Phenotyping of myocardial involvement by cardiac magnetic resonance in idiopathic inflammatory myopathies

• Published in: European radiology 2021-07, Vol.31 (7), p.5077-5086
• Main Authors: Sun, Jianhong, Yin, Geng, Xu, Yuanwei, Wu, Yang, Yu, Liuyu, Li, Weihao, Wan, Ke, Sun, Jiayu, Zhang, Qing, Xie, Qibing, Chen, Yucheng
• Format: Article
• Language: English
• Subjects: Antibodies; Body mass; Body mass index; Body size; Cardiac; Cardiovascular disease; Damage; Dermatomyositis; Diabetes; Diagnostic Radiology; Dyspnea; Ejection fraction; Gadolinium; Heart; Hospitals; Hypertension; Imaging; Immunology; Inflammation; Inflammatory diseases; Internal Medicine; Interventional Radiology; Kinases; Laboratories; Magnetic resonance; Medicine; Medicine & Public Health; Multivariate analysis; Musculoskeletal diseases; Myositis; Neuroradiology; Patients; Peptides; Phenotyping; Polymyositis; Radiology; Resonance; Rheumatology; Risk analysis; Risk factors; Subgroups; Ultrasound; Ventricle
• ISSN: 0938-7994; 1432-1084
• DOI: 10.1007/s00330-020-07448-7

Objectives Cardiac dysfunction is commonly noted in patients with idiopathic inflammatory myopathies (IIMs). This study aimed to investigate the characteristics of cardiac dysfunction using cardiac magnetic resonance (CMR) in polymyositis (PM), dermatomyositis (DM) and necrotising myositis (NM). Methods Fifty-one patients with IIMs and 20 matched healthy controls (HCs) were assessed using CMR examination. The clinical data, cardiac serum markers and autoimmune antibodies were determined for all patients. Cardiac involvement was identified by myocardial native T1, extracellular volume (ECV), late gadolinium enhancement (LGE) and left ventricular ejection fraction (LVEF). Results Different subtypes of IIMs showed different patterns of LGE and varying degrees of myocardial damage. The PM subgroup showed higher native T1 ( p  = 0.010) and ECV ( p  = 0.000) than the HCs. The prevalence of LGE was comparable between the PM and DM subgroups (40.0% vs. 31.6%, p  = 0.741); however, it was higher in the PM subgroup than in the NM subgroup (40% vs. 0.0%, p  = 0.014). Patients with positive LGE in the PM subgroup showed a higher proportion of positive LGE ( p  = 0.018) and lower LVEF ( p  = 0.024) than those with positive LGE in the DM subgroup. In multivariate analysis, the presence of LGE could be predicted by increased NT-proBNP ( p  = 0.036, OR = 1.001) and anti-MDA-5 antibody positivity ( p  = 0.011, OR = 12.4). The risk factors associated with native T1 were NT-proBNP ( p  = 0.016, β  = 0.353) and body mass index (BMI) ( p  = 0.024, β  = − 0.331). Conclusions Distinct cardiac involvements in different subtypes of IIMs were identified using CMR. Elevated NT-proBNP and a low BMI were the risk factors associated with LGE and elevated native T1. Key Points • The characteristics of cardiac involvement in different subtypes of IIMs could be identified with cardiac magnetic resonance. • The NT-proBNP levels could reflect focal and diffuse myocardial damage in patients with IIMs.