To assess the change of regional and global left ventricular (LV) myocardial systolic function in patients with Kawasaki disease (KD) in different phases by two-dimensional speckle tracking imaging (2DSTI) and analyze the correlation of systolic peak strain (S) and strain rate (SR) of LV with other conventional indices.

50 KD children in acute phase were enrolled as KD group, while 30 age and gender-matched children who had no cardiac abnormalities on echocardiography were included as control group from January 2016 to December 2016 in children hospital of Lanzhou university second hospital. The KD group were divided into coronary artery dilation (CAD) subgroup (n=12) and coronary artery normal (CAN) subgroup (n=38). Conventional echocardiography parameters, peak systolic LV myocardial regional and global S and SR were obtained using 2DSTI before intravenous immunoglobulin (IVIG) in the acute phase, 1 week after IVIG in the subacute phase, and 8 weeks after the onset of KD in the convalescent phase. The laboratory parameters including C-reactive protein (CRP), erythrocyte sedimentation rate (ESR), white blood cell (WBC), platelet (PLT), alanine transaminase (ALT) and aspartatetransaminase (AST) were also acquired from KD patients in acute phase. The Comparisons on above parameters were made between controls and the KD group in different phases, and between CAD subgroup and CAN subgroup as well. The correlation of LV systolic global longitudinal strain (GLS) with conventional echocardiography parameters and laboratory parameters was analyzed.

Compared with controls, The KD group had higher left ventricular mass index (LVMI), left coronary artery (LCA) diameter, right coronary artery (RCA) diameter and lower GLS, basal longitudinal S, middle longitudinal S, peak systolic global circumferential strain (GCS) and basal circumferential S (t=4.823, 4.123, 3.018, 3.982, 5.135, 4.753, 3.012, 4.6351 respectively, P<0.01 or 0.05) in acute phase. And LVMI, LCA diameter and lower GLS, middle longitudinal S remained at higher level in subacute phase (t=4.786, 4.387, 4.895, 4.031 respectively, P<0.01). But all indices recovered to normal level in convalescent phase. The LVMI in KD group decreased significantly and The GLS, basal longitudinal S, GCS and basal circumferential S increased remarkably in subacute phase compared with acute phase (t=3.99, 2.976, 4.5362, 4.428, 5.327, respectively, P<0.01 or 0.05). Compared with CAN subgroup, KD patients with CAD had higher ESR, CRP, ALT, and AST (t=3.127, P=0.003; t=928, P=0.006; t=3.201, P=0.003; t=3.174, P=0.004, respectively). The GLS was associated with LVMI and CRP negatively (r=-0.795, P=0.000; r=-0.67, P=0.041, respectively) in acute phase.

The LV longitudinal and circumferential systolic S significantly decreased during the acute phase of KD, improved immediately at subacute phase and continued to improve at the convalescent phase. 2DSTI was a simple and accurate approach in evaluating LV myocardium function during different phases of KD. Despite normal LV systolic function by routine echocardiographic measurements in acute phase, KD patients already had reduced LV systolic S which may be a more sensitive indicator of myocardial inflammation and may provide supportive criteria to avoid delayed diagnosis of KD. The CAD may not be a risk factor for LV dysfunction in acute phase. This new technology may provide clinician with valuable information on assistive diagnosis, guidance for drug use, long-term follow-up and reduction in the incidence of coronary artery lesions.