Intermittent Dialysis Was Not Noninferior to Continuous Therapy in Critically Ill Patients With AKI. The Trial Stopped Early.
In the 554-patient ICRAKI trial in NEJM, death by day 90 occurred in 63.1 percent of patients started on intermittent hemodialysis and 53.5 percent of those started on continuous therapy. The authors urge caution because the trial stopped early.
A randomized trial published online Oct. 14 in the New England Journal of Medicine found that starting critically ill adults with acute kidney injury on intermittent hemodialysis was not shown to be noninferior to starting them on continuous renal-replacement therapy (CRRT). The trial was stopped early after sequential interim analyses showed a sustained signal of higher mortality with intermittent dialysis, according to the published abstract.
Many ICUs choose between the two modalities by habit, equipment and staffing rather than by trial data. The trial's registry entry notes that earlier randomized comparisons were done more than 15 years ago, before recent trials reshaped when renal-replacement therapy is started. This one tested the choice directly against a patient-centered composite outcome.
What the trial showed
The paper, "Intermittent or Continuous Renal-Replacement Therapy in Acute Kidney Injury", is led by Stéphane Gaudry of Assistance Publique-Hôpitaux de Paris for the ICRAKI Trial Group. It was a multicenter, open-label, noninferiority trial in French intensive care units (NCT06032884), funded by the French Ministry of Health. It enrolled critically ill adults with acute kidney injury who had an indication for renal-replacement therapy: 272 were assigned to intermittent hemodialysis and 282 to CRRT as the initial treatment, and 535 were included in the intention-to-treat analysis. The registry entry had planned for 1,000 patients.
The primary outcome was a major adverse kidney event by day 90, defined as death, dependence on renal-replacement therapy or a serum creatinine more than 25 percent above baseline. It occurred in 78.1 percent of the intermittent group and 71.6 percent of the continuous group, an adjusted absolute risk difference of 6.28 percentage points (95% CI, -1.22 to 13.79). The authors concluded that intermittent hemodialysis was not shown to be noninferior.
Mortality and the early stop
Death by day 90 occurred in 63.1 percent of patients started on intermittent hemodialysis and 53.5 percent of those started on CRRT. Kaplan-Meier estimates were 63.3 percent and 53.6 percent (hazard ratio, 1.30; 95% CI, 1.04 to 1.63). Hypotension, the most common complication potentially related to the kidney injury or the therapy, occurred in 68.8 percent and 66.5 percent.
"Because the trial was stopped early, the magnitude of the mortality difference should be interpreted with caution," the authors write, as quoted by Medical Xpress. Trials halted for an apparent harm signal tend to overestimate effect size, and the confidence interval on the primary composite still includes no difference.
A stopped-early, open-label trial is a reason to review a default, not to rewrite a protocol overnight.
Limits
The design was open-label, so clinicians knew the assigned modality. All participating centers were in France, so applicability to units with different staffing, hybrid therapies such as prolonged intermittent dialysis or different thresholds for starting therapy is uncertain. The trial compared initial modality; it does not say when, or whether, patients should be switched.
What it means in clinic
NEJM paired the paper with an editorial by Paul M. Palevsky titled "Continuous or Intermittent Renal-Replacement Therapy — What Is the Question?", a sign that the field will debate how to read the result. Unit directors can start with a narrower question: who in your ICU is routinely started on intermittent dialysis, and was that a deliberate choice? For another recent critical care trial with a sobering mortality result, see our report on inhaled sedation in hypoxemic respiratory failure.
Reported, fact-checked and edited by The Script Pad's editorial staff.
This article is for professional education and does not replace clinical judgment. Treatment decisions should be based on the individual patient and current guidelines.
