Effect of frequent or extended hemodialysis on cardiovascular parameters: a meta-analysis

Abstract

Background: Increased left ventricular (LV) mass is a risk factor for cardiovascular mortality in patients with chronic kidney failure. More frequent or extended hemodialysis (HD) has been hypothesized to have a beneficial effect on LV mass.

Study design: Meta-analysis.

Setting & population: MEDLINE literature search (inception to April 2011), Cochrane Central Register of Controlled Trials and ClinicalTrials.gov using the search terms "short daily HD," "daily HD," "quotidian HD," "frequent HD," "intensive HD," "nocturnal HD," and "home HD."

Selection criteria for studies: Single-arm cohort studies (with pre- and post-study evaluations) and trials examining the effect of frequent or extended HD on cardiac morphology and function and blood pressure parameters. Studies of hemofiltration, hemodiafiltration, and peritoneal dialysis were excluded.

Intervention: Frequent (2-8 hours, >3 times weekly) or extended (>4 hours, 3 times weekly) HD compared with conventional (≤4 hours, 3 times weekly) HD.

Outcomes: Absolute changes in cardiac morphology and function, including LV mass index (LVMI; primary) and blood pressure parameters (secondary).

Results: We identified 38 single-arm studies, 5 crossover trials, and 3 randomized controlled trials. By meta-analysis of 23 study arms, frequent or extended HD significantly reduced LVMI from baseline (-31.2 g/m(2), 95% CI, -39.8 to -22.5; P < 0.001). The 3 randomized trials found a less pronounced net reduction in LVMI (-7.0 g/m(2); 95% CI, -10.2 to -3.7; P < 0.001). LV ejection fraction improved by 6.7% (95% CI, 1.6% to 11.9%; P = 0.01). Other cardiac morphologic parameters showed similar improvements. There also were significant decreases in systolic, diastolic, and mean blood pressure and mean number of antihypertensive medications.

Limitations: Paucity of randomized controlled trials.

Conclusions: Conversion from conventional to frequent or extended HD is associated with improvements in cardiac morphology and function, including LVMI and LV ejection fraction, respectively, and several blood pressure parameters, which collectively might confer long-term cardiovascular benefit. Trials with long-term clinical outcomes are needed.

Figure 1

Study selection flow diagram.

Figure 2

Forest plot displaying the effect of frequent or extended hemodialysis on change in the left ventricular mass index (LVMI). P < 0.001; I² index = 84% *The study by Chan et al reported 2 studies in the same years whereas Fagugli et al and Weinreich et al included 2 different HD prescriptions in the same study.

Figure 3

Univariate meta-regression analyses displaying the effect of frequent or extended hemodialysis on change in the left ventricular mass index (LVMI) stratified by duration of follow-up periods (3A), reader variability (3B), and cardiac imaging acquisition technique (3C). Where indicated, P values refer to the univariable meta-regression comparing studies of M-mode echocardiography vs. cardiac MRI († P = 0.009), unspecified mode echocardiography vs. cardiac MRI (‡ P < 0.001), and blinded vs. non-blinded readers (* P = 0.005).

Figure 4

Univariate meta-regression analyses displaying the effect of frequent of extended hemodialysis on change in the systolic blood pressure (SBP) stratified by the study measurement periods (4A), the mean number of anti-hypertensive medications (4B), and the blood pressure recording technique (4C).