sporadically empty, with some verbal paraphasias and
circumlocutions. Word-finding difficulties, however, were most conspicuous during object
naming tasks. Repetition of spoken language was only mildly disturbed by mnestic
difficulties which resulted in occasional word omissions or word order errors in longer
sentences. The hemianopia had remained unchanged.
Language sample. Spontaneous language was elicited in a free interview situation by means
of standard open-ended questions concerning the patient's interests (holidays, school or
professional activities, pets, etc.). The interviewer's task was to prompt the patient or to
initiate conversation. This was the conversational discourse. In addition, samples of narrative
discourse were obtained by asking the patient to tell the well-known Little Red Riding Hood's
tale. Both the conversational and the narrative discourse were video- or audiotaped. Three
language samples were obtained: sample LI (229 utterances; 19 min 27 sec) was a videotaped
recording made in the acute phase (approximately 6 days after surgery), samples L2 (177
utterances; 14 min 23 sec) and L3 (114 utterances; 12 min 36 sec) were audiotaped recordings
at lO'A and 12 years, respectively. The patient was alert and co-operative during the
recordings.
Transcription. The recordings of the conversational and narrative discourses were

orthographically transcribed by one of the examiners (VVM) according to the TOAST
procedure (Taal-Onderzoek via Analyse van Spontane Taal - Language Assessment through
Analysis of Spontaneous Language) (Moerman-Coetsier and Van Besien, 1987). Neologisms
and unintelligible productions were transcribed using the International Phonetic Alphabet.
One week later, the language samples were completely re-listened to and, where necessary,
corrections were made to the first transcripts. Another week later, if necessary, final
adaptations were made after listening again to transcripts. Then, the patient's and
interviewer's turns were segmented into utterances, an utterance being considered as "a unity
of expression that utters a thought, a feeling, an intention, etc." (Moerman-Coetsier and Van
Besien, 1987, p. 45). Criteria for segmentation were changes in intonation and the occurrence
of pauses, as determined by the TOAST procedure.
Reliability of transcription. An independent and naive judge, who by reason of her profession
was familiar with orthographic transcription, was asked to transcribe five minutes of
spontaneous language selected at random from each of the three language samples. Inter-
judge agreement was then determined by comparing her transcripts with the corresponding
three original transcripts, and calculated according to a frequency ratio in which the number
of similarly transcribed items was divided by the total number of transcribed items and
multiplied by 100. The mean inter-judge percent agreement was 9 1 % (94% for LI, 91% for
L2, and 88% for L3). These percentages are relatively high, which indicates a fair inter-judge
agreement.
Analysis of spontaneous language. Mean length of utterance (MLU) was rated according to
the TOAST procedure. As the easiest way to measure MLU, the MLU length expressed in
words (MLUW) was considered (Beheydt, 1983). MLUW and MLU of the three longest
utterances expressed in words (MLU3) were calculated according to Wagenaar et al.'s (1975)
psycholinguistic analysis of spontaneous language in adult aphasics2. We also calculated the
length expressed in morphemes (MLUm) as several studies indicate that MLUm can be
considered a measure of morphosyntactic development, the length of the utterances also
depending on the acquisition level of morphology, at least in young children (Brown, 1973;
Bloom and Lahey, 1978; Moerman-Coetsier and Van Besien, 1987). There is a high
correlation between MLU m and MLUW, especially in younger children (Moerman-Coetsier
and Van Besien, 1987).
Other variables of study included the presence of neologisms, circumlocutions, conduites
d'approche, dyssyntaxis, hedging, false starts, and anaphors. hi line with Kerschensteiner et
al. (1972), we also rated a number of variables relevant for the analysis of adult spontaneous
aphasic language, which have subsequently been rated in children with ACA too (Van
Dongen et al, 2001): rate of speaking, prosody, effort, articulation, perseveration, word
choice (such as passe-partout words), pauses, verbal paraphasias, and literal paraphasias (for
a description, see the Appendix).
Application of TOAST procedure to language samples. The work samples used to compute
MLU were defined following the TOAST procedure. According to the assessment manual,
first a sample consisting of at least 110 usable patient's utterances - 55 coming from
conversational discourse and 55 from narrative discourse - has to be selected from the
complete language samples. Then, the sample is reduced to 100 usable utterances (2 x 50, viz.
the 6th up to the 55 th utterance included) by omitting the first five utterances from each
discourse (because utterances tend to be shorter at the beginning of verbal interaction).
As our recordings did not allow us to apply this procedure literally, we changed it as
follows: three work samples were chosen to contain the same number of usable utterances as
those present in the smallest language sample (in our case, 101 usable utterances in L3) with a
minimum of 100 utterances. Consequently, MLU was calculated as follows:
MLU m = 2 morphemes / 101
MLUW = S w o r d s / 1 0 1
MLU3 = Z words of 3 longest utterances / 3
2
Although direct comparisons between data collected in aphasic adults and children with this method are not
possible for methodological reasons, we would like to point out that the computation of MLUW is identical in
TOAST and Wagenaar et a/.'s (1975) study.
The remaining (non-TOAST) variables were rated on the complete language samples.
RESULTS
To ascertain the representativeness of the work samples and, thus, the validity of the MLU
analysis we checked the distribution of the utterance lengths throughout the work samples
according to the TOAST procedure. Language samples that are representative of the patient's
overall verbal output, indeed, show a normal distribution of utterance lengths (Moerman-
Coetsier and Van Besien, 1987). Kolmogorov-Smirnov test results showed that utterance
lengths are normally distributed throughout the three language samples (p > 0.10) (Figure 1);
a reliable interpretation of MLU is thus possible.
Table 1 displays the different MLU values for the three work samples. Table 2 shows the
non-TOAST spontaneous language characteristics for the three language samples. A Bartlett's
test for homogeneity of variances did not find any significant difference in standard
deviations of MLU m across the three work samples (B = 0.27; d.f. = 2; p > 0.05). A one-way
analysis of variance (ANOVA) indicated that there was no significant difference across the
three MLUm (VR = 2.86; d.f. = 2,300; p > 0.05) (Table 1). MLUW remained stable across the
three work samples, whereas MLU3 slightly increased (Table 1).
Table I — MLU values computed for the three work samples
MLU LI L2 L3
In morphemes (MLUm) 9.6 11.0 9.6
In words (MLUW) 8 9 8
MLUwof the 3 longest utterances (MLU3) 16 19 18
SDofMLU,,, 4.7 4.8 4.6
Note: MLU = mean length of utterance; L = language sample; SD = standard deviation
Regarding the remaining (non-TOAST) language characteristics, Table 2 shows that

paraphasias were observed at onset of aphasia but they decreased dramatically in the course of
follow-up. Twelve years post-onset only two semantic paraphasias were noted. Neologisms
were not observed in the acute phase or at follow-up. In the acute phase, passe-partout words
mainly concerned the class of verbs and, to a lesser degree, that of substantives; 12 years later
they were sporadically recorded as empty substantives. Circumlocutions and conduites
d'approche only made up a small proportion of all utterances in the acute phase and at follow-
up. However, the percentage of false starts against the total number of utterances was
markedly more important both in the acute phase and 12 years post-onset.
In the acute phase and at follow-up, the patient experienced mild syntactic difficulties
which sporadically manifested themselves as dyssyntaxis characterized by a wrong word
sequence within sentences, concord errors between subject and verb form, a wrong article use,
a wrong verb tense use, erroneous application of the double negation rule, and confusions
between substantives and nominal forms.
The rate of speaking increased from 108 words per minute in the acute phase to 129 and
121 words per minute lO'A and 12 years post-onset. Based on a perceptual analysis of speech
characteristics, no anomalies of prosody, effort, articulation, and pauses were observed in the
acute phase or at follow-up.
Figure I - Distribution of utterance lengths expressed in morphemes.
In the acute phase literal and verbal perseverations were observed, as well as a small
number of word group perseverations. At follow-up 12 years post-onset, perseverations had
almost completely disappeared (except for one verbal perseveration).
Table 2 — Non-TOAST spontaneous language characteristics rated throughout the three language
samples (percentages are calculated against total number of utterances)
Speech and language variables LI L2 L3
Verbal paraphasias: Morphemic 0.9% — —

Semantic 1.3% 2.8% 1.8%
Unrelated 1.3% — —
Total 3.5% 2.8% 1.8%
Literal paraphasias 2.6% — —
Neologisms — — —
Circumlocutions 3.1% — 1.8%
Conduites d'approche 1.7% — 0.9%
Dyssyntaxis + + +
Hedging + + —
False starts 15.7% 9.0% 10.5%
Anaphors: Erroneous use 2.6% 4.5% 3.5%
Indefinite 4.8% 2.8% 2.6%
Omission 2.2% 1.1% 0.9%
Total 9.6% 8.4% 7.0%
Rate of speaking (wpm) 108 129 121
Prosody N N N
Effort — — —
Articulation N N N
Perseverations: Literal 5.2% 0.6% —
Verbal 3.1% 3.4% 0.9%
Word group 0.9% 0.6% —
Total 9.2% 4.6% 0.9%
Word choice: Empty + + +
Cliches — — —
Passe-partout words 8.7% — 1.8%
Pauses N N N
Note: L = language sample; wpm = words per minute; N = normal; + = present; — = absent
Finally, especially in the acute phase but also at follow-up the patient experienced
difficulties with anaphors (Table 2). Anaphoric difficulties were classified as indefinite
anaphors (anaphors without a referent), erroneous use of anaphors, and omission of anaphors.
During the acute phase, indefinite anaphors occurred more often than omissions and
erroneous uses of anaphors (Table 2). At follow-up, the number of indefinite anaphors and
anaphoric omissions decreased, while erroneous use of anaphors slightly increased.
To summarize, evaluations in the acute phase and at follow-up did not show any
significant difference across MLUm. The patient used long utterances and spoke fluently. The
analysis further revealed the presence of a varying number of language characteristics both in
the acute phase and at follow-up, including semantic verbal paraphasias (including
superordinate substitutions), passe-partout words, circumlocutions, conduites d'approche,
dyssyntaxis, false starts, inappropriate use of anaphors, and empty speech. At follow-up, a
quantitative improvement in the patient's verbal output was observed but no new language
features or anomalies.
DISCUSSION
Methods for the analysis of spontaneous language in aphasic subjects are scarce in the Dutch-
speaking area, especially with regard to a pediatric population (Evy Visch-Brink, personal
communication). Moreover, no methods have been developed which allow similar analyses
and reliable comparisons of spontaneous language in both child and adult aphasic
populations. With these practical constraints and the unavailability of comparable norms in
mind, we chose the TOAST procedure (Moerman-Coetsier and Van Besien, 1987) to analyze
aspects of our Dutch-speaking patient's language that were related to utterance length, as a
number of linguistic variables assessed by TOAST (e.g., MLU) are universally recognized in
language research (Wagenaar et ai, 1975; Vermeulen et al, 1989; Eisenberg et ai, 2001;
Bol, 2003). However, a major methodological limitation due to this choice is the impossibility
to judge the normality or abnormality of our patient's performance levels on the assessed
TOAST variables, as norms corresponding to her chronological age are not available. Taking
these restrictions into account, we considered the TOAST analysis as true intellectual exercise
aimed at collecting potentially interesting information that might add value to the analysis of
the other linguistic and non-linguistic variables.
Our patient's MLUm in the acute phase (9.55) is exceeding the mean MLU m of the oldest
norm group, i.e., 48-54 months (mean 5.25, s.d. 0.58), thus preventing any firm interpretation
as to the normality of her MLU m . It can only be stated that her MLU m in the acute phase is
comparable to that at follow-up, which might suggest that her expressive morphosyntactic
ability did not significantly change since onset of aphasia. Whether this invariability reflects
an acquired aphasic morphosyntactic disorder persisting in the long-term or a normal
morphosyntactic acquisition level already present in the acute stage is perhaps less a matter of
speculation than it may appear at first sight. Indeed, one may assume that the comparable
MLUm values indicate an age-adequate acquisition level of morphology already present at
onset of aphasia for at least three reasons: (a) unlike the early stages of language acquisition
during which any morphosyntactic progress causes utterance lengthening, an increase in

structural complexity may cause no change in utterance length in later acquisition stages
(Feyereisen et al., 1991); (b) the active development of syntax and morphology is generally
complete between the age of VA and 814 (Gillis and Schaerlaekens, 2000), whereas our patient
became aphasic beyond that period; and (c) the overall clinical course of our patient's aphasic
language disorder evolved quite favorably over the years, making the existence of a specific,
chronic, and stable deficit of morphosyntaxis since onset of aphasia most unlikely.
When considering the other language variables of study, we observed that word-finding
difficulties were prominent in the acute phase, and more subtle at follow-up. At both times
they appeared to be the cause of false starts, repetitions, conduites d'approche (the active
search for the right word), circumlocutions, verbal and literal paraphasias, and passe-partout
words. Semantic verbal paraphasias also occurred at follow-up. The use of circumlocutions,
being object or function descriptions of words, delayed the transfer of information. Word-
finding difficulties also gave rise to semantically empty spontaneous language. At follow-up,
sporadic passages of empty speech were caused by the use of vague or generic substantives,
passe-partout words (e.g., thing, thingummy, stuff), and all-purpose verbs (e.g., to do) as
substitutes for information-loaded words. The use of circumlocutions and passe-partout
words might not only be the mere manifestation of word-finding difficulties, but also an
adaptive strategy to compensate for these difficulties.
Interestingly enough, both in the acute phase and at follow-up the patient gave evidence of
metalinguistic awareness, as she often self-corrected verbal paraphasias even before
pronouncing the word completely, or interrupted her utterances to re-edit them (so-called false
starts). The use of hedges also reflected the patient's metalinguistic awareness, as in doing so
she admitted not being sure about certain statements of hers. It is clear that the patient was not
anosognosic.
Especially in the acute phase but also at follow-up, at times the patient's message was
hardly intelligible owing to the lack of explicit and accurate information. For instance, she
frequently used pronouns without antecedents instead of substantives. The lack of accuracy
was also evident in the disordered cohesion between co-referring words. There is cohesion
between sentences when two words (e.g., a pronoun and its antecedent) co-refer to the same
entity. The most frequently studied types of cohesion are pronominal and lexical co-reference.
Lexical co-reference occurs when the definite article 'the' precedes a noun to signal that this
noun refers to a previously mentioned entity (Davis et al, 1997). Pronominal co-reference
occurs when a pronoun such as a personal or possessive pronoun refers to an antecedent. It
appears that the patient only had difficulties with pronominal co-reference. In the acute phase,
incomplete cohesion resulted from the use of indefinite anaphors (the information referred to
by the anaphors was not always provided by the patient), whereas other anaphors were
omitted or erroneously used. Again, at follow-up we observed this inadequate use of
anaphors, which resulted in an inappropriate pragmatic interaction as the global meaning of
the patient's message could not sufficiently be inferred from the context. Anaphoric errors
and omissions of relevant information often resulted in a disjointed and fragmented discourse.
A few pragmatic aspects of her communicative behavior were also manifest at follow-up.
The patient behaved as if she assumed the listener had sufficient foreknowledge to understand
the message (presupposition), and omitted significant parts of the information. She did not
always succeeded in conveying the content and purpose of her message. A relation between
this pragmatic aspect of communication and her anaphoric difficulties seems plausible.
The paralinguistic aspects of the patient's utterances (the so-called suprasegmental

