The association of hysterectomy with or without ovarian conservation with subclinical atherosclerosis progression in healthy postmenopausal women

Abstract

Objective: While the deleterious associations of surgical menopause after bilateral oophorectomy with cardiovascular disease are documented, less is specifically known concerning subclinical atherosclerosis progression.

Methods: We used data from 590 healthy postmenopausal women randomized to hormone therapy or placebo in the Early versus Late Intervention Trial with Estradiol (ELITE), which was conducted from July 2005 to February 2013. Subclinical atherosclerosis progression was measured as annual rate of change in carotid artery intima-media thickness (CIMT) over a median 4.8 years. Mixed-effects linear models assessed the association of hysterectomy and bilateral oophorectomy compared with natural menopause with CIMT progression adjusted for age and treatment assignment. We also tested modifying associations by age at or years since oophorectomy or hysterectomy.

Results: Among 590 postmenopausal women, 79 (13.4%) underwent hysterectomy with bilateral oophorectomy and 35 (5.9%) underwent hysterectomy with ovarian conservation, a median of 14.3 years before trial randomization. Compared with natural menopause, women who underwent hysterectomy with and without bilateral oophorectomy had higher fasting plasma triglycerides while women who underwent bilateral oophorectomy had lower plasma testosterone. The CIMT progression rate in bilaterally oophorectomized women was 2.2 μm/y greater than natural menopause ( P = 0.08); specifically, compared with natural menopause, the associations were significantly greater in postmenopausal women who were older than 50 years at the time of bilateral oophorectomy ( P = 0.014) and in postmenopausal women who underwent bilateral oophorectomy more than 15 years before randomization ( P = 0.015). Moreover, the CIMT progression rate in hysterectomized women with ovarian conservation was 4.6 μm/y greater than natural menopause ( P = 0.015); in particular, compared with natural menopause, the association was significantly greater in postmenopausal women who underwent hysterectomy with ovarian conservation more than 15 years before randomization ( P = 0.018).

Conclusions: Hysterectomy with bilateral oophorectomy and ovarian conservation were associated with greater subclinical atherosclerosis progression relative to natural menopause. The associations were stronger for later age and longer time since oophorectomy/hysterectomy. Further research should continue to examine long-term atherosclerosis outcomes related to oophorectomy/hysterectomy.

Figure 1.. Differential Timing of Bilateral Oophorectomy Association with CIMT Progression Rate (μm/y)

A. compared women who underwent concurrent bilateral oophorectomy and hysterectomy, and women who underwent hysterectomy with ovarian conservation, with naturally postmenopausal women. B. compared women who had bilateral oophorectomy ≤50 years old, and >50 years old with naturally menopausal women. C. compared women who had bilateral oophorectomy ≤15 years and >15 years prior to randomization with naturally menopausal women. D. compared women who had hysterectomy with ovarian conservation ≤50 years old, and >50 years old with naturally menopausal women. E. compared women who had hysterectomy with ovarian conservation ≤15 years and >15 years prior to randomization with naturally menopausal women. The mean CIMT progression rate in each stratum was calculated from mixed-effects models with adjustment for age (years), treatment assignment (hormone therapy, placebo), BMI, and time since menopause (< 6 years, ≥ 10 years). Lines indicate 95% confidence intervals of the mean CIMT progression rate (μm/y); P-value from the Wald test.

Figure 1.. Differential Timing of Bilateral Oophorectomy Association with CIMT Progression Rate (μm/y)

A. compared women who underwent concurrent bilateral oophorectomy and hysterectomy, and women who underwent hysterectomy with ovarian conservation, with naturally postmenopausal women. B. compared women who had bilateral oophorectomy ≤50 years old, and >50 years old with naturally menopausal women. C. compared women who had bilateral oophorectomy ≤15 years and >15 years prior to randomization with naturally menopausal women. D. compared women who had hysterectomy with ovarian conservation ≤50 years old, and >50 years old with naturally menopausal women. E. compared women who had hysterectomy with ovarian conservation ≤15 years and >15 years prior to randomization with naturally menopausal women. The mean CIMT progression rate in each stratum was calculated from mixed-effects models with adjustment for age (years), treatment assignment (hormone therapy, placebo), BMI, and time since menopause (< 6 years, ≥ 10 years). Lines indicate 95% confidence intervals of the mean CIMT progression rate (μm/y); P-value from the Wald test.

Figure 1.. Differential Timing of Bilateral Oophorectomy Association with CIMT Progression Rate (μm/y)

A. compared women who underwent concurrent bilateral oophorectomy and hysterectomy, and women who underwent hysterectomy with ovarian conservation, with naturally postmenopausal women. B. compared women who had bilateral oophorectomy ≤50 years old, and >50 years old with naturally menopausal women. C. compared women who had bilateral oophorectomy ≤15 years and >15 years prior to randomization with naturally menopausal women. D. compared women who had hysterectomy with ovarian conservation ≤50 years old, and >50 years old with naturally menopausal women. E. compared women who had hysterectomy with ovarian conservation ≤15 years and >15 years prior to randomization with naturally menopausal women. The mean CIMT progression rate in each stratum was calculated from mixed-effects models with adjustment for age (years), treatment assignment (hormone therapy, placebo), BMI, and time since menopause (< 6 years, ≥ 10 years). Lines indicate 95% confidence intervals of the mean CIMT progression rate (μm/y); P-value from the Wald test.