REVIEW ARTICLE

Decompressive craniectomy:
indications and techniques
Craniotomia descompressiva: indicações e técnicas
Rodrigo Moreira Faleiro¹, Luanna Rocha Vieira Martins²

DOI: 10.5935/2238-3182.20140143

ABSTRACT
1
MD. Coordinator of the Neurology and Neurosurgery Introduction: Decompressive craniectomy (DC) is a surgical method indicated for im-
Service at the João XXIII Emergency Room Hospital in
Belo Horizonte, MG – Brazil mediate reduction of intracranial pressure (ICP) in general facing brain swelling, acute
2
Medical School student at the Medical School subdural hematoma, and some non-traumatic diseases. The technique consists of crani-
of the Federal University of Minas Gerais – UFMG.
Belo Horizonte, MG – Brazil. ectomy and expansion of the dura-mater to accommodate the swollen brain. Literature
review: several studies with DC indications show distinct results. There is no evidence
to justify its routine use in adults with serious traumatic brain injury (TBI); however, it is
indicated in children. Discussion: the refractoriness of intracranial hypertension (ICH)
towards clinical measures is an indirect finding of cerebral auto-regulation loss, which
configures the presence of Brain Swelling (BS), or swelling of the brain. DC is indicated
in cases of BS, being bi- (Marshall III) or unilateral (Marshall IV). The technique requires
broad bone removal, greater than the usually used in bruises and contusions. The incision
in ‘T’- Kemp is preferred for allowing broad access to the frontotemporoparietal regions
and great bone decompression. After craniectomy, dural diaeresis followed by duroplastia
is conducted and, finally with the replacement of the bone flap to its original location after
bulging reduction. Conclusion: DC is effective in immediately reducing ICP, avoiding ICH,
which is a factor for poor prognosis. Its usefulness is especially observed if carried out
prematurely in TBI victims with ICH when compared to being carried out late and with
clinical measures of ICP control such as barbiturate coma and hypothermia.
Key words: Craniocerebral Trauma; Craniotomy; Intracranial Hypertension.

RESUMO

Introdução: a craniotomia descompressiva (CD) é método cirúrgico indicado para a

redução imediata da pressão intracraniana (PIC) em geral, diante de tumefação cerebral,
hematoma subdural agudo e algumas doenças não traumáticas. A técnica consiste em
craniotomia e ampliação da dura-máter para se acomodar o cérebro tumefeito. Revisão
da literatura: vários estudos com indicações de CD apresentam resultados distintos entre
si. Não existem evidências que justifiquem sua aplicação de rotina em adultos com trau-
matismo cranioencefálico (TCE) grave, mas em crianças parece que sim. Discussão: a
refratariedade da hipertensão intracraniana (HIC) às medidas clínicas é achado indireto
de perda da autorregulação cerebral, o que configura a presença de um Brain Swelling
(BS), ou tumefação cerebral. A CD está indicada nos casos de BS, seja bi (Marshall III)
ou unilateral (Marshall IV). A técnica do procedimento requer a retirada óssea ampla,
Submitted: 2014/09/07 maior que as usualmente utilizadas nos hematomas e contusões, sendo preferencial
Approved: 2014/11/10
a incisão em ‘T’- Kemp, por permitir acesso amplo às regiões frontotemporoparietal e
Institution: grande descompressão óssea. Após a craniotomia, procede-se à diérese dural seguida
João XXIII Emergency Room Hospital
Belo Horizonte, MG – Brazil
de duroplastia e, por fim, recolocação do flap ósseo ao seu local original, após redução
do abaulamento. Conclusão: a CD é eficaz em reduzir imediatamente a PIC, evitando-se
Corresponding Author:
Rodrigo Moreira Faleiro
a HIC, que é fator de mau prognóstico. Sua utilidade é especialmente observada se real-
E-mail: r.m.faleiro@hotmail.com izada precocemente em vítimas de TCE e com HIC quando comparada à sua realização

