Small dense LDL, clinical characteristics and survival - an analysis of LIPIDOGRAM2004-2015 cohort studies

Background LDL cholesterol (LDL-C) is a well-established causal factor and treatment target in cardiovascular (CVD) prevention. Small dense LDL cholesterol (sdLDL-C) particles are known to be more atherogenic than large buoyant particles. Purpose To assess the clinical characteristics associated with sdLDL and its relationship with patient survival. Methods LIPIDOGRAM studies were conducted in Poland in 2004, 2006, and 2015, involving a population representative of patients in a primary care setting. The median follow-up duration was 5,570 days and survival data up to December 2021 were available. sdLDL-C was calculated using common lipid parameters with an equation published by Sampson et al., with the 80th percentile used as the cut-off for elevated sdLDL (as suggested by Sampson et al.). Multivariable analysis for clinical characteristics was conducted by iteratively removing variables with a variance inflation factor exceeding 5. Survival analyses excluded patients with post-myocardial infarction (MI) status, follow-up below one year, or those on lipid-lowering drugs, using a spline-based approach. Multivariable analysis was adjusted for HDL-C, age, sex, waist circumference (WC), diabetes mellitus, education level, hypertension, and smoking. Results A total of 45,591 patients were analyzed. The median sdLDL-C was 1.042 mmol/l (IQR: 0.823 - 1.307). The 80th percentile of sdLDL-C was 1.378 mmol/l. In general, the elevated sdLDL group had higher BMI, WC, and fasting glucose, as well as increased prevalence of hypertension, obesity, and smoking, along with a lower prevalence of previous MI (all p<0.001) and no significant difference in statin use (p=0.405). In the univariate regression analysis, HbA1c, glucose, and metabolic syndrome were strongly correlated with sdLDL-C, while HDL-C, higher education, and female sex negatively correlated with sdLDL-C. Multivariable analysis in the 2015 cohort indicated that factors positively associated with sdLDL-C included smoking, CVD family history, and age, while negative correlations were observed for diabetes, atrial fibrillation, statin use, previous MI, and HDL-C. Spline-based survival analysis revealed an almost linear relationship between sdLDL-C and the hazard of death from all causes (p<0.001), although multivariable analysis produced a U-shaped curve. Conclusions sdLDL-C may serve as a prognostic marker for cardiovascular disease or overall mortality, but further research is needed to understand its clinical usefulness and impact on major cardiovascular events.Figure 1.Results of regression analysis.Figure 2.Results of survival analysis.