Incremental value of high-sensitivity cardiac troponin T for risk prediction in patients with suspected acute myocardial infarction

BACKGROUND High-sensitivity cardiac troponin assays have better analytical precision and sensitivity than earlier-generation assays when measuring cardiac troponin at low concentrations. We evaluated whether use of a high-sensitivity assay could further improve risk stratification compared with a standard cardiac troponin assay. METHODS We enrolled consecutive patients presenting with acute chest pain, 30% of whom were diagnosed with acute coronary syndrome. Blood samples were drawn at the time of presentation. We measured cardiac troponin T with a standard fourth-generation assay (cTnT) and a high-sensitivity assay (hs-cTnT) (both Roche Diagnostics) and followed the patients for 24 months. RESULTS Of the 1159 patients, 76 died and 42 developed an acute myocardial infarction (AMI). Prognostic accuracy of hs-cTnT for death was significantly higher [area under ROC curve (AUC) 0.79, 95% CI 0.74–0.84] than that of cTnT (AUC 0.69, 95% CI 0.62–0.76; P < 0.001). After adjustment for Thrombolysis in Myocardial Infarction (TIMI) risk score (that included the cTnT assay result), hs-cTnT above the 99th percentile (0.014 μg/L) was associated with a hazard ratio for death of 2.60 (95% CI 1.42–4.74). Addition of hs-cTnT to the risk score improved the reclassification of patients (net reclassification improvement 0.91; 95% CI 0.67–1.14; P < 0.001). Subgroup analyses showed that this effect resulted from the better classification of patients without AMI at time of testing. hs-cTnT outperformed cTnT in the prediction of AMI during follow-up (P=0.02), but was not independently predictive for this endpoint. CONCLUSIONS Concentrations of hs-cTnT >0.014 μg/L improve the prediction of death but not subsequent AMI in unselected patients presenting with acute chest pain.