Omitting face-mask ventilation before tracheal intubation in low-risk adults: a randomized non-inferiority trial

Face-mask ventilation (FMV) is routinely performed between the induction of general anesthesia and tracheal intubation, but it can cause gastric insufflation, and its necessity in low-risk patients is uncertain. We examined whether omitting FMV before intubation is non-inferior to routine FMV for oxygen desaturation, and whether the two approaches differ in gastric insufflation.

In this single-center, randomized, group-sequential non-inferiority trial, low-risk adults undergoing elective thyroid surgery were assigned to tracheal intubation with FMV (rocuronium 0.6 mg/kg) or without FMV (rocuronium 1.2 mg/kg). The primary outcome was the incidence of desaturation (peripheral oxygen saturation [SpO₂] ≤ 95%) from induction to the first intubation attempt, with a non-inferiority margin of 2% (absolute risk difference). The pre-specified secondary outcome was the gastric antral cross-sectional area (CSA) after intubation, measured by ultrasonography. The trial began as a fixed-sample design; a single interim analysis at 50% of enrollment (O'Brien-Fleming spending function) was added by protocol amendment after enrollment had begun. Gastric insufflation, defined by a qualitative ultrasonographic sign and by an antral CSA increase greater than 30%, was analyzed post hoc.

The trial was stopped at the interim analysis with 288 participants (144 per group), a decision made in the absence of any events rather than by a stopping rule. No patient in either group reached the desaturation threshold (0% vs. 0%; risk difference 0%, 95% confidence interval - 2.6% to 2.6%). The gastric antral CSA after intubation did not differ between groups (median 3.7 cm² in both groups; P = 0.99). In post hoc analyses, gastric insufflation occurred only in the FMV group, both by the qualitative sign (8/144 [5.6%] vs. 0/144; P = 0.007) and by the antral CSA threshold (9/144 [6.2%] vs. 0/144; P = 0.003).

In low-risk adults undergoing elective thyroid surgery, no patient reached the desaturation threshold whether or not FMV was performed before intubation. Because the trial was stopped at the interim analysis with zero events, the upper limit of the confidence interval for the risk difference exceeded the pre-specified 2% margin, and formal non-inferiority was not established. In exploratory post hoc analyses, gastric insufflation occurred only in the FMV group. These hypothesis-generating findings warrant confirmation in adequately powered trials before routine FMV is omitted in low-risk patients.

ClinicalTrials.gov, NCT02400931. Registered 23 March 2015.