Abstract
During operative treatment for ovarian tumours assistance is frequently required to make decisions regarding malignancy status and the extent of the ensuing procedure. Intra-operative frozen section analysis may be useful, provided there is adequate acquaintance with the correlation between using frozen sections and permanent histopathological sections for diagnosis at the institution where the operation is being undertaken. This retrospective study aimed to determine this correlation. Findings from 131 intraoperative frozen sections were compared with the subsequent diagnosis from permanent histopathological sections for women with benign, borderline and malignant ovarian tumours at the Maribor Teaching Hospital (now the University Clinical Centre Maribor) between 1 January 1993 and 31 December 2001. Frozen-section findings corresponded to histopathological findings in 84.7% of cases, with 15.3% false-negative and no false-positive results. For benign, borderline and malignant ovarian tumours, sensitivity was 100.0%, 76.1% and 89.0%, respectively, and specificity was 90.6%, 90.6% and 100.0%, respectively. The majority of errors occurred in diagnosing mucinous borderline tumours. Precise preoperative diagnosis is extremely important in the treatment of ovarian tumours.