[Treatment of intracranial cysts in children: peritoneal derivation or endoscopic fenestration?].

OBJECTIVE

The goal of this study is to evaluate the indications, benefit and complications of shunts and endoscopic fenestrations in the treatment of malformative intracranial cysts.

MATERIAL AND METHOD

The records of 172 consecutive children (mean age of 4 years) were reviewed. All had a malformative cyst. Dandy Walker malformation, mega cisterna magna, and cysts from tumoral or porencephalic origin were excluded from the study. The cysts were diagnosed either in utero (n=64) or postnatally (n=108). Most of them were unique (94.8%) and localized in the posterior fossa (26.2%) or at the convexity (23.2%). Indication for surgery was based on clinical symptoms (n=101; 86.3%) or size of the lesion (n=16; 13.7%). Endoscopy was the treatment of choice when cysts were in closed relationship with enlarged ventricles. Shunting procedures were indicated when endoscopy was not feasible and craniotomies when shunt insertion was unsafe or diagnosis uncertain. Fifty children underwent an endoscopic fenestration, 55 a shunting procedure, 7 the puncture or the external drainage of a pericerebral collection and 5 a direct surgical approach. The mean follow-up was 5.5 years. Psycho-motor, intellectual and school performances were evaluated in 93 children (54%). Success was defined by both the disappearance of symptoms of increased intra cranial pressure and regression of the cyst.

RESULTS

Compared to shunts, endoscopic fenestrations were more frequently successful (70% vs 61.8%), led to less complications (6% vs 61.8%) and to a lesser number of reoperations (in average 1.6 operation per child vs 2.2). Median developmental and intellectual quotients for the whole series were respectively 98 and 97 and did not depend upon the type of treatment.

CONCLUSION

The study of this series shows that treatment modalities necessarily vary according to the site of the cysts but that endoscopic fenestrations are preferable to shunts whenever feasible.