Association Between Statin Use and Sex Hormone in the Multi-Ethnic Study of Atherosclerosis Cohort

Abstract

Purpose

Based on the 2018 American College of Cardiology/American Heart Association cholesterol guidelines, the number of individuals eligible for statin therapy to reduce atherosclerotic cardiovascular disease risk has greatly expanded. Statins inhibit cholesterol biosynthesis, which can impair gonadal steroidogenesis. We evaluated the effect of statins on endogenous sex hormones in a large epidemiological study.

Methods

A total of 6814 Multi-Ethnic Study of Atherosclerosis (MESA) participants underwent the baseline examination. Of these, 6171 had measurements of serum sex hormones available: dehydroepiandrosterone (DHEA), SHBG, estradiol, and total and bioavailable testosterone. Multivariable linear regression models were used to assess the relationship of statin use with each sex hormone.

Results

A total of 345 women (17.4%) and 464 men (14.7%) were statin users (mean age, 67 years; 41% white, 29% black, 11% Chinese, and 19% Hispanic). Among the users vs nonusers of statins, the mean SHBG was 3.54 nmol/L (P < 0.01) lower in women and 3.37 nmol/L (P < 0.001) lower in men; the mean DHEA was 1.06 nmol/L (P < 0.05) lower in women and 0.70 nmol/L (P < 0.01) lower in men, after adjustment for potential confounders. With further propensity score adjustment, the mean DHEA and SHBG levels were 0.67 nmol/L (P < 0.05) and 3.49 nmol/L (P < 0.001) lower, respectively, for statin users vs nonusers. No statistically significant association was noted between estradiol, total testosterone, and bioavailable testosterone and statin use.

Conclusion

Statin users have lower levels of SHBG and DHEA. This is especially relevant owing to the increasing use of statin therapy.

The 2018 American College of Cardiology (ACC)/American Heart Association (AHA) cholesterol guidelines recommend statin therapy to reduce risk of atherosclerotic cardiovascular disease (1). Using these guidelines, many older subjects have qualified for the initiation and continuation of statin therapy (2–4). A recent Multi-Ethnic Study of Atherosclerosis (MESA) study showed that the number of participants eligible for statin therapy doubled using the 2013 ACC/AHA cholesterol guidelines compared with the 2001 National Cholesterol Education Program/Adult Treatment Program III guidelines (5). Once initiated, statins are generally indicated throughout an individual’s life until we have data to indicate a more appropriate length of treatment. Cholesterol is a required intermediate in sex steroid synthesis. With the increased use of statins, especially moderate- and high-intensity statin therapy, concern has been raised regarding impaired gonadal steroidogenesis. This might be related to inhibition of the 3-hydroxy-3-methylglutaryl coenzyme A reductase enzymes in the steroid pathway or due to reduced circulating cholesterol.

The effect of statins on endogenous sex hormones has been studied in some small clinical trials but with inconsistent results (6–11). Similarly, observational studies have provided heterogeneous results on sex hormone levels with the use of statin therapy (12–14). The characteristics of both experimental and observational studies of statins and sex hormones are summarized in Table 1. These studies were limited by small sample sizes and did not thoroughly accounted for potential confounders.

Therefore, to better characterize the association of statin therapy with serum sex hormone levels, we undertook the present cross-sectional study in the population-based MESA cohort.

Methods

Study population

The MESA design has been described previously (15). In brief, MESA was a population-based study of 6814 white, black, Hispanic, and Chinese participants, aged 45 to 84 years without clinical cardiovascular disease, who had been recruited from 2000 to 2002. The participants lived in six US communities: Baltimore, MD; Chicago, IL; Forsyth County, NC; Los Angeles County, CA; northern Manhattan, NY; and St. Paul, MN. Each participant had provided informed consent, and the institutional review board of the individual institutions had approved the study.

Study variables

Serum sex hormones

A total of 6814 MESA participants had undergone the baseline examination. Of these, 6171 had measurements of the serum sex hormones available: dehydroepiandrosterone (DHEA), SHBG, estradiol (E2), and total and bioavailable testosterone. Excluded from the analyses were 1021 women receiving hormone replacement therapy, for a resulting analysis data set of 1987 women and 3163 men.

Fasting blood samples taken between 7:30 am and 10:30 am were used for the assays. Serum samples were extracted by centrifugation at 2000g for 15 minutes or 3000g for 10 minutes and stored at −70°C. The serum hormone concentrations were measured at the University of Massachusetts Medical Center (Worcester, MA).

SHBG was measured by chemiluminescent enzyme immunometric assay using Immulite kits (Diagnostic Products Corp., Los Angeles, CA). The RIA kit was used to measure DHEA and total testosterone directly. E2 was measured using an ultrasensitive RIA kit (Beckman Coulter, Webster, TX). Bioavailable testosterone was calculated using the method reported by Vermeulen et al. (16) The intra-assay coefficients of variation for SHBG, DHEA, total testosterone, and E2 were 9.0%, 11.2%, 12.3%, and 10.5%, respectively.

