Genetic liability to age at first sex and birth in relation to cardiovascular diseases: a Mendelian randomization study

Abstract

Background: Growing evidence suggests that various reproductive factors, including early menarche, early menopause, and age at first birth, may increase the risk of developing cardiovascular disease (CVD) later in life. However, the associations between reproductive factors and CVDs are inconsistent and controversial. Therefore, we conducted a two-sample Mendelian randomization (MR) analysis to explore the potential links between age at first sex (AFS) and age at first birth (AFB) and several CVDs.

Methods: We obtained summary statistics for exposure from the largest genome-wide association studies of AFS and AFB. To serve as instrumental variables, we selected 259 SNPs associated with AFS and 81 SNPs associated with AFB at the genome-wide significance level. We employed a random-effects inverse-variance weighted method to pool estimates, and conducted multivariable MR analysis to determine the direct association between AFS and AFB with CVDs, while accounting for the effects of confounders.

Results: The genetic liability to later AFS was associated with decreased risks of heart failure (odd ratio [OR] 0.700; 95% confidence interval [CI] 0.639-0.767; p = 2.23 × 10^-14), coronary artery disease (OR 0.728; 95% CI 0.657-0.808; p = 1.82 × 10^-9), myocardial infarction (OR 0.731; 95% CI 0.657-0.813; p = 8.33 × 10^-9), stroke (OR 0.747; 95% CI 0.684-0.816; p = 6.89 × 10^-11), and atrial fibrillation (OR 0.871; 95% CI 0.806-0.941; p = 4.48 × 10^-4). The genetic liability to later AFB was also associated with decreased risks of CVDs, including myocardial infarction (OR 0.895; 95% CI 0.852-0.940; p = 8.66 × 10^-6), coronary heart disease (OR 0.901; 95% CI 0.860-0.943; p = 9.02 × 10^-6), heart failure (OR 0.925; 95% CI 0.891-0.961; p = 5.32 × 10^-5), and atrial fibrillation (OR 0.944; 95% CI 0.911-0.978; p = 0.001). However, no association was found between AFB and stroke. The associations remained independent from the effects of AFS and AFB on potential confounders, including smoking, alcohol intake, body mass index, and depression. Mediation analysis suggested that education attainment partly mediates the link from AFS and AFB to CVD outcomes.

Conclusion: Our results observed a causal relationship between later AFS, AFB and lower CVDs risk; it emphasizes the importance of providing sex education since early sex and birth may have undesirable effects. Cardiovascular risk stratification that considers reproductive factors may help address CVD risk.

Keywords: Age at first birth; Age at first sex; Cardiovascular diseases; Mendelian randomization.

Fig. 1

Mendelian randomization association of genetic liability to age at first sex with cardiovascular diseases. OR odds ratio; CI confidence interval. Estimates are from the random-effects inverse variance weighted method. Significant at the Bonferroni-corrected threshold of p-value < 0.01

Fig. 2

Multivariable Mendelian randomization was performed to estimate the direct effect of the association of age at first sex with cardiovascular disease after accounting for the effects of age at first sex on other confounders

Fig. 3

Mendelian randomization association of genetic liability to age at first birth with cardiovascular diseases. OR odds ratio; CI confidence interval. Estimates are from the random-effects inverse variance weighted method. Significant at the Bonferroni-corrected threshold of p-value < 0.01

Fig. 4

Multivariable Mendelian randomization was performed to estimate the direct effect of the association of age at first birth with cardiovascular disease after accounting for the effects of age at first birth on other confounders