Association of high-density lipoprotein cholesterol with all-cause and cause-specific mortality in a Chinese population of 3.3 million adults: a prospective cohort study.
Level 3 - non-randomized controlled study
Prospective cohort study
PubMed 38357392 · doi:10.1016/j.lanwpc.2023.100874
What was done
The China Health Evaluation And risk Reduction through nationwide Teamwork (ChinaHEART) prospective cohort study evaluated 3,397,547 Chinese adults aged 35–75 years with a median follow-up of 3.9 years. Baseline HDL-C concentrations were measured, and all-cause, cardiovascular disease (CVD), and cancer mortality were tracked via China's National Mortality Surveillance System and Vital Registration.
What was found
HDL-C exhibited a U-shaped association with all-cause, CVD, and cancer mortality, with the lowest risk observed between 50 and 79 mg/dL. Compared to the lowest-risk groups: - HDL-C <30 mg/dL had adjusted hazard ratios (HR) of 1.23 (95% CI 1.17–1.29) for all-cause mortality, 1.33 (95% CI 1.23–1.45) for CVD mortality, and 1.18 (95% CI 1.09–1.28) for cancer mortality. - HDL-C >90 mg/dL had adjusted HRs of 1.10 (95% CI 1.05–1.15) for all-cause mortality, 1.09 (95% CI 1.01–1.18) for CVD mortality, and 1.11 (95% CI 1.03–1.19) for cancer mortality. - Similar U-shaped associations occurred for ischemic heart disease, ischemic stroke, and liver cancer. - Abnormal HDL-C accounted for an estimated 3.25% of all-cause mortality. The primary mortality contributor was ischemic heart disease (16.06 deaths per 100,000 persons, 95% UI 10.30–22.67) for HDL-C <40 mg/dL and esophageal cancer (2.29 deaths per 100,000 persons, 95% UI 0.57–4.77) for HDL-C >70 mg/dL.
Why it matters
In a massive cohort of over 3 million individuals, very high HDL-C was paradoxically linked to higher mortality, challenging the assumption that higher HDL-C is always protective.
Limits
Follow-up was relatively short at a median of 3.9 years, which raises the possibility of reverse causality. HDL-C was measured only once at baseline, unmeasured residual confounding cannot be ruled out in an observational cohort, and findings may not generalize to non-Chinese populations.
Cited by
- supports HDL cholesterol levels exceeding 100 mg/dL are frequently dysfunctional or ineffective as cardiovascular bioindicators.