Statin therapy

Statin medication type, frequency, and dosages during the 2-week period before the baseline visit were determined by participant self-report and a review of the pharmacy containers the participants had brought to the examinations. Statin use was defined as having brought in a container indicating the use of a statin drug.

Covariates

The following self-reported covariates were evaluated: age, sex, and race/ethnicity (i.e., white, black, Hispanic, Chinese). The body mass index (BMI) was calculated as the weight in kilograms divided by the square of the height in meters. Total cholesterol, high-density lipoprotein cholesterol, and triglyceride measurements were performed at the Collaborative Studies Clinical Laboratory at Fairview University Medical Center (Minneapolis, MN) from blood samples obtained after the participants had fasted for 12 hours. Diabetes was defined as fasting blood glucose ≥7.0 mmol/L (126 mg/dL), self-reported diabetes, or the use of hypoglycemic drugs. Hypertension was defined as untreated diastolic blood pressure ≥90 mm Hg, systolic blood pressure ≥140 mm Hg, or the use of antihypertensive medication. Smoking was defined as the current use of cigarettes.

Statistical analysis

We have reported the mean for the continuous variables and counts with percentages for the categorical variables. Sex-specific multivariable linear regression models were used to assess the association of statin use with the serum concentrations of each sex hormone. Although the recent ACC/AHA cholesterol guidelines have recommended the use of statins for patients with diabetes, this was not a guideline indication for statin use in 2000 to 2002 when the MESA participants were recruited (17). However, the overall metabolic picture associated with diabetes might be the sort of background in which clinicians could have been motivated to prescribe statins on their own. Therefore, to reduce confounding by indication, regression analyses were performed with inclusion of an estimate of the propensity for statin use (18). Three regression models were performed. Model 1 was adjusted for age and race. Model 2 was also adjusted for high-density lipoprotein cholesterol, total cholesterol, BMI, and diabetes. Because no significant interaction use was found between sex and statin use, for both DHEA and SHBG levels, the women and men were pooled into a single sample for the regression analyses adjusting for the propensity score. The models for total testosterone and bioavailable testosterone were for men only and the model for estradiol was for women only, adjusting for the propensity score. Additional regression models were performed stratified by statin type to explore the association of SHBG with glucose and insulin levels in those without diabetes (19). The natural log insulin and glucose levels were used, because the data were skewed. Patients with diabetes were excluded from this analysis, because the treatment of diabetes has major effects on both glucose and insulin levels. P values < 0.05 denoted statistical significance. All analyses were performed using Stata, version 15 (StataCorp, College Station, TX).

Results

The mean age of the participants receiving statin therapy was 67 years; 57% were men, 26% had diabetes, 41% were white, 29% were black, 11% were Chinese, and 19% were Hispanic (Table 2). The total testosterone levels in men and the SHBG and DHEA levels in both men and postmenopausal women were inversely associated with statin use, even after multivariate adjustment (P < 0.05; Table 3). For the statin users compared with the nonusers of statins, the mean SHBG was 3.54 nmol/L (P < 0.05) lower in women and 3.37 nmol/L (P < 0.05) lower in men, or approximately one fifth of 1 SD less. Similarly, the mean DHEA was 1.06 nmol/L (P < 0.05) lower in the women and 0.70 nmol/L (P < 0.05) lower in the men, or approximately one tenth of 1 SD less, among the statin users compared with the nonusers of statins. In addition, total testosterone was 0.81 nmol/L (P < 0.05) lower in the male statin users compared with the nonusers. No significant association was noted between statin use and E2 or bioavailable testosterone among either women or men. Although the propensity score analysis to some degree attenuated the association of DHEA with statin use, the inverse association of SHBG levels with statin use remained strong. The DHEA level was 0.67 nmol/L (P < 0.05) lower and the mean SHBG level was 3.47 nmol/L (P < 0.05) lower, no significant association was found for total testosterone in men among statin users compared with nonusers (Table 4).

Among the statin users, 7.5% were taking atorvastatin, 4.8% were taking simvastatin, and 2% were taking pravastatin (19). After adjustment, no statistically significant evidence was found that a specific statin drug was consistently related to any of the sex hormones (19). Furthermore, a strong negative association was found between the glucose and SHBG levels in men and women and insulin levels in women (19).

Discussion

The key finding of our cross-sectional epidemiological study was the lower levels of serum sex hormone concentrations in statin users compared with nonusers of statins. A significant inverse association of statin use was found with SHBG levels among statin users compared with nonusers of statin. In both men and women, this inverse association remained statistically significant even after controlling for BMI, diabetes, and other covariates. To eliminate confounding by indication, we also conducted a propensity score analysis but did not find meaningful changes to the SHBG results.

In a clinical trial of men with hyperlipidemia who had been randomized to receive simvastatin 80 mg or placebo, Dobs et al. (7) concluded that statin therapy had no effect on SHBG levels. A reduction of −0.08% of SHBG was found in the simvastatin group at 12 weeks compared with baseline but the difference did not meet statistical significance. This likely resulted from the small sample size (n = 37), which provided little power to detect a difference; thus, these results were inconclusive (7). Our findings concur with the findings from a randomized placebo-controlled trial of men with hypercholesterolemia in which the investigators noted a small, but significant, reduction in SHBG for the simvastatin 20-mg group compared with placebo after 24 weeks of treatment (8). Our findings are also consistent with the results from the BACH (Boston Area Community Health) survey, an epidemiological study, which showed 11% lower SHBG levels among statin users compared with nonusers of statins (12).

