HEART Score Recalibration Using Higher Sensitivity Troponin T

We examined the diagnostic performance of a recalibrated History, Electrocardiogram, Age, Risk factors, Troponin (HEART), and Thrombolysis in Myocardial Infarction (TIMI) score in patients with suspected acute cardiac syndrome (ACS). Recalibration of troponin thresholds was performed, including shifting from the 99th percentile to the limit of detection (LOD) or to the limit of quantification (LOQ) We compared the discharge potential and safety of the recalibrated composite scores using a single presentation high-sensitivity cardiac troponin (hs-cTn) T to the conventional scores and with a LOD/LOQ troponin strategy alone.

We undertook a 2-center prospective cohort study in the United Kingdom (UK) (2018) (Clinicaltrials.gov NCT03619733) to specifically assess recalibrated risk scores (shifting the troponin subset scoring from 99th percentile to LOD [UK]) and combined the results of this with secondary analyses of 2 prospective cohort studies in the UK (2011) and the United States (2018, using LOQ rather than LOD). The primary outcome was major adverse cardiovascular events (MACE), defined as adjudicated type 1 myocardial infarction (MI), urgent coronary revascularization, and all-cause death, at 30 days. We evaluated the original scores using hs-cTn below the 99th percentile and recalibrated scores using hs-cTn

We studied 3,752 patients (3,003 in the UK and 749 in the United States). Median age was 58 years, and 48% were female. At 30 days, 330/3,752 (8.8%) experienced MACE. The sensitivities of the original HEART less or equal to 3 and recalibrated HEART less or equal to 3 scores for rule-out were 96.1% (95% confidence interval [CI], 93.4 to 97.9) and 98.6% (95% CI, 96.5 to 99.5) respectively; the original TIMI less or equal to 1 and recalibrated TIMI less or equal to 1 scores' sensitivities were 79.7% (95% CI, 74.9 to 83.9) and 96.1% (95% CI, 93.4 to 97.9) respectively; and nonischemic ECG with hs-cTn T below the 99th percentile and hs-cTn T less than LOD/LOQ was 79.7% (95%CI, 0.749 to 0.839) and 99.1% (95% CI, 0.974 to 0.998), respectively. Recalibrated HEART less or equal to 3 was projected to discharge 14% more patients than hs-cTn T less than LOD/LOQ. The improved sensitivity of rule-out for recalibrated HEART less than or equal to 3 came at the cost of reduced specificity (50.8% versus 53.8% for recalibrated HEART and conventional HEART respectively).

This study indicates that recalibrated HEART score of less or equal to 3 is a feasible and safe early discharge strategy using a single presentation hs-cTnT. This finding should be further tested using competitor hs-cTn assays in independent prospective cohorts before implementation.