Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Carotid Wall Thickness and Years Since Bilateral Oophorectomy: The Los Angeles Atherosclerosis Study

Carotid Wall Thickness and Years Since Bilateral Oophorectomy: The Los Angeles Atherosclerosis Study Abstract Evidence that coronary heart disease risk increases with surgical menopause is consistent. However, findings concerning atherosclerosis and surgical menopause are inconsistent. The Los Angeles Atherosclerosis Study (1995–1996) assessed the cross-sectional relation at baseline between years since bilateral oophorectomy and common carotid artery intima-media thickness (IMT). Participants included 269 employed California women asymptomatic for cardiovascular disease and aged 45–60 years. Ninety-seven women reported a hysterectomy: 42 without oophorectomy or a unilateral oophorectomy and 55 with a concurrent bilateral oophorectomy. IMT was measured bilaterally with B-mode ultrasound and was regressed on age, height, and years since hysterectomy in each group. Among women who had undergone bilateral oophorectomy, IMT was significantly related to years since hysterectomy (β = 0.042 (standard error, 0.018) mm/10 years, p = 0.02). However, IMT was unrelated to years since hysterectomy in the no bilateral oophorectomy group (β = 0.005 (standard error, 0.023) mm/10 years, p = 0.82). Adjustment for high density lipoprotein or low density lipoprotein cholesterol attenuated the association between IMT and years since hysterectomy by about a fourth in the bilateral oophorectomy group. Since over 90% of this group had a history of hormone replacement therapy use, the finding that years since bilateral oophorectomy was associated with increasing atherosclerosis conflicts with a well-known finding that such therapy reverses the adverse effect of bilateral oophorectomy on coronary heart disease. atherosclerosis; cardiovascular diseases; carotid arteries; hysterectomy; ovariectomy Abbreviations: ARIC, Atherosclerosis Risk in Communities; HRT, hormone replacement therapy; IMT, intima-media thickness; SE, standard error. Received for publication October 25, 2001; accepted for publication April 24, 2002. Cardiovascular disease is a leading cause of morbidity and mortality among women in the United States, with the disease presenting 10–15 years later than in men (1). One hypothesis to explain this difference is the protective role that ovarian hormones play in delaying the initiation and progression of atherosclerosis in women, a benefit that may be reduced with both surgical (2) and natural (3) menopause. Several studies have reported that women who have undergone bilateral oophorectomy are at increased risk of developing cardiovascular disease relative to premenopausal women and women who underwent simple hysterectomy (2, 4–8). While these findings implicate endogenous ovarian hormones in preventing atherothrombotic events, the failure of opposed hormone replacement therapy (HRT) in randomized trials to provide protection against such events (9) or progression of coronary stenosis (10) in randomized trials points to the need for a better understanding of the effects of endogenous ovarian hormones on atherosclerosis and thrombosis. The current study examined the association between intima-media thickness (IMT) of the common carotid artery and number of years since hysterectomy in two groups: 1) women who had undergone a hysterectomy and bilateral oophorectomy and 2) women who had undergone a hysterectomy but had an intact ovary. MATERIALS AND METHODS Subjects The Los Angeles Atherosclerosis Study comprised employees of a major utility company in southern California and has been described previously (11). Briefly, the sample included 269 women aged 45–60 years who reported no history of heart attack, angina, revascularization, or stroke at enrollment. Participants were randomly sampled from all employees, with oversampling of smokers and Hispanics. The participation rate was 85 percent. Participants were primarily non-Hispanic Whites and Hispanics. The study protocol was approved by the Institutional Review Board of the University of Southern California Keck School of Medicine (Los Angeles). All participants signed an approved informed consent. Measurements All participants completed a baseline examination (1995–1996), which included a detailed nurse-assisted questionnaire, physical examination (including height and weight), blood pressure measurements (using a standard sphygmomanometer), venipuncture, smoking history, alcohol intake history, and 24-hour dietary recalls. Abdominal dimensions were assessed by the sagittal and transverse diameters. Participants were instructed to fast overnight, and examinations occurred during the morning in a mobile unit driven to the worksite. Carotid IMT Common carotid artery IMT was measured in the far wall of the right and left arteries by using high-resolution B-mode ultrasound (Ultramark 4 plus, 7.5-MHz linear transducer; Advanced Technology Laboratories, Bothell, Washington). The protocols for image acquisition and processing have been reported previously (12). Briefly, average thickness over a 1-cm segment of the common carotid artery 0.25-cm proximal to the dilatation of the bulb was measured using automated edge-tracking software. Reproducibility of IMT was high, with an average absolute difference of 0.022 mm (coefficient of variation, 2.8 percent) between repeated scans by two sonographers (12). Maximum IMT was measured as the average of the four maximums in the two arteries in the two body positions (supine and lateral). Reproducibility was lower than for average IMT (mean absolute difference between repeated scans, 0.043 mm; coefficient of variation, 4.6 percent) (12). Reproductive history and hormone intake Date of hysterectomy, type of concurrent ovarian surgery (unilateral or bilateral oophorectomy), and history of exogenous hormone intake were assessed by self-report. Of the 269 women in the sample, 97 reported a history of hysterectomy. Fifty-five of these women (56.7 percent) reported having undergone a bilateral oophorectomy, and 42 (43.3 percent) had had either a unilateral oophorectomy (n = 7) or no oophorectomy (n = 35). The correlation between self-reported year of hysterectomy for two examinations (separated by 3 years) was 0.96, indicating strong reliability for this measure. Duration of hormone replacement medication use was ascertained. Reproducibility for this measure was also high (r = 0.90). For the current analysis, women were grouped as current, former, or never users of opposed or unopposed medications. Statistical analysis Mean and maximum IMT were regressed on age and years since hysterectomy. Years since hysterectomy was equivalent to years since bilateral oophorectomy. To assess possible confounding with other determinants of atherosclerosis, separate models were estimated by including smoking status (current and former), HRT use (current and former), current alcohol intake, and use of blood pressure- or lipid-lowering medications. Additional models were also estimated that included potential mediators of an effect of bilateral oophorectomy on atherosclerosis: total, low density lipoprotein, and high density lipoprotein cholesterol; ratio of sagittal to transverse abdominal diameters; body mass index (kg/m2); and systolic blood pressure and pulse pressure. In this paper, metric regression coefficients with standard errors are reported. RESULTS Mean values and standard deviations of age, sagittal and transverse diameter measurements of body fat, systolic blood pressure, lipid levels, and other covariates, by oophorectomy group, are reported in table 1. No significant differences were found between the groups who had and had not undergone bilateral oophorectomy regarding age, smoking status, current alcohol intake, former or current hormone use, and use of antihypertensive or cholesterol-lowering medications. Mean total, high density lipoprotein, and low density lipoprotein cholesterol levels; estimates of body fat composition; average systolic blood pressure and IMT; and years since hysterectomy were also comparable in the two groups. Mean pulse pressure was significantly higher in the bilateral oophorectomy group (p = 0.03). Notably, over 70 percent of both groups reported current HRT use. Means, by oophorectomy status and years since hysterectomy, are presented in table 2. Years since hysterectomy was collapsed into three groups: less than 10 years, 10–20 years, and more than 20 years. Years since bilateral oophorectomy was significantly correlated with mean IMT and, more strongly, with maximum IMT. Years since bilateral oophorectomy was also positively correlated with the sagittal to transverse measure of central adiposity (r = 0.27) and negatively correlated with high density lipoprotein cholesterol (r = –  0.27). Ethnicity was not significantly correlated with years since hysterectomy nor with mean and maximum IMT. Since maximum IMT, compared with mean IMT, was more strongly correlated with years since bilateral oophorectomy, subsequent analyses were confined to examining this measure of early atherosclerosis. To further investigate these relations, maximum IMT