492 Rev Med Minas Gerais 2014; 24(4): 492-497

 Decompressive craniectomy: indications and techniques

tardia e com medidas clínicas de controle da PIC, como

coma barbitúrico e hipotermia. group. The surgical group showed a statistically sig-
nificant improvement compared to the conservative
Palavras-chave: Traumatismos Craniocerebrais; Crani-
otomia; Hipertensão Intracraniana. group control. The pediatric population evolved bet-
ter and those operated after 48 hours evolved unfavor-
ably. The mortality rate was reported as 23%, and fa-
INTRODUCTION vorable evolution as 37%. The procedure is indicated
early in patients with ICP of less than 40 mmHg and
Decompressive craniectomy (DC) is a surgical with more than three points on the Glasgow Coma
method used for the immediate reduction of intracra- Scale (GCS). We highlight the episode of 10 patients
nial pressure (ICP). It is generally indicated in cases of (28.5%) who developed hydrocephalus, dependent
brain swelling, acute subdural hematoma (ASDH) and of ventricular derivation, after the procedure.
non-traumatic diseases.1 It consists of craniectomy and Guerra et al.12, in 1999, evaluated 57 patients un-
expansion of the dura mater to accommodate the swol- dergoing DC (26 patients with bilateral craniectomy
len brain. The bone can be temporarily housed in the and 31 with unilateral craniectomy). They obtained
abdominal subcutaneous tissue or be ignored until a better results than the previous series, probably be-
subsequent cranioplasty with acrylic. In these cases, cause of a rigorous selection of candidates and early
the correct term would be decompressive craniectomy. performance of DC. They reported a mortality rate of
Conducts in relation to trauma were systematized 19%, with good functional rehabilitation in 58%. ECG
in 1996 by classifying them according to scientific and ICP in the first postoperative day were the most
evidences in guidelines or options for assistance in important prognostic indicators. The complications
traumatic brain injury (TBI).2 These guidelines were reported were: 14% hydrocephalus, 26% hygroma,
updated in 2000, with DC being suggested as the sec- 7% post-operative seizures, and 2% infection. There
ond option (second tier) for the treatment of intracra- was no mortality directly related to the technique.
nial hypertension (ICH), however, without validation Therefore, they indicate DC in patients younger than
by class I studies. 50 years old, with signs of brain swelling, without pri-
This technique was widely used in the 70s and mary irreversible lesion, and with refractory ICH or
80s, however, with disappointing results. It was ap- responsible for neurological worsening.
plied only when all clinical attempts to reduce ICH Munch et al.16 retrospectively evaluated 49 patients
clinics had already failed.3,4 The control of ICH was undergoing unilateral DC by pairing them with a control
tried through clinical measures (barbiturate coma, group of barbiturate coma from the TCDB. The prog-
hypothermia), which also showed reduced effective- nosis (Glasgow Outcome Score-GOS) at ICU discharge
ness and frequent clinical complications.5-9 was similar in both groups, however, with significant
Recently, studies have been assessing the effective- improvement in the surgical group when evaluated af-
ness of DC when applied early in patients who are vic- ter six months. An improved evolution was observed in
tims of severe TBI with ICH, showing superior results patients under the age of 50 years and operated early in
in late application or when compared to specific clini- the surgical group. The appearance of the mesencephal-
cal measures (barbiturate coma and hypothermia).10-14 ic cistern in brain injury correlated with the prognosis,
even when the midline deviation was unchanged. They
obtained 20% of good results, with 33% mortality.
REVIEW OF THE LITERATURE Faleiro et al.17 evaluated prognostic factors and
complications in 89 patients undergoing unilateral DC
There is no doubt that DC is an effective technique in closed brain injury cases. The statistically significant
to immediately reduce ICP, avoiding ICH, which in it- prognostic factor was the ECG on admission, and the
self is a bad prognostic factor. major complications were liquor malabsorption, such as
Polin et al.15, in an attempt to a more homogeneous hydrocephalus (7.9%) and subdural collection (11.2%).
comparison between the two groups, retrospectively The Cochrane systematic review concluded that
studied 35 patients with refractory brain swelling and there is no scientific evidence to justify the routine
undergoing bi-frontal DC. The authors used patients implementation of DC in adults with severe TBI, em-
treated with barbiturate coma extracted from the phasizing only the study by Taylor, in 2001, with posi-
Traumatic Coma Data Bank (TCDB) as the control tive evidence in the pediatric population.18,19

