The successful management of an unanticipated difficult airway is critically important in the practice of emergency medicine and is achieved frequently. In emergency medicine literature, difficult intubation occurs in 3.1% of patients and failed intubation occurs in 1.1% to 2.7% of patients.1^–3

Although most cases of difficult intubations are managed satisfactorily, the results of failed intubation are catastrophic to physicians as well as patients. Resulting from these uncommon yet critical failures, many strategies including rapid sequence intubation and new devices have been developed for airway management.4^–6

The GlideScope video laryngoscope (GVL; Saturn Biomedical System, Burnaby, British Columbia, Canada) is a novel video laryngoscope that was developed for managing difficult airways (fig 1). The GlideScope is a laryngoscope with a high-resolution camera embedded within the blade and a light source mounted beside the camera. The image is displayed on a 7-inch liquid crystal display screen. The blade design differs from a standard laryngoscope blade in that it is not detachable, has a maximum width of 18 mm at any point and has a 60° midline angle. This design provides a view of the supraglottic anatomy with minimal distortion compared with the conventional Macintosh laryngoscope (Welch Allyn Inc, New York, USA).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1 GlideScope video laryngoscope.

At present, the GVL has been used mainly in operating rooms by anaesthesiologists7 8 but experience with this device in rescue situations in the emergency department is limited. Therefore, the performance of the GVL in emergency airway management (both normal and difficult) remains to be determined. The purpose of this study was to compare the standard Macintosh laryngoscope with the GVL to ascertain whether an alternative device is useful in a standardised difficult intubation scenario. We hypothesised that the GVL would perform comparably with the Macintosh laryngoscope in the normal airway and prove to be superior to the Macintosh laryngoscope in difficult airway management cases.

MATERIALS AND METHODS

Study design and participants

We performed a prospective, crossover and randomised study to evaluate the usefulness of the GVL when used by experienced emergency physicians in anatomically correct manikins. After Institutional Review Board approval was obtained, 25 emergency physicians of varying experience were voluntarily recruited in two emergency medicine residency programmes. All participants were briefed on the study and written informed consent was obtained. We included emergency physicians with at least one year of clinical experience and 50 intubations using the Macintosh laryngoscope. We excluded emergency physicians who had previous experience using the GVL.

Study protocol

Before beginning the study, participants were given 2-minute demonstrations and instructions on intubation with the GVL by one investigator (HJK). Participants were then allowed 3 minutes to familiarise themselves with the GVL and practice intubation on a Laerdal airway management trainer (Laerdal Medical AS, Stavanger, Norway).

Each emergency physician performed tracheal intubation in four scenarios on the Airsim manikin (Trucorp, Belfast, Northern Ireland): normal airway; cervical spine immobilisation using a manual in-line manoeuvre; tongue oedema and a combination of cervical spine immobilisation with tongue oedema. They practised intubation in each scenario using the GVL and Macintosh size 3 blade. All intubations were performed using a 7.0-mm cuffed tracheal tube (Mallinckrodt Medical, Athlone, Ireland). A stylet was inserted into the tracheal tube in all intubations. Before using the GVL the tube curvature was configured to that of the GlideScope blade to increase the chance of successful intubation according to the manufacturer’s instructions.

Participants were assigned at random by a sequence generator computer program to start with one of the four scenarios. They were further randomly assigned to start with either the GVL or the Macintosh laryngoscope for each case. In the same manner, we assigned the other scenarios to each participant over one-week intervals. The interval of one week was determined at the discretion of the investigators so as not to familiarise participants with the GVL.

The primary endpoint was the time required for successful tracheal intubation, which was defined as the time in seconds from touching the laryngoscope to passage of the tracheal tube through the vocal cords. The verification of passage was performed by direct visual and manual confirmation by gripping the tracheal tube passing through the trachea. Additional endpoints included the rate of successful tube placement in the trachea and the number of intubation attempts. A failed intubation attempt occurred when the trachea was not intubated or when the intubation attempt required more than 120 seconds. After each practice session, participants were asked to report the vocal cord visualisation by the percentage of glottic opening (POGO) visible9 and the subjective ease of intubation on a visual analogue scale (from 0, extremely easy to 10, extremely difficult). We also surveyed participants’ choice of intubation devices for each scenario.

