Introduction: Although maintenance hemodialysis (MHD) in end-stage renal disease (ESRD) appears to induce some risk factors and strengthen cardiac function, the morbidity of ESRD patients receiving hemodialysis remains high. This study aimed to identify left ventricular (LV) structural and functional abnormalities in ESRD patients on MHD using three-dimensional speckle-tracking imaging (3D-STI). Methods: Eighty-five ESRD patients with normal LV ejection fraction (LVEF >50%) participated in this study, including 55 MHD patients comprising the chronic kidney disease (CKD) V-D group and 30 nondialysis patients comprising the CKD V-ND group. Thirty age- and sex-matched control participants who had normal kidney function were enrolled as the N group. Conventional echocardiography and 3D-STI were conducted, and global longitudinal strain (GLS), global circumferential strain (GCS), global area strain (GAS), and global radial strain (GRS) values were measured. Results: No substantial differences in two-dimensional LVEF were observed among the three groups, and LV hypertrophy was the most common abnormality in patients with ESRD, irrespective of whether they had received or not received MHD. There were no significant differences in the 3D LV mass index between the CKD V-ND and N groups (p > 0.05). Conversely, the 3D LV mass index was considerably higher in the CKD V-D group than in both the N and CKD V-ND groups. The GLS, GAS, and GRS values were significantly lower in the CKD V-ND group than in the N group (p < 0.05). Furthermore, the CKD V-D group had significantly lower GLS, GCS, GAS, and GRS values than the N and CKD V-ND groups (p < 0.05). The interventricular septal thickness and E/e’ ratio were independently associated with LV strain values in all patients with ESRD. Conclusions: MHD can exacerbate LV deformation and dysfunction in ESRD patients with preserved LVEF, and 3D-STI can be potentially useful for detecting these asymptomatic preclinical abnormalities.

Chronic kidney disease (CKD) is a global health problem, and end-stage renal disease (ESRD) is the final manifestation of chronic renal insufficiency, affecting approximately 15% of the adult population and causing high morbidity and mortality [1]. Cardiovascular disease (CVD) is the major cause of mortality in ESRD patients; approximately 58% of these patients die of CVD, and heart failure and sudden cardiac death account for a large number of CVD events, particularly at advanced CKD stages [2, 3]. Currently, hemodialysis is the major treatment method for ESRD patients who cannot undergo kidney transplantation [4]. Long-term hemodialysis has been demonstrated to be effective in reducing clinical symptoms and enhancing the quality of life of ESRD patients. However, the mortality and morbidity of ESRD patients on hemodialysis remain high, and their quality of life remains subpar [5].

Therefore, early detection of cardiac function in patients on maintenance hemodialysis (MHD) has major clinical relevance. Conventional echocardiography is not sensitive enough to detect early cardiac dysfunction in CKD patients [6]. Moreover, two-dimensional speckle-tracking imaging (2D-STI) has been used for evaluating left ventricular (LV) function. Yan et al. [7] showed that longitudinal, radial, and circumferential strains were reduced in MHD patients with preserved LV ejection fraction (LVEF). Compared with 2D-STI, three-dimensional speckle-tracking imaging (3D-STI) is a highly accurate and reproducible technique for the evaluation of LV volume and function [8] and can assess myocardial motion more adequately, accurately, and comprehensively (without out-of-plane speckle loss); offer strain information closer to the truth; and more accurately and reliably evaluate LV myocardial function [9]. Currently, studies on 3D-STI are primarily applied to coronary heart disease, cardiomyopathy, valvular heart disease, and other heart diseases [10‒12]. It can accurately evaluate myocardial motor function and has significant value in the differential diagnosis of diseases, risk stratification, and predicting adverse events. Therefore, herein, we used 3D-STI, a newly developed approach, for evaluating preclinical LV abnormalities in ESRD patients on MHD and analyzed the effect of MHD on their cardiac structure and function.

Participants

Between March 2021 and June 2022, 97 ESRD patients with an LVEF of ≥50% were diagnosed in our hospital. All patients had a history of chronic renal failure, consistent with stage 5 CKD as per the KDIGO 2012 clinical practice guidelines for the evaluation and treatment of CKDs. Their estimated glomerular filtration rate (eGFR) was <15 mL/min/1.73 m2. MHD patients underwent MHD treatment thrice weekly for ≥3 months or more. The distal forearm radial artery and cephalic vein fistula were developed. Patients who satisfied any one of the following criteria were excluded: (1) patients with heart diseases (congenital heart disease, heart valve disease, coronary heart disease, cardiomyopathy, heart failure, arrhythmia, etc.); (2) patients who had acute renal failure; (3) patients with a poor acoustic window and fuzzy 3D images; and (4) patients with arteriovenous fistula complications, such as stenosis, thrombosis, infection, and aneurysm. Finally, 85 patients with ESRD were enrolled in this study and categorized into two subgroups according to whether they received MHD through the forearm fistula: 55 MHD patients in the CKD V-D group (33 men and 22 women) and 30 nondialysis patients in the CKD V-ND group (18 men and 12 women).

We also enrolled 30 age- and sex-matched healthy volunteers as the control group, namely, the N group (18 men and 12 women). There was no evidence of cardiac or kidney disease and any other diseases on laboratory, ultrasound, or radiological tests in these individuals. The patients had good acoustic window and high-quality 3D images.

Research Content

Instruments and Methods

All conventional tests were conducted using a Vivid E95 scanner (GE Ultrasound) and cardiac transducers (4Vc: frequency 1.4–5.2 MHz; M5Sc: frequency 1.4–4.6 MHz). All parameters were obtained before dialysis. Routine measurement of systolic blood pressure (SBP), diastolic blood pressure (DBP), heart rate (HR), weight, height, body surface area (m2), and body mass index in the case and control groups was conducted. The participants were directed to lie on their left side and to breathe calmly, while a routine electrocardiogram was recorded in a quiet state.

