ApoB, LDL-C, and non-HDL-C as markers of cardiovascular risk.

BACKGROUND: Conventional statistical approaches are not designed to compare highly correlated variables such as low-density lipoprotein cholesterol (LDL-C), non-high density lipoprotein cholesterol (non-HDL-C), and apolipoprotein B (apoB). Discordance analysis was designed to overcome this limitation by creating groups in which the predictions of 2 markers differ. OBJECTIVE: This systematic review compiled all discordance studies that compare the predictive powers of LDL-C and non-HDL-C vs LDL particle number (LDL P) or apoB as markers of atherosclerotic disease risk to determine which is the most accurate marker of cardiovascular risk. METHODS: A PubMed search completed September 30, 2024, identified 15 studies involving 593,354 participants. These studies encompassed diverse populations, and included patients with and without statin therapy. Several variations of discordance analysis were used including median-based, percentile-based, residual-based, and variance-based approaches. RESULTS: ApoB outperformed LDL-C in 9 of 9 studies whereas LDL P was superior to LDL-C in 2 of 3 comparisons. In 1 study, non-HDL-C was superior to apoB, in 1 study apoB and non-HDL-C were equivalent, whereas in 7 studies, apoB, overall, was a significantly more accurate marker of atherosclerotic cardiovascular disease risk than non-HDL-C. CONCLUSION: Discordance analysis provides robust evidence that apoB is a more accurate marker of cardiovascular risk than either LDL-C or non-HDL-C, notwithstanding these variables are highly intercorrelated. Thus, neither LDL-C nor non-HDL-C are adequate clinical surrogates for apoB. Accordingly, apoB should be the primary measure in clinical care to estimate the cardiovascular risk attributable to the apoB lipoproteins and the adequacy of lipid-lowering therapy to reduce this risk.

Duke Scholars

Author Michael J Pencina Biostatistics & Bioinformatics, Division of Biostatistics