Bone mineral density and cardiovascular diseases: a two-sample Mendelian randomization study

Abstract

The link between BMD and cardiovascular disease (CVD) remains a topic of extensive debate in observational studies, with inconsistent reports regarding the causality of this relationship. This study implements robust methodologies to evaluate the causal relationship between BMD and various CVDs. Two sample Mendelian randomization (MR) method was used to estimate the relationship between genetically predicted BMD and seven key CVDs: atrial fibrillation and flutter, angina, ischemic heart disease, heart failure, hypertension, myocardial infarction, and non-ischemic cardiomyopathy. Data were obtained from independent publicly available genome-wide association studies (GWAS) for BMD and CVDs, using two separate datasets for the cardiovascular outcomes: the UK Biobank cohort (primary analysis) and the FinnGen cohort (validation analysis). The MR Pleiotropy RESidual Sum and Outlier test assessed the heterogeneity and pleiotropy of selected instrumental variables (IVs). We applied the inverse variance weighted model (IVW), weighted median, weighted mode method, and MR-Egger regression model to estimate causal effects. MR results indicate no relationship between BMD and atrial fibrillation and flutter (IVW, beta-estimate: 0.011, SE: 0.03, p = .73), angina (IVW, beta-estimate: 0.04, SE: 0.03, p = .17), chronic ischemic heart disease (IVW, beta-estimate: 0.009, SE: 0.03, p = .74), heart failure (IVW, beta-estimate: 0.004, SE: 0.04, p = .91), hypertension (IVW, beta-estimate: -0.01, SE: 0.01, p = .44), myocardial infarction (IVW, beta-estimate: 0.02, SE: 0.03, p = .36), or non-ischemic cardiomyopathy (IVW, beta-estimate: 0.1, SE: 0.08, p = .20). These findings remained consistent across all complementary analyses (MR-Egger, weighted median and weighted mode) and were validated using the FinnGen cohort GWAS dataset. This comprehensive analysis identified no evidence for a causal link between genetically predicted BMD and a range of key CVDs. Previously reported observational associations between bone and cardiovascular health likely represent shared risk factors rather than direct causal mechanisms.

Keywords: BMD; Mendelian randomization; cardiovascular diseases; causality; genome-wide association study; osteoporosis.

Graphical Abstract

Figure 1

Overall study design. Schematic representation of the MR analysis performed SNPs: single nucleotide polymorphisms. Abbreviations: GWAS: genome-wide association study; IVW: inverse variance weighted; LD: linkage disequilibrium; MR: Mendelian randomization; SE: standard error.

Figure 2

Impact of SNPs on BMD and cardiac diseases. The scatter plots from the Mendelian randomization (MR) analysis illustrate the statistical relationship between genetically predicted BMD and various cardiac diseases. The x-axis represents the odds ratio and 95% confidence intervals for the genetic association between BMD and outcomes, derived from both the UK Biobank (UKBB) and FinnGen cohorts. The y-axis lists the cardiac outcomes analyzed, including atrial fibrillation and flutter, chronic ischemic heart disease, heart failure, hypertension, myocardial infarction, and non-ischemic cardiomyopathy. The MR methods used include inverse variance weighted, MR-Egger, simple mode, weighted median, and weighted mode. Results are presented separately for each cohort to validate findings across independent datasets.