Longitudinal, radial, circumferential and rotational mechanics in acute myocarditis; a 2D speckle-tracking echocardiography study

Abstract

Abstract Objectives The aim of our study was to assess the utility of echocardiographic deformation imaging of the left ventricle in patients with CMR-proven acute myocarditis (AM). Background Often conventional echocardiography reveals no obvious abnormalities in AM. While longitudinal strain is known to reflect the subendocardial muscle layer deformation, in AM myocardial damage is predominantly confined to the subepicardium. The latter is involved in circumferential deformation and rotational mechanics. Methods Data were obtained from a prospectively maintained database of patients (n=44) with AM who had an echocardiography and CMR within 2 weeks of hospital presentation. All patients underwent a comprehensive echocardiographic study including measurement of left ventricular (LV) global longitudinal strain (GLS), basal and apical radial strain (RS), basal and apical circumferential strain (CS), basal and apical rotation, peak twist and peak twist rate. Furthermore, LV deformation was analysed in an age-matched control group (CG) of individuals (n=27) with no evidence of cardiovascular disease. Results There was no significant difference in demographics and most echocardiographic parameters (Table 1). LV EF (57±11% vs 63±4%, p=0.005) was lower in AM. Difference in wall thickness (LVPWd 10.6±1.9mm vs 9.5±1.5mm, p=0.002) and diastolic function (E/E' lateral 6.7±3.1 vs 4.9±0.9, p=0.015) were significant but not specific. Deformation imaging revealed significant differences between the 2 groups in parameters reflecting longitudinal and circumferential deformation, and rotational mechanics (Table 2); GLS (−15.4±4.2% vs −18.9±2.1%, p<0.001), apical CS (−26.1±10.2% vs −34.8±9.4%, p<0.001), basal CS (−20.8±7.4% vs −24.4±6.0%, p=0.017), apical rotation (4.5±3.2° vs 7.9±4.0°, p<0.001), peak twist (9.8±5.8° vs 14.5±5.1°, p<0.001) and peak twist rate (69.5±32.7deg/sec vs 89.3±32.7deg/sec, p<0.001). In patients with preserved systolic function (EF ≥50%, n=35) the only statistically significant 2D parameter between groups was LVPWd thickness (10.8±1.7mm vs 9.2±1.5mm, p=0.001), whereas differences in most strain and rotational parameters persisted; GLS (−16.5±3.3% vs −18.9±2.2%, p=0.003), apical CS (−28.0±8.2% vs −34.8±9.4%, p=0.030), apical rotation (4.7±3.2° vs 7.9±4.0°, p<0.001), peak twist (10.3±6.2° vs 14.5±5.5°, p=0.008), peak twist rate (71.1±33.3 deg/sec vs 89.3±32.7 deg/sec, p=0.048). ROC analysis showed fair classification accuracy for apical rotation (0.760, p<0.001), LV PWd (AUC 0.760, p<0.001), peak twist (0.752, p=0.01) and GLS (AUC 0.706, p<0.001). Conclusion Apical rotation and myocardial twist can objectively quantify myocardial mechanics in AM. Circumferential and rotational mechanics are mostly affected, even in patients with preserved ejection fraction, possibly reflecting subclinical LV dysfunction associated with subepicardial muscle layer inflammation. Funding Acknowledgement Type of funding sources: None.