Does Maximum Troponin T Value Correlate With Acute Coronary Occlusion Among Patients Treated as NSTEMI?

Abstract

Background: Occlusion myocardial infarction (OMI) is increasingly recognized among NSTEMI patients, yet current diagnostic paradigms may fail to detect coronary occlusion early. In clinical situations, the diagnosis of OMI requires a super electrocardiogram (ECG) expert or use of artificial intelligence like the Queen of hearts. For research purposes, we try to use high-sensitivity cardiac troponin T (hs-cTnT) to help retrospectively classify ECGs as indicating OMI or not among patients treated as NSTEMI.

Aims: To evaluate the association between hs-cTnT and angiographic occlusion in patients treated as NSTEMI. The results of this study could help to guide future studies of ECG findings of acute occlusion, with the goal of identifying acute occlusion before troponin results return.

Methods: This retrospective, single-center study analyzed data from patients treated as NSTEMI who underwent coronary angiography at Beauvais Hospital, France, between January to December 2024. Among these patients, we initially selected all patients with TIMI flow 0-2, constituting the « OMI » group; subsequently, we included consecutively those patients admitted within the first 3 months with TIMI flow 3, constituting the « NOMI » comparison group. ROC curve analysis was performed to identify hs-cTnT thresholds, and predefined cutoffs (1000 ng/L and 10,000 ng/L) were evaluated. The value of hs-cTnT used was the highest prior to coronary angiography. The setting of hs-cTnT to define coronary occlusion was inspired by previous works in the literature done for research purposes.

Results: 200 patients were included (31 OMI, 169 NOMI) among patients treated as NSTEMI in 2024. In multivariable analysis, persistent chest pain (OR 2.6; 95% CI: 1.06-6.40; p = 0.037) and de novo wall motion abnormality (OR 3.41; 95% CI: 1.30-8.89; p = 0.012) were independently associated with coronary occlusion. ROC analysis identified 678 ng/L as the optimal threshold of hs-cTnT (AUC = 0.58) with 71% sensitivity (Se) and 65% specificity (Sp). The negative predictive value (NVP) and positive predictive value (PPV) were respectively 91% and 22% in the patient population studied.

Conclusion: Coronary occlusion was found in 15% of patients treated as NSTEMI in our cohort. Subject to limitations such as the non-use of peak troponin levels and the exclusion of TIMI-3 flow from the OMI group, which led to an underestimation of prevalence, the hs-cTnT threshold of 678 ng/L was associated with coronary occlusion in patients treated as NSTEMI, with modest sensitivity (Se), good NPV, modest Sp, and poor PPV. As troponin testing is non-specific, the troponin threshold should not be used independently to classify patients treated as NSTEMI into OMI or NOMI. An early ECG interpreted using artificial intelligence is currently the most reliable method for diagnosing OMI.