was regressed on years since bilateral oophorectomy in various models with differing potential confounding and mediating factors as covariates (table 3). For these models, we report the unstandardized regression coefficient for an interval of 10 years since bilateral oophorectomy. For example, in the first model, maximum IMT was regressed on years since bilateral oophorectomy only; a significant association was found between the two variables among the women who had undergone bilateral oophorectomy, indicating that maximum IMT increased on average β = 0.045 (standard error (SE), 0.018) mm per 10 years of time since surgery (p = 0.01; table 3, last two columns). After adjustment for age, the slope for years since surgery was reduced only slightly for the women reporting bilateral oophorectomy (β = 0.042 (SE, 0.018) mm/10 years, p = 0.02). In contrast, years since surgery was not significantly associated with maximum IMT in any of the models for the women who had not undergone bilateral oophorectomy. After adjustment for age, the slope for maximum IMT regressed on years since surgery was very close to the null (β = 0.005 (SE, 0.023) mm/10 years, p = 0.82) in the group that had not undergone bilateral oophorectomy. The covariate that accounted for the largest proportion of the association between maximum IMT and years since surgery was high density lipoprotein cholesterol. Adjustment for this covariate reduced the coefficient by 29 percent (from 0.042 to 0.030 mm/10 years, table 3). Despite the significant difference in pulse pressure between the two groups (table 1), adjustment for pulse pressure did not explain a substantial portion of the relation between IMT and years since surgery in the bilateral oophorectomy group (table 3) because of the weak correlation between pulse pressure and years since surgery in this group (Spearman r = 0.19, p = 0.17). The effect of time since bilateral oophorectomy on maximum IMT is illustrated in figure 1. After adjustment for age and smoking, maximum IMT in the women who had undergone bilateral oophorectomy increased significantly over 10-year categories of years since bilateral oophorectomy (p for trend = 0.03). Across these same categories, no increase in maximum IMT was observed in the women who had not undergone bilateral oophorectomy. Figure 2 illustrates the hypothetical case of two groups of women of comparable ages who underwent hysterectomy at age 49 years. Note that no IMT measurements were made prior to surgery, so the portion of the graph corresponding to ages prior to 49 years (the equivalent progression rates in the two groups) is speculative. After adjustment for HRT use and other potential confounding variables, maximum IMT increased more rapidly in the women who had undergone bilateral oophorectomy than in those who had had a hysterectomy but no bilateral oophorectomy. Figure 2 depicts estimates from a regression model for those women who had and had not undergone bilateral oophorectomy, with a term for interaction between years since hysterectomy and group (without vs. with bilateral oophorectomy, p for interaction = 0.07). This model was also adjusted for age and smoking. DISCUSSION Since subclinical atherosclerosis is a strong predictor of cardiovascular events in women (13–15), we examined carotid IMT in two groups of women who had undergone hysterectomy: those with and without concurrent bilateral oophorectomy. Among the women who had had a bilateral oophorectomy, IMT was significantly related to years since surgery in an age-adjusted model (β = 0.042 (SE, 0.018) mm/10 years, p = 0.02). However, IMT was unrelated to years since hysterectomy in the group that had not undergone bilateral oophorectomy (β = 0.005 (SE, 0.023) mm/10 years, p = 0.82). Adjustment for high density lipoprotein or low density lipoprotein cholesterol attenuated this association by about a fourth in the bilateral oophorectomy group. The finding that years since bilateral oophorectomy was significantly related to carotid IMT supports the hypothesis that ovarian hormones are protective against development of atherosclerosis. The finding that years since hysterectomy was not related to IMT among women who had not undergone bilateral oophorectomy supports the interpretation that the relation of IMT with time since bilateral oophorectomy was not due to confounding by factors associated with hysterectomy and years since surgery. Increased risk of coronary heart disease among women who have undergone bilateral oophorectomy has been observed in several studies (2, 4–8). Furthermore, Colditz et al. found that the increased risk of coronary heart disease associated with bilateral oophorectomy was ameliorated for women using HRT (2), suggesting that loss of estrogen production may mediate the adverse cardiovascular effect of bilateral oophorectomy. However, findings concerning the impact of bilateral oophorectomy on preclinical atherosclerosis are inconsistent (16, 17), raising the possibility that ovarian hormone effects on events occur via a thrombotic or hemodynamic pathway. A study of aortic calcium (an indicator of intimal atherosclerosis) in women found bilateral oophorectomy to be associated with a fivefold increase in the likelihood of detectible calcium (16). On the other hand, the Atherosclerosis Risk in Communities (ARIC) Study of carotid IMT failed to detect an increase in IMT in women who had experienced surgical menopause (17). In contrast to the null findings in the ARIC Study, we found that duration of time since surgical menopause was associated with increased IMT in women who had undergone bilateral oophorectomy relative to women who had had a hysterectomy but still had one or both ovaries. In the current study, use of a comparison hysterectomy group plus use of years since surgery as a measure of exposure may have enabled us to detect an adverse effect of bilateral oophorectomy not evident in the group comparisons used in the ARIC analysis. The ARIC analysis did investigate duration since menopause and IMT (17) but not in the surgical menopause group. Ziegler-Johnson et al. found an increase in IMT in Black women who had undergone hysterectomy but no oophorectomy compared with Black women who had not had a hysterectomy (18). These authors speculated that hysterectomy alone may have an atherogenic effect due to the loss of hormones released by the uterus (19). In the current study, the effect of hysterectomy was not investigated, since all women included had a history of hysterectomy. However, the failure to find a relation between years since surgery and IMT among women with one or two ovaries argues against a strong hysterectomy effect. Several large epidemiologic studies found a reduced risk of cardiovascular disease among women using HRT (20–22), and a meta-analysis estimated a 50 percent reduction in the risk of coronary heart disease (23). Fifteen of 16 prospective studies included in the meta-analysis observed a protective association of HRT with risk of coronary heart disease. In contrast to the consistent epidemiologic evidence, two recent randomized trials found no impact of HRT on secondary prevention of coronary events (9) or progression of coronary stenosis (10). The Heart and Estrogen/progestin Replacement Study (9) and a large epidemiologic cohort study (24) suggested that short-term (1-year) cardiovascular effects of HRT may be adverse while longer-term effects are beneficial. Our finding that bilateral oophorectomy was associated with increased IMT, even though over 90 percent of these women had a history of HRT use, suggests that HRT, as implemented in the general population, may not entirely compensate for the loss of estrogen (25) (or loss of the overall balance of multiple sex hormones (26)) that occurs with a bilateral oophorectomy. This finding tends to conflict with results reported from the large Nurses’ Health Study (2, 27). Strengths of the current study include use of a highly reproducible measurement of IMT (12) and random sampling from a healthy population with a high participation rate. Such sampling and participation avoided biases that arise with self-selection. Finally, selection of women free from symptomatic cardiovascular disease, in combination with a measure of preclinical atherosclerosis, avoided biases due to differential treatments. Observational studies are subject to inherent limitations. Oophorectomy status may be confounded with unmeasured factors such as other diseases that promote atherosclerosis. If the onset of such a disease was proximate in time to surgery, then the duration-related increase in IMT in the group that had undergone bilateral oophorectomy could be due to progress of that disease rather than ovary loss. However, if the bilateral oophorectomy was due to a chronic condition, then it would be expected to produce a larger association with age (rather than years since surgery) in this group. Years since surgery could also be confounded with hormone use. Given the high prevalence of HRT among the women who underwent bilateral oophorectomy (93 percent), atherosclerosis would increase in this group only if HRT had an adverse effect. Thus, the adverse effect of years since surgery in