Rev Med Minas Gerais 2014; 24(4): 492-497 493

Decompressive craniectomy: indications and techniques

DECRA or venous. The cerebral self-regulation needs to be

intact so that such measures, such as hyperventila-
The DECRA was a multicenter study (15 tertiary tion and hypertonic solution or mannitol exert their
hospitals) that randomized 155 patients to evalu- effects. The refractoriness of ICH to the clinical mea-
ate the effectiveness of bi-frontal craniectomy (BC) sures is an indirect finding of self-regulation loss, i.e.
in relation to the prognosis (modified GOS). These Brain Swelling (BS) or swelling of the brain occurs.
patients were randomized for BC or conventional DC is indicated in cases of BS, be it bilateral
clinical treatment if they showed refractory ICH (> 20 (Marshall III) or unilateral (Marshall IV). This patho-
mmHg for more than 15 minutes). In summary, the physiological concept also helps to understand the
study showed that in the BC group, the time of me- appropriate time to complete DC. See Marshall clas-
chanical ventilation and ICU stay were shorter, how- sification (Table 1).
ever, when evaluated belatedly, these patients were
functionally worst (modified GOS) than those in the Table 1 - Marshall classification
conservative treatment group. Type Tomographic description Prognosis
This study should be evaluated with caution because I Tomography without visible pathologies. 9.6% mortality
its definition of refractory ICH (20 mmHg for 15 minutes) Cisterns present; midline deviation 0-5
II 13.5% mortality
is fairly low; current centers tolerate ICP up to 25 mmHg, mm; absence of lesions > 25 ml.

especially if the cerebral perfusion pressure (CPP) is at Compressed cisterns or absent;

III midline deviation 0-5 mm; 34% mortality
optimal levels. Therefore, many patients received BC absence of lesions > 25 ml.
when it was not needed. Midline deviation > 5 mm;
IV 56.2% mortality
Another criticism lies on the chosen technique. absence of lesions > 25 ml.
The authors describe a similar technique to that used Source: Marshall SB, Marshall LF, Klauber MR et al. The new classification of
head injury based on computerized tomography. J Neurosurg 1991; 75: S14-S20.
in the study by Polin et al.18 but without cutting the
brain scythe. It is known that BC is related to poor
functional results, especially when the brain scythe is On the Marshall IV CT (Figure 1), there is unilat-
cut, because the cerebral frontal lobes advance and eral loss of self-regulation and any clinical measure-
are harmed against the dural fold with patients becom- ment that act in the blood compartment will act in the
ing frontalized lately. The study shows that BC is not healthy hemisphere, worsening the midline deviation
a good technique to be used, and therefore, bilateral (MLD). In these cases, the DC conduct is early, with
craniectomy is preferable to bifrontal craniectomy. the placement of an ICP monitor after the procedure,
A careful analysis of the statistical data shows that and patient transfer to the ICU.
18% of patients in the conservative group were craniec-
tomized as a life-saving measure; however, their results
were analyzed as conservative. More importantly, 27%
of patients from the craniectomy group had bilateral
mydriasis against only 12% in the conservative group.
There was no statistical difference in long-term func-
tionality between the two groups when the patients
with mydriasis were removed from both groups.
Unfortunately, DECRA did not respond to the
main issues related to DC.

DISCUSSION

Indications

Most clinical measures for the treatment of ICH Figure 1 - Marshall IV (MLD = 6 mm).
act in the intracranial blood compartment – arterial Source: author’s collection.

494 Rev Med Minas Gerais 2014; 24(4): 492-497

 Decompressive craniectomy: indications and techniques

In the Marshall III CT (Figure 2), the MLD is less Technique

than 5 mm, however, there is a closure of the basal
cisterns and lateral ventricles. These patients should There are two types of incision described in the
receive ICP monitoring measures and ICH aggres- literature: one as an inverted question mark, or Beck-
sive treatment in the ICU. In cases of treatment re- er type, also known as trauma flap for being used in
fractoriness, one must resort to DC, which can be the drainage of most hematomas or post-traumatic
unilateral in cases of MLD. In cases where there is contusions. In the decompression technique, a wide
neither the least deviation in the pellucidum sep- bone removal is indicated (12 to 14 cm in its great-
tum, the bilateral DC is chosen. This technique is est diameter), greater than those usually applied in
preferred over the bifrontal because it is associated hematomas and contusions. The occurrence of skin
to worsen functional prognosis as a consequence of necrosis is not rare in the attempt to increase this inci-
frontalized patients or akinetic mutism, especially sion after the auricle (Figure 3).
when the brain scythe is not cut (DECRA).

