Biochemical and clinical predictors of long-term outcome in patients with nonspecific chest pain and nondiagnostic electrocardiograms

Background There are few data assessing the relative value of clinical factors and sensitive cardiac markers in determining the long-term prognosis of patients with chest pain. Likewise, little information exists about the long-term outcome of patients with chest pain who have negative markers of myocardial cell necrosis. This study addresses these issues in a cohort of patients with nonspecific chest pain and nondiagnostic electrocardiograms (ECGs). Methods Eligible subjects (n = 501) had experienced > 15 minutes chest pain at rest during the previous 24 hours, but were found to be at low-risk for acute myocardial infarction (AMI) by means of a well-validated clinical algorithm. Cardiac troponin 1, creatine kinase MBmass, myoglobin, and myosin light chain-1 were collected at presentation and 3, 6, and 12 hours later. Patients were observed for a median of 3 1 months. The composite end point was death or AMI subsequent to the index admission. Results Cardiac troponin I was the best single biochemical predictor of outcome (risk ratio 2.34, 95% CI 1.31-4.17, P = .004), but was of less independent prognostic value than age and an abnormal presenting ECG. It was also inferior to a combination strategy, using all 4 markers tested (risk ratio 2.37, 95% CI 1.44-3.91, P < .001). Fifty of 428 patients (12%) with a cardiac troponin I level less than or equal to 0.2 ng/mL and 25 of 287 patients (9%) without elevation of any marker tested sustained an adverse event during follow-up. Conclusions Cardiac troponin I is the most useful single biochemical predictor of long-term outcome, but the best determinants are age, an abnormal presenting ECG, and an "any marker positive" strategy. Patients without elevated cardiac markers have an adverse event rate of approximately 10% in the subsequent 31 months.