this group could be due, in part, to an adverse effect of HRT. However, 83 percent of the women who had undergone hysterectomy but no bilateral oophorectomy also had a history of HRT use, yet no association between years since surgery and IMT was observed in this group. It remains possible that HRT use in the bilateral oophorectomy group was somehow different in terms of composition, prescribed dose, or compliance over several decades. These alternative explanations could be investigated in a prospective design in which IMT was determined prior to surgery, with follow-up IMT and monitoring of exogenous hormone intake over multiple years. Another potential limitation is the narrow age range of the sample at examination, such that years since surgery was strongly associated with age at surgery and calendar year of surgery. This limited age range potentially confounded hysterectomy with stages of aging or changes in surgical procedures and prescription practices. However, all of these factors should have been distributed similarly in the two groups (those who had and had not undergone bilateral oophorectomy), so this potential confounding within groups may not have introduced bias into estimates of group differences. Another limitation of the current study design was the small sample size; however, this limitation was compensated to some extent by the precision of the IMT measurement (12). The results from our study suggest that production of endogenous ovarian hormones in women who have undergone hysterectomy (but not bilateral oophorectomy) may play a role in protection against the development of subclinical atherosclerosis, which may then increase the risk of subsequent cardiovascular disease. Our finding that IMT was reduced in women who had undergone hysterectomy but not bilateral oophorectomy (figure 2) relative to women who had had a hysterectomy and bilateral oophorectomy, despite the high prevalence of HRT use, suggests that endogenous hormones may offer protection that currently prescribed replacement hormones do not. Adjustment for high density lipoprotein cholesterol accounted for the largest portion of this association, further suggesting that loss of ovarian hormones impacts atherosclerosis via pathways that involve high density lipoprotein cholesterol. ACKNOWLEDGMENTS This work was supported by grant HL-49910 from the National Heart, Lung, and Blood Institute. The authors acknowledge the contributions of Lynne Mellett, who died prior to completion of this paper. Reprint requests to Dr. Kathleen M. Dwyer, Department of Preventive Medicine, Keck School of Medicine, 1000 South Fremont Avenue (U.S. Post = Unit 8, Courier = Bldg. 6A, Room A6123), Alhambra, CA 91803 (e-mail: kdwyer@hsc.usc.edu or jimdwye@usc.edu). View largeDownload slide FIGURE 1. Carotid atherosclerosis and number of years since hysterectomy, The Los Angeles Atherosclerosis Study, 1995–1996. Error bars represent standard errors. The sample size in each group, according to years since hysterectomy, is as follows: Women without bilateral oophorectomy (hatched bars): <10, n = 12; 10–20, n = 22; >20, n = 8. Women with bilateral oophorectomy (white bars): <10, n = 21; 10–20, n = 19; >20, n = 15. IMT, intima-media thickness. View largeDownload slide FIGURE 1. Carotid atherosclerosis and number of years since hysterectomy, The Los Angeles Atherosclerosis Study, 1995–1996. Error bars represent standard errors. The sample size in each group, according to years since hysterectomy, is as follows: Women without bilateral oophorectomy (hatched bars): <10, n = 12; 10–20, n = 22; >20, n = 8. Women with bilateral oophorectomy (white bars): <10, n = 21; 10–20, n = 19; >20, n = 15. IMT, intima-media thickness. View largeDownload slide FIGURE 2. Idealized time path of common carotid wall thickness for two women, both of whom underwent a hysterectomy at age 49 years (y): one had a bilateral oophorectomy (dashed line) and one did not (solid line), The Los Angeles Atherosclerosis Study, 1995–1996. The increase in intima-media thickness (IMT) with age was not derived from longitudinal data but rather from the following cross-sectional regression model: maximum IMT = α + β1Age + β2YrHyst + β3Ooph + γOoph × YrHyst + covariates, where Age is chronologic age at the time of examination, YrHyst is years since hysterectomy, Ooph is bilateral oophorectomy status (1 = yes, 0 = no), and Ooph × YrHyst is the interaction term between oophorectomy status and years since hysterectomy. Covariates were cigarette smoking (current/former/never), alcohol intake (current g/day), use of hormone replacement therapy (current/former/never), and interactions between oophorectomy status and these factors. View largeDownload slide FIGURE 2. Idealized time path of common carotid wall thickness for two women, both of whom underwent a hysterectomy at age 49 years (y): one had a bilateral oophorectomy (dashed line) and one did not (solid line), The Los Angeles Atherosclerosis Study, 1995–1996. The increase in intima-media thickness (IMT) with age was not derived from longitudinal data but rather from the following cross-sectional regression model: maximum IMT = α + β1Age + β2YrHyst + β3Ooph + γOoph × YrHyst + covariates, where Age is chronologic age at the time of examination, YrHyst is years since hysterectomy, Ooph is bilateral oophorectomy status (1 = yes, 0 = no), and Ooph × YrHyst is the interaction term between oophorectomy status and years since hysterectomy. Covariates were cigarette smoking (current/former/never), alcohol intake (current g/day), use of hormone replacement therapy (current/former/never), and interactions between oophorectomy status and these factors. TABLE 1. Characteristics of the study sample,* by bilateral oophorectomy status, The Los Angeles Atherosclerosis Study, 1995–1996 Characteristic  No bilateral oophorectomy (n = 42)  Bilateral oophorectomy (n = 55)  p for difference between groups  Age (years)  51.2 (4.2)  52.6 (4.4)  0.14  Body mass index (kg/m2)  27.0 (4.8)  26.9 (4.2)  0.89  Ratio of sagittal to transverse abdominal diameters  0.63 (0.05)  0.64 (0.05)  0.57  Systolic blood pressure (mmHg)  125.0 (17.0)  130.1 (15.4)  0.12  Pulse pressure (mmHg)  36.7 (10.8)  42.1 (12.5)  0.03  Average IMT† (mm)  0.66 (0.07)  0.66 (0.08)  0.74  Maximum IMT (mm)  0.80 (0.1)  0.79 (0.1)  0.80  Total cholesterol (mmol/liter)  5.6 (0.8)  5.6 (1.0)  0.83  Low density lipoprotein cholesterol (mmol/liter)  3.3 (0.8)  3.3 (0.9)  0.99  High density lipoprotein cholesterol (mmol/liter)  1.7 (0.3)  1.7 (0.4)  0.51  Current smoker  28.6  18.2  0.23  Former smoker  33.3  34.5  0.71  Current alcohol intake (g/day)  2.6 (5.3)  2.7 (7.3)  0.91  Blood pressure medication  21.4  12.7  0.26  Cholesterol medication  9.5  3.6  0.24  No. of years since hysterectomy  14.2 (7.4)  12.9 (8.4)  0.42  Current HRT†   71.4  81.8  0.23  Former HRT   11.9  10.9  0.88  Characteristic  No bilateral oophorectomy (n = 42)  Bilateral oophorectomy (n = 55)  p for difference between groups  Age (years)  51.2 (4.2)  52.6 (4.4)  0.14  Body mass index (kg/m2)  27.0 (4.8)  26.9 (4.2)  0.89  Ratio of sagittal to transverse abdominal diameters  0.63 (0.05)  0.64 (0.05)  0.57  Systolic blood pressure (mmHg)  125.0 (17.0)  130.1 (15.4)  0.12  Pulse pressure (mmHg)  36.7 (10.8)  42.1 (12.5)  0.03  Average IMT† (mm)  0.66 (0.07)  0.66 (0.08)  0.74  Maximum IMT (mm)  0.80 (0.1)  0.79 (0.1)  0.80  Total cholesterol (mmol/liter)  5.6 (0.8)  5.6 (1.0)  0.83  Low density lipoprotein cholesterol (mmol/liter)  3.3 (0.8)  3.3 (0.9)  0.99  High density lipoprotein cholesterol (mmol/liter)  1.7 (0.3)  1.7 (0.4)  0.51  Current smoker  28.6  18.2  0.23  Former smoker  33.3  34.5  0.71  Current alcohol intake (g/day)  2.6 (5.3)  2.7 (7.3)  0.91  Blood pressure medication  21.4  12.7  0.26  Cholesterol medication  9.5  3.6  0.24  No. of years since hysterectomy  14.2 (7.4)  12.9 (8.4)  0.42  Current HRT†   71.4  81.8  0.23  Former HRT   11.9  10.9  0.88  * Values are expressed as mean (standard deviation) or percent. † IMT, intima-media thickness (common carotid); HRT, hormone replacement therapy. View Large TABLE 2. Mean (standard deviation) values of variables, by oophorectomy status and number of years since hysterectomy (10-year intervals), The Los Angeles Atherosclerosis Study, 1995–1996 Variable  No. of years since hysterectomy  <10  10–20  >20  Hysterectomy without bilateral oophorectomy        No. of subjects  12  22  8  Age (years) at examination  48.3 (3.5)  52.5 (3.7)  52.3 (4.7)  Age (years) at hysterectomy  41.9 (3.4)  38.0 (4.2)  26.8 (4.9)  No. of years since hysterectomy  6.3 (2.7)  14.5 (2.7)  25.5 (6.1)  Hormone use (no. of years)  1.6 (2.3)  8.4 (5.8)  4.9 (6.5)  Average IMT* (mm)  0.655 (0.088)  0.653 (0.070)  0.662 (0.061)  Maximum IMT (mm)  0.792 (0.127)  0.799 (0.090)  0.817 (0.094)          Hysterectomy with bilateral oophorectomy        No. of subjects  21  19  15  Age (years) at examination  51.2 (4.3)  52.7 (3.9)  54.3 (4.8)  Age (years) at hysterectomy  46.8 (5.6)  38.8 (5.4)  30.7 (5.3)  No. of years since hysterectomy  4.4 (2.5)  13.9 (3.1)  23.5 (4.6)  Hormone use (no. of years)  5.1 (3.0)  9.9 (6.1)  14.0 (9.7)  Average IMT (mm)  0.644 (0.092)  0.664 (0.067)  