By Rebecca H. Allen, MD, MPH, Editor
SYNOPSIS: In this population-based retrospective cohort study, among 44,549 adult women undergoing hysterectomy in Ontario, Canada, there was marked variation between surgeons in bilateral salpingo-oophorectomy (BSO) rates after controlling for patient age and other factors. Approximately 41% of patients had no indication for the bilateral salpingo-oophorectomy in their records.
SOURCE: Cusimano MC, Moineddin R, Chiu M, et al. Practice variation in bilateral salpingo-oophorectomy at benign abdominal hysterectomy: A population-based study. Am J Obstet Gynecol 2021;224:585.e1-585.e30.
Second to cesarean delivery, hysterectomy is one of the most common surgeries performed on women. Although removal of both fallopian tubes at the time of hysterectomy, also known as opportunistic salpingectomy, has increasingly become an accepted procedure to reduce the risk of future epithelial ovarian cancer, the question of whether to remove the ovaries has long been debated.1 In recent years, evidence has emerged that routine removal of the ovaries has negative consequences on subsequent cardiovascular health and mortality, especially if removed in premenopausal women.2 The authors of this study sought to determine variation in bilateral salpingo-oophorectomy (BSO) practice at the surgeon level.
This was a population-based, retrospective, cohort study using Ontario, Canada’s health research databases that include any residents who access healthcare. Women ≥ 20 years of age undergoing benign hysterectomy from Jan. 1, 2014, to Dec. 31, 2018, were included. Exclusion criteria included non-Ontario residents; hysterectomy performed for emergent, malignant, or genetic susceptibility to cancer indications; and a history of previous breast or gynecologic cancer. The primary outcome was BSO at the time of hysterectomy. A secondary outcome was the “avoidable” BSO, which was defined as BSO performed without a documented indication, such as endometrial hyperplasia, endometriosis, and/or ovarian cysts. Patient demographic and clinical characteristics as well as surgeon characteristics were collected. Primary analyses were stratified by age (< 45 years, 45-54 years, and ≥ 55 years old), and a multivariable regression analysis was performed.
A total of 44,549 women were included in the study, with 706 unique surgeons, and 39.9% of patients underwent concurrent BSO. Approximately 48.3% of cases were open laparotomy and 51.7% were via the laparoscopic approach. The median age of the cohort was 44 years in women without a BSO compared to 51 years in the BSO group. The rates of BSO varied by age group, with 17%, 44%, and 84% of patients aged < 45, 45-54, and ≥ 55 years, respectively, undergoing the procedure. Among women who underwent BSO, 41% did not have a documented indication, and patients aged younger than 45 years of age were less likely to have a potential avoidable BSO compared to patients aged 45-54 years and ≥ 55 years (31.1% vs. 44% vs. 43.2%, P < 0.001). There was significant variation between surgeons in the group aged 45-54 years, with surgeons accounting for 22% of the residual variation in that group even after adjustment for patient and surgeon factors, such as age, surgical approach, obesity, presence of other gynecologic conditions (abnormal uterine bleeding, fibroid tumors, endometriosis, ovarian cysts, premalignant conditions, and prolapse), and surgeon factors (gender, specialty, country of graduation, year of graduation, and annual hysterectomy volume). Multivariable logistic regression revealed that the individual surgeon had the strongest influence on whether BSO was performed, followed by patient factors, such as benign ovarian cysts, premalignant disease, and morbid obesity.
The decision to remove the ovaries at the time of hysterectomy is a complex one and depends on multiple factors. The principle findings of this study are that BSO at benign hysterectomy is common but varies markedly between surgeons even after controlling for patient differences. This variance was most pronounced in the perimenopausal group aged 45-54 years. This is one of the first studies looking at differences at the surgeon level in the rates of BSO. The authors’ interpretation was that there still is a level of uncertainty among clinicians about whether to conserve or remove ovaries, especially in the perimenopausal age group. The study had several strengths since it included both inpatient and outpatient hysterectomies performed in the past decade and had the ability to analyze surgeon-level data. However, there were some limitations, including the lack of data regarding actual menopausal status, intraoperative findings, family history, and genetic testing.
In the United States, it is estimated that the ovaries are removed during 40% to 50% of routine hysterectomies.2 Traditionally, arguments in favor of removing the ovaries at the time of hysterectomy have included a decreased risk of breast and ovarian cancer in the future as well as a reduced risk of future surgery to remove the adnexa. However, we know now that surgical menopause is associated with increased mortality as the result from all causes, cardiovascular disease, decreased sexual and cognitive functioning, and osteoporosis.3 It turns out that the postmenopausal ovary continues to secrete androgens and plays an important role in overall health that cannot always be replaced by postoperative hormone therapy.
In 2005, a Markov decision analysis to estimate the risk of prophylactic oophorectomy in women at average risk of ovarian cancer was published.4 The analysis balanced the risks of overall mortality and mortality from cardiovascular disease, hip fracture, stroke, ovarian cancer, and breast cancer. The model calculated that oophorectomy before age 55 years had an 8.6% risk (and a 3.9% risk from ages
56-59 years) of excess mortality as the result of cardiovascular disease. Ovarian preservation always was favored in the model because the relative risk of dying from ovarian cancer is dwarfed by the risk of cardiovascular disease and hip fracture. The authors of this decision analysis recommended that routine prophylactic oophorectomy should not be performed before age 65 years. A more recent systematic review concluded, “Bilateral salpingo-oophorectomy offers the advantage of effectively eliminating the risk of ovarian cancer and reoperation but can be detrimental to other aspects of health, especially among women younger than age 45 years.”5
Factors that are important to consider when counseling patients about whether prophylactic oophorectomy is best for them include age, menopausal status, and the presence of known pathogenic gene mutations or a family history of ovarian/breast cancer.3 According to the authors of one clinical algorithm, oophorectomy should not be recommended at age ≤ 45 years for patients at average risk of ovarian cancer unless there is pathology.2 In contrast, it is reasonable to offer BSO if the patient is 65 years of age or older. For the middle age group, ages 46-64 years, the authors suggest evaluating menopausal status. If patients are premenopausal, the risks of BSO may outweigh any benefits. For those who are menopausal, if they have a significant personal or family history of cardiovascular disease, dementia, or osteoporosis, they also may want to defer oophorectomy. Ultimately, the decision should be arrived at through a process of shared decision-making between the surgeon and patient, taking into account the patient’s individual preferences and risk factors.
- [No authors listed]. ACOG Committee Opinion No. 774: Opportunistic salpingectomy as a strategy for epithelial ovarian cancer prevention. Obstet Gynecol 2019;133:e279-e284.
- Huber-Keener KJ, Pearlman MD. When should prophylactic oophorectomy be recommended at the time of elective hysterectomy? Clin Obstet Gynecol 2020;63:337-348.
- Adelman MR, Sharp HT. Ovarian conservation vs removal at the time of benign hysterectomy. Am J Obstet Gynecol 2018;218:269-279.
- Parker WH, Broder MS, Liu Z, et al. Ovarian conservation at the time of hysterectomy for benign disease. Obstet Gynecol 2005;106:219-226.
- Casiano Evans E, Matteson KA, Orejuela FJ, et al. Salpingo-oophorectomy at the time of benign hysterectomy: A systematic review. Obstet Gynecol 2016;128:476-485.