Closed-loop oxygen keeps ward patients in their saturation range a fifth more of the time

Design

Systematic review and meta-analysis of 7 randomised trials

Population

635 hospitalised adults on non-invasive oxygen

Primary outcome

Proportion of time within target SpO2 range

Effect

Mean difference +21.2 percentage points (95% CI 18.5 to 23.8) favouring closed-loop control

A systematic review and meta-analysis pooled seven randomised trials (635 hospitalised adults) comparing automated closed-loop oxygen control, where a device adjusts flow continuously from pulse oximetry, with manual adjustment by staff. Ventilated patients were excluded.

Patients spent 81% to 95% of the time within their target saturation range with closed-loop control against 51% to 84% with manual titration, a mean difference of 21.2 percentage points (95% CI 18.5 to 23.8). The gain was largest with low-flow oxygen (31 points) and smaller with high-flow nasal oxygen (13 points), which explained most of the variation between trials.

Time in range is a process measure: these trials were not designed to show fewer deaths, intubations or hypercapnic episodes. But too much and too little oxygen both harm, and manual titration on a busy ward drifts. Most Indian wards do not have these devices; the transferable point is that whatever titrates oxygen needs a written target range and someone checking it.

• Prescribe a target saturation range for every patient on oxygen, not just a flow rate.
• Use 88-92% for patients at risk of hypercapnic respiratory failure unless a blood gas supports otherwise.
• Check saturation against target at each observation round and adjust promptly in both directions.
• Where closed-loop devices are available, they are most useful for patients on low-flow oxygen.

Why it matters

Over-oxygenation on general wards is common and largely invisible unless someone is measuring time in range.

Don't overread it

The outcome is time in target range; no patient-outcome benefit was established.

The statistics, in plain English

A mean difference of 21 percentage points with a tight interval means the gain in time-in-range is consistent and large. The high heterogeneity (I-squared 87.9%) is mostly explained by oxygen delivery type, so the size of benefit depends on what device you are replacing.

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.