Data analysis

SPSS 12.0 for Windows (SPSS Inc, Chicago, Illinois, USA) was used for statistical analyses. The sample size was calculated according to our primary endpoint variable (time required to intubate). A difference of 5 seconds between the times required for tracheal intubation using the GVL and Macintosh laryngoscopes under the normal airway scenario and 10 seconds for the cervical spine immobilisation scenario were selected as the minimum clinically significant values.10 Using these values, we calculated that a sample size of 20 participants was sufficient to detect an effect value of 0.68 (mean difference/common SD) for the time required to intubate between devices at a significance level of 0.05 (two sided) with 80% power.11 Parametric data are presented as mean (±SD). Non-parametric data are presented as the median and interquartile range (IQR). The time required for the first attempt and successful intubation attempt were analyzed using a paired t test for two device comparisons. Given the paired nature of these data and the fact that the data were not normally distributed, Wilcoxon signed-rank tests were used for POGO and subjective ease. Binomial tests with the test proportion set at 0.5 were used to assess participants’ choice of device. Friedman’s test was used for POGO analysis among the scenarios. A p value less than 0.05 was considered statistically significant.

RESULTS

All 25 participants successfully completed the study. The median experience as an emergency physician of the participants was 3.2 (2.2 to 4.5) years, and 21 participants (84%) were male. There was a significant difference in the POGO score for the Macintosh laryngoscope among the scenarios (p = 0.002). This result showed that four airway scenarios varied in the degree of difficulty.

In the normal airway scenario, all emergency physicians successfully intubated the trachea in the first attempt using both devices. The mean times required for successful tracheal intubation using the GVL and Macintosh laryngoscope were 10.9 seconds (SD 5.5) and 9.6 seconds (SD 3.3), respectively (p = 0.136; table 1). There were no differences in the POGO score and subjective ease of intubation between the two devices (p = 0.550 and p = 0.280, respectively). For the cervical spine immobilisation case, all participants successfully performed tracheal intubation in the first attempt with both laryngoscopes. The median (IQR) POGO scores were 80.0% (55.0 to 95.0) for the GVL and 60.0% (45.0 to 85.0) for the Macintosh laryngoscope, which was a significant difference (p = 0.027). There were, however, no differences in the mean time required for intubation and ease of intubation between the two devices (p = 0.775 and p = 0.052, respectively). After their experience with the manikin, 76% of participants thought the GVL would be a useful device for cervical spine immobilisation in their own clinical practice (p = 0.015). In the tongue oedema scenario, all initial tracheal intubation attempts using both laryngoscopes were successful. Statistical analyses for the time required for complete intubation, POGO and ease of intubation showed no differences between the two devices. For the combined scenario, one participant failed to intubate the trachea with the Macintosh laryngoscope. Two emergency physicians required two attempts with the Macintosh laryngoscope to intubate the trachea successfully and one physician required two attempts with the GVL. The duration of tracheal intubation, POGO and ease of intubation were, however, similar for both devices.

DISCUSSION

Although the curved laryngoscope blade described by Macintosh in 1943 remains the most popular device for tracheal intubation,12 there has been extensive research into the development of new airway devices for difficult airway management. The ideal device would be a tool that is designed to be used as a standard laryngoscope but provides improved views of the larynx in an unanticipated difficult airway. The GVL may be one such device. The GVL is similar in design to a conventional Macintosh laryngoscope, so emergency physicians would be familiar with using this device and easily adapt to it. The GVL has a high-resolution camera embedded within the blade and provides a wider field of view than the conventional laryngoscope.

This study demonstrates that the GVL provides comparable intubation conditions in the normal airway and a higher quality view of the glottis in difficult scenarios compared with the Macintosh laryngoscope. All participants had previously performed at least 50 intubations using the Macintosh laryngoscope, so competence in conventional intubation techniques was assumed. No participating emergency physicians had, however, had previous experience using the GVL. Participants were given only 2-minute demonstrations and were allowed 3 minutes to practice intubation with the GVL. Nonetheless, there was no failure in tracheal intubation using the GVL for normal and difficult scenarios. This finding shows that the GVL can be used easily and learned rapidly.

