Validation of a 0/2-hour high sensitivity cardiac troponin algorithm for suspected acute coronary syndrome in the emergency department

BackgroundWe implemented a high-sensitivity cardiac troponin I (hs-cTnI)-based algorithm for emergency department (ED) evaluation of possible non-ST elevation acute coronary syndromes (NSTE-ACS) within an integrated health system (The Kaiser Permanente Northern California [KPNC] NSTE-ACS algorithm). MethodsRetrospective study of adult (18+ years) ED encounters for chest pain/discomfort with hs-cTnI testing (Access hsTnI, Beckman) at 21 KPNC medical centers between January 1, 2023 and June 30, 2024. Exclusion criteria were ST-elevation myocardial infarction (MI), leaving the ED against medical advice, lack of active KP health plan coverage, or an included encounter within 30 days prior. The primary outcome was 30-day MI or death. Sensitivity, specificity, negative predictive value (NPV), positive predictive value (PPV), and likelihood ratios (LR) were reported, with subgroup analyses by age, sex, coronary artery disease (CAD), chronic kidney disease (CKD), and ED disposition. ResultsThere were 104,025 encounters in the final study cohort. Median age was 59 years, 45% were male, 18% had CAD, and 13% had CKD. The primary outcome occurred in 5.5% of encounters. Rule-out criteria were present in 70% of encounters with a sensitivity of 95.4% (95% CI: 94.8-96.0%), a NPV of 99.7% (95% CI: 99.6-99.7%) and an LR-of 0.05, while 7% of encounters met rule-in criteria with a specificity of 96.7% (95% CI: 96.6-96.8%), a PPV of 60.2% (95% CI: 59.3-61.1%) and an LR+ of 24.4. In subgroup analyses, rule-out criteria NPV was statistically below 99% in stage 4+ CKD (96.1%; 95% CI: 94.6-97.6%) and ischemic CAD (98.6%; 95% CI: 98.3-98.9%), though not among those selected for ED discharge (98.4%; 95% CI: 96.7-99.2% and 99.1%; 95% CI: 98.8-99.4%, respectively). ConclusionsThe KPNC NSTE-ACS evaluation algorithm demonstrated excellent overall performance. NPV was modestly diminished in ischemic CAD or advanced CKD, but this excess risk was largely mitigated by ED discharge disposition decisions.