Earlier-start versus usual-start dialysis in patients with community-acquired acute kidney injury: a randomized controlled trial.

Background Optimum timing of the initiation of dialysis therapy in acute kidney injury is not clear. Study Design Prospective, open label, 2-arm, randomized, controlled trial. Setting & Participants 208 adults with acute kidney injury with progressively worsening azotemia at the artificial kidney dialysis unit of a tertiary-care referral center in western India. Intervention Earlier-start dialysis was initiated when serum urea nitrogen and/or creatinine levels increased to 70 and 7 mg/dL, respectively, whereas the usual-start dialysis patients (control group) received dialysis when clinically indicated as judged by treating nephrologists. Outcomes Primary outcome was in-hospital mortality and dialysis dependence at 3 months. Secondary outcome in patients receiving dialysis was time to recovery of kidney function, computed from time of enrollment to the last dialysis session. Results Of 585 screened patients, 102 were assigned to earlier-start dialysis, and 106 to usual-start dialysis. Baseline characteristics were similar between randomized groups. 93 (91.1%) and 88 (83.1%) participants received dialysis in the intervention and control groups, respectively. Mean serum urea nitrogen and serum creatinine levels at dialysis therapy initiation were 71.7 ± 21.7 (SD) and 7.4 ± 5.3 mg/dL, respectively, in the intervention group versus 100.9 ± 32.6 and 10.41 ± 3.3 mg/dL in the control group. Data on primary outcome were available for all patients. In-hospital mortality was 20.5% and 12.2% in the intervention and control groups, respectively (relative risk, 1.67; 95% CI, 0.88-3.17; P  = 0.2). 4.9% and 4.7% of patients in the intervention and control groups, respectively, were dialysis dependent at 3 months (relative risk, 1.04; 95% CI, 0.29-3.7; P  = 0.9). Limitations Study was not double blind, event rate (ie, mortality) was less than predicted, wide CIs preclude definitive findings. Conclusions Our data do not support the earlier initiation of dialysis therapy in community-acquired acute kidney injury.