Tracheal intubation is a vital skill for anaesthesiologists in daily practice. Since direct laryngoscopy was introduced in the 1940s to facilitate tracheal intubation with visualisation of the glottic entrance, it has become the standard device and technique.1 Despite rapid improvements in medical technology, tracheal intubation around the world is still accomplished using this traditional method. However, the viewing angle of the direct laryngoscope has been measured at 15° and is restricted by oropharyngeal structures and the position of the patient.2 Intubation using direct laryngoscopy can be a challenging task when a difficult airway (e.g. patients with cervical spine immobilisation) or challenging conditions (e.g. confined spaces) are encountered. Many techniques such as fibreoptic laryngoscopes, laryngeal tube and laryngeal mask airway have been developed as alternative airway management strategies for difficult intubation, but none of them has proved to be a suitable substitute for tracheal intubation by direct visualisation of the larynx.
In recent years, several types of video laryngoscopes have been introduced into clinical practice. The prices are much higher (e.g. $6500 to $7000 for the GlideScope) than for a standard laryngoscope. All of them share similar features: they consist of a handle and laryngoscope blade with a video camera built into the end of the blade. The images are displayed on a video system such as a liquid crystal display (LCD). As a result, video laryngoscopes allow a wider viewing angle and make alignment of the oral, pharyngeal and tracheal axes unnecessary.3 Moreover, their ease of use, short learning curves4,5 and flexibility make them potential substitutes for conventional direct laryngoscopy.6,7
Several trials have been published comparing video laryngoscopes with direct laryngoscopy for tracheal intubation. However, the results were varied and there has been no meta-analysis of the results. Consequently, we conducted the first meta-analysis based on the published outcomes of randomised trials.
Data sources and search strategy
We searched PubMed and EMBASE (up to 24 September 2010) with no language restrictions. Keywords (‘video’, ‘laryngoscope’ and ‘intubation’), Medical Subject Headings (MeSH) (‘laryngoscopes’, ‘videotape recording’ and intubation’) and Emtree terms (‘laryngoscope’, ‘videorecording’ and ‘respiratory tract intubation’) were used to achieve the results in a Boolean search strategy. References of retrieved articles were also examined for relevant publications.
We included studies comparing video laryngoscopes with direct laryngoscopy regarding time to intubation (TTI), the success rate of intubation and quality of the glottic view. Either device can be used as a tool for oral intubation. Among the publications identified, we considered only randomised clinical trials with at least two independent groups. We excluded observational studies because they are subject to unmeasured confounding factors. Manikin studies were excluded because they may not completely simulate the real situation. Studies with crossover design were excluded because the ‘carry-over’ effects cannot be avoided when repeated intubations were performed during a short period. Finally, we excluded non-original studies (e.g. case reports, editorials, review articles, etc.).
Two independent investigators (SYC and LYK) extracted the data from eligible articles using a standardised form. Different perspectives were resolved by group discussions. The corresponding authors of three trials2,8,9 were contacted to provide additional data on study outcomes for meta-analysis. Variables that were assessed included country; publication year; study outcomes (comparison of the glottic view, TTI and success rate for intubation); sex; types of video laryngoscopes; mean age; intubation scenario (normal situation or difficult intubation); intubators (experienced anaesthesiologists only, or not); and sample sizes of both groups.
Glottic views were examined based on the Cormack and Lehane grade or percentage of glottis opening (POGO) score and were recorded as ‘improved’ or ‘not improved’ based on the comparison between groups using the statistical tests in each trial. Definitions of the TTI and failed intubation were specified. Every procedure was recorded as a successful intubation or a failed intubation (binary outcome) in each trial. TTI was recorded as mean and standard deviation (SD). In three trials, only median times with either interquartile range2,8 or range9 were available. The authors were contacted and the original dataset was retrieved from one corresponding author for computing the estimates.9 For another two trials,2,8 the measured time was converted into mean and SD under the assumption of normal distribution.
Outcome data for successful intubation were summarised using basic descriptive statistics (counts and proportions). Successful intubation rate ratios were calculated and pooled using DerSimonian and Laird random effects models.10 Because the definitions of TTI varied among trials, comparisons between video laryngoscopes and direct laryngoscopy were summarised using the standardised mean difference.11 Between-study heterogeneity was evaluated with the I2 statistic.12 Both the Egger test and Begg test were applied for assessment of potential publication bias.13,14 We also conducted sensitivity analysis to evaluate the influence of each study on the overall pooled estimate.
