Neuropathological diagnostic accuracy.

Abstract

This study investigated variations in neuropathological diagnosis when histopathological slides are evaluated with access to all information pertinent to a case, compared with evaluation of H & E stained slides with only limited clinical information. The aim of the study is to evaluate the role of 'local' factors in reaching a definitive, neuropathological diagnosis. Each of 227 cases was reviewed by the local neuropathologist at one of three European centres with access to all clinical, radiological and histopathological data. Each case was also reviewed by a second, independent neuropathologist at another European unit, who was blinded to the local pathologist's diagnosis, and only had access to the age and sex of the patient and the location of the lesion. For the histopathological evaluation, initially he had access to H & E and GFAP stains only. Diagnoses were made using the WHO 2000 classification system and the two evaluations were classified into one of the following four categories: complete agreement; minor disagreement (not affecting treatment); minor disagreement (affecting treatment); complete disagreement. In 175 (77.1%) cases there was complete agreement between the two neuropathologists and in 18 (7.9%) cases there were non-treatment altering discrepancies in the diagnoses. Overall, there was disagreement which might potentially affect treatment in 34 (15%) of the cases. Histopathological diagnosis of the majority (85%) of brain tumours is accurate and uniform. However, histopathology is subject to a degree of interpretation. The availability of full, 'local', clinical information may be of considerable importance to the diagnostic process.