Automated oxygen flow titration for infants with bronchiolitis: a multicentre randomised controlled trial

To compare automated closed-loop oxygen therapy using the FreeO2 device versus conventional manual oxygen therapy with regard to length of hospital stay among spontaneously breathing infants with bronchiolitis.

Multicentre randomised controlled trial conducted from 2018 to 2023.

Ten paediatric departments across France.

103 infants, 1-12 months of age, hospitalised for acute bronchiolitis requiring oxygen therapy with SpO₂ <92%, excluding severe bronchiolitis.

Patients were randomised to receive either automated oxygen administration via the FreeO2 device or conventional manual administration.

The primary endpoint was length of hospital stay in hours from emergency admission to discharge. Secondary endpoints included re-hospitalisation at 7 and 30 days, use of non-invasive ventilation (NIV), transfer to intensive care within 3 days and time spent in hypoxia (SpO2<92%) or hyperoxia (SpO₂>98%).

The median hospital stay was 71.00 hours (IQR 45.8-97.6) in the FreeO2 group and 69.6 hours (IQR 47.5-116.7) in the manual group (p=0.39). The percentage of time within the oxygen target zone was 89.4% in the FreeO2 group vs 74.9% in the manual group (p<0.05). Median oxygen flow was 0.1 L/min (IQR 0.1-0.2) in the FreeO2 group vs 0.3 L/min (IQR 0.2-0.5) in the manual group (p<0.05). The groups did not significantly differ in re-hospitalisation at D7 (p=1.00) or D30 (p=1.00) or NIV use (p=0.47).

While we did not find evidence of a reduced length of stay, which may be due to low study recruitment, we did find the FreeO2 device yielded a positive reduction in flow rates of oxygen.