X3.1 Controlled oxygen delivery

In the emergency setting, supplemental oxygen may be required to relieve hypoxaemia. Oxygen flow should be carefully titrated to achieve a target SpO₂ range of 88 to 92%. Nasal cannulae deliver a variable concentration of oxygen, but a flow of 0.5–2.0 L per minute is usually sufficient. The TSANZ position paper on acute oxygen use in adults highlights the importance of assessment of hypoxia, prescription of oxygen and always implementing SpO2 targets to prevent over-oxygenation (Barnett 2022).

High flow oxygen via a Hudson mask or non-rebreather mask should be avoided, as it is rarely necessary and may lead to hypoventilation and worsening respiratory acidosis and increased mortality. A randomised study has demonstrated that in the pre-hospital emergency setting titrated oxygen via nasal cannula compared with high flow oxygen reduced mortality by 78% in COPD patients (NNH=14) (Austin 2010) [evidence level II]. In an observational study from the UK of 1027 patients admitted across 6 hospitals with an exacerbation of COPD and receiving supplemental oxygen, Echevarria et al reported that in-hospital mortality was lowest in those with admission oxygen saturations between 88 and 92%.  This mortality effect was dose-responsive with mortality rates highest in the sub-group with oxygen saturations 97-100%.  The effect was also present in patients with normocapnia.  The authors recommend that all patients with COPD receiving supplemental oxygen should have an oxygen saturation target of 88-92% independent of the presence of hypercapnia (Echevarria 2020).  In a Victorian retrospective case file emergency department audit of patients admitted to hospital with exacerbation of COPD between Jan 2012 and March 2013, 84.4% had a final ambulance oxygen saturation reading of > 93% (95% CI 79.5 to 88.3%) (Chow 2016). A retrospective Australian study examined oxygen use in 111 patients admitted with hypercapnia due to an exacerbation of COPD. Over-oxygenation was common and was significantly more likely to occur on non-respiratory ward admissions (76% vs 57%, p=0.03) (Anderson 2020). In Wellington, New Zealand, an audit of patients with an exacerbation of COPD transferred by ambulance to hospital was undertaken, before and after an education program to reduce high concentration oxygen delivery was undertaken (Pilcher 2015). Significantly fewer patients received high concentrations of oxygen in 2010; however, concern was voiced by the authors about the continued use of high concentration oxygen to drive nebulisers. Education may be the key to changing practice.

Where there is evidence of acute respiratory acidosis (or a rise in PaCO₂) on arterial blood gas (ABG), together with signs of increasing respiratory fatigue and/or obtunded conscious state, assisted ventilation should be considered. Early non-invasive positive pressure ventilation (NIV) may reduce the need for endotracheal intubation (see below for more detail).