Abstract Background The development of high-sensitivity cardiac troponin (hs-cTn) assays has enabled precise measurements below the 99th percentile upper reference limit, allowing serial measurements to detect subtle elevations of cTn at low concentrations. Therefore, analytical performance of the hs-cTn assay and the definition of “change” in cTn concentration in serial measurements are important factors for patient classification when using accelerated diagnostic pathways for patients with chest pain in the emergency setting. For the Abbott ARCHITECT hs-cTnI assay, the European Society of Cardiology (ESC) 0h/1h rapid ‘rule-out’ algorithm defines 4 ng/L (limit of quantification (LoQ); 3.5 ng/L for Abbott ARCHITECT hs-cTnI assay in the United States) to rule-out at presentation and a delta of 2 ng/L for values slightly exceeding the 4 ng/L cutoff on serial measurements; emphasizing the need for robust analytical performance at these low concentrations. Unfortunately, in the United States, laboratories are not able to measure concentrations below the LoQ preventing assessment of analytical performance. In Canada, we can measure the Abbott ARCHITECT hs-cTnI assay at concentrations below the LoQ and accordingly we sought to assess imprecision using commercial quality control material below the LoQ ( 4 ng/L) across seven different hospital sites. Methods The Thermo Scientific MAS Omni-CARDIO™ QC Ultra Low material (package insert value 10 ng/L for the Abbott ARCHITECT hs-cTnI assay) was tested daily on seven different Abbott ARCHITECT analyzers (i1000 and i2000), generating 8685 measurements from 2022-2025 from different hospital sites: Hamilton General Hospital (i1000), St. Joseph’s Healthcare Hamilton (i1000), Milton District Hospital (i1000), Juravinski Hospital and Cancer Center (i2000), Georgetown Hospital (i1000), Oakville Trafalgar Memorial Hospital (i2000), and West Lincoln Memorial Hospital (i1000). Analytical performance at this level was evaluated against a standard deviation (SD) =1 ng/L as a precision goal, based on the recently published allowable performance limit (APL) of ±3 ng/L below 10 ng/L for hs-cTn assays (Clin Chem. 2025 Feb 3;71(2):332-334) Results The overall average concentration of the QC material was 3.1 ng/L with a SD of 0.9 ng/L, meeting the precision goal of SD =1 ng/L. Only 0.9% of QC results fell outside the APL (i.e. 0.1 and 6.1 ng/L), further supporting the validity of the proposed APL. Next, we further evaluated the impact of analytical performance at this level on the ESC ‘rule-out’ algorithm (i.e. 4 ng/L for the initial hs-cTnI concentration). Assuming each QC result corresponds to an initial hs-cTnI test result (in whole number), 73.8% of all QC measurements would be 4 ng/L. Conclusion This is the first study to evaluate daily QC performance below the LoQ of a hs-cTn assay. Daily QC measurements (mean = 3.1 ng/L for the Abbott ARCHITECT hs-cTnI assay) can be achieved with acceptable analytical precision (SD =1 ng/L) in small and large community and academic hospital settings. Despite robust analytical performance, nearly a quarter of results exceed 4 ng/L, potentially contributing to patient misclassification using a single low hs-cTnI cutoff as per the ESC pathway.
Choi et al. (Wed,) studied this question.