Date on Master's Thesis/Doctoral Dissertation


Document Type

Doctoral Dissertation

Degree Name

Ph. D.


Epidemiology and Population Health

Committee Chair

Hornung, Carlton

Committee Co-Chair (if applicable)

Winters, Stephen J.

Committee Member

Baumgartner, Richard N.

Committee Member

Allison, Matthew A.

Committee Member

Groves, Frank D.

Committee Member

Myers, John

Author's Keywords

Cardiovascular disease; Womens health; Bilateral oophorectomy; Diabetes


Cardiovascular system--Diseases; Women--Health and hygiene; Diabetes in women


Cardiovascular Disease (CVD) is the leading cause of death in women and the risk is increased threefold in diabetics. In postmenopausal diabetics, the ovary responds to hyperinsulinemia by increasing secretion of testosterone precursors which increases the risk of CVD. Data from the National Health and Nutrition Examination Survey (1988-2010) and the Study of Osteoporotic Fractures were used to test the hypothesis that a history of bilateral oophorectomy -surgical removal of both ovaries-would decrease the risk of CVD among postmenopausal diabetic women due to reduced androgen levels. Logistic regression, discrete time logit model, B-Spline regression and Cox proportional hazards models were performed with adjusted estimates and 95% confidence intervals (CIs) calculated. Overall, the studies comprised of 17,549 postmenopausal women with approximately 17.1 % having a history of diabetes Mellitus. Additionally, 24.1% of diabetic women had undergone bilateral oophorectomy with 55% of these having the surgery before age 45 years. Diabetics were more likely to be obese, live a more sedentary lifestyle, have dyslipidemia and were more likely to report a higher prevalence of stroke and myocardial infarction than non diabetics at baseline. Women who had oophorectomy before age 45 years were more likely to be nulliparous and report a family history of myocardial infarction. Diabetic women with oophorectomy had lower levels of total testosterone compared to diabetic women with intact ovaries or naturally menopausal non-diabetic women, which was independent of age and body mass index (p=0.016). In multivariable regression models, bilateral oophorectomy was predictive of prevalent CVD in diabetics (Odds ratio: 1.63 (95% CI: 1.16 - 2.30) with the odds limited to women who had this procedure before age 45 years (OR: 2.11, CI: 1.45-3.08). Although ovarian status did not influence incident CVD in diabetic women (OR: 0.78, CI: 0.56-1.09), women with oophorectomy before the average age at menopause (51 years), with or without diabetes were observed to have elevated CVD risk in spline regression models. Furthermore, the risk of CVD mortality was elevated for diabetic women irrespective of ovarian status, although women with no ovaries had an increased risk (Hazard ratio: 2.57, CI: 1.18-3.67) than those with intact ovaries (HR: 1.99, CI: 1.65-2.39). This present analyses consisting of three different population-based samples of postmenopausal women did not support oophorectomy having a protective effect on cardiovascular health in diabetic women; however, our interpretation of these data is that oophorectomy was performed more often in younger women who inherited a risk factor for heart disease.

Included in

Epidemiology Commons