Minimum Myocardial Injury defined by High-Sensitivity Troponin I detectable on CMR in MINOCA patients.

Background. Myocardial infarction with non-obstructive coronary arteries (MINOCA) represents a clinically relevant and heterogeneous entity, accounting for approximately 6–8% of all myocardial infarctions (1). Because MINOCA is a 'working diagnosis', the identification of the underlying mechanism is crucial for appropriate management and prognosis (2). Together with intracoronary imaging, cardiac magnetic resonance (CMR) is a key tool for differentiating between aetiological causes(2). However, the accessibility, cost, and variable diagnostic yield of CMR highlight the need for reliable biomarkers to guide patient selection and optimize timing. High-sensitivity cardiac troponin I (hs-cTnI) is routinely available and may serve as a gatekeeper to prioritize early CMR in MINOCA. Methods. This retrospective, single-centre study included 391 consecutive patients with MINOCA who underwent CMR between January 2021 and September 2024. Hs-cTnI was measured using Abbott Architect (January 2021-September 2023) and Beckman Coulter Access (October 2023 - September 2025) assays. CMR findings, in terms of MINOCA diagnosis, were classified as 'coronary causes', 'non-coronary causes', 'non-diagnostic findings', or 'normal'. Receiver operating characteristic (ROC) analyses were performed to identify hs-cTnI thresholds predicting diagnostic CMR according to imaging timing (≤14 vs. >14 days from hs-cTni peak). Results. Hs-cTnI concentrations were markedly higher in patients with pathological CMR findings compared with those showing normal or non-diagnostic scans [Abbott: median 3,313 (IQR 676-12561) ng/L for pathological diagnostic findings vs. 219 (IQR 77- 530) ng/L for normal scans; Beckman: median 1808 (IQR 253-5433) ng/L for pathological diagnostic findings vs. 197(IQR 66-510) ng/L for normal scans; p < 0.001 for both. When CMR was performed within 14 days from the hs-cTnI peak, the biomarker demonstrated strong discriminatory ability for identifying clinically relevant findings (AUC = 0.79; 95% CI 0.74– 0.85; p < 0.001). The optimal diagnostic threshold was approximately 500 ng/L across both assays (sensitivity 73%, specificity 74%). In contrast, when imaging was delayed beyond 14 days, the predictive accuracy of hs-cTnI decreased (AUC = 0.69), and higher thresholds (~700 ng/L) were required to maintain specificity. Conclusions. Hs-cTnI is a reliable and easily accessible biomarker that can predict the diagnostic yield of CMR in patients with MINOCA. A threshold around 500 ng/L across both Abbott and Beckman assays identifies patients with a high probability of structural myocardial abnormalities when CMR is performed within 14 days from the hs-cTnI peak. Integrating hs-cTnI concentration with CMR timing provides a simple and clinically actionable strategy to optimize diagnostic accuracy, reduce unnecessary imaging, and harmonize patient management across different laboratory platforms. These results support a biomarker-guided and time-sensitive approach to the diagnostic work-up of MINOCA.