A-028 Real-world comparative clinical performance of high-sensitivity cardiac troponin T Gen5 and Gen6 in routine acute care testing

Abstract

Small analytical differences between generations of high-sensitivity cardiac troponin T (hs-cTnT) assays can shift whether and when results cross clinical decision thresholds, potentially influencing downstream care including diagnoses, serial testing needs, and time to disposition. As health systems consider transitioning from hs-cTnT Gen5 to Gen6, paired real-world comparisons are needed to quantify how analytical differences impact clinical behavior across acute care settings. We performed a retrospective, encounter-level analysis of patients undergoing hs-cTnT testing as part of routine care, predominantly in the Emergency Department (83.1%; 1443/1737 patients). Paired Gen5 and Gen6 results from the same specimen draw were evaluated longitudinally within encounters. We compared time to first elevation above sex-specific 99th percentile upper reference limits (URLs) between assays. To estimate clinical impact, we applied a standardized 0/1/3-hour rule-in/rule-out pathway to each assay’s results and assessed agreement in classification and implications for timing of key decisions (baseline classification vs serial testing; time to rule-out). Among 1847 encounters, Gen6 crossed the sex-specific 99th percentile URL earlier than Gen5 substantially more often than the reverse among encounters with any URL elevation (11.6% vs 2.8%). When both assays exceeded the URL, Gen6 crossed earlier or at the same time in 98.9% of encounters; Gen6 preceded Gen5 in 2.6%, whereas Gen5 preceded Gen6 in 0.1%. Timing differences clustered near clinical decision limits, where small absolute concentration differences resulted in earlier threshold crossing by Gen6 while paired Gen5 values remained just below the URL. Targeted adjudication of encounters with the greatest discordance supported enhanced low-range analytical sensitivity as the primary driver of earlier Gen6 elevation; apparent earlier Gen5 elevation was uncommon and more consistent with isolated outlier behavior (“fliers”) when interpreted alongside clinical context and concomitant laboratory results. Applying the 0/1/3-hour pathway, Gen5 and Gen6 demonstrated high overall agreement in categorical classification (∼85%), with discordance concentrated in borderline cases near decision thresholds. Median time to abnormal classification was 0 hours for both assay generations, reflecting frequent baseline results already above the URL at presentation. Among encounters requiring serial testing, Gen6 enabled modestly earlier rule-out than Gen5 (median 1.03 hours; 95% CI 1.00-1.30 vs. 1.08 hours; 95% CI 1.02-3.00), indicating a small but measurable shift toward earlier rule-out assignment in a subset of patients. In routine real-world acute care testing, hs-cTnT Gen6 crosses the sex-specific 99th percentile URL earlier than Gen5 in a substantial minority of encounters and only rarely later, consistent with improved low-concentration analytical performance. While overall pathway classification is largely preserved, Gen6 yields modestly earlier rule-out among encounters requiring serial testing, an effect that may translate into incremental improvements in ED throughput and time-to-disposition for borderline cases. These findings support evidence-based assay transitions and underscore that generation changes can alter threshold behavior beyond numeric harmonization alone.