Endoscopic Third Ventriculostomy Vs Cerebrospinal Fluid Shunt in the Treatment of Hydrocephalus in Children: A Propensity Score-Adjusted Analysis

BACKGROUND: Endoscopic third ventriculostomy (ETV) has preferentially been offered to patients with more favorable prognostic features compared with shunt. OBJECTIVE: To use advanced statistical methods to adjust for treatment selection bias to determine whether ETV survival is superior to shunt survival once the bias of patient-related prognostic factors is removed. METHODS: An international cohort of children (<= 19 years of age) with newly diagnosed hydrocephalus treated with ETV (n = 489) or shunt (n = 720) was analyzed. We used propensity score adjustment techniques to account for 2 important patient prognostic factors: age and cause of hydrocephalus. Cox regression survival analysis was performed to compare time-to-treatment failure in an unadjusted model and 3 propensity score-adjusted models, each of which would adjust for the imbalance in prognostic factors. RESULTS: In the unadjusted Cox model, the ETV failure rate was lower than the shunt failure rate from the immediate postoperative phase and became even more favorable with longer duration from surgery. Once patient prognostic factors were corrected for in the 3 adjusted models, however, the early failure rate for ETV was higher than that for shunt. It was only after about 3 months after surgery did the ETV failure rate become lower than the shunt failure rate. CONCLUSIONS: The relative risk of ETV failure is initially higher than that for shunt, but after about 3 months, the relative risk becomes progressively lower for ETV. Therefore, after the early high-risk period of ETV failure, a patient could experience a long-term treatment survival advantage compared with shunt. It might take several years, however, to realize this benefit.