Conventional Echocardiography Acquisition

Conventional echocardiography was performed. Left atrial diameter (LAD), LV end-diastolic diameter (LVDd), interventricular septal thickness (IVST), posterior wall thickness (PWT), and LVEF were analyzed according to the American Society of Echocardiography recommendations. LVEF was evaluated using the Simpson’s biplane method. Transmitral inflow diastolic velocities (E and A) were collected using the pulsed wave Doppler ultrasound. Mitral annular diastolic velocities (e’ and a’) were measured using tissue Doppler imaging. Subsequently, E/A, e’/a’, and E/e’ ratios were determined.

3D-STI

We used a fully sampled matrix array transducer (GE 4Vc) for 3D data acquisition. The depth and angle were modified moderately such that the frame rate could be >40% of the participant’s HR. A standard A4C was obtained in the full-volume mode and stored for later examination. Subsequently, the GE EchoPAC software was used to evaluate the stored 3D full-volume images. First, the points of mitral valve closure and the LV apex were calculated manually in A4C. The tracks of endocardial and epicardial borders in a single cycle were then detected automatically by the software. Necessary adjustments were made manually to achieve optimal LV delineation. Finally, the LV volume-time curve, LVEF, and regional and global strain values in several directions were generated by the system. The 3D routine data and strain parameters were as follows: LV end-diastolic volume (EDV), end-systolic volume (ESV), 3D LVEF, LV end-diastolic mass index (EDMassI), and end-systolic mass index (ESMassI). Global longitudinal strain (GLS), global circumferential strain (GCS), global area strain (GAS), and global radial strain (GRS) values were determined using the software as the weighted averages of these mental peak strain values (Fig. 1, 2). All parameter measurements were performed by senior physicians.

Fig. 1.

Demonstration of 3D-STI in a 34-year-old woman with ESRD. a The 4D Automated Left Ventricular Quantification software presented the LV volume-time curve, end-diastolic volume (EDV), end-systolic volume (ESV), ejection fraction (EF), and stroke volume (SV) after identifying the track of endocardial border in a single cycle. b The 4D Automated Left Ventricular Quantification software traced the epicardial border and collected an entire delineation of the LV wall. EDV, end-diastolic volume; EF, ejection fraction; ESV, end-systolic volume; HR, heart rate; SpI, spherical index; SV, stroke volume.

Fig. 1.

Demonstration of 3D-STI in a 34-year-old woman with ESRD. a The 4D Automated Left Ventricular Quantification software presented the LV volume-time curve, end-diastolic volume (EDV), end-systolic volume (ESV), ejection fraction (EF), and stroke volume (SV) after identifying the track of endocardial border in a single cycle. b The 4D Automated Left Ventricular Quantification software traced the epicardial border and collected an entire delineation of the LV wall. EDV, end-diastolic volume; EF, ejection fraction; ESV, end-systolic volume; HR, heart rate; SpI, spherical index; SV, stroke volume.

Close modal
Fig. 2.

Strain-time curve and peak global systolic strain values were evaluated using the software in a 34-year-old woman with ESRD. a GLS. b GCS. c GAS. d GRS.

Fig. 2.

Strain-time curve and peak global systolic strain values were evaluated using the software in a 34-year-old woman with ESRD. a GLS. b GCS. c GAS. d GRS.

Close modal

Other Examinations

Fasting venous blood was collected within 1 h before dialysis for routine blood biochemical tests including creatinine, blood urea nitrogen (BUN), N-terminal pro-brain natriuretic peptide (NT-proBNP), glomerular filtration rate, calcium, phosphorus, hemoglobin (Hb), and parathormone.

Evaluation of Intraobserver and Interobserver Variability

We randomly selected 30 participants from the N, CKD V-ND, and CKD V-D groups and evaluated the differences between two observations made by the same observer after a few days. The differences between two independent observers who were unaware of the patient groupings were also examined. Intraclass correlation coefficients (ICCs) were used to assess the intraobserver and interobserver variability. The criteria for ICCs were as follows: “good” if ≥0.75, “moderate” if 0.41–0.74, and “poor” if ≤0.40.

Statistical Analyses

All statistical analyses were performed using SPSS version 22.0 statistical software. Continuous variables were expressed as means ± standard deviations for normally distributed data or as medians (25th–75th percentile) for non-normally distributed data. One-way analysis of variance was used for multiple-group comparisons. Fisher’s least significant difference (LSD) test or Tamhane’s T2 was used for multiple comparisons between groups for normally distributed data, or Kruskal-Wallis H tests were used for multiple-group comparisons and Mann-Whitney U test was used for two-group comparisons for non-normally distributed data. Categorical variables were presented as frequencies and compared using χ2 test or Fisher’s exact test. Multivariate linear regression analysis was performed to calculate the independent predictors of GLS, GCS, GAS, and GRS in patients with ESRD. A p value of <0.05 was considered statistically significant.

Baseline Demographic and Clinical Data

Table 1 summarizes the baseline demographic and clinical characteristics of the study participants. A total of 55 (64.7%) patients were on MHD, and 30 (35.2%) were on nondialysis. The former had a significantly longer history of renal failure than the latter (p < 0.05). The median dialysis vintage of the CKD V-D group was 24 (12, 51) months. Blood tests revealed that ESRD patients had anemia, hypertension, and hyperparathyroidism. In this study, the most common etiology was chronic glomerulonephritis, followed by diabetes. The CKD V-D group had considerably higher creatinine, phosphorus, and lower eGFR levels than the CKD V-ND and N groups. Additionally, the CKD V-ND group had higher creatinine and phosphorus and lower eGFR levels than the N group (p < 0.05). Compared with normal controls, ESRD patients had significantly higher levels of BUN, NT-proBNP, SBP, and DBP and lower levels of Hb (p < 0.05). BUN, NT-proBNP, Hb, SBP, and DBP levels did not differ significantly between the CKD V-D and CKD V-ND groups (p > 0.05). The CKD V-D group had a significantly higher HR than the normal group (p < 0.05); however, compared with the CKD V-ND group, there were no significant differences (p > 0.05). There were no statistically significant differences in the remaining variables among the groups (Table 1).

