Cardiac MR and CTA in suspected non-ST-elevation acute coronary syndrome: a head-to-head comparison in the ESC troponin observe range

Abstract Background Accurately diagnosing patients with acute chest pain and non-diagnostic electrocardiographic findings is challenging, particularly when patients are categorised into the ‘observe’ group using the ESC high sensitivity cardiac troponin T (hs-cTnT) 0/1hr algorithm. These patients are often empirically treated as non-ST-elevation acute coronary syndrome (NSTE-ACS) and referred for invasive coronary angiography (ICA), although the absence of obstructive coronary artery disease (CAD) is common. Early non-invasive imaging using cardiac magnetic resonance (CMR) or cardiac computed tomography angiography (CCTA) may effectively rule out obstructive CAD, uncover alternative diagnoses, and improve referral to ICA. Aim This study aimed to investigate the diagnostic accuracy of early CMR and CCTA in identifying obstructive CAD compared to ICA in patients with suspected NSTE-ACS in the ESC ‘observe’ group. Methods Patients with suspected NSTE-ACS scheduled for ICA and hs-cTnT levels in the ESC ‘observe’ range were prospectively included. Patients underwent CMR and/or CCTA imaging before ICA within 72 hours of presentation. Images were reviewed in consensus by a blinded imaging cardiologist and cardiovascular radiologist. The main outcome was the diagnostic accuracy of CMR and CCTA for detecting obstructive CAD, defined by ICA as a stenosis of ≥70% in any epicardial artery, ≥50% in the left main artery or based on significant functional invasive measurements. Secondary endpoints included the prevalence of obstructive CAD and clinical diagnosis of NSTE-ACS, as well as the diagnostic accuracy of CMR and CCTA in diagnosing NSTE-ACS. Results A total of 51 patients were included. Of these, 50 patients underwent CMR and 38 patients CCTA imaging before ICA. Obstructive CAD was observed on ICA in 45.1%. CMR demonstrated sensitivity of 87.0% (95%CI: 66.4-97.2), specificity of 77.8% (95%CI: 57.7-91.4), positive predictive value of 76.9% (95%CI: 61.8-87.3) and negative predictive value of 87.5% (95%CI 70.5 - 95.4) for detecting obstructive CAD compared to ICA. CCTA showed sensitivity of 100% (95%CI: 80.5-100), specificity of 61.9% (95%CI 38.4-81.9), positive predictive value of 68.0% (95%CI: 55.2-75.6) and negative predictive value of 100% (95%CI: 75.3-100). NSTE-ACS was clinically diagnosed in 31 patients (60.8%). Compared to the final clinical diagnosis of NSTE-ACS, CMR demonstrated a sensitivity of 73.3% (95% CI: 54.1-87.8) and specificity of 80.0% (95%CI: 56.3-94.3), while CCTA showed a sensitivity of 79.2% (95%CI: 57.9-92.9) and specificity of 57.1% (95%CI: 28.9-82.3). Conclusion These findings indicate that CMR and CCTA are effective initial diagnostic modalities for patients presenting with suspected NSTE-ACS and mildly elevated hs-cTnT levels in the ESC ‘observe’ range, avoiding the need for invasive testing in over half of these patients. Routine ICA may consequently no longer be necessary in the early evaluation of this patient population.