Nasal high-flow preoxygenation for endotracheal intubation in the critically ill patient: a randomized clinical trial

PurposePreoxygenation with high-flow therapy by nasal cannulae (HFNC) is now widespread in the intensive care unit (ICU). However, no large randomized study has assessed its relevance in non-severely hypoxemic patients. In a randomized controlled trial (PROTRACH study), we aimed to evaluate preoxygenation with HFNC vs. standard bag-valve mask oxygenation (SMO) in non-severely hypoxemic patients during rapid sequence intubation (RSI) in the ICU.MethodsRandomized controlled trial including non-severely hypoxemic patients requiring intubation in the ICU. Patients received preoxygenation by HFNC or SMO during RSI. HFNC was maintained throughout the intubation procedure whereas SMO was removed to perform laryngoscopy. The primary outcome was the lowest pulse oximetry (SpO(2)) throughout the intubation procedure. Secondary outcomes included drop in SpO(2), adverse events related to intubation, and outcome in the ICU.ResultsA total of 192 patients were randomized. In the intent-to-treat analysis, 184 patients (HFNC n=95; SMO n=89), the median [IQR] lowest SpO(2) was 100% [97; 100] for HFNC and 99% [95; 100] for the SMO group (P=0.30). Mild desaturation below 95% was more frequent with SMO (23%) than with HFNC (12%) (RR 0.51, 95% CI 0.26-0.99, P=0.045). There were fewer adverse events in the HFNC group (6%) than in the SMO group (19%) (RR 0.31, 95% CI 0.13-0.76, P=0.007), including fewer severe adverse events, respectively 6 (6%) and 14 (16%) with HFNC and SMO (RR 0.38, 95% CI 0.15-0.95, P=0.03). ConclusionsCompared with SMO, preoxygenation with HFNC in the ICU did not improve the lowest SpO(2) during intubation in the non-severely hypoxemic patients but led to a reduction in intubation-related adverse events.Trial registrationClinical trial Submission: 7 March 2016. Registry name: Benefits of high-flow nasal cannulae oxygen for preoxygenation during intubation in non-severely hypoxemic patients: the PROTRACH study. Clinicaltrials.gov identifier: NCT02700321. Eudra CT: 2015-A00145-44. CPP: 15/13-975 (Comite de protection des personnes de Rennes). URL registry: https://clinicaltrials.gov/ct2/show/record/NCT02700321.