Table 1.

Baseline demographic and clinical data for patients with ESRD and controls

ParameterN group (n = 30)CKD V-ND group (n = 30)CKD V-D group (n = 55)F//H/Z/χ2p value
Age, years 45.70±10.17 49.97±11.38 51.11±14.22 1.841 0.163 
Sex (female), n (%) 12 (40) 12 (40) 22 (40) 0.000 1.000 
Height, cm 165.13±8.41 164.43±7.92 163.02±8.09 0.731 0.484 
Weight, kg 60.07±10.07 59.78±10.79 56.24±10.37 1.798 1.7 
BSA, m2 1.64±0.17 1.66±0.19 1.59±0.15 1.705 0.186 
BMI, kg/m2 21.93±2.58 22.03±3.16 21.19±3.88 0.781 0.46 
HR, beats/min 71.14±9.11 75.30±9.73 80.14±16.81a 5.175 0.008 
SBP, mm Hg 117.57±13.43 146.13±17.57a 142.98±21.32a 33.86 0.000 
DBP, mm Hg 75.17±11.69 86.80±13.57a 83.80±12.11a 7.344 0.001 
Cre, μmol/L 65 (59–79) 749.5 (619.5–1,083.7)a 992 (786–1,333)a,b 72.02 0.000 
BUN, mmol/L 4.8 (3.6–5.2) 24.49 (19.54–31.7)a 24 (18.8–30.64)a 65.98 0.000 
NT-proBNP, pg/mL 43 (33.5–54.75) 2,420 (735–12,295.5)a 6,962 (2,312–33,039)a 63.26 0.000 
eGFR, mL/min 106.5 (90.75–149.75) 6 (4–7.25)a 3 (3–4)a,b 77.52 0.000 
Hb, g/L 145.13±13.01 91.9±21.84a 104.32±27.32a 85.26 0.000 
Ca, mmol/L 2.30±0.09 2.19±0.24 2.26±0.23 2.64 0.079 
P, mmol/L 1.21±0.11 1.68±0.42a 2.05±0.91a,b 37.76 0.000 
CKD etiology, n (%) 
 Glomerulonephritis 10 (33.3) 27 (49.1) 1.961 0.161 
 Hypertensive 1 (3.3) 5 (9.1) 0.3 0.584 
 Diabetic 5 (16.7) 7 (12.7) 0.746 
 IgA 4 (13.3) 0 (0) 5.009 0.025 
 Polycystic 2 (6.7) 2 (3.6) 0.009 0.925 
 Obstructive 0 (0) 5 (9.1) 1.488 0.222 
 Hyperuricemic 2 (6.7) 0 (0) 0.122 
 SLE 1 (3.3) 2 (3.6) 0.000 1.000 
 ANCA 0 (0) 1 (1.8) 1.000 
 Other/unknown 5 (16.7) 6 (10.9) 0.174 0.676 
HD 30 (35.2) 55 (64.7)   
Dialysis vintage, months 24 (12–51)   
history of renal failure, months 12 (6–25) 48 (20–84) −3.677 0.000 
Hypertension, n (%) 26 (86.7) 49 (89) 0.000 1.000 
Diabetes, n (%) 8 (26.7) 10 (18.2) 0.837 0.36 
HPT, n (%) 20 (66.7) 45 (81.8) 2.477 0.116 
β-RB, n (%) 13 (43.3) 21 (38.2) 0.215 0.643 
RAAS inhibitors, n (%) 11 (36.7) 23 (41.8) 0.215 0.643 
α-RA, n (%) 7 (23.3) 10 (18.2) 0.322 0.57 
SGLT2i, n (%) 2 (6.7) 4 (7.3) 0.000 1.000 
CCB, n (%) 23 (76.7) 41 (74.5) 0.047 0.828 
Severity of anemia, n (%) 
 No anemia 2 (6.7) 13 (23.6) 3.846 0.05 
 Mild 15 (50) 23 (41.8) 0.526 0.468 
 Moderate 12 (40) 16 (29.1) 1.046 0.306 
 Severe 1 (3.3) 3 (5.4) 0.000 1.000 
ParameterN group (n = 30)CKD V-ND group (n = 30)CKD V-D group (n = 55)F//H/Z/χ2p value
Age, years 45.70±10.17 49.97±11.38 51.11±14.22 1.841 0.163 
Sex (female), n (%) 12 (40) 12 (40) 22 (40) 0.000 1.000 
Height, cm 165.13±8.41 164.43±7.92 163.02±8.09 0.731 0.484 
Weight, kg 60.07±10.07 59.78±10.79 56.24±10.37 1.798 1.7 
BSA, m2 1.64±0.17 1.66±0.19 1.59±0.15 1.705 0.186 
BMI, kg/m2 21.93±2.58 22.03±3.16 21.19±3.88 0.781 0.46 
HR, beats/min 71.14±9.11 75.30±9.73 80.14±16.81a 5.175 0.008 
SBP, mm Hg 117.57±13.43 146.13±17.57a 142.98±21.32a 33.86 0.000 
DBP, mm Hg 75.17±11.69 86.80±13.57a 83.80±12.11a 7.344 0.001 
Cre, μmol/L 65 (59–79) 749.5 (619.5–1,083.7)a 992 (786–1,333)a,b 72.02 0.000 
BUN, mmol/L 4.8 (3.6–5.2) 24.49 (19.54–31.7)a 24 (18.8–30.64)a 65.98 0.000 
NT-proBNP, pg/mL 43 (33.5–54.75) 2,420 (735–12,295.5)a 6,962 (2,312–33,039)a 63.26 0.000 
eGFR, mL/min 106.5 (90.75–149.75) 6 (4–7.25)a 3 (3–4)a,b 77.52 0.000 
Hb, g/L 145.13±13.01 91.9±21.84a 104.32±27.32a 85.26 0.000 
Ca, mmol/L 2.30±0.09 2.19±0.24 2.26±0.23 2.64 0.079 
P, mmol/L 1.21±0.11 1.68±0.42a 2.05±0.91a,b 37.76 0.000 
CKD etiology, n (%) 
 Glomerulonephritis 10 (33.3) 27 (49.1) 1.961 0.161 
 Hypertensive 1 (3.3) 5 (9.1) 0.3 0.584 
 Diabetic 5 (16.7) 7 (12.7) 0.746 
 IgA 4 (13.3) 0 (0) 5.009 0.025 
 Polycystic 2 (6.7) 2 (3.6) 0.009 0.925 
 Obstructive 0 (0) 5 (9.1) 1.488 0.222 
 Hyperuricemic 2 (6.7) 0 (0) 0.122 
 SLE 1 (3.3) 2 (3.6) 0.000 1.000 
 ANCA 0 (0) 1 (1.8) 1.000 
 Other/unknown 5 (16.7) 6 (10.9) 0.174 0.676 
HD 30 (35.2) 55 (64.7)   
Dialysis vintage, months 24 (12–51)   
history of renal failure, months 12 (6–25) 48 (20–84) −3.677 0.000 
Hypertension, n (%) 26 (86.7) 49 (89) 0.000 1.000 
Diabetes, n (%) 8 (26.7) 10 (18.2) 0.837 0.36 
HPT, n (%) 20 (66.7) 45 (81.8) 2.477 0.116 
β-RB, n (%) 13 (43.3) 21 (38.2) 0.215 0.643 
RAAS inhibitors, n (%) 11 (36.7) 23 (41.8) 0.215 0.643 
α-RA, n (%) 7 (23.3) 10 (18.2) 0.322 0.57 
SGLT2i, n (%) 2 (6.7) 4 (7.3) 0.000 1.000 
CCB, n (%) 23 (76.7) 41 (74.5) 0.047 0.828 
Severity of anemia, n (%) 
 No anemia 2 (6.7) 13 (23.6) 3.846 0.05 
 Mild 15 (50) 23 (41.8) 0.526 0.468 
 Moderate 12 (40) 16 (29.1) 1.046 0.306 
 Severe 1 (3.3) 3 (5.4) 0.000 1.000 