Meta-regression was then performed to explore possible heterogeneity based on predefined study variables, including intubators, age group (paediatric or adult), intubation scenario and types of video laryngoscope. The Monte Carlo permutation test for single covariate meta-regressions was applied to the adjusted results. Significant variables identified were further assessed with subgroup analyses.
Analyses were all conducted using STATA version 11.0 (StataCorp, College Station, Texas, USA). All statistical tests were two-sided and were considered significant when the P value was 0.05 or less.
Based on the initial search results, 260 titles and abstracts were examined, and among these, we discarded eight publications which were identified by both databases. We rejected 94 because they were not relevant to our study focus. Of the remaining articles, 149 did not meet our inclusion criteria. Two additional publications were retrieved by hand search of the references. Therefore, 11 trials were included in our analysis.2,3,8,9,15–21 The selection process is summarised in Fig. 1.
Description of studies
The analysis included 11 trials2,3,8,9,15–21 with a total of 1196 participants. All trials were conducted in the operating room setting. Participants were patients who were to undergo scheduled surgical interventions, and anaesthesiologists (either attending physicians who were experienced with video laryngoscopes and direct laryngoscopy, or novices) performed the intubations. The characteristics of the trials are summarised in Table 1.
Comparison of glottic views
Nine of the 11 trials compared the glottic view between video laryngoscopy and direct laryngoscopy during intubation.2,8,9,15–17,19–21 The Cormack and Lehane grade was used for comparison in all nine trials; the POGO score was also applied in two trials.2,15 Significantly improved glottic views were obtained under video laryngoscopy compared to direct laryngoscopy in eight of the nine trials;2,8,9,16,17,19–21 in the trial conducted by Komatsu et al.,15 the grade of glottic view was similar in each group. Intubation-related trauma, if mentioned,15,21 was uncommon and did not cause serious adverse consequences.
Comparison of successful intubation
The success rates for tracheal intubation were computed and pooled to evaluate possible heterogeneity. Despite different definitions of failed intubation among trials, the success rate was nearly 100% in both groups in every trial (Fig. 2 and Table 2). The pooled successful intubation rate ratio was 1.0 [95% confidence interval (CI) 0.99–1.01]. The I2 statistic was 0.0% with a P value of 0.608, indicating identical outcomes among trials.
Comparison of time to intubation
Using standardised mean differences, the pooled mean difference in TTI between video laryngoscopy and direct laryngoscopy was 0.26 (95% CI −0.27–0.78; P = 0.341; Fig. 3 and Table 3). The result revealed that there was no statistical difference in TTI between the groups. However, the I2 statistic was 94.6% with a P value less than 0.0001, indicating heterogeneity among trials.
A sensitivity analysis regarding TTI was conducted by removing one trial at a time. The pooled result seemed to be robust. For example, removing the most influential study conducted by Komatsu et al.15 changed the pooled estimate from 0.26 to 0.43 (95% CI −0.04–0.89; P = 0.072; Fig. 4).
Both the Egger test and Begg test were performed and the P values were 0.671 and 0.876, respectively. These results indicate there was no statistical evidence of publication bias.
Subgroup analysis and meta-regression
The results were homogeneous for comparison of successful intubation (P = 0.608) and significantly heterogeneous for comparison of TTI (P < 0.0001). We explored factors which might explain this heterogeneity by subgroup analysis and meta-regression. Factors assessed in our meta-regression model included age group (paediatric or adult), intubators (experienced anaesthesiologist only, or not), difficult intubation (yes or no) and types of video laryngoscope (Glidescope, X-Lite, Storz, Pentax-AWS and McGrath). Although the basic airway anatomies of adults and children are quite distinct, we found that neither age group nor type of video laryngoscope influenced TTI. However, there were only 11 trials in our study and the power may be low. Difficult intubation was the only factor which significantly affected TTI. In a subgroup analysis, the pooled standardised mean difference was −0.75 (95% CI −1.24 to −0.25) for trials with difficult intubation and 0.86 (95% CI 0.59–1.13) for trials without difficult intubation. Thus, video laryngoscopes may be associated with a shorter TTI in a setting of difficult intubation but a longer TTI in a straightforward intubation. Trials that included difficult intubation scenarios were conducted only among adults but the result still holds even when trials which focused on paediatric groups were excluded (Fig. 5). The analyses are summarised in Table 4.