characteristics of language: intelligibility, loudness, voice quality, prosody, fluency, and rate
of speaking) were always adequate, both in the acute stage and at follow-up. From a video
recording made in the acute stage we were able to analyze the nonverbal aspects of the
patient's communication such as posture, gestures, eye contact, and facial expressions. All
these facets were adequate, and they still were 12 years post-onset. Both in the acute phase
and at follow-up, the patient was aware of the social rules of conversation as she adequately
waited for her turn without interrupting the interlocutor. When she did not understand a
question, she asked for clarification.
After having discussed our patient's aphasiological features in the acute stage and at
follow-up, we would like to address some methodological issues that are of importance in a
clinical setting. The lack of significant difference in MLU values between the first assessment
and follow-ups raises some questions about the clinical utility of this measure in
aphasiological practice. Indeed, MLU did not differ significantly between both assessment
times, although our patient showed an indisputable clinical recovery, even if she still
presented with some residual anomic aphasia in the long term. Consequently, it is
questionable whether MLU measures are capable of objectivizing such a favorable evolution.
Wagenaar et al. (1975) demonstrated that MLU measures are helpful in differentiating
nonfluent from fluent adult aphasics, nonfluent patients producing utterances shorter than
6.25 words and fluent patients having utterances longer than 7.33 words. In other words,
fluent aphasics are capable of producing utterances, the length of which equals that of normal
speakers, even if their sentences are semantically empty and morphosyntactically inadequate.
Consequently, it is not surprising that within the group of fluent aphasics, MLU
measurements performed at two different points in time in the same patient do not show
significant differences despite a significant recovery, as these measurements are not indicative
of the intrinsic characteristics of semantic adequacy or morphosyntactic complexity and
exactness. MLU measures are not able to capture positive signs such as paraphasias and
dyssyntaxis. As a matter of fact, any improvement in structural complexity or semantic
adequacy may cause no change in utterance length at all once the early stages of language
acquisition are completed (Feyereisen et al, 1991). Consequently, MLU measures reflect
nothing but utterance length, which is known to be normal in fluent aphasics (Benson and
Ardila, 1996). Regarding our patient, similar MLU values in the acute stage and 12 years later
only reveal that the mean length of her utterances remained comparable but they are not
indicative of the nature of the underlying defective linguistic mechanisms or the qualitative
improvements in content. As a consequence, it seems that in our patient MLU measurements
would only have contributed to the characterization of her aphasic deficit if a change in
utterance length (i.e., a quantitative change) would have occurred between the two
assessments. The clinical utility of MLU measurements in aphasic patients who are fluent
from the very beginning of their aphasia appears, thus, to be quite limited.
CONCLUSION
Our patient presented with a fluent type of aphasia since the onset of her language problems
(Paquier and Van Dongen, 1991). In a psycholinguistic study on spontaneous language in
adult aphasics, Wagenaar et al. (1975) concluded that patients can be classified as fluent or
nonfluent on the basis of two variables, viz. MLU and rate of speaking. They found that fluent
patients had utterances longer than 7.33 words and that they uttered more than 540 words in 6
minutes (90 wpm). Although their analysis does not apply to a pediatric population of aphasic
subjects, and, consequently, does not allow reliable comparisons between adults and children,
we would yet like to remark that already in the acute phase of aphasia (and, later on, during
follow-up assessments) our patient's performance levels on MLUW and speech rate exceeded
both cut-off values proposed by Wagenaar et al. (1975). No anomalies of prosody, effort,
articulation, and pauses were observed. Positive signs (Van Hout et al, 1985) such as
dyssyntaxis and paraphasias were noted at the onset of aphasia. The patient's verbal output
features, which are specific to fluent aphasia, were confirmed in a recent study on
conversational speech fluency in the acute phase of ACA (Van Dongen et al, 2001). The
patient's clinical recovery is indisputable, yet not complete. Twelve years post-onset she
presented with residual anomic aphasia which was most evident during naming tasks.
In our opinion, the patient's verbal output was still recognizable as aphasic 12 years post-
onset because of a combination of factors. As most salient feature we would suggest the
anaphoric difficulties and the lack of explicit information and semantic accuracy which, along
with a disturbed cohesion between discourse elements, led to an inadequate pragmatic
interaction and a disjointed and fragmented discourse. In addition, paralinguistic aspects of
the patient's verbal output such as intelligibility, voice quality, prosody, and rate of speaking
did not differ subjectively across the three recorded language samples. Also, the patient still
displayed word-finding difficulties with similar strategic adaptations at follow-up. As MLU
measures are not capable of capturing any relevant changes in morphosyntactic complexity
and exactness or in semantic adequacy, we do not feel that the similar MLU values in the
acute stage and 12 years later add much value to the recognition of our patient's verbal
output, except perhaps that they demonstrated that her utterances did not substantially
increase or decrease in length over time. As a consequence, we believe that our findings shed
but a subdued light on the clinical utility of MLU measures in aphasiological practice, at least
in fluent aphasic subjects.
Finally, the present study confirms that subtle but lasting language difficulties can persist
as long-term outcome of ACA. In cases of tumor-related etiology, the course and outcome of
aphasia usually parallel the growth of the tumor, except when aphasia is caused by acute
bleeding or sudden cystic enlargement (Van Hout, 1990). Surgical resection of the tumor can
cause aphasia in a previously non-aphasic subject, and post-surgical epileptic disorders can
impair the recovery process by causing transient aggravations of the language impairment.
ACKNOWLEDGEMENTS
The authors are indebted to John van Borsel, PhD, Speech and Hearing Rehabilitation Center,
University of Ghent, for co-supervising the second author's MA dissertation (Van
Maldeghem, 1999; Van Maldeghem et al, 2001).
APPENDIX
Description of the variables used in the analysis of spontaneous language samples (Monrad-
Kxohn, 1947; Benson, 1967; Kerschensteiner et al., 1972; Benson and Ardila, 1996).
Anaphors
A word that has an antecedent occurring before or after the word to which it refers. Words
that can be used as anaphors are definite articles, relative pronouns, and possessive pronouns.
Aphasics often use anaphors for which there is no referent (indefinite anaphors). For instance,
a patient may say "I saw it" but the key referent ('book', 'car', 'accident', etc) has not been
mentioned; in this example, 'it' is an indefinite anaphor.
Articulation
The attributes of articulation and phonation which are necessary to produce readily intelligible
word speech patterns.
Circumlocution
A substitution of object description or instrumental function for a word.
Cliche
An accepted and stereotyped utterance which is little specific and carries different meanings,
and can thus be used in different situations. Cliches are used as fillers, and as such they are
redundant and empty.
Conduite d 'approche
A patient's repeated and conscious attempts to correct an erroneous utterance and to
pronounce a word correctly.
Dysssyntaxis
Grammatical deviation characterized by a verbal output that violates the normative rales of
morphosyntactic convention. Dyssyntaxis may result from: (a) erroneous selection of
grammatical elements; (b) overuse of grammatical elements (particularly connectors)
associated with a decrease of nouns; and (c) lack of defining limits in the sentences, correlated
with an excessive verbal output.
Effort
Two patterns are recognized, one with marked difficulty in starting to speak, and a second in
which the patient readily begins a word or a short phrase but becomes tied up and labored in
attempting to continue, hi the former group, the evidence of effort is often dramatic with
facial grimacing, puffing of the chest, and body movements.
Empty speech
A fluent verbal output that conveys very little explicit information because of the lack of
information-loaded words as opposed to the use of a number of relational words, vague and
generic adjectives and adverbs.
False start
Evidence of metalinguistic awareness, characterized by a patient's interruptions, alterations or
adjustments of his/her production in order to re-edit it.
Hedging
An expression of uncertainty (derived from the verb 'to hedge') such as: "I think", "I guess",
"according to me", etc. The difference between a 'hedge' and a 'cliche' is that the latter does
not necessarily carry a connotation of uncertainty.
Literal (or phonological) paraphasia

An incorrectly produced word caused by omissions, additions, displacements, or substitutions
of phonemes within the desired word (e.g., 'nitroben' for "nitrogen", 'porlity' for "policy").
Neologism
A meaningless phonological form resulting from the combination of appropriate and
reproducible phonemes carrying no meaning in the speaker's vocabulary. The listener is not
able to identify the target word.
Passe-partout word
Vague or generic, all-purpose word.
Pauses
The hesitations occurring within a phrase; pauses are not the intervals in the flow of speech
which interrupt phrases. Pauses may be due to: great effort to produce words resulting in breaks
preceding and following phonation; articulation difficulties as the patient struggles to pronounce
or goes back to correct improperly produced syllables; paraphasic substitutions as the patient
attempts to correct or decides to continue despite the error; word-finding difficulties as the
patient finds he is unable to produce the word necessary to continue.
Perseveration
The needless repetition of a phoneme, syllable, word or even a short phrase.
Prosody
Correct placing of stress upon syllables and words within the sentence (including
prolongations); natural rhythm, pauses and rate of speaking; natural shifting of pitch from
syllable to syllable and from word to word, some being pronounced on a higher note, some on
a lower note, varying from sentence to sentence (gradual or abrupt rising and falling of the
pitch).
Rate of speaking
The number of words uttered in one minute (wpm); the normal rate being more than 90 wpm.
Verbal paraphasia
The erroneous use of a word belonging to an inventory of the language in place of another word
that also belongs to one of the language inventories. Morphemic verbal paraphasias refer to
inadequate words that have been assembled by using morphemes belonging to the language
inventory (e.g., "winterly"). Semantic verbal paraphasias designate aphasic transformations in
which the desired and the substituted words are close in meaning (e.g., fork-knife).
Morphological verbal paraphasias are transformations in which the substituting word and the
substituted word are similar in form but not in meaning (e.g., house-mouse). Unrelated verbal
paraphasias are word substitutions that, in the given context, are neither phonologically nor
semantically related to the word that seems to be required (e.g., "It costs a fortune to fly first
carbuncle").
Word choice
The division into predominantly substantive word use (classic telegraphic speech or one-word
sentences, i.e. the use of nouns but also action verbs and significant modifiers expressing an
entire idea) or predominantly relational word use (empty speech, i.e. the use of many
relational words, adjectives, adverbs carrying very little information because of the lack of
specific substantive words). The use of many relational words, cliches and empty speech has
been shown to be peculiar of fluent aphasia.
REFERENCES
Bates, E. (1999a). Plasticity, localization, and language development. In: The Changing
Nervous System: Neurobehavioral Consequences of Early Brain Disorders (S.H.
Broman and J. M. Fletcher, eds.), pp. 214-253. Oxford University Press, New York.
Bates, E. (1999b). Language and the infant brain. JComm Disord, 32, 195-205.
Bates, E., J. Reilly, B. Wulfeck, N. Dronkers, M. Opie, J. Fenson, S. Kriz, R. Jeffries, L.
Miller and K. Herbst (2001). Differential effects of unilateral lesions on language
production in children and adults. Brain Lang, 79, 223-265.
Bates, E., D. Thai, D. Trauner, J. Fenson, D. Aram, J. Eisele and R. Nass (1997). From first
words to grammar in children with focal brain injury. Dev Neuropsychol, 13, 275-343.
Beheydt, L. (1983). Kindertaalonderzoek: Een Methodologisch Handboek. Cabay, Louvain-
la-Neuve.
Benson, D. F. (1967). Fluency in aphasia: correlation with radioactive scan localization.
Cortex, 3, 373-394.
Benson, D. F. and A. Ardila (1996). Aphasia: A Clinical Perspective. Oxford University
Press, New York.
Bloom, L. and M. Lahey (1978). Language Development and Language Disorders. Wiley,
New York.
Bol, G. W. (2003). MLU-matching and the production of morphosyntax in Dutch children
with specific language impairment. In: Language Competence across Populations:
Toward a Definition of Specific Language Impairment (Y. Levy and J. Schaeffer,
eds.), pp. 259-271. Lawrence Erlbaum Associates, Mahwah.
Brown, R. (1973). A First Language. Harvard University Press, Cambridge.
Cooper, J.A. and C. R. Flowers (1987). Children with a history of acquired aphasia: residual
language and academic impairments. J Speech Hear Disord, 52, 251-262.
Cranberg, L. D., C. M. Filley, E. J. Hart and M. P. Alexander (1987). Acquired aphasia in
childhood: clinical and CT investigations. Neurology, 37, 1165-1172.
Davis, G. A., Th. M. O'Neil-Pirozzi and M. Coon (1997). Referential cohesion and logical
coherence of narration after right hemisphere stroke. Brain Lang, 56, 183-210.
Deonna, T., C. Peter and A. L. Ziegler (1989). Adult follow-up of the acquired aphasia-
epilepsy syndrome in childhood: report of seven cases. Neuropediatrics, 20, 132-138.
Eisenberg, S.L., T. Fersko and C. Lundgren (2001). The use of MLU for identifying language
impairment in preschool children: a review. Am J Speech Lang Pathol, 10, 323-342.
Feyereisen, P., A. Pillon and M. P. De Partz (1991). On the measures of fluency in the
assessment of spontaneous speech production by aphasic subjects. Aphasiology, 5, 1-
21.
Gilles, S. and A. M. Schaerlaekens (2000). Kindertaalverwerving: Een Handboek voor het
Nederlands. Nijhoff, Groningen.
Hecaen, H. (1983). Acquired aphasia in children: revisited. Neuropsychologia, 21, 581-587.
Kerschensteiner, M., K. Poeck and E. Brunner (1972). The fluency-non fluency dimension in
the classification of aphasic speech. Cortex, 8, 233-247.
Klein, S.K., D. Masur, K. Farber, S. Shinnar and I. Rapin (1992). Fluent aphasia in children:
definition and natural history. J Child Neurol, 7, 50-59.
course and prognosis. Neurology, 30, 524-529.
Martins, I. P. and J. Ferro (1992). Recovery of acquired aphasia in children. Aphasiology, 6,
431-438.
Moerman-Coetsier, L. and F. Van Besien, (1987). Toast - TaalOnderzoek via Analyse van
Spontane Taal. Acco, Amersfoort/Leuven.
Monrad-Krohn, G. H. (1947). Dysprosody or altered melody of language. Brain, 70, 405-415.
Nass, R. (1997). Language development in children with congenital strokes. Semin Pediatr
Neurol, 4, 109-116.
Paquier, P. and H. R. Van Dongen (1991). Two contrasting cases of fluent aphasia in
children. Aphasiology, 5, 235-245.
Paquier, P. and H. R. Van Dongen (1996). Review of research on the clinical presentation of
acquired childhood aphasia. Acta Neurol Scand, 93, 428-436.
Paquier, P. and H. R. Van Dongen (1998). Is acquired childhood aphasia atypical? In:
Aphasia in Atypical Populations (P. Coppens, Y. Lebrun and A. Basso, eds.), pp. 67-
115). Lawrence Erlbaum Associates, Mahwah.
Pitchford, N. J. (2000). Spoken language correlates of reading impairments acquired in
childhood. Brain Lang, 72, 129-149.
Pitchford, N. J., E. Funnell, A. W. Ellis, S. H. Green and S. Chapman (1997). Recovery of
spoken language processing in a 6-year-old child following a left hemisphere stroke: a
longitudinal study. Aphasiology, 11, 83-102.
Reilly, J. S., E. A. Bates and V. A. Marchman (1998). Narrative discourse in children with
early focal brain injury. Brain Lang, 61, 335-375.
Van Dongen, H. R., J. Meulstee, M. Blauw-Van Mourik and F. Van Harskamp (1989).
Landau-Kleffner syndrome: a case study with a 14-year follow-up. Eur Neurol, 29,
109-114.
Van Dongen, H. R., P. Paquier, W. L. Creten, J. Van Borsel and C. Catsman-Berrevoets
(2001). Clinical evaluation of conversational speech fluency in the acute phase of
acquired childhood aphasia: does a fluency/nonfluency dichotomy exist? J Child
Neurol, 16, 345-351.
Van Hout, A. (1990). Acquired Aphasia in Childhood. Its Impact on the Conception of
Functional Maturation of the Brain and its Implication for Pediatric
Neuropsychology. Unpublished doctoral thesis. Catholic University of Louvain
(UCL): School of Medicine.
Van Hout, A., P. Evrard and G. Lyon. (1985). On the positive semiology of acquired aphasia
in children. Dev Med Child Neurol, 27, 231-241.
Van Maideghem, V. (1999). Spontane Taalanalyse bij Verworven Kinderafasie na Resectie

van een Temporo-Parietale Tumor: een Gevalsstudie met een 12 Jaar Follow-up.
Unpublished master's dissertation. University of Ghent: School of Medicine.
Van Maldeghem, V., P. Paquier, H. R. Van Dongen and J. Van Borsel (2001). Analyse van de
spontane taal bij verworven kinderafasie: een gevalsstudie met een 12 jaar follow-up.
Stem-, Spraak- en Taalpathologie, 10, 49-70.
Vermeulen, J., R. Bastiaanse and B. Van Wageningen (1989). Spontaneous speech in aphasia:
a correlational study. Brain Lang, 36, 252-274.
Wagenaar, E., C. Snow and R. Prins. (1975). Spontaneous speech of aphasic patients: a
psycholinguistic analysis. Brain Lang, 2, 281-303.
Watamori, T. S., S. Sasanuma and S. Ueda. (1990). Recovery and plasticity in child-onset
aphasics: ultimate outcome at adulthood. Aphasiology, 4, 9-30.
Woods, B. T. and S. Carey (1979). Language deficits after apparent clinical recovery from
childhood aphasia. Ann Neurol, 6, 405-409.
Woods, B.T. and H. L. Teuber (1978). Changing patterns of childhood aphasia. Ann Neurol,
3, 273-280.
Long-term language recovery in an aphasic Czech child 199
11
RECOVERY FROM APHASIA AFTER

