What does this research mean for the field?

Only 56.2% of patients were classified consistently among hs-cTnT and hs-cTnI assays, indicating significant discordance affecting myocardial infarction diagnosis decisions. Novelty: ClaimNovelty.NOVEL_FINDING. Consensus alignment: ConsensusAlignment.CHALLENGES_CONSENSUS.

What question did this study set out to answer?

To evaluate the agreement of single-sample classification of cardiac troponin assays in diagnosing acute coronary syndromes per ESC guidelines.

February 8, 2026

To agree, or to disagree, that is the question (when applying European Society of Cardiology single-sample diagnostic criteria to four high-sensitivity cardiac troponin assays)

Key Result

Only 56.2% of patients had concordant 0h/1h ESC guideline classification across four hs-cTn assays, showing poor agreement affecting ACS diagnosis.

Key Points

To evaluate the agreement of single-sample classification of cardiac troponin assays in diagnosing acute coronary syndromes per ESC guidelines.
Analyzed data from hs-cTnT and hs-cTnI assays across four manufacturers.
Classified baseline results as rule-out, rule-in, or requiring more samples according to 2023 ESC guidelines.
Calculated weighted Cohen’s Kappa values for classification agreement.
Only 56.2% of individuals received the same classification across all assays.
Roche Elecsys and Siemens Atellica ruled out fewer individuals than Abbott Alinity and Quidel TriageTrue methods.
Method-related differences led to poor concordance in MI diagnosis classifications.

Structured PICO

Does the choice of high-sensitivity cardiac troponin assay affect single-sample diagnostic classification for acute coronary syndrome according to ESC guidelines?

Population

3251 individuals in the MACROS2 patient cohort evaluated for acute coronary syndrome, 53.1% male, 46.9% female, with an observed MI incidence of 6.3% (n=206).

Intervention

Application of 2023 ESC guideline single-sample diagnostic criteria (0h/1h algorithm) using four high-sensitivity cardiac troponin assays (Abbott Alinity hs-cTnI, Quidel TriageTrue POC hs-cTnI, Roche Elecsys hs-cTnT, Siemens Atellica hs-cTnI).

Comparator

Pairwise comparison against Abbott Alinity hs-cTnI as the reference standard.

Outcome

Concordance of single-sample classification (rule-out, rule-in, or additional sample required) across all four assays.surrogate

There is poor agreement between different high-sensitivity cardiac troponin assays when applying ESC single-sample diagnostic criteria, which can substantially affect patient triage.

Main Result

Absolute Event Rate: 0% vs 0%

Abstract

Abstract Background Analysis of high-sensitivity cardiac troponin T and I subunits (hs-cTnT and hs-cTnI, respectively) is integral to the 4th Universal Definition of Myocardial Infarction (MI). However, fundamental differences between the cTnT and cTnI subunits, as well as lack of analytical standardisation of hs-cTnI assays, complicates clinical decision making. Purpose To investigate the concordance between single-sample classification according to the 2023 European Society of Cardiology (ESC) guideline for the management of acute coronary syndromes (ACS) between hs-cTnT and hs-cTnI assays from four manufacturers (three central laboratory and one point-of-care) within the Mersey Acute Coronary syndrome Rule Out Study (MACROS2) patient cohort. Methods Inclusion criteria were concurrent data from Abbott Alinity hs-cTnI, Quidel TriageTrue point-of-care hs-cTnI (in whole blood and plasma matrices), Roche Elecsys hs-cTnT and Siemens Atellica hs-cTnI. Baseline results were classified as single-sample rule-out (SSRO), single-sample rule-in (SSRI) or additional sample required (ASR) according to assay-specific rule-in and rule-out thresholds from the 2023 ESC guideline for the management of ACS (0h/1h algorithm). Weighted Cohen’s Kappa values (R Studio) were used for pairwise comparison of classification by each assay against the Abbott Alinity hs-cTnI (the gold-standard in MACROS2, clinicaltrials.gov: NCT05322395). Kappa values of 0.41-0.60 were deemed moderate and 0.61-0.80 substantial. Concordance of classification across all assays was also considered. Results The population investigated comprised 3251 individuals (53.1% male, 46.9% female) with an observed MI incidence of 6.3% (n=206). Table 1 summarises agreement in SSRO, SSRI and ASR classification for each assay. Both Roche Elecsys and Siemens Atellica methods ruled-out fewer patients on a single sample with a greater proportion requiring an additional sample to make a 2023 ESC ACS guideline diagnosis than Abbott Alinity and Quidel TriageTrue POC methods. Notably, only 56.2% of individuals (n=1828) were allocated to the same SSRO, SSRI or ASR category by all assays. Conclusions Cardiac subunit differences between hs-cTnT and hs-cTnI, as well as method-related differences in hs-cTnI result in poor agreement between assays and considerably influences classification to ESC SSRO, SSRI or ASR groups on the baseline sample. The choice of hs-cTn can substantially affect the diagnostic journey for a patient.

Bookmark