0.679 (0.085)  Maximum IMT (mm)  0.758 (0.115)  0.807 (0.096)  0.832 (0.126)  Variable  No. of years since hysterectomy  <10  10–20  >20  Hysterectomy without bilateral oophorectomy        No. of subjects  12  22  8  Age (years) at examination  48.3 (3.5)  52.5 (3.7)  52.3 (4.7)  Age (years) at hysterectomy  41.9 (3.4)  38.0 (4.2)  26.8 (4.9)  No. of years since hysterectomy  6.3 (2.7)  14.5 (2.7)  25.5 (6.1)  Hormone use (no. of years)  1.6 (2.3)  8.4 (5.8)  4.9 (6.5)  Average IMT* (mm)  0.655 (0.088)  0.653 (0.070)  0.662 (0.061)  Maximum IMT (mm)  0.792 (0.127)  0.799 (0.090)  0.817 (0.094)          Hysterectomy with bilateral oophorectomy        No. of subjects  21  19  15  Age (years) at examination  51.2 (4.3)  52.7 (3.9)  54.3 (4.8)  Age (years) at hysterectomy  46.8 (5.6)  38.8 (5.4)  30.7 (5.3)  No. of years since hysterectomy  4.4 (2.5)  13.9 (3.1)  23.5 (4.6)  Hormone use (no. of years)  5.1 (3.0)  9.9 (6.1)  14.0 (9.7)  Average IMT (mm)  0.644 (0.092)  0.664 (0.067)  0.679 (0.085)  Maximum IMT (mm)  0.758 (0.115)  0.807 (0.096)  0.832 (0.126)  * IMT, intima-media thickness. View Large TABLE 3. Regression of maximum intima-media thickness on number of years since hysterectomy, with potential confounding and mediating factors as covariates, The Los Angeles Atherosclerosis Study, 1995–1996 Covariate  No bilateral oophorectomy    Bilateral oophorectomy  β* (standard error)  p value    β* (standard error)  p value  None (no. of years since surgical menopause only)  0.026 (0.021)  0.23    0.045 (0.018)  0.01  Age  0.005 (0.023)  0.82    0.042 (0.018)  0.02  Age, current/former smoking  –0.001 (0.022)  0.97    0.042 (0.019)  0.03  Age, current/former HRT† use  0.001 (0.022)  0.95    0.042 (0.020)  0.04  Age, current alcohol intake   –0.009 (0.022)  0.69    0.039 (0.019)  0.04  Age, use of blood pressure/cholesterol medications  0.012 (0.024)  0.63    0.036 (0.019)  0.06  Age, total cholesterol  0.009 (0.022)  0.67    0.042 (0.018)  0.03  Age, low density lipoprotein cholesterol  –0.005 (0.024)  0.85    0.031 (0.020)  0.13  Age, high density lipoprotein cholesterol  0.009 (0.021)  0.66    0.030 (0.018)  0.11  Age, ratio of sagittal to transverse abdominal diameters  0.005 (0.023)  0.84    0.033 (0.019)  0.10  Age, body mass index  0.006 (0.023)  0.79    0.038 (0.017)  0.03  Age, systolic blood pressure  0.002 (0.023)  0.92    0.035 (0.018)  0.05  Age, pulse pressure  0.003 (0.023)  0.90    0.037 (0.018)  0.05  Covariate  No bilateral oophorectomy    Bilateral oophorectomy  β* (standard error)  p value    β* (standard error)  p value  None (no. of years since surgical menopause only)  0.026 (0.021)  0.23    0.045 (0.018)  0.01  Age  0.005 (0.023)  0.82    0.042 (0.018)  0.02  Age, current/former smoking  –0.001 (0.022)  0.97    0.042 (0.019)  0.03  Age, current/former HRT† use  0.001 (0.022)  0.95    0.042 (0.020)  0.04  Age, current alcohol intake   –0.009 (0.022)  0.69    0.039 (0.019)  0.04  Age, use of blood pressure/cholesterol medications  0.012 (0.024)  0.63    0.036 (0.019)  0.06  Age, total cholesterol  0.009 (0.022)  0.67    0.042 (0.018)  0.03  Age, low density lipoprotein cholesterol  –0.005 (0.024)  0.85    0.031 (0.020)  0.13  Age, high density lipoprotein cholesterol  0.009 (0.021)  0.66    0.030 (0.018)  0.11  Age, ratio of sagittal to transverse abdominal diameters  0.005 (0.023)  0.84    0.033 (0.019)  0.10  Age, body mass index  0.006 (0.023)  0.79    0.038 (0.017)  0.03  Age, systolic blood pressure  0.002 (0.023)  0.92    0.035 (0.018)  0.05  Age, pulse pressure  0.003 (0.023)  0.90    0.037 (0.018)  0.05  * Unstandardized regression coefficient for an interval of 10 years since hysterectomy; for example, a coefficient of 0.026 (0.021) indicates that intima-media thickness increased on average 0.026 mm per 10 years of time since hysterectomy. † HRT, hormone replacement therapy. View Large References 1. Eaker ED, Chesebro JH, Sacks FM, et al. AHA Medical/Scientific Statement: cardiovascular disease in women. Circulation  1993; 88(4 pt 1): 1999–2009. Google Scholar 2. Colditz GA, Willett WC, Stampfer MJ, et al. Menopause and the risk of coronary heart disease in women. N Engl J Med  1987; 316: 1105–10. Google Scholar 3. Kannel WB, Hjortland MC, McNamara PM, et al. Menopause and risk of cardiovascular disease: The Framingham Study. Ann Intern Med  1976; 85: 447–52. Google Scholar 4. Oliver M, Boyd G. Effect of bilateral ovariectomy on coronary-artery disease and serum-lipid levels. Lancet  1959; 2: 690–4. Google Scholar 5. Wuest J, Dry T, Edwards J. The degree of coronary atherosclerosis in bilaterally oophorectomized women. Am Heart J  1953; 7: 801–9. Google Scholar 6. Parrish H, Carr C, Hall D, et al. Time interval from castration in premenopausal women to development of excessive coronary atherosclerosis. Am J Obstet Gynecol  1967; 99: 155–62. Google Scholar 7. Gordon T, Kannel WB, Hjortland MC, et al. Menopause and coronary heart disease. The Framingham Study. Ann Intern Med  1978; 89: 157–61. Google Scholar 8. Punnonen R, Ikalainen M, Seppala E. Premenopausal hysterectomy and risk of cardiovascular disease. (Letter). Lancet  1987; 1: 1139. Google Scholar 9. Hulley S, Grady D, Bush T, et al. Randomized trial of estrogen plus progestin for secondary prevention of coronary heart disease in postmenopausal women. Heart and Estrogen/progestin Replacement Study (HERS) Research Group. JAMA  1998; 280: 605–13. Google Scholar 10. Herrington DM, Reboussin DM, Brosnihan KB, et al. Effects of estrogen replacement on the progression of coronary-artery atherosclerosis. N Engl J Med  2000; 343: 522–9. Google Scholar 11. Sun P, Dwyer KM, Merz CN, et al. Blood pressure, LDL cholesterol, and intima-media thickness: a test of the “response to injury” hypothesis of atherosclerosis. Arterioscler Thromb Vasc Biol  2000; 20: 2005–10. Google Scholar 12. Dwyer JH, Sun P, Kwong-Fu H, et al. Automated intima-media thickness: The Los Angeles Atherosclerosis Study. Ultrasound Med Biol  1998; 24: 981–7. Google Scholar 13. Chambless LE, Heiss G, Folsom AR, et al. Association of coronary heart disease incidence with carotid arterial wall thickness and major risk factors: the Atherosclerosis Risk in Communities (ARIC) Study, 1987–1993. Am J Epidemiol  1997; 146: 483–94. Google Scholar 14. O’Leary DH, Polak JF, Kronmal RA, et al. Carotid-artery intima and media thickness as a risk factor for myocardial infarction and stroke in older adults. Cardiovascular Health Study Collaborative Research Group. N Engl J Med  1999; 340: 14–22. Google Scholar 15. Bots ML, Hoes AW, Hofman A, et al. Cross-sectionally assessed carotid intima-media thickness relates to long-term risk of stroke, coronary heart disease and death as estimated by available risk functions. J Intern Med  1999; 245: 269–76. Google Scholar 16. Witteman JC, Grobbee DE, Kok FJ, et al. Increased risk of atherosclerosis in women after the menopause. BMJ  1989; 298: 642–4. Google Scholar 17. Nabulsi A, Folsom A, Szklo M, et al. No association of menopause and hormone replacement therapy with carotid artery intima-media thickness. Circulation  1996; 94: 1857–63. Google Scholar 18. Zeigler-Johnson CM, Holmes JL, Lassila HC, et al. Subclinical atherosclerosis in relation to hysterectomy status in black women. Stroke  1998; 29: 759–64. Google Scholar 19. Chalmers C. Does hysterectomy in a premenopausal woman affect ovarian function? Med Hypotheses  1996; 46: 573–5. Google Scholar 20. Stampfer MJ, Willett WC, Colditz GA, et al. A prospective study of postmenopausal estrogen therapy and coronary heart disease. N Engl J Med  1985; 313: 1044–9. Google Scholar 21. Bush TL, Barrett-Connor E, Cowan LD, et al. Cardiovascular mortality and noncontraceptive use of estrogen in women: results from the Lipid Research Clinics Program Follow-up Study. Circulation  1987; 75: 1102–9. Google Scholar 22. Henderson BE, Paganini-Hill A, Ross RK. Decreased mortality in users of estrogen replacement therapy. Arch Intern Med  1991; 151: 75–8. Google Scholar 23. Stampfer MJ, Colditz GA. Estrogen replacement therapy and coronary heart disease: a quantitative assessment of the epidemiologic evidence. Prev Med  1991; 20: 47–63. Google Scholar 24. Grodstein F, Manson JE, Stampfer MJ. Postmenopausal hormone use and secondary prevention of coronary events in the Nurses’ Health Study. A prospective, observational study. Ann Intern Med  2001; 135: 1–8. Google Scholar 25. Utian WH, Katz M, Davey DA, et al. Effect of premenopausal castration and incremental dosages of conjugated equine estrogens on plasma follicle-stimulating hormone, luteinizing hormone, and estradiol. Am J Obstet Gynecol  1978; 132: 297–302. Google Scholar 26. Liu Y, Ding J, Bush TL, et al. Relative androgen excess and increased cardiovascular risk after menopause: a hypothesized relation. Am J Epidemiol  2001; 154: 489–94. Google Scholar 27. Grodstein F, Stampfer MJ. Estrogen for women at varying risk of coronary disease. Maturitas  1998; 30: 19–26. Google Scholar http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png American Journal of Epidemiology Oxford University Press

Carotid Wall Thickness and Years Since Bilateral Oophorectomy: The Los Angeles Atherosclerosis Study

Loading next page...
