Low-Density Lipoprotein Cholesterol Attributable Cardiovascular Disease Risk Is Sex Specific

Abstract

Background Epidemiological studies show that women are generally at lower risk for cardiovascular disease than men. Here, we investigated the sex-specific differential effect of genetically increased low-density lipoprotein cholesterol (LDL-C) on cardiovascular disease (CVD) and other lipid-associated diseases. Methods and Results This is a 2-sample Mendelian randomization study that uses individual participant data from 425 043 participants from the UK Biobank, including 229 279 female participants. An 80-variant LDL-C weighted genetic score was generated. Linear and logistic regression models with interactions were used to identify differences between sex-specific LDL-C effects on lipids, carotid-intima media thickness, and multiple cardiovascular outcomes such as CVD, ischemic heart disease, peripheral artery disease, heart failure, aortic valve disease, type 2 diabetes, atrial fibrillation, and aortic aneurysm and dissection. After correction for multiple testing, we observed that the genetically increased LDL-C effect on CVD events was sex specific: per SD genetically increased LDL-C, female participants had a higher LDL-C increase but an attenuated CVD risk increase compared with male participants (LDL-C: female participants 0.71 mmol/L, 95% CI, 0.70-0.72 and male participants 0.57 mmol/L, 95% CI, 0.56-0.59. P for interaction: 5.03×10^-60; CVD: female participants: odds ratio [OR], 1.32; 95% CI 1.24-1.40 and male participants: OR, 1.52; 95% CI, 1.46-1.58. P for interaction: 9.88×10^-5). We also observed attenuated risks for ischemic heart disease and (nominally for) heart failure in female participants, and genetically increased LDL-C results in higher risk for aortic valve disease in female participants compared with male participants. Genetically increased LDL-C was also associated with an attenuated carotid-intima media thickness increase in female participants. We did not observe other significant attenuations. Sensitivity analyses with an unweighted genetic score and sex-specific weighted genetic scores showed similar results. Conclusions We found that genetically increased LDL-C has a sex-specific differential effect on the risk for cardiovascular disease, ischemic heart disease, heart failure, and aortic valve stenosis. Our observations provide evidence that LDL-C might be a less important determinant of CVD in women compared with men, suggesting that male patients might benefit more from LDL-C targeted therapies for CVD management than female patients and warranting investigations into the sex-specific relative contribution of risk factors for CVD.

Keywords: cardiovascular disease; genetics; risk factor; sex‐differences.

Figure 1. Genetic instrument distribution and effect on lipids and CVD, stratified for sex.

A, difference in lipids due to 1 SD genetically increased LDL‐C in female participants and in male participants, with 95% CIs. Values on the top represent P value for interaction. B, Boxplot summarizing the distribution of the genetic instrument in the cohort (in median and interquartile range), stratified for female and male participants. ApoA indicates apolipoprotein A; ApoB, apolipoprotein B; CVD, cardiovascular disease; HDL‐C, high‐density lipoprotein cholesterol; LDL‐C, low‐density lipoprotein cholesterol; and Lp(a), lipoprotein(a).

Figure 2. The effects of an SD genetically increased LDL‐C on clinical outcome, stratified for sex.

Forest plot with odds ratios per 1 SD genetically increased LDL‐C for CVD, its subcomponents and other cardiovascular outcome, stratified for female participants and male participants. Values on the right side of the panel represent P values for interaction. LDL‐C indicates low‐density lipoprotein cholesterol; and OR, odds ratio.

Figure 3. Effect of 1 SD genetically increased LDL‐C on mean IMT.

Bar plot comparing the difference in mean carotid intima media thickness (cIMT) between male and female participants per 1 SD genetically increased LDL‐C, Including 95% CI. Values on the top represent P‐value for interaction. LDL‐C indicates low‐density lipoprotein cholesterol.

Figure 4. Visualization of directional concordance of genetic data with clinical trial data.

Forest plot illustrating the concordance of the genetic data and statin trial data on the significant difference between male and female participants on the risk ratio (genetic) and rate ratio (statin trials) for cardiovascular events, scaled to 1 mmol/L LDL‐C reduction, as defined by the CTT. The genetic data is based on dichotomization of the genetic risk instrument, effectively creating 2 groups within the separate male and female groups, similar to 2 groups in a clinical trial. CTT indicates Cholesterol Treatment Trialists’ Collaboration; LDL‐C, low‐density lipoprotein cholesterol; OR, odds ratio; and RR, rate ratio.