This trial randomised adults with acute hypercapnic respiratory failure from COPD exacerbation, who had failed or were ineligible for extubation within 24 hours of intubation, to extracorporeal CO2 removal or continued standard ventilation. The primary endpoint was death or severe disability at day 60.
It was terminated early by the sponsor for financial reasons after 18 of a planned 192 patients. The primary endpoint occurred in 0 of 8 treated patients against 3 of 9 evaluable controls (33%), a risk difference of -33% with a 95% CI from -65% to 6%. Ventilation duration was shorter with CO2 removal, at 7.1 against 24.3 days (p = 0.043), with more device-free days on post-hoc analysis. Ventilator-associated pneumonia occurred in none against three patients. One treated patient had severe bleeding.
This is worth knowing about and worth acting on not at all. Eighteen patients is under a tenth of the planned enrolment, the primary endpoint interval crosses zero, and the ventilation-duration finding rests on nine patients per arm where a single prolonged wean moves the mean. A trial stopped for funding rather than for futility or harm carries no information about the intervention itself, which is precisely why the result should not be read as encouraging or discouraging. Adequately powered trials are needed and this does not substitute for one.
- Terminated by the sponsor for financial reasons at 18 of 192 planned patients
- Primary endpoint 0/8 vs 3/9; interval -65% to +6%, crossing zero
- Ventilation duration 7.1 vs 24.3 days, on nine patients per arm
- Not a basis for practice change in either direction
The statistics, in plain English
With eight and nine patients per arm, a single event shifts the result by more than 10 percentage points, which is why the confidence interval spans 71 points and includes both large benefit and modest harm. The significant p-value on ventilation duration, 0.043, is fragile in the same way: means from tiny groups are dominated by outliers, and one patient ventilated for 60 days can create it. Early termination for non-clinical reasons leaves an underpowered trial, not a negative or positive one.
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.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for emergency & critical care, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free