Abstract 12239: C-reactive Protein as a Predictor of Coronary Events and Coronary Artery Disease Prevalence, Extent, or Severity by Coronary CT Angiography

Abstract

Background. The association of C-reactive protein (CRP) with coronary calcium score, coronary artery disease (CAD) assessed by invasive angiography, and adverse coronary events has been studied with widely variable results. CRP as a predictor of Coronary Computed Tomography Angiographic (CCTA) findings and their association with downstream outcomes has not been extensively evaluated.

Methods. Retrospective cohort of all CCTA scans at two large tertiary hospitals within Partners Healthcare system (Boston, Massachusetts) from Sep 2004 to Oct 2011. We included all consecutive patients > 18 years old without known prior CAD who underwent CCTA and CRP assessment within 6 months. Diagnostic performance of CRP for prediction of any CAD, >50% and >70% stenosis, Segment Stenosis Score (SSS), Segment Involvement Score (SIS) was evaluated using sensitivity (Se), specificity (Sp), area under receiver operating characteristic curve (AUC), and likelihood ratios (-LR and +LR). Survival analysis using Cox proportional hazards was used to evaluate CRP as a predictor of adverse events (death, non-fatal myocardial infarction, late revascularization [>90 days post CCTA], or unstable angina).

Results. 250 patients with a mean age 56+/-13 and median follow up of 3.6 years. 96 (38%) had normal CCTA, 91(36%) had 1.0 mg/dL had a 74% Se and 20% Sp; while for CRP>3, 32% Se and 56% Sp; and for CRP > 10, 12% Se and 86% Sp. AUC for CRP was 0.44. CRP was similarly not predictive of >50% stenosis, >70% stenosis, SIS, or SSS. There were 9 (3.6%; 1.1% per year) combined adverse cardiac events (4 deaths and 5 late revascularizations) over median 3.6 years. Median CRP differed when stratified by mortality (median CRP 1.9 [1 - 4.8] alive versus 8.5 [4.1 - 11.2] for death, p = 0.046), but not other events given low event rates. Unadjusted hazard ratio (HR) for CRP > 3 to predict death was 5.3 (0.55 - 51.1, p = 0.1) and combined events was 0.88 (0.22 - 3.53, p = 0.9). Conclusion. CRP had no association with angiographic CAD. A trend for elevated CRP was noted in association with death suggesting CRP may predict outcomes independently of CAD, but events were rare. CRP’s prognostic value in this population may be low due to low sensitivity and low event rates.