POLYTRAUMA IN A CZECH CHILD: WHAT I S
LOST AND WHAT IS LEFT
Helena Leheckovd
University of Helsinki, Finland
Abstract—A case study is presented of a Czech-speaking child with acquired aphasia. The patient (a
right-handed female) was 11 years old when she sustained multiple severe injuries in a car accident.
Long-lasting intensive care began when she was in a state of traumatic shock accompanied by multiple
organic failures. There was an extensive craniocerebral contusion. Most of the lesions were situated in
both frontal regions, with left prevalence and a bilateral subdural haematoma. In the beginning, the
patient was in a deep comatose state. When she showed signs of regaining consciousness, many
neurological deficits manifested: hemiparesis on the right side, cerebellar symptomatology, severe
psychosyndrome and global aphasia. The peripheral traumatic lesions were compensated for after 3-4
months, which was followed by a slow amelioration of cerebral functions. The aim of this study is to
describe the aphasic symptoms in the child during her re-acquisition of language and to compare them
with the aphasic symptoms in Czech adults.
The conclusion is that the aphasic symptoms of the child are identical to those of comparable cases of
adult aphasia and to aphasic symptoms in Czech in general. What is lost, 3'A years after the accident,
is fluent reading and writing, smooth access to the whole lexicon and sensitivity to some grammatical
features, especially those with a formal character only. What is re-achieved after an almost hopeless
state is the ability to use language for communication and reflection. While two years post-onset the
comparable adult patient reached a stage after which her language skills no longer improved, this child
continued to develop her linguistic abilities.
Key words: child aphasia, Czech aphasic symptoms, long-term recovery

CASE DESCRIPTION
Aphasia is always manifested in a certain impairment of the morphosyntactic structure. All

aphasic patients, independently of the type of impairment, make mistakes in grammar.
Generally speaking, grammatical problems in language production are mainly reflected in:
(1) omission of grammatical morphemes; and
(2) substitution of incorrect grammatical morphemes for correct ones.
In inflecting languages only free grammatical morphemes are omitted. Bound morphemes are
substituted because word stems cannot stand as separate items.
Czech is a strongly inflecting language with a great number of different grammatical
forms for each inflected word. Grammatical relations in Czech are mostly expressed by bound
morphemes. The typical structure of inflected words consists of a stem and an ending, neither
of which exist as separate words.
In Leheckova (2001) I have described the typical manifestations of agrammatism in Czech
adults:
(a) omission of some grammatical words (mainly prepositions, auxiliaries and pronouns);
(b) concentration of errors in some grammatical categories (especially case, tense and person);
(c) overuse of certain forms of grammatical categories (masculine gender, singular number,
nominative case, 3rd person, present tense, indicative mood, active voice and imperfective
aspect).
In this paper I present a detailed description of a longitudinal study of a Czech child with
acquired aphasia. The patterns of aphasia in childhood have been believed not to be exactly
the same as those in adulthood. The aim of this paper is to describe the aphasic symptoms in a
child during her re-acquisition of language and to compare them with aphasic symptoms in
adults. The results may shed some light not only on child aphasia but also on the functioning
of language in general.
Diagnosis
The patient (born 1988, right-handed, female) was 11 years old when she sustained multiple
severe injuries in a car accident. The long-lasting intensive care began during the severe
traumatic shock state and was accompanied by multiple organic failures. First, it was
necessary to compensate for the exudative pulmonary contusion with traumatic hepatic and
lienal lesions and to stabilize multiple osseous fractures: clavicular, scapular and pelvic. In
addition, the patient had an extensive craniocerebral contusion. The largest of the lesions was
situated in both frontal regions, with prevalence on the left side and a bilateral subdural
haematoma. During her prolonged intensive care, an infectious endocarditis appeared,
requiring protracted treatment with different antibiotics. A hemorrhagic gastrointestinal ulcer
also developed. Both subdural haematomas were treated by repeated surgical evacuation.
Long-term language recovery in an aphasia Czech child 201
Three months after she sustained the injury, computer tomography showed post-contusion
and post-ischemic changes in her left frontal, temporal and parietal lobes, significant
supratentorial brain atrophy with a dilated ventricular system -the 3rd ventricule was 10 mm
wide. Subdural collection occurred on both sides with the character of a chronical haematoma
that was 11 mm thick. For three months after the accident, the patient remained in a deep
coma which was followed by an apallic state. The peripheral traumatic lesions were
compensated for after 3-4 months, followed by a slow amelioration of her cerebral functions.
When the child exhibited signs of regaining consciousness, about four months post-onset,
more local neurological deficits manifested. Central spastic quadriparesis changed into a
hemiparesis on the right side, and the patient was diagnosed with cerebellar symptomatology
and severe psychosyndrome. The initial global aphasia developed into a mixed aphasia with a
predominant expressive component (Broca's aphasia). The clinical neurological amelioration
was accompanied by the amelioration of diffuse EEG abnormalities and improved evoked
potentials.
After six months of intensive care, the patient could be transferred to the Institute of
Rehabilitation for physiotherapy and logopaedic training. Two months of these procedures
markedly ameliorated the symptoms.
Computerized tomography conducted six months post-onset revealed that the subdural
collection had receded on both sides. Thus, only smaller residual changes could be observed
frontally on both sides. Post-contusion and post-ischemic changes remained in the frontal,
temporal and parietal regions. The signs of supratentorial brain atrophy, located mainly in the
left hemisphere, had not changed. The neurological report stated that the patient had a central
right-sided hemiparesis, tremor and apraxia. The patient was cooperative, though motorically
and mentally apparently slow.
Rehabilitation1
Six months after her accident, the patient was admitted to the Institute of Rehabilitation. She
was in a wheelchair, accompanied by her mother. She was insecure and frightened in her new
environment. Her articulatory organs showed no pathology, her sight and hearing were
normal. Her voice was high and clear, and she breathed calmly. She had difficulties
coordinating her articulatory movements.
The patient communicated only by gestures and mimics in answers to yes-no-questions.
She understood basic instructions and tended to show objects and pictures correctly even
though her achievement was restricted by her apraxia. On the other hand, she was confused
by instructions containing more complicated grammatical structures (constructions containing
prepositions, prefixes, etc.):
1
In this section I shall present an overview of medical reports made by neurologists, phoniatrists and speech
therapists during the three years after the accident.
(1) Dej papir na knihu. 'Put the paper on the book.'

Dejpapir do knihy. 'Put the paper in the book.'
Dej papir pfed knihu. 'Put the paper in front of the book.'
Vyndej papir z knihy. 'Take the paper out of the book.'
(2) Lezi kniha na stole? 'Does the book lie on the table?'
Nelezi kniha pod stolem ? 'Doesn't the book lie under the table?'
Nelezi kniha vedle stolu? 'Doesn't the book lie beside the table?'
The patient had problems in establishing logico-grammatical relations such as the opposition
between 'first - last', 'big - small', and 'biggest - smallest'. She made mistakes in recognizing
colours. Furthermore, she was unable to repeat and used the perseverance mama instead. She
could not name objects.
She read globally, put inscriptions to pictures but used only short simple words. She made
mistakes eventually and corrected herself occasionally. She was unable to read letters or
nonsense words. She made mistakes in recognizing numbers and she was more certain of the
100-1000 scale than in smaller numbers. She had apraxia.
She began speech therapy by doing phonic exercises during which she produced all
vowels (la/, /i/, /u/\ with some difficulties lo/, /e/). Later, they were combined with consonants
in syllables (pa - ba - ma, pi - bi - mi, etc.), reduplicated syllables (jo-jo, je-je, pi-pi, etc.) and
simple words (bdba, ndna, pipd, pes, pas). Short sentences were then introduced:
(3) Mimi hajd. 'Baby sleeps.'

Tata void. 'Daddy calls.'
Her repetition gradually improved. The most difficult phonemes proved to be Iv/, /f/, /// and
M.
Six weeks after having started intensive speech therapy the patient could repeat short
sentences, especially when divided into syllables. She named spontaneously about 20 familiar
pictures. She recognized some letters in the alphabet but was uncertain about them. She could
count to 10. Speech production improved, although it was not fluent. She provided mostly
one-word answers to questions and rarely used a simple sentence. Her speech was very slow,
her intonation flat and she expressed herself with no emotional charge. She had hypomimia.
Her understanding of grammatical structures, e.g. (1) or (2), improved but was far from
perfect.
At one year and three months post-onset, all symptoms of Gerstmann's syndrome were
still present (i.e. amnestic aphasia, alexia, agraphia, acalculia, agnosia of fingers and problems
with left — right orientation). Whereas spontaneous drawing was impossible for her, copying
was preserved. Her spontaneous speech was slow and laborious with apparent word-finding
difficulties. Understanding of speech was relatively good as far as global information was
concerned. However, isolated information (without context) caused problems.
At one and a half years post-onset Gerstmann's syndrome was still observable but had
improved. She had much better left - right orientation and finger gnosia with only a slight
uncertainty. Her drawing had improved substantially, both in copying and as spontaneous
activity. Her spontaneous speech was still marked by word-finding difficulties, but to a lesser
degree. Acalculia and alexia persisted.
Three years after the accident the character of aphasia is reported to have changed both
qualitatively and quantitatively: from global through Broca's to amnestic aphasia. Finger
agnosia had disappeared. Her drawing further developed both as to form and content.
Moreover, her spontaneous speech improved considerably. Latencies caused by word-finding
difficulties almost disappeared. The patient needed a longer time for reading, remembering
and reproducing a text. Understanding of logico-grammatical structures still caused problems
but they were smaller than before.
School attendance
Before the injury, the patient was an excellent pupil in the 6th form of a primary school. She
resumed school attendance one year after the accident, although she still could neither read
nor write. She had hemiparesis on the right-hand side and her psychomotor performance was
slower. However, her intellectual capacity (in the sense of finding connections and solutions)
was reported to be undisturbed. As a result, it was recommended that she not be transferred to
a special school but be kept in her previous class, even though she could not fully participate
in the schoolwork.
Three and a half years post-onset, the patient is attending the last form of compulsory
primary school. She is not marked and follows an individual plan. Her family faces the
serious problem of her further education.
APHASIC SYMPTOMS 2 IN SPONTANEOUS SPEECH
Six months post-onset
The patient produced no spontaneous speech except for perseverance on two words: mama
('Mom') and ne ('no'). After six weeks of intensive therapy, she used simple sentences such as:
(4) Jajsem DN. 'I am DN.'

Jdjedu na hole. 'I ride a bike.'
Jd mam pannu. 'I have a doll.'
Her spontaneous speech was not fluent.
Two and a half years post-onset
1 examined the patient two and a half years after the polytrauma. While organic and motor
problems had mostly disappeared by that time, the patient's linguistic functions were not fully
restored. I recorded the spontaneous speech of the patient during a long visit at home. I asked
her about her illness, school, friends and hobbies. I also asked her to retell the Red Riding
Hood story and the content of a film she had seen.
The patient cooperated eagerly and spoke fluently. Subjectively she did not feel she had
any problems in understanding or expressing herself:
(5) Dfivjsemjako nevedela, co mam fikat, a co patri k sobe, ale ted' toje normalni.
'Earlier I did not know what to say and what belongs together but now it is normal.'
It is only the linguistic analysis that revealed an impairment of language. The material
consists of about one hour's recording. There were 3256 words in 543 clauses, 14% of which
were ungrammatical. Of all errors, 80% can be classified as omissions and 20% as
substitutions. It is not always easy to draw the line between omissions and substitutions, as
the same error can be classified as both (for explanation, cf. Leheckova 2001). The criterion
that I have applied was the system of Czech language (i.e. in nouns with a zero ending in
nominative, I classified a missing ending in another case as substitution by nominative, not an
omission of a case ending, because it was analogical to errors in nouns that do have an ending
in nominative).
The speech of the patient was rather fluent, with normal melody and intonation, without
articulatory difficulties. As long as the conversation was confined to small talk with a quick
exchange of questions and answers, no abnormalities were observable. Furthermore, the
patient understood all questions and re-used many words from the question in her answer:
(6) Hralajsi nejakou hru?

'Did you play any game?'
Hrdla. Basket a ... tenis.

'I played. Basketball and ... tennis.'
(7) Jakou hudbu rdda poslouchds?

'Which music do you like to listen to?'
2
In what follows I shall present my own results obtained from testing the patient.
No takposlouchdm rdda takovou moderni...

'Well, I like to listen to a kind of modern ...'
However, longer answers revealed substantial word-finding difficulties and the use of fillers
and empty words.
(8) To mi kamarddka nahrdla, to kamardd nekde sehnal, takze to ... takze tyhlety takhle.
'My friend recorded it, my friend found it somewhere, thus that... thus these so.'
(9) To ted' vsechno mdme asi moderni... to takze takovyhle... jd nevim tfeba ndky
skupiny... tfeba ndky kluci...
'Thus now we have everything modern ... so thus such ... I don't know let's say some
groups... let's say some boys ...'
(10) Vsakse fikd, ze muzikajako leci, takjd to to...

'They say that music heals, so I that that...'
The patient encountered difficulties with questions that demanded more complicated
formulations and words that were not originally in the question. The speech rate decreased,
the syntactic structure of the utterances was disrupted and content words were missing:
(11) Ctes si knizky?

'Do you read books?'
Jo, to si ctu. Jdjsem ted' zacala, ted' sem dfiv necetla, ndky holky mi tfeba pucily vo
ndkejch... jd nevim, tarn bylo tfeba... tarn bylo vo ndky ... z knihovny jsem tfeba si
pucovala, tak tarn je ndky pro nds uz tak ... sem si tfeba cetla .. no ale zatim... to je
hrozny.
'Yes, I do read. I have started now, now I have not read earlier, some girls lent me for
example about some... I don't know, there was for example ... there was about some ...
in the library I borrowed, so there is some for us already so ... I read for example ...
well but still... that is terrible.'
When the subject of the question was unexpected, the patient reacted with semantically
almost empty utterances. Interestingly enough, the syntactic structure was preserved but
content words were missing:
(12) Mohla bys mi vypravovat Cervenou Karkulku?

'Could you tell me Red Riding Hood?'
No jo, Cervend Karkulka, to jo, ale vyprdvet, to rnoc teda... to nevim, co tarn bylo a co
tarn nebylo ... tfeba von si dycky ...no ... co fikd ...ale tfeba co si vymyslim ...moznd
bysem si vzpomnela, ale ted' vubec takovyhle pohddky nebo tak...to ted' vubec...
'Oh yes, Red Riding Hood, oh yes, but to tell, well that ... I don't know, what was
there and what wasn't there ... for example he always ... well ... what he says ... but
maybe what I make out... maybe I would remember, but now such stories or so ... well
now completely ...'
When the question was more personal and familiar, the patient did much better:
(13) Vzpomnela by sis vubec na nejakou pohddku nebo pfibeh, kterej se ti libil? Melas rdda
pohddky, kdyzjsi by la mala?
'Could you remember any story that you liked? Did you like fairy tales when you were
little?'
Jo, tak jo. Jd nevim, deda mi vzdycky rikal nejlepsi. To von mel jako pro sebe, o
ndkym drakovi, ze byli ctyri nohy, teda ... jd nevim, jak mne to deda rikal... asi tak by la
asi takovd nejlepsipohddka, co deda vzdycky...
'Well, yes. I don't know, Grandfather always told the best ones. He had it out of
himself, about a dragon, that there were four legs, actually ... I don't know how
grandfather told it... about that was about the best story, what grandfather always ...'
Table 1 — Omissions in spontaneous speech.
Omitted item Number of Percent of all

errors errors
Content words Finite verb in pres 17 28%
Finite verb in past 11 18%
Infinitive 3 5%
Past participle 5 8%
Predicative adjective 8 13%
Noun after preposition 2 3%
Adverbial 3 5%
Content / Pro/noun in object 9 15%

Grammatical words
Grammatical Preposition 2 3%
words Modal verb before inf. 1 2%
Total (omitted items) Number of Percent of all

errors errors
Content words 49 80%
Content/grammatical words 9 15%
Grammatical words 3 5%
Omissions. Relatively few grammatical errors were found in the material. The patient even
used grammatical items that are most difficult for children, i.e. comparative/superlative, or the
conditional. The syntactic structure was mostly preserved, even though some arguments were
missing.
As expected, only free morphemes were omitted. Surprisingly, the omissions affected
mostly content words (49 errors, i.e. 80% of omissions), not grammatical words. There were
only 3 examples of missing purely grammatical items, while in 9 cases, the missing word
could be a pronoun as well as a noun.
Substitutions. Errors caused by substitution were rarer than those caused by omission. Few
lexical substitutions occurred:
(14) pisnicka ('song') instead of pismenko ('letter')
(15) drak ma styri nohy ('dragon has four legs') instead of hlavy ('heads')
(16) tak dve nebo tak ('about two or so') instead oinekolik ('several')
Table 2 — Substitutions in spontaneous speech
Category Replacing item Replaced item IN[umber of errors

Gender MASC FEM 1
FEM NEUTR 1
Number SG PL 3
PL SG 1
Case NOM ACC 2
NOM INSTR 1
NOM GEN 1
GEN INSTR 1
Other PREP PREP 2
Total (uncorrected errors) 13

