Risk of Diabetes After Hysterectomy With or Without Oophorectomy in Postmenopausal Women

Juhua Luo; JoAnn E. Manson; Rachel Peragallo Urrutia; Michael Hendryx; Erin S. LeBlanc; Karen L. Margolis


Am J Epidemiol. 2017;185(9):777-785. 

In This Article

Abstract and Introduction


The aim of this study was to determine the associations between hysterectomy, bilateral salpingo-oophorectomy (BSO), and incidence of diabetes in postmenopausal women participating in the Women's Health Initiative (WHI), a series of trials conducted in the United States, during the period 1993–1998. A total of 67,130 postmenopausal women aged 50–79 years were followed for a mean of 13.4 years. Among them, 7,430 cases of diabetes were diagnosed. Multivariable Cox proportional hazards models were used to assess the association between hysterectomy/oophorectomy status and diabetes incidence. Compared with women without hysterectomy, women with hysterectomy had a significantly higher risk of diabetes (hazard ratio = 1.13, 95% confidence interval: 1.06, 1.21). The increased risk of diabetes was similar for women with hysterectomy only and for women with hysterectomy with concomitant BSO. Compared with hysterectomy alone, hysterectomy with BSO was not associated with additional risk of diabetes after stratification by age at hysterectomy and hormone therapy status. In our large, prospective study, we observed that hysterectomy, regardless of oophorectomy status, was associated with increased risk of diabetes among postmenopausal women. However, our data did not support the hypothesis that early loss of ovarian estrogens is a risk factor for type 2 diabetes. The modest increased risk of diabetes associated with hysterectomy may be due to residual confounding, such as the reasons for hysterectomy.


Hysterectomy is the second most common surgery among women in the United States, and approximately 600,000 hysterectomies are performed in the United States annually.[1] Almost 90% of surgeries are for benign gynecologic conditions, including symptomatic uterine fibroids or abnormal uterine bleeding.[2,3] Approximately 44% of women have concomitant bilateral salpingo-oophorectomy (BSO) at the time of hysterectomy in order to prevent the subsequent development of ovarian cancer, treat medical conditions, or prevent the need for future adnexal surgery.[4] In premenopausal women, BSO induces menopause, but hormonal effects may also be present in postmenopausal women.[5,6] In addition, hysterectomy without BSO has been associated with a shorter time to menopause[7] and has been found to have similar, but less dramatic, hormonal changes as BSO.[5,8,9] Increased knowledge about the long-term effects of hysterectomy and BSO on women's health will improve medical decision-making for women and their providers.[10]

BSO before age 50 years is significantly associated with death from all causes.[11] Some have attributed this to a higher risk of cardiovascular disease.[12] It has been postulated that women who undergo early BSO may subsequently experience a higher risk of type 2 diabetes mellitus (hereafter referred to as diabetes) compared with women who do not undergo oophorectomy.[13] According to Mauvais-Jarvis et al.,[14] ovarian hormones regulate both insulin secretion and survival of pancreatic beta cells. In animal studies, an absence of female sex hormones after BSO leads to decreased whole-body insulin-mediated glucose uptake.[15,16] Oophorectomy worsens glucose tolerance and insulin resistance in mice.[17] Furthermore, several randomized clinical trials have indicated that exogenous menopausal hormone therapy (either estrogen alone or estrogen plus progestin) reduces the risk of diabetes.[18–20] Thus, we hypothesized that early BSO would be associated with an increased risk of diabetes and that hysterectomy without BSO might be associated with an intermediate risk compared with undergoing neither BSO nor hysterectomy.

Epidemiologic studies investigating the associations between hysterectomy, BSO, and diabetes incidence are sparse. Appiah et al.,[21] using data from 2,597 postmenopausal women enrolled in the National Health and Nutrition Examination Survey I Epidemiologic Follow-up Study, reported that hysterectomy with BSO was significantly associated with diabetes risk (hazard ratio (HR) = 1.57, 95% confidence interval (CI): 1.03, 2.41), while hysterectomy alone was associated with a nonsignificantly increased risk of diabetes (HR = 1.38, 95% CI: 0.94, 2.04). In another small, prospective study (only 33 women with BSO), Lejsková et al.[22] found that women with BSO had a significant increase in fasting glycemia compared with women with natural menopause. In a third study, a secondary analysis of a randomized controlled trial among glucose intolerant adults, Kim et al.[23] reported no association between diabetes risk and bilateral oophorectomy. In that study, analysis of the subgroup that was randomized to a lifestyle intervention indicated that diabetes risk was lower among women with BSO compared with premenopausal women.[23] However the population was women at high risk of diabetes, and investigators were unable to assess the effects of hormone use due to the lack of diabetes cases.

The purpose of this study was to use a large prospective data set (the Women's Health Initiative (WHI)) to examine the associations between hysterectomy, BSO, and incidence of diabetes while adjusting for important confounders. We also tested the hypothesis that early loss of ovarian estrogens is a risk factor for diabetes by comparing diabetes risk among women who had hysterectomy with BSO with the risk among women who underwent hysterectomy only, stratified by age at hysterectomy and hormone therapy status.