Figure 3 - Skin necrosis in Becker type incision.

Source: author’s collection.

Figure 2 - Marshall III (MLD < 5 mm and base cisterns

erased). For this reason, the incision in ‘T’ is preferred, de-
Source: author’s collection.
scribed for hemispherectomies by Ludwig G. Kempe.
This type of incision ensures secure access to fronto-
There are two types of patients with ASDH. The temporoparietal regions, allowing wide bone decom-
first refers to the young, trauma victim with great ac- pression.4 Special attention should be given to the
celeration-deceleration. A laminar hematoma is usu- removal of bones, up to the base of the middle fossa,
ally found in these cases, with MLD disproportion- because this measure will enable the opening of the
ate to the hematoma size. This suggests unilateral BS basal cistern.
and indication for DC. The other type is concerned Once craniectomy is performed, the dural di-
an older patient, alcoholic, or with blood dyscra- aeresis is followed by duroplasty. There are sev-
sia, with low kinetic energy trauma (fall from own eral incision and dura mater plasty techniques. At
height, stairs), who develops voluminous hematoma this point, it should be emphasized that whatever
by cerebral atrophy and great MLD, proportional to technique is employed, it must allow homogeneous
the hematoma size. In these cases, craniectomy is expansion of the brain tissue. The pericranium is
indicated for the hematoma drainage with replace- used in the duroplasty. The hermetic dural suture
ment of the bone flap in its original position. with prolene 4 or 5 thread is preferable (Figure 5).

Rev Med Minas Gerais 2014; 24(4): 492-497 495

Decompressive craniectomy: indications and techniques

cranioplasty, the lower the incidence of hydrocepha-

lus or subdural collections, in addition to a proven im-
provement in the subjacent cerebral cortex function.21

Figure 4 - Broad craniectomy with resection of the

temporal bar. Figura 6 - Umbilical incision.
Source: author’s collection. Source: author’s collection.

CONCLUSION

Hence, it is evident that DC is an effective tech-

nique to immediately reduce the ICP, preventing ICH,
which is a poor prognosis factor itself. It applies to pa-
tients with refractory clinical measures ICH (Marshall
III) or those who arrive with major midline deviation
(Marshall IV). The surgery should be avoided in pa-
tients with three points in the ECG, no reflections in
the brainstem, and cranial computed tomography
showing irreversible lesions.
The main complications from the technique are
Figure 5 - Hermetic duroplasty. liquor malabsorption (such as hydrocephalus and
Source: author’s collection.
subdural collection), hygroma, postoperative convul-
sion, and infection.
The detection of hydrocephalus or subdural collec- The early performance of DC in victims with se-
tions after DC is frequent, and the hermetic dural vere TBI and ICH seems to be associated with superior
suture decreases the incidence of liquor fistula.17,20 results compared to its late application or specific clin-
Finally, the bone must be stored in the abdominal ical measures (barbiturate coma and hypothermia).
subcutaneous tissue. The transversal infraumbilical
incision, Pfannestiel type, was initially performed for
that purpose, however, it was observed that a direct REFERENCES
contact with the diaper and secretions could be as-
sociated with high rates of infection, which made the 1. Carter BS, Ogilvy CS, Candia GJ, Rosas HD, Buonanno F. One year
outcome after decompressive surgery for massive nondominant
paraumbilical incision preferred (Figure 6).
hemisferic infarction. Neurosurgery. 1997; 40:1168-76.
The bone flap should be replaced in its original lo-
2. Bullock R, Chessnut RM, Clifton G, Ghajar J, Marion DW, Naray-
cation as soon a reduction in bulging, in the craniecto-
an RK, et al. Guidelines for the management of severe traumat-
my area, is observed. It is believed that the earlier the

