Abstract 12641: Absolute Change in Troponin Outperforms Relative Change in Troponin for Discrimination of Acute Myocardial Infarction: The Atherosclerosis Risk in Communities (ARIC) Study

Abstract

Introduction: The rise or fall of cardiac troponin plays a central role in classification of acute myocardial infarction (AMI) by the Third Universal Definition. Clinical Practice Guidelines (CPG) advocate using a 20% change in serial troponin measurements for AMI detection. However, it is uncertain whether absolute or relative change in troponin has greater diagnostic accuracy. Hypothesis: Absolute and relative change in troponin will have differing abilities to detect AMI in patients hospitalized with chest pain and elevated troponin. Methods: Since 1987, the ARIC study has conducted surveillance of hospitalized events in 4 US communities. Clinical and laboratory data were abstracted, including troponin values from 1996 onward. AMI was classified by ARIC with a validated algorithm, using elevated biomarkers, presence of chest pain, and ECG evidence. Our analysis was based on patients with chest pain and serial troponin measurements (day 1 and day 2 of hospitalization), with elevated troponin on day 1. Patients undergoing revascularization procedures <24 hours after event onset were excluded. The absolute and relative change in troponin between days 1 and 2 were calculated, and outcome discrimination was analyzed using the area under the receiver operating characteristic curve (AUC). All analyses were weighted by the inverse of the sampling probability. Results: A total of 6,232 hospitalizations spanning January 1 1996 - December 31, 2012 met our inclusion criteria. Most patients were white (64%) and male (60%), with a mean age of 63 years. Troponin I was more often assayed (93%) than troponin T. A total of 5,408 (87%) were classified as AMI by the ARIC algorithm, with 10% as STEMI. Absolute change in troponin outperformed relative change in troponin for detection of AMI (AUC: 0.69 vs. 0.64), NSTEMI (AUC: 0.68 vs. 0.64), and STEMI (AUC: 0.77 vs. 0.65); p <0.001 for all comparisons. The recommended 20% cut-point endorsed by CPG diagnosed AMI, NSTEMI, and STEMI similarly, with a sensitivity of 75% and specificity of 44%. Conclusion: In patients hospitalized with chest pain and elevated troponin, absolute and relative change in serial troponin values detected AMI with moderate diagnostic accuracy; however, absolute change in troponin was a better discriminator.