Lipoprotein(a) defines a calcific coronary phenotype independent of LDL cholesterol in the Lp(a)-lowering era
Y Iemura, S Ueshima, K Kikkawa, S Naito, Y Horizawa, K Miyoshi, M Ogata, N Okuda, Y Kawahara, A Tani, T Murakami, T Ando, S Ishihara, H Komiyama, K JujoAbstract
Introduction
Despite optimal LDL cholesterol (LDL-C) control, genetically determined lipoprotein(a) [Lp(a)] remains a major source of residual cardiovascular risk1-2. Epidemiologic studies consistently link elevated Lp(a) to coronary events; however, whether this risk is mediated by a distinct calcific plaque phenotype in vivo—particularly one associated with adverse outcomes and procedural complexity—has not been clearly demonstrated3.
Purpose
We sought to determine whether elevated Lp(a) independently identifies a severe calcific coronary phenotype assessed by intravascular ultrasound (IVUS), and to define a clinically actionable threshold relevant to risk stratification and interventional planning.
Methods
We performed a retrospective analysis of 235 consecutive patients undergoing coronary angiography at a single tertiary center. Among them, 151 patients with both IVUS assessment and Lp(a) measurement were included. Coronary calcification was quantified using an IVUS-derived calcification score (CAC 0–4), and severe calcification was defined as CAC ≥3. Associations between Lp(a) and calcification severity were evaluated using correlation analysis and multivariable logistic regression adjusting for age, sex, LDL-C, diabetes, renal function, and acute coronary syndrome (ACS) presentation. To depict potential nonlinearity and clinical risk gradients, restricted cubic spline functions were fitted within the logistic model to generate adjusted probabilities of severe calcification across the range of Lp(a) and LDL-C. Discrimination for severe calcification was assessed by receiver operating characteristic (ROC) analysis, with the optimal cutoff determined by the Youden index.
Results
The mean age was 72.4±11.6 years, and 77% were male. Median Lp(a) was 19.0 mg/dL and median LDL-C was 88.0 mg/dL. Lp(a) increased stepwise across CAC categories and demonstrated a strong positive correlation with calcification severity (Spearman’s r=0.60, p<0.001; Figure 1). In multivariable analysis, Lp(a) remained independently associated with severe calcification after adjustment for LDL-C and other clinical covariates (adjusted OR 1.78 [1.42–2.22] per 10 mg/dL, p<0.001). Spline-derived adjusted probability curves showed a progressive increase in the probability of severe calcification with higher Lp(a), whereas the corresponding LDL-C curve remained essentially flat (Figure 2). ROC analysis showed good discrimination for severe calcification (AUC 0.81), with an optimal Lp(a) threshold of 30 mg/dL.
Conclusions
Elevated Lp(a) independently identifies a high-risk calcific coronary phenotype beyond LDL-C levels. These findings provide in vivo structural insight linking genetically mediated lipoprotein biology to plaque remodeling. Lp(a) identifies a distinct calcific coronary phenotype independent of LDL-C, providing structural mechanistic support for Lp(a)-targeted therapy.