We performed a diagnostic accuracy study on the accuracy of frozen section diagnosis in the assessment of the adnexal mass. Final histopathologic diagnosis used as the gold standard. The diagnostic accuracy indices were estimated for frozen section. In the present study, the overall accuracy of frozen section diagnosis of ovarian masses was 98.9%.
The high precision of the frozen section helps gyn-oncologists make reasonable intraoperative decisions and thus in the vast majority of benign cases prevent unnecessary morbidity of the procedure. The accuracy of the frozen section for ovarian tumors varies across different institutions. Subbian et al. found an overall accuracy of 84.2% in a study involving 135 cases of ovarian tumors (13). Several studies have shown that frozen section diagnosis of ovarian tumors is a robust technique with overall accuracy ranging from 71.9–97% (10, 14). In this study, the overall accuracy of the frozen section diagnosis by intraoperative ovarian mass was 98.9%, which is consistent with previous reports (90–97%) (1, 6, 15–18).
The sensitivity of frozen section to a benign tumor from our observation was 100%. Previous studies also reported a high sensitivity ranging from 92.8% − 100% (10, 11, 19–21). In the present study, one borderline tumor was diagnosed as benign by frozen section and that was mucinous tumor. Five borderline tumors, diagnosed as benign on the frozen section, were reported in a tertiary center experience by (21). Several explanations have been given for the relative inaccuracy of the frozen section in the diagnosis of borderline tumors. There may be just a few focal points of frank malignancy in a large borderline tumor which may require a large number of frozen section samples for diagnosis. This is quite labor intensive and typically beyond most laboratory abilities (22). It is remarkable that for borderline tumor diagnosis, most studies have reported low sensitivity values. Given the low accuracy rate for borderline ovarian tumors, care and attention are required to develop this field (10, 11, 13, 14, 18). Md Arshad et al. observed that the borderline tumor sensitivity was 76.2% (21). Subbian et al., on the other hand, recorded the lowest borderline tumor sensitivity (31.2%), especially in the mucinous classification (13). In a review of 60 patients of ovarian tumors by (11), frozen section had low sensitivity (75%) and PPV (50%) for borderline tumors. In a further retrospective study of 282 cases, the sensitivity of the frozen section to borderline tumors was 95.8% and the specificity 97.6% (23). For the borderline category, they found the lowest positive predictive value (79.3%), all with epithelium of mucinous type. Our results showed 89% for sensitivity and 80% for positive predictive value, which were in range of similar studies. In the present study, three discordant cases were identified during diagnosis; two cases were incorrectly diagnosed as borderline epithelial tumor and borderline mucinous tumor using a frozen section. The first was a malignant endometrioid tumor, and the second was a malignant mucinous tumor of Final HPE. Mucinous tumors are usually large with a heterogeneous component including benign, borderline and malignant constituents in the same tumor, making diagnosis difficult (18, 21, 24). In addition, Gultekin et al. suggested that the size of the tumor, the presence of a solid component and preoperative CA 125 levels may have an impact on diagnosis (25). However, Palakkan et al. found that there was no statistically significant association between raised CA 125 level and ovarian malignancy (11).
Most studies typically reported a sensitivity of 71–100% for the detection of malignancies by frozen section, and a specificity ranging from 96–100% (11, 19, 21, 23, 26, 27). In our study, the sensitivity of malignant ovarian tumor detection was 93.3% and the specificity was 100%, reflecting that the frozen section examination was highly sensitive and specific to ovarian malignancy. It is very important to correctly detect malignant ovarian tumors, as it determines the type of surgery that should be performed. A study by (28) indicated that, if the pathologist manager had under-specialty training in gynecological pathology, prevention of sampling errors and misinterpretation, along with effective communication could be ensured. If a frozen section is required during surgery, the Gyne-oncology team should be able to communicate preoperatively with the anesthetist and pathologist. A detailed clinical history and intraoperative observations should be communicated to the pathologist as they may help the pathologist to get the correct diagnosis, especially in problematic cases. Despite advances in histological and molecular techniques, frozen section evidently remains an effective tool to be considered during operational procedures.
Frozen section procedure could reduce the possibilities of doing incomplete surgery for benign tumors (oophorectomy or cystectomy) or radical surgery for malignant conditions, in a majority of patients. Frozen section is sometimes referred to as intraoperative consultation, as its practice usually involves a liaison between the surgeon and gynecological pathologist, rather than the mere provision of a histological diagnosis. Accuracy of frozen section depends on a number of factors such as the number of specimens processed, number of slices of specimen analyzed and nature of tissue. Of note is that the report of benign histology on frozen section is only a report of that percentage of tissue analyzed.18 Therefore, intraoperative frozen section is necessary, both as a quick histologic diagnosis and as a guide for the surgeons in planning an appropriate management.
Besides the nature of the tumor affecting frozen section examination, frozen artifact may also cause misinterpretation of frozen section. The ovarian tumor that was sent to the laboratory is preferably frozen in liquid nitrogen. It is important that the tissue be frozen as soon as possible to prevent ice crystal formation resulting in artifact and poor morphological preservation of the tissue, rendering poor slide quality and difficulty to obtain a diagnosis. Another factor that may contribute to misinterpretation is the presence of necrosis. A tumor that has become necrotic following torsion might have limited tissue available for frozen section evaluation.
The strength of this study was the fact that all slides from both frozen section and final HPE for all the unmatched cases were traced and reevaluated by the pathologist to determine the possible reason for any difference in the result or interpretation. However, several limitations were identified in this study. Firstly, the sample size was limited. Secondly, we did not have all the data on the progress of the patients after the intraoperative frozen section and surgery. Thirdly, the cost effectiveness of intraoperative frozen section was not assessed in our study. Finally, if another blinded pathologist was recruited to determine the reasons for the mismatch, this might decrease our study bias.