Abstract
Background The predictive value of lipoprotein(a) (Lp[a]), high-sensitivity C-reactive protein (hsCRP), and remnant cholesterol (RC) beyond low-density lipoprotein cholesterol (LDL-C) varies across cardiovascular disease (CVD) outcomes. This analysis evaluates the extent to which concentrations of these non-LDL-C biomarkers improve MI-specific risk prediction in a primary prevention population. Methods We analyzed 306,183 UK Biobank participants free of cardiovascular disease at baseline with available Lp(a), RC, and hsCRP measurements. RC was calculated as total cholesterol minus LDL-C minus high-density lipoprotein cholesterol (HDLC). Adjusted hazard ratios (HRs) and 95% confidence intervals (CIs) were estimated using Cox regression across biomarker quintiles and cumulative biomarker burden. The primary endpoint was first MI. Results Over 15 years of follow-up, 10,824 MI events occurred. In fully adjusted models comparing quintile 5 with quintile 1, HRs (95% CI) were 1.09 (1.08-1.11) for Lp(a), 1.14 (1.13-1.16) for RC, and 1.08 (1.06-1.10) for hsCRP. Per-SD increases were associated with higher MI risk for RC 1.22 (1.20-1.25), Lp(a) 1.16 (1.13-1.18), and hsCRP 1.13 (1.10-1.15). The risk of MI increased stepwise with cumulative biomarker burden; compared with individuals with no biomarker in the top quintile, HRs (95% CI) were 1.45 (1.39-1.51), 2.14 (2.02-2.26), and 2.83 (2.48-3.24) for those with one, two, or all three elevated biomarkers, respectively. Conclusions Lp(a), RC, and hsCRP each provide independent and complementary information for MI risk. Their combined elevation identifies individuals at higher MI risk, suggesting selective testing of all three biomarkers in primary prevention.</p>