| Objective:Left ventricular hypertrophy (LVH) is is an independent predictor of cardiovascular morbidity and mortality in the general population and in patients with CKD. Recently, some studies indicate that prolonged or frequent hemodialysis has a beneficial influence on cardiac morphology and function, can improve the life-quality of hemodialysis patients.Methods:In the database of Pubmed% Cochrane Library, using the search terms "daily HD", " quotidian HD ", "frequent HD", "nocturnal HD" and "extend HD" in English and Chinese. "HD" is short of "hemodialysis or haemodialysis". Selected the full text of documents according to the inclusion criteria and exclusion criteria, update end to September 2015. Trial quality was assessed by the the Newcastle-Ottawa Scale and the Jadad score. System evaluation was performed by using STATA 12.0.Results:11 cohort studies and 4 RCTs were included, and one RCT was a combined study of another two, our data extracted only once from them.1. In an analysis of 11 studies that assessed LVMI (503 analyzable patients), baseline mean LVMI by meta-analysis was 141.0 g/m2 (95% confidence interval [CI], 122.4 to 159.6, conventional hemodialysis is not included). Conversion from conventional HD to frequent or extended HD resulted in a significant decrease in LVMI (-23.8; 95% CI,-32.2 to-15.5; P<0.001). To compare in groups, group 1: Prolonged (nocturnal) hemodialysis (NHD, thrice weekly, 7h to 8h sessions) vs. Conventional hemodialysis (CHD, thrice weekly, 4h to 5h sessions), group 2: Frequent and Prolonged (nocturnal) hemodialysis (FNHD, five times or more a week, 7h to 8h sessions) vs. Conventional hemodialysis (FHD, five times or more a week, 1.5h to 3.5h sessions ), we can not judge which one is better in decreasing of LVMI directly in group 1, or in group 2 neither (NHD: -25.8; 95%CI, -34.5 to -17.1; PO.001. CHD: -7.7; 95%CI, -21.6 to 6.2; P=0.28. FNHD: -17.3; 95%CI, -28.1 to -6.5; P=0.002. FHD: -31.6; 95%CI, -62.0 to -1.3; P=0.041). However, when these studies of small sample size were abandoned, statistical significant was found in group 1, but not group 2 (NHD: -25.8; 95%CI, -34.5 to -17.1; PO.001. CHD: -0.03; 95%CI, -4.23 to 4.18; P=0.991. FNHD: -15.9; 95%CI, -27.0 to -4.9; P=0.005. FHD: -25.8; 95%CI,-63.9 to 12.3; P=0.185).2. In an analysis of 6 studies that assessed LVM (274 analyzable patients), baseline mean LVM by meta-analysis was 160.1 g/m2 (95% CI, 128.3 to 191.8, CHD is not included). Conversion from conventional HD to frequent or extended HD resulted in a significant decrease in LVM (-16.1; 95% CI, -23.5 to -8.6 ; P O.001). To compare in the same groups, we can not judge which one is better in decreasing of LVM directly in neither of the groups (group 1: NHD: -22.7; 95%CI, -54.1 to 8.7; P=0.156. CHD: 0.8; 95%CI, -2.8 to 4.4; P=0.67. group 2: FNHD: -14.6; 95%CI, -25.7 to -3.6; P=0.009. FHD: only one study, -16.3; 95%CI, -23.2 to -9.4).3. In an analysis of 2 studies that assessed EF (143 analyzable patients), baseline mean EF by meta-analysis was 63% (95%CI, 54.3 to 71.7, CHD is not included). Conversion from conventional HD to extended HD resulted in an increase in EF (NHD: 4.2; 95% CI, 2.3 to 6.1 ; P <0.001. CHD: 1.4; 95%CI, 0.1 to 2.6; p = 0.035).4. In an analysis of 7 studies that were accessible for meta-analysis (1140 analyzable patients), out of 9 studies detailed in the deaths, we found that frequent or extended HD was associated with decreased relative risk of mortality (RR=0.52; 95%CI, 0.28 to 0.94; P = 0.030). There was no Statistical significance at percentage of diabetes patients in each group (RR=1.03; 95%CI, 0.85 to 1.24;P = 0.760).To compare in the same groups, we can found that NHD had an advantage over CHD in group 1 on survival possibly, but not in group 2 (FNHD vs.CHD:RR=2.20; 95%CI, 0.34 to 14.48; P= 0.411. NHD vs.CHD:RR=0.31; 95%CI,0.12 to 0.76;P= 0.011. FHD vs.CHD:RR=0.66; 95%CI,0.25 to 1.74; P= 0.402).In an analysis of 3 RCTs (383 analyzable patients), it showed a pooled Relative risk of 0.724 for all-cause mortality compared to CHD (RR= 0.774; 95%CI,0.32 to 1.87;P= 0.570;I2=0.0%, x 2 P=0.433).5. Different measurement methods (ultrasonic cardiogram vs. Cardiac magnetic resonance) lead to different results. No matter in which group, the decreased amount of LVM or LVMI is more larger by ultrasonic cardiogram than by Cardiac magnetic resonance respectively with Statistical significance (group LVM:MRI:-12.7; 95%CI,-17.0 to-8.3; P <0.001. UCG:-40.0; 95%CI,-56.5 to-23.5;P <0.001. Group LVMI: MRI:-6.5; 95%CI,-9.5 to-3.5;P<0.001. UCG:-31.5; 95%CI,-38.9 to-24.1; P <0.001).6. To analysis again conversely by group UCG and group MRI, we found that FHD is better than both FNHD and NHD in improving LVMI, and the effect of FNHD and NHD is similar, we can not judge which one is better directly (FNHD:-28.0; 95%CI,-35.7 to-20.2; P<0.001. NHD:-25.8; 95%CI,-34.5 to-17.1; P<0.001. FHD:-48.2; 95%CI,-59.9 to-36.5; P<0.001). The other indicators is not allowed because of few studies.Conclusions:1. Conversion from conventional to frequent or extended HD is associated with improvements in left ventricular hypertrophy. Frequent HD is better than extended arm of frequent HD in improving the left ventricular hypertrophy at subgroup analysis.2. NHD is associated with improvements in EF and reduction in mortality.3.Whether convert from conventional to frequent or extended arm of frequent HD is better in reducing all-cause mortality or not is not clear.4. Measurement methods maybe a reason of heterogeneity. Large, multicenter randomized, controlled trials are needed to confirm our results, better with long-terms. |
PDF Full Text Request