BACKGROUND: Early rule-in/rule-out of myocardial infarction (MI) in patients presenting to the emergency department (ED) is important for patient care and resource allocation. Given that dysglycemia is a strong risk factor for MI, we sought to explore and compare different combinations of cardiac troponin (cTn) cutoffs with glycemic markers for the early rule-in/rule-out of MI. METHODS: We included ED patients (n ϭ 1137) with symptoms suggestive of acute coronary syndrome (ACS) who had cTnI,

more »

... h-sensitivity cTnI (hs-cTnI), hs-cTnT, glucose, and hemoglobin A 1c (Hb A 1c ) measurements. We derived rule-in/rule-out algorithms using different combinations of ROC-derived and literature cutoffs for rule-in and rule-out of MI within 7 days after presentation. These algorithms were then tested for MI/ cardiovascular death and ACS/cardiovascular death at 7 days. ROC curves, sensitivity, specificity, likelihood ratios, positive and negative predictive values (PPV and NPV), and CIs were determined for various biomarker combinations. RESULTS: MI was diagnosed in 133 patients (11.7%; 95% CI, 9.8 -13.8). The algorithms that included cTn and glucose produced the greatest number of patients ruled out/ruled in for MI and yielded sensitivity Ն99%, NPV Ն99.5%, specificity Ն99%, and PPV Ն80%. This diagnostic performance was maintained for MI/cardiovascular death but not for ACS/cardiovascular death. The addition of hemoglobin A 1c (Hb A 1c ) (Ն6.5%) to these algorithms did not change these estimates; how-ever, 50 patients with previously unknown diabetes may have been identified if Hb A 1c was measured. CONCLUSIONS: Algorithms incorporating glucose with cTn may lead to an earlier MI diagnosis and rule-out for MI/cardiovascular death. Addition of Hb A 1c into these algorithms allows for identification of diabetes. Future studies extending these findings are needed for ACS/ cardiovascular death. ClinicalTrials.gov identifier: NCT01994577 Chest pain is one of the most common presenting complaints to emergency departments (EDs) 7 worldwide. In Canada, the combination of abdominal and chest pain represents the primary reason for ED visits and results in longer ED times than less emergent conditions (1 ). Part of the reason for these longer wait times may be attributed to guidelines which recommend a formal workup for suspected acute coronary syndrome (ACS), including serial measurements of cardiac troponin (cTn) and electrocardiograms (ECGs) for the diagnosis of myocardial infarction (MI) (2 ). Identification of low-risk patients is an attractive option to reduce wait times and unburden the ED. The introduction of high-sensitivity cTn (hs-cTn) assays offers the opportunity to not only diagnose MI, but also to rapidly rule out MI in the ED. We have previously demonstrated the utility of a dual-panel approach for deter-