SOHO trial in NEJM: 🏙️ Multicenter RCT comparing high-flow nasal cannula vs. low-flow oxygen among patients with acute hypoxemic respiratory failure (mostly pneumonia) 🏙️ Inclusion criteria required a respiratory rate >25 and substantial hypoxemia (P/F <200) 🏙️ No difference in the primary endpoint of mortality (not surprising, the mortality endpoint is nearly always neutral in modern RCTs evaluating the nuances of supportive care). 🏙️ HFNC reduced the rate of intubation and increased the number of ventilator-free days. 🏙️ No significant signals of harm from HFNC. This is similar to the FLORALI trial (by the same group of investigators). However, FLORALI showed *improvement* in mortality, whereas the current study didn't. Why do these studies disagree about mortality? The FLORALI trial found no mortality difference between HFNC versus conventional oxygen among ALL PATIENTS. However, it DID find a mortality difference in the *subgroup* of patients with P/F <200. So this is a classic tale in critical care - a subgroup analysis from one trial subsequently fails to replicate. (At least, in terms of the mortality benefit.) But chasing mortality in these trials is a fool's errand. Overall, the SOHO trial still supports the use of HFNC in acute hypoxemic respiratory failure (in terms of reduced intubation and improved ventilator-free days). Using HFNC for sick pneumonia patients is already standard care, so I don't see this changing practice very much. Keep calm and HFNC on. 🌬️