Association of Recorded Cerebrospinal Fluid Diversion with Registry Outcome Status in Pediatric Medulloblastoma: An Exploratory 10-Year Single-Centre Cohort Study

Objectives. To examine, in an exploratory institutional cohort, whether a registry record of cerebrospinal fluid (CSF) diversion was associated with the available binary registry outcome status in children with medulloblastoma, while distinguishing this intervention-related variable from radiographic hydrocephalus. Methods. We performed a retrospective single-institution cohort study of 85 patients younger than 18 years with histologically confirmed medulloblastoma who underwent surgical treatment at the Neurosurgery Unit of Bagdasar-Arseni Emergency Hospital. The exposure was any source-record indication of external ventricular drainage, shunting, or ventriculocisternostomy. The analytic file did not provide the timing, indication, or standardized type of diversion; therefore, the exposure could not be characterized as occurring at presentation or before tumor resection. The outcome was the database field labelled “five-year survival,” coded as alive or death/event. Dates of diagnosis, death, progression, and last follow-up were unavailable, so uniform five-year ascertainment and the meaning of “event” could not be verified. Descriptive analyses, univariable comparisons, and a parsimonious logistic model including recorded CSF diversion and metastatic disease were performed. Results. The median age was 6 years (IQR, 3–11), and 47 patients were male (55.3%). Hydrocephalus was recorded in 67 patients (78.8%), any CSF diversion in 40 (47.1%), metastatic disease in 12 (14.1%), and gross-total resection in 64 (75.3%). The registry outcome was recorded as alive in 56 patients (65.9%) and death/event in 29 (34.1%). The recorded death/event category occurred in 24/40 patients with diversion versus 5/45 without diversion (60.0% vs. 11.1%; p < 0.001), and in 9/12 patients with metastasis versus 20/73 without metastasis (75.0% vs. 27.4%; p = 0.002). In the mutually adjusted exploratory model, recorded CSF diversion was associated with the death/event category (OR, 12.71; 95% CI, 3.75–43.05; p < 0.001), as was metastatic disease (OR, 8.93; 95% CI, 1.66–47.91; p = 0.011). Conclusions. Recorded CSF diversion was associated with adverse registry outcome status after mutual adjustment for metastatic disease. Given uncertainty regarding timing and indication, possible outcome misclassification, the small number of recorded events, and absent clinical, treatment, and molecular covariates, this finding should be interpreted as an exploratory institutional association that may reflect disease severity and management selection, not procedural causality or validated prognostic value.