Imprecision and Delta Criteria for a New ESC 0/2-Hour Algorithm

One advantage to using the European Society of Cardiology (ESC) 0/2-h algorithm over the 0/1-h algorithm are the larger deltas (i.e., absolute change in concentrations) used, which in part may mitigate misclassification due to imprecision and may improve performance (1, 2). However, not all high-sensitivity cardiac troponin (hs-cTn) assays have 0/2-h algorithms (1, 2). In this regard, the Advantageous Predictors of Acute Coronary Syndrome Evaluation Study (APACE) study investigators have developed a 0/2-h algorithm for the Ortho Clinical Diagnostics hs-cTnI assay with a listed limit of detection of 0.4 ng/L, limit of quantification of 1.2 ng/L, and overall 99th percentile of 11 ng/L (3). Compared to the other published 0/2-h algorithms, a conspicuous difference is the 2-h delta to rule-in criterion, where the delta for the Ortho hs-cTnI assay only increases by 1 ng/l from ≥4 ng/L to ≥5 ng/L, while for the other published hs-cTn assays the delta values are increased by at least 5 ng/L (1–3). However, this difference is not evident for the rule-out delta, which increases from <1 ng/L for the 0/1-h algorithm to <3 ng/L for the 0/2-h algorithm. To assess the impact of imprecision on possible misclassification using the 0/2-h algorithm deltas, long-term QC results over several reagent lots were obtained and analyzed on an in-use instrument that clinically reported Ortho hs-cTnI results. Briefly, data were obtained from 3 levels of Thermo Fisher OMNI QC material that were measured from May 5, 2020, to November 30, 2021, on a VITROS 7600 XT analyzer. For the QC values, 2 different analyses were performed only for level 1 (normal concentration) and for level 2 (abnormal concentration below 40 ng/L). First, data were analyzed for normality (Shapiro–Wilk) with the median (interquartile; IQR) and percentage of misclassified results calculated. Here, results were misclassified if for the normal QC level (level 1) the individual result was ≥2.5 ng/L from the central estimate (i.e., <3 ng/L used for rule-out) and for the abnormal level (level 2) the individual result was ≥4.5 ng/L (i.e., ≥5 ng/L used for rule-in) from the central estimate. A frequency histogram for both levels of QC was also derived. Second, the mean, SD, and CV% were derived separately for the 8 different reagent lots that were used during this time period for both levels of QC. The target imprecision was a SD ≤0.8 ng/L for level 1 and a CV ≤10% for level 2 in agreement with laboratory recommendations. A final analysis was performed using patient samples (n = 40) on 4 different lot-to-lot comparisons (10 samples per lot-to-lot comparison) in 2021 to determine the absolute difference between hs-cTnI results between lots for concentrations <40 ng/L and percent difference for concentrations ≥40 ng/L. Over 19 months, the median (IQR) concentration for level 1 was 4.3 ng/L (3.8 ng/L to 4.8 ng/L) (n = 1229, Shapiro–Wilk test P-value <0.01) and for level 2 was 26.3 ng/L (24.0 ng/L to 29.2 ng/L) (n = 1263, Shapiro–Wilk test P-value <0.01) (Fig. 1). At the low end (level 1) only 2% of the results were ≥2.5 ng/L from the median value as compared to 25% of the results that were ≥4.5 ng/L from the median value in the abnormal range (level 2). Another published estimate for the delta determined using pooled patient plasma samples across 11 different analyzers and 2 reagent lots for rule-in for the Ortho hs-cTnI assay is ≥9 ng/L (4). Applying this criterion only 3% of the results for level 2 QC would be ≥8.5 ng/L from the median value. No individual lot achieved the target precision for both level 1 and level 2 (Fig. 1). For the lot-to-lot comparison 25 samples had hs-cTnI concentrations <40 ng/L with the largest absolute difference being 4 ng/L (37 ng/L vs 33 ng/L), with the largest percent difference being 7.7% (13450 ng/L vs 14480 ng/L) between the lots. Histogram displaying the percent frequency of results for both level 1 and level 2 QC materials obtained with the Ortho hs-cTnI assay via one instrument over 19 months. The imprecision for both levels of QC for each reagent lot (n = 8) is also provided. The achieved Ortho hs-cTnI long-term imprecision using commercial QC does not support the delta values to rule-in using the 0/2-h or the 0/1-h algorithm (i.e., ≥5 ng/L or ≥4 ng/L). Lot-to-lot comparisons using patient samples also indicate misclassification using the 0/1-h algorithm. Using a larger delta (i.e., ≥9 ng/L) may mitigate potential misclassification due to imprecision. Commercial QC does have limitations; however, the Thermo QC material did identify a reagent problem with the Ortho hs-cTnI assay (5). Finally, there are other patient related variables that may further result in misclassification of an elevated hs-cTnI result with the Ortho assay so additional clinical judgment is a necessity (5). All authors confirmed they have contributed to the intellectual content of this paper and have met the following 4 requirements: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; (c) final approval of the published article; and (d) agreement to be accountable for all aspects of the article thus ensuring that questions related to the accuracy or integrity of any part of the article are appropriately investigated and resolved. Authors’ Disclosures or Potential Conflicts of Interest:Upon manuscript submission, all authors completed the author disclosure form. Disclosures and/or potential conflicts of interest: Employment or Leadership: None declared. Consultant or Advisory Role: P.A. Kavsak, Abbott Point of Care, Beckman Coulter, Roche Diagnostics, Quidel, and Siemens Healthcare Diagnostics. Stock Ownership: None declared. Honoraria: P.A. Kavsak, Beckman Coulter, Roche Diagnostics, Siemens Healthcare Diagnostics, and Thermo Fisher Scientific. Research Funding: P.A. Kavsak, grants/reagents from Abbott Laboratories, Beckman Coulter, Ortho Clinical Diagnostics, Randox Laboratories, Roche Diagnostics, and Siemens Healthcare Diagnostics. Expert Testimony: None declared. Patents: McMaster University has filed patents with P.A. Kavsak and A. Worster listed as an inventor in the acute cardiovascular biomarker field, in particular, a patent has been awarded in Europe (EP 3 341 723 B1) on a Method of determining risk of an adverse cardiac event. McMaster University has also filed a patent with P.A. Kavsak listed as an inventor on Quality Control Materials for Cardiac Troponin Testing. Other Remuneration: P.A. Kavsak, support for attending meetings and/or travel from Randox Laboratories.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.