May 31, 2019

EJRC - Debating the ideal time of RRT

An RCT to address the question when to initiate in patients in septic patients suffering from severe AKI


The ideal time point of initiation of Renal Replacement Therapy (RRT) is still a matter of debate. To address the question, on when to initiate in patients in septic patients suffering from severe acute kidney injury (AKI), Barbar et al. performed a randomised, controlled, open-label, multi-centre trial (IDEAL-ICU trial). The trial was performed in multiple French hospitals.

Patients with early stage of septic shock with AKI in the failure stage of the RIFLE-criteria were randomised into an early (within 12 hours after reaching failure stage) and delayed (after 48 hours delay) strategy of initiation of RRT. After randomising 488 patients, the trial was stopped early for futility.

There was no significant difference in the primary outcome of mortality between patients in the early and delayed strategy of RRT initiation (58% vs. 54%; p=0.38). While in the delayed strategy group, 17% of the patients required emergency initiation of RRT, 38% in this group did not receive RRT at all. Of the patients not requiring RRT in this group, 75.3% did not require it because of spontaneous recovery of renal function.

Unsurprisingly, the number of days free of RRT was significantly higher in the delayed group than in the early group (median days; 16 vs. 12; p=0.006). There were no significant differences in the other secondary outcomes (mortality at 28 days and 180 days, number of days free of mechanical ventilation and vasopressors, length of ICU and hospital stay and rate of dependence on RRT at 28 days).



This was a well-planned and conducted study randomising a significant number of patients. However, some limitations must be noted. The authors used the RIFLE classification instead of the newer KDIGO AKI classification. This, as is noted by the authors, is due to the fact that RIFLE was the most commonly used definition at the time the trial was designed. Furthermore, reaching of the failure stage is not necessarily an indication for the initiation of RRT.

In the second interim analysis, the data safety monitoring board – based on a sample size reassessment – determined that 9,669 patients per arm would be required to accurately compare early versus delayed initiation of RRT and therefore recommended to terminate the trial for futility.


  • In patients with septic shock and AKI, delayed initiation of RRT did not have an impact on mortality. However, it did allow a significant number of patients to recover renal function without ever requiring RRT.
  • Significantly more patients in the delayed group developed hyperkalaemia (10 vs. 0 patients; p=0.03). When applying a delayed strategy, patients should be closely monitored for electrolyte disbalances.


This article review was prepared and submitted by Sebastian J Klein, Division of Intensive Care and Emergency Medicine, Department of Internal Medicine, Medical University Innsbruck, on behalf of the ESICM AKI Section and the Journal Review Club.



1) Barbar, S.D., et al., Timing of Renal-Replacement Therapy in Patients with Acute Kidney Injury and Sepsis. N Engl J Med, 2018. 379(15): p. 1431-1442.

2) Forni, L.G. and M. Joannidis, IDEAL timing of renal replacement therapy in critical care. Nat Rev Nephrol, 2019. 15(1): p. 5-6.

3) Klein, S.J., M. Joannidis, and L. Forni, Sepsis: early interventions count but not RRT! Journal of Thoracic Disease, 2019. 11(Suppl 3): p. S325-S328.

Comment on this news