High flow nasal cannula compared with conventional oxygen therapy for acute hypoxemic respiratory failure: a systematic review and meta-analysis

We included 9 RCTs (n = 2093 patients). We found no difference in mortality in patients treated with HFNC (relative risk [RR] 0.94, 95% confidence interval [CI] 0.67–1.31, moderate certainty) compared to conventional oxygen therapy. We found a decreased risk of requiring intubation (RR 0.85, 95% CI 0.74–0.99) or escalation of oxygen therapy (defined as crossover to HFNC in the control group, or initiation of non-invasive ventilation or invasive mechanical ventilation in either group) favouring HFNC-treated patients (RR 0.71, 95% CI 0.51–0.98), although certainty in both outcomes was low due to imprecision and issues related to risk of bias. HFNC had no effect on intensive care unit length of stay (mean difference [MD] 1.38 days more, 95% CI 0.90 days fewer to 3.66 days more, low certainty), hospital length of stay (MD 0.85 days fewer, 95% CI 2.07 days fewer to 0.37 days more, moderate certainty), patient reported comfort (SMD 0.12 lower, 95% CI 0.61 lower to 0.37 higher, very low certainty) or patient reported dyspnea (standardized mean difference [SMD] 0.16 lower, 95% CI 1.10 lower to 1.42 higher, low certainty). Complications of treatment were variably reported amongst included studies, but little harm was associated with HFNC use.