Accuracy of grading gliomas on CT-guided stereotactic biopsies: a survival analysis

The accuracy of using a combination of cytopathologic and histopathologic techniques to diagnose stereotactically guided brain biopsies was investigated in 74 patients. Diagnostic accuracy was assessed by determining whether classification of the biopsies as gliosis, astrocytoma (A), anaplastic astrocytoma (AA), or glioblastoma multiforme (GBM) predicted survival. The utility of on-site evaluation using Diff-Quik-stained crush preparations was also assessed. The patients ranged in age from 5 to 88 years (mean, 55 years) and were followed for over 2 years in most cases. Four cases (5%) were classified as gliosis (G), 7 (9%) as atypical gliosis (AG), 4 (5%) as high-grade mixed oligodendroglioma/astrocytoma (OA), 11 (15%) as astrocytoma (A), 21 (28%) as anaplastic astrocytoma (AA), and 27 (36%) as glioblastoma multiforme (GBM). Median survival was 11 months in patient with OA, 57 months in patients with A, 10 months in patients with AA, and 5 months in patients with GBM. Diagnosis of Diff-Quik-stained crush preparations made during the biopsy procedure was highly correlated with the final diagnosis and survival. We conclude that the diagnosis of stereotactic brain biopsies using cytopathology with on-site evaluation in combination with histopathological evaluation of needle cores is accurate based on a survival analysis. However, A and G may be difficult to distinguish.