Delta creatine kinase-MB outperforms myoglobin at two hours during the emergency department identification and exclusion of troponin positive non-ST-segment elevation acute coronary syndromes.

STUDY OBJECTIVE Limited information is available about the diagnostic performance of creatine kinase (CK)-MB and myoglobin levels during the early evaluation of chest pain patients using cardiac troponins as the criterion standard for diagnosing acute myocardial infarction. In this study, we compare the sensitivity and specificity of the baseline, 2-hour absolute, and 2-hour delta values of myoglobin and CK-MB mass assay for detection of acute myocardial infarction using cardiac troponin I (troponin) as the sole marker of myocardial necrosis. METHODS A prospective observational study was conducted of 975 chest pain patients with a baseline troponin level of 1.0 ng/mL or less (Abbott Axsym Assay) and an initial ECG nondiagnostic for injury. CK-MB, myoglobin, and troponin levels were all measured on the Abbott Axsym immunoassay. Acute myocardial infarction was diagnosed if there was at least 20 minutes of chest pain and any one of the following criteria within 24 hours of ED presentation: a serial increase in troponin to more than 1.0 ng/mL, new Q-wave formation in 2 contiguous leads, or patient death by cardiac or unknown cause. The optimal values of CK-MB and myoglobin were chosen at the most accurate value on the receiver operating characteristic (ROC) curve (ie, value with lowest false-negative and false-positive rate) of the 2-hour absolute and 2-hour delta value for predicting acute myocardial infarction. RESULTS Acute myocardial infarction was diagnosed in 44 (4.5%) of the 975 study patients. ROC curve analysis revealed no statistically significant differences in areas for myoglobin and CK-MB values at baseline and 2 hours for determination of acute myocardial infarction. However, the ROC curve area of the delta CK-MB level significantly outperformed the ROC curve area of the delta myoglobin level for early identification of acute myocardial infarction (0.97 versus 0.81; 95% confidence interval [CI] for difference between areas 0.09 to 0.24). At the most accurate cutoff value, a 2-hour delta CK-MB level more than 0.7 ng/mL had a sensitivity of 93.2% (95% CI 81.3% to 98.5%), a specificity of 94.4% (95% CI 92.7% to 95.8%), a positive likelihood ratio of 16.7, and a negative likelihood ratio of 0.07. CONCLUSION A 2-hour delta CK-MB level outperforms myoglobin level in the early identification and exclusion of acute myocardial infarction in non-ST-segment elevation chest pain patients. This finding suggests that myoglobin may no longer be the optimal early marker of acute myocardial infarction when troponins are used as the criterion standard.