Adverse Fetal Outcomes and Maternal Mortality Following Nonobstetric Abdominopelvic Surgery in Pregnancy: Systematic Review and Meta-analysis

Abstract

INTRODUCTION: Uterine leiomyosarcoma (LMS) is rare and is estimated to comprise just 1% of all uterine malignancies. Preoperative detection of LMS is challenging. There are limited data on the accuracy of frozen section analysis for uterine LMS. The goal of this study is to review the utility of frozen section analysis in uterine LMS.METHODS: IRB approval was obtained for this retrospective cohort study. Women with a diagnosis of uterine LMS who underwent hysterectomy and had frozen section analysis obtained at the time of surgery were included. Data were summarized with frequency and percentage; Fisher exact tests were used for statistical analysis.RESULTS: There were 28 women analyzed, of whom only 14 (50%) had an accurate frozen section analysis reported as LMS or smooth muscle tumor favoring LMS. Another 14.3% were reported as malignant but not LMS on frozen section. The surgical plan changed based on the intraoperative pathology results in 14 of 28 (50%) cases. Three of 28 (10.7%) patients underwent a second surgery due to the inaccuracy of the frozen section results.CONCLUSION: Our data are consistent with previous case series that frozen section analyses on hysterectomy specimens in LMS are less reliable compared with other gynecologic cancers. Malignancy was correctly predicted in more than half of the cases; however, more than 10% of patients still required a second procedure due to under-calling malignancy on frozen section analysis. Further investigation is needed to help increase the accuracy of frozen section for LMS.