A-005 In use analytical performance of a high sensitivity cardiac troponin I whole blood analyser when performed at the point of care

Abstract Background Measurement of high sensitivity cardiac troponin (hs-cTn) I using cartridge-based systems at the point of care (POC) requires an appropriate quality control system. The optimal strategy for this currently remains unclear. We report the results of our experience with such a system with measurement over 842 days. Methods The Mersey Acute Coronary syndrome Rule Out Study (MACROS-2). MACROS-2 is a two-centre, pragmatic, randomised controlled trial (RCT) comparing the rule out of non ST-elevation myocardial infarction (NSTEMI), using the European Society of Cardiology (ESC) 0/1 or 0/3 hour diagnostic algorithms (ClinicalTrials.gov Identifier: NCT05322395). The inclusion period was 13/4/22-29/07/2024. Upon presentation with symptoms compatible with NSTEMI, the patient had blood drawn at presentation for central laboratory high sensitive cardiac troponin(hs-cTn) T (roche , elecsys [clinical biomarker in use]), but an additional research sample was drawn for immediate analysis in the ED (emergency department) of whole blood (EDTA tube) Quidel TriageTrueTM hs-cTn I test on the Quidel Triage Meterpro device. This process was repeated for any serial samples. The device includes two on-board quality control (QC) checks on the analytical process. Internal liquid QC was performed using manufacturer supplied EDTA plasma-based QC material. Both alow control (target 25 ng/L) and high control (target 500 ng/L) were use by expert laboratory operators with regular checks on a weekly basis. For plasma, the analytical range is 0.1-1000 ng/L, limit of detection (LoD) is 1.5 ng/L, limit of quantitation (LoQ) at 20% coefficient of variation (CV) 2.1 ng/L and at 10% CV 4.6ng/L. 99th percentile overall 20.5 ng/L. All IQC was downloaded to a central repository for analysis. Results Over 842 days 539triage true POC tests were run on patient samples and 382 QC samples. An invalid test result on first analysis occurred in 56/5391 (1%). On repeat analysis there were 24 analytical failures giving a total failure rate of 24/5291 (0.4%). 328 (165 low, 162 high) QC samples were run. There were 7 lots for the low QC and 9 lots for a high QC over 21 reagent lots. eight expert laboratory operators undertook IQC. For the low control, 4/165 (2.5%) samples exceeded the two standard deviation (SD) limit and none exceeded 3 SD. Mean value was 25.4 ng/L with a mean CV of 9.7%. There was only one analytical failure in the high control with only 2 (1.2% exceeded 2 SD and none 3 SD’s. The mean value was 519 ng/L with a mean CV of 9.5%. Time series regression analysis across reagent batches showed no evidence of significant lot to lot variation or evidence of QC drift across the entire period of the study. Conclusion Analytical reliability of the system was robust. QC stability with time suggests that following lot verification, QC frequency can be such that only the midpoint and towards the end of a particular batch is measured.