Accuracy of an integrated evaluation including clinical presentation, laboratory data and global longitudinal strain by echocardiography for diagnosing acute myocarditis

Abstract

Abstract Funding Acknowledgements Type of funding sources: None. BACKGROUND Cardiac magnetic resonance imaging (MRI) represents the gold standard for the diagnosis of acute myocarditis. However, several institutions are not able to perform such imaging test, especially in the early phase of the disease, with possible missing or delaying in diagnosis. On the other hand, the penetration of the trans-thoracic echocardiographic (TTE) assessment of global longitudinal strain (GLS) by speckle tracking echocardiography (STE) is rapidly growing. An impairment in GLS may occur in the early phase of acute myocarditis, with the degree of such impairment being related to the amount of oedema. Few studies have been published on the role of STE in the diagnosis of acute myocarditis and current evidence on the topic is limited. We present here a multicentre, retrospective study on the diagnostic accuracy of a non-invasive model including GLS for acute myocarditis. PURPOSE Aim of the study was the evaluation of the accuracy of 3-layer left ventricular (LV) GLS data (epicardial, midwall and endocardial) coupled with clinical and laboratory assessment for the diagnosis of acute myocarditis, using cardiac MRI as reference diagnostic tool. METHODS A total of 70 patients with clinical suspect of acute myocarditis were identified. Clinical parameters, results of laboratory tests and data from both TTE and STE on admission were recorded. Cardiac MRI was performed in all patients. GLS assessment was reported as absolute value. A total of 13 patients were excluded because of poor acoustic window, unconfirmed diagnosis or missing data. RESULTS 57 patients with MRI-confirmed acute myocarditis were included in the final analysis (age 38.8 ± 15.6 years, 49 males). Twenty-one patients (37.5%) had fever on admission. Mean white blood cell (WBC) count was 10.92/10^3 ± 1.7 and C-reactive protein levels were 4.9 ± 5.2 mg/dL. At TTE, a mild reduction of LV systolic function was overall observed (LV ejection fraction 50.1% ±11.2), without impairment of the diastolic function (E/A Ratio 1.31 ± 0.55, E/e’ average ratio 7.59 ± 3.4). Mean epicardial GLS of the LV was 14.5 ± 4.3%, midwall GLS was 16.5 ± 4.7% and endocardial GLS 18.6 ± 5%. The best localization agreement between regional strain decrease and late gadolinium enhancement at cardiac MRI was found for an epicardial GLS <18% (43 patients, 77%). The integration of STE data (epicardial GLS <18%), clinical signs (body temperature >37.5°) and laboratory findings (WBC >10/10^3) was able to identify all patients (100%) with MRI-diagnosed acute myocarditis (Figure 1). CONCLUSION STE of the LV, especially for epicardial GLS, presents a high agreement with cardiac MRI to detect acute myocarditis. The integration of GLS assessment, body temperature and WBC on admission is highly sensitive for an early diagnosis of acute myocarditis. This model could be more extensively tested in those settings where cardiac MRI is not immediately available. Abstract Figure 1