Prognostic factors for cerebellar astrocytomas in children: A study of 102 cases

Objective: To evaluate the prognostic factors influencing the length of survival of pediatric patients with cerebellar astrocytomas. Methods: The clinical data of 102 patients under the age of 12 years with cerebellar astrocytomas were retrospectively analyzed. The radiological features, surgical findings, histology and adjuvant radiotherapy were reviewed. Patients were followed up on an outpatient basis. Computerized tomography (CT) scans of the brain were performed to detect recurrence. Results: The age of the patients at presentation varied from 10 months to 12 years. The mean age at presentation was 7 years and 11 months. The average duration of symptoms was 5.8 months. The clinical features were predominantly related to intracranial hypertension and the location of the tumor. Twenty-six tumors were located in the vermis and 76 in the cerebellar hemisphere. The brain stem was involved in 20 patients. All 102 patients had a preoperative contrast-enhanced CT scan. Midline vermian tumors were predominantly solid and enhancing, whilst the hemispheric tumors were cystic and nonenhancing. The tumors were graded into three subgroups based on histological characteristics: pilocytic astrocytoma, low-grade fibrillary astrocytoma and high-grade fibrillary astrocytoma. Total, radical and subtotal excision, as determined by the surgical impression, were performed in 82, 16 and 4 patients, respectively. Postoperative radiotherapy was administered to 12 patients. CSF diversion was carried out in 11 of 79 cases who had moderate to severe hydrocephalus. Conclusions. Our study has shown that the location of the tumor (p < 0.05), histological grade (p < 0.001) and the extent of tumor resection (p < 0.001) have a significant and definitive relationship to the length of survival of pediatric patients with cerebellar astrocytomas. The patient's age (p > 0.05) does not influence the outcome. We could not evaluate the effect of brain stem involvement and adjuvant radiotherapy in our study because of the small sample size. Copyright (C) 2001 S. Karger AG, Basel.