Tracheal intubation is required for mechanical ventilation during general anesthesia and to prevent aspiration of secretions. Commonly, direct laryngoscopy (DL) is used, but this technique may fail in patients with a difficult airway (1), e.g., in patients scheduled for otorhinolaryngologic or oral and maxillofacial surgery. Besides intubation with a bronchoscope that is still regarded as the gold standard, videolaryngoscopy (VL) has evolved as a valuable alternative technique in patients with an expected or unexpected difficult airway (2–4). However, VL has some limitations and may fail due to insufficient visualization of the larynx or impossible tube advancement. Recently, a tracheal tube with an integrated camera has been introduced (VST, VivaSight-SL, Ambu A/S, Ballerup, Denmark) that may allow for direct guidance of the tube and may aid in tracheal intubation in patients with difficult airways (5). This tube has been evaluated under the clinical conditions in intensive care patients (6) and patients with morbid adiposity compared with DL (7), but there is a paucity of data in patients with difficult airways, so far. As opposed to VL, the camera axis of VST may be adjusted individually by modifying the angulation of a stylet and by direct steering of the tip of the tube during tube advancement which may provide a benefit in the cases of difficult airways.

Therefore, we assessed the feasibility of intubation with the VST in patients with a predicted difficult airway compared with VL in a prospective randomized non-inferiority trial. Due to the paucity of data in these patients, we chose end-tidal oxygen fractions (etO₂) after intubation as an important safety parameter for the primary outcome measure and to ensure adequate oxygenation throughout the intubation procedure.

From August 27, 2020, to February 12, 2021, 48 patients receiving tracheal intubation were randomized to either camera-assisted intubation with the VST or intubation with VL in a 1:1 ratio (as shown in Figure 2). The baseline characteristics of patients are shown in Table 1. Groups had similar values for baseline saturation of oxygen, and end-tidal fractions of oxygen and carbon dioxide after pre-oxygenation (as shown in Table 2). All patients received intubation in deep anesthesia and were paralyzed.

Concerning the oxygenation after airway management, intubation with the VST was non-inferior to VL with a post-intubation fraction of oxygen of 0.79 ± 0.08 (95% CIs: 0.75–0.82) in the VST group versus 0.81 ± 0.06 (0.79–0.84) in the VL group with a mean difference of 0.02 (95% CIs: −0.07; 0.02), p = 0.234. The first attempt success rate was 79% in the VST group and 88% in the control group, p = 0.449. The overall success rate in the VST group was 100 and 96% in the VL group (p = 0.312) with one patient requiring a laryngoscopy with a hyperangulated blade. POGO was 89 ± 21% in the VST-group versus 60 ± 36% in the VL group, p = 0.002. While time to successful intubation did not statistically differ between the groups, time to successful intubation in the first attempt was prolonged in the VST group (84 ± 31 s vs. 49 ± 14 s, p < 0.001). A graphical depiction is given in Figure 3. Tube advancement was rated to be more difficult for the VST (visual analog scale 0–100, lower values better: 47 ± 23 vs. 21 ± 25, p = 0.001). Other parameters did not differ between the groups. Regurgitation, aspiration, or accidental esophageal intubation did not occur in any patient. An overview on results is given in Table 2. Anesthetists professional experience was 16 ± 5 years in the VST and 15 ± 5 years in the VL group, p = 0.277. The pre-trial VST angulation studies with a manikin yielded an angle of 85 degrees. Results of an evaluation of this setup in a manikin trial are reported in Supplementary Table 1.

In this prospective randomized trial comparing tracheal intubation assisted by a tube-mounted camera (VST) with intubation by VL in patients with a predicted difficult airway, we found no difference concerning oxygenation during airway management or the first attempt success rate. While time to intubation with the VST was prolonged compared with VL, visualization of the larynx by the VST was improved as compared with VL. However, tube advancement was rated more difficult using the VST. No significant differences were found for other parameters, particularly not for complications, such as hypoxia, hypotension, regurgitation, or accidental esophageal intubation.

So far, the VST has been evaluated for tracheal intubation in patients with a predicted non-difficult airway in intensive care (6), in patients with morbid obesity (7), via supraglottic airway devices (15, 16), in manikins (17–21), in a cadaver study (22), and for the guidance of percutaneous dilatational tracheostomy (23, 24).

Awake intubation via bronchoscopy is often regarded as the gold standard technique for the management of the expected difficult airway (3), but this method may be complex, difficult to learn, expensive, time-consuming, and requires thorough patient preparation. Awake videolaryngoscopy has been found to be a safe and effective alternative in carefully selected patients (25). Videolaryngoscopy plays a key role in unexpected difficult airways (4, 26), but the angle of the attached camera is fixed in both Macintosh style and hyperangulated blades. Therefore, visualization of the larynx may result in inadequate viewing conditions. In particular, a restricted epiglottis motion with the epiglottis being adherent to the pharynx [grade 3 view according to Cook (27)] is a pitfall and possible limitation of VL and rather a domain of flexible intubation methods (28). Beyond bronchoscopic and videolaryngoscopic intubation techniques further video-guided techniques using tube or stylet mounted cameras have been introduced but their role for the management of the expected or unexpected difficult intubation remains unclear (29).

