Abstract: Objective To compare clinical effects of intermittent high volume predilution on-line hemofiltration (HF) and standard hemodialysis (HD) in the treatment of critically ill patients with acute kidney injury (AKI). Methods This was a prospective, randomized controlled, and single-centered clinical study. Mortality and recovery of kidney function were compare in critically ill patients with AKI treated with HF (n=50) and HD (n=45). The outcome included all-cause mortality and in- hospital all-cause mortality, and recovery rate from AKI within 60 days. Results Baseline characteristics were similar in the HF group and HD group.
All-cause mortality within 60 days was 68.0% (34/50) in the HF group and 82.2% (37/45) in the HD group (hazard ratio 0.75; 95% confidence interval 0.80～1.28; P=0.036), while in- hospital mortality and recovery rate from AKI were statistically indifferent between the two groups (P=0.274 and 0.565, respectively). The recovery time from AKI and the ratio of patients required dialysis were lower in HF group than in HD group but without statistical significance (P=0.223 and 0.687, respectively). Conclusion HF was better than HD in lowering mortality in critically ill patients with AKI. HF also reduced the recovery time from AKI and the requirement for dialysis support, but without statistical significances as compare with those in HD group.

Key words: acute kidney injury, hemodialysis, intermittent high-volume predilution on-line haemofiltration, mortality