Excess atherosclerotic cardiovascular burden associated with elevated lipoprotein(a) across age, sex, and cardiometabolic risk factors: insights from 16-year follow-up of the HUNT Study
Y Khalil, E K Vesterbekkmo, K B Holven, H Dalen, K Retterstol, S Blom, B O Asvold, E B HaugAbstract
Background
Lipoprotein(a) [Lp(a)] is a causal risk factor for atherosclerotic cardiovascular disease (ASCVD), but prevention strategies should be guided by where the absolute excess burden of events is greatest.
Purpose
To test additive interaction between elevated Lp(a) and other ASCVD risk factors, and to estimate absolute excess ASCVD hazard associated with elevated Lp(a) across cardiometabolic profiles at different ages.
Methods
We included 19 495 adults without ASCVD at baseline from the Trøndelag Health Study (2006–08) and followed them for up to 16 years for first ASCVD event. Lp(a) was dichotomised as >105 versus ≤105 nmol/L, and pre-specified risk factors were classified into higher and lower-risk strata. Multivariable cause-specific Cox models were used to estimate hazard ratios (HRs) and to evaluate additive interaction using the relative excess risk due to interaction (RERI). Sex-specific marginal hazards and hazard rate differences (HRDs) of ASCVD per 1000 person-years were estimated at age 60 across cardiometabolic markers. Sensitivity analyses used Lp(a) ≥96th versus 1–74th percentiles.
Results
HRs were generally highest when elevated Lp(a) coexisted with the higher-risk stratum of the risk factor. Additive interaction was observed for older age (RERI 0.60; 95% CI: 0.10–1.10) and male sex (RERI 0.63; 95% CI: 0.27–0.99). Using ≥96th versus 1–74th percentiles, smoking also showed additive interaction (RERI 0.57; 95% CI: 0.05–1.08). RERI estimates for remaining risk factors were small. At age 60, hazard rates of ASCVD were higher for elevated versus low Lp(a) across all risk factors and were generally larger in men. Across mean systolic blood pressure, the excess burden associated with elevated Lp(a) increased with higher levels in men, reaching 7.88 additional events per 1000 person-years at 180 mmHg (95% CI: -1.38–17.14), whereas women showed a smaller increase (HRD 3.89; 95% CI: -0.74–8.53). Across triglycerides-to-high-density lipoprotein cholesterol (TG:HDL-C) ratio, the excess burden in men peaked around a TG:HDL-C ratio of ~1.4, reaching 7.41 additional events per 1000 person-years (95% CI: 4.42–10.41), while in women the HRDs were relatively constant. In contrast, HRDs varied little across other risk factors, including C-reactive protein, low-density lipoprotein cholesterol, and non-high-density lipoprotein cholesterol. Similar patterns were observed at older ages.
Conclusions
Elevated Lp(a) shows clinically relevant additive interaction with age and sex, while interaction with most other cardiometabolic risk factors appears limited. Nevertheless, absolute excess ASCVD burden varies across selected cardiometabolic profiles and is consistently greater in men. These findings support a clear communication of the absolute excess burden associated with elevated Lp(a) and may help inform the most relevant target groups for future Lp(a)-lowering therapies. Larger cohorts are needed to confirm these findings.