Search

Question In patients with chest pain, how well do the NICE algorithms, based on high-sensitivity troponin (hsTn) I and T assays, rule out acute myocardial infarction (AMI)? Methods Design 3 cohort studies for post hoc validation of the recommended National Institute for Health and Care Excellence (NICE) algorithms. ACTRN12611001069943, NCT00470587. Setting 3 tertiary emergency departments (EDs) in Europe, Australia, and New Zealand. Patients 3580 patients ࣙ 18 years of age (mean age 55 to 65 y, 67% men) who presented to the ED with ࣙ 5 min of possible cardiac symptoms (acute chest, epigastric, neck, jaw or arm pain; or discomfort or pressure without an apparent noncardiac source) and received orders for serial biomarker testing (2 cohorts); or presented with symptoms of AMI (acute chest pain and angina pectoris), with onset or peak in the previous 12 hours (1 cohort). Exclusion criteria included ST elevation on electrocardiogram at presentation. Description of prediction guides hsTnI algorithm: AMI was ruled out if baseline hsTnI was ࣘ 2 ng/L (limit of detection) or if baseline and hsTnI repeated at 3 hours was < 26 ng/L (99th percentile of a healthy reference population, coefficient of variation [CV] < 5%). hsTnT algorithm: AMI was ruled out if baseline hsTnT was < 3 ng/L (limit of blank) or hsTnT repeated at 2 hours was < 14 ng/L (99th percentile, CV 10% at 13 ng/L) and absolute change within 2 hours was < 20%. Outcomes Primary outcome was cardiac death or AMI within 24 hours of presentation. Secondary outcome was major adverse cardiac events (MACE) comprising cardiac death, cardiac arrest, acute coronary syndromes (AMI or unstable angina), cardiogenic shock, or coronary revascularization at 30 days and sensitivity, specificity, and negative predictive value of the assays. Main results 15% of patients had AMI at presentation. 51 of 2506 (2.0%) patients ruled out by hsTnI and 7 of 1794 (0.4%) ruled out by hsTnT had AMI. 267 (11%) and 153 (8.5%) patients ruled out by hsTnI and hsTnT, respectively, had MACE at 30 days. Accuracy of the algorithms for ruling out AMI are shown in the Table. Conclusions In patients presenting with chest pain, the NICE algorithms, based on baseline and serial high-sensitivity troponin I and T assays, had > 89% sensitivity and 62% to 93% specificity for detecting acute myocardial infarction (AMI). > 8.5% of patients with AMI "ruled out" had major adverse cardiac events. [ABSTRACT FROM AUTHOR]