Self-corrected errors:
Category Replacing item Replaced item Number of

cases
Number PL SG 1
Total (self-corrected errors) 1
Grammatical substitutions affected mainly bound morphemes. There were only two
substitutions of free morphemes (prepositions), otherwise one grammatical ending was
replaced by another ending belonging to the same category. The number of substitutions is
too small to draw any definitive conclusion. Nevertheless, the replacing forms are mostly
those that I have defined as best preserved in Czech adult aphasics: i.e. masculine gender,
singular number, nominative case, 3 rd person, present tense, indicative mood, active voice and
the imperfective aspect (for the hierarchy of replacing forms, cf. Leheckova, 2001).
Neologisms. There were 10 neologisms in the recording: 4 were nonsense-words both

lexically and structurally, while 6 were non-existing words with correct grammatical endings:
(17) tak me tak docela pobuzdili (instead of povzbudili)

'thus they rather supported me1 - 3rd PERS PL PAST
(18) dycky me tak ndk pozdrej (instead ofpodrzej)

'they always somehow keep me' - 3rd PERS PL PRES
(19) voni sou ted' v kutdku (instead of e.g. v prvdku)

'they are now in (e.g. first form)' - LOC SG
(20) ale snazilajsem se to vsechno narazit (instead of e.g. nahradit)

'I tried to replace that all' - INF
(21) nejtezsijsou tydlety prirodnice (instead of e.g. prdce)

'the most difficult are these subjects' - NOM PL FEM
The patient overused some words expressing uncertainty (jd nevim 'I don't know', tfeba 'for
example', moznd 'maybe', etc.).
Three and a half years post-onset
A year after the first recording, I again recorded the patient's responses to the same questions.
During the second session, she produced 1981 words which were divided into 342 clauses. Of
these, 6% were erroneous. Most of her errors were omissions (62%). As before, she omitted
mainly content words, especially verbs. In the second recording, she only omitted 2
grammatical words. The substitutions again concentrated in the category of case (75%). While
in the previous testing the patient substituted mostly nominative case for other cases, during
the second testing she replaced cases without a clear preference (genitive for dative,
instrumental for genitive, dative for accusative, genitive for locative, nominative for
accusative etc.). There was only one neologism in the material, which was used with the
proper verbal ending.
The grammaticality of the spontaneous speech improved quantitatively while qualitatively
showed roughly the same tendencies as a year ago.
Table 3 — Errors in spontaneous speech at 2 14 and 3 Vi years post-onset
2!4 years post-onset 3/4 years post-onset
Ungrammatical clauses 14% 6%

Errors caused by omission 80% 62%
Errors caused by substitution 20% 38%
Number of neologisms 10 1
GRAMMATICAL TESTS
Two and a half years post-onset
Two and a half years post-onset the spontaneous speech of the patient seemed to be only
mildly impaired. However, grammatical tests revealed a more serious deficit. There were two
grammatical issues that I tested with the patient: prepositions and prefixes. The reason for this
choice is based on my experience with aphasics on the one hand and language learners on the
other. Even patients with mild impairments as well as foreigners speaking Czech have
problems both with prepositions and prefixes.
Prepositions. The use of prepositions in Czech is influenced by two factors: either by their
own (mostly spatial) meaning as in (22), or by the government (mostly of the verb), as in
(23):
(22) Zahradaje za domem. 'The garden is behind the house.1

Knihaje na stole. 'The book is on the table.'
(23) Dekuji ti za dopis. 'Thank you for the letter.'

Odpovez na otdzku. 'Answer the question.'
The patient was administered a test consisting of written sentences in which the preposition
was missing and the noun was given in its nominative form.
(24) Moji rodice bydli (Praha).

'My parents live (Prague).'
Test 1 is comprised of 34 sentences with missing prepositions having their own lexical
meaning. It took the patient 25 minutes to fill in the gaps. There were two wrong prepositions
substituted for the correct ones and one preposition was omitted. Thus, 9% of the sentences
were erroneous.
Test 2 consisted of 33 sentences with missing prepositions having grammatical meaning only.
The test took the patient 25 minutes to accomplish. One preposition was replaced by a wrong
one and three prepositions were omitted. Thus, 12% of sentences were erroneous.
It seems to be easier to omit and even substitute prepositions that have only a grammatical
function. Some of the errors were caused by choosing a wrong, but in another context
possible, government:
(25) Rodice poslali doktora.

'The parents sent the doctor.'
instead of
Rodice poslali pro doktora.
'The parents sent for the doctor.'
However, it would be possible to say:

Rodice poslali telegram.
'The parents sent a wire.'
(26) Vybral si bohatou nabidku.

'He chose a rich offer.'
instead of
Vybral si z bohate nabidky.
'He chose from a rich offer.'
However, it would be possible to say:

Vybral si bohatou nevestu.
'He chose a rich bride.'
The patient did not have many prepositional errors in her spontaneous speech (two
prepositions omitted, two substituted for by another preposition). The grammatical test
revealed a more serious problem in this category, especially with prepositions without lexical
meaning.
Prefixes. Czech verb can be modified by different prefixes. Prefixes have a purely
grammatical function, i.e. they form the perfective aspect from imperfective:
(27) Psal dopis. 'He was writing a letter.'

Napsal dopis. 'He has written a letter.'
Prefixes can also modify the meaning of the verb:
(28) Dopsal dopis . 'He finished writing a letter.'

Rozepsal dopis. 'He started to write a letter.'
Pfepsal dopis. 'He wrote the letter again.'
Podepsal dopis. 'He signed a letter.'
The patient was administered three tests with missing prefixes. Test 3 consisted of 20
sentences with verbs that demanded unmotivated grammatical prefixes:
(29) Je tfeba to poradne ... .pldnovat.

'It must be well planned.'
Je tfeba to pofddne napldnovat.
(30) Tatinek se rdno zapomnel... holit.

'Daddy forgot to shave himself in the morning.'
Tatinek se rdno zapomnel oholit.
The patient completed the test in seven minutes with one error only (omission of prefix).
Test 4 had the same structure but the missing prefixes had their own meaning, that of
direction:
(31) Na pfisti zastdvce musite vystoupit. 'You must get off on the next stop.'
Dobehl do cilejako prvni. 'He ran into the goal as the first.'
The patient usually used the same prefix as the preposition in the sentence. This is often
correct:
(32) Auto zajelo za roh.

'The car disappeared.'
Sometimes the following is possible, though not as a first option:
(33) Dojeli jsme vytahem do pfizemi.

'We descended to the ground floor by elevator.'
instead of
Sjelijsme vytahem do pfizemi.
Sometimes it is wrong:
(34) Dopsaljsem se do sesitu.

'I wrote my name in the notebook.'
instead of
Zapsaljsem se do sesitu.
The patient needed 25 minutes for this test and made 8 mistakes: three omissions and five
substitutions. That means that 40% of the responses were incorrect.
Test 5 consisted of 20 sentences with missing prefixes in a more abstract, mostly idiomatic
meaning:
(35) Maji pofad obsazeno, nemuzu se tarn dovolat.

'The phone is busy, I cannot get through.'
Ktere noviny mate pfedplacene?

'Which newspaper do you subscribe to?'
The patient completed this task in 20 minutes with five errors (two omissions and three
substitutions), thus 25% of the responses were incorrect.
In her spontaneous speech the patient made no errors in prefixes but did not use many
prefixes (only 19). Grammatical tests revealed the uncertainty of the patient as far as prefixes
were concerned. I expected that the biggest problem would be caused by purely grammatical
prefixes that cannot be chosen on the basis of their meaning. However, they were almost
always used correctly. The most difficult proved to be the use of prefixes with directional
meaning. In choosing them, both their meaning and the structure of the sentence must be
taken into account. The use of prefixes in idioms was better because the expressions
concerned are very frequent in everyday use.
Three and a half years post-onset
One year later I administered the grammatical tests again to the patient. She was asked both to
fill in missing grammatical items and to judge the grammaticality of the test sentences.
Compared to the previous testing, she was much quicker and self-assured in her responses.
However, the error rate in the grammatical categories in question was still much higher than
in her spontaneous speech generally, hi contrast to the previous testing, the differences in test
scores diminished.
Table 4 — Errors on grammatical tests
2'A years 3/2 years

post-onset post-onset
Test Grammatical features tested Error Time Error Time
No. rate rate
1 Prepositions with lexical meaning 9% 25 min 21% 8 min
2 Prepositions with grammatical 12% 25 min 27% 12 min
meaning
3 Grammatical prefixes 5% 7 min 11% 4 min
4 Directional / spatial prefixes 40% 25 min 26% 15 min
5 Idiomatic prefixes 25% 20 min 15% 10 min
As the patient was doing well enough and was cooperative, I could test other grammatical
issues as well. I used similar tasks as those proposed in Paradis (1989). The patient had no
problems with determining thematic roles (both for nouns and pronouns) as long as there was
the canonic word order (SVO).
(36) Divka strka chlapce.'

'The girl (NOM) pushes the boy (ACC).'
Divka ho strka.
'The girl (NOM) pushes him (ACC).'
When the word order changed into OVS, which is quite possible in Czech, she assigned
the role of the agent to the first noun, although it was in accusative:
(37) Chlapce strka divka.

'The boy (ACC) pushes the girl (NOM).'
It is the girl who pushes the boy.
The patient also got confused by negation:
(38) Chlapce divka nestrkd.

'The boy (ACC) the girl (NOM) not push.'
The boy is not pushed by the girl.
She could not construct simple sentences from given words in basic forms:
(39) zeleny / list / videt

'green / a leaf/ to see'
Her reading was laborious and slow, she obviously did not understand what she was
reading. When asked about less frequent words, she did not know their meaning. Moreover,
she was unable to find synonyms and antonyms to isolated words but she was doing much
better with words in context or with whole sentences. Another problem she encountered was
that she could not divide the text written without punctuation into sentences.
Recent grammatical tests show that the patient's linguistic performance is far from being
perfect. The main difficulty seems to concentrate in her syntactic (structural) organization of
language, both in production and perception.
WRITING AND DRAWING
The patient suffered from a complete agraphia after the accident. Six months later, she tried to
use her left hand for writing block letters. She found that could not make a cross, a circle or
connect two points. She was also unable to sign her name.
After six weeks of intensive therapy she switched back to the right hand, although her left
hand was still more dexterous. She could now spontaneously produce the following block
letters: D, T, A, E, I, M. She could draw a cross and a circle, which reflected her
improvement.
Two months later, she was able to draw simplified pictures of her house and of a human
being. At this point, she could write her name, but she was not able to write spontaneously
any other word. On the other hand, she could copy writing.
One year and three months post-onset, the patient began to use block letters and simple
pictures spontaneously:
(40)
Long-term language recovery in an aphasic 215
Czech child
(41)
One and a half years post-onset she still could not manage handwriting but her
drawing was improving.
Two years post-onset, the patient could write the names of the weekdays with only
slight errors {uteri instead of utery). In this instance, both forms would be pronounced
in the same way but an orthographic rule dictates that the word should be written
utery. According to this rule letter r cannot be followed by i/i but only by y/y and this
is already mastered by nine-year-old children.
(44) utery 'Tuesday'

stfeda 'Wednesday'
ctvrtek Thursday1
pdtek 'Friday'
sobota 'Saturday'
nedele 'Sunday'
She wrote spontaneously the following sentence:
(45)Cesta z Podebrad do Prahy byla ndrocna.

'The journey from P. to Prague was difficult.'
Pronounced as [cesta s pod'ebrat do prahi bila na:rocna:]
This sentence resembles a phonetic transcription, in that it reflects more or less the
phonetic value of the spoken form but neglects all rules of orthography which had been well
known to the patient before the accident. In fact, she did not even start the sentence with a
capital letter or end it with a period.
Her drawing continued to develop:
(46)
Two years post-onset the patient used a computer to type spontaneously a longer text (94
words) about Christmas. This text is surprisingly consistent and contains no major lexical or
grammatical mistakes. On the other hand, it completely neglects orthographic rules and
reflects only the phonetic representation. It is worth noting that at that time, the patient could
not have delivered a coherent speech of this type orally. This text starts with a capital letter
but contains no punctuation except for one period in an inappropriate place. She uses capitals
for the kinship terms 'Mum, Dad, Grandmother, Grandfather', which is incorrect in Czech, but
she does not begin the sentence with a capital letter. There are many misspellings, usually
reflecting what is pronounced instead of what should be written. Often two words are merged
into one.
The phonetically motivated mistakes were:
(47) gdis instead of kdyz pronounced as [gdis] 'when'

du instead ofjdu pronounced [du] 'I go'
pride instead ofpfijde pronounced [pfi:de] 'it comes'
druhi instead ofdruhy pronounced [druhi:] 'second'
g Dedovi instead of A; dedovi pronounced [gd'edovi] 'to Grandfather'
nejraci instead of nejradsi pronounced [nejraci] 'I like most'
The phonetically unmotivated mistakes were:
(48) chamilie insead offamilie 'family'

kukuju instead ofkupuju 'I buy'; repeated in the next sentence:
koukim instead koupim 'I shall buy'
Vdvdnoce instead of Vdnoce 'Christmas' (reduplication)
Two years and four months post-onset, the patient made the same errors in listing weekdays
as four months earlier.
Two years and seven months post-onset, the patient handwrote a text about her summer
holiday (111 words):
(50)
This text is divided into sentences, although punctuation is often incorrect. For example,
commas are mostly missing, periods are used in illogical places and are not followed by
capital letters, and conjunctions do not correspond to the logical structure of the text:
(51) Kdyz jsme pfijeli takjsme nevefili vlastnim ocim ale potom jsme to pfezili. ale koupili
jsme si dve deky
'When we came we did not believe our eyes but then we coped, but we bought two
blankets'
Her lexicon and grammar do not exhibit major mistakes. Her orthography has improved
dramatically, as there is only one misspelling:
(52) podkala instead of potkala pronounced [potkala] 'I met'
Three years post-onset, the patient's handwriting improved slightly but she still
occasionally forgot letters. The patient preferred to write on a computer. She wrote
spontaneously an introspective text of 488 words. This text is divided into long sentences
starting with a capital letter and ending with a period. However, clauses within sentences are
not divided by obligatory punctuation so that 60 commas or periods are missing. Otherwise
the patient made only four grammatical mistakes:
— one missing relative pronoun:
(53) a ptaljak mu to slusi

'and he asked how he looked'
instead of
a ptal se, jak mu to slusi
— one missing subordinate conjunction 'that':
(54) byla st'astna jsme si psali kazdy den

'she was happy that we wrote every day'
instead of
byla st'astna, zejsme si psali kazdy den
— one missing object of a transitive verb:
(55) ale nebudu psatprotoze si mislim

'but I shall not write it because I think'
instead of
ale nebudu to psdt, protoze si myslim
— one substituted preposition:
(56) s autobusu instead of k autobusu 'to the bus'
There were 21 misspellings, which were mainly the incorrect representations of the
phoneme HI. In Czech I'll can be written either as i or y, depending on the morphophonological
context. The patient would certainly not have made similar mistakes before the accident as
she had been taught the correct orthography at school by the age of 11 and she had been an
excellent pupil.
Examples of the letter i used instead oiy are:
(57) druchi instead of druhy 'another'

pizamo instead oipyzamo 'pyjamas'
holky zarlili instead oiholky zdrlily 'girls were jealous'
mislim instead otmyslim 'I think' (2 errors)
The use of the letter >> instead of/:

(5 8) lavycce instead of lavicce 'a bench'
libym instead of libim 'I like'
stavyl instead oistavil 'he stopped'
vydet instead of videt 'to see' (3 errors)
nevym instead of nevim 'I don't know1
The letter ch (in Czech ch is taken for one letter representing the voiceless /h/) instead ofh:
(59) druchi instead of druhy 'another'
pomdchat instead ofpomdhat 'to help'
dloucho instead of dlouho 'long'
The letter ch instead of/?

(60) douchdm instead of doufdm 'I hope'
The letter z instead off:

(61) zikat instead offikat 'to say' (6 errors)
Her drawing again improved:

(62)
The patient's spontaneous writing seemed to adjust to spontaneous speech, with regard to
both the orthography of single words and the text structure. She has made a tremendous
development in re-achieving her writing skills but text subdivision into structured
concatenation of clauses and correct spelling remained out of reach. When I asked the patient
whether she remembered the orthographic rules that she had learned at school, she answered
affirmatively. Nevertheless, she completely neglects these in her writing, even when she is
copying texts (cf. zeleny - zeleni 'green', videt - vydet 'to see')
(63)
zeleny / list / videt
and especially in spontaneous writing (commas missing, an impossible combination of f+y):
(64)
Jirka zjistil, ze ho boli zuby, tak se rozhodl, ze pujde k lekafi.

