Otvos · Journal of clinical lipidology 2011 · prospective cohort study · n=6814

Clinical implications of discordance between low-density lipoprotein cholesterol and particle number.

Cited 387 times in the scientific literature.

Level 3 - non-randomized controlled study

Prospective cohort study

PubMed 21392724 · doi:10.1016/j.jacl.2011.02.001 · record verified 2026-08-30

What was done

Baseline nuclear magnetic resonance (NMR) spectroscopy-measured LDL particle number (LDL-P), calculated LDL cholesterol (LDL-C), and carotid intima-media thickness (IMT) were evaluated in 6,814 participants free of clinical cardiovascular disease (CVD) from the Multi-Ethnic Study of Atherosclerosis (MESA). Participants were followed for incident CVD events over 5.5 years (319 events). Discordance was defined as LDL-P and LDL-C percentile values differing by ≥ 12 percentile units. Associations per 1 standard deviation difference in LDL-C or LDL-P were evaluated with adjustment for age, sex, and race.

What was found

Overall, both LDL-C (hazard ratio [HR] 1.20, 95% CI 1.08–1.34) and LDL-P (HR 1.32, 95% CI 1.19–1.47) were associated with incident CVD. Among participants with discordant LDL-C and LDL-P values, only LDL-P remained associated with incident CVD (HR 1.45, 95% CI 1.19–1.78), whereas LDL-C was not significantly associated (HR 1.07, 95% CI 0.88–1.30). Carotid IMT also tracked with LDL-P: adjusted mean IMT was 958 µm in the LDL-P > LDL-C subgroup, 932 µm in the concordant group, and 917 µm in the LDL-P < LDL-C subgroup. This IMT difference persisted after adjusting for LDL-C (P = .002), but not after adjusting for LDL-P (P = .60).

Why it matters

These findings indicate that cardiovascular risk and subclinical atherosclerosis are driven more directly by the circulating number of atherogenic particles than by total cholesterol content when the two metrics diverge.

Limits

The study is observational, precluding causal inference. Follow-up was relatively short (5.5 years with 319 CVD events). Calculations relied on baseline lipid measurements, and the abstract does not report whether subsequent lipid-lowering therapies or other confounders beyond age, sex, and race were adjusted for.

Cited by