An appraisal of smear diagnosis in neurosurgical biopsies - abstract

ABSTRACT

The neurosurgeon needs rapid tissue diagnosis during surgery to plan the extent of the operation if the exact nature of the lesion is known, to take additional specimens if a previous attempt has failed to be diagnostic and to decide on management if the lesion is non-neoplastic. Rapid tissue diagnosis is mostly offered by frozen sections in various fields of surgery. However, neuro-surgical biopsies are generally multiple, tiny fragments of gelatinous tissue, mostly inappropriate for frozen sections. Smear technique for tissue diagnosis has been in use since 1930 and is routinely used for neurosurgical diagnosis in many European centres. It was started at the University Hospital of the West Indies in 1981. To evaluate the accuracy of the technique we analyzed smear diagnoses and compared them with paraffin section diagnoses. The results were compared with similar studies from other parts of the world. One hundred and forty-four biopsies, comprising 133 intracranial and 11 spinal biopsies, were analyzed. The correct diagnosis was made in 93 percent of cases. The conditions encountered were mainly neoplastic. In 10 cases, there were discrepancy between smear and paraffin section diagnosis. In 8 of the 10 cases, the error was in properly classifying the tumour and did not have any immediate significance on the management. Only in two cases could the error have had a management implication. There was no false negative and only one false positive when a meningioma was diagnosed on superficial biopsy material and on subsequent analysis of more specimen proved to be an inflammatory pseudo-tumour. The diagnostic accuracy (93 percent) was in keeping with similar studies from Europe and North America. The technique is simple, rapid, sensitive, accurate and appropriate for rapid diagnosis of neurosurgical biopsies. (AU)