Avoidable bilateral salpingo-oophorectomy at hysterectomy: a large retrospective study.

Abstract

Evaluate the proportion of justified bilateral salpingo-oophorectomy (BSO) at hysterectomy, based on pathologic diagnosis, and determine prevalence of avoidable BSO based on pre- and intraoperative considerations and pathologic diagnosis. Retrospective review of hysterectomies at seven Ontario, Canada hospitals from 2016 to 2019. Surgeries completed by oncologists or for invasive placentation were excluded. Patient, case, and surgeon characteristics were recorded along with pathologic diagnoses. Avoidable BSO criteria were: preoperative diagnosis of cervical dysplasia or benign diagnosis other than endometriosis, gender dysphoria, risk reduction or premenstrual dysphoric disorder; age < 51 years; absence of intraoperative endometriosis and adhesions; unjustified pathology (where "justified" pathology was endometriosis or (pre)malignant diagnosis except for cervical dysplasia). Patients with avoidable BSO were compared to those having at least one criterion for BSO. Binary logistic regression identified factors most strongly associated with avoidable BSO. Four thousand one hundred ninety-one hysterectomies were completed with 1,422 (33.9%) patients having concomitant BSO. Pathologic diagnosis justified BSO in most patients (1,035/1,422, 72.8%) with endometrial cancer being most common (439/1,422, 30.9%). When preoperative characteristics, intraoperative findings, and pathologic diagnoses were considered, 79 of 1,422 (5.6%) BSOs were avoidable. Compared to cases with at least one criterion for BSO, avoidable BSOs were more frequently completed by generalists (OR 1.80, 95% CI 1.10-2.99, P = 0.021), for preoperative diagnoses of abnormal uterine bleeding/menorrhagia (OR 3.82, 95% CI 2.35-6.30, P = 0.001) and fibroids (OR 4.25, 95% CI 2.63-6.92, P < 0.001). Pathologic diagnosis justified most BSOs at hysterectomy. BSO was avoidable in 5.6% of patients, underscoring the need to standardize practice of BSO.