> ## Content Index
> Fetch the complete content index at: https://nephspace.com/llms.txt
> Use this file to discover other available public pages before exploring further.

# Intermittent or continuous dialysis in the ICU: the ICRAKI trial stopped early after more deaths with intermittent hemodialysis
- URL: https://nephspace.com/icraki-intermittent-continuous-dialysis-icu/
- Published: 2026-10-20T15:14:25.000Z
- Updated: 2026-10-20T15:14:25.000Z
- Description: The trial set out to show intermittent hemodialysis was no worse than continuous therapy and couldn't. Patients who started on intermittent hemodialysis had higher 90-day mortality, and the trial stopped at about half its planned size. The size of the gap is uncertain; its direction is harm.
- Author: Dr Wael Hussein
- Tags: AKI and critical care, #story, #has-audio, #research, #full-story, Stories

**🎧 Listen:** [Weekly · Oct 20](https://nephspace.com/weekly-oct-20/), from 01:27.

### The background

When a critically ill patient with acute kidney injury needs dialysis, ICUs choose between intermittent hemodialysis (a session of a few hours, repeated as needed) and continuous therapy run around the clock. Earlier trials were small and mostly more than 15 years old, and meta-analyses found no consistent survival difference. The KDIGO guideline gives a weak suggestion to favor continuous therapy in hemodynamically unstable patients. So many of us thought the question was settled: pick whichever method your unit does best. Intermittent hemodialysis also has practical appeal. As the editorial notes, it corrects the chemistry and clears toxins faster, and one machine can treat several patients in turn; the trial's authors add that, after the first few days, it could make it easier to get patients moving.

### What the trial did

ICRAKI was an investigator-initiated, open-label, randomized noninferiority trial in 38 French ICUs (October 2023 to July 2025), funded by the French Ministry of Health. It enrolled critically ill adults with AKI, at least one recognized indication to start dialysis, and invasive ventilation, catecholamines or both. Patients with conditions that favor one method were excluded, such as brain injury or raised intracranial pressure, and poisoning with a dialyzable toxin. Patients started on either intermittent hemodialysis or continuous therapy and stayed on the assigned method for at least the first four days; switching was allowed after that. The primary outcome was a major adverse kidney event by day 90: death, dependence on dialysis, or a creatinine more than 25% above baseline. The trial was designed to show that intermittent hemodialysis was no worse, within 7.5 percentage points.

### What they found

- The trial was stopped early on the safety board's recommendation, after interim looks at 28-day mortality showed a survival difference favoring continuous therapy. 554 of a planned 1,000 patients were randomized; 535 were analyzed.
- Most patients were very sick: about 84% were ventilated and about 79% were on catecholamines on the day of randomization.
- Primary outcome: 78.1% with intermittent hemodialysis vs 71.6% with continuous therapy (adjusted difference 6.3 percentage points; 95% CI, −1.2 to 13.8). Noninferiority was not shown, and the difference favoring continuous therapy was not statistically significant.
- Death by day 90, a secondary outcome: 63.1% vs 53.5% (Kaplan–Meier estimates 63.3% vs 53.6%; hazard ratio 1.30; 95% CI, 1.04 to 1.63), roughly one extra death for every ten patients in this trial (a crude difference, likely overstated by the early stop). The whole confidence interval lies on the side of harm. Confidence intervals for secondary outcomes were not adjusted for multiple comparisons.
- Dialysis dependence at day 90 was rare in both groups (2 vs 3 patients). In post hoc exploratory analyses, transient catecholamine escalations were more frequent on dialysis days with intermittent hemodialysis, and daily catecholamine doses were higher in that group (hypothesis-generating).

### What the editorial adds

In an accompanying editorial, Dr. Paul Palevsky notes that early termination left the trial underpowered and that the mortality difference was unexpected given earlier trials. Overall mortality was higher than in older trials, which may reflect the newer, stricter criteria for starting dialysis. He asks whether we are asking the right question: in practice the choice isn't random. Some health systems use continuous therapy for almost all ICU patients; others reserve it for the most unstable. The real question, he argues, is how best to combine the available therapies.

### Why it matters

The trial set out to show that intermittent hemodialysis was no worse, and it couldn't show that. Its best estimate is about a 30% higher risk of death with intermittent hemodialysis, and because the trial stopped early, the true difference may be smaller. The 95% confidence interval runs from about 4% to about 63% higher. It wasn't adjusted for multiple comparisons, and stopping early can exaggerate the gap, but the whole interval sits on the side of harm. Our take: for the sickest ICU patients with a clear indication for dialysis, at least in the first few days, intermittent hemodialysis is likely not the safer option, and this trial leans toward starting continuous therapy. After the first few days, it doesn't tell us much.

### Caveats

The authors write that stopping early can overstate the size of an effect, so the direction of the mortality difference is more reliable than its magnitude. The trial was open-label, the delivered intermittent dose was slightly below target, switching after day 3 means it doesn't say which method is better after the first few days, and the early stop left the kidney outcomes underpowered.

Funding and ties: funded by the French Ministry of Health; no industry funding was reported. Authors' disclosure forms are available at NEJM.org.

---

**The source:** Gaudry S, Boubaya M, Louis G, et al., for the ICRAKI Trial Group. Intermittent or Continuous Renal-Replacement Therapy in Acute Kidney Injury. N Engl J Med. Published online October 14, 2026\. doi:10.1056/NEJMoa2608397 [Read the original](https://doi.org/10.1056/NEJMoa2608397?ref=nephspace.com)

**Also discussed:**

- Palevsky PM. Continuous or Intermittent Renal-Replacement Therapy — What Is the Question? N Engl J Med. Published online October 14, 2026\. doi:10.1056/NEJMe2612910 [Link](https://doi.org/10.1056/NEJMe2612910?ref=nephspace.com)

**🎧 Listen:** [Weekly · Oct 20](https://nephspace.com/weekly-oct-20/), from 01:27.

*Physician-led, AI-assisted. Dr. Wael Hussein chooses every item NephSpace covers and reviews and edits every story before it is published. The first draft was written in our own words with AI (Claude, by Anthropic) from the full text of the paper and checked against our reading notes; Dr. Hussein then reviewed and edited it. For education and information only, not medical advice. Please talk to your own physician or care team about your health. Clinicians: read the original before changing practice. See our* [*disclaimer*](https://nephspace.com/disclaimer/) *and* [*disclosures*](https://nephspace.com/disclosures/)*. Spotted an error?* [*Tell us*](mailto:hello@nephspace.com?subject=Correction)*.*