 
/lp/oxford-university-press/carotid-wall-thickness-and-years-since-bilateral-oophorectomy-the-los-bo3TGCZPE7

References (29)

Publisher
Oxford University Press
ISSN
0002-9262
eISSN
1476-6256
DOI
10.1093/aje/kwf051
Publisher site
See Article on Publisher Site

Abstract

Abstract Evidence that coronary heart disease risk increases with surgical menopause is consistent. However, findings concerning atherosclerosis and surgical menopause are inconsistent. The Los Angeles Atherosclerosis Study (1995–1996) assessed the cross-sectional relation at baseline between years since bilateral oophorectomy and common carotid artery intima-media thickness (IMT). Participants included 269 employed California women asymptomatic for cardiovascular disease and aged 45–60 years. Ninety-seven women reported a hysterectomy: 42 without oophorectomy or a unilateral oophorectomy and 55 with a concurrent bilateral oophorectomy. IMT was measured bilaterally with B-mode ultrasound and was regressed on age, height, and years since hysterectomy in each group. Among women who had undergone bilateral oophorectomy, IMT was significantly related to years since hysterectomy (β = 0.042 (standard error, 0.018) mm/10 years, p = 0.02). However, IMT was unrelated to years since hysterectomy in the no bilateral oophorectomy group (β = 0.005 (standard error, 0.023) mm/10 years, p = 0.82). Adjustment for high density lipoprotein or low density lipoprotein cholesterol attenuated the association between IMT and years since hysterectomy by about a fourth in the bilateral oophorectomy group. Since over 90% of this group had a history of hormone replacement therapy use, the finding that years since bilateral oophorectomy was associated with increasing atherosclerosis conflicts with a well-known finding that such therapy reverses the adverse effect of bilateral oophorectomy on coronary heart disease. atherosclerosis; cardiovascular diseases; carotid arteries; hysterectomy; ovariectomy Abbreviations: ARIC, Atherosclerosis Risk in Communities; HRT, hormone replacement therapy; IMT, intima-media thickness; SE, standard error. Received for publication October 25, 2001; accepted for publication April 24, 2002. Cardiovascular disease is a leading cause of morbidity and mortality among women in the United States, with the disease presenting 10–15 years later than in men (1). One hypothesis to explain this difference is the protective role that ovarian hormones play in delaying the initiation and progression of atherosclerosis in women, a benefit that may be reduced with both surgical (2) and natural (3) menopause. Several studies have reported that women who have undergone bilateral oophorectomy are at increased risk of developing cardiovascular disease relative to premenopausal women and women who underwent simple hysterectomy (2, 4–8). While these findings implicate endogenous ovarian hormones in preventing atherothrombotic events, the failure of opposed hormone replacement therapy (HRT) in randomized trials to provide protection against such events (9) or progression of coronary stenosis (10) in randomized trials points to the need for a better understanding of the effects of endogenous ovarian hormones on atherosclerosis and thrombosis. The current study examined the association between intima-media thickness (IMT) of the common carotid artery and number of years since hysterectomy in two groups: 1) women who had undergone a hysterectomy and bilateral oophorectomy and 2) women who had undergone a hysterectomy but had an intact ovary. MATERIALS AND METHODS Subjects The Los Angeles Atherosclerosis Study comprised employees of a major utility company in southern California and has been described previously (11). Briefly, the sample included 269 women aged 45–60 years who reported no history of heart attack, angina, revascularization, or stroke at enrollment. Participants were randomly sampled from all employees, with oversampling of smokers and Hispanics. The participation rate was 85 percent. Participants were primarily non-Hispanic Whites and Hispanics. The study protocol was approved by the Institutional Review Board of the University of Southern California Keck School of Medicine (Los Angeles). All participants signed an approved informed consent. Measurements All participants completed a baseline examination (1995–1996), which included a detailed nurse-assisted questionnaire, physical examination (including height and weight), blood pressure measurements (using a standard sphygmomanometer), venipuncture, smoking history, alcohol intake history, and 24-hour dietary recalls. Abdominal dimensions were assessed by the sagittal and transverse diameters. Participants were instructed to fast overnight, and examinations occurred during the morning in a mobile unit driven to the worksite. Carotid IMT Common carotid artery IMT was measured in the far wall of the right and left arteries by using high-resolution B-mode ultrasound (Ultramark 4 plus, 7.5-MHz linear transducer; Advanced Technology Laboratories, Bothell, Washington). The protocols for image acquisition and processing have been reported previously (12). Briefly, average thickness over a 1-cm segment of the common carotid artery 0.25-cm proximal to the dilatation of the bulb was measured using automated edge-tracking software. Reproducibility of IMT was high, with an average absolute difference of 0.022 mm (coefficient of variation, 2.8 percent) between repeated scans by two sonographers (12). Maximum IMT was measured as the average of the four maximums in the two arteries in the two body positions (supine and lateral). Reproducibility was lower than for average IMT (mean absolute difference between repeated scans, 0.043 mm; coefficient of variation, 4.6 percent) (12). Reproductive history and hormone intake Date of hysterectomy, type of concurrent ovarian surgery (unilateral or bilateral oophorectomy), and history of exogenous hormone intake were assessed by self-report. Of the 269 women in the sample, 97 reported a history of hysterectomy. Fifty-five of these women (56.7 percent) reported having undergone a bilateral oophorectomy, and 42 (43.3 percent) had had either a unilateral oophorectomy (n = 7) or no oophorectomy (n = 35). The correlation between self-reported year of hysterectomy for two examinations (separated by 3 years) was 0.96, indicating strong reliability for this measure. Duration of hormone replacement medication use was ascertained. Reproducibility for this measure was also high (r = 0.90). For the current analysis, women were grouped as current, former, or never users of opposed or unopposed medications. Statistical analysis Mean and maximum IMT were regressed on age and years since hysterectomy. Years since hysterectomy was equivalent to years since bilateral oophorectomy. To assess possible confounding with other determinants of atherosclerosis, separate models were estimated by including smoking status (current and former), HRT use (current and former), current alcohol intake, and use of blood pressure- or lipid-lowering medications. Additional models were also estimated that included potential mediators of an effect of bilateral oophorectomy on atherosclerosis: total, low density lipoprotein, and high density lipoprotein cholesterol; ratio of sagittal to transverse abdominal diameters; body mass index (kg/m2); and systolic blood pressure and pulse pressure. In this paper, metric regression coefficients with standard errors are reported. RESULTS Mean values and standard deviations of age, sagittal and transverse diameter measurements of body fat, systolic blood pressure, lipid levels, and other covariates, by oophorectomy group, are reported in table 1. No significant differences were found between the groups who had and had not undergone bilateral oophorectomy regarding age, smoking status, current alcohol intake, former or current hormone use, and use of antihypertensive or cholesterol-lowering medications. Mean total, high density lipoprotein, and low density lipoprotein cholesterol levels; estimates of body fat composition; average systolic blood pressure and IMT; and years since hysterectomy were also comparable in the two groups. Mean pulse pressure was significantly higher in the bilateral oophorectomy group (p = 0.03). Notably, over 70 percent of both groups reported current HRT use. Means, by oophorectomy status and years since hysterectomy, are presented in table 2. Years since hysterectomy was collapsed into three groups: less than 10 years, 10–20 years, and more than 20 years. Years since bilateral oophorectomy was significantly correlated with mean IMT and, more strongly, with maximum IMT. Years since bilateral oophorectomy was also positively correlated with the sagittal to transverse measure of central adiposity (r = 0.27) and negatively correlated with high density lipoprotein cholesterol (r = –  0.27). Ethnicity was not significantly correlated with years since hysterectomy nor with mean and maximum IMT. Since maximum IMT, compared with mean IMT, was more strongly correlated with years since bilateral oophorectomy, subsequent analyses were confined to examining this measure of early atherosclerosis. To further