Acute kidney injury (AKI) is a common yet possibly fatal complication among critically ill patients in intensive care units (ICU). Although renal replacement therapy (RRT) is an important supportive management for severe AKI patients, the optimal timing of RRT initiation for these patients is still unclear.
In this systematic review, we searched all relevant randomized controlled trials (RCTs) that directly compared accelerated with standard initiation of RRT from PUBMED, MEDLINE, EMBASE, and Cnki.net published prior to July, 20, 2020. We extracted study characteristics and outcomes of being free of dialysis, dialysis dependence and mortality. We rated the certainty of evidence according to Cochrane methods and the GRADE approach.
We identified 56 published relevant studies from 1071 screened abstracts. Ten RCTs with 4753 critically ill AKI patients in intensive care unit (ICU) were included in this meta-analysis. In our study, accelerated and standard RRT group were not associated with all-cause mortality (log odds-ratio [OR]: - 0.04, 95% confidence intervals [CI] - 0.16 to 0.07, p = 0.46) and free of dialysis (log OR: - 0.03, 95% CI - 0.14 to 0.09, p = 0.65). In the subgroup analyses, accelerated RRT group was significantly associated with lower risk of all-cause mortality in the surgical ICU and for those who received continuous renal replacement therapy (CRRT). In addition, patients in these two subgroups had higher chances of being eventually dialysis-free. However, accelerated initiation of RRT augmented the risk of dialysis dependence in the subgroups of patients treated with non-CRRT modality and whose Sequential Organ Failure Assessment (SOFA) score were more than 11.
In this meta-analysis, critically ill patients with severe AKI would benefit from accelerated RRT initiation regarding all-cause mortality and being eventually free of dialysis only if they were surgical ICU patients or if they underwent CRRT treatment. However, the risk of dialysis dependence was increased in the accelerated RRT group when those patients used non-CRRT modality or had high SOFA scores. All the literatures reviewed in this study were highly heterogeneous and potentially subject to biases. Trial registration CRD42020201466, Sep 07, 2020. https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=201466 ....

Accelerated versus standard initiation of renal replacement therapy for critically ill patients with acute kidney injury: a systematic review and meta-analysis of RCT studies

10.1186/s13054-020-03434-z