Validating the modified Canadian Preoperative Prediction Rule for Hydrocephalus for accurate hydrocephalus prediction in a statewide pediatric brain tumor cohort.

purposeThe modified Canadian Preoperative Prediction Rule for Hydrocephalus (mCPPRH) was developed to predict the need for permanent CSF diversion in children with posterior fossa tumors (PFT). This study aimed to externally validate the mCPPRH in a cohort of 113 pediatric patients with PFTs.

methodsWe conducted a retrospective analysis, calculating the mCPPRH score for each patient and performing receiver operating characteristic (ROC) curve analysis to assess the tool's discriminative ability. Sensitivity, specificity, predictive values, and likelihood ratios were calculated using a cutoff score of ≥ 5. Multivariable logistic regression with bidirectional stepwise selection was used to evaluate individual components of the score. The mCPPRH components were modified and the performance of adjusted tools was compared to the original.

resultsOf the 113 patients, 35 (31.0%) required permanent CSF diversion. The mCPPRH demonstrated acceptable discriminative ability (AUC = 0.701, 95% CI 0.608-0.795, p < 0.0003). Sensitivity was 34.1%, specificity 89.7%, positive predictive value 60%, and negative predictive value 75.3%. Initial regression identified no significant predictors. In stepwise regression, moderate-severe hydrocephalus independently predicted permanent CSF diversion (OR 6.37, 95% CI 1.71-41.55, p = 0.02). Increasing the age cutoff to < 5 years, removing tumor diagnosis, and modifying hydrocephalus weighting improved performance (AUC = 0.768, sensitivity 71.4%, specificity 75.6%).

conclusionsThe mCPPRH demonstrates acceptable discriminative ability (AUC 0.701) in our cohort, with particular utility in identifying low-risk patients. However, its poor sensitivity (34.1%) and variable predictor performance suggest that additional clinical factors should be considered for treatment planning, particularly in higher-risk cases. Further modification of mCPPRH components is suggested to improve its utility.