Use of troponin T and creatine kinase-MB subunit levels for risk stratification of emergency department patients with possible myocardial ischemia.

STUDY OBJECTIVES

To evaluate and compare the utility of measurement of troponin T and the creatine kinase MB subunit (CK-MB) for risk stratification of ED patients with possible myocardial ischemia.

METHODS

Prospective observational study of ED patients with symptoms of possible myocardial ischemia with early, single sample serologic testing for cardiac troponin T and CK-MB using an identity-unlinked process. Chart review (ED, inpatient, outpatient), and telephone and mail surveys identified adverse events (AEs) during the 14 days following enrollment. AEs recorded included death, respiratory or cardiac arrest, myocardial infarction (MI), atrial and ventricular arrhythmias, pulmonary edema, conduction disturbances, and recurrent angina. Measures of the predictive ability for AEs were calculated for troponin T, CK-MB, and a combined troponin T/CK-MB index (defined as positive if either troponin T or CK-MB levels exceeded threshold values).

RESULTS

Among 292 study patients, 45 (15.4%) experienced at least one AE, including seven deaths and 12 MIs. The troponin T result was positive in 34 patients, and the CK-MB result was positive in 15 patients; 6 patients had positive results for both markers and 43 patients had a positive combined troponin T/CK-MB index. Odds ratios (ORs) for occurrence of AEs among all patients were 4.4 (1.8 to 10.2), 10.0 (3.0 to 36.0), and 4.5 (2.0 to 9.8) for troponin T, CK-MB and the troponin T/CK-MB index, respectively. Both markers were individually predictive of AEs (troponin T = 4.3; CK-MB = 7.5) among all those with chest pain. Only the CK-MB level was significantly predictive of AEs among those presenting with symptoms other than chest pain (OR = 24.3 [1.1, 1448]), whereas only the troponin T level was significantly predictive among patients representing a disposition dilemma for the emergency physician (OR = 5.7 [1.4, 20.7]). When compared, the ORs for troponin T and CK-MB were not significantly different for any patient subgroup. The troponin T/CK-MB index did not have a higher prognostic value than either troponin T or CK-MB alone in any subgroup studied.

CONCLUSION

A positive test result for either troponin T or CK-MB in the ED successfully identified patients at significantly higher risk of adverse events during the 2 weeks following their ED visit. The two markers may complement each other in that each appears to have prognostic ability among a unique patient subgroup. ED marker measurement can provide useful prognostic information for patients with a broad spectrum of presentations consistent with possible myocardial ischemia.