Continuous renal-replacement therapy beat intermittent haemodialysis on survival

Design

Multicentre, open-label, randomised noninferiority trial (ICRAKI), stopped early

Population

554 critically ill adults with acute kidney injury and an indication for renal-replacement therapy

Primary outcome

Major adverse kidney event by day 90 (death, dialysis dependence or creatinine over 25% above baseline)

Effect

78.1% vs 71.6% (adjusted difference 6.28 points, 95% CI -1.22 to 13.79); day-90 mortality HR 1.30 (95% CI 1.04 to 1.63)

ICRAKI was a French multicentre, open-label, noninferiority trial in critically ill adults with acute kidney injury and an indication for renal-replacement therapy. They were randomised to start with intermittent haemodialysis (272) or continuous renal-replacement therapy (282). The primary outcome was a major adverse kidney event by day 90: death, continued dependence on renal-replacement therapy, or creatinine more than 25% above baseline.

The trial was stopped early because interim analyses showed a sustained signal of higher mortality with intermittent dialysis. In the intention-to-treat population (535 patients) major adverse kidney events occurred in 78.1% with intermittent and 71.6% with continuous treatment (adjusted absolute difference 6.28 percentage points, 95% CI -1.22 to 13.79), so noninferiority was not shown. Death by day 90 occurred in 63.1% vs 53.5% (Kaplan-Meier hazard ratio 1.30, 95% CI 1.04 to 1.63). Hypotension was common and similar in both groups (68.8% vs 66.5%).

The result favours continuous therapy as the starting approach in unstable, critically ill patients. Where continuous machines or trained staff are not available, the trial does not say that intermittent dialysis should be withheld.

• Starting with intermittent haemodialysis was not shown to be noninferior to continuous therapy for 90-day major adverse kidney events.
• Day-90 mortality was 63.3% with intermittent vs 53.6% with continuous therapy (HR 1.30, 95% CI 1.04 to 1.63).
• Hypotension was frequent in both arms (about two thirds), so it does not separate the modalities.
• Consider continuous therapy as the initial modality in the sickest, haemodynamically unstable ICU patients where it is available.
• The trial stopped early, which can overstate effects, so read the size of the mortality gap with caution.

Why it matters

It challenges the idea that the two modalities are interchangeable at the start of treatment.

Don't overread it

The trial was open-label and stopped early, and mortality was a secondary signal rather than the primary endpoint.

The statistics, in plain English

The main outcome combined death, dialysis dependence and persistent creatinine rise. Its confidence interval (-1.22 to 13.79) includes zero, so that comparison alone is not conclusive; the trial's stated finding is that noninferiority was not shown. The mortality hazard ratio (1.30, lower bound 1.04) is statistically significant but only just, and early stopping tends to exaggerate effects.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.