investigate these relations, maximum IMT was regressed on years since bilateral oophorectomy in various models with differing potential confounding and mediating factors as covariates (table 3). For these models, we report the unstandardized regression coefficient for an interval of 10 years since bilateral oophorectomy. For example, in the first model, maximum IMT was regressed on years since bilateral oophorectomy only; a significant association was found between the two variables among the women who had undergone bilateral oophorectomy, indicating that maximum IMT increased on average β = 0.045 (standard error (SE), 0.018) mm per 10 years of time since surgery (p = 0.01; table 3, last two columns). After adjustment for age, the slope for years since surgery was reduced only slightly for the women reporting bilateral oophorectomy (β = 0.042 (SE, 0.018) mm/10 years, p = 0.02). In contrast, years since surgery was not significantly associated with maximum IMT in any of the models for the women who had not undergone bilateral oophorectomy. After adjustment for age, the slope for maximum IMT regressed on years since surgery was very close to the null (β = 0.005 (SE, 0.023) mm/10 years, p = 0.82) in the group that had not undergone bilateral oophorectomy. The covariate that accounted for the largest proportion of the association between maximum IMT and years since surgery was high density lipoprotein cholesterol. Adjustment for this covariate reduced the coefficient by 29 percent (from 0.042 to 0.030 mm/10 years, table 3). Despite the significant difference in pulse pressure between the two groups (table 1), adjustment for pulse pressure did not explain a substantial portion of the relation between IMT and years since surgery in the bilateral oophorectomy group (table 3) because of the weak correlation between pulse pressure and years since surgery in this group (Spearman r = 0.19, p = 0.17). The effect of time since bilateral oophorectomy on maximum IMT is illustrated in figure 1. After adjustment for age and smoking, maximum IMT in the women who had undergone bilateral oophorectomy increased significantly over 10-year categories of years since bilateral oophorectomy (p for trend = 0.03). Across these same categories, no increase in maximum IMT was observed in the women who had not undergone bilateral oophorectomy. Figure 2 illustrates the hypothetical case of two groups of women of comparable ages who underwent hysterectomy at age 49 years. Note that no IMT measurements were made prior to surgery, so the portion of the graph corresponding to ages prior to 49 years (the equivalent progression rates in the two groups) is speculative. After adjustment for HRT use and other potential confounding variables, maximum IMT increased more rapidly in the women who had undergone bilateral oophorectomy than in those who had had a hysterectomy but no bilateral oophorectomy. Figure 2 depicts estimates from a regression model for those women who had and had not undergone bilateral oophorectomy, with a term for interaction between years since hysterectomy and group (without vs. with bilateral oophorectomy, p for interaction = 0.07). This model was also adjusted for age and smoking. DISCUSSION Since subclinical atherosclerosis is a strong predictor of cardiovascular events in women (13–15), we examined carotid IMT in two groups of women who had undergone hysterectomy: those with and without concurrent bilateral oophorectomy. Among the women who had had a bilateral oophorectomy, IMT was significantly related to years since surgery in an age-adjusted model (β = 0.042 (SE, 0.018) mm/10 years, p = 0.02). However, IMT was unrelated to years since hysterectomy in the group that had not undergone bilateral oophorectomy (β = 0.005 (SE, 0.023) mm/10 years, p = 0.82). Adjustment for high density lipoprotein or low density lipoprotein cholesterol attenuated this association by about a fourth in the bilateral oophorectomy group. The finding that years since bilateral oophorectomy was significantly related to carotid IMT supports the hypothesis that ovarian hormones are protective against development of atherosclerosis. The finding that years since hysterectomy was not related to IMT among women who had not undergone bilateral oophorectomy supports the interpretation that the relation of IMT with time since bilateral oophorectomy was not due to confounding by factors associated with hysterectomy and years since surgery. Increased risk of coronary heart disease among women who have undergone bilateral oophorectomy has been observed in several studies (2, 4–8). Furthermore, Colditz et al. found that the increased risk of coronary heart disease associated with bilateral oophorectomy was ameliorated for women using HRT (2), suggesting that loss of estrogen production may mediate the adverse cardiovascular effect of bilateral oophorectomy. However, findings concerning the impact of bilateral oophorectomy on preclinical atherosclerosis are inconsistent (16, 17), raising the possibility that ovarian hormone effects on events occur via a thrombotic or hemodynamic pathway. A study of aortic calcium (an indicator of intimal atherosclerosis) in women found bilateral oophorectomy to be associated with a fivefold increase in the likelihood of detectible calcium (16). On the other hand, the Atherosclerosis Risk in Communities (ARIC) Study of carotid IMT failed to detect an increase in IMT in women who had experienced surgical menopause (17). In contrast to the null findings in the ARIC Study, we found that duration of time since surgical menopause was associated with increased IMT in women who had undergone bilateral oophorectomy relative to women who had had a hysterectomy but still had one or both ovaries. In the current study, use of a comparison hysterectomy group plus use of years since surgery as a measure of exposure may have enabled us to detect an adverse effect of bilateral oophorectomy not evident in the group comparisons used in the ARIC analysis. The ARIC analysis did investigate duration since menopause and IMT (17) but not in the surgical menopause group. Ziegler-Johnson et al. found an increase in IMT in Black women who had undergone hysterectomy but no oophorectomy compared with Black women who had not had a hysterectomy (18). These authors speculated that hysterectomy alone may have an atherogenic effect due to the loss of hormones released by the uterus (19). In the current study, the effect of hysterectomy was not investigated, since all women included had a history of hysterectomy. However, the failure to find a relation between years since surgery and IMT among women with one or two ovaries argues against a strong hysterectomy effect. Several large epidemiologic studies found a reduced risk of cardiovascular disease among women using HRT (20–22), and a meta-analysis estimated a 50 percent reduction in the risk of coronary heart disease (23). Fifteen of 16 prospective studies included in the meta-analysis observed a protective association of HRT with risk of coronary heart disease. In contrast to the consistent epidemiologic evidence, two recent randomized trials found no impact of HRT on secondary prevention of coronary events (9) or progression of coronary stenosis (10). The Heart and Estrogen/progestin Replacement Study (9) and a large epidemiologic cohort study (24) suggested that short-term (1-year) cardiovascular effects of HRT may be adverse while longer-term effects are beneficial. Our finding that bilateral oophorectomy was associated with increased IMT, even though over 90 percent of these women had a history of HRT use, suggests that HRT, as implemented in the general population, may not entirely compensate for the loss of estrogen (25) (or loss of the overall balance of multiple sex hormones (26)) that occurs with a bilateral oophorectomy. This finding tends to conflict with results reported from the large Nurses’ Health Study (2, 27). Strengths of the current study include use of a highly reproducible measurement of IMT (12) and random sampling from a healthy population with a high participation rate. Such sampling and participation avoided biases that arise with self-selection. Finally, selection of women free from symptomatic cardiovascular disease, in combination with a measure of preclinical atherosclerosis, avoided biases due to differential treatments. Observational studies are subject to inherent limitations. Oophorectomy status may be confounded with unmeasured factors such as other diseases that promote atherosclerosis. If the onset of such a disease was proximate in time to surgery, then the duration-related increase in IMT in the group that had undergone bilateral oophorectomy could be due to progress of that disease rather than ovary loss. However, if the bilateral oophorectomy was due to a chronic condition, then it would be expected to produce a larger association with age (rather than years since surgery) in this group. Years since surgery could also be confounded with hormone use. Given the high prevalence of HRT among the women who underwent bilateral oophorectomy (93 percent), atherosclerosis would increase in this group only if HRT had an