Demographic and baseline clinical characteristics of participants
We first compared the demographic and baseline clinical characteristics of participants in the control group (n=35; men: 19; women: 16), the NLVH group (n=35; men: 18, LVMI ≤ 115 g/m2; women: 17, LVMI ≤ 95 g/m2), and the LVH group (n=33; men: 19, LVMI > 115 g/m2; women: 14, LVMI > 95 g/m2). As shown in Table 1, no significant differences were observed with regard to age, gender, heart rate, and body mass index (BMI) between the groups (all P > 0.05). Further, there were no significant differences in remove volume, dry weight, cause of disease, and medication history between these three groups. As expected, the NLVH and LVH groups had significantly higher systolic blood pressure (SBP), diastolic blood pressure (DBP), and creatinine levels than the control group (P < 0.05), but no significant differences in these parameters were observed between the NLVH and LVH groups (P > 0.05). The N-terminal pro-brain natriuretic peptide (NT-proBNP) values in the NLVH group and LVH group were significantly higher than those in the control group, and the NT-proBNP levels in the LVH group were significantly higher than those in the NLVH group. In addition, the LVH group had a significantly longer duration of dialysis than the NLVH group (P < 0.05).
Conventional echocardiographic parameters for participants
We next compared the conventional echocardiographic parameters between these three groups. Compared with the control group, the NLVH and LVH groups had significantly higher LVIDD, LVIDS, LAD, LVEDV, and LVESV (P < 0.05), but there were no significant differences in these parameters between the NLVH and LVH groups (P > 0.05). In addition, the LVH group had a significantly lower LVEF than the control group (P < 0.05). The LVH group had significantly worse E/A and E/e' than the control and NLVH groups, while the NLVH group had significantly worse E/A and E/e' than the control group, indicating that MHD patients, especially those in the LVH group, had diastolic dysfunction (Table 2).
Comparison of MW and strain parameters of participants
The NLVH and LVH groups had lower GWE and GLS but higher GWW than the control group (P < 0.05). The LVH group had lower GWE and GLS but higher GWW and PSD than the NLVH group (P < 0.05). In addition, the LVH group had significantly lower GWI and GCW than both the control and NLVH groups (P < 0.01), and had higher PSD than the control group. However, the control and NLVH groups had comparable GWI, GCW, and PSD (P > 0.05) (Table 3).
The control group had the largest PSL area and the smallest global average wasted work. The color of the bull’s eye in the control group was uniformly green, suggesting high work efficiency. The PSL area and global average wasted work for the NLVH group were between those for the control and LVH groups. The bull's eye was unevenly light yellow, indicating a slightly lower work efficiency. The color of the bull's eye in the LVH group was messy, with yellow and orange colors, indicating low work efficiency (Figure 1).
Definition of abnormal left ventricular function in LVH group
According to the diagnostic criteria for heart failure with preserved ejection fraction (HFpEF) proposed by the European Society of Cardiology in 2019 [14], the LVEF in the LVH group was normal or slightly reduced. However, these patients had cardiac function classes of II-IV according to the New York Heart Association (NYHA) cardiac function classification, and NT-proBNP > 125 pg/ml, GLS < 16%, LVH and/or left atrium enlargement, and diastolic dysfunction. Therefore, the patients in the LVH group had abnormal LV function.
Correlation between MW parameters and LVEF in MHD with LVH patients
We performed correlation analyses between the MW parameters and LVEF in the LVH group; the latter is a widely used indicator of cardiac function [15]. We found that LVEF was positively correlated with GWI, GWE, and GCW (r = 0.643, 0.523, 0.505; all P < 0.05), but negatively correlated with GWW (r =-0.506, P =0.003) (Figure 2).
Determination of the cutoff values, sensitivity, and specificity of MW parameters and
GLS for identifying LV function in MHD with LVH patients
We next used AUC curves to determine the sensitivity and specificity of MW parameters and GLS for identifying LV function in MHD with LVH patients. GWE, GWW, and GLS had AUCs of 0.982, 0.972, and 0.971, respectively (Figure 3a, b, c), with respective cutoff values of 94%, 123 mmHg%, and -18.44%. GWE had higher sensitivity than GLS and GWW (97.0% vs. 93.0% and 84.9%). In addition, GWW has higher specificity than GLS and GWE (97.1% vs. 94.3% and 91.4%). The AUCs for GWI (Figure 3d) and GCW (Figure 3e) were 0.835 and 0.782, respectively, both of which were lower than the AUCs for GWE, GWW, and GLS.
Intra- and inter-observer variability
We also examined the intra- and inter-observer variability in this study. The intra-observer differences in the ICC values of GLS, PSD, LVEF, GWI, GCW, GWW, and GWE were 0.94, 0.93, 0.84, 0.86, 0.88, 0.96, and 0.88, respectively. The differences in the ICC values for these parameters between the observers were 0.93, 0.82, 0.79, 0.91, 0.89, 0.93, and 0.82, respectively. The ICC value and 95% confidence interval for each parameter are shown in Table 4. Our results suggested that our study generated reliable and consistent observations.