'Jirka found out that he has tooth-ache, thus he decided to go to the dentist.'
COMPARISON WITH AN ADULT CASE
In what follows I shall compare the language production of the child with that of an adult who
was, just like the child, classified two and a half years post-onset as having amnestic aphasia
with about the same severity of language impairment, i.e. she was able to accomplish the
same linguistic tasks with roughly the same error rate. My aim is to capture the type of their
errors3.
The patient (born 1945, female, right-handed, computer maintenance person) suffered a
haemorrhage in the left parietal region with a right-sided hemiparesis and mixed aphasia in
February 1997. The haematoma was aspirated. Angiography revealed a residual arteriovenous
malformation (AVM), which was irradiated by a Gamma Knife in April 1998. Eight months
3
As a linguist based in Finland, I am not in permanent touch with Czech aphasic patients. Their choice and
diagnosis (including the determination of the type of aphasia) are made by speech pathologists and neurologists
with whom I have been co-operating. I concentrate on the linguistic analysis of the material.
post-onset, the patient began speech therapy. An amelioration of the hemiparesis and aphasia
was observed during the rehabilitation.
I examined the patient two and a half years post-onset. She was diagnosed as having
residual hemiparesis on the right side, agraphia, alexia and amnestic aphasia. Her sight and
hearing, as well as orientation as to time and space, were normal. The patient still could not
write with her right hand. She used her left hand to compensate, but only for a short time
before she got tired. She had difficulties with reading, especially long words. She had special
problems with letters v and /, and she could not differentiate letters a and e. Her spontaneous
speech was relatively fluent, though slow. The impairment of grammatical structure was mild.
The patient often noticed that she had made a mistake and tried desperately to correct herself.
When she did not succeed, she expressed her dissatisfaction:
(65) Mohl bych vets prosit ... laskavost? Ale neco tarn chybi, ne? Jajsempak
takovd nest'astnd.
'Could I ask you for (omitted PREP) something? But something is missing there, isn't
it?
I am so unhappy then.'
The patient had occasional word-finding difficulties, especially with members of the
following automated sequences:
(66) bylo to pondeli, utery, ne, stfeda

'it was Monday, Tuesday, no, Wednesday'
(67) chci deset, patndct, ne, dvacet deka

'I want ten, fifteen, no, twenty decagrammes'
Spontaneous speech
I met the patient in one session, and our conversation was recorded. I inquired about her
illness, her work and her hobbies. The analysed material consists of 824 clauses containing
4126 words. The average number of words per clause (5) is rather high, because repetitions
were counted as separate words.
Of the total 824 clauses, 89 were erroneous, i.e. 11% of clauses were ungrammatical.
Among the errors, 71% were classified as omissions and 29% were substitutions.
Omissions. It is interesting to note that in both cases, most of the omissions are such that
the omitted element is a content word, not a grammatical one. Furthermore, most of the
omissions affect finite verbs (46% for the child and 45% for the adult). Curiously enough, no
auxiliaries were missing, while infinitives and participles after auxiliaries were omitted in
several cases.
(68) no, takjsem ..., kdyz jsem nemohla vubec ...-, takjsemji ukdzala tohleto
'well, so I have ..., when I couldn't at all..., so I showed it to her'
There were as many cases of omitted prepositions before nouns as omitted nouns after
prepositions:
(69) takjsem furt chodila do ..., kdyz topfislo

'so I used to go to ... (omitted NOUN) when it occured'
(70) kdyz ...- ty operace seprobudim

'when ... (omitted PREP) the operation I wake up'
Table 5 — Transcribed material of the child and the adult (2V2 years post-onset)
The child The adult

Words 3256 4126
Clauses 543 824
Ungrammatical clauses 76 14% 94 11%
Errors: Omissions 61 80% 67 71%

Substitutions 15 20% 27 29%
Table 6 — Comparison of omissions between the child and the adult patients
Omitted items The child The adult

Number ]Percent of all Number of Percent of all
of errors errors errors errors
Finite verb in PRES 17 28% 22 33%
Finite verb in PAST 11 18% 7 10%
Finite verb in FUT 1 1%
Infinitive 3 5% 9 13%
Past participle 5 8% 7 10%
Predicative adjective 8 13% 6 9%
Noun after PREPOSITION 2 3% 3 5%
Adverbial 3 5%
Pro/noun in OBJECT 9 15% 6 9%
Pro/noun in SUBJECT 3 5%
Preposition 2 3% 3 5%
Modal verb before INFINITIVE 1 2%
Total (omitted items) The child Th e adult
Number of Percent of Number Percent of all

errors all errors of errors errors
Content words 49 80% 55 82%
Content/Grammatical words 9 15% 9 13%
Grammatical words 3 5% 3 5%
Substitutions. Interestingly enough, the self-corrected errors do not correspond to the

substitution hierarchy presented in Leheckova (2001), whereas the uncorrected errors do
follow the suggested hierarchical pattern. The majority of the substitution errors were existing
forms of the same category, and as to the order of substitution, they showed the same
tendency as described in Leheckova (2001). The number of errors was nevertheless too small
to make generalizations concerning the distribution among different forms.
Table 7 — Comparison of substitutions between the child and the adult patients
The child The adult
Category Replacing item Replaced item Number of Number of

errors errors
Gender MASC FEM 1 2
FEM NEUTR 1
Number SG PL 3 2
PL SG 1
Case NOM ACC 2 2
NOM INSTR 1
NOM GEN 1 2
NOM LOC 4
GEN INSTR 1
Person 3SG 1SG 2
1SG 2SG 1
Tense PRES PAST 4
Moorf IND COND 1
Other
PREP PREP 2
INDEF. PRON. FINITE VERB 4
Total 13 24
(uncorrected errors)
Self-corrected errors:
The child The adult

Category Replacing item Replaced item Number of Number of
errors errors
Gender NEUTR MASC 1
Number PL SG 1
Person 2SG 1SG 1
Other INF PAST PART 1
Total (self-corrected errors) 1 3
Neologisms. The adult patient did not use neologisms. When she could not find the correct
word, she used indefinite pronouns or adverbs instead (which sometimes occurs even in
normal spontaneous speech):
(71) tak to me tak tohleto

'so it me so like this'
(72) no tak tohleto tak, todleto tadyhle

'well so this so, this here'
The adult patient seemed to have greater word-finding difficulties than the child, although
the latter used neologisms that appeared to be correct grammatically in analogical situations.
Grammatical tests
The most significant grammatical problem in the spontaneous speech of the adult patient
seems to be the use of a preposition with a noun phrase. The adult patient omitted either a
preposition or a noun, or put the noun in a wrong case (nominative singular). The material
contained 68 occurrences of the PREPOSITION + NOUN combination, out of which 46 were
correct and 22 incorrect. This error rate (32%) is much higher than in other categories
(median 11%). It is also noteworthy that during the following sessions, the patient did not
make any preposition errors in grammatical tests that were administered in the written form
(cf. 2.2.). She succeeded in achieving 100% accuracy, although she displayed hesitation.
(73) Vesnice byla za - na - ee -po zemetreseni uplne znicena.

'The village was behind - on - ee- completely destroyed after the earthquake.'
The adult patient made many preposition errors in spontaneous speech but none in
grammatical tests. The child, on the contrary, made few errors in prepositions in spontaneous
speech but many on tests. As for prefixes, the adult had neither errors in her spontaneous
speech nor on the grammatical tests two years post-onset, while the child continued to make
mistakes (median error rate 17%) even three and a half years, after her injury.
Writing and reading
The writing samples of the child and adult could not be compared due to the adult patient's
lasting right hemiparesis.
Both patients reported difficulties in reading. The child expressed these sentiments:
(74) To je hrozny, jd vzdycky kdyz si to pfectu, ale moc tomu nerozumim, tak si to musim
vod zacdtku vsechno, takze to je delsi doba... Nekdy tfeba... kdyz j'a ctu a nekdo na me
promluvi, tak to musim cist uplneznova, a toje...
'That is awful, I always when I have read it, but I do not understand it very much, thus
I have to ... from the beginning everything, thus it is a longer time ... Sometimes
maybe ... when I am reading and somebody speaks to me, so I have to reread it from
the very beginning again, and that is ...'
The adult:
(75) Mne se ted' stdvd to, ze viibec, jako ted'... nevim, co ctu i co je to na ... ee ... za
pismenko, tfeba a musim to odlozit, ze chci taky cist, jo. Nahlas mne to nejde, to se
nejak zarazim a nemuzu, tak abych vedela, co ctu, zejo, no takjenom tohle.
' It happens to me that I at all, as now ... do not know what I read and what is on ... ee
... which letter, maybe I have to put it down, that I want to read too, yes. I cannot read
aloud, I get somehow stuck and cannot, so that I know what I am reading, all right, so
thus only this.'
APHASIA IN CHILDHOOD AND IN ADULTHOOD
Acquired childhood aphasia is a relatively rare syndrome and its clinical description has
changed substantially over the last 15 years (Van Hout, 2000). The patterns of aphasia in
childhood had been reported not to be exactly like those of adulthood. The similarity of all
aphasias in children suggested that language abilities might be more diffusely organized for
them, at least within the language areas, than in adults. Obler (2000) states that there are no
reports of fluent aphasias in children. Even when the localization of the lesion would suggest
a Wernicke's aphasia, the child will produce slow, laborious speech with reduced syntactic
complexity, if not outright agrammatism. Most recently, however, neurological studies have
observed that aphasic syndromes in children are somewhat similar to adult aphasic
syndromes. Van Hout (2000) maintains that all aphasic subtypes described in adults also
apply to children. Prognosis is less favourable than what was originally thought, both in terms
of language sequelae and academic failure (Pavao Martins, 1991; Van Hout, 1992).
Lenneberg (1967) studied children with unilateral brain injury to analyse its effects on
language and concluded that the two hemispheres are initially equally able to control
language. This is known as the equipotentiality hypothesis. He maintained that in the age
group consisting of 3-10 year-olds aphasic symptoms tend to be fully recovered (except in
reading and/or writing). At that age, there is evidence of both hemispheres' involvement in
language and it is thus possible to re-establish language in the right hemisphere if the left is
damaged. In the age-group of 11-14 year-olds, some aphasia symptoms are not reversible,
particularly after traumatic lesions. Lateralization is formally established already at this age
and it is usually irreversible.
More recent studies have shown that the right hemisphere is not entirely able to take over
language functions, even in childhood. The language of children who had had left-brain
damage looked normal on the surface as they participated in daily conversation. However,
grammatical tests revealed below-normal performance (Dennis & Kohn, 1975).
Aram (1988) analysed the spontaneous speech of aphasic children and concluded that left-
hemisphere-damaged children had more difficulty with simple and complex sentences than
normal controls. Both left- and right-hemisphere-damaged children showed persistent
difficulty in naming objects. Children with a left-sided injury answered questions more slowly
but more accurately than children with a right-sided injury. Aram found no effect of age at the
time of brain injury in any of her analyses.
Murdoch (1990) points out that although the rate of spontaneous recovery in children
following closed head injury is often described as excellent, persistent long-term language
disorders have been reported. Especially naming difficulties in children with acquired aphasia
are more common than first thought. Written language disorders are often more severe than
oral language impairment, and expressive disorders more frequent than the receptive disorders
of written language.
The present case of child aphasia in Czech confirms these findings. On the one hand, the
patient recovered remarkably well from a serious polytrauma and from being in a comatose
state for four months. After a global aphasia, she progressed into a mild amnestic aphasia and
re-achieved most of the linguistic functions, except for fluent writing and reading. On the
other hand, a linguistic analysis uncovers grammatical and lexical problems that do not
necessarily appear in everyday conversation but which nevertheless restrict the patient's
language.
Van Hout (2000) points out that, although aphasic symptoms in children have been found
to be similar to those of adults, children may show atypical symptoms. Their frequency might
provide an explanation for some of the discrepancies found between classic and new
symptom descriptions in childhood aphasia. Some of these symptoms consist in a dissociation

between the different fluency parameters, or in general hypospontaneity.
Hypospontaneity was not manifested in the present case of a Czech child. Compared to an
analogous case of adult aphasia, the child's spontaneous speech appeared to be more normal.
The child neither noticed nor corrected her errors, whereas the adult patient was very upset
about her mistakes. While the adult compensated for her word-finding difficulties using a
very slow speech rate and long pauses while she searched for words from the same lexical
field, the child replaced the missing lexemes either by neologisms that have the formal
characteristics of the target word, or by indefinite expressions. In other respects, the linguistic
deficit of the child corresponds both quantitatively and qualitatively to the comparable adult
case.
Moreover, the type of linguistic errors made by the child is in accordance with the
manifestation of aphasic symptoms in Czech adults in general. My synoptic study of the
spontaneous speech of Czech aphasics (cf. Leheckova, 2001) showed that:
a. the morpho-syntactic structure of language production is impaired for all patients;
b. grammatical errors result from both omissions and substitutions;
c. free grammatical morphemes are only omitted, not substituted;
d. bound grammatical morphemes are only substituted, not omitted;
e. morpho-syntactic errors caused by substitutions are more frequent than those caused
by omissions of grammatical morphemes;
f. grammatical words are omitted less than content words;
g. in substitutions, the replacing forms are existing grammatical morphemes, mainly
belonging to the same grammatical category (e.g. one case ending substituting for
another);
h. the richer a category is in forms, the more it is liable to errors;
i. there is a clear hierarchical order in the substitution of some grammatical forms for
others.
Both the child and the adult patient differ from the general tendency in (e) only. This
difference may be caused by their type of aphasia (i.e. amnestic) because the general
tendencies were investigated mainly in Broca's aphasics.
CONCLUSION
To answer the question in the title: What is lost three and a half years after the accident is
fluent writing and reading, smooth access to the whole lexicon and sensitivity to some more
complicated grammatical features, especially to those with a formal, not semantic character.
What is re-achieved after an almost hopeless state is the ability to use language for
communication and reflection. The child's aphasia is basically manifested in the same way as
the adult's aphasia with one exception: while the adult patient two years post-onset reached a
stage after which her language skills no longer improved, the child has continued to develop
her linguistic abilities.
ACKNOWLEDGEMENTS
I am grateful to Dr. Eva Skodova and Dr. Jitka Stejskalova for allowing me to work with their
patients and providing me with their documentation.
REFERENCES:
Aram, D. M., B.L. Ekelman and S.A. Whitaker (1986). Spoken syntax in children with
acquired unilateral hemisphere lesions. Brain Lang, 27, 75-100.
Aram, D. M. (1988). Language sequelae of unilateral brain lesions in children. In: Language
communication and the brain (F. Plum, ed.), pp. Raven Press, New York.
Bishop, V. M. and L. B. Leonard. (2000). Speech and language impairments in children:
causes, characteristics, intervention and outcome. Psychology Press, Hove.
Dennis, M. and B. Kohn (1975). Comprehension of syntax in infantile hemiplegics after
cerebral hemidecortication: Left hemisphere superiority. Brain Lang, 2, 472-482.
Eisenson, J. (1972). Aphasia in children. Harper and Row, New York.
Grodzinsky, Y., L.P. Shapiro and D. Swinney (2000). Language and the brain:
representation and processing. Academic Press, San Diego.
Jakobson, R. (1968). Child language, aphasia and phonological universals. Mouton, The
Hague.
Joanette, Y. and H. H. Brownell (1990). Discourse ability and brain damage: theoretical and
empirical perspectives. Springer, New York.
Leheckova, H. (1988). Linguistic theories and the interpretation of agrammatism. Clin
Linguist Phon, 2, 271-289.
Leheckova, H. (1997). Jazyk, komunikace a fecove poruchy. In: Afdzie (P. Kulistak, H.
Leheckova, M. Mimrova and J. Nebudova, eds.) pp. 125-175. Triton, Praha.
Leheckova, H. (2001). Manifestation of Aphasic Symptoms in Czech. Journal of
Lenneberg, E.H. (1967). Biological foundations of language. John Wiley and Sons, New
York.
Murdoch, B. E. (1990). Acquired neurological speech/language disorders in childhood.
Taylor and Francis, London.
Obler, L. K. and K. Gjerlow (1999). Language and the brain. Cambridge University Press,
Cambridge.
Paradis, M. (1989). Bilingual aphasia test. Czech version. Lawrence Erlbaum Associates,
Inc., New Jersey.
Pavao Martins, I., A. Castro-Caldas, H.R. Van Dongen, A. Van Hout (1991). Acquired
Aphasia in Children. Martinus Nijhoff Publishers, Dordrecht.
Van Hout, A. (1992). Acquired aphasias in children. In: Handbook in Neuropsychology (F.
Boiler and J. Grafman, eds.), Volume 7, pp. 139-161. Elsevier, Amsterdam.
Van Hout, A. (2000). An outline of acquired aphasia in children. Saggi CD&D, 26, 13-21.
Persistent Acquired Childhood Aphasia 231
12
PERSISTENT ACQUIRED CHILDHOOD APHASIA
Isabel Pavao Martins