adverse effect. Thus, the adverse effect of years since surgery in this group could be due, in part, to an adverse effect of HRT. However, 83 percent of the women who had undergone hysterectomy but no bilateral oophorectomy also had a history of HRT use, yet no association between years since surgery and IMT was observed in this group. It remains possible that HRT use in the bilateral oophorectomy group was somehow different in terms of composition, prescribed dose, or compliance over several decades. These alternative explanations could be investigated in a prospective design in which IMT was determined prior to surgery, with follow-up IMT and monitoring of exogenous hormone intake over multiple years. Another potential limitation is the narrow age range of the sample at examination, such that years since surgery was strongly associated with age at surgery and calendar year of surgery. This limited age range potentially confounded hysterectomy with stages of aging or changes in surgical procedures and prescription practices. However, all of these factors should have been distributed similarly in the two groups (those who had and had not undergone bilateral oophorectomy), so this potential confounding within groups may not have introduced bias into estimates of group differences. Another limitation of the current study design was the small sample size; however, this limitation was compensated to some extent by the precision of the IMT measurement (12). The results from our study suggest that production of endogenous ovarian hormones in women who have undergone hysterectomy (but not bilateral oophorectomy) may play a role in protection against the development of subclinical atherosclerosis, which may then increase the risk of subsequent cardiovascular disease. Our finding that IMT was reduced in women who had undergone hysterectomy but not bilateral oophorectomy (figure 2) relative to women who had had a hysterectomy and bilateral oophorectomy, despite the high prevalence of HRT use, suggests that endogenous hormones may offer protection that currently prescribed replacement hormones do not. Adjustment for high density lipoprotein cholesterol accounted for the largest portion of this association, further suggesting that loss of ovarian hormones impacts atherosclerosis via pathways that involve high density lipoprotein cholesterol. ACKNOWLEDGMENTS This work was supported by grant HL-49910 from the National Heart, Lung, and Blood Institute. The authors acknowledge the contributions of Lynne Mellett, who died prior to completion of this paper. Reprint requests to Dr. Kathleen M. Dwyer, Department of Preventive Medicine, Keck School of Medicine, 1000 South Fremont Avenue (U.S. Post = Unit 8, Courier = Bldg. 6A, Room A6123), Alhambra, CA 91803 (e-mail: kdwyer@hsc.usc.edu or jimdwye@usc.edu). View largeDownload slide FIGURE 1. Carotid atherosclerosis and number of years since hysterectomy, The Los Angeles Atherosclerosis Study, 1995–1996. Error bars represent standard errors. The sample size in each group, according to years since hysterectomy, is as follows: Women without bilateral oophorectomy (hatched bars): <10, n = 12; 10–20, n = 22; >20, n = 8. Women with bilateral oophorectomy (white bars): <10, n = 21; 10–20, n = 19; >20, n = 15. IMT, intima-media thickness. View largeDownload slide FIGURE 1. Carotid atherosclerosis and number of years since hysterectomy, The Los Angeles Atherosclerosis Study, 1995–1996. Error bars represent standard errors. The sample size in each group, according to years since hysterectomy, is as follows: Women without bilateral oophorectomy (hatched bars): <10, n = 12; 10–20, n = 22; >20, n = 8. Women with bilateral oophorectomy (white bars): <10, n = 21; 10–20, n = 19; >20, n = 15. IMT, intima-media thickness. View largeDownload slide FIGURE 2. Idealized time path of common carotid wall thickness for two women, both of whom underwent a hysterectomy at age 49 years (y): one had a bilateral oophorectomy (dashed line) and one did not (solid line), The Los Angeles Atherosclerosis Study, 1995–1996. The increase in intima-media thickness (IMT) with age was not derived from longitudinal data but rather from the following cross-sectional regression model: maximum IMT = α + β1Age + β2YrHyst + β3Ooph + γOoph × YrHyst + covariates, where Age is chronologic age at the time of examination, YrHyst is years since hysterectomy, Ooph is bilateral oophorectomy status (1 = yes, 0 = no), and Ooph × YrHyst is the interaction term between oophorectomy status and years since hysterectomy. Covariates were cigarette smoking (current/former/never), alcohol intake (current g/day), use of hormone replacement therapy (current/former/never), and interactions between oophorectomy status and these factors. View largeDownload slide FIGURE 2. Idealized time path of common carotid wall thickness for two women, both of whom underwent a hysterectomy at age 49 years (y): one had a bilateral oophorectomy (dashed line) and one did not (solid line), The Los Angeles Atherosclerosis Study, 1995–1996. The increase in intima-media thickness (IMT) with age was not derived from longitudinal data but rather from the following cross-sectional regression model: maximum IMT = α + β1Age + β2YrHyst + β3Ooph + γOoph × YrHyst + covariates, where Age is chronologic age at the time of examination, YrHyst is years since hysterectomy, Ooph is bilateral oophorectomy status (1 = yes, 0 = no), and Ooph × YrHyst is the interaction term between oophorectomy status and years since hysterectomy. Covariates were cigarette smoking (current/former/never), alcohol intake (current g/day), use of hormone replacement therapy (current/former/never), and interactions between oophorectomy status and these factors. TABLE 1. Characteristics of the study sample,* by bilateral oophorectomy status, The Los Angeles Atherosclerosis Study, 1995–1996 Characteristic  No bilateral oophorectomy (n = 42)  Bilateral oophorectomy (n = 55)  p for difference between groups  Age (years)  51.2 (4.2)  52.6 (4.4)  0.14  Body mass index (kg/m2)  27.0 (4.8)  26.9 (4.2)  0.89  Ratio of sagittal to transverse abdominal diameters  0.63 (0.05)  0.64 (0.05)  0.57  Systolic blood pressure (mmHg)  125.0 (17.0)  130.1 (15.4)  0.12  Pulse pressure (mmHg)  36.7 (10.8)  42.1 (12.5)  0.03  Average IMT† (mm)  0.66 (0.07)  0.66 (0.08)  0.74  Maximum IMT (mm)  0.80 (0.1)  0.79 (0.1)  0.80  Total cholesterol (mmol/liter)  5.6 (0.8)  5.6 (1.0)  0.83  Low density lipoprotein cholesterol (mmol/liter)  3.3 (0.8)  3.3 (0.9)  0.99  High density lipoprotein cholesterol (mmol/liter)  1.7 (0.3)  1.7 (0.4)  0.51  Current smoker  28.6  18.2  0.23  Former smoker  33.3  34.5  0.71  Current alcohol intake (g/day)  2.6 (5.3)  2.7 (7.3)  0.91  Blood pressure medication  21.4  12.7  0.26  Cholesterol medication  9.5  3.6  0.24  No. of years since hysterectomy  14.2 (7.4)  12.9 (8.4)  0.42  Current HRT†   71.4  81.8  0.23  Former HRT   11.9  10.9  0.88  Characteristic  No bilateral oophorectomy (n = 42)  Bilateral oophorectomy (n = 55)  p for difference between groups  Age (years)  51.2 (4.2)  52.6 (4.4)  0.14  Body mass index (kg/m2)  27.0 (4.8)  26.9 (4.2)  0.89  Ratio of sagittal to transverse abdominal diameters  0.63 (0.05)  0.64 (0.05)  0.57  Systolic blood pressure (mmHg)  125.0 (17.0)  130.1 (15.4)  0.12  Pulse pressure (mmHg)  36.7 (10.8)  42.1 (12.5)  0.03  Average IMT† (mm)  0.66 (0.07)  0.66 (0.08)  0.74  Maximum IMT (mm)  0.80 (0.1)  0.79 (0.1)  0.80  Total cholesterol (mmol/liter)  5.6 (0.8)  5.6 (1.0)  0.83  Low density lipoprotein cholesterol (mmol/liter)  3.3 (0.8)  3.3 (0.9)  0.99  High density lipoprotein cholesterol (mmol/liter)  1.7 (0.3)  1.7 (0.4)  0.51  Current smoker  28.6  18.2  0.23  Former smoker  33.3  34.5  0.71  Current alcohol intake (g/day)  2.6 (5.3)  2.7 (7.3)  0.91  Blood pressure medication  21.4  12.7  0.26  Cholesterol medication  9.5  3.6  0.24  No. of years since hysterectomy  14.2 (7.4)  12.9 (8.4)  0.42  Current HRT†   71.4  81.8  0.23  Former HRT   11.9  10.9  0.88  * Values are expressed as mean (standard deviation) or percent. † IMT, intima-media thickness (common carotid); HRT, hormone replacement therapy. View Large TABLE 2. Mean (standard deviation) values of variables, by oophorectomy status and number of years since hysterectomy (10-year intervals), The Los Angeles Atherosclerosis Study, 1995–1996 Variable  No. of years since hysterectomy  <10  10–20  >20  Hysterectomy without bilateral oophorectomy        No. of subjects  12  22  8  Age (years) at examination  48.3 (3.5)  52.5 (3.7)  52.3 (4.7)  Age (years) at hysterectomy  41.9 (3.4)  38.0 (4.2)  26.8 (4.9)  No. of years since hysterectomy  6.3 (2.7)  14.5 (2.7)  25.5 (6.1)  Hormone use (no. of years)  1.6 (2.3)  8.4 (5.8)  4.9 (6.5)  Average IMT* (mm)  0.655 (0.088)  0.653 (0.070)  0.662 (0.061)  Maximum IMT (mm)  0.792 (0.127)  0.799 (0.090)  0.817 (0.094)          Hysterectomy with bilateral oophorectomy        No. of subjects  21  19  15  Age (years) at examination  51.2 (4.3)  52.7 (3.9)  54.3 (4.8)  Age (years) at hysterectomy  46.8 (5.6)  38.8 (5.4)  30.7 (5.3)  No. of years since hysterectomy  4.4 (2.5)  13.9 (3.1)  23.5 (4.6)  Hormone use (no. of years)  