A Negative High-Sensitivity Troponin and Nonischemic ECG Can Rule Out AMI

Authors: Pickering JW et al., Ann Intern Med 2017 Apr 18;

However, samples for the troponin assay should be drawn at least 3 hours after symptom onset, and a positive troponin does not rule in acute myocardial infarction.

Several studies have demonstrated the potential utility of high-sensitivity cardiac troponin T (hs-cTnT) assays combined with electrocardiograms (ECGs) in the rapid rule-out of acute myocardial infarction (AMI) in adults presenting to the emergency department with chest pain (NEJM JW Emerg Med May 2010 and Circulation 2010; 121:1227; NEJM JW Emerg Med Jul 2016 and JAMA Cardiol 2016; 1:397; NEJM JW Emerg Med Sep 2016 and Ann Emerg Med 2016; 68:649). Researchers have now conducted a meta-analysis to determine the sensitivity of a single negative hs-cTnT plus a nonischemic ECG for ruling out AMI in these patients.

A total of 11 studies with 9241 patients (64% men, mean age 61 years) met inclusion criteria. Of included patients, 2825 (30.6%) were deemed to be low risk (hs-cTnT<0.005 µg/L and nonischemic ECG). Only 14 low-risk patients (0.5%) had AMI during the index hospital admission, and in half of them blood was drawn for the single hs-cTnT test less than 3 hours after symptom onset. Pooled test characteristics showed sensitivity of 98.7%, specificity of 64.3%, negative predictive value of 99.3%, and positive predictive value of 21.8%. However, 9 of the 11 studies were judged to have a high or unclear risk of bias due to patient selection and missing data.

Comment

These results support the use of a single negative hs-cTnT and nonischemic ECG to rule out AMI in adult ED patients with chest pain. However, the hs-cTnT level must be measured at least 3 hours after symptom onset (which can be difficult to define), and we must consider the overtesting that will result from false positive results, given the PPV of only 22%. Notably, the hs-cTnT cutoff used in this meta-analysis (<0.005 µg/L) is slightly lower than that approved by the FDA (<0.006 µg/L), so the sensitivity may be slightly lower for the U.S. version of the assay.