Abstract 9256: Integrating High-Sensitivity C-Reactive Protein Into Risk Equations for Individuals with Established Atherosclerotic Cardiovascular Disease
Abstract
Background: Cardiovascular (CV) inflammation, measured by high-sensitivity C-reactive protein (hsCRP), is associated with higher residual CV risk in patients with atherosclerotic CV disease (ASCVD). HsCRP is included in the secondary CV risk prediction tool Secondary Manifestations of ARTerial disease 2 (SMART2) in Europe. However, risk prediction tools are not consistently used in US secondary prevention, and primary prevention risk prediction tools, such as the Predicting Risk of Cardiovascular Disease Events (PREVENT) equation, do not account for hsCRP. Research question: Does inclusion of hsCRP improve the ability of existing CV risk scores to assess risk of CV events in patients with ASCVD? Methods: This retrospective cohort study identified patients with established ASCVD, an hsCRP measurement, and all necessary data for the risk equations from Komodo’s Healthcare Map (01/01/2017-7/31/2025). Patients were followed for subsequent major adverse CV events (revised 3-point MACE: composite of all-cause mortality, nonfatal MI, and nonfatal stroke). Cox proportional hazards models were used to examine the association between hsCRP and CV events. Nested prediction models with and without hsCRP were compared via likelihood ratio tests. Net reclassification improvement (NRI) and integrated discrimination improvement (IDI) were used to compare performance of models using SMART2 and PREVENT risk factors, with and without hsCRP to explore the potential applicability of PREVENT risk factors in secondary prevention. Results: Of 7,285 included patients, 62.2% had coronary artery disease, 36.1% had peripheral artery disease, and 6.7% had ischemic stroke. The mean age was 64.3 years, and the mean follow-up time was 38.6 months. The majority of patients were White (60.2%) and 51.9% were male. At baseline, 37.2% of patients had diabetes, 73.8% were on antihypertensives, 71.1% were on statins, and 36.1% were on aspirin. After controlling for SMART2 risk factors, hsCRP was independently associated with higher risk of revised 3-point MACE (hazard ratio [HR] per 1 mg/L hsCRP increase, 1.04; 95% CI, 1.01-1.06) and all-cause mortality (HR, 1.07; 95% CI, 1.03-1.11). Similarly, hsCRP was independently associated with higher risk of revised 3-point MACE (HR, 1.04; 95% CI, 1.01-1.07) and all-cause mortality (HR, 1.08; 95% CI, 1.04-1.12) when controlling for PREVENT risk factors (Figure). Likelihood ratio tests indicated that models with hsCRP demonstrated significantly better model fit than models without hsCRP. The addition of hsCRP significantly improved the NRI and IDI for MACE prediction (Table). Conclusions: This study reaffirms the association of CV inflammation with CV risk. Current US risk scores do not account for inflammation and may underestimate patient risk. Integration of hsCRP into risk equations could better stratify patient risk and refine clinical decision-making.