Although MHD is the primary treatment for ESRD patients who are not candidates for kidney replacement, the mortality rate of MHD patients remains high and cardiovascular complications, such as heart failure and coronary heart disease, are the main factors associated with the high mortality of MHD patients (18). Therefore, early detection of cardiac dysfunction in MHD patients has important clinical applications. In the present study, we used LST to evaluate the LS and CS of different LV myocardial layers of control and MHD patients and found that, despite the comparable LVEF between these two groups, MHD patients had altered LS and CS at different myocardial layers, higher TTP of 17 LV segments, and higher PSD compared to control subjects. Our findings suggest that LST holds value for early identification of LV dysfunction of MHD patients.
A previous study showed that hemodialysis did not significantly improve LV remodeling and systolic dysfunction in patients (19). Consistent with that report, we first employed traditional echocardiography to assess the cardiac function of control subjects and MHD patients. We found that controls and MHD patients had a relatively comparable LVEF; however, MHD patients had significantly increased cardiac functional indexes, including LAD, LVIDD, LVIDS, IVSD, LVPWD, and LVMI. These findings suggest that ESRD patients exhibit LV remodeling with potential LV systolic dysfunction even after MHD treatment. Indeed, a previous study showed that ESRD patients on MHD had extensive pre- and post-loading factors such as sodium and water retention, anemia, malnutrition, valvular insufficiency, and hypertension (20). The presence of a long-term arteriovenous fistula increased the pre-cardiac load in MHD patients (21). To overcome increases in pre- and post-load, the LV develops physiological cardiac hypertrophy at an early stage to maintain normal cardiac output in MHD patients (22–23). In line with the above observations, we found that MHD patients had significantly lower E/A but higher E/e’ compared to healthy subjects, indicating that MHD patients exhibited slightly abnormal early stage LV diastolic function, which was likely associated with LV hypertrophy (24–25), myocardial fibrosis (14), and/or left atrial dysfunction (26).
Early studies have shown that the overall LS obtained by 2D-STE has become an objective and sensitive indicator for quantitative analysis of small changes in LV systolic function (27), which can be used to detect changes in regional myocardial blood supply earlier than LVEF (28). LST based on 2D-STE technology is a new modality for evaluating wall motion. Animal experiments showed that LST can better assess the degree of myocardial infarction-induced damage at an early stage (29). Previous studies showed that the mid- and epi- myocardia of LV are sensitive to post-load changes (30–32), and that the endocardial myocardium is sensitive to changes in volumetric load (30,33), indicating that different LV myocardial layers exhibit different responses to myocardial insults. Thus, measuring layer-specific strain could be helpful to accurately assess the subtle changes in LV function during the progression of cardiac diseases. The present study used the layer-specific strain by 2D speckle tracking to appraise cardiac strain parameters of MHD patients, and revealed that MHD patients had significantly decreased myocardial LS of all layers compared to the control group. The impaired myocardial LS of all layers could be attributed to endothelial dysfunction and vascular injury (15), and to the observation that ESRD patients receiving MHD commonly present with anemia, secondary hyperparathyroidism, and arteriovenous fistula, all of which increase volume load and result in LV remodeling, hypertrophy, and fibrosis (34). These pathological changes aggravate the progression of atherosclerosis and further heighten LV wall stress and stiffness (35), eventually leading to decreased endocardial LS (36). As the disease progresses and dialysis time increases, myocardial fibrosis and LV hypertrophy is exacerbated, leading to a decrease in myocardial strain of each layer.
The cardiac myocardium can be divided into three layers: endocardial, middle, and epicardial. Approximately, 75% of the endocardial and epicardial layers consist of longitudinal myocardium, while 25% of the middle layer consists of ring-shaped myocardium (37). During myocardial contraction, the middle annular myocardium produces motion in the short axis direction. Compared with the control group, we found that MHD patients had comparable CS at the apical and the middle endocardial segments, indicating that LS was first impaired in MHD patients, while CS was only partially impaired at the early stage of cardiac impairment. Moreover, impaired CS was mainly observed in the middle and epicardial myocardium, which may be related to the fact that the middle myocardium has more ring-shaped myocardium. Since hemodialysis removes excess fluid from the body, the heart consequently increases LV contraction to maintain normal cardiac output and blood pressure (38). However, MHD patients often have compromised cardiac function, and present with fluctuating blood pressure after hemodialysis. Since the mid- and epi-myocardium is sensitive to afterload (30–32), a large range of blood pressure fluctuation may underlie reduced CS in the mid- and epi-cardial myocardium of MHD patients.
LV TTP is the time from the R-wave of the electrocardiogram to the longitudinal peak strain of the LV 17 segments, and PSD measures the dispersion of the TTP of the LV 17 segments. Both TTP and PSD can be used to evaluate LV myocardial synchrony (39). Compared with the control group, MHD patients had significantly delayed TTP and increased PSD, indicating decreased LV myocardial synchrony in MHD patients. This decreased LV synchrony could be attributed to increased sodium retention and high blood pressure. Indeed, some studies have shown that hypertension causes LV remodeling and alters myocardial electrophysiological properties, such as increased cardiomyocyte autorhythmicity and potential instability, myocardial electrophysiological conduction block, and myocardial excitation-contraction coupling dysregulation, all of which may lead to impaired LV systolic synchrony (40–41). Due to excessive volume overload, elevated peripheral blood pressure, and lack of ATP, LV contraction time is extended as a compensatory mechanism to insure adequate cardiac output (42). Indeed, we found that PSD was positively correlated with GLS, which also indicates that the decrease in LV contraction in MHD patients is accompanied by a decrease in contractile synchrony. We also found that PSD was positively correlated with sub-endocardial and mid-myocardial GLS, but not with sub-epicardial GLS. This finding was probably due to the distribution of His bundle, left bundle branch, and Purkinje fiber mainly in sub-endocardium and mid-myocardium, thus rendering these myocardia more susceptible to the above factors (43).
Previously, Shi et al (44) studied LV strain of young and middle-aged patients undergoing peritoneal dialysis, and found that sub-endocardial GLS was more sensitive to blood perfusion than the epicardium. Sun et al (45) used LST technology to study LV function in ESRD patients, and reported that the curvature radius of the annular myocardial fiber was smaller than the curvature radius of the longitudinal myocardial fiber. Moreover, they found that the tension of the circumferential myocardial fiber was lower when deformed. Thus, they concluded that the LS of the myocardial fiber was more sensitive than CS. Leng et al (46) also found that the decrease in LS occurred earlier than CS in response to external insults. In the present study, we analyzed the efficacy of three-layer strain parameters for predicting LV systolic function in MHD patients, and found that the AUC of the sub-endocardial LS was the largest with a cut-off value of –21.15%, indicating that the sub-endocardial LS was the best index to evaluate LV systolic function. The AUC of the mid- and sub-epicardial CS was greater than that of the sub-endocardial CS. Also, the AUC of the sub-endocardial LS was greater than that of the mid- and epi-cardial CS. Thus, our findings are in line with the previous reports.