Abstract
Despite the causal role of cigarette smoking in atherosclerotic cardiovascular disease (ASCVD), the underlying mechanisms are not fully understood. We evaluated the joint relation between smoking and inflammatory markers with ASCVD risk. We tested cross-sectional associations of self-reported smoking status (never, former, current) and intensity (packs/day) with lipoprotein-associated phospholipase A(2) (Lp-PLA(2)) activity and high-sensitivity C-reactive protein (hsCRP) in 10,506 Atherosclerosis Risk in Communities participants at Visit 4 (1996 to 1998). Using Cox hazard models adjusted for demographic and traditional ASCVD risk factors, we examined the associations of smoking status and intensity with incident adjudicated ASCVD events (n = 1,745 cases) over an average of 17 years, stratified by Lp-PLA(2) and hsCRP categories. Greater packs/day smoked was linearly associated with higher levels of both Lp-PLA(2) and hsCRP among current smokers. Compared with never smokers, the hazard ratio for incident ASCVD in current smokers was 2.04 (95% CI 1.76 to 2.35). Among current smokers, the risk for ASCVD per 1 pack/day greater was 1.39 (1.10 to 1.76). Both Lp-PLA(2) activity ≥253 nmol/min/ml and hsCRP >3 mg/L identified current smokers at the highest risk for incident ASCVD, with similar hazard ratios. hsCRP risk-stratified current smokers better based on intensity. Among current smokers, hsCRP improved ASCVD prediction beyond traditional risk factors better than Lp-PLA(2) (C-statistic 0.675 for hsCRP vs 0.668 for Lp-PLA2, p = 0.001). In this large cohort with long follow-up, we found a dose-response relation between smoking intensity with Lp-PLA(2) activity, hsCRP, and ASCVD events. Although both Lp-PLA(2) activity and hsCRP categories identified high risk among current smokers, hsCRP may better stratify risk of future ASCVD.