Videolaryngoscopy or fibreoptic bronchoscopy for awake intubation of bariatric patients with predicted difficult airways – a randomised, controlled trial

Awake fibreoptic bronchoscopy is the gold standard for securing a predicted difficult airway. However, it requires expensive equipment, and expertise that is difficult to maintain 1. After institutional ethical approval, written participant consent and trial registration (www.clinicaltrials.gov NCT02169232), we compared awake intubation using videolaryngoscopy with fibreoptic bronchoscopy (FOB) in patients scheduled for bariatric surgery with a simplified airway risk index of 4 or higher 2. Having administered intravenous glycopyrrolate 0.4 mg and midazolam 0.02 mg.kg−1 ideal body weight (IBW) 3, 20 ml lidocaine 2% was aerosolised onto the airway during deep respiration, followed by placement of cotton gauze soaked with viscous lidocaine 2% against the tonsillar pillars. If insertion of an oral airway resulted in a reaction score >2 4, an additional 10–20 ml of lidocaine was aerosolised. Once the reaction score was < = 2, remifentanil 0.06 μg.kg−1 IBW was administered intravenously, before tracheal intubation one minute later. Patients were randomly allocated to an awake Glidescope® (GVL, Verathon Inc, Bothell, WA, USA)-assisted tracheal intubation performed in a standard fashion 5 by one of the two authors, AM or RH, or an awake fiberoptic bronchoscope assisted tracheal intubation performed in a standard fashion 6 by TS or RL, each operator having performed >40 procedures. Our primary outcome was the duration of the intubation procedure, recorded by an independent observer as the time in seconds from initial introduction of an airway instrument into the mouth to the time tube placement was confirmed by capnography. Secondary outcomes were: the number of attempts taken for tracheal tube placement; reaction score; change in heart rate during the procedure; difficulty of the procedure rated by both the operator and an independent observer on a 10 cm visual analogue scale (VAS) from very easy to very difficult; and patient-rated gagging severity and overall satisfaction using a VAS. We wished to demonstrate the non-inferiority of videolaryngoscope-assisted tracheal intubation, and calculated a sample size of 36 patients (α 0.05, 90% power, non-inferiority limit dNI 30 s, standard deviation 30 s 7. We randomly allocated 5 patients to the videolaryngoscope and 6 patients to the FOB groups, with mean (SD) body mass indices of 60 (12) and 60 (14) kg.m−2, respectively, before the bariatric surgery service moved from our institution, effectively terminating our study. To account for early termination of our study, the p value and the difference in group means (including confidence interval) for the primary outcome were adjusted using a Pocock-type alpha spending function 8. Data were analysed using Stata 14 (StataCorp LP, College Station, TX, USA). The adjusted difference in mean time for tracheal intubation was 60.9 (95% CI 20.5–101.3; p = 0.003) seconds faster in the videolaryngoscope group. We found no other statistically significant differences. Although this study lacks the statistical power to compare secondary outcomes between groups, we are confident that videolaryngocopy enables significantly faster tracheal intubation in bariatric patients with predicted difficult airways compared to fibreoptic bronchoscopy. It is unlikely that this resulted from differences in sedation or topical anesthesia, as the bispectral indices and airway reaction scores were comparable between the groups. We would welcome confirmation of our findings by other researchers.