Early and late outcome prediction of death in the emergency room setting by point-of-care and laboratory assays of cardiac troponin I.

BACKGROUND: Point-of-care (POC) assays of cardiac troponins are common in the emergency department setting. The question raised was as follows: What is the clinical impact of the results of POC assays of cardiac troponins as compared with sensitive laboratory assays?
METHODS: Patients admitted consecutively to the emergency department (N = 1,069) and on whom cardiac troponins were requested as part of their clinical work-up were included. Cardiac troponin I (cTnI) was measured by the POC assays-i-Stat (Abbott Diagnostics, Abbott Park, IL) and Stratus CS (Siemens Healthcare Diagnostics, Deerfield, IL)-and by the laboratory assays-Access AccuTnI (Beckman Coulter, Fullerton, CA) and Architect cTnI (Abbott Diagnostics). Results were related to early (14 days) and late outcome (median 3.3 months, range 0.1-35) as to death.
RESULTS: The laboratory assays identified more patients (P < .001) with elevated levels than the two POC assays (39%-74% vs 20%-27%). Adopting the 99th percentiles upper reference limit, the Access AccuTnI identified 88% and Architect cTnI identified 81% of all patients who died of cardiovascular disease as compared with 50% and 54% for i-Stat and Stratus CS, respectively (P < .001). Negative predictive values for the laboratory assays were 97% as compared with 89% to 93% for the POC assays. Negative likelihood ratios were 0.25 (CI 0.15-0.041) and 0.59 to 0.68 (CI 0.47-0.79), respectively.
CONCLUSIONS: The current POC cTnI assays are less sensitive for outcome prediction of patients with myocardial injury. The clinical judgment of the patient with suspected myocardial ischemia should not solely rely on results from POC assays. If a clinical suspicion of myocardial injury remains despite negative cTnI results with the POC assays, such results should be complemented by results from sensitive laboratory assays.