Longitudinal changes and borderline reclassification of Lipoprotein(a) compared with conventional lipids in over 230,000 adults

Background and aims: Lipoprotein(a) [Lp(a)] is a genetically determined cardiovascular risk factor. Although generally considered stable, data on intra-individual reclassification of Lp(a) categories over time compared with conventional lipid markers remain limited, particularly in East Asian populations. Methods: We conducted a retrospective cohort study including 230,018 Korean adults (mean age, 38·5 years; 51·4 % male) who underwent routine health examinations with at least two Lp(a) measurements between July 1, 2015, and December 30, 2022. Individuals with a history of coronary artery disease or those receiving lipid-lowering therapy were excluded. The primary outcomes were intraclass correlation coefficients (ICCs), reclassification rates across predefined Lp(a) categories (<30, 30–50, 50–100, and ≥100 mg/dL), and comparisons with traditional lipid parameters. Results: Lp(a) demonstrated excellent reproducibility (ICC ≥0·99) and greater long-term stability than LDL-C, ApoB, and triglycerides. Among participants with baseline Lp(a) levels <30 mg/dL or ≥100 mg/dL, 93 % and 91 % remained in the same category at follow-up, respectively. However, among those with baseline Lp(a) levels of 30–50 mg/dL, 35 % transitioned to higher-risk categories, as did 19 % of those with levels of 50–100 mg/dL. By comparison, 28 % of individuals with LDL-C ≥160 mg/dL and 22 % with triglycerides ≥200 mg/dL moved to lower categories during follow-up. Conclusions: Lp(a) demonstrates greater stability than conventional lipid markers. While a single measurement may suffice for clearly low (<30 mg/dL) or high (≥100 mg/dL) Lp(a) levels, periodic reassessment may be clinically warranted for individuals in intermediate Lp(a) levels (30–100 mg/dL).