Association of lipoprotein(a) with coronary artery disease presence and severity according to CAD-RADS on CCT
A Rosca, B T Theodora Benedek, B B M Botond Barna Matyas, I B Imre BenedekAbstract
Background
Lipoprotein(a) (Lp(a)) is a genetically determined lipoprotein particle recognized as an independent risk factor for atherosclerotic cardiovascular disease. However, its relationship with coronary plaque burden and anatomical disease severity assessed by coronary computed tomography (CCT) remains incompletely defined.
Purpose
To investigate the association between circulating Lp(a) levels and coronary artery disease (CAD) presence, plaque composition, coronary artery calcium score (CACs), and CAD-RADS classification in patients undergoing CCT.
Methods
This cross-sectional study included 106 consecutive patients undergoing CCT. Patients were stratified into Group 1 (Lp(a) < 30 mg/dL, n = 67) and Group 2 (Lp(a) > 30 mg/dL, n = 39). CAD was defined as the presence of any atherosclerotic plaque. Plaque type was classified as non-calcified (NCP), mixed (MP), or calcified (CP). CACs and CAD-RADS categories (0–1, 2–3, 4–5) were recorded. ROC analysis evaluated the diagnostic performance of Lp(a) for CAD detection, and regression analysis assessed the association between Lp(a) levels and CAD-RADS score.
Results
Baseline demographic and cardiovascular risk factors were comparable between groups, including age (58.49 ± 9.54 years), male sex (61.3%), BMI (median 27.8 kg/m²), hypertension (55.7%), dyslipidemia (67.0%), diabetes (17.9%), and smoking (34.0%) (all p > 0.05). Overall, CAD was present in 64 patients (60.4%). CAD prevalence was higher in Group 2 compared with Group 1 (69.2% vs. 55.2%), though not statistically significant (p = 0.2166). Regarding plaque phenotype, Group 2 exhibited higher proportions of NCP (28.2% vs. 14.9%) and MP (30.8% vs. 20.9%), and a lower proportion of CP (10.3% vs. 19.4%); however, these differences did not reach statistical significance (p = 0.0981 for NCP). Mean CACs were similar between groups (54.64 ± 115.9 vs. 36.54 ± 79.70; p = 0.3446), and CACs category distribution showed no significant intergroup differences. CAD-RADS distribution demonstrated a non-significant trend toward more advanced categories in Group 2, with CAD-RADS 4–5 observed in 20.5% versus 14.9% in Group 1 (p = 0.4600). ROC analysis demonstrated good discriminatory performance of Lp(a) for CAD presence (AUC = 0.801; 95%CI: 0.71 – 0.87; p < 0.0001). The optimal cut-off value for CAD detection was > 10.7 mg/dL (Youden index 0.58), yielding 93.75% sensitivity and 64.29% specificity. Linear regression analysis revealed a significant positive association between Lp(a) and CAD-RADS score (r = 0.1039, p = 0.0008).
Conclusions
Elevated Lp(a) levels are associated with a higher prevalence of CAD and a trend toward more non-calcified and mixed plaque phenotypes on CCT. Lp(a) demonstrated good discriminatory ability for CAD detection and was independently associated with increasing CAD-RADS score. These findings support the role of Lp(a) as a relevant biomarker linked to coronary disease burden and severity by CCT.Lp(a) vs CAD-RADS regression analysis.ROC curve of Lp(a) for CAD detection.