Endoscopic third ventriculostomy with choroid plexus cauterization (ETV/CPC) offers an alternative to shunt placement for infant hydrocephalus, avoiding shunt-related complications when successful. However, approximately 40% of procedures fail, with 30% ultimately requiring shunt placement. Current risk stratification relies primarily on age and prior interventions, limiting ability to optimize treatment selection. Therefore, this study aimed to determine whether preoperative cerebrospinal fluid volume (CSFV) measurements independently predict shunt-dependence following ETV/CPC in infant hydrocephalus. This retrospective cohort study was conducted from 2008 to 2024, with follow-up through 2024 and statistical analysis performed in January 2025. Infants who underwent ETV/CPC as the primary treatment for hydrocephalus and had magnetic resonance imaging within two weeks before surgery were included. The primary outcome was shunt-dependence, defined as progressive ventriculomegaly requiring shunt placement after ETV/CPC. Imaging measures included traditional frontal-occipital horn ratio (FOHR) and novel volumetric assessments: CSFV and CSFV/intracranial volume (ICV) ratio estimated using deep learning-based segmentation methods and the corresponding z-scores using age- and sex- adjusted normative curves. Cox proportional hazards regression identified independent predictors of shunt-dependence. A total of 188 infants (median age at surgery 60 days, 42.5% female) were included. Sixty-six (35.8%) patients became shunt-dependent over a median follow-up of 3.8 years as of January 2025. In multivariable analysis adjusting for age and prior temporary CSF diversion, both CSFV z-score (HR 1.10, 95%CI 1.02–1.20, p = 0.02) and CSFV/ICV ratio z-score (HR 1.07, 95%CI 1.01–1.13, p = 0.02) independently predicted shunt-dependence, while FOHR did not achieve statistical significance in adjusted models. Younger infants with higher CSFV z-scores demonstrated elevated failure rates. Preoperative CSF volume measurements provide superior risk stratification compared to traditional linear assessments for predicting shunt-dependence in infant hydrocephalus. These findings support integration of volumetric brain imaging analysis into clinical decision-making to optimize treatment selection and improve outcomes through more precise identification of candidates most likely to benefit from ETV/CPC versus primary shunt placement.
Yoshikawa et al. (2026) studied this question.