Background Oxygen may be delivered by various non-invasive methods to infants. Currently there is limited evidence about the actual fraction of inspired oxygen (FiO2) available via these different methods and how these various methods compare with each another. Head boxes are used in children with conditions such as bronchiolitis, with removable lids for accessing patients. Venturi adapters are simple devices that generate a consistent flow of gas at a constant FiO2.

Aims We aim to assess the FiO2 within the head box and the variation under 3 principle conditions; 1) different flow rates of oxygen 2) opening the lid and 3) delivery of gas via a venturi.

Methods Bench study using infant head box. The FiO2 was measured using saturation probes placed at different points within the head box and under the various conditions above.

Results The FiO2 within the head box varied by 1–5% depending on location, and was associated with higher flow rates. Higher flow rates increased to higher steady state FiO2with (1 l/min 25–28%, 5 l/min 32–38%, 15 litres 48–55%) and reached this steady state earlier (1 l/min 6 min, 15 l/min 3 min). Higher flow rates also resulted in more rapid recovery of FiO2 following removal of the head box lid. The use of venturi mask adapter resulted in less variation of head box FiO2 compared to delivery without the venturi (1 l/min 1% vs 3.5%, 15 l/min 2.4% vs 6.7%), and quicker times to reach steady state (1 l/min 3 min, 15 l/min 1.5 min).

Conclusion This bench study demonstrates variation of oxygen concentration within the head box, and the time taken to reach equilibrium of the FiO2 which varies depending on flow rates.

We suggest monitoring of head box oxygen concentrations (especially on accessing the box) with probes placed close to the infants head. Consider the use of higher flow rates when there is a need for tighter oxygen control such as in unwell children when they are more susceptible to changes in oxygen concentration. Further studies exploring the delivery of oxygen using other non invasive methods are indicated to assess the concentrations of oxygen delivered.