Abstract 16287: Predictive Value of High Sensitivity Troponin T for Patients With Suspected Acute Coronary Syndromes Undergoing Invasive Coronary Angiography

Abstract

Introduction: High sensitivity troponin T (hs-cTnT) assays are used to risk stratify patients with suspected acute coronary syndromes(ACS), but the predictive value of hs-cTnT in patients undergoing invasive coronary angiography (ICA) has not been well described. Hypothesis: Hs-cTnT levels at baseline and the delta change with serial sampling will correlate with the findings of ICA, and the diagnostic performance of hs-cTnT may be affected by patient specific variables. Methods: This is a retrospective study of 1,037 patients that presented to the emergency department with suspected ACS who had hs-cTnT measured and underwent ICA. We collected patient demographics, medical history and results of cardiac testing. Our primary endpoint was a diagnosis of obstructive coronary artery disease (CAD) defined as a stenosis > 70% or > 50% in the left main. We excluded cases of arrhythmias, cardiac arrests, or STEMI. Data analyzed using multivariate regression models. Results: We analyzed data from 728 patients with a mean age of 68 years old, 62.5% were male and 37.5% female. 312 patients were diagnosed with no obstructive CAD and 413 with obstructive CAD, of which 286 underwent PCI, 43 underwent CABG and 84 were medically managed. There were significant (p<0.001) differences in average baseline hs-cTnT levels between no obstructive CAD (78.8 ng/L) and obstructive CAD (201.1 ng/L) patients, but no significant difference in the 1 hour hs-cTnT delta (30.4 ng/L vs 33.3ng/L) or the 3 hour delta (56.4ng/L vs 66.6ng/L). Being male, White race, or having diabetes were independent predictors of obstructive CAD with odd ratios of 2.42 (95% CI 1.75-3.34), 1.64 (95% CI 1.12-2.41), and 1.47 (95% CI 1.05-2.06) respectively. Conclusion: In this study, elevated baseline hs-cTnT was predictive of obstructive CAD but trending hs-cTnT had no additionalpredictive value. Average baseline hs-cTnT for no obstructive CAD was higher than the manufacture suggested cutoff (52 ng/L) to identify myocardial injury. Baseline variables of sex, race and history of diabetes were predictive of obstructive CAD regardless of the hs-cTnT. Future analysis will determine whether specific hs-cTnT thresholds in these populations could improve hs-cTnT performance for the identification of obstructive CAD.