To our knowledge, this is the first meta-analysis of randomised trials evaluating the video laryngoscope in comparison with direct laryngoscopy. Video laryngoscopes are designed to offer a better view of the glottis without the need to align oral, pharyngeal and tracheal axes. In most of the trials included in our meta-analysis, the view of glottis was either significantly improved or similar using the video laryngoscope when compared with direct laryngoscopy.2,8,9,15–17,19–21 The potential benefits of an improved view of the glottis are that adverse events associated with a blind intubation may be avoided and that novices may gain a better understanding of the anatomy of the airway.6,7,22–27
Although the definitions of failed intubation were different among studies, we found an extremely high successful intubation rate (almost 100%) in both video laryngoscope and direct laryngoscopy groups. The success rate was essentially the same in both groups, and the results were consistent across trials. This is expected because participants enrolled in the studies were patients scheduled for elective surgery, and all the intubations were undertaken in a well prepared situation (operating room) by anaesthesiologists who were familiar with airway management. Although some other studies had been conducted either in the emergency room28 or a prehospital setting,29 they were not randomised trials and were, therefore, not included in our analysis. As a result, the pooled outcome may not be generalised to locations other than the operating theatre suite.
The pooled TTI was slightly longer in the video laryngoscope group, but the difference was not statistically significant. The increased time might result from the technique required to manipulate the tracheal tube in order to pass it through the vocal cords and did not appear to increase adverse outcomes, such as deoxygenation or airway trauma.20 In addition, anaesthesiologists enrolled in all trials were more familiar with intubation under direct laryngoscopy. Although video laryngoscopes have shorter learning curves,4,5 the increased TTI may still be the result of less experience with video laryngoscopes.
The results relating to TTI were highly heterogeneous, and the meta-regression model may not have had enough power to evaluate some possible factors such as specific types of video laryngoscope. Nevertheless, the difficult intubation scenario may, at least partly, explain this heterogeneity. In our subgroup analyses, video laryngoscopy was associated with a significantly shorter TTI in a difficult intubation setting but a longer TTI when intubation was not difficult. Because a good view of the glottis cannot always be obtained easily with direct laryngoscopy, it is not surprising that video laryngoscopes appear to be advantageous when difficulty is encountered using conventional techniques. The difficult airway is always challenging for clinicians and tracheal intubation can cause airway trauma and even life-threatening incidents if not managed appropriately.30 The potential benefits found in our analysis indicate that video laryngoscopes are a good option when tracheal intubation is difficult.
We need to acknowledge some limitations of our study. First, the time to visualise the vocal cords and the time from visualisation to intubation were recorded only in the trial conducted by Vlatten et al.2 As a result, we cannot speculate as to whether an angulated or a curved video laryngoscope may result in more rapid tracheal intubation. Second, none of the randomised trials included in our analysis was double-blinded because it was impossible to make the anaesthesiologist unaware of the devices they would use for intubation. Third, the costs of video laryngoscopes are much higher than those of standard laryngoscopes. However, due to the limited data, we could not conduct a cost-effectiveness analysis based on the costs of intubation and the number of complications. Fourth, the high heterogeneity among the selected trials made it difficult to pool all data together. In order to evaluate the validity of video laryngoscopes, we propose that determinants such as intubators, devices (curved or angulated video laryngoscopes), definitions of TTI and failed intubation, and number of attempts should be more clearly specified in future studies (Table 5). Finally, tracheal intubation in locations other than the operating theatre suite (e.g. medical wards or emergency departments) are usually undertaken as urgent, emergency procedures and it would be difficult or impossible to obtain informed consent for a randomised trial. In order to enrol high-quality trials, we had strict inclusion criteria. As a result, all randomised trials included in our meta-analysis were conducted in the operating theatre suite and intubations were performed by anaesthesiologists.
Based on the randomised trials that we included, our meta-analysis found that, compared with direct laryngoscopy, the video laryngoscope achieved a better view of the glottis, a similarly high rate of successful intubation and a shorter TTI when difficulty is encountered.
The authors would like to thank Dr Dan Baraclough for assistance with the study.
There was no financial support or sponsorship for this work.
None of the authors has any conflict of interest.
