OBJECTIVE Endoscopic third ventriculostomy (ETV) is a surgical alternative to placing a CSF shunt in certain patients with hydrocephalus. The ETV Success Score (ETVSS) is a reliable, simple method to estimate the success of the procedure by 6 months of postoperative follow-up. The highest score is 90, estimating a 90% chance of the ETV effectively treating hydrocephalus without requiring a shunt. Treatment with ETV fails in certain patients, despite their being the theoretically best candidates for the procedure. In this study the authors attempted to identify factors that further predicted success in patients with the highest ETVSSs. METHODS A retrospective review was performed of all patients treated with ETV at 3 institutions. Demographic, radiological, and clinical data were recorded. All patients by definition were older than 1 year, had obstructive hydrocephalus, and did not have a prior shunt. Failure of ETV was defined as the need for a shunt by 1 year. The ETV was considered a success if the patient did not require another surgery (either shunt placement or a repeat endoscopic procedure) by 1 year. A statistical analysis was performed to identify factors associated with success or failure. RESULTS Fifty-nine patients met the entry criteria for the study. Eleven patients (18.6%) required further surgery by 1 year. All of these patients received a shunt. The presenting symptom of lethargy statistically correlated with success (p = 0.0126, odds ratio [OR] = 0.072). The preoperative radiological finding of transependymal flow (p = 0.0375, OR 0.158) correlated with success. A postoperative larger maximum width of the third ventricle correlated with failure (p = 0.0265). CONCLUSIONS The preoperative findings of lethargy and transependymal flow statistically correlated with success. This suggests that the best candidates for ETV are those with a relatively acute elevation of intracranial pressure. Cases without these findings may represent the failures in this highly selected group.