Low Predictive Value of Cardiac Troponin-T in Low-Risk Chest Pain Patients

abstract & commentary

Source: Peacock WF, et al. Prediction of short- and long-term outcomes by troponin-T levels in low-risk patients evaluated for acute coronary syndromes. Ann Emerg Med 2000;35:213-220.

The authors of this study sought to determine the ability of serum cardiac troponin-T (cTnT) to predict both short- and long-term complications in patients admitted to an ED observation unit for chest pain.

Patients were eligible if they presented to the ED with chest pain and were judged to be at low risk for acute myocardial infarction (MI). All had stable vital signs, no ECG evidence of acute MI, and initial CK-MB index less than 4%. Serum cTnT levels were obtained at the time of enrollment and again at four, eight, and 16 hours after presentation. Adverse outcomes were defined as MI, unstable angina, need for revascularization procedure, and death from any cause either during the initial hospitalization period or within six months of the index visit.

Of the 266 patients in the study group, two were diagnosed with an acute MI during the initial hospitalization; only one had an elevated cTnT. cTnT was positive in one of the 18 patients diagnosed with unstable angina and in two of the five patients undergoing revascularization.

Overall, cTnT was positive in only two of the 20 patients (10%) sustaining an adverse outcome during the initial hospitalization and in none of the 20 patients with an adverse outcome within six months of the initial period. The authors conclude that cTnT has very low sensitivity in predicting adverse cardiac outcomes among low-risk chest pain patients and should not be exclusively relied upon for making disposition decisions.

Comment by David J. Karras, MD, faaem, facep

Most prior studies of the utility of cTnT had found that the test approaches 100% sensitivity for diagnosing acute MI within 12 hours after symptom onset. The finding of only 50% sensitivity for cTnT in diagnosing acute MI must be taken in context; only two patients ruled-in for MI during the initial visit. The poor overall sensitivity of cTnT in predicting other adverse cardiac events is substantially lower than that documented by other studies. Nonetheless, these findings remind us that no serum cardiac marker is a valid substitute for clinical judgment.

While positive cTnT (or troponin-I) values are highly predictive of adverse cardiac outcomes, negative values at even four hours or longer after symptom onset appear to have limited prognostic value and must be interpreted in light of other clinical information.