Congenital hydrocephalus and ventriculoperitoneal shunts: influence of etiology and programmable shunts on revisions Clinical article

Object. Hydrocephalus is a notorious neurosurgical disease that carries the adage once a shunt always a shunt. This study was conducted to review the treatment results of pediatric hydrocephalus. Methods. Pediatric patients who underwent ventriculoperitoneal shunt surgery over the past 14 years were reviewed for shunt revisions. Variables studied included age at shunt placement, revision, or replacement; programmable shunts; infection; obstruction; and diagnosis (congenital, posthemorrhagic, craniospinal dysraphism, and others including trauma, tumors, and infection). Multiple regression analysis methods were used to determine independent risk factors for shunt failure and the number of shunt: revisions. The Kaplan-Meier method of survival analysis was used to compare etiologies on the 5-year survival (revision-free) rate and the median 5-year survival time. Results. A total of 253 patients were studied with an almost equal sex distribution. There were 92 patients with congenital hydrocephalus. 69 with posthemorrhagic hydrocephalus, 48 with craniospinal dysraphism, and 44 with other causes. Programmable shunts were used in 73 patients (other types of shunts were used in 180 patients). A total of 197 patients (78%) underwent revision surgeries due to shunt failures. The mortality rate was 1.6%. Age at first revision, the 5-year survival rate, and the median 5-year survival time were significantly less for both posthemorrhagic and craniospinal dysraphism than for either the congenital or other group (p < 0.05). The failure rate and number of revisions were not significantly reduced with programmable shunts compared with either pressure-controlled or no-valve shunts (p > 0.5). Conclusions. Posthemorrhagic hydrocephalus and craniospinal dysraphism hydrocephalus had significantly earlier revisions than congenital and other etiologies. Programmable systems did not reduce the failure rate or the average number of shunts revisions. (DOI: 10.3171/2009.7.PEDS08371)