Effect of monthly vitamin D supplementation on cardiac biomarkers: A post-hoc analysis of a randomized controlled trial.

Abstract

The effects of vitamin D supplementation on cardiovascular diseases are controversial. Data on effects of vitamin D upon cardiac biomarkers, as surrogate endpoints of cardiovascular diseases, are limited and inconclusive. Therefore, we carried out a post-hoc analysis of sub-samples of a randomized, double-blinded, placebo-controlled trial with community-based older adults who were randomized to receive monthly 100,000-IU vitamin D or placebo, to determine effect of monthly vitamin D supplementation on high-sensitivity cardiac troponin I (hs-cTnI), troponin T (hs-cTnT) and N-terminal-pro-B-type natriuretic peptide (NT-proBNP). Adjusted relative difference (aRD) of follow-up geometric mean of biomarkers and adjusted relative risk (aRR) of elevated biomarkers between two groups were calculated. A total of 779 participants aged 50-84 y, randomized to vitamin D (n = 395) or placebo (n = 384) groups underwent sampling for measurement of plasma biomarkers at baseline and after one or two years treatment. Over a mean follow-up of 1.6 years, we did not find significant relative difference of geometric mean levels of three biomarkers at follow-up between vitamin D and placebo groups: hs-cTnI (aRD=1.03, 95%CI=0.97-1.09), hs-cTnT (aRD=0.99, 95%CI=0.95-1.04), and NT-proBNP (aRD=1.01, 95%CI=0.92-1.10). No significant differences were found in likelihood of clinically elevated biomarkers between two groups: hs-cTnI (aRR=0.92, 95%CI=0.51-1.69), hs-cTnT (aRR=1.11, 95%CI=0.86-1.42), and NT-proBNP (aRR=1.03,95%CI=0.89-1.20). However, among participants with initial low vitamin D status (<50 nmol/L, n = 200), follow-up NT-proBNP were significantly lower in the vitamin D group compared to placebo (geometric mean 75.9 vs 94.5 pg/mL, respectively; aRD=0.84, 95%CI=0.71-<1.00). The same results were observed for the NT-proBNP levels change from baseline between two groups. Overall, in older adults, monthly vitamin D supplementation did not reduce concentrations of hs-cTnI, hs-cTnT, and NT-proBNP. In those with low vitamin D status, vitamin D treatment was associated, on follow up and change from baseline, with lower plasma NT-proBNP compared with placebo. This potentially signals reduced risk of subsequent heart failure within this sub-group. However, we acknowledge that these findings need to be considered exploratory. Further research is required to replicate them.