Acute hypoxemic respiratory failure is treated with oxygen. Standard oxygen therapy delivered via a nasal cannula or a face mask is limited by the fraction of inspired oxygen delivered. Low-flow nasal cannula deliver dry oxygen at 1 to 6 litres per minute, face mask can deliver 5 to 10 litres of oxygen per minute. Non-rebreather masks can deliver 10 to 15 litres per minute of oxygen. High-Flow Nasal Cannulae can deliver oxygen at a higher rate with warmed and humified gas flow, and it generates a low level of positive pressure in the upper airways due to a high flow of gas, this also does “wash out” the dead space in the upper airways.
Recently, several systematic reviews with incorporated meta-analysis were published on the effectiveness of High-Flow Nasal Cannula oxygen therapy to treat acute respiratory failure (Maitra, Som et al. 2016, Frat, Coudroy et al. 2017, Huang, Sun et al. 2017, Monro-Somerville, Sim et al. 2017, Nedel, Deutschendorf et al. 2017, Ni, Luo et al. 2017, Ou, Hua et al. 2017, Yue, Zhang et al. 2017, Zhao, Wang et al. 2017, Zhu, Yin et al. 2017). The results are not conclusive and these reviews do not make full use of the published literature. Classical meta-analytical approaches do only take into account the randomised trials that are directly comparing two treatments. What is missing is a network meta-analysis, which can consider more trials by integrating the indirect evidence (i.e. also studies not evaluating HFNC could be used).
Therefore, we set out to systematically search the literature for randomised trials comparing different forms of oxygen therapy in patients with acute hypoxemic respiratory failure and to summarize the evidence by the means of a network meta-analysis.