In cases of insufficient visualization of the larynx for intubation, a hyperangulated blade or flexible bronchoscopy may be used to obtain visualization of the larynx with a more advantageous viewing angle. With the VST, the axis of the camera may be individually adjusted by angulation of the stylet, presumably explaining the improved laryngeal viewing conditions with a higher POGO score for the VST found in our study. For visualization during VL, it has been shown that insufficient laryngeal views above Cormack/Lehane grade 2 are associated with an increased failure rate of the method (30). Failure rates for VL depend on the type of VL, as well, with first pass success rates ranging from below 40% to above 90% (31). Although visualization was superior for the VST group in our study, we could not show a decrease in failure rate. Interestingly, the most common cause for failure of intubation with VL is the inability to direct the tube toward the laryngeal inlet (32). Presumably, the VST provides a benefit in this respect as the camera provides direct guidance.

The number of intubation attempts did not differ in our study. Previously, more than two attempts have been associated with an increase in complications (33) and strategies have been suggested to increase the rate of first attempt success of intubation, one strategy focusing on the selection of an adequate device for the respective predicted difficult airway (34).

For VL, an overall intubation time of approximately 50–60 s has been reported in human subjects with a difficult airway (31). For the VST, approximately 30 s have been reported previously, but these results were obtained in patients without difficult airways (6). This contrasts with the required duration for intubation in our study with the VST needing nearly three times longer, only counting intubations that were successful on first attempt. We could not find any negative effects as desaturation or a higher rate of hypotension in our patients, but we presume that the prolonged intubation increases complications in patients prone to hypoxia, i.e., in critically ill patients or patients with morbid obesity (35, 36). This may limit the usefulness of the VST to elective cases without risk for hypoxia.

We chose to use a Macintosh-style VL for the control group, angulated at approximately 30 degrees, while hyperangulated blades are angulated at approximately 60 degrees (28, 37). For the VST, we used an angulation of approximately 85 degrees as obtained prior to the beginning of our study that was more pronounced than typical hyperangulated laryngoscope blades. For VL with hyperangulated blades, laryngoscopic view is often improved compared with Macintosh blades, but a common problem arises when attempting to advance the tube through the laryngeal inlet because the tube may hit the anterior tracheal wall at a nearly perpendicular angle, hindering further tube advancement (37), explaining why tube advancement in our study was rated more difficult with the VST as compared with VL.

Recently, a video stylet inserted into a tracheal tube providing camera guidance has been evaluated for intubation in patients with cervical immobilization showing a significantly lower first attempt success rate and prolonged time to intubation as compared with VL (29). As the basic principles of intubation with camera guidance resemble the VST approach, and the results are comparable with our study results, tube-camera guided intubation might indeed be inferior to a laryngoscope approach in the clinical settings, but further data are required.

Our study has the following limitations. Our choice of calculating the sample size for etO₂ as the primary endpoint may be questionable as first attempt success rates are widely accepted as endpoints for studies evaluating airway devices. However, sufficient oxygenation is paramount in airway management and an important safety parameter and thus chosen purposely. Our sample size was adequately powered to show non-inferiority for oxygenation but may be underpowered for the first attempt success rate. However, we provide the first data in human subjects for use of the VST in the predicted difficult airways. The learning curve for the VST may not have had reached its peak after the manikin training, but all participating anesthetists were highly experienced in airway management, such as the management of predicted and unexpected difficult airway scenarios.

Tracheal intubation with the VST in patients with an expected difficult airway is feasible, and oxygenation was not inferior as an important safety parameter in the VST group. The first pass success rate in our cohort did not differ between groups. Visualization of the larynx with the VST was superior to VL but time to intubation was prolonged with the VST. We believe further studies are warranted to define which cohorts of patients might benefit from intubation with the VST.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by Ethics Committee of the Hamburg Chamber of Physicians (No. PV7276, June 6, 2020). The patients/participants provided their written informed consent to participate in this study.

JG designed the study, wrote the manuscript, performed the statistical analysis, and interpreted the data. EM recruited the patients, acquired the data, helped to write the manuscript, and helped with the statistical analysis. PB and TP helped to recruit the patients and to acquire the data. PT helped with the statistical analysis and to interpret the data. MAP helped to acquire the data and to revise the manuscript. SK helped to interpret the data and to revise the manuscript. MP designed the study, helped to acquire the data, interpreted the data, and revised the manuscript. All authors read and approved the final manuscript.

This study was funded exclusively from the departmental resources including the acquisition of the VivaSight-SL tracheal tubes.

JG has received research support from Adroit Surgical, Ambu, ETView, and Infectopharm, and received consultant and lecture fees from Drägerwerk, Fresenius Medical, GE Healthcare, and Smith Medical; SK received research support from Ambu, Daiichi Sankyo, ETView Ltd., Fisher & Paykel, Pfizer, and Xenios, lecture fees from Astra, C.R.Bard, Baxter, Biotest, Cytosorbents, Daiichi Sankyo, Fresenius, Gilead, Mitsubishi Tanabe Pharma, MSD, Pfizer, Philips, and Zoll, and consultant fees from Bayer, Fresenius, Gilead, MSD, and Pfizer; MP received a research grant awarded by Verathon. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.