A randomized clinical trial of high volume peritoneal dialysis versus extended daily hemodialysis for acute kidney injury patients

Abstract

Background: Acute kidney injury (AKI) requiring dialysis in critically ill patients is associated with an in-hospital mortality rate of 50-80 %. Extended daily hemodialysis (EHD) and high volume peritoneal dialysis (HVPD) have emerged as alternative modalities.

Methods: A double-center, randomized, controlled trial was conducted comparing EHD versus HVPD for the treatment for AKI in the intensive care unit (ICU). Four hundred and seven patients were randomized and 143 patients were analyzed. Principal outcome measure was hospital mortality, and secondary end points were recovery of renal function and metabolic and fluid control.

Results: There was no difference between the two groups in relation to median ICU stay [11 (5.7-20) vs. 9 (5.7-19)], recovery of kidney function (26.9 vs. 29.6 %, p = 0.11), need for chronic dialysis (9.7 vs. 6.5 %, p = 0.23), and hospital mortality (63.4 vs. 63.9 %, p = 0.94). The groups were different in metabolic and fluid control. Blood urea nitrogen (BUN), creatinine, and bicarbonate levels were stabilized faster in EHD group than in HVPD group. Delivered Kt/V and ultrafiltration were higher in EHD group. Despite randomization, there were significant differences between the groups in some covariates, including age, pre-dialysis BUN, and creatinine levels, biased in favor of the EHD. Using logistic regression to adjust for the imbalances in group assignment, the odds of death associated with HVPD was 1.4 (95 % CI 0.7-2.4, p = 0.19). A detailed investigation of the randomization process failed to explain the marked differences in patient assignment.

Conclusions: Despite faster metabolic control and higher dialysis dose and ultrafiltration with EHD, this study provides no evidence of a survival benefit of EHD compared with HVPD. The limitations of this study were that the results were not presented according to the intention to treat and it did not control other supportive management strategies as nutrition support and timing of dialysis initiation that might influence outcomes in AKI.