Layer-specific fast strain-encoded cardiovascular magnetic resonance in suspected acute coronary syndrome: a prospective study

Clinical Research in Cardiology https://doi.org/10.1007/s00392-026-02936-0 ORIGINAL PAPER Layer‑specific fast strain‑encoded cardiovascular magnetic resonance in suspected acute coronary syndrome: a prospective study Lukas D. Weberling1,2 · Deborah Siry1,2 · Ailís C. Haney1,2 · Janek Salatzki1,2 · Andreas Ochs1,2,3 · Evangelos Giannitsis1,2 · Norbert Frey1,2 · Florian André1,2 Received: 24 March 2026 / Accepted: 4 May 2026 © The Author(s) 2026 Abstract Background High-sensitivity troponin assays detect myocardial injury early, but only a subset reflects coronary plaque rupture requiring invasive therapy. Cardiovascular magnetic resonance (CMR) is the non-invasive reference standard for tissue characterization and functional assessment. Fast Strain-ENCoded imaging (fSENC) enables rapid strain analysis, and layer-specific strain may help identify ischemic causes. Methods In this prospective, single-centre observational study, patients presenting with chest pain and low-to-intermediate pretest probability for major adverse cardiac events (MACE) underwent rapid CMR prior to further therapy. Based on final clinical diagnosis, patients were classified as Healthy, acute coronary syndrome (ACS), or other cardiac abnormality (OCA). Global longitudinal strain (GLS) and the difference between endocardial and epicardial GLS (GLS-Dif) were quantified using fSENC. Follow-up assessed death and MACE. Results A total of 108 patients (50 female; age 57.5 ± 17.4 years) were included. Total scanning time was 14 ± 3 min. GLS was significantly reduced in ACS (− 17.1%) and OCA (− 15.8%) compared with Healthy patients (− 20.1%; both p < 0.001), predicting any cardiac abnormality with an AUC of 86.5%. GLS-Dif was negative in ACS (− 0.6%) and significantly lower than in Healthy patients (1.5%) and OCA (0.95%), predicting ACS with an AUC of 91.3%. Combining GLS and GLS-Dif yielded excellent discrimination for cardiac abnormality (AUC 93.0%) and ACS (AUC 93.2%; sensitivity 100%; specificity 71.4%) and would have potentially reduced 35.3% of performed invasive coronary angiographies. Over a median followup > 5 years, death and MACE rates were low and unrelated to strain parameters. Conclusions Layer-specific CMR strain analysis using fSENC reliably identifies ischemic myocardial injury in suspected ACS, supporting clinical decision-making and potentially reducing unnecessary invasive procedures. * Lukas D. Weberling 1 Department of Cardiology, Angiology and Pneumology, Heidelberg University Hospital, Heidelberg, Germany 2 DZHK (German Centre for Cardiovascular Research), Heidelberg / Mannheim, Germany 3 Department of Cardiology and Angiology, Frankfurt University Hospital, Frankfurt, Germany Vol.:(0123456789) Clinical Research in Cardiology Graphical Abstract Keywords CMR · Cardiovascular imaging · NSTEMI · FSENC · Strain Abbreviations ACS Acute coronary syndrome AMI Acute myocardial injury AUC Area under the curve CAD Coronary artery disease CMR Cardiovascular magnetic resonance fSENC Fast Strain ENCoded GLS Global longitudinal strain hsTnT High-sensitivity Troponin T NSTEMI Non-ST-elevation myocardial infarction OCA Other cardiac abnormality Background Modern troponin assays have revolutionized emergency medicine through their fast and sensitive identification of myocardial injury. However, only a fraction of acute myocardial injuries (AMI) is caused by coronary plaque rupture and therefore treatable through invasive coronary angiography (ICA) [1]. Therefore, the inherent risk of a potentially unnecessary invasive procedure must be weighed against the possibility of a potentially untreated plaque rupture and further decision support tools are needed. Cardiovascular magnetic resonance (CMR) is the reference standard for myocardial tissue and functional evaluation [2]. So far, its long acquisition time has hindered its usage in emergency medicine [3]. Modern CMR strain sequences like fast Strain-ENCoded (fSENC), however, can be performed in a single heartbeat per slice, making it a suitable candidate for a clinical support tool for decision making in the emergency department. Despite their fast acquisition times, fSENC strain measurements show high reliability and are not derived from images (like so-called feature or speckle tracking) but obtained from the physical properties of the myocardium itself [4]. This necessitates additional dedicated sequences in the CMR protocol, but results in a high predictive power for cardiac outcomes in both healthy individuals and a wide range of ischemic and non-ischemic cardiac diseases that outperforms clinical data and conventional CMR parameters [5, 6]. Also, the high spatial resolution of fSENC (< 2 mm) enables measurements of not only the global longitudinal strain (GLS), but also its components segmental and layer-specific strain. Whilst the substantial predictive value of GLS has been shown in both healthy cohorts and a large variety of patient groups alike, little data exists on the potential of a layer-specific or segmental strain [7–15]. GLS as diagnostic support tool has only rarely been tested [16–18]. A Clinical Research in Cardiology recent study however indicates its usefulness in the differentiation of AMI [4]. When compared to healthy controls, the GLS identified patients with an AMI, but was unable to differentiate its underlying causes [4]. Here, a difference of the myocardial strain between the epicardium and the endocardium, the GLSDif, proved to be able to differentiate patients with a non-ST elevation myocardial infarction (NSTEMI) from other AMIs, since only these patients had a negative GLS-Dif and therefore a diminished endocardial GLS [4]. However, this was only a proof-of-concept study evaluating 64 patients that were already ruled-in by serial high-sensitivity troponin T (hsTnT) measurements [4]. Moreover, patients with previous cardiac events were not excluded and healthy controls were drawn from a proven healthy reference cohort, not from patients in the emergency department with an inconspicuous work-up [4]. The aim of the present study therefore is to assess the diagnostic potential of the fSENC-derived strain measurements GLS and GLS-Dif on patients at a low-to-intermediate risk with suspected coronary syndrome in a real-world setting. Methods Study population and design This single-centre, real-world, prospective, observational CMR study included patients presenting with acute chest pain in our emergency department between September 2018 and August 2019. The study design and results of the global strain measurements have been published before [19]. Inclusion criteria were (I) acute chest pain, (II) a low or intermediate HEART score of ≤ 6 and (III) signed informed consent [19]. Exclusion criteria involved (I) fast rule-in or rule-out based on initial hsTnT at 0 h (< 5 or ≥ 52 pg/nl as per our assay) without the need for a 1 h value, (II) hemodynamically unstable patients, (III) a history of CAD, (...truncated)