By Samuel Nadler, MD, PhD

Critical Care, Pulmonary Medicine, The Polyclinic Madison Center, Seattle; Clinical Instructor, University of Washington, Seattle

Dr. Nadler reports no financial relationships relevant to this field of study.

SYNOPSIS: Initiation of continuous renal replacement therapy for patients with Kidney Disease: Improving Global Outcomes stage 2 renal failure reduced 90-day all-cause mortality.

SOURCE: Zarbock A, Kellum JA, Schmidt C, et al. Effect of early vs. delayed initiation of renal replacement therapy on mortality in critically ill patients with acute kidney injury: The ELAIN Randomized Clinical Trial. JAMA 2016;315:2190-2199.

The decision to start continuous renal replacement therapy (CRRT) in critically ill patients requires careful analysis of risks and benefits. The timing of initiating CRRT remains unclear. The ELAIN trial sought to inform this decision in a single center, randomized trial of early vs. late initiation of CRRT. Inclusion criteria were: Kidney Disease: Improving Global Outcomes (KDIGO) stage 2 acute kidney injury (AKI) (i.e., two-fold increase in serum creatinine or urine output < 0.5 mL/kg/hr for > 12 hours), neutrophil gelatinase-associated lipocalin (NGAL) > 150 ng/mL, 18-90 years of age, and one additional condition such as severe sepsis, vasopressor requirement, refractory fluid overload, PaO2/FiO2 < 300, or increase in Sequential Organ Failure Assessment (SOFA) score > 2. Patients who presented with chronic kidney disease, dialysis, previous AKI, pregnancy, or kidney transplantation were excluded.

Patients were randomized to early vs. late initiation of CRRT. Early CRRT started within eight hours of diagnosis of KDIGO stage 2 AKI. Delayed CRRT started within 12 hours of patients progressing to KDIGO stage 3 AKI (urine output < 0.3 mL/kg/hr for 24 hours and/or three-fold increase in serum creatinine or serum creatinine > 4 mg/dL with an acute increase of at least 0.5 mg/dL within 48 hours), or if any of the criteria for renal replacement therapy (RRT) were met: blood urea level > 100 mg/dL, potassium > 6 mEq/L and/or with ECG changes, magnesium > 8mEq/L, urine output < 200 mL per 12 hours, or organ edema resistant to diuretics. Once started, all patients received identical prescriptions for CRRT that continued until urine output exceeded 400 mL/24 hours without diuretic treatment or 2,100 mL/24 hours with diuretics.

The primary endpoint was 90-day mortality with secondary outcomes, including 28- and 60-day mortality, SOFA scores, recovery of renal function, need for hemodialysis after day 28, duration of CRRT, hospital length of stay (LOS), and biologic markers of inflammation. Overall, 231 patients were included in the intention-to-treat analysis, although 11 patients in the delayed group did not receive dialysis. Many patients were recruited after surgical procedures, including coronary artery bypass grafting (CABG), valve replacements, trauma, bowel resection, and liver transplantation. Most were mechanically ventilated and required vasopressors. Early initiation of CRRT was associated with decreased 90-day mortality (39.3% vs. 54.7%; odds ratio, 0.66; P = 0.03). In contrast, 28-day and 60-day mortality were not statistically different between the early and late groups, although there was a trend toward benefit for the early group. There were reductions in hospital LOS and duration of mechanical ventilation for those who were randomized to early initiation of CRRT.


The ELAIN trial was published just before the AKIKI study group published a similar study1 with contrasting results, and any analysis should consider both studies. The AKIKI study was a multicenter, randomized trial of 630 patients that examined when to start CRRT. In that trial, early initiation occurred within six hours of the diagnosis of stage 3 AKI while delayed initiation occurred if oliguria/anuria lasted for > 72 hours or severe electrolyte abnormalities occurred. In the AKIKI trial, no benefit was seen in 28- or 60-day mortality, although patients in the early initiation arm did exhibit higher rates of catheter-related infections.

Three factors may explain the differences between the ELAIN and AKIKI results. First, the patient populations in each study were different. As noted, much of the recruitment for the ELAIN trial involved post-surgical patients, while the AKIKI trial involved primarily medical patients. Although many co-morbidities were similar, surgical and medical ICU patients often follow different hospital courses. Many of the AKIKI patients in the delayed initiation group never received dialysis, and it is likely many of the patients in the early initiation group would not have required CRRT. In contrast, the inclusion of NGAL criteria in the ELAIN trial was intended to improve discernment for those who would ultimately require CRRT. Second, the ELAIN trial started CRRT sooner than AKIKI. The “delayed” group in ELAIN started at a similar time to the “early” group in AKIKI. A subgroup analysis of the ELAIN trial comparing patients in the delayed arm that started CRRT due to stage 3 disease (similar to AKIKI “early” group) and patients that required CRRT for electrolyte disturbances (similar to the AKIKI “delayed” group) found no difference in duration of CRRT, ICU, or hospital stay. Thus, earlier initiation of CRRT may have an effect on outcomes. Third, while ELAIN reported a 90-day mortality benefit for early initiation of CRRT, both studies failed to show 28- and 60-day mortality benefits. It is unclear why there was only a 90-day mortality benefit as neither trial seemed to show less dependence on long-term dialysis, although the ELAIN trial indicated less dependence when excluding patients who died during the trial.

Comparing the ELAIN and AKIKI trials informs the decision when to start CRRT in the ICU. In patients who will require CRRT, early initiation may have benefits. However, identifying these patients is challenging as commonly used endpoints such as KDIGO stage are not reliable predictors. Clinical judgment and patient-specific factors remain important in the risk-benefit analysis of patients presenting with AKI in the ICU.


  1. Gaudry S, Hajage D, Schortgen F, et al. Initiation strategies for renal-replacement therapy in the intensive care unit. N Engl J Med 2016;375:122-133.