Moreover, endogenous sex hormones might differentially modulate glycemic status and the risk of type 2 diabetes in men and women. High testosterone levels have been associated with a greater risk of type 2 diabetes in women but a lower risk in men. The risk of type 2 diabetes is lower with greater levels of SHBG, with this inverse association stronger in women than in men (20). Our data showing a strong negative association of glucose with SHBG in men and women is consistent with this finding. Genetic studies have strengthened the evidence that SHBG and sex hormones are involved in the etiology of type 2 diabetes (21, 22). This is of particular interest given the data from both observational studies and randomized statin trials that incident diabetes increased with statin use (23–25).

Statin use was also associated with lower levels of DHEA, in both men and women, in our study. DHEA is one of the major hormones secreted by the adrenal gland, and it declines linearly with age (26). Our finding is consistent with a previous study by Dobs et al. (6), which showed lower levels of DHEA with pravastatin use. A study by Hall et al. (12) demonstrated lower levels of DHEA with statin use in older men but not in younger individuals. A meta-analysis by Yang et al. (27) also indicated that statins significantly reduced the levels of DHEA compared with placebo in patients with polycystic ovary syndrome. A study evaluating testicular function in men with hypercholesterolemia after simvastatin therapy did not reveal any variations in the DHEA levels (28). DHEA has been associated with an increased risk of impaired fasting glucose in postmenopausal women (29), although no such association has been noted in men.

Statin use was associated with lower total testosterone levels in men. However, after propensity score analysis, the association was not statistically significant. Changes in bioavailable testosterone correlated highly with changes in total testosterone (30). Cholesterol is the main substrate for androgen synthesis of testosterone and dihydrotestosterone and a reduction in cholesterol might ultimately affect these levels (12). However, a number of studies have evaluated the association of statin use and testosterone levels and yielded null results. Our findings concur with observational studies, including the BACH survey and the NHANES III (National Health and Nutrition Examination Survey), which found no cross-sectional association between statin use and total testosterone levels in men (12, 13). Similarly, clinical trials have shown that statin use had no significant effect on total testosterone levels (8, 9, 11).

We observed no association between statin use and E2 levels, consistent with the results from previous studies. The Women’s Ischemia Syndrome Evaluation study enrolled 453 women in the pre, peri-, and postmenopausal stages and assessed the reproductive hormone levels after the initiation of statins (14). They found no significant differences in the E2 levels with statin use (14). Another study of peri- and postmenopausal women revealed no changes in E2 levels after 5 years of pravastatin treatment compared with nonusers of statin (31).

The strengths of the present study included the large population-based, ethnically and racially diverse cohort without clinical cardiovascular disease at baseline and the use of a central research laboratory for the hormone measurements. In addition, we were able to reduce confounding by indication by including in our regression models an estimate of the probability (propensity) that a participant would have been prescribed a statin. Another limitation of our study was that the sex hormones were only measured at a single time point, which might not reflect usual levels, given the significant biological variability in serum hormone concentrations over time. However, this type of variability has typically tended to weaken observed associations. The sex hormone levels could also vary by statin type or dose, which was not adjusted for. In our study, the number of participants using specific statins was inadequate to investigate these effects. In addition, the present study was a cross-sectional study with the usual limitations of all epidemiological studies and the inability to determine the time effects of statin use (e.g., when the participant had started taking the statin was not known).

Conclusion

In the present cross-sectional study, the SHBG and DHEA levels in both men and women were inversely associated with statin use. These changes give insight into other pathways affected by statins and might necessitate monitoring of sex hormone levels during statin therapy. Further prospective studies or clinical trials will be relevant to elucidate this.

Acknowledgments

The authors thank the other investigators, staff, and participants of the MESA study for their valuable contributions. A full list of the participating MESA investigators and institutions can be found at http://www.mesa-nhlbi.org.

Financial Support: Measurement of the hormone concentrations was supported by the National Institutes of Health (Grants R01-HL-074406 and R01-HL-074338), and MESA was supported by the National Heart, Lung, and Blood Institute (Contracts HHSN268201500003I, N01-HC-95159, N01-HC-95160, N01-HC-95161, N01-HC-95162, N01-HC-95163, N01-HC-95164, N01-HC-95165, N01-HC-95166, N01-HC-95167, N01-HC-95168, and N01-HC-95169).

Additional Information

Disclosure Summary: The authors have nothing to disclose.

Data Availability:

The data sets generated during and/or analyzed in the present study are not publicly available but are available from the corresponding author on reasonable request.

Abbreviations:

Abbreviations:
 
• ACC

  American College of Cardiology

 
• AHA

  American Heart Association

 
• BMI

  body mass index

 
• DHEA

  dehydroepiandrosterone

 
• E2

  estradiol

 
• MESA

  Multi-Ethnic Study of Atherosclerosis

References and Notes