1. Sakles JC, Rodgers R, Keim SM. Optical and video
laryngoscopes for emergency airway management. Intern Emerg Med
2. Vlatten A, Aucoin S, Litz S, et al. A comparison of the STORZ video
laryngoscope and standard direct laryngoscopy
in the pediatric airway: a randomized clinical trial. Paediatr Anaesth
3. Redel A, Karademir F, Schlitterlau A, et al. Validation of the glidescope video
laryngoscope in pediatric patients. Paediatr Anaesth
4. Kaplan MB, Ward DS, Berci G. A new video
laryngoscope: an aid to intubation
and teaching. J Clin Anesth
5. Savoldelli GL, Schiffer E, Abegg C, et al. Learning curves of the Glidescope, the McGrath and the Airtraq laryngoscopes: a manikin study. Eur J Anaesthesiol
6. You JS, Park S, Chung SP, et al. The usefulness of the GlideScope video
laryngoscope in the education of conventional tracheal intubation
for the novice. Emerg Med J
7. Low D, Healy D, Rasburn N. The use of the BERCI DCI video
laryngoscope for teaching novices direct laryngoscopy
and tracheal intubation
8. Walker L, Brampton W, Halai M, et al. Randomized controlled trial of intubation
with the McGrath Series 5 videolaryngoscope by inexperienced anaesthetists. Br J Anaesth
9. Macnair D, Baraclough DAN, Wilson G, et al. Pediatric airway management: comparing the Berci-Kaplan video
laryngoscope with direct laryngoscopy
. Paediatr Anaesth
10. DerSimonian R, Laird N. Meta-analysis
in clinical trials. Control Clin Trials
11. Egger M, Smith GD, Altman DG. Systematic reviews in healthcare: meta-analysis in context
. 2nd ed. London, England: BMJ Books; 2001. pp. 285–312.
12. Higgins JP, Thompson SG, Deeks JJ, et al. Measuring inconsistency in meta-analyses. BMJ
13. Begg CB, Mazumdar M. Operating characteristics of a rank correlation test for publication bias. Biometrics
14. Egger M, Davey Smith G, Schneider M, et al. Bias in meta-analysis
detected by a simple, graphical test. BMJ
15. Komatsu R, Kamata K, Sessler DI, et al. Airway Scope and Macintosh laryngoscope for tracheal intubation
in patients lying on the ground. Anesth Analg
16. Jungbauer A, Schumann M, Brunkhorst V, et al. Expected difficult tracheal intubation
: a prospective comparison of direct laryngoscopy
and video laryngoscopy
in 200 patients. Br J Anaesth
17. Kim JT, Na HS, Bae JY, et al. GlideScope video
laryngoscope: a randomized clinical trial in 203 paediatric patients. Br J Anaesth
18. Xue FS, Zhang GH, Li XY, et al. Comparison of hemodynamic responses to orotracheal intubation
with the GlideScope videolaryngoscope and the Macintosh direct laryngoscope. J Clin Anesth
19. Marrel J, Blanc C, Frascarolo P, et al. Videolaryngoscopy improves intubation
condition in morbidly obese patients. Eur J Anaesthesiol
20. Sun DA, Warriner CB, Parsons DG, et al. The GlideScope video
laryngoscope: randomized clinical trial in 200 patients. Br J Anaesth
21. Lim Y, Yeo SW. A comparison of the GlideScope with the Macintosh laryngoscope for tracheal intubation
in patients with simulated difficult airway. Anaesth Intensive Care
22. Ray DC, Billington C, Kearns PK, et al. A comparison of McGrath and Macintosh laryngoscopes in novice users: a manikin study. Anaesthesia
23. Sadamori T, Kusunoki S, Otani T, et al. Airway Scope for emergency intubations: usefulness of a new video
-laryngoscope. Hiroshima J Med Sci
24. Howard-Quijano KJ, Huang YM, Matevosian R, et al. Video
-assisted instruction improves the success rate for tracheal intubation
by novices. Br J Anaesth
25. Boedeker BH, Hoffman S, Murray WB. Endotracheal intubation
training using virtual images: learning with the mobile telementoring intubating video
laryngoscope. Stud Health Technol Inform
26. Iwase Y, Matsushima H, Nemoto M, et al. The efficacy of video
intubating laryngoscope for novice residents. Masui
27. Levitan RM, Goldman TS, Bryan DA, et al. Training with video
imaging improves the initial intubation
success rates of paramedic trainees in an operating room setting. Ann Emerg Med
28. Wayne MA, McDonnell M. Comparison of traditional versus video laryngoscopy
in out-of-hospital tracheal intubation
. Prehosp Emerg Care
29. Brown CA, Bair AE, Pallin DJ, et al. Improved glottic exposure with the Video
Macintosh Laryngoscope in adult emergency department tracheal intubations. Ann Emerg Med
30. Langeron O, Amour J, Vivien B, et al. Clinical review: management of difficult airways. Crit Care
Keywords:© 2011 European Society of Anaesthesiology
intubation; laryngoscopy; meta-analysis; video