Validation of the ESC 0/3h-algorithm with a novel high-sensitivity cardiac troponin I assay in patients with suspected myocardial infarction

Abstract Background There is uncertainty among physicians regarding the optimal clinical application of the novel high-sensitivity cardiac troponin I (hs-cTnI)-VITROS assay within the widely implemented upper reference limit (URL)-based ESC 0/3h-algorithm. Methods This prospective, international, multicenter study aimed to validate and compare different modifications of the URL-based ESC 0/3h-algorithm among adult patients presenting with suspected acute myocardial infarction (AMI) to the emergency department. Final diagnoses of AMI were centrally adjudicated by two independent cardiologists, blinded to hs-cTnI-VITROS concentrations. Discrimination was compared to the best validated hs-cTnT/I-assays. Clinical follow-up for cardiovascular death and AMI was conducted until five years. Results Among 1225 eligible patients (median age 63 years (IQR 52-75), 31.9% women), 18.9% were adjudicated as NSTEMI. Diagnostic discrimination of hs-cTnI-VITROS at 0h was very high (AUC 0.949, 95% CI 0.941–0.958), and comparable to hs-cTnI-Architect and hs-cTnT-Elecsys. Using the ESC 0/3h algorithm, 30.9% of patients were classified as rule-out (sensitivity 99.1%, NPV 99.5%) and 19.0% as rule-in (specificity 96.0%, PPV 82.8%). Substituting the HEART score for the GRACE score increased rule-out efficacy to 43.9%, while maintaining very high safety (sensitivity 99.6%; NPV 99.8%). Findings were consistent across predefined subgroups, in sensitivity analyses restricted to type 1 NSTEMI, using the 2h study blood sample (n=2249), and for the composite of cardiovascular death or AMI during five-year follow-up. Conclusions The ESC 0/3h-algorithm demonstrated excellent performance for rule-out of NSTEMI with greater efficacy when combined with the HEART-instead of the GRACE score. Registration https://clinicaltrials.gov/study/NCT00470587 Clinical Perspective What Is New? In a prospective, international multicenter cohort, the URL-based ESC 0/3h algorithm using the FDA-cleared hs-cTnI-VITROS assay achieved very high rule-out safety (sensitivity 99.1%, NPV 99.5%) but limited efficacy (30.9% rule-out) when combined with GRACE score