- Design
- Single-centre retrospective cohort
- Population
- 185,455 adults having noncardiac surgery under ventilated GA, France
- Primary outcome
- In-hospital mortality
- Effect
- Adjusted OR 1.63 (1.36–1.86) per 5 mmHg lower end-tidal CO2
This single-centre French retrospective cohort included 185,455 adults having noncardiac surgery under ventilated general anaesthesia between 2010 and 2020. In-hospital mortality was 0.85%. Investigators asked whether low end-tidal CO2 still predicted death after adjusting for minute ventilation and the depth and duration of hypotension.
It did. Each 5 mmHg fall in mean end-tidal CO2 below the median was associated with higher odds of death (adjusted OR 1.63, 95% CI 1.36–1.86), with a non-linear relationship. There was no interaction with hypotension, and the result held across sensitivity analyses.
End-tidal CO2 reflects cardiac output and dead space as well as ventilation. A persistently low value with unchanged ventilation may be an early sign of low flow that blood pressure does not yet show. This is association, not proof that targeting a higher value helps, but it argues for treating an unexplained low reading as a haemodynamic clue rather than a ventilator setting.
- Treat an unexplained fall in end-tidal CO2 at constant ventilation as a possible low-output state
- Check for hypovolaemia, embolism, air leak and sampling problems when end-tidal CO2 drops
- Avoid routine hyperventilation that drives end-tidal CO2 down without a reason
- Consider low intraoperative end-tidal CO2 when flagging patients for closer postoperative review
Why it matters
Challenges the habit of reading the capnograph only as a ventilation monitor.
Don't overread it
This retrospective association does not show that raising end-tidal CO2 lowers mortality.
The statistics, in plain English
An adjusted odds ratio of 1.63 per 5 mmHg means each such fall was associated with about 60% higher odds of in-hospital death, after accounting for ventilation and hypotension. Because baseline mortality was under 1%, the absolute increase for most patients is small, and residual confounding by illness severity remains possible.
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 anaesthesiology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free