The effect of on-line high-flux hemofiltration versus low-flux hemodialysis on mortality in chronic kidney failure: a small randomized controlled trial

Abstract

Background: Given the paucity of prospective randomized controlled trials assessing comparative performances of different dialysis techniques, we compared on-line high-flux hemofiltration (HF) with ultrapure low-flux hemodialysis (HD), assessing survival and morbidity in patients with end-stage renal disease (ESRD).

Study design: An investigator-driven, prospective, multicenter, 3-year-follow-up, centrally randomized study with no blinding and based on the intention-to-treat principle.

Setting & participants: Prevalent patients with ESRD (age, 16 to 80 years; vintage > 6 months) receiving renal replacement therapy at 20 Italian dialysis centers.

Interventions: Patients were centrally randomly assigned to HD (n = 32) or HF (n = 32).

Outcomes & measurements: All-cause mortality, hospitalization rate for any cause, prevalence of dialysis hypotension, standard biochemical indexes, and nutritional status. Analyses were performed using the multivariate analysis of variance and Cox proportional hazard method.

Results: There was significant improvement in survival with HF compared with HD (78%, HF versus 57%, HD) at 3 years of follow-up after allowing for the effects of age (P = 0.05). End-of-treatment Kt/V was significantly higher with HD (1.42 +/- 0.06 versus 1.07 +/- 0.06 with HF), whereas beta(2)-microglobulin levels remained constant in HD patients (33.90 +/- 2.94 mg/dL at baseline and 36.90 +/- 5.06 mg/dL at 3 years), but decreased significantly in HF patients (30.02 +/- 3.54 mg/dL at baseline versus 23.9 +/- 1.77 mg/dL; P < 0.05). The number of hospitalization events for each patient was not significantly different (2.36 +/- 0.41 versus 1.94 +/- 0.33 events), whereas length of stay proved to be significantly shorter in HF patients compared with HD patients (P < 0.001). End-of-treatment body mass index decreased in HD patients, but increased in HF patients. Throughout the study period, the difference in trends of intradialytic acute hypotension was statistically significant, with a clear decrease in HF (P = 0.03).

Limitations: This is a small preliminary intervention study with a high dropout rate and problematic generalizability.

Conclusion: On-line HF may improve survival independent of Kt/V in patients with ESRD, with a significant decrease in plasma beta(2)-microglobulin levels and increased body mass index. A larger study is required to confirm these results.