Combination of heliox and CPAP without a ventilator: bench test and clinical observations.

EDITOR: Failure to restore an adequate airway in patients with acute upper airway obstruction can rapidly lead to hypoxic injury or even death. Helium–oxygen mixtures (heliox) are less dense than air–oxygen mixtures and can improve the flow of gas through partially obstructed airways. Since Barach described this principle 70 years ago, heliox has at times been applied in spontaneously breathing patients with acute upper airways obstruction [1]. Heliox has also been combined with (non-) invasive mechanical ventilation, often with specifically adapted apparatus [2]. However, the correct function of ventilators relies on the precise measurement of flow, pressure, temperature and oxygen fraction for which the necessary sensors are calibrated for use with air–oxygen. The unique physical properties of heliox have an important impact on all these measurements. Thus the correct application of heliox in ventilators is far from trivial [3]. This technical problem becomes especially relevant under emergency circumstances. Continuous positive airway pressure (CPAP) by face mask can reduce upper airway collapse [4] and the combination of mask CPAP and heliox might be useful. The Boussignac continuous positive airway pressure (BCPAP) system (Vygon, 95440 Ecouen, France) for mask CPAP is a simple and lightweight (10 g) disposable plastic device without moving parts [5]. This device may thus avoid the problems with ventilators and heliox. In a bench test, we evaluated whether heliox works with the BCPAP system and how much more heliox flow was needed compared to oxygen to achieve similar pressures. We then evaluated the effect of this combination in selected patients with acute upper airways obstruction. We chose a helium–oxygen mixture of 60/40 to ensure a sufficient oxygen delivery under all circumstances. For the bench test we used 10 randomly selected, commercially available BCPAP systems and an analyser specifically designed to measure pressure or flow for helium–oxygen mixtures (VT Plus, Fluke Biomedical, Carson City, NV, USA). Since the variable orifice flow meters mounted on the heliox cylinders were oxygen-calibrated, we first recalibrated these meters by comparing the indicated flow of oxygen or heliox with the true flow as measured with the VTplus analyser. This was done 10 times for each of the following levels of indicated flow: 5, 10, 15, 20, 25 and 30 L min. The observed relation between indicated and actual flow was also compared to the theoretically predicted flow [6]. After this flowcalibration, we then measured the pressures generated by the BCPAP system for the various indicated flows (Fig. 1). The measured flow rates (true flow) of the variable orifice flow meters showed a 256 6% (SD) higher flow for heliox compared to oxygen. When we calculated the predicted actual flow for the variable orifice flow meters for heliox 60/40 based on