High-flow nasal oxygen versus face-mask ventilation for rapid sequence induction in non-elective surgical patients: a randomized controlled trial

Citation:

Lam ND, Son LDT, Phat TM, Thu ND, Nga NT, Son VT, Hong BM. High-flow nasal oxygen versus face-mask ventilation for rapid sequence induction in non-elective surgical patients: a randomized controlled trial. BMC Anesthesiol. 2026 Jan 31. doi:10.1186/s12871-026-03654-w. (Trial: NCT06879600)

Study at a glance

- Design and setting: Prospective, single-centre, single-blind (outcome assessor) parallel-group RCT (1:1) in adult non-elective OR cases at Phu Tho Provincial General Hospital, Vietnam. Randomised 216; analysed 214 (107 per arm).

- Population: Adults ≥18 years, ASA I–II, non-elective surgery requiring modified RSI + tracheal intubation; mostly abdominal surgery (~97–99%). Excluded ASA ≥III, anticipated difficult airway, respiratory disease, pregnancy, etc.

- Intervention (HFNC): Heated/humidified 100% O₂ via HFNC at 60 L·min⁻¹ (37°C) for 3 min preoxygenation, continued through induction/apnoea/intubation.

- Comparator (FMV): Tight face mask preoxygenation FiO₂ 1.0 with fresh gas flow 10 L·min⁻¹ for 3 min (no PEEP/inspiratory assist), then mask removed; no routine mask ventilation during apnoea (rescue only).

- Common protocol: Modified RSI with fentanyl 2 µg·kg⁻¹ + propofol 2 mg·kg⁻¹ + rocuronium 1 mg·kg⁻¹; cricoid pressure after LOC; direct laryngoscopy (Macintosh); max 3 attempts; gentle rescue mask ventilation only if SpO₂ ≤94%. Monitoring included SpO₂/EtCO₂/TcCO₂ + arterial line for ABGs; gastric ultrasound before and after.

Key results

- Primary (SpO₂ desaturation <94% during apnoea/intubation):

- HFNC: 0/107 (0%)

- FMV: 13/107 (12.1%)

- Absolute risk difference: −12.1 percentage points (HFNC lower). Approx NNT ≈ 9 to prevent one desaturation event (using 12.1% baseline risk).

- Oxygenation (arterial oxygenation and first-breath oxygen reserve):

- PaO₂ after 3 min preoxygenation: 445.7 ± 16.8 vs 314.2 ± 14.5 mmHg (HFNC higher)

- PaO₂ immediately after intubation: 405.5 ± 17.7 vs 236.5 ± 58.5 mmHg (HFNC higher)

- EtO₂ first breath after intubation: 86.0 ± 11.4% vs 75.7 ± 12.3% (HFNC higher)

- CO₂ / ventilation surrogate outcomes (post-intubation):

- PaCO₂ immediately after intubation: 43.8 ± 1.83 vs 47.43 ± 1.82 mmHg (HFNC lower)

- TcCO₂ at end of intubation: 42.5 ± 2.1 vs 44.3 ± 2.1 mmHg (HFNC lower)

- EtCO₂ first breath after intubation: 39.3 ± 2.9 vs 43.6 ± 3.4 mmHg (HFNC lower)

- Gastric insufflation (ultrasound-derived change from baseline to after intubation):

- ΔGRV (ml): 1.726 ± 0.455 vs 1.785 ± 1.005 (small, similar)

- ΔCSA (cm²): 0.172 ± 0.092 vs 0.192 ± 0.078 (small, similar)

- Comfort (0–10 NRS; higher=better): median 3 (IQR 1–5) HFNC vs 4 (IQR 3–5) FMV (subjective and unblinded → higher bias risk).

- Reported adverse events:

- Nasal dryness/mucosal irritation: 7/107 HFNC vs 1/107 FMV.

- Aspiration: Table reports 1/107 in each group, but the Results text reportedly says “no aspiration events” (internal inconsistency to resolve).

Bias / certainty (as provided)

- RoB 2 overall: Some concerns (mainly “selection of reported result” due to limited visibility of prespecified outcomes/SAP and multiplicity; other domains judged low risk).

- GRADE (selected outcomes): Moderate certainty for the key oxygenation/CO₂ outcomes (downgraded 1 level for study-level reporting/analysis-set concerns); Low certainty for comfort (unblinded subjective measure).

Practical takeaway

- In low-risk (ASA I–II) non-elective OR RSI cases using a “no routine mask ventilation” approach, HFNC (60 L·min⁻¹) during apnoea/intubation was associated with markedly fewer desaturation events and higher peri-intubation oxygenation, with small differences in CO₂ measures and no clear signal of increased gastric insufflation—while minor nasal irritation was more common and aspiration reporting needs clarification.