No mortality difference between early and delayed initiation of renal replacement therapy in critically ill patients with AKI

Référence

Gaudry S, Hajage D, Benichou N, et al. Delayed versus early initiation of renal replacement therapy for severe acute kidney injury: a systematic review and individual data meta-analysis of randomised clinical trials. Lancet 2020;395(10235):1506-1515.

Sommaire

These investigators searched MEDLINE, EMBASE, and the Cochrane Central Register of Controlled Trials, as well as conference proceedings and reference lists of selected studies to find randomized controlled trials that examined the effect of early versus delayed RRT initiation in critically ill adult patients who had severe AKI without life-threatening complications such as severe acidosis or pulmonary edema. Two investigators independently selected studies and assessed for risk of bias. The definitions of early and delayed initiation strategies varied among the studies. For most studies, early initiation occurred 2 hours to 8 hours after randomization. Delayed initiation ranged from 25 hours to 57 hours after randomization; in some cases, delayed RRT was only initiated if a fixed criterion, such as reaching a more severe stage of AKI, was achieved. The authors performed a meta-analysis using individual patient data obtained from 9 of the 10 included studies (N = 1879; 946 delayed RRT patients and 944 early RRT patients). Overall, 42% of the patients assigned to the delayed group ultimately did not receive RRT, presumably because of renal recovery or death prior to initiation. Despite this, there was no significant between-group difference in the number of RRT-free days. For the 1664 patients with data available regarding 28-day mortality, the rates did not significantly differ between groups (44% in the delayed group vs 43% in the early group; P = .80). Additionally, there were no significant differences in length of hospital stay, RRT dependence on discharge, in-hospital mortality, 60-day or 90-day mortality, or adverse events such as severe hyperkalemia or cardiac arrhythmias. There was no heterogeneity across studies and no evidence of publication bias.