Analysis of Evolution of Hydrocephalus in Posterior Fossa Tumors and Validation Study of the Modified Canadian Preoperative Prediction Rule for Hydrocephalus in Children and Frankfurt Grading System for Prediction of Cerebrospinal Fluid Diversion in Adults with Posterior Fossa Tumors

Abstract

We sought to analyze the evolution of hydrocephalus and enumerate its predictive factors in posterior fossa tumors in children and adults. We also validated the modified Canadian Preoperative Prediction Rule for Hydrocephalus (mCPPRH) and Frankfurt grading systems as tools to predict cerebrospinal fluid (CSF) diversion in children and adults, respectively. A retrospective review of patients with posterior fossa tumors operated between 2012 and 2019 was performed. Clinical, radiologic, and operative data were obtained. Validation was performed for both scores via receiver operating characteristic (ROC) curves and evaluation of the area under the curve (AUC). We included 116 children and 343 adults in the study. Of the adults, 141 patients had intraaxial tumors and 172 had extraaxial tumors. The insertion of external ventricular drain (EVD), its duration, papilledema was noted to have significant influence on the need for permanent CSF diversion (P < 0.05) in children. The ROC for mCPPRH score was 0.659 (0.501-0.816), Age- 0.496 (0.334-0.658) and Evans index- 0.788 (0.654-0.922). In adults- Intraaxial tumours Age, Frankfurt score, duration of EVD, Diagnosis, Extent of resection and periventricular capping significant predictors and ROC age AUC 0.300 (0.193-0.407), Evans index 0.939 (0.888-0.990), and Frankfurt score 0.908 (0.853-0.964) (P < 0.05), whereas in extraaxial tumors Frankfurt grading, sex, duration of EVD, presence of perilesional edema and extent of resection (P < 0.05) ROC age AUC 0.439 (0.344-0.534), Evans index 0.941 (0.906-0.977), and Frankfurt score 0.847 (0.782-0.912). This is the first external validation study for the 2 predictive systems in use. mCPPRH demonstrated poor predictive accuracy, and Frankfurt grading system demonstrated good accuracy. EVD insertion and its duration was significantly predictive of the need for permanent CSF diversion.