High-sensitivity troponin I magnitude predicts occlusive myocardial infarction and critical coronary lesions: a tool for invasive strategy timing?
M Arantes Prata, L Nakayama Ohe, T Schumann Munhoz, A Labrunie, I Menezes Ferreira, V Costa, K Franchini, F Feres, R StaicoAbstract
Background
In acute coronary syndromes (ACS), high-sensitivity troponin (hs-cTn) elevation prompts an early (<24h) invasive strategy1. However, this binary approach treats all elevations similarly, potentially leading to unnecessary procedures for minor elevations while failing to identify patients with very high hs-cTn who may have an underlying occlusive myocardial infarction (OMI) and require emergent catheterisation2. We hypothesised that hs-cTnI magnitude could refine risk stratification, as previously demonstrated for hs-cTnT 3.
Purpose
To evaluate the diagnostic accuracy of admission hs-cTnI and its early dynamics in predicting OMI and critical coronary lesions in acute myocardial infarction (AMI).
Methods
Single-centre retrospective cohort of 396 consecutive AMI patients who underwent coronary angiography within 7 days (September 2024–August 2025). hs-cTnI was measured using a high-sensitivity troponin I assay (99th percentile URL 11 ng/L). Two primary endpoints were assessed: OMI(TIMI flow 0–1 in culprit artery) and Critical Lesion (CL: TIMI flow ≤ 2, or stenosis ≥90% in epicardial vessel, or ≥70% with plaque instability features). Absolute and relative delta troponin were analysed in a subgroup with serial sampling per ESC 0/1-hour protocol 4 (n=318). ROC curves were constructed; Youden index, ≥90% sensitivity (Se) and ≥90% specificity (Sp) cut-offs were determined.
Results
Mean age 64.1±11.4 years; 69.2% male. OMI occurred in 23.5% and CL in 57.6%. 42.4% of AMI patients had neither CL nor OMI by angiography. Median admission hs-cTnI was significantly higher in patients with OMI (11,040 vs 495 ng/L, p<0.001) and CL (3,006 vs 307 ng/L, p<0.001). Admission hs-cTnI showed moderate accuracy for OMI (AUC 0.736; 95%CI 0.674–0.799) and CL (AUC 0.672; 95%CI 0.620–0.725). Youden-optimal cut-offs were ≥ 2,991 ng/L for OMI and ≥ 1518 ng/L for CL. A high hs-cTnI cutoff for OMI (≥ 35,050 ng/L) had 90.4% Sp and a positive likelihood ratio (LR+) of 3.7. For CL, a cutoff of ≥ 11,790 ng/L yielded 90.5% Sp and a LR+ of 3.56. Absolute hs-cTnI change also predicted OMI (AUC 0.725; 95%CI 0.647–0.802); notably, a change < 28.6 ng/L had 95.8% NPV for ruling out OMI (LR- 0.188). In contrast, relative (%) hs-cTnI change showed no discriminative ability for either endpoint (AUC 0.442 and 0.469).
Conclusion
The magnitude of hs-cTnI elevation and its early absolute change provide moderate, yet clinically valuable, predictive accuracy for acute coronary occlusion and critical coronary anatomy. Very high hs-cTnI levels can identify patients with a high probability of OMI, supporting an emergent invasive approach. Conversely, a minimal early rise in hs-cTnI could contribute to ruling out OMI, potentially allowing for safe de-escalation of care. These findings suggest that hs-cTnI quantification, beyond its current binary interpretation, may serve as an additional tool to refine invasive strategy timing in AMI.Central Figure: ROC curves and key findingsDiagnostic Performance of hs-cTnI