5.1 (3.0)  9.9 (6.1)  14.0 (9.7)  Average IMT (mm)  0.644 (0.092)  0.664 (0.067)  0.679 (0.085)  Maximum IMT (mm)  0.758 (0.115)  0.807 (0.096)  0.832 (0.126)  Variable  No. of years since hysterectomy  <10  10–20  >20  Hysterectomy without bilateral oophorectomy        No. of subjects  12  22  8  Age (years) at examination  48.3 (3.5)  52.5 (3.7)  52.3 (4.7)  Age (years) at hysterectomy  41.9 (3.4)  38.0 (4.2)  26.8 (4.9)  No. of years since hysterectomy  6.3 (2.7)  14.5 (2.7)  25.5 (6.1)  Hormone use (no. of years)  1.6 (2.3)  8.4 (5.8)  4.9 (6.5)  Average IMT* (mm)  0.655 (0.088)  0.653 (0.070)  0.662 (0.061)  Maximum IMT (mm)  0.792 (0.127)  0.799 (0.090)  0.817 (0.094)          Hysterectomy with bilateral oophorectomy        No. of subjects  21  19  15  Age (years) at examination  51.2 (4.3)  52.7 (3.9)  54.3 (4.8)  Age (years) at hysterectomy  46.8 (5.6)  38.8 (5.4)  30.7 (5.3)  No. of years since hysterectomy  4.4 (2.5)  13.9 (3.1)  23.5 (4.6)  Hormone use (no. of years)  5.1 (3.0)  9.9 (6.1)  14.0 (9.7)  Average IMT (mm)  0.644 (0.092)  0.664 (0.067)  0.679 (0.085)  Maximum IMT (mm)  0.758 (0.115)  0.807 (0.096)  0.832 (0.126)  * IMT, intima-media thickness. View Large TABLE 3. Regression of maximum intima-media thickness on number of years since hysterectomy, with potential confounding and mediating factors as covariates, The Los Angeles Atherosclerosis Study, 1995–1996 Covariate  No bilateral oophorectomy    Bilateral oophorectomy  β* (standard error)  p value    β* (standard error)  p value  None (no. of years since surgical menopause only)  0.026 (0.021)  0.23    0.045 (0.018)  0.01  Age  0.005 (0.023)  0.82    0.042 (0.018)  0.02  Age, current/former smoking  –0.001 (0.022)  0.97    0.042 (0.019)  0.03  Age, current/former HRT† use  0.001 (0.022)  0.95    0.042 (0.020)  0.04  Age, current alcohol intake   –0.009 (0.022)  0.69    0.039 (0.019)  0.04  Age, use of blood pressure/cholesterol medications  0.012 (0.024)  0.63    0.036 (0.019)  0.06  Age, total cholesterol  0.009 (0.022)  0.67    0.042 (0.018)  0.03  Age, low density lipoprotein cholesterol  –0.005 (0.024)  0.85    0.031 (0.020)  0.13  Age, high density lipoprotein cholesterol  0.009 (0.021)  0.66    0.030 (0.018)  0.11  Age, ratio of sagittal to transverse abdominal diameters  0.005 (0.023)  0.84    0.033 (0.019)  0.10  Age, body mass index  0.006 (0.023)  0.79    0.038 (0.017)  0.03  Age, systolic blood pressure  0.002 (0.023)  0.92    0.035 (0.018)  0.05  Age, pulse pressure  0.003 (0.023)  0.90    0.037 (0.018)  0.05  Covariate  No bilateral oophorectomy    Bilateral oophorectomy  β* (standard error)  p value    β* (standard error)  p value  None (no. of years since surgical menopause only)  0.026 (0.021)  0.23    0.045 (0.018)  0.01  Age  0.005 (0.023)  0.82    0.042 (0.018)  0.02  Age, current/former smoking  –0.001 (0.022)  0.97    0.042 (0.019)  0.03  Age, current/former HRT† use  0.001 (0.022)  0.95    0.042 (0.020)  0.04  Age, current alcohol intake   –0.009 (0.022)  0.69    0.039 (0.019)  0.04  Age, use of blood pressure/cholesterol medications  0.012 (0.024)  0.63    0.036 (0.019)  0.06  Age, total cholesterol  0.009 (0.022)  0.67    0.042 (0.018)  0.03  Age, low density lipoprotein cholesterol  –0.005 (0.024)  0.85    0.031 (0.020)  0.13  Age, high density lipoprotein cholesterol  0.009 (0.021)  0.66    0.030 (0.018)  0.11  Age, ratio of sagittal to transverse abdominal diameters  0.005 (0.023)  0.84    0.033 (0.019)  0.10  Age, body mass index  0.006 (0.023)  0.79    0.038 (0.017)  0.03  Age, systolic blood pressure  0.002 (0.023)  0.92    0.035 (0.018)  0.05  Age, pulse pressure  0.003 (0.023)  0.90    0.037 (0.018)  0.05  * Unstandardized regression coefficient for an interval of 10 years since hysterectomy; for example, a coefficient of 0.026 (0.021) indicates that intima-media thickness increased on average 0.026 mm per 10 years of time since hysterectomy. † HRT, hormone replacement therapy. View Large References 1. Eaker ED, Chesebro JH, Sacks FM, et al. AHA Medical/Scientific Statement: cardiovascular disease in women. Circulation  1993; 88(4 pt 1): 1999–2009. Google Scholar 2. Colditz GA, Willett WC, Stampfer MJ, et al. Menopause and the risk of coronary heart disease in women. N Engl J Med  1987; 316: 1105–10. Google Scholar 3. Kannel WB, Hjortland MC, McNamara PM, et al. Menopause and risk of cardiovascular disease: The Framingham Study. Ann Intern Med  1976; 85: 447–52. Google Scholar 4. Oliver M, Boyd G. Effect of bilateral ovariectomy on coronary-artery disease and serum-lipid levels. Lancet  1959; 2: 690–4. Google Scholar 5. Wuest J, Dry T, Edwards J. The degree of coronary atherosclerosis in bilaterally oophorectomized women. Am Heart J  1953; 7: 801–9. Google Scholar 6. Parrish H, Carr C, Hall D, et al. Time interval from castration in premenopausal women to development of excessive coronary atherosclerosis. Am J Obstet Gynecol  1967; 99: 155–62. Google Scholar 7. Gordon T, Kannel WB, Hjortland MC, et al. Menopause and coronary heart disease. The Framingham Study. Ann Intern Med  1978; 89: 157–61. Google Scholar 8. Punnonen R, Ikalainen M, Seppala E. Premenopausal hysterectomy and risk of cardiovascular disease. (Letter). Lancet  1987; 1: 1139. Google Scholar 9. Hulley S, Grady D, Bush T, et al. Randomized trial of estrogen plus progestin for secondary prevention of coronary heart disease in postmenopausal women. Heart and Estrogen/progestin Replacement Study (HERS) Research Group. JAMA  1998; 280: 605–13. Google Scholar 10. Herrington DM, Reboussin DM, Brosnihan KB, et al. Effects of estrogen replacement on the progression of coronary-artery atherosclerosis. N Engl J Med  2000; 343: 522–9. Google Scholar 11. Sun P, Dwyer KM, Merz CN, et al. Blood pressure, LDL cholesterol, and intima-media thickness: a test of the “response to injury” hypothesis of atherosclerosis. Arterioscler Thromb Vasc Biol  2000; 20: 2005–10. Google Scholar 12. Dwyer JH, Sun P, Kwong-Fu H, et al. Automated intima-media thickness: The Los Angeles Atherosclerosis Study. Ultrasound Med Biol  1998; 24: 981–7. Google Scholar 13. Chambless LE, Heiss G, Folsom AR, et al. Association of coronary heart disease incidence with carotid arterial wall thickness and major risk factors: the Atherosclerosis Risk in Communities (ARIC) Study, 1987–1993. Am J Epidemiol  1997; 146: 483–94. Google Scholar 14. O’Leary DH, Polak JF, Kronmal RA, et al. Carotid-artery intima and media thickness as a risk factor for myocardial infarction and stroke in older adults. Cardiovascular Health Study Collaborative Research Group. N Engl J Med  1999; 340: 14–22. Google Scholar 15. Bots ML, Hoes AW, Hofman A, et al. Cross-sectionally assessed carotid intima-media thickness relates to long-term risk of stroke, coronary heart disease and death as estimated by available risk functions. J Intern Med  1999; 245: 269–76. Google Scholar 16. Witteman JC, Grobbee DE, Kok FJ, et al. Increased risk of atherosclerosis in women after the menopause. BMJ  1989; 298: 642–4. Google Scholar 17. Nabulsi A, Folsom A, Szklo M, et al. No association of menopause and hormone replacement therapy with carotid artery intima-media thickness. Circulation  1996; 94: 1857–63. Google Scholar 18. Zeigler-Johnson CM, Holmes JL, Lassila HC, et al. Subclinical atherosclerosis in relation to hysterectomy status in black women. Stroke  1998; 29: 759–64. Google Scholar 19. Chalmers C. Does hysterectomy in a premenopausal woman affect ovarian function? Med Hypotheses  1996; 46: 573–5. Google Scholar 20. Stampfer MJ, Willett WC, Colditz GA, et al. A prospective study of postmenopausal estrogen therapy and coronary heart disease. N Engl J Med  1985; 313: 1044–9. Google Scholar 21. Bush TL, Barrett-Connor E, Cowan LD, et al. Cardiovascular mortality and noncontraceptive use of estrogen in women: results from the Lipid Research Clinics Program Follow-up Study. Circulation  1987; 75: 1102–9. Google Scholar 22. Henderson BE, Paganini-Hill A, Ross RK. Decreased mortality in users of estrogen replacement therapy. Arch Intern Med  1991; 151: 75–8. Google Scholar 23. Stampfer MJ, Colditz GA. Estrogen replacement therapy and coronary heart disease: a quantitative assessment of the epidemiologic evidence. Prev Med  1991; 20: 47–63. Google Scholar 24. Grodstein F, Manson JE, Stampfer MJ. Postmenopausal hormone use and secondary prevention of coronary events in the Nurses’ Health Study. A prospective, observational study. Ann Intern Med  2001; 135: 1–8. Google Scholar 25. Utian WH, Katz M, Davey DA, et al. Effect of premenopausal castration and incremental dosages of conjugated equine estrogens on plasma follicle-stimulating hormone, luteinizing hormone, and estradiol. Am J Obstet Gynecol  1978; 132: 297–302. Google Scholar 26. Liu Y, Ding J, Bush TL, et al. Relative androgen excess and increased cardiovascular risk after menopause: a hypothesized relation. Am J Epidemiol  2001; 154: 489–94. Google Scholar 27. Grodstein F, Stampfer MJ. Estrogen for women at varying risk of coronary disease. Maturitas  1998; 30: 19–26. Google Scholar

Journal

American Journal of EpidemiologyOxford University Press

Published: Sep 1, 2002

There are no references for this article.