Calcium/vitamin D supplementation, serum 25-hydroxyvitamin D concentrations, and cholesterol profiles in the Women's Health Initiative calcium/vitamin D randomized trial

Abstract

Objective: The objective of this study was to evaluate whether increased serum 25-hydroxyvitamin D3 (25OHD3) concentrations, in response to calcium/vitamin D (CaD) supplementation, are associated with improved lipids in postmenopausal women.

Methods: The parent trial was a double-blind, randomized, placebo-controlled, parallel-group trial designed to test the effects of CaD supplementation (1,000 mg of elemental calcium + 400 IU of vitamin D3 daily) versus placebo in postmenopausal women. Women from the general community, including multiple sites in the United States, were enrolled between 1993 and 1998. This cohort included 300 white, 200 African-American, and 100 Hispanic participants who were randomly selected from the Women's Health Initiative CaD trial. Serum 25OHD3 and lipid (fasting plasma triglycerides [TG], high-density lipoprotein cholesterol [HDL-C], and calculated low-density lipoprotein cholesterol [LDL-C]) levels were assessed before and after CaD randomization.

Results: There was a 38% increase in mean serum 25OHD3 concentrations after 2 years (95% CI, 1.29-1.47, P < 0.001) for women randomized to CaD (24.3 ng/mL postrandomization mean) compared with placebo (18.2 ng/mL). Women randomized to CaD had a 4.46-mg/dL mean decrease in LDL-C (P = 0.03). Higher concentrations of 25OHD3 were associated with higher HDL-C levels (P = 0.003), along with lower LDL-C and TG levels (P = 0.02 and P < 0.001, respectively).

Conclusions: Supplemental CaD significantly increases 25OHD3 concentrations and decreases LDL-C. Women with higher 25OHD3 concentrations have more favorable lipid profiles, including increased HDL-C, lower LDL-C, and lower TG. These results support the hypothesis that higher concentrations of 25OHD3, in response to CaD supplementation, are associated with improved LDL-C.

Figure 1

This figure outlines the eligibility and selection process, along with the inclusion and exclusion determination, for the women in the CaD trial.

Figure 2

This figure illustrates the availability of core biomarkers and 25OHD₃ for the analytic sample (n=576) by visit year. 25OHD₃ was not measured at baseline and Y6, so are 100% missing by design. Baseline and Y6 values of 25OHD₃ were imputed by carrying the post (pre)-randomization measures forward (backwards) and by multiple imputation. The mediation analysis included 25OHD₃ as a time-dependent variable and used both measured and imputed values of 25OHD₃.

Figure 3

(A) Total effect of CaD on LDL-C; path c. (B) CaD is hypothesized to lower LDL-C through 25OHD₃. It is believed that randomization to CaD will increase circulating levels of 25OHD₃; path a. Consequently, higher levels of 25OHD₃, the mediating variable, will be associated with lower levels of LDL-C; path b. CaD will also exert a direct effect on LDL-C after adjustment for paths a and b; path c′.

Figure 4

Multivariable adjusted Generalized Additive Mixed Model (GAMM) estimates of the mean LDL-C, HDL-C, and Triglycerides (95% confidence interval in shaded region) as a smoothed function of 25OHD₃. Lipids and 25OHD₃ measurements at both Year 1 and Year 3 were included in the GAMMs to incorporate both cross-sectional and longitudinal information. GAMMs included an indicator variable for pre/post-CaD randomization visit and were adjusted for age, race/ethnicity, body mass index, smoking status, history of high blood cholesterol, diabetes mellitus, prior HT use, physical activity, total calcium intake, HT randomization arm, DM randomization arm, CaD randomization arm, and CaD randomization × visit interaction. To account for within participant correlation a random intercept was also included. The smoothness of each spline fit was chosen objectively by generalized cross-validation.