Performance of a sensitive troponin assay in the early diagnosis of acute myocardial infarction in the emergency department

Aim: Troponin assays have high diagnostic value for myocardial infarction (MI), but sensitivity has been weak early after chest pain onset. New, so-called 'sensitive' troponin assays have recently been introduced. Two studies report high sensitivity for assays taken at ED presentation, but studied selected populations. Our aim was to evaluate the diagnostic performance for MI of a sensitive troponin assay measured at ED presentation in an unselected chest pain population without ECG evidence of ischaemia.

Methods: This is a sub-study of a prospective cohort study of adult patients with potentially cardiac chest pain who underwent evaluation for acute coronary syndrome. Patients with clear ECG evidence of acute ischaemia or an alternative diagnosis were excluded. Data collected included demographic, clinical, ECG, biomarker and outcome data. A 'positive' troponin was defined as >99th percentile of the assay used. MI diagnosis was as judged by the treating cardiologist. The outcomes of interest were sensitivity, specificity and likelihood ratios (LR) for positive troponin assay taken at ED presentation. Data were analysed by clinical performance analysis.

Results: Totally 952 were studied. Median age was 61 years; 56.4% were male and median TIMI score was 2. There were 129 MI (13.6, 95% CI 11.5-15.9). Sensitivity of TnI at ED presentation was 76.7% (95% CI 68.5-83.7%), specificity 93.6% (95% CI 91.7-95.1%), with LR positive 11.92 and LR negative 0.25.

Conclusion: Sensitive TnI assay at ED presentation has insufficient diagnostic accuracy for detection of MI. Serial biomarker assays in patients with negative initial TnI are required.