Thoracic surgery often requires one-lung ventilation with a double-lumen tracheal tube (DLT) for better surgical exposure. DLTs are larger in outer diameter and less compliant than single-lumen tubes. Tracheal intubation can therefore be challenging even in patients without anatomical airway difficulties.1
Videolaryngoscopes and indirect laryngoscopes are dedicated devices for managing difficult tracheal intubation and are recommended in the guidelines of the Difficult Airway Society for management of difficult tracheal intubation.2 They are more efficient than direct laryngoscopy in managing both non-difficult and difficult orotracheal intubation with DLTs3–6 but information is lacking about the most efficient device.
The Glidescope (Verathon Inc., Bothell, Washington, USA) and Airtraq (Prodol Meditec S.A., Vizcaya, Spain) are the two most commonly used devices in the United Kingdom and probably in many other countries.7 Both are reliable for DLT insertion.8–11 However, their design is very different. The Glidescope requires a rigid stilette inside the DLT to guide it through the glottis, whereas the Airtraq double lumen (DL) uses a guiding channel to advance a 35 to 41 French gauge (FG) DLT through the glottis. These different designs could lead to different efficacy and safety. Thus, it is important to compare them.
Two previous studies have compared the Glidescope and Airtraq for DLT insertion. In a manikin, simulating easy and difficult airways were not different between Glidescope and Airtraq DL in terms of successful DLT insertion rate, but the use of the Airtraq DL was associated with a longer intubation time.12 In patients without a predicted difficult airway, successful DLT insertion rates were not different between the two techniques, but use of the Airtraq DL required a shorter intubation time and was associated with less haemodynamic response.10 There is no previously published study comparing the efficacy and safety of the Glidescope and Airtraq DL for DLT insertion in patients with a predicted difficult airway. Our hypothesis was that the Glidescope would be superior to the Airtraq for DLT insertion in patients with a predicted or known difficult airway.
The study protocol was approved by the Institutional Research Board of Tours – Région Centre – Ouest 1 [no. 2014-S5 (2014-A00143-44)] on 18 February 2014 (chairperson Dr P Bertrand) and recorded on the national register of the French National Agency for Drug and Health Products Safety (no. 2014-A00143-44). The study complies with the CONSORT 2010 statement for randomised studies.
Patients were eligible for inclusion if they were adults scheduled for thoracic elective thoracic surgery requiring one-lung ventilation, had a predicted difficult intubation score of at least 7 (Arné risk index,13 Appendix 1) and gave informed written consent. During the preoperative assessment, one of the three investigators (all anaesthesiologists) documented the variables used in the calculation of the Arné risk