Clues and pitfalls in stereotactic biopsy of the central nervous system.

We present a 6-year experience on 307 stereotactic biopsy specimens of the central nervous system using Leksell's and Talairach's systems independently and either Leksell or Sedan needles. Patients with deep cerebral lesions (basal ganglia, parasellar, pineal, or third ventricle), those located in highly functional areas or those poorly defined on imaging studies, as well as candidates for brachytherapy, were selected. Smear examination during surgery was a routine procedure followed by conventional histologic methods. Ages ranged from 8 months to 81 years (mean, 33.64 years). The series comprised 258 tumors, 28 nonneoplastic cases, and 21 nondiagnostic samples. Of the 258 tumors, 179 were supratentorial, 28 were infratentorial, 36 were of the pineal area, and 15 were from sellar and suprasellar regions. Results of the histologic examination showed the following: astrocytic tumors, 148 (57.36%); oligodendroglial, 25 (9.68%); ependymal, six (2.32%); primitive neuroectodermal tumors, 17, including 14 pineoblastomas (5.45%) and three medulloblastomas (1.16%), seven lymphomas (2.71%), seven meningiomas (2.71%), four schwannomas (1.55%), eight craniopharyngiomas (3.10%), 12 germinomas (4.65%), and 20 metastases (7.78%). Nontumoral cases included six arteriovenous malformations, six pyogenic lesions, seven infarcts, two hematomas, one multiple sclerosis plaque, one Fahr, one progressive multifocal leukoencephalopathy, one tuberculosis, one cysticercosis, and one Chagas' encephalitis. Awareness of the cerebellar granular layer in infratentorial targets as well as glial reaction around craniopharyngiomas is essential to avoid misdiagnosis. Difficulties were basically differential diagnosis between well-differentiated astrocytomas vs glial reaction, as well as poorly differentiated neoplasms vs metastases.(ABSTRACT TRUNCATED AT 250 WORDS)