Morphologic evaluation of stereotactic brain tumour biopsies.

The validity of morphologic diagnosis of stereotactic brain tumour biopsies was evaluated in a series of 600 patients treated since 1977 at the University Hospital, Freiburg. Combined cytological (smear preparations) and histological examination of paraffin-embedded samples revealed the tumour type and approximate grading in 492 (82%) of cases. In 66 patients a clinically suspected neoplasm could be ruled out. In the remaining 42 cases (7%), the presence of a tumour was confirmed but the available samples did not allow an unequivocal classification of the neoplasm. Inaccurate diagnoses were most frequently due to sampling errors in non-homogeneous tumours, i.e. biopsies taken from sites not representative for the entire neoplasm (tumour necroses, infiltration zone). In the future, the use of immunohistochemical methods for the identification of tumour markers and cytoskeleton proteins may partially compensate for the limited size of stereotactic biopsy samples.