Decreasing utilization of vaginal hysterectomy: an analysis by candidacy for vaginal approach

Although vaginal hysterectomy (VH) is recommended as the preferred approach for benign hysterectomy whenever feasible, the rate of VHs performed is declining. It is unclear whether the number of appropriate candidates for VH is a contributing factor. We sought to assess trends in hysterectomy routes by patients who are likely and unlikely candidates for a vaginal approach using two independent surgical registries. We performed a retrospective cohort study of patients who underwent vaginal, abdominal, or laparoscopic hysterectomy between 2017-2020 in two surgical registries with chart-abstracted hysterectomy data: the National Surgical Quality Improvement Program (NSQIP) database and the Michigan Surgical Quality Collaborative (MSQC). Patients undergoing hysterectomy for a primary diagnosis of benign uterine pathology, dysplasia, abnormal uterine bleeding, pelvic floor disorder, or pelvic pain based on International Classification of Diseases codes were eligible for inclusion; patients undergoing hysterectomy for malignancy, endometriosis, adnexal indications, or other indications were excluded. Based on an algorithm developed to guide surgical approach, patients were deemed likely VH candidates if they were parous, had no history of pelvic or abdominal surgery, and had a uterine weight < 280 gm on pathology. Average annual changes in the proportion of likely VH candidates and route of hysterectomy were assessed using linear regression and trends were compared. A total of 81,194 patients in NSQIP and 16,910 patients in MSQC met inclusion criteria. Of these patients in NSQIP and MSQC, 14,278 (17.6%) and 4,068 (24.1%) were deemed likely VH candidates, respectively. Among likely VH candidates, the proportion who underwent VH was 34.1% and 29.2%; among unlikely VH candidates, it was 13.8% and 10.6%, respectively. There was a decrease in the overall proportion of VH in both datasets (NSQIP: -1.2%/year, p<0.01; MSQC -0.5%/year; p=0.02). This decreasing trend was more pronounced among likely VH candidates (NSQIP: -1.9%/year, p<0.01; MSQC -1.5%/year, p=0.04) compared to unlikely VH candidates (NSQIP: -1.1%/year, p<0.01; MSQC -0.2%/year, p=0.53); the difference reached statistical significance in the NSQIP dataset (p<0.01) but not the MSQC dataset (p=0.08). There was no significant change in the proportion of likely VH candidates in the NSQIP dataset (p=0.33) while a decrease of -0.7%/year was identified in the MSQC dataset (p=0.02). The proportion of VH performed for eligible indications decreased between 2017-2020 in two independent surgical registries. This negative trend was present and more pronounced among patients who were likely candidates for VH based on favorable parity, surgical history, and uterine weight. These trends are important to consider, as they may have implications on preservation of vaginal surgery skills, surgical training, and outcomes.View Large Image Figure ViewerDownload Hi-res image Download (PPT)