The impact of surgeon volume on perioperative adverse events in women undergoing minimally invasive hysterectomy for the large uterus

Abstract

There are currently sparse data on the relationship between surgeon- and patient-related factors and perioperative morbidity in the setting of elective hysterectomy for the larger uterus. We sought to evaluate the impact of surgeon case volume on perioperative adverse events in women undergoing minimally invasive hysterectomy for uteri >250 g. This is a retrospective cohort study of all women who underwent total vaginal, total laparoscopic, laparoscopic-assisted vaginal, or robotic-assisted total laparoscopic hysterectomy from January 2014 through July 2016. Hysterectomy was performed for: fibroids, pelvic pain, abnormal uterine bleeding, or prolapse. Patients were identified by Current Procedural Terminology codes and the systemwide electronic medical record was queried for demographic and perioperative data. Perioperative adverse events were defined a priori and classified using the Clavien-Dindo scale. Surgeon case volume was defined as the mean number of minimally invasive hysterectomy cases performed per month by each surgeon during the study period. In all, 763 patients met inclusion criteria: 416 (54.5%) total laparoscopic hysterectomy, 196 (25.7%) robotic-assisted total laparoscopic hysterectomy, 90 (11.8%) total vaginal hysterectomy, and 61 (8%) laparoscopic-assisted vaginal hysterectomy. Mean (±SD) age was 47.3 ± 6.1 years, and body mass index was 31.1 ± 7.4 kg/m2. In all, 66 surgeons performed minimally invasive hysterectomy for uteri >250 g during the study period, and the median rate of minimally invasive hysterectomy cases for large uteri per month was 3.4 (0.4-3.7) cases/month. The median (IQR) uterine weight was 409 (308-606.5) g. The rate of postoperative adverse events Dindo grade >2 was 17.8% (95% confidence interval, 15.2-20.7). The overall rate of intraoperative adverse events was 4.2% (95% confidence interval, 2.9-5.9). The rate of conversion to laparotomy was 5.5% (95% confidence interval, 4.0-7.4). There was no significant difference in adverse event rates between the routes of minimally invasive hysterectomy cases (25.6% vs 17.5% vs 18.0% vs 14.8% for total laparoscopic hysterectomy, robotic-assisted laparoscopic hysterectomy, total vaginal hysterectomy, and laparoscopic-assisted vaginal hysterectomy, respectively, P= .2). In a logistic regression model controlling for age, body mass index, uterine weight, operating time, and history of laparotomy, higher monthly minimally invasive hysterectomy volume was significantly associated with the likelihood that a patient would experience a postoperative adverse event (adjusted odds ratio, 1.1 for each additional minimally invasive hysterectomy case for large uteri per month; 95% confidence interval, 1.0-1.3). When controlling for the same variables, a higher incidence of intraoperative complications was significantly associated with monthly minimally invasive hysterectomy case volume (adjusted odds ratio, 1.5 for each additional minimally invasive hysterectomy case for large uteri per month; 95% confidence interval, 1.20-2.08). Increasing age was associated with a lower incidence of complications (adjusted odds ratio, 0.9 for each additional year; 95% confidence interval, 0.8-0.9). Higher monthly minimally invasive hysterectomy volume was associated with a lower rate of conversion from a minimally invasive approach to laparotomy (adjusted odds ratio, 0.4 for each additional minimally invasive hysterectomy case for large uteri per month; 95% confidence interval, 0.2-0.5). The overall rate of serious adverse events associated with minimally invasive hysterectomy for uteri >250 g was low. Higher monthly minimally invasive hysterectomy case volume was associated with a higher rate of intraoperative and postoperative adverse events but was associated with a lower rate of conversion to laparotomy.