Comparison of hemodialysis, hemofiltration, and acetate-free biofiltration for ESRD: Systematic review

Abstract

We performed a systematic review of randomized controlled trials (RCTs) comparing hemodialysis (HD), hemofiltration (HF), hemodiafiltration (HDF), and acetate-free biofiltration (AFB) in the treatment of patients with end-stage renal disease to assess their clinical effectiveness. The Cochrane CENTRAL Registry, MEDLINE, EMBASE, CINAHL, the American College of Physicians Database, Database of Abstracts of Reviews of Effectiveness, and reference lists were searched for randomized trials of HF, HDF, and AFB compared with HD; HDF compared with AFB; and HF compared with HDF. Two reviewers extracted data for all-cause mortality; hypotension, headache, nausea, vomiting, and any other adverse symptoms; quality of life (QoL); hospitalization; dialysis adequacy; and end-of-treatment beta 2 -microglobulin levels. Analysis was by means of a random-effects model, and results are expressed as relative risk (RR) and weighted mean difference (WMD) with 95% confidence intervals (CIs). Eighteen eligible trials (588 patients) were identified. HDF was associated with significantly greater mortality risk than HD (4 trials, 326 patients; RR, 3.52; 95% CI, 1.37 to 9.47). Risk for mortality was not different among the other comparisons. Risks for hypotension episodes and dialysis-related symptoms were not significantly different with HD, HF, HDF, and AFB (18 trials, 583 patients). QoL, assessed by using an unvalidated scoring tool, appeared to be significantly improved in patients on HDF therapy than those on HD therapy (1 trial, 67 patients; WMD, 0.6; 95% CI, 0.3 to 0.9), but this was not evident when validated QoL assessment tools were used. Use of AFB compared with HDF was not associated with a significant difference in risk for hospitalization (1 trial, 11 patients; WMD, -0.45; 95% CI, -1.42 to 0.52). HDF in comparison to HD did not reduce the risk for carpal tunnel syndrome (1 trial, 67 patients; RR, 2.04; 95% CI, 0.59 to 7.00). Kt/V was significantly different with HDF compared with HD (3 trials, 124 patients; WMD, 0.14; 95% CI, 0.05 to 0.22). No other substantial data for these interventions and their impact on major patient-centered outcomes were available. The trials assessed were not powered adequately and had suboptimal method quality. It is not possible on the basis of effectiveness to prefer one extracorporeal renal replacement therapy modality to the other for end-stage kidney disease because significant differences in clinically important outcomes have not been shown by available published RCTs.