Long Term Follow-up of Endometriosis Surgery in Ontario: A Population-Based Cohort Study

Abstract Objective To characterize the long term outcomes, including recurrence of symptoms, fertility outcomes and need for reoperation, of patients undergoing surgical management of endometriosis. Methods Population-based cohort study using universal coverage health database for the province of Ontario, Canada of women aged 18 to 50 undergoing surgery for endometriosis from April 1, 2002 through March 31, 2018. Surgery was classified as diagnostic laparoscopy, conservative/uterine preserving (minor or major, with and without ovarian preservation) and hysterectomy (with and without ovarian preservation). Outcomes occurred 30 days after index surgery until the end of the study period or censoring. Cox Proportional-Hazards regression models were used to estimate hazard ratios (aHR) between exposures and outcomes adjusted for confounders. Results A total of 84,885 women (3.2% diagnostic laparoscopy, 25.4% minor conservative surgery, 33.6% major conservative with ovarian preservation, 2.5% major conservative without ovarian preservation, 25.5% hysterectomy with ovarian preservation, and 9.9% hysterectomy without ovarian preservation) were included in the cohort and followed for a median of 10 (IQR:6-13) years. In the first postoperative year, women who underwent diagnostic laparoscopy were significantly more likely to require repeat surgery (aHR1.68, 95% CI: 1.51, 1.87), while those having major conservative surgery were significantly less likely to require repeat surgery (with ovarian preservation aHR 0.44, 95% CI: 0.41, 0.48 and without ovarian preservation aHR 0.05, 95% CI: 0.03, 0.09). Among women who did not receive repeat surgery in the first year, those having diagnostic laparoscopy (aHR 0.85, 95% CI: 0.76, 0.95) and major conservative surgery without ovarian preservation were less likely to undergo repeat surgery (aHR 0.12, 95% CI: 0.09, 0.18) than those who initially had minor surgery. Compared to those initially having minor surgery, patients who underwent other treatment modalities were less likely to undergo hysterectomy (diagnostic laparoscopy aHR 0.85, 95%CI 0.75, 0.96, major with ovarian preservation aHR 0.60, 95% CI: 0.57, 0.64, and major without ovarian preservation aHR 0.05 95% CI: 0.03, 0.08). Following minor and major conservative with ovarian preservation surgery, 38.6% and 33.3% of patients sought an infertility consult, respectively. By five years after index surgery, 29.4% of patients who had minor conservative surgery and 20.7% of major conservative with ovarian preservation surgery patients had given birth at least once. Comment Our study suggests that few endometriosis patients who undergo hysterectomy require repeat surgery; however, up to 1 in 4 who undergo minor surgery and 1 in 5 who undergo major conservative surgery with ovarian preservation require additional endometriosis surgery. Up to 1 in 3 patients who had uterine sparing endometriosis surgery subsequently seek infertility assessment. These findings may inform preoperative counseling with regards to recurrence of symptoms, fertility outcomes and need for reoperation for women seeking surgical management of endometriosis. Future studies should consider outcomes of patient satisfaction and quality of life under the current practices for management of endometriosis.