Objective: Elevations of serum troponin T and I values are being used to diagnose acute myocardial infarction (AMI) and to rule out the condition in patients before their discharge from the emergency department (ED). However, the sensitivity and specificity of these tests vary considerably. Our goal was to systematically review the data on the accuracy of troponin T and I for the diagnosis of AMI in the ED.
Search strategy: We searched the MEDLINE database using the following strategy: troponin (text word) and diagnosis (medical subject heading [MeSH]) or troponin/diagnostic use (MeSH). The references of articles meeting our inclusion criteria were searched for additional articles.
Selection criteria: We evaluated each study for quality. Only prospective blinded cohort studies with an adequate reference standard were included in the analysis.
Data collection/analysis: Data from each study were abstracted by 2 investigators. We graphed sensitivity and specificity for different points in time from arrival in the ED or from the onset of pain and calculated summary estimates when appropriate and possible.
Main results: Sensitivity increases for both troponin T and I from 10% to 45% within 1 hour of the onset of pain (depending on the cutoff) to more than 90% at 8 or more hours. Specificity declines gradually from 87% to 80% from 1 to 12 hours after the onset of chest pain for troponin T and is approximately 95% for troponin I. The peak abnormal value in the first 24 hours after admission to the ED has an area under the receiver operating characteristic curve of 0.99 and is very useful at ruling out AMI if negative.
Conclusions: Although troponin T and I values are useful tools for the diagnosis of AMI, they must be interpreted according to the number of hours from the onset of chest pain. The test is particularly useful at ruling out MI when the value is negative at 8 or more hours after the onset of chest pain.