Endoscopic third ventriculostomy (ETV)/choroid plexus cauterization (CPC) has become an increasingly common technique for the treatment of infant hydrocephalus. Both flexible and rigid neuroendoscopy can be used, with little empirical evidence directly comparing the two. Therefore, the authors used a propensity score–matched cohort and survival analysis to assess the comparative efficacy of flexible and rigid neuroendoscopy.
Individual data were collected through retrospective review of infants younger than 2 years of age, treated at 1 of 2 hospitals: 1) Boston Children's Hospital, exclusively utilizing flexible neuroendoscopy, and 2) Nicklaus Children's Hospital-Jackson Memorial Hospital, exclusively utilizing rigid neuroendoscopy. Patient characteristics and postoperative outcomes were assessed. A propensity score model was developed to balance patient characteristics in the case mix.
A propensity score model for neuroendoscope type was developed with 5 independent variables: chronological age, sex, hydrocephalus etiology, prior CSF diversion, and prepontine scarring. Propensity score decile-adjusted and 1-to-1 nearest-neighbor matching analysis revealed that compared with flexible neuroendoscopy, rigid neuroendoscopy had an ETV/CPC failure odds ratio (OR) of 1.43 (p = 0.31) and 1.31 (p = 0.47), respectively, compared with an unadjusted OR of 2.40 (p = 0.034). Furthermore, in a Cox regression analysis controlled by propensity score, rigid neuroendoscopy had a hazard ratio (HR) of 1.10 (p = 0.70), compared with an unadjusted HR of 1.61 (p = 0.031).
Although unadjusted analysis suggested worse ETV/CPC outcomes for infants treated by rigid neuroendoscopy, much of the difference could be attributed to the case mix and other predictors of outcome. A larger sample observational study or randomized controlled trials are required to provide evidence-based guidelines on ETV/CPC technique.