Abstract
Background and Aims
Cardiovascular disease is the leading cause of mortality in patients with end-stage kidney disease (ESKD). Evidence on the possible echocardiographic differences between patients undergoing different dialysis modalities is scarce. This study aimed to evaluate differences in left (LA) and right atrial (RA) and left (LV) and right ventricular (RV) geometry, systolic and diastolic function, as well as lung water content in hemodialysis and peritoneal dialysis (PD) patients.
Method
A total of 38 hemodialysis and 38 PD patients receiving treatment for ≥3 months, matched in a 1:1 ratio for age, sex and dialysis vintage were included in this study. Lung ultrasound, two-dimensional and tissue-Doppler echocardiography were performed during an interdialytic day in hemodialysis and before a programmed follow-up visit in PD patients. To identify factors possible associated with LVH (left ventricular hypertrophy), we performed univariate and multivariate linear regression analyses in the total population studied.
Results
No significant differences were evidenced in ultrasound B-lines (4.00 [6.00] vs 3.00 [4.25]; p=0.623) between the two groups. Vena cava diameter (11.09±4.53 vs 14.91±4.30 mm; P<0.001) was significantly lower in hemodialysis patients. Indices of LA, RA, LV and RV dimensions were similar between the two groups. LVMi (116.91 [38.56] vs 122.83 [52.33] g/m2; P=0.767) was similar, but relative wall thickness (RWT) was marginally (0.40 [0.14] vs 0.45 [0.15] cm; P=0.055) lower in hemodialysis patients. LV hypertrophy prevalence, defined as LVMi values >95 or >115 g/m2 for female and male patients, was similar between groups (73.7% vs 71.1%; p=0.798), but relative wall thickness (RWT) was numerically lower (0.40 [0.14] vs 0.45 [0.15] cm; P=0.055) and fractional shortening (29.12±7.07% vs 23.37±8.84%; P=0.003) was significantly higher in patients under hemodialysis compared to those under PD. Hemodialysis patients presented mainly eccentric (normal RWT and increased LVMi), while PD patients presented mainly concentric LVH (increased RWT and increased LVMi). Left atrial (LA), right atrial (RA) and ventricular (RV) echocardiographic indices were again similar between the two study groups. Ventricular systolic function was similar between-groups, except for stroke volume (78.97 [24.24] vs 64.66 [27.35] ml; P=0.030) and cardiac output (5.75 [2.29] vs 4.93 [2.10] L/min; P=0.036) which were higher in hemodialysis. With regards to RV systolic function indices, RV systolic pressure (RVSP) was significantly lower in the hemodialysis compared to the PD group (20.37 [22.54] vs 27.68 [14.32] mmHg; P=0.009). All diastolic function indices were similar between the two groups. Prevalence of mitral valve (MV) regurgitation was significantly lower in the hemodialysis group (10.5% vs 39.5%; p=0.004). According to the results of multivariate linear regression analysis, only male gender (β=20.677, 95%CI: 3.479 to 37.874; P=0.019) and number of US-B lines (β=0.892, 95%CI:0.071 to 1.713; P=0.034) were independently associated with LVMi.
Conclusion
Hemodialysis and PD patients present similar volume overload, evaluated with lung ultrasound, and no significant differences in echocardiographic indices reflecting cardiac geometry, but different patterns of abnormal LV remodeling was evident in each dialysis modality, with hemodialysis presenting eccentric and PD concentric LVH. These results clearly support that PD is no better than HD with regards to cardiovascular stress, despite the fact that they experience a more stable volume status.
Lung ultrasound and BNP to detect hidden pulmonary congestion in euvolemic hemodialysis patients: a single centre experience
