Clinicopathological evaluation of 78 astrocytomas in Taiwan with emphasis on a simple grading system.

The grading systems of astrocytic tumors have long been the subject of controversy. A simple, objective and effective grading method is urgently needed for the evaluation of prognosis and the planning of treatment. This study investigated the relationship of clinical prognostic factors to the grading method of Daumas-Duport. This method determines the grade of tumor based on the presence or absence of four morphological criteria: nuclear atypia, mitosis, endothelial proliferation, and necrosis. A total of 143 astrocytic tumors were reviewed and screened, of which 65 ordinary and 13 pilocytic astrocytomas were selected for grading and comparison. Among ordinary astrocytomas, the grading method distinguished 9.2% grade 1, 26.2% grade 2, 36.9% grade 3, and 27.6% grade 4. At least 2-year follow-up was available on all surviving patients. Median survival was 57, 32, 12.5, and 8 months in grades 1, 2, 3, and 4 tumors, respectively. By a multiple regression model and analysis of variance, grade is significantly associated with survival (total regression coefficient r = 0.711). Age lost its significance on survival after multiple regression analysis. Sex, location, and surgical procedure were all unrelated to survival after regression. The age distribution and survival of patients with pilocytic astrocytomas revealed that this is a distinct disease entity and should not be admixed with ordinary astrocytomas in a grading scheme.