496 Rev Med Minas Gerais 2014; 24(4): 492-497

 Decompressive craniectomy: indications and techniques

ic brain injury. The Brain Trauma Foundation (New York), The 12. Guerra WKW, Gaab MR, Dietz H, Mueller JU, Piek J, Fritsch MJ. Sur-
American Association of Neurological Surgeons (Park Ridge, gical decompression for traumatic brain sweelling: indications
Illinois), and The Joint Section of Neurotrauma and Critical and results. J Neurosurg. 1999; 90:187-96.
Care; 1995. 13. Clifton GL, Miller ER, Choi SC, Levin HS, McCauley S, Smith KR, et
3. Ransohoff J, Benjamin MV, Epstein F. Hemicraniectomy for acute al. Lack of effect of induction of hypothermia after acute brain
subdural hematoma. J Neurosurg. 1971; 34:70-1. injury. N. Engl J Méd 2001; 344:556-63.
4. Seelig JM, Becker DP, Miller JD, Greenberg RP, Ward JD, Choi SC. 14. Narayan RK. Hypothermia for traumatic brain injury: a good
Traumatic Acute Subdural Hematoma: Major mortality reduc- idea proved ineffective. N Engl J Med. 2001; 344:602-3.
tion in comatose patients treated within four hours. N Engl J 15. Polin RS, Shaffrey ME, Bogaev CA, Tisdale N, Germanson T, Boc-
Med. 1981; 304:1511-8. chicchio B, et al. Decompressive bifrontal craniectomy in the
5. Da Silva JA, Da Silva CE, Sousa MB. Decompression craniotomy treatment of severe refractory posttraumatic cerebral edema.
in acute brain edema: Apropos of 30 operated cases. Arq Neu- Neurosurgery. 1997; 41(1):84-94.
ropsiquiatr. 1976 34: 232-40. 16. Munch E, Horn P, Schurer L, Piepgras A, Paul T, Schmiedek P. Man-
6. Pereira WC, Neves VJ, Rodrigues Y. Bifrontal decompressive crani- agement of traumatic brain injury by decompressive craniec-
otomy as the treatment for severe cerebral edema. Arq Neurop- tomy. Neurosurgery. 2000; 47(2):315-22.
siquiatr. 1977; 35:99-111. 17. Faleiro RM, Faleiro LCM, Caetano E, Gomide I, Pita C, Coelho G,
7. Eisenberg H, Frankowski R, Contant C. Comprehensive central et al. Decompressive Craniotomy. Prognostic factors and compli-
nervous system trauma centers: High-dose barbiturate control cations in 89 patients. Arq Neuropsiquiatr. 2008; 66(2-B):369-73.
of elevated intracranial pressure in patients with severe head 18. Sahuquillo J. Decompressive craniectomy for the treatment of
injury. J Neurosurg. 1988; 69:15-23. refractory high intracranial pressure in traumatic brain injury.
8. Schalen W, Messeter K, Nordstrom CH. Complications and side Cochrane Database of Systematic Reviews. 2006; CD003983(1).
effects during thiopentone theraphy in patients with severe 19. Taylor A, Butt W, Rosenfeld J, Shann F, Ditchfield M, Lewis E, et al.
head injuries. Acta Anaesthesiol Scand. 1992; 36:369-77. A randomized trial of very early decompressive craniectomy in
9. Shiozaki T, Sugimoto H, Taneda M, Yoshida H, Iwai A, Yoshioka T, children with traumatic brain injury and sustained intracranial
Sugimoto T. Effect of mild hypothermia on uncontrollable intra- hypertension. Childs Nerv Syst. 2001; 17:154-62.
cranial hypertension after severe head injury. J Neurosurg. 1993; 20. Adamo MA, Drazin D, Waldman JB. Decompressive craniectomy
79:363-8. and postoperative complication management in infants and
10. Gower DJ, Lee Stuart, McWhorter JM. Role of subtemporal de- toddlers with severe traumatic brain injuries. J Neurosurg Pedi-
compression in severe closed head injury. Neurosurgery. 1988; atr. 2009; 3:334-9.
23:417-22. 21. Aarabi B, Chesler D, Maulucci C, Blacklock T, Alexander M. Dy-
11. Whitfield P, Guazzo E. ICP reduction following decompressive namics of subdural hygroma following decompressive craniec-
craniotomy. Stroke. 1995; 26:1125-6. tomy: a comparative study. Neurosurg Focus. 2009; 26(6):E8.

Rev Med Minas Gerais 2014; 24(4): 492-497 497