Effect of tumor size on the accuracy of frozen section in the evaluation of mucinous borderline ovarian tumors

Abstract

ObjectiveThis study aims to evaluate the accuracy of frozen section (FS) in mucinous borderline ovarian tumors (BOTs) and to examine the factors associated with misdiagnosis. MethodsIn this retrospective study, cases diagnosed as mucinous BOTs by FS or final pathologic (FP) results were studied. The results of FS and FP were compared, and the factors associated with misdiagnosis were analyzed. ResultsSeventy-nine cases were examined. The median tumor diameter was 16 (6–50) cm, and 89.9 % of cases had tumors ≥10 cm. The overall agreement ratio between FS and FP was 79.7 %. Over-diagnosis and under-diagnosis rates were 3.8 % and 16.5 %, respectively. The sensitivity and positive predictive values were both 88.7 %. None of the underdiagnosed patients (13 cases) had recurrence during the 100-month median follow-up (9–222). Misdiagnosis was more common in tumors <10 cm (p = 0.025). The under-diagnosis rate for tumors <10 cm was 30.8 %. Tumor size <10 cm was significantly associated with misdiagnosis in univariate and multivariate analyses (Odds ratio {OR} 4.92, 95 % Confidence Interval {CI} (1.08–22.45) p = 0.040; OR 5.17, 95 % CI (1.07–25.05) p = 0.041, respectively). Laterality and preoperative CA 125 levels were not associated with misdiagnosis. ConclusionTumor size <10 cm is associated with misdiagnosis in mucinous BOTs. Laterality and CA 125 levels do not affect diagnostic accuracy. The evaluation of FS by gynecologic pathologists can help to increase the accuracy of FS.