Rationale & Objective:
Uremic symptoms and frailty in kidney failure may be driven by the accumulation of uremic toxins that are variably cleared by dialysis. We conducted a systematic review and meta-analysis of randomized controlled trials (RCTs) that evaluated the effect of hemodiafiltration (HDF) compared to hemodialysis (HD) on uremic symptoms, physical function, and quality of life (QOL).
Study Design:
Systematic review and meta-analysis.
Setting & Study Populations:
We searched Medline, EMBASE, Cochrane Library, CINAHL, and PROSPERO up to September 2024 for RCTs comparing maintenance HDF vs HD in adults with kidney failure.
Selection Criteria:
RCTs with the following outcomes: uremic symptoms as measured by patient-reported outcome measures (PROMs), physical function, and QOL.
Data Extraction:
Two reviewers independently screened abstracts and full texts, extracted data, and assessed risk of bias.
Analytical Approach:
Random effects meta-analysis.
Results:
Of 1,226 identified articles, 117 were included for full-text review, of which 24 RCTs (16 parallel, 8 crossover) met inclusion criteria. Studies were mostly from Europe or Asia and recruited a total of 2,382 participants. HDF was primarily delivered for 4 hours, 3 times weekly, in post-dilution mode (15/24 studies). The prescribed effluent dose was reported in 14 of 24 studies and ranged from 15-24 L. Uremic symptoms including cramps, fatigue, gastrointestinal symptoms, headache, neuropathy, pain, itch, restless legs, and sleep disturbance were variably reported using general or disease-specific PROMs. In meta-analyses, HDF did not improve sleep, itch, pain, fatigue, or cramps compared with HD. Meta-analyses for physical function and QOL were not performed.
Limitations:
RCTs were not representative of the general dialysis population, poor effluent dose reporting, lack of blinding, variability in PROMs and heterogeneity.
Conclusions:
The evidence for HDF over HD for uremic symptoms and physical function remains limited due to the variability in HDF/HD prescriptions and PROMs used in clinical trials, risk of bias, and heterogeneity in treatment effects. Large, adequately powered, high-quality RCTs that use standardized, validated, reliable, and responsive PROMs are needed.
Registration:
The protocol was registered on PROSPERO (CRD42022316562).