OBJECTIVE: This is a retrospective study to identify risk factors for failure in the treatment of obstructive hydrocephalus with endoscopic third ventriculostomy (ETV).
METHODS: The records for 89 patients, including 32 with ventriculoperitoneal or ventriculoatrial shunt malfunctions or infections, who underwent ETVs between 1993 and 1998, at our institution, were examined. Multiple variables possibly related to failure were considered. These included age, sex, cause of hydrocephalus, presence and function of ventriculoperitoneal/ventriculoatrial shunts, history of shunt revisions or infections, symptoms, preoperative imaging results, presence of retained shunt catheters, postoperative meningitis, and postoperative ventricular size.
RESULTS: Twenty-nine patients (32.6%) required subsequent shunt replacement and/or ETV revision. Of these 29 reoperations, 12 procedures (41.4%) were performed within 2 weeks and only 3 were performed more than 10 months after the initial ETV procedure. The ventricular size remained unchanged in 75% of the cases on the day after ETV, in 57.4% at 3 months, in 48.2% at 6 months, and in 41.8% at 1 year. Cine phase-contrast magnetic resonance imaging findings were consistent with postoperative symptomatic resolution in 96.3% of the cases. Seven patients (7.9%) experienced complications related to ETV, all of which were transient. Significant risk factors in univariate analyses were as follows: presence of Chiari Type I malformation (P = 0.003), shunt infection at presentation (P = 0.014), history of shunt infections (P = 0.0004), three or more previous shunt revisions (P = 0.0018), and postoperative meningitis (P = 0.0001). Late-onset idiopathic aqueductal stenosis was a significant predictor of good outcomes (P = 0.044). These factors were reanalyzed in a multivariate analysis, which confirmed a history of shunt infections and postoperative meningitis as independent risk factors.
CONCLUSION: The risk of failure increases with intracerebral infection, likely because of obliteration of cerebrospinal fluid pathways.