OBJECTIVE: Endoscopic third ventriculostomy (ETV) is currently the principal alters native to cerebrospinal fluid shut placement in the management of pediatric hydrocephalus. Cost-effectiveness analysis can help determine the optimal strategy for integrating these different approaches. METHODS: All patients (n = 28) who underwent ETV at British Columbia's Children's Hospital between 1989 and 1998 were matched for age, pathogenesis, and. number of previous shunt procedures, with, patients treated with cerebrospinal fluid shunts. To performs' a cost-effectiveness analysis, hydrocephalus-related resource consumption and outcome (determined asthe'nuimber of hydrocephalus treatment-free days during follow-up) were then retrospectively identified. Cost data were linked to resource use. provide a total cost for all resources used. Costs and outcomes were discounted annually at 5% by standard economic analysis methods. RESULTS: Twenty-four of 28, ETV patients had obstructive hydrocephalus. Over equivalent follow-up periods (median, 35 mo), the ETV success rate (defined by need-for reoperation) was 54%. One hydrocephalus-related death and one hemiparesis occurred in the ETV group. No permanent procedure-related morbidity or mortality was seen in the shunt group. The cost/effect ratios for the two groups were similar. The additional incremental resource use by the shunt group included six readmissions and eight reoperations. ETV mean costs per patient were $10,510 +/- $7628, versus $10,922 +/- $8722 for the shunt group (Canadian dollars for the year 2000). Costs accrued more quickly for the shunt group as time passed. The additional incremental, outcome benefit to the endoscopy group was 86 treatment-free days (3.07 d per patient [95% confidence interval, -7.56 to 13.70 d]). Neither of these differences was, statistically significant. CONCLUSION: In this matched cohort, ETV was not significantly less cosily or more effective over a median 35 months of follw-up, with a 54% initial ETV success rate, even before the additional morbidity 'and' mortality encountered were taken into ' account. The time course for the accrued Cost's suggests that a larger cohort, longer follow-up, or higher success rates are needed to demonstrate the cost-effectiveness,of this therapy.