Preoxygenation strategies for intubation of patients who are critically ill: a systematic review and network meta-analysis of randomised trials.

Summary

Across 15 randomized trials (n=3420), preoxygenation with NIPPV probably reduces hypoxemia during intubation compared with HFNC (RR 0.73, 95% CI 0.55–0.98) and reduces hypoxemia versus facemask oxygen (RR 0.51, 0.39–0.65). HFNC also reduces hypoxemia versus facemask (RR 0.69, 0.54–0.88). First-pass success and all-cause mortality were not different; NIPPV likely lowers serious adverse events versus facemask and possibly versus HFNC.

Key Findings

• NIPPV reduced hypoxemia versus HFNC (RR 0.73, 95% CI 0.55–0.98) and versus facemask oxygen (RR 0.51, 0.39–0.65).
• HFNC reduced hypoxemia versus facemask oxygen (RR 0.69, 0.54–0.88).
• No differences were found in first-attempt intubation success or all-cause mortality among strategies.
• NIPPV probably reduced serious adverse events versus facemask oxygen and may reduce them versus HFNC.

Clinical Implications

When feasible, use NIPPV for preoxygenation of critically ill adults requiring intubation; HFNC is preferable to facemask oxygen when NIPPV is unavailable or contraindicated. Protocols should prioritize staff training and equipment availability.

Limitations

• Heterogeneity in trial protocols and definitions of serious adverse events and mortality
• No observed effects on first-pass success or mortality despite reductions in hypoxemia

Future Directions

Head-to-head pragmatic RCTs in high-risk subgroups (e.g., severe hypoxemia, obesity), evaluation of combined strategies (NIPPV+HFNC), and cost-effectiveness and implementation studies.