Universidade de Lisboa, Portugal
Abstract — It is recognized that acquired childhood aphasia has a favourable prognosis, but the long-
term outcome of chronic cases is not known. Eleven patients are described (mean age = 17.5 years, +
5.2), who remained chronically aphasic for 2 or more years following a non-progressive brain lesion
sustained in childhood (mean age at onset = 9.5 years). These patients belong to a series of 50 children
with aphasia followed for an average of 6.9 years after lesion onset. Factors associated with persistent
aphasia were traumatic or infectious etiology, poor verbal comprehension at onset, epilepsy and
lesions involving the classical language areas of the left hemisphere. Six patients had mild forms of
anomic aphasia, and the majority continued to improve their language test scores, changing their
aphasic profile on consecutive evaluations and showing that recovery continues in the chronic period.
In adult age these patients had educational, social and professional problems. Children who recovered
were compared with a control group consisting of their healthy siblings of identical age. Recovered
aphasics had significantly lower scores on short-term and working memory tests and also had
educational difficulties. Our findings stress the need for rehabilitation and educational/professional
support in the chronic stages of childhood aphasia.
Key words: acquired aphasia in children, recovery, prognosis, chronic aphasia in children.
INTRODUCTION
Language is a well lateralized and localized function in the adult brain, and the study of
acquired aphasia in children allows us to understand the development of its cerebral
organization with time and cognitive development.
In the last two decades, systematic studies of acquired aphasia in children have dismissed
traditional beliefs about this disorder showing that: a) crossed aphasia is rare in children and,
therefore, in normal circumstances language develops in the left hemisphere from the outset
(Martins, 1997; Marien et al, 2001); b) all types of aphasia may affect children, namely
fluent aphasia, jargon aphasia and disturbances of oral comprehension (Van Hout, 1986;
Klein et al, 1992; Paquier et al, 1996; Van Dongen et al, 2001); c) positive aphasic
symptoms may occur such as paraphasias, circumlocutions, conduites d'approche,
preserverations, etc. (Van Hout et al, 1985); and d) anatomo-clinical correlations are identical
to those found in adults, suggesting a modular developmental organization (Van Dongen et
al, 1985; Cranberg et al, 1987; Martins & Ferro, 1992; Martins & Ferro, 1993; Nass et al,
1998; Van Dongen et al, 2001). The last evidence, however, is based on few studies.
The most striking differences between child and adult acquired aphasia concern the high
frequency of mutism and nonfluent types of speech in children (Basso & Scarpa, 1990; Van
Hout, 1991; Martins, 1997), and prognosis, which is more favourable in young children.
Although evidence of good recovery has been reported in a number of studies (Cranberg et
al, 1987; Satz & Bullard-Bates, 1981; Loonen & Van Dongen, 1990; Martins & Ferro,
1992b), only few reports address the long-term outcome in these children. Some authors
described subtle language sequelae and learning disabilities after apparent clinical recovery of
acquired aphasia (Woods & Carey, 1979; Riva & Cazzaniga, 1986; Cooper & Flowers, 1987;
Cranberg et al, 1987; Eisele & Aram, 1993), but only a few reported chronic aphasia.
Watamori et al. (1990) described in detail three patients with acquired aphasia followed
up into adult life. Three other studies (Cooper & Flowers, 1987; Basso & Scarpa, 1990; Lees,
1997) reported cases of permanent language sequelae of variable types and degrees, 2 or more
years after acquired aphasia. These series includes patients with Landau-Kleffner syndrome,
which is not a good model to study recovery as its underlying pathology may be progressive
or fluctuating and thus interfere with the usual recovery mechanisms.
The purpose of this report is to describe the long-tem outcome in adulthood of patients
who suffered aphasia during childhood, focusing on those that did not recover.
METHOD
Population
Children included in this study are part of a series of 120 children with acquired brain lesions
referred to our Centre since 1979.
Inclusion criteria for the present study were: 1) lesion onset after some language
development (minimum age was 18 months) and before 15 years of age; 2) documented
diagnosis of acquired aphasia (with language testing) during the acute or chronic stage; 3)
brain lesions due to non-progressive pathology; vascular, traumatic and infectious etiologies
were included, but not patients with a diagnosis of tumor or Landau-Kleffner syndrome; 4)
follow-up time longer than two years for children that did not recover.
Procedure
Clinical data concerning lesion onset, etiology, epilepsy, biographical, educational and social
information were obtained from medical records and through an interview with parents.
Children were assessed by a test battery including: 1) the Lisbon Aphasia battery (Castro-
Caldas, 1979) comprising an analysis of spontaneous speech features such as fluency, syntax,
prosody, anomia, paraphasias and dysarthria; and tests of visual object naming, auditory
comprehension and word and sentence repetition; 2) the aphasia severity rating scale from the
Boston Diagnosis Aphasia Examination (Goodglass & Kaplan, 1983); 3) fluency grades from
the Western Aphasia Battery (Kertesz, 1979); 4) a short version of the Token Test (De Renzi
& Vignolo, 1962; Benton, 1969); 5) WAIS or WISC forward and backward digit span
(Wechsler, 1949, 1970). Children younger than 6 years were also assessed by the Reynell
Developmental Language Scales (Reynell, 1977).
Children who recovered from aphasia and were more than 10 years of age, were also
tested by a more extensive battery including the WAIS/WISC (Wechsler, 1949, 1970) and the
Wechsler Memory Scale (Wechsler, 1969), among other tests, and are now being compared
with the'ir healthy brothers or sisters in an on-going case-control study.
Lesion localization was determined either by Computerized Tomography (CT) or brain
Magnetic Resonance Imaging (MRI) and judged by two independent observers who were
asked to verify the presence or absence of lesions in selected brain areas of the left
hemisphere, according to brain maps (Matsui & Hirano, 1978; Damasio, 1995). Areas of
interest included the classical language areas of the left hemisphere (Broca and Wernicke's
areas), insula, superior temporal cortex (Brodmann areas 41, 42), supramargynal gyrus and
angular gyrus (Brodmann areas 40 and 39). Extension of lesion in Wernicke's area was
scored according to a 5-point scale: grade 0 = no lesion; 1 = lesion involving less than 25% of
this area; 2 = 25-50% of Wernicke's area was damaged; 3 = 50-75% of area was involved;
and 4 = more than 75% of the area was damaged. Since this study began before imaging
studies were systematically performed, the number of imaging exams is lower than the
number of children included. In the remaining cases, lesion side and localization were
determined by clinical evaluation, angiography or surgery. Only CT and MRI data were
analysed.
Socio-economic status of patients who reached adult age (or their controls) was classified
in 4 points, according to their occupation (or their spouses', if married): 1 - Unskilled manual
labour; 2 - Skilled manual work, 3 - Clerk or technical work and small trade, 4 -
Professionals, large trade or land owners.
Operational criteria used to define recovery were the following: a) normal speech
(concerning fluency, syntax, prosody, effort and articulation); b) normal performance on the
four cardinal tests of the Lisbon battery (naming, repetition, fluency and auditory
comprehension) or on both Reynell subscales; c) maximum score on the aphasia severity
rating scale and, d) normal (age-adjusted) performance on the Token Test.
A neurological examination was also performed to quantify motor impairment at the time
of language assessment.
RESULTS
Fifty children (27 girls and 23 boys) had sustained aphasia due to a static brain lesion.
Their age at lesion onset was, on average, 8.7 (+3.9) years. Thirteen patients were pre-
schoolers and 37 attended school, their school grade being on average 4.05 (+2.6) years.
Lesion etiology was varied: vascular in 19 cases, traumatic in 23 and infectious in 8 cases.
The lesion was localized unilaterally to the left hemisphere in 40 cases, bilateral in 6 and
localized to the right hemisphere in 4 patients (including one left-handed child). Thirty-four
(34) patients received the first neuropsychological evaluation in the first month, six (6) during
the first six months and ten (10) after the 6th month. Children were re-evaluated and followed
for variable periods of time, ranging from 6 months to 27 years (6.9 years on average): 39
(78%) patients recovered from aphasia, while 11 (22%) did not. Forty-two patients had
imaging exams (CT scan in 30 cases and MRI in 12). Only left hemisphere lesions were
compared between the two outcome groups.
Prognostic factors. Children who recovered were not significantly different from those
that did not recover in terms of (Tables 1 and 2) age at aphasia onset (8.4 years for those who
recovered versus 9.5 years for those who remained aphasic [t = - 0.80, p=0.42, n.s.]), gender
(X2= 0.41, p= n.s.), mean follow-up time (6.8 and 7.5 years, respectively [t=-.27, p= n.s.]), age
at last evaluation (14.9 and 16.8 years, [t=-.82, p= n.s.]), school grade of school attenders
(4.04 and 4.11 years, respectively [t=-.75, p= n.s.]), knowledge of reading and writing (pre-
schoolers and first graders versus those mastering reading and writing [x2= 0.78, p= n.s.]),
mutism during the acute period (yes or no [x2=-00, p= n.s]), or speech fluency at onset (fluent
or nonfluent [x2 = 1.8, p=n.s.]). Lesion side (left, right or bilateral) was not different in the
two outcome groups (x 2= 0.13, p= n.s.) neither was the occurrence of hemiparesis (x = .47,
p= n.s.), change of handedness (yes or no [x =.36, p= n.s]), or convulsions during the acute
stage (yes or no [x2=3.4, p= n.s.]).
Table 1 — Recovery from aphasia
Aphasia Handedness Sex Mean age Lesion side Follow-up Age at last
recovery N at onset Left Right Bilat time evaluation
mean (sd)
R L A M F (yrs) (years) mean (sd) yrs
Yes 39 35 2 2 17 22 8.4(3.8) 31 3 5 6.8(7.5) 14.9(7.5)
No 11 11 0 0 6 5 9.5(4.2) 9 1 1 7.5(4.5) 16.8(4.4)
X2=41 t=-.8O X2=.13 t=- .27 t=-.82

p=n.s. p=n.s. p=n.s. p=n.s. p=n.s.
Note: R = right, L = left; A = hand preference undefined; M = male; F = female
Factors associated with persistent aphasia included impaired verbal auditory

comprehension (simple verbal commands) in the acute stage (%2= 6.88, p=.008) and infectious
or traumatic etiology, compared to stroke (x2=13.3, p=.001). Patients who later developed
epilepsy (x =6.20, p=.01) had a worse outcome. Two children developed intractable epilepsy,
being candidates for surgery: one boy was in the chronic aphasia group, while the other (with
a left latero-temporal vascular malformation) was in the recovered group.
Speech therapy was inversely related to outcome (x2=21.18, p<.000) since the majority of
children who recovered were never referred to speech therapy, while chronic aphasics were.
Thirty-four patients with demonstrable left hemisphere (unilateral left or bilateral) lesions
were studied by imaging. There was a high agreement between two independent observers in
terms of presence or absence of lesions in areas of interest of the left hemisphere (Cohen's
kappa coefficient= 0.88). Lesions of the classical left language areas were associated with a
poor outcome, that is damage to Broca's area (cortical and/or subcortical [x 2= 3.85, p=. 04]),
superior temporal cortex (auditory cortex, BA 41,42 [x 2= 13.46, p= .0002]), Wernicke's area
(cortical and or subcortical [x2= 14.24, p= .0002]), and the insula (y}= 15.9, p= .0001) (Table
3). In contrast, lesions involving the prefrontal or the parietal cortex (BA 39 and 40), and
subcortical lesions (basal ganglia and subcortical white matter fiber tracts) were not
associated with a worse outcome. The extension of lesion into Wernicke's area was associated
with the outcome, being worse when lesions were more extensive [x = 29.9, p< .0000](see
Table 3).
Cases that did not recover. Eleven children (6 boys and 5 girls) (Table 4) remained
chronically aphasic for 2 or more years after the lesion onset. Mean follow-up time was 7.5 (+
4.5) years, ranging from 2 to 15 years. Six children had severe head injuries with depressed
fractures, and five had infectious diseases of the CNS (cerebral malaria in one case, herpes
encephalitis in three and tuberculous meningitis in one). All but two children (pre-schoolers)
were attending school at the time of injury. Four children had persistent motor weakness
mostly involving the right hand, and became left-handed. All eleven children walked
independently at the last examination.
Table 2 — Recovery from aphasia - Prognostic factors
N Recove red Urn-eco1*/ered

N=39 N=ll
Etiology
stroke 19 19 0
head trauma 23 17 6 X2=13.2
infection 8 3 5 p= .001
Mutism*
yes 28 23 5 X2=.OO
no 11 9 2 p=n.s.
Speech fluency*
nonfluent 21 20 1 x2=1.80
fluent 16 8 8 p=n.s.
Auditory comprehension*
normal 23 23 0 X2=6.88
poor 11 8 3 p=.008
Token Test score*
normal 5 5 0 X2=2.35
poor 33 22 11 p=n.s.
Hemiparesis*
yes 40 32 8 X2=0.47
no 10 7 3 p=n.s.
School grade at lesion onset
Preschooler +lst grader 20 16 4 X2=0.78
More than one year 30 23 7 p=n.s.
Speech Therapy
yes 17 7 10 X2=21.18
no 28 28 0 p< .0000
Chronic epilepsy
yes 3 1 2 X2=6.2
no 41 36 5 p=.O13
1
Results of the evaluation performed during the acute period
Table 3 — Recovery from aphasia - Damaged areas
Area of interest Lesion ]Recovered Not recovered X2= p=

Left hemisphere
Prerolandic yes 8 3 1.04 n.s.
no 20 3
Broca (cortical/ yes 3 3 3.85 0.04
/subcortical) no 24 4
Insula yes 3 6 15.9 0.0001
no 24 1
Temporal cortex yes 4 6 13.46 0.0002
(areas 41, 42) no 23 1
Wernicke's area yes 6 7 14.24 0.0002
(cortical/subcortical) no 21 0
Parietal cortex yes 11 5 2.1 n.s.
(BA 39, 40) no 16 2
Caudate nucleus yes 4 1 0.00 n.s.
no 23 6
Lenticular nucleus yes 10 1 1.31 n.s.
internal capsule no 17 6
Extent of lesion in Wernicke's area
No lesion 21 0 29.92 0.0000

Lesion < 25% 5 0
Lesion 25-50% 1 2
Lesion 50-75% 0 2
Lesion>75% 0 3
Nine patients were studied by imaging (MRI in four, CT scan in five cases). One child
(patient 10) had an extensive contusion of the right temporal lobe (including both the
hippocampus and the neocortex) and no visible lesions of the left hemisphere on MRI.
However, and due to its traumatic etiology, this case cannot be considered a crossed aphasia
case since traumatic lesions may be unapparent on imaging. All other patients had left
hemisphere lesions involving the temporal cortex (areas 41 and 42), Wernicke's area and
additional lesions of the insula and Broca's area in some cases. In 6 patients BA 39 and 40
were also damaged (Table 5).
Table 4 — Chronic Aphasics: clinical data and follow-up
N Sex Age at Etiology Chronic Change in Epilepsy

1esion onse:t Hemiparesis laterality*
1 M 8.2 Cerebral malaria yes(hand) yes(L) yes
2 M 8.6 Herpes encephalitis no no yes**
3 F 13.7 Herpes encephalitis no no no
4 M 14 Herpes encephalitis no no no
Tuberculous
5 F 11 meningitis yes(hand) yes(L) no
HT +
6 M 12 Parietal DF no no no
HT+DF+
7 F 6.7 FTP contusion no no no
HT+ DF+TP
8 M 15.4 contusion yes(hand) yes(L) no
HT+DF+
9 F 1.5 T contusion no no no
Closed HT+
10 F 5.7 RT contusion no no no
HT+DF+
11 M 7.7 FP contusion yes (hand) yes(L) no
N Ileturn to Special Repeated Final Follow-up Final age

school education grade grade (years) (years)
1 yes yes yes 1 11 19
2 yes yes 9 12 20
3 yes no 3 8 5 18
4 no - - 7 3.3 17
5 yes yes yes 5 2 12.6
6 no - - 4 8 20
8 no - - 4 4 20
9 yes no no 2 7.7 8.9
11 yes yes yes 9 2.3 10.2
Note: *present handedness in brackets; ** intractable epilepsy, requiring surgery
Evolution of chronic aphasia. Although these eleven patients remained aphasic two or
more years after lesion onset, their aphasic profile, aphasia severity and taxonomic diagnosis
continued to change in the chronic period (Table 6). For instance, patient 6 changed from
Broca's to conduction aphasia (improving fluency) more than 5 years after lesion onset.
Another patient (case 9) evolved from transcortical sensory aphasia to anomic aphasia more
than three years after onset and patient 7 was still improving after more than ten years. Her
diagnosis evolved from conduction aphasia to anomic aphasia.
Table 5 - Unrecovered patients-lesion localization in the left hemisphere
Imaging lesionPre-ro Temporal Wernicke's area Lesion angular supra