In this study, the GVL was not obviously better than the conventional Macintosh laryngoscope with regard to the time required for successful tracheal intubation. This result was different from the result of the previous study. Lim et al13 reported that the GVL required more time for intubating the trachea than the Macintosh laryngoscope did in easy scenarios but less time in difficult scenarios. This difference may be attributed to the fact that the physician feels resistance to the advancement of the tracheal tube while the tube is passing the vocal cords. The difficulty with tube manipulation under indirect vision probably accounts for a major proportion of the time required for successful tracheal intubation. Also, this difficulty could be related to complications such as oropharyngeal injuries.14 These issues are not confined solely to the GVL and are well recognised in videolaryngoscopy with other devices.15 The resistance could be managed by withdrawing the stylet by approximately 5 cm and withdrawing the blade by 1–2 cm as recommended by the manufacturer. All participants were given 5-minute instructions, including this withdrawing manoeuvre, on intubation with the GVL. This manoeuvre in manipulating instruments under indirect vision would, however, require more practice in order to become skilled at it. There are several tips that help learners improve rapidly with videolaryngoscopy, such as midline placement, insertion of the tube into the mouth before insertion of the videolaryngoscope and insertion of the tube as close as possible to the laryngoscope blade,14 so it is likely that the times to intubation could decrease with a short period of training.

In the cervical spine immobilisation scenario, the POGO score when using the GVL was significantly higher than that when using the Macintosh laryngoscope. This is probably due to the fact that the GVL is designed to provide a view of the glottis without alignment of the oral, pharyngeal and tracheal axes. This merit may reduce the upward lifting force required to expose the glottis and require less movement of the neck. Moreover, poor glottic exposure is more likely to require prolonged or multiple attempts that are associated with increased trauma to the airway and fatal complications.16^–18 This improvement of vocal cord visualisation was not, however, observed in the tongue oedema and combined scenarios. It seems that the 60° angle present in the GlideScope blade limits movement in the relatively small oral cavity. Consequently, it is difficult to place the blade tip in the optimal position on the tongue in the presence of oedema.

Despite their experience and familiarity with the Macintosh laryngoscope, participants preferred the GVL to the conventional Macintosh laryngoscope in the cervical spine immobilisation situation. This finding also supports the belief that novice users could easily be familiarised with the GVL.

In order to reflect the circumstance of most emergency physicians, we included participants who were familiar with using the conventional Macintosh laryngoscope but had no previous experience using the GVL. Nonetheless, our study showed that the performance of the GVL was similar to that of the Macintosh laryngoscope. These results could be changed if participants had been skilled in using the GVL or novice users were sampled for both devices.

Unanticipated difficult airways occur with a low incidence, although emergency physicians consistently encounter them. According to the widespread use and high success rate of rapid sequence intubation, emergency physicians are not familiar with alternative rescue devices that are as effective and easy to master. Reeder et al19 reported that most residency programmes have several rescue airway devices available in the emergency department but these alternatives are rarely utilised in actual practice. The authors suggested that residency programmes should consider using these alternative tools in primary airway control techniques to ensure the proficiency of these adjuncts. Therefore, the GVL could be considered not as an alternative device stocked in the emergency department but as a primary option.

There were several limitations to this study. First, we conducted this study using the Airsim, an anatomically correct airway manikin and therefore applying the findings presented here to actual clinical situations might be questionable. It is, however, impractical to recruit a sufficient number of patients with a consistent degree of airway difficulty and laryngoscopy grade in clinical practice. Furthermore, there are ethical and practical issues in conducting trials with new equipment on patients. Second, we did not divide the process of intubation into multiple parts and the time to best view and time to intubation were not recorded as this would determine where the delay in intubation with the GVL occurs. Consequently, we do not know why tracheal intubation took more time than predicted using the GVL in spite of a better glottic view. Finally, the success rates with the Macintosh laryngoscope were unexpectedly too high to reflect the degree of difficulty as we intended, although the POGO scores were different between the devices. We are aware that this is a preliminary study and further studies are needed to explore these limitations in manikins as well as in a clinical context.

CONCLUSIONS

This study suggested that the performance of the GVL was similar to that of the Macintosh laryngoscope for tracheal intubation in simulated models. The GVL, however, provided a better glottic view in cervical spine immobilisation airway applications when used by emergency physicians with no formal training. We consider the GVL to be a useful airway device to manage the difficult airway as well as the normal airway.