The clinical value of estimated small dense LDL in primary and secondary prevention

Background Small-dense low-density lipoprotein (sdLDL) is more atherogenic than larger LDL particles, but its measurement in routine clinical practice has not been validated. Recent evidence reports good agreement between measured and estimated sdLDL levels using a readily applicable formula, which includes standard lipoprotein fractions. We evaluated the clinical value of estimated sdLDL (esdLDL) in primary and secondary prevention settings for atherosclerotic cardiovascular disease (ASCVD). Methods Analysis was conducted across three cohorts: an outpatients ASCVD primary and secondary prevention cohort (Athens Angiometabolic Cohort, n=1,327, (AAC)), a general population cohort (ATTICA study, n=1,590) and an acute coronary syndrome cohort (GREECS study, n=1,296). We examined the association between esdLDL and markers of subclinical atherosclerosis in AAC [carotid and femoral artery wall thickness (WT) and Ankle-Brachial Index (ABI)] and assessed its predictive value for major adverse cardiovascular events (MACE) including a composite of CV death, acute myocardial infarction, heart failure, revascularization and stroke in all cohorts after adjustment for established risk scores recommended in ESC guidelines. esdLDL was calculated using the formula: [esdLDL=LDL-C- 1.43*LDL-C+ 0.14*(LN(Triglyceride)*LDL-C) + 8.99.] Results esdLDL levels in the highest quintile were significantly associated with increased carotid (OR 1.60, p<0.001), and femoral WT (OR 1.37, p=0.043), with ABI<0.9 (OR 2.37 p=0.04) and number of diseased coronary artery vessels (OR: 0.606 95% CI: 0.309, 0.804, p<0.001) (Table 1). esdLDL associated with MACE in all three cohorts independently of SCORE2 or SMART2 risk scores based on presence of ASCVD (AAC, HR: 2.07 per highest vs lower quintiles of esdLDL95% CI 1.17–3.67, p=0.012, median follow-up 2.5 years, ATTICA, HR:1.31 95%CI: 1.07, 1.61, P=0.008, median follow-up 20 years, GREECS HR: 1.27, 95% CI: 1.07, 1.52, P=0.008, , median follow-up 14 years). Adding esdLDL to SCORE2 and SMART2 risk model improved the C-statistic (AUC: 0.629 to 0.655 for SCORE2 and AUC: 0.613 to 0.624 for SMART2) and yielded a significant Net Reclassification Improvement (NRI) of 0.360 (p<0.001) in the ATTICA study and NRI 0.115 (p=0.036) in the GREECS study. While esdLDL provided incremental value over LDL-C in ATTICA study ΔAUC 0.011, P=0.011, NRI: 0.148, p=0.003), this improvement was not significant in the ACS population (ΔAUC 0.002, P=0.497, NRI 0.08, P=0.12) Conclusion Estimated sdLDL calculated by a readily available formula, is associated with subclinical and clinically overt atherosclerosis and incrementally predicts long-term adverse cardiovascular events in 3 independent cohorts representative of the general population, outpatients assessed in ASCVD prevention clinics and ACS population. These findings should trigger further investigation regarding its utility in risk stratification in primary and secondary prevention.