N Exam side landicBrocalnsula Cortex (BA 22) Ext. gyrus marginal
cortex BA 41,42 B/W W Sc Wernicke (BA 39) (BA40)
1 CT/MRI L + + + - + + + 3 + +
2 CT/MRI L - - - + + + + 4 + +
3 CT L - - - + + - + 4
4 CT L - - + + + + + 4
5 CT/MRI Bilat - + + + + + + 2 - +
6 CT L + - + + + + + 3 + +
7 CT L + + + + + + + 2 - +
8 no L
9 no L
10* CT/MRI R . . . 0
11 CT L + + + + + + + 4 + +
CT=Computerized Tomography; MRI-Magnetic Resonance Imaging
R=right hemisphere*, L=left hemisphere, Bilat=bilateral
BA=Brodmann area; + lesion involving that area; -no lesion
Changes were observed in all aspects of language abilities, namely visual naming, word
repetition, auditory comprehension of simple verbal commands and speech fluency scales
often switching from non-fluent to fluent types of speech (Table 7). On the last evaluation,
most children (6/11) had mild forms of anomic aphasia. Language scores declined in only one
patient. This was related to the development of an intractable epilepsy requiring surgery.
School achievement. Although these children had mild aphasic syndromes, three of them
never returned to school. Five of them attended special education classes. All but one (the
younger child at lesion onset) repeated grades and eventually left school (Table 4).
Socio-professional achievement. By the time the last evaluation was performed, 9 patients
had reached adult age (> 18 years of age) (Table 8). Four were unemployed, often after trying
several jobs. Two patients (cases 6 and 7) specifically mentioned that they were unable to
keep their jobs because of their inability to maintain a conversation with clients (while
working as a salesman in a shop and helping in a hairdressers', respectively). Patient 6 was
now working on a voluntary basis (being unpaid) in a garage, helping to load and unload
materials. Others complained of reading or writing difficulties as their main handicap for
work. Only three subjects were still studying (in two cases, special education classes).
Although four had a paresis of the right upper limb, involving the hand, none of them
mentioned it as a cause for their professional or school difficulties. Compared to their parents,
these young adults had a lower socio-economic status. Most had brothers or sisters who were
still studying. Only one patient was economically independent, all others depended upon their
parents or spouses.
Table 6 — Chronic aphasic patients - Type and Severity of Aphasia: changes at follow-up
Time Post-onset
N First 1-6 6-12 1-3 4-5 5-10 >10
month months months years years years years

1 0 G 0 G 1 B 2 Tm 2 Tm
1 Tm
2 4 W 5 Ts 5 Ts 5 A 5 A 5 A
3 1 Ts 2 Ts 4 A 4 A
4 4 W 4 A 4 A 4 A
4 Ts
5 1 B 1 B
6 5 B 5 B 5 C
7 4 C 4 C 4 A
8 1 G 1 G 1 G
9 3 Ts 4 A
10 4 Tm 4 A 4 A 5 A
11 0 Mutism 0 G 2 G 4 Tm
Note: T= transcortical aphasia; Tm = Transcortical motor aphasia; Ts =

Transcortical sensory aphasia; G = global aphasia; A = anomic aphasia; C =
conduction aphasia; B = Broca's aphasia; W = Wemicke's aphasia. Numbers
represent severity of aphasia by the Boston Diagnosis Aphasia Examination
(Maximum =5)
Two patients developed epilepsy. In one case (patient 2) it was a chronic intractable epilepsy
requiring surgery (left anterior temporal lobectomy) which did not lead to any consistent
improvement. Another patient (patient 1) developed epilepsy when he was 17 years old but
easily controlled it with sodium valproate.
Family life. Only the three female patients in this adult group got married and had
children. Since they did not work, they were economically dependent on their husbands or
parents. All the male patients continued to live with their parents, were economically
dependent and did not establish new stable emotional relationships.
Table 7 — Performances of chronic aphasics on Naming, Word Repetition, Verbal Comprehension,

Speech Fluency (changes at follow-up)
Naming Performance
1 month 1 to 6 6-12 1 to 3 4 to 5 5 to 10 >10yrs
months months years years years
1 0 0 3 6 7
2 0 11 12 13 14 11
3 0 3 9 11
4 0 9 10 14
5 0 0
6 15 15 15
7 13 13 13
8 0 0 0
9 10 12 13
10 10 13 16 15
11 0 8 14
Maximum score = 16
Word Repetition Test
1 month 1 to 6 6-12 1 to 3 4 to 5 5tol0 >10yrs
months months years years years
1 0 0 21 28 30
2 20 30 30 30 30 30
3 28 30 30 30
4 11 30 29 30
5 0 15
6 26 25 27
7 22 24 26
8 0 0 2
9 30 30 29
10 30 30 30 30
11 0 15 28
Maximum score = 30
Ve rbal comiprehensiian
1 1 to 4 to 5 >10
N month 1 to 6, 6-12 3 5 tolO yrs
nnonths month,s years years years
1 7 6 7 8 6
2 5 7.5 7 8 8 8
3 1 3.5 7 7
4 6.5 7.5 8 8
5 8 8
6 8 8 8
7 7.5 8 8
8 2 3.5 6
9 5.5 8 8
10 8 8 8 8
11 3 5 8
Maximum score = 8
Speech fluency
1 lto 4 to 5 >10
N month 1 tof> 6-12 3 5 tolO yrs
rnonths months years years years
1 Mutism NFO NFO NF3 NF4 NF4
2 FL 3 FL4 FL5 FL5 FL5 FL5
3 FL 3 FL3 FL3 FL3
4 FL3 FL4 F14 FL4
5 NF3 NF3
6 NF5 NF5 FL4
7 FL4 FL4 FL4
8 NF 0 NF1 NF1
9 FL4 FL4 FL4
10 NF5 NF5 FL5 FL5
11 NFO NF3 NF5
NF= non fluent; FL= Fluent. Figures represent fluency grades
(from the Western Aphasia Battery ranging from 0 to 5 in
either scale, fluent and nonfluent) at different follow-up times.
Note: Numbers represent raw scores obtained at different
follow-up times.
Recovered Cases. Thirty-nine children (17 boys and 22 girls) recovered from aphasia. The
majority (19 cases) had vascular lesions, but there were also traumatic (17) and infectious
cases (3). At the time of the last evaluation their age was, on average, 14.9 years (+ 7.5).
To understand their cognitive and educational achievement, those children were compared
with their healthy siblings of identical age in a case-control study: 16.8 years (+ 5.5) in
patients and 16.6 years (+4.5) in controls.
Table 8 — Chronic aphasics - Evaluation in adult age ( age > 18 years)
folio hand SES economic

Pt w-up ed motor Independent Job or pati pare dependen Family
N Age time ness defect Walking Occupation ent nts cy Life
special
RHemi education lives with
1 19 11 L paresis Yes gardening 1 yes mother
lives with
2* 20 12 R no Yes unemployed 3 yes mother
married
3 18 5 R no yes student 4 yes one child
unskilled lives with
4 18 3.3 R no yes work 1 2 no parents
R Upper
limb married
5 23** 13 R paresis yes unemployed 1 yes one child
voluntary
manual job lives with
6 20 8 R no yes unpaid 1 3 yes an aunt
married
pregnant
lives with
7 22 15 R no yes unemployed 1 2 yes husband
R
Upper
limb lives with
8 20 5 L paresis yes unemployed yes parents
R
Upper
limb special lives with
11 18** 10 L paresis yes education 3 yes parents
Pt= patient, L=left, R right;
Figures represent socio-economic status, estimated from jobs or from spouses' occupation
*intractable epilepsy, submitted to epilepsy surgery
** information obtained at present from clinical records and by telephone interview
An analysis of the first eleven patient-control pairs has shown that:

a) There are no significant differences between the two groups concerning verbal IQ
(patients: VIQ = 96.2, controls: VIQ= 106.2 [paired t test = -1.4, p = n.s.]),
performance IQ (patients: PIQ = 100.8, controls: PIQ = 111.3 [paired t = -0.6, p =
n.s.]) or full scale IQ (patients: FIQ= 97.6, controls: FIQ=109.4 [paired t = -1.4, p =
n.s.]) neither on Wechsler verbal episodic memory subtests, logic memory (patients =
9.3, controls = 11.8 [paired t= -.91, p= n.s.]), word pair associates (patients = 16.3,
controls = 17.9 [paired t= -1.6, p=n.s.]) or visual memory (patients = 12.0, controls =
14.3 [paired t= -.91, p= n.s.]), although controls had higher average scores on all tests.
b) Ex-aphasics have significantly lower scores in forward and backward digit span
(patients' average = 8.2, controls' average =12.5 [paired t= -6.14, pO.OOO, df =9]).
c) Although patients had a poor school achievement, they did not differ significantly from
controls in grade repetitions (average number of grade repetitions of patients was 1.5
versus 1.0 of controls) (excluding the year of illness or missing school because of
illness) or final school grade.
DISCUSSION
Focal brain lesions in childhood are associated with a favourable prognosis for language
recover}' and subsequent development, as demonstrated in several longitudinal studies of
children with congenital or perinatal lesions of the left cerebral hemisphere (Eisele & Aram,
1993; Bates et al, 1999; Vargha-Khadem et al, 1994; Trauner et al, 2001), acquired aphasia
(Cranberg et al, 1987; Loonen et al, 1990; Satz , 1981; Martins & Ferro, 1992b), and also by
the follow-up of patients with left hemispherectomy (Dennis & Whitaker, 1976; Ogden, 1988;
Vargha-Kadhem et al, 1991; Devlin et al, 2003). These studies have shown that patients
with left hemisphere damage have a nearly normal development of language, although a fine-
grain analysis of their linguistic abilities may reveal subtle defects. This indicates a
remarkable plasticity of the developing brain for functional re-organization as well as
language resilience after brain lesions.
The present study suggests that plasticity declines after language acquisition. While
almost 80% of children recovered - a higher percentage than the one reported in adult patients
- this outcome is worse than when lesions occur before language acquisition, where clinical
aphasia is unusual. Of 50 children who developed aphasia between the ages of 1.5 and 15
years, 11 (22%) remained chronically aphasic after a long follow-up. However, age was not a
significant factor for prognosis in this particular group. Indeed, it was found that lesion-
related variables (site and nature, late epilepsy) but not subject-related variables (age, gender,
grade of education) were the main factors responsible for the outcome. This corroborates
results from a previous research on a smaller sample of patients (Martins & Ferro, 1992b) and
studies of adult aphasia (Basso, 1992) suggesting that individual variables cannot compensate
for the degree and type of structural damage. This study also showed that albeit mild, child or
adolescent chronic aphasia has an impact on the educational, socio-professional and family
life, just as would be expected in adults.
Regarding recovery, there are two main differences between these findings and findings
related to adult aphasics. First, unlike adults, the overall recovery was better in children and,
secondly, children improved in the chronic period. In different series of stroke patients,
recovery rates of adult aphasia vary between 40 and 50% (Kertesz, 2002). In the present
study, all vascular cases recovered completely, even when lesions were localized bilaterally.
The same advantage for children compared to adults was found by Bates et al. (2001) in a
direct comparison of language abilities in patients with left hemisphere lesions sustained in
the perinatal period or during adulthood. However, besides the brain plasticity hypothesis,
other factors may account for a better recovery, namely the specific topography and
pathogenesis of stroke in young children (Ganesan et al, 2003). Regarding the other two
etiologies, traumatic aphasia in general has a better prognosis than stroke (Basso, 1992), but
the outcome of head injury and CNS infections depends upon multiple factors (type and
severity of injury, infectious agent, etc.), and differences between children and adults are
more disputable (Basso & Scarpa, 1990).
Another difference between child and adult aphasics, was the timing of recovery. Curves
of aphasia recovery in adults (again, mostly stroke patients) have shown that the greatest
improvement is seen in the first 3 months (especially in the first three weeks) (Hartman,
1981). There may be some recovery during the first year (Basso, 1992), but afterwards there
are no significant changes, and functional decline has occasionally been reported (Hanson,
1989). Although children who recovered did so during the first year, in our group of chronic
aphasics improvement of different language abilities was found not only after the first year,
but also 5 or more years after lesion onset, causing changes in the aphasic profile and
taxonomic diagnosis. Unusual patterns of recovery and long periods of aphasia improvement
in children have also been described by other authors (Watamori et al., 1990; Ogden, 1988;
Ikeda et al, 1993). It is difficult to attribute these changes to simple developmental
progression as some of these patients were already in the adult age. The long follow-up period
of children may explain some of this advantage, since adults are not usually followed for so
long.
Recovery mechanisms are different during the acute and chronic stages of brain damage.
During the first weeks after stroke, for instance, functional improvement is associated with
reversal of diachisis and ischemic penumbra, and better perfusion after volume reabsorption
(Ferro et al, 1999). During the chronic period, on the other hand, recovery depends upon four
basic forms of neuroplasticity (Grafman & Litvan, 1999), some of which have been
demonstrated in children with left hemisphere lesions. These mechanisms are: 1) Language
transfer to the homologous areas of the right hemisphere following cortical lesions in
childhood (especially associated with epilepsy) which explains recovery after left
hemispherectomy. This has been demonstrated by fMRI (Hertz-Pannier et al, 2002; Staudt et
al, 2002), electrical stimulation (Duchowny, 1996) and Wada testing (Miller et al, 2003). 2)
Map expansion, i.e. the use of neighbouring areas to perform the same function. This was
shown for language recovery (De Vos et al, 1995) and might be facilitated in childhood
because normal children activate more extensive neuronal networks than adults, while
performing verbal tasks (Gaillard et al, 2000), thus, naturally using "neighbouring" areas. 3)
Compensatory masquerade. This mechanism has been suggested by Stiles-Davis et al. (1988)
in children with right hemisphere damage who, although fully recovered, had difficulty
performing non-canonical drawing tasks, which suggested a rigid pattern of performance. 4)
Cross-modal reassignment, which is more often observed after congenital sensory
deprivation.
The lack of recovery in our 11 patients compared to those that recovered suggests that
lesions may have hampered some of these possible recovery mechanisms. Infectious and
traumatic lesions may cause bilateral hemispheric damage, even if not visible on imaging
exams. This could interfere with homologous area transfer. The extensive, or cumulative,
destruction of the left hemisphere language areas may have impaired map expansion. There
are some indications that the left temporal cortex may be particularly vulnerable to persistent
functional impairment. Left temporal lobe lesions have been associated with a worse outcome
even in congenital lesions. According to Bates et al. (1999), "this pattern is more evident
during the early stages of language acquisition". In our patients, the lesion extension in
Wernicke's area was a negative factor as was a poor verbal comprehension in the acute stage,
a symptom of left temporal involvement. This area may be of particular importance either
because of its aptitude to handle "perceptual detail" (Bates et al, 1999) or because it is one of
the first areas to commit to language, as demonstrated by fMRI in infants (Elman et al,
1996). This early commitment and eventual automatization of function, may enhance its
innate cytoarchitectonic aptitude to process speech sounds, reduce its representation and make
it difficult to transfer this function to neighbouring areas. However, the continuing recovery
of function in the chronic period, observed in these patients, suggests that plasticity
phenomena, possible behavioural compensation or compensatory masquerade, are still taking
place many years after injury, at least in the child and adolescent brain. They may be sensitive
to rehabilitation mesures.
Differences between congenital left hemisphere lesions and the cases just described
suggest that language acquisition may impair functional re-organization. These differences
support the theory of computational nativism [60] in the sense that the progressive
commitment of an area to a function may transform a small computational difference into a
major bias for functional processing to the point of no return.
The observed language impairment was sufficient to cause educational and socio-
professional handicaps. It is well-known that adult aphasia is associated with poor
professional and social adjustment, poor quality of life, depression and social impairment. In
this respect, these children were not different from adult aphasics, although their aphasic
syndrome was relatively mild as were their motor impairments, and only one had refractory
epilepsy. Children failed or left school. In adult age, they were economically dependent and
most of them did not live on their own nor had a family. This stresses the specific educational
needs of these patients. However, it is important to note that behavioural difficulties are
common not only to children with trauma but also after stroke (Blom et al, 2003),
independently of aphasia, and this may have an additional impact on the outcome that was not
specifically sought.
Although this is a very small sample, it is striking that women had an advantage
concerning family life compared to men, which may be related to the traditional gender effect
on social roles. These women were not impaired in social or family life but, as they were
unemployed, they were just as handicapped as men. Men are expected to support the family,
economically. However, we acknowledge that we need more information from the control
group to understand this data.
Children who recovered from aphasia might also have specific educational and
professional needs. Compared with the control group, they had lower scores on measures of
IQ and long-term verbal memory and visual memory. These differences reached statistical
significance on the phonological/working memory tests, which may underlie some of their
school difficulties often described as sequelae of acquired aphasia. Although these are
preliminary data, they stress the educational needs of recovered patients.
CONCLUSION
The prognosis of acquired childhood aphasia secondary to static brain lesions is generally
favourable, with a higher rate of recovery than in adults. Yet, language recovery is less
complete than that reported after congenital lesions, suggesting a reduced plasticity during the
first years of life. In a subset of patients, recovery is partial, even after long periods of follow-
up. Impaired verbal comprehension of simple verbal commands and lesion etiology and
localization are prognostic indicators. Although chronic aphasics tend to have mild forms of
aphasia and their linguistic abilities may continue to improve in the chronic stage, persistent
aphasia has a negative impact on their educational, social and professional life.
ACKOWLEDGMENTS
The author is indebted to Professor Jose Ferro, MD, for his help in the analysis of the imaging
exams and wishes to thank Drs. Alexandra Reis and Susana Rodrigues who participated in the
evaluation of controls.
REFERENCES
Basso, A. and M.T. Scarpa (1990). Traumatic aphasia in children and adults: a comparison of
clinical features and evolution. Cortex, 26, 501-514.
Basso, A. (1992). Prognostic factors in aphasia. Aphasiology, 6(4), 337-348.
Bates, E., J. Reilly, B. Wulfeck, N. Dronkerrs, M. Opie, J. Fenson, S. Kriz, R. Jeffries, L.
Miller and K. Herbs (2001). Differential effects of unilateral lesions on language
production in children and adults. Brain Lang, 79(2), 223-265.
Bates, E., S. Vicari and D. Trauner (1999). Neural Mediation of Language Development:
Perspectives from lesion studies of infants and children. In: Neurodevelopmental
Disorders (H. Tager-Fluberg, ed.), pp. 533-581. The MIT Press, Cambridge.
Benton, A. L. (1969). Development of a Multilingual Aphasia Battery. Progress and
Problems. JNeurol Sci, 9, 39-48.
Blom, I., E. L. L. M. De Schnyver, L. J. Kapelle, G. J. E. Rinkel, A. Jennekens-Schinkel and
A. C. B. Peter (2003). Prognosis of haemorrhagic stoke in childhood: A long term
follow-up study. Dev Med Child Neurol, 45, 233-239.
Castro-Caldas, A. (1979). Diagnostico e evolucao das afasias de causa vascular. Tese de
Doutoramento. Faculdade de Medicina de Lisboa.
Cooper, J. A. andC. R. Flowers (1987). Children with a history of acquired aphasia: Residual
language and academic impairment. J Speech Hear Dis, 52, 251-262.
Cranberg, L. D., C. M. Filley, E. J. Hart and M. P. Alexander (1987). Acquired aphasia in
childhood: Clinical and CT investigations. Neurology, 37, 1165-1172.
Damasio, H. (1995). Human Brain Anatomy in Computerized Images. Oxford University
Press, Oxford.
De Renzi, E. and L. A. Vignolo (1962). The Token Test: a sensitive test to detect receptive
disturbances in aphasics. Brain, 85, 665-678.
De Vos, K. J., E. Wyllie, C. Geckler, P. Kotagal and Y. Comair (1995). Language Dominance
in patients with early childhood tumors near left hemisphere language areas.
Neurology, 45, 349-356.
Dennis, M. and H. Whitaker (1976). Language acquisition following hemidecortication.
Linguistic superiority of the left over the right hemisphere. Brain Lang, 3, 404-433.
Devlin, A. M., J. H. Cross, W. Harkness, W. K. Chong, B. Harding, F. Vargha-Khadem and
B. G. R. Neville (2003). Clinical outcomes of hemispherectomy for epilepsy in
childhood and adolescence. Brain, 126, 556-566.
Duchowny, M., P. Jayakar, A. S. Harvey, T. Resnick, L. Alvarez, P. Dean and B. Levin
(1996). Language cortex representation: Effects of Developmental versus Acquired
Pathology. Ann Neurol, 40, 31-38.
Eisele, J. and D. Aram (1993). Differential effects of early hemisphere damage on lexical
comprehension and production. Aphasiology, 7, 513-523.
Elman, J., E. Bates, M. Johnson, A. Karmiloff-Smith, D. Parisi and K. Plunkett (1996). Brain
Development. In J.Elman & E.Bates (Eds): Rethinking innatness: Development in a
connectionistperspective, pp. 310-311. MIT/ Bradford Books, Cambridge, MA.
Ferro, J. M. and S. Madureira (2002). Recovery from Cognitive and Behavioral deficits. In:
Long term effects of stroke (J. Bogousslavsky, ed.), pp. 149-181. Marcel Dekker, Inc.,
New York.
Gaillard. W., L. Hertz-Pannier, S. Mott, A. Barnett, D. LeBihan and W. Theodore (2000).
Functional anatomy of cognitive development - fMRI of verbal fluency in children
and adults. Neurology, 54, 180-185.
Ganesan, V., M. Prengler, M. A. McShane, A. M. Wade and F.J. Kirkham (2003).
Investigations of risk factors in children with arterial ischemic stroke. Ann Neurol, 53,
167-173.
Goodglass, H. and E. Kaplan (1983). The Assessment of Aphasia and Related Disorders. Lea
and Fabiger, Philadelphia.
Grafman, J. and I. Litvan (1999). Evidence for four forms of Neuroplasticity. In: Neuronal
Plasticity: Building a Bridge from the Laboratory to the clinic (J. Grafman and X.
Chisten, eds.), pp. 131-139. Springer-Verlag, Berlin.
Hanson, W. R., J. Metter and E. H. Riege (1989). The course of chronic aphasia. Aphasiology,
3, 19-29.
Hartman, J. (1981). Measurement of early spontaneous recovery from aphasia with stroke.
Ann Neurol, 9, 89-91.
Hertz-Pannier, L., C. Chiron, I. Jambaque, V. Renaux ,. Kieffer, P. F. Van de Moontele, O.
Delande, M. Fohlen, F. Brunelle and D. Le Bihan (2002). Late plasticity for language
in a child's nondominant hemisphere. A pre- and post-surgery fMRI study. Brain, 125,
361-372.
Ikeda, M., H. Tanabe, K. Yamada, T. Yoshimine, T. Hayakawa, K. Hashikawa and T.
Nishimura (1993). A case of acquired childhood aphasia with evolution of global into
transcortical sensory aphasia. Aphasiology, 7, 497-502.
Kertesz, A. (1979). Aphasia and associated disorders: Taxonomy, localization and recovery.
Grune and Stratton, New York.
Kertesz, A. (2002). Behavioural and Cognitive Disorders. In: Prognosis of Neurological
Disorders (R. W. Evans, D. B. Baskin and F. M. Yatsu, eds), pp. 610-621. Oxford
University Press, Oxford.
Klein, S., D. Masur, K. Farber, S. Shinnar and I. Rapin (1972). Fluent aphasia in children.
Definition and natural history. J Child Neurol, 7, 50-59.
Lees, J. (1997). Long-term effects of acquired aphasias in childhood. Pediatr Rehabil, 1 (1),
pp. 45-49.
Loonen, M. C. B. and H. R. Van Dongen (1990). Acquired childhood Aphasia. Outcome 1
year after onset. Arch Neurol, 47, 1324-1328.
Marien, P., P. Paquier, S. Engelborghs and P. P. Deyn (2001). Acquired crossed aphasia in
dextral children revisited. Brain Lang, 79, 426-443.
Martins, I. P. and J. M. Ferro (1993). Acquired childhood aphasia: A clinicoradiological study
of 11 stroke patients. Aphasiology, 7, 489-495.
Martins, I. P. and J. M. Ferro (1992). Acquired subcortical lesions in children. In:
Neuropsychological disorders associated with subcortical lesions (G. Vallar, S.F.
Cappa and C.-W. Wallesch, eds), pp. 381-396. Oxford University Press, Oxford.
Martins, I. P. and J. M. Ferro (1997). Recovery of aphasia in children. Aphasiology, 6, 431-
438.
Martins, I. P. Lesoes cerebrals adquirldas na infdncia: dissolucao, recuperacao e
reorganizagdo das capacidades cognitivas na crlanca. Doctoral thesis. Faculty of
Medicine, Lisbon.
Matsui, T. and H. Hirano (1978). An Atlas of the human brain for Computerized
Tomography. Igaku-Shoin, New York.
Miller, J. W., C. B. Dodrill, D. E. Born and G. A. Ojemann (2003). Atypical speech is rare in
individuals with normal developmental histories. Neurology, 60, 1042-1044.
Nass, R., F. Leventhal, B. Levine, D.Lebron, C.,Maxfield P.McCaul, A.George and J. Allen
(1998). Conduction aphasia in a 3 year old with a left posterior cortical/subcortica
abscess. Brain Lang, 62, 70-88.
Ogden, J. A. (1988). Language and Memory Functions after long recovery periods in left-
hemispherectomized subjects. Neuropsychologia, 26, 645-659.
Paquier, P. and H. R. Van Dongen (1996). Review of research on the clinical presentation of
acquired childhood aphasia. Ada Neurol Scand, 93, 428-436.
Reynell, J K. (1977). Manual for the Reynell Developmental Language Scales (Revised).
NFER Publishing Company Ltd, Windsor.
Riva, D. and L. Cazzaniga (1986). Late effects of unilateral brain lesions sustained before and
after age one. Neuropsychologia, 24, 423-428.
Satz, P. and C. Bullard-Bates (1981). Acquired Aphasia in children. In: Acquired aphasia (M.
T. Sarno, ed.). Academic Press, New York.
Staudt, M., K. Lidzba, W. Groodd, D. Wildgruber, M. Erb and I. Krageloh-Mann (2002).
Right-hemispheric organization of language following early left sided brain lesion:
Functional MRI topography. Neurolmage, 16, 954-967.
Stiles-Davis, J., J. Janowsky, M. Engel and R.Nass (1988). Drawing ability in four young
children with congenital unilateral brain lesions, Neuropsychologia, 26, 359-371.
Trauner, D. A., R. D. Nass and A. O. Ballantyne (2001). Behavioural profiles of children and
adolescents after prenatal or perinatal unilateral brain damage. Brain, 124, 995-1002.
Van Dongen, H., P. Paquier, W. Creten, J. Borsel and C. Catsman-Berrevoets (2001). Clinical
Evaluation of Conversational Speech Fluency in the Acute Phase of Acquired
Childhood Aphasia: Does a Fluency/Nonfluency Dichotomy Exist? J Child Neurol,
16(5); 345-351.
Van Dongen, H. R., C. B. Loonen and K. J. Van Dongen (1985). Anatomical Basis for
Acquired fluent aphasia in children. Ann Neurol, 17, 306-309.
Van Hout, A. and G. Lyon (1986). Wemicke's Aphasia in a 10-year-old boy. Brain Lang, 9,
268-285.
Van Hout, A. (1991). Characteristics of Language in Acquired Aphasia in Children. In:
Acquired aphasia in children: Acquisition and Breakdown of Language in the
Developing Brain (I. P. Martins, A. Castro-Caldas, H.R. van Dongen and A. van Hout,
eds), pp. 117-124. Kluwer Academic Publishers, Dordrecht.
Vargha-Kadhem, F., E. Isaacs and V. A. Muter (1994). A review of cognitive outcome after
unilateral lesions sustained during childhood. J Child Neurol, 9, 67-73.
Vargha-Kadhem, F., E. B. Isaacs, H. Papaleloudi, C. E. Polkey and J. Wilson (1991).
Development of language in six hemispherectomized patients. Brain, 114, 473-495.
Watamori, T. S., S. Sasanuma and S. Veda (1990). Recovery and plasticity in child-onset
aphasics: Ultimate outcome at adulthood. Aphasiology, 4, 9-30.
Wechsler, D. (1970). Echelle D Intelligence de Wechsler pour adultes. Manuel. Centre de
Psychologie Appliquee, Paris.
Wechsler, D. (1969). Manuel de I'echelle clinique de memoire. Centre de psychologie
appliquee, Paris.
Wechsler, D. (1949). Wechsler Intelligence Scale for children. Manual. New York University
School of Medicine, New York.
Woods, B. T. and S. Carey (1979). Language deficits after apparent clinical recovery from
childhood aphasia. Ann Neurol, 6, 405-409.
253
INDEX
acquired aphasias 26, 27-8, 182-3 focal lesions 53-4

acquired aphasia in children 232-3 Fused Dichotic Words Test 54-5
acquired childhood aphasia 148,
150-2, 160-7 hemispheric lateralization 58-9
acquired speech and language
dysfunction 40-3 Landau-Kleffner Syndrome (LKS) 10, 13-15,
autism 29-30 17-18, 27-8, 72, 74, 80-1
language development 57-8, 133–4,
brain imaging 15-16 136, 137, 138
brain 112-20 language disorders 26
language dominance 165-7
cerebellar aplasia 129-30 language 13, 104
cerebellar malformations 128-9 long-term recovery 203-20
cerebellum 88-90, 128
child aphasia 200, 225-6 NREM sleep 70, 75, 78, 80, 82
childhood epilepsy 10-11,15,16
children 112-20, 182-3 Operculum syndrome 38
chronic aphasia in children 238–4
congenital aphasia 68-9 paroxysmal abnormalities 71, 73, 82
Continuous Spike-Waves during pathophysiology 15-19
Slow Sleep (CSWS) 26, 28-9 phonological processing deficits 44
crossed aphasia 148-9, 152-3, 156-60 phonological short-term
Czech aphasic symptoms 203-9 memory 14-15
po sterior fo s sa 88-90
development dysphasia 80 prognosis 234-5
early brain lesion 66-8 recovery 191, 235-7, 242-4, 245-6

Electrical Status Epilepticus during regression 29-30
Slow-Wave Sleep 26, 28-9
epilepsy 28-9,43–4 tumor 112-20,192

Neurogenic Language Disorders in Children (Fabbro)

Uploaded by

Copyright:

Available Formats

Neurogenic Language Disorders in Children (Fabbro)

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

Neurogenic Language Disorders in Children (Fabbro)

Uploaded by

Copyright:

Available Formats

What books and journals are mentioned related to neurogenic language disorders in children?

What books and journals are mentioned related to neurogenic language disorders in children?

What information is provided about copyright?

What information is provided about copyright?

Neurogenic Language Disorders

Title: The Sciences of Aphasia: From Therapy to Theory

Title: Speech-Language Pathology

Title: Manifestations of Aphasia Symptoms in Different Languages

Title: Foundations of Stuttering

Title: Clinical Phonetics & Phonology: Making Sense of Disordered Speech

Title: Concise Encyclopedia of Language Pathology

Journal of Fluency Disorders

Journal of Communication Disorders

Brain & Language

For further information on these titles, visit www.elsevier.com

Amsterdam - Boston - Heidelberg - London - New York - Oxford - Paris

© 2004 Elsevier Ltd. All rights reserved.

Electronic Storage or Usage

First edition 2004

Library of Congress Cataloging in Publication Data

British Library Cataloguing in Publication Data

2 Pathophysiological Basis of Aphasia and Verbal Outome in Landau- 9-23

3 Acquired Language Disorders and Epilepsy: From Landau -Kleffner 25-35

4 Persistent Subtle Language and Learning Deficits in a Child with 37-48

5 Cerebral Language Lateralization and Early Linguistic In Children 49-63

6 Language Disorders Associated With Paroxysmal Abnormalities 65-85

7 Language and Phonological Awareness Abilities of Children treated 87-126

8 Language Development in Children with Cerebellar Malformations.

9 Crossed Aphasia in Children.

10 Recognizable Spontaneous Language Characteristics in a Young 181 -197

11 Recovery From Aphasia After Polytrauma in a Czech Child: 199-229

12 Persistent Acquired Childhood Aphasia. 231-251

NEUROGENIC LANGUAGE DISORDERS IN

ACQUIRED LANGUAGE DISORDERS IN CHILDREN

Acquired childhood aphasia

Language disorders following PCF lesions

Acquired Epileptiform Aphasias

LANGUAGE DISORDERS DUE TO PRE-PERINATAL LESIONS

DEVELOPMENTAL LANGUAGE DISORDERS

Developmental language disorders (DLDs) — also known as Specific Language Impairment

LANGUAGE DISORDERS IN GENETIC SYNDROMES WITH MENTAL

An important line of research in modern clinical neurolinguistics is the study of language

PATHOPHYSIOLOGICAL BASIS OF APHASIA AND

Keywords: childhood epilepsy, Landau-Kleffner Syndrome, language, phonological short-

CLINICAL FEATURES OF ACQUIRED EPILEPTIC APHASIA

Figure 1 — A Landau-Kleffner Syndrome typical EEG recorded during wakefulness showing

Typical neuropsychological impairments

LATE OUTCOME OF LKS

PATHOPHYSIOLOGICAL BASIS OF EPILEPTIC APHASIA

A focal epileptic dysfunction in the temporal lobe

PATHOPHYSIOLOGICAL BASIS OF VERBAL OUTCOME IN LKS

In this paper, we attempted to provide a provisional pathophysiological account of aphasic

Steve Majerus is a Postdoctoral Researcher at the Fonds National de la Recherche

Neurol, 12, 489-495.

associated with Landau-Kleffner Syndrome: a long-term study of auditory

ACQUIRED LANGUAGE DISORDERS AND

ACQUIRED APHASIA WITH CONVULSIVE DISORDER OR LANDAU-

ELECTRICAL STATUS EPILEPTICUS DURING SLEEP (ESES): CONTINUOUS

AUTISTIC REGRESSION WITH AN EPILEPTIFORM EEG (AREE)

MEDICAL AND SURGICAL MANAGEMENT OF CHILDREN WITH ACQUIRED