Timing of renal replacement therapy initiation for acute kidney injury in critically ill patients: a systematic review of randomized clinical trials with meta-analysis and trial sequential analysis

Abstract

Abstract Background Acute kidney injury (AKI) is a common serious complication in critically ill patients. AKI occurs in up to 50% patients in intensive care unit (ICU), with poor clinical prognosis. Renal replacement therapy (RRT) has been widely used in critically ill patients with AKI. However, in patients without urgent indications such as acute pulmonary edema, severe acidosis, and severe hyperkalemia, the optimal timing of RRT initiation is still under debate. We conducted this systematic review of randomized clinical trials (RCTs) with meta-analysis and trial sequential analysis (TSA) to compare the effects of early RRT initiation versus delayed RRT initiation. Methods We searched databases (PubMed, EMBASE and Cochrane Library) from inception through to July 20, 2020, to identify eligible RCTs. The primary outcome was 28-day mortality. Two authors extracted the data independently. When the I2 values < 25%, we used fixed-effect mode. Otherwise, the random effects model was used as appropriate. TSA was performed to control the risk of random errors and assess whether the results in our meta-analysis were conclusive. Results Eleven studies involving 5086 patients were identified. Two studies included patients with sepsis, one study included patients with shock after cardiac surgery, and eight others included mixed populations. The criteria for the initiation of RRT, the definition of AKI, and RRT modalities existed great variations among the studies. The median time of RRT initiation across studies ranged from 2 to 7.6 h in the early RRT group and 21 to 57 h in the delayed RRT group. The pooled results showed that early initiation of RRT could not decrease 28-day all-cause mortality compared with delayed RRT (RR 1.01; 95% CI 0.94–1.09; P = 0.77; I2 = 0%). TSA result showed that the required information size was 2949. The cumulative Z curve crossed the futility boundary and reached the required information size. In addition, early initiation of RRT could lead to unnecessary RRT exposure in some patients and was associated with a higher incidence of hypotension (RR 1.42; 95% CI 1.23–1.63; P < 0.00001; I2 = 8%) and RRT-associated infection events (RR 1.34; 95% CI 1.01–1.78; P = 0.04; I2 = 0%). Conclusions This meta-analysis suggested that early initiation of RRT was not associated with survival benefit in critically ill patients with AKI. In addition, early initiation of RRT could lead to unnecessary RRT exposure in some patients, resulting in a waste of health resources and a higher incidence of RRT-associated adverse events. Maybe, only critically ill patients with a clear and hard indication, such as severe acidosis, pulmonary edema, and hyperkalemia, could benefit from early initiation of RRT.