There is limited data on the impact of frozen pathology on intraoperative management and patient outcomes in gynecologic surgery. We aimed to determine the rate of discordance between intraoperative frozen and final pathology and to calculate sensitivity, specificity, positive and negative predictive value. We also examined the impact on perioperative outcomes.METHODS:
All cases of frozen ovarian pathology between 2004–2014 at a single institution were reviewed. The primary outcome was the discordance rate between benign frozen and final pathology. Pathology was categorized as benign, malignant (borderline+malignant pathology) or uncertain. Need for reoperation was determined when final pathology was malignant.RESULTS:
1139 cases were submitted for frozen ovarian pathology. Four of the 797 benign frozen cases had malignant final pathology consistent with a 0.5% discordance rate. The sensitivity, specificity, positive and negative predictive value of frozen ovarian pathology was 82.6% (242/293), 92.2% (781/847), 98.4% (242/246), and 98% (781/797), respectively. Three of the discordant cases required reoperation. Of these, two cases were fibroma-thecomas on frozen and granulosa cell tumors on final. The third specimen was consistent with a benign cyst on frozen pathology, and a serous adenocarcinoma on final pathology. The final case was consistent with mature cystic teratoma on frozen pathology and immature cystic teratoma on final pathology and opted for observation in place of reoperation.CONCLUSION:
Discordance between benign frozen and final pathology is low. Rarely, when discordance occurs is subsequent surgery required. Further study is necessary to determine the effect of discordance on treatment outcomes.