BSA, body surface area; BMI, body mass index; HR, heart rate; SBP, systolic blood pressure; DBP, diastolic blood pressure; Cre, creatinine; BUN, blood urea nitrogen; GFR, glomerular filtration rate; NT-proBNP, N-Terminal Pro-Brain Natriuretic Peptide; CKD, chronic kidney disease; HD, hemodialysis; CCB, calcium channel blockers; β-RB, beta-receptor blocker; α-RA, alpha-receptor antagonists; SGLT2i, sodium-dependent glucose transporters 2 inhibitors; RAAS, renin-angiotensin-aldosterone system; Hb, hemoglobin; Ca, calcium; P, phosphorus; HPT, hyperparathyroidism.

ap < 0.05 versus N group.

bp < 0.05 versus CKD V-ND group.

Conventional Echocardiographic Data

LV wall thickness was significantly higher in the CKD V-D group than in the CKD V-ND and N groups (p < 0.05). ESRD patients had a significantly higher LAD and E/e’ ratio and lower e’ values and e’/a’ ratio than normal controls (p < 0.05). No significant differences were observed in LAD, E/e’, e’/a’ ratio, and e’ values between the CKD V-D and CKD V-ND groups. LVDd values were significantly higher in the CKD V-ND group than in the N group (p < 0.05). No significant differences in two-dimensional ejection fraction and the peak transmitral flow velocity in early diastole, namely, E, were observed among the groups (Table 2).

Table 2.

Conventional echocardiographic data for patients with ESRD and controls

ParameterN group (n = 30)CKD V-ND group (n = 30)CKD V-D group (n = 55)F/Hp value
LAD, mm 29.83±3.08 35.90±5.93a 36.91±6.74a 27.63 0.000 
LVDd, mm 43.23±3.55 47.55±5.07a 45.71±6.55 7.62 0.001 
IVST, mm 8.53±0.97 12.10±1.58a 12.91±2.16a,b 107.67 0.000 
PWT, mm 8.50±1.20 11.97±1.61a 12.80±2.07a,b 59.55 0.000 
2D EF, % 63.10±8.39 62.30±7.21 61.09±9.07 0.59 0.557 
E, cm/s 80 (70–100) 80 (70–102) 80 (70–90) 0.18 0.913 
e’, cm/s 10 (8–11.5) 4 (4–6)a 4 (4–5)a 52.38 0.000 
E/e’ 7 (6.18–9.09) 12.86 (10.66–20.63)a 14.29 (10–20)a 44.09 0.000 
e’/a’ 1.22±0.39 0.67±0.23a 0.58±0.22a 33.46 0.000 
ParameterN group (n = 30)CKD V-ND group (n = 30)CKD V-D group (n = 55)F/Hp value
LAD, mm 29.83±3.08 35.90±5.93a 36.91±6.74a 27.63 0.000 
LVDd, mm 43.23±3.55 47.55±5.07a 45.71±6.55 7.62 0.001 
IVST, mm 8.53±0.97 12.10±1.58a 12.91±2.16a,b 107.67 0.000 
PWT, mm 8.50±1.20 11.97±1.61a 12.80±2.07a,b 59.55 0.000 
2D EF, % 63.10±8.39 62.30±7.21 61.09±9.07 0.59 0.557 
E, cm/s 80 (70–100) 80 (70–102) 80 (70–90) 0.18 0.913 
e’, cm/s 10 (8–11.5) 4 (4–6)a 4 (4–5)a 52.38 0.000 
E/e’ 7 (6.18–9.09) 12.86 (10.66–20.63)a 14.29 (10–20)a 44.09 0.000 
e’/a’ 1.22±0.39 0.67±0.23a 0.58±0.22a 33.46 0.000 

LA, left atrium; LVDd, LV end-diastolic diameter; IVST, interventricular septum thickness; PWT, posterior wall thickness; 2D EF, 2D ejection fraction; E, peak transmitral flow velocity in early diastole; e’ and a’, peak early diastolic and late peak diastolic of myocardial velocities the mitral annulus.

ap < 0.05 versus normal group.

bp < 0.05 versus CKD V-ND group.

Three-Dimensional Morphologic Data and Strain Parameters

EDV and ESV values were significantly higher in ESRD patients than in normal controls (p < 0.05). No significant differences in EDV and ESV were observed between the CKD V-D and CKD V-ND groups. The 3D LVEF was lower in ESRD patients than in normal controls (p < 0.05). No significant differences in the 3D LV mass index were noted between the CKD V-ND and N groups. Conversely, the 3D LV mass index was significantly higher in the CKD V-D group than in both the N and CKD V-ND groups (p < 0.05). GLS, GAS, and GRS values were significantly lower in the CKD V-ND group than in the N group (p < 0.05). Furthermore, the CKD V-D group had significantly lower GLS, GCS, GAS, and GRS values than the N and CKD V-ND groups (p < 0.05; Table 3).

Table 3.

Three-dimensional speckle-tracking parameters for patients with ESRD and controls

ParameterN group (n = 30)CKD V-ND group (n = 30)CKD V-D group (n = 55)F/Hp value
EDV, mL 80.47±18.59 110.18±18.58a 109.04±35.00a 18.73 0.000 
ESV, mL 30.47±9.54 47.55±14.77a 51.60±23.24a 24.76 0.000 
3D EF, % 62.38±6.23 56.90±7.58a 53.74±9.76a 10.21 0.000 
EDmassI, g/m2 73 (68.75–80.5) 78 67.5–85) 87 (75–99)a,b 21.84 0.000 
ESmassI, g/m2 73.5 (69.75–82.50) 78 (67.75–85.25) 87 (75–99)a,b 22.74 0.000 
GLS, % −18.51±3.29 −15.55±2.94a −13.60±4.33a,b 17.21 0.000 
GCS, % −18.57±3.14 −17.60±3.87 −14.82±4.09a,b 11.06 0.000 
GAS, % −31.23±4.14 −27.93±5.21a −24.16±6.37a,b 16.21 0.000 
GRS, % 54.37±11.42 47.17±12.44a 39.04±13.13a,b 15.05 0.000 
ParameterN group (n = 30)CKD V-ND group (n = 30)CKD V-D group (n = 55)F/Hp value
EDV, mL 80.47±18.59 110.18±18.58a 109.04±35.00a 18.73 0.000 
ESV, mL 30.47±9.54 47.55±14.77a 51.60±23.24a 24.76 0.000 
3D EF, % 62.38±6.23 56.90±7.58a 53.74±9.76a 10.21 0.000 
EDmassI, g/m2 73 (68.75–80.5) 78 67.5–85) 87 (75–99)a,b 21.84 0.000 
ESmassI, g/m2 73.5 (69.75–82.50) 78 (67.75–85.25) 87 (75–99)a,b 22.74 0.000 
GLS, % −18.51±3.29 −15.55±2.94a −13.60±4.33a,b 17.21 0.000 
GCS, % −18.57±3.14 −17.60±3.87 −14.82±4.09a,b 11.06 0.000 
GAS, % −31.23±4.14 −27.93±5.21a −24.16±6.37a,b 16.21 0.000 
GRS, % 54.37±11.42 47.17±12.44a 39.04±13.13a,b 15.05 0.000 

EDV, end-diastolic volume; ESV: end-systolic volume; EF, ejection fraction; EDmassI, end-diastolic mass index; ESmassI, end-systolic mass index; GLS, global longitudinal strain; GCS, global circumferential strain; GAS, global area strain; and GRS, global radial strain.

ap < 0.05 vs. normal group.

bp <0.05 vs. CKD V-ND group.

Predictive Factors for 3D Strain in Patients with ESRD

Based on the multivariate linear regression analysis, the IVST and E/e’ ratio were associated with GLS, GCS, GAS, and GRS, which were used to identify the risk factors for preclinical myocardial systolic dysfunction. IVST and E/e’ ratio were independently associated with the GLS (β = 0.48, p = 0.022; β = 0.327, p = 0.001), GCS (β = 0.511, p = 0.032; β = 0.29, p = 0.006), GAS (β = 0.603, p = 0.006; β = 0.334, p = 0.001), and GRS (β = −0.542, p = 0.014; β = −0.312, p = 0.002) values (Table 4).

Table 4.

Predictive factors for 3D strain in patients with ESRD

GLSGCSGASGRS
βp valueβp valueβp valueβp value
SBP 0.013 0.901 −0.071 0.562 −0.032 0.778 −0.003 0.978 
DBP 0.160 0.104 0.185 0.098 0.163 0.112 −0.212 0.041* 
IVST 0.48 0.022* 0.511 0.032* 0.603 0.006* −0.542 0.014* 
PWT −0.188 0.374 −0.409 0.090 −0.378 0.090 0.367 0.101 
E/e’ 0.327 0.001* 0.290 0.006* 0.334 0.001* −0.312 0.002* 
GLSGCSGASGRS
βp valueβp valueβp valueβp value
SBP 0.013 0.901 −0.071 0.562 −0.032 0.778 −0.003 0.978 
DBP 0.160 0.104 0.185 0.098 0.163 0.112 −0.212 0.041* 
IVST 0.48 0.022* 0.511 0.032* 0.603 0.006* −0.542 0.014* 
PWT −0.188 0.374 −0.409 0.090 −0.378 0.090 0.367 0.101 
E/e’ 0.327 0.001* 0.290 0.006* 0.334 0.001* −0.312 0.002* 

SBP, systolic blood pressure; GLS, global longitudinal strain; GCS, global circumferential strain; GAS, global area strain; GRS, global radial strain; IVST, interventricular septum thickness; PWT, posterior wall thickness; E, peak transmitral flow velocity in early diastole; e’, peak early diastolic of myocardial velocities the mitral annulus.

*p < 0.05.

Intraobserver and Interobserver Variability

We also evaluated the intraobserver and interobserver variability in this study. The intraobserver differences in the ICC values of GLS, GCS, GAS, and GRS were 0.90, 0.95, 0.94, and 0.97, respectively, while the interobserver differences were 0.88, 0.80, 0.88, and 0.89, respectively. All ICC values were >0.75, which indicated good uniformity. Table 5 depicts the ICC values and 95% confidence interval for each parameter. Our results indicated that our study generated reliable and consistent observations.

Table 5.

Intra- and interobserver variability

IntraobserverInterobserver
ICC95% CIICC95% CI
GLS, % 0.90 0.81–0.95 0.88 0.76–0.94 
GCS, % 0.95 0.89–0.97 0.80 0.63–0.90 
GAS, % 0.94 0.88–0.97 0.88 0.77–0.94 
GRS, % 0.97 0.96–0.98 0.89 0.79–0.94 
IntraobserverInterobserver
ICC95% CIICC95% CI
GLS, % 0.90 0.81–0.95 0.88 0.76–0.94 
GCS, % 0.95 0.89–0.97 0.80 0.63–0.90 
GAS, % 0.94 0.88–0.97 0.88 0.77–0.94 
GRS, % 0.97 0.96–0.98 0.89 0.79–0.94 

ICC, intraclass correlation coefficient; 95% CI, 95% confidence interval; GLS, global longitudinal strain; GCS, global circumferential strain; GAS, global area strain; and GRS, global radial strain.

In this study, we used conventional echocardiography and 3D-STI to evaluate LV structure and function in ESRD patients on/not on MHD and examined the effect of MHD on their LV structure and function. The major outcomes of this study were as follows: (1) significant differences in LV patterns were found among the three groups. (2) LVEF was normal in the CKD V-D group; however, GLS, GAS, and GRS values were significantly lower in the CKD V-ND group than in the N group. Furthermore, the CKD V-D group had significantly lower GLS, GCS, GAS, and GRS values than the N and CKD V-ND groups. (3) Increased 3D LV mass index was found only in MHD patients and not in nondialysis patients. (4) IVST and E/e’ ratio were independently associated with LV strain values in all ESRD patients. These results indicate that even in ESRD patients with normal EF, the myocardial strain has reduced, and cardiac function had already been impaired, particularly in those undergoing dialysis. 3D-STI can help detect these anomalies early.

Herein, we discovered that although the patients with ESRD had normal two-dimensional ejection fraction, MHD patients had significantly elevated cardiac functional indices, including LAD, IVST, PWT, and E/e’. Compared with controls, nondialysis patients had significantly higher LVDd and lower e’ values and e’/a’ ratio. These results indicate that ESRD patients exhibit LV remodeling, which affects LV diastolic function, even after MHD treatment. According to studies, LV remodeling is associated with the development of a peculiar CKD-related cardiac remodeling, progressive LV hypertrophy (LVH), fibrosis [13], and myocardial edema [14]; is followed mainly by diastolic LV dysfunction; and is associated with worse prognosis.

Novel 2D-STI is a sensitive and accurate method for detecting preclinical changes in LV function, even when LVEF is normal. Studies demonstrated that changes in longitudinal strain can help detect early preclinical LV dysfunction and is a significant prognostic predictor in ESRD patients with preserved LV EF [15]. This technique can evaluate myocardial motion and myocardial torsional motion from three directions (longitudinal, radial, and circumferential), thereby reflecting the local and overall function of the myocardium [16]. However, because 2D-STI tracking information is derived from a two-dimensional plane and the heart is a three-dimensional structure, the tracking particles may be scanned outside the plane or cannot be completely scanned within the plane, resulting in missing information and an inability to fully reflect myocardial movement. In contrast, the recently developed 3D-STI can assess movements of LV segments in three spatial directions simultaneously, overcoming the limitations of 2D-STI. Moreover, the accuracy and reproducibility of 3D-STI in the evaluation of LV volume and systolic function have been previously verified [9, 17].

There were different responses to the strain analysis after hemodialysis. According to a previous study in preserved LVEF patients, 2D strain analysis revealed that hemodialysis patients had better LV function than advanced CKD patients [18, 19]. Kovács et al. [18] reported a significant increase in all myocardial strain directions after hemodialysis. The improvement of myocardial deformation may be explained by the removal of overhydration gained during the interdialytic period. Liu et al. [19] found that compared with moderate-to-advanced CKD patients, global peak systolic longitudinal strain, circumferential strain, and strain rate were better in ESRD patients receiving MHD. They believe that the improvement in LV function in the hemodialysis patients may have resulted from factors such as improvements in their anemia, malnutrition, calcium-phosphorus metabolism disorder, acid-base balance, etc. These factors may decrease the LV volume load and thereby improve the anoxic condition of tissues and organs.

However, our results were substantially different from the abovementioned findings. In this study, we found that GLS, GCS, GAS, and GRS values were significantly lower in the MHD group than in both the N and CKD V-ND groups. The GLS, GAS, and GRS values were also significantly lower in the CKD V-ND group than in the N group. The 3D LV mass index was significantly higher in the CKD V-D group than in both the N and CKD V-ND groups. Herein, we found that the Hb level was lower in the CKD V-D group than in the N group; however, there was no statistical significance between the CKD V-D and CKD V-ND groups. The value of phosphorus was higher in the CKD V-D group than in the CKD V-ND and N groups. Moreover, there was no significant difference in LAD, LVDd, EDV, and ESV between the CKD V-D and CKD V-ND groups. Therefore, anemia and calcium-phosphorus metabolism disorder did not improve, and the LV volume load did not decrease, which may be one of the reasons why cardiac function did not improve but rather got worse, even if the patients have received hemodialysis. These findings suggest that in ESRD patients with preserved LVEF whose cardiac systolic function had already been impaired, MHD not only failed to provide protection but may further exacerbate the damage to cardiac function. These results support the data published by other authors. For instance, Guler et al. [20] reported that hemodialysis treatment results in deterioration in all LV strain directions. Sun et al. [21] revealed that MHD patients had significantly lower GLS, GRS, and LVEF as well as enlarged LV volume (EDV and ESV) and LV mass index. These results are consistent with our findings.

These results demonstrate the importance of 3D-STI in detecting early LV deformation and dysfunction. The impaired myocardial strain of all directions could be attributed to endothelial dysfunction and vascular injury [22] and the observation that ESRD patients receiving MHD frequently present with anemia, hypertension, secondary hyperparathyroidism, and arteriovenous fistula, all of which increase pre- and post-loading. Moreover, long-term dialysis does not achieve the expected improvement of these worsening conditions and causes LV remodeling, hypertrophy, and fibrosis [13]. Although these pathological changes maintain normal cardiac output in MHD patients, they exacerbate the progression of atherosclerosis and further increase LV wall stress and stiffness [23]. Furthermore, ESRD patients may have pathophysiological changes, such as cardiomyocyte apoptosis, fibrosis, and myocardial calcifications. Overhydration and accumulation of uremic toxins may induce the development of LVH and LV dysfunction in CKD [7]. Furthermore, patients undergoing hemodialysis generally have a long history of kidney failure. In our study, the median history of kidney failure in the dialysis group was 48 (20, 84) months, far greater than that in the nondialysis group. Additionally, the duration of dialysis vintage was 3–144 months, with a median of 24 (12, 51) months. All these factors may lead to reduced myocardial strain value. As the disease progresses and hemodialysis time increases, myocardial fibrosis and LVH are exacerbated, resulting in LV diastolic restriction, impaired diastolic and systolic function, and decreased myocardial strain.

In this study, hemodialysis patients had thicker IVST and PWT, higher LV mass index, and lower myocardial strain than nondialysis patients. Furthermore, multiple linear regression analysis revealed IVST as an independent predictor of global LV systolic function. A previous study showed that LVH was a major predictor of adverse cardiovascular outcomes in patients undergoing chronic hemodialysis [3]. Our findings indicated that LVH could be partly associated with the presence of preclinical LV systolic dysfunction in these patients.

In this study, conventional echocardiography was able to identify changes in cardiac structure and function in patients with ESRD, such as left atrial enlargement, LV wall thickening, increased E/e’, and decreased e’ and e’/a; however, most parameters were not significantly different between the case groups. Moreover, the strain value in 3D-STI could reflect the difference between the case groups. In ESRD patients with preserved LVEF, 3D-STI could identify changes in cardiac function at an early stage.

This study has some limitations. First, this was a single-center study with a small sample size, which may suggest a limitation in the validation of the efficacy of the methods and data. Second, spatial and temporal resolutions are relatively lower in 3D-STI than in 2D-STI. Therefore, several patients had to be excluded from the study because the image quality in one or more segments was inadequate for STI analysis. Third, the rate of long-term cardiovascular events was not determined, and survival evaluations were not performed in MHD patients in this study. Finally, because of the small sample size, we were unable to further categorize MHD patients into subgroups; therefore, the association between dialysis vintage duration and cardiac function should be further corroborated in future studies.

MHD can exacerbate LV deformation and dysfunction in ESRD patients with preserved LVEF, and 3D-STI can be potentially useful for detecting these asymptomatic preclinical abnormalities.

This study was approved by the Medical Ethics Committee of the Second Affiliated Hospital of Hainan Medical University. The data are anonymous, and the requirement for informed consent was therefore waived by the Ethics Committee.

The authors declare that they have no competing interests.

This work was supported by Hainan Provincial Natural Science Foundation of China (No.: 822RC841) and Hainan Provincial Medical and Health Research Project of China (No.: 21A200226).

Meihua Chen and Xuning Huang designed the study. Xiaojuan Chen, Meihua Chen, Lehua Wang, and Hanyin Huang contributed to clinical information collection and literature search. Meihua Chen, Xiaojuan Chen, and Yunpeng Wei performed the data analysis and drafted the manuscript. All the authors revised and approved the final manuscript.

All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.

1.
Gansevoort
RT
,
Matsushita
K
,
van der Velde
M
,
Astor
BC
,
Woodward
M
,
Levey
AS
.
Lower estimated GFR and higher albuminuria are associated with adverse kidney outcomes. A collaborative meta-analysis of general and high-risk population cohorts
.
Kidney Int
.
2011 Jul
80
1
93
104
.
2.
Sarnak
MJ
,
Amann
K
,
Bangalore
S
,
Cavalcante
JL
,
Charytan
DM
,
Craig
JC
.
Chronic kidney disease and coronary artery disease: JACC state-of-the-art review
.
J Am Coll Cardiol
.
2019 Oct 8
74
14
1823
38
.
3.
Ravera
M
,
Rosa
GM
,
Fontanive
P
,
Bussalino
E
,
Dorighi
U
,
Picciotto
D
.
Impaired left ventricular global longitudinal strain among patients with chronic kidney disease and end-stage renal disease and renal transplant recipients
.
Cardiorenal Med
.
2019 Jan
9
1
61
8
.
4.
Shibiru
T
,
Gudina
EK
,
Habte
B
,
Derbew
A
,
Agonafer
T
.
Survival patterns of patients on maintenance hemodialysis for end stage renal disease in Ethiopia: summary of 91 cases
.
BMC Nephrol
.
2013 Jun
14
127
.
5.
Bansal
N
,
Keane
M
,
Delafontaine
P
,
Dries
D
,
Foster
E
,
Gadegbeku
CA
.
A longitudinal study of left ventricular function and structure from CKD to ESRD: the CRIC study
.
Clin J Am Soc Nephrol
.
2013 Mar
8
3
355
62
.
6.
Edwards
NC
,
Hirth
A
,
Ferro
CJ
,
Townend
JN
,
Steeds
RP
.
Subclinical abnormalities of left ventricular myocardial deformation in early-stage chronic kidney disease: the precursor of uremic cardiomyopathy
.
J Am Soc Echocardiogr
.
2008 Dec
21
12
1293
8
.
7.
Yan
P
,
Li
H
,
Hao
C
,
Shi
H
,
Gu
Y
,
Huang
G
.
2D-speckle tracking echocardiography contributes to early identification of impaired left ventricular myocardial function in patients with chronic kidney disease
.
Nephron Clin Pract
.
2011 Jun
118
3
c232
240
.
8.
Huang
BT
,
Yao
HM
,
Huang
H
.
Left ventricular remodeling and dysfunction in systemic lupus erythematosus: a three-dimensional speckle tracking study
.
Echocardiography
.
2014 Oct
31
9
1085
94
.
9.
Amzulescu
MS
,
Langet
H
,
Saloux
E
,
Manrique
A
,
Slimani
A
,
Allain
P
.
Improvements of myocardial deformation assessment by three-dimensional speckle-tracking versus two-dimensional speckle-tracking revealed by cardiac magnetic resonance tagging
.
J Am Soc Echocardiogr
.
2018 Sep
31
9
1021
33
.e1.
10.
Dogdus
M
,
Simsek
E
,
Cinar
CS
.
3D-speckle tracking echocardiography for assessment of coronary artery disease severity in stable angina pectoris
.
Echocardiography
.
2019 Feb
36
2
320
7
.
11.
Orta Kilickesmez
K
,
Baydar
O
,
Bostan
C
,
Coskun
U
,
Kucukoglu
S
.
Four-dimensional speckle tracking echocardiography in patients with hypertrophic cardiomyopathy
.
Echocardiography
.
2015 Oct
32
10
1547
53
.
12.
Bi
X
,
Yeung
DF
,
Salah
HM
,
Arciniegas Calle
MC
,
Thaden
JJ
,
Nhola
LF
.
Dissecting myocardial mechanics in patients with severe aortic stenosis: 2-dimensional vs 3-dimensional-speckle tracking echocardiography
.
BMC Cardiovasc Disord
.
2020 Jan
20
1
33
.
13.
Chen
M
,
Arcari
L
,
Engel
J
,
Freiwald
T
,
Platschek
S
,
Zhou
H
.
Aortic stiffness is independently associated with interstitial myocardial fibrosis by native T1 and accelerated in the presence of chronic kidney disease
.
Int J Cardiol Heart Vasc
.
2019 Jun
24
100389
.
14.
Arcari
L
,
Hinojar
R
,
Engel
J
,
Freiwald
T
,
Platschek
S
,
Zainal
H
.
Native T1 and T2 provide distinctive signatures in hypertrophic cardiac conditions-comparison of uremic, hypertensive and hypertrophic cardiomyopathy
.
Int J Cardiol
.
2020 May
306
102
8
.
15.
Li
C
,
Li
K
,
Yuan
M
,
Bai
W
,
Rao
L
.
Peak strain dispersion within the left ventricle detected by two-dimensional speckle tracking in patients with uncomplicated systemic lupus erythematosus
.
Int J Cardiovasc Imaging
.
2021 Jul
37
7
2197
205
.
16.
Suzuki
K
,
Kato
T
,
Koyama
S
,
Shinohara
T
,
Inukai
S
,
Sato
J
.
Influence of percutaneous occlusion of atrial septal defect on left atrial function evaluated using 2D speckle tracking echocardiography
.
Int Heart J
.
2020 Jan
61
1
83
8
.
17.
Ran
H
,
Zhang
PY
,
Zhang
YX
,
Zhang
JX
,
Wu
WF
,
Dong
J
.
Assessment of left ventricular myocardial viability by 3-dimensional speckle-tracking echocardiography in patients with myocardial infarction
.
J Ultrasound Med
.
2016 Aug
35
8
1631
8
.
18.
Kovács
A
,
Tapolyai
M
,
Celeng
C
,
Gara
E
,
Faludi
M
,
Berta
K
.
Impact of hemodialysis, left ventricular mass and FGF-23 on myocardial mechanics in end-stage renal disease: a three-dimensional speckle tracking study
.
Int J Cardiovasc Imaging
.
2014 Oct
30
7
1331
7
.
19.
Liu
YW
,
Su
CT
,
Huang
YY
,
Yang
CS
,
Huang
JW
,
Yang
MT
.
Left ventricular systolic strain in chronic kidney disease and hemodialysis patients
.
Am J Nephrol
.
2011 Jan
33
1
84
90
.
20.
Guler
HS
,
Tulunay Kaya
C
,
Kumru
G
,
Kosku
H
,
Ozyuncu
N
,
Sengul
S
.
Acute stunning effect of hemodialysis on myocardial performance: a three-dimensional speckle tracking echocardiographic study
.
Artif Organs
.
2020 Oct
44
10
1081
9
.
21.
Sun
M
,
Kang
Y
,
Cheng
L
,
Pan
C
,
Cao
X
,
Yao
H
.
Global longitudinal strain is an independent predictor of cardiovascular events in patients with maintenance hemodialysis: a prospective study using three-dimensional speckle tracking echocardiography
.
Int J Cardiovasc Imaging
.
2016 May
32
5
757
66
.
22.
Krishnasamy
R
,
Hawley
CM
,
Stanton
T
,
Pascoe
EM
,
Campbell
KL
,
Rossi
M
.
Left ventricular global longitudinal strain is associated with cardiovascular risk factors and arterial stiffness in chronic kidney disease
.
BMC Nephrol
.
2015 Jul
16
106
.
23.
Widmer
RJ
,
Lerman
A
.
Endothelial dysfunction and cardiovascular disease
.
Glob Cardiol Sci Pract
.
2014 Oct
2014
3
291
308
.