Validation of hs-cTnI-Alinity in the early diagnosis of acute myocardial infarction

OBJECTIVES: It is unknown whether high-sensitivity cardiac troponin I (hs-cTnI) measured on the Alinity platform (hs-cTnI-Alinity), can be used with the same clinical decision values as the Architect platform (hs-cTnI-Architect). METHODS: In an international prospective study, hs-cTnI-Alinity and hs-cTnI-Architect were measured from the same blood samples obtained at 0 h, 1 h, and 2 h from patients presenting to the emergency department with acute chest discomfort. The final diagnosis was centrally adjudicated by two cardiologists. We tested four hypotheses: hs-cTnI-Alinity and hs-cTnI-Architect have (1) excellent analytical agreement, (2) high concordance at common analytical benchmarks (limit-of-detection and upper-reference-limit); (3) high diagnostic accuracy for AMI; and (4) comparable clinical performance for early rule-out and rule-in of AMI. RESULTS: Among 1,573 eligible patients (median age 60 years, 34.7 % women) 7,438 measurements of hs-cTnI-Alinity and hs-cTnI-Architect from the same samples were performed. The prevalence of AMI was 17.7 % (279/1,573). Analytical agreement between hs-cTnI-Alinity and hs-cTnI-Architect was excellent (mean bias -3.6 ng/L [95 % CI -17.3 to 10.1]). Using limit-of-detection and upper-reference-limit as benchmarks, agreement was high. However, major disagreement occurred in 2.5 %, including only one AMI. Diagnostic accuracy for AMI was very high and comparable between hs-cTnI-Alinity and hs-cTnI-Architect (area under the receiver-operating-characteristics curve 0.95 [95 % CI 0.95-0.96] and 0.96 [95 % CI 0.95-0.96]). Sensitivity for ruling-out AMI were >99 % and specificity >93 % and comparable between both assays when applying the ESC 0/1h and ESC 0/2h algorithm derived for hs-cTnI-Architect. CONCLUSIONS: Analytic benchmarks and ESC 0/1h and 0/2h algorithm cut-offs originally derived for hs-cTnI-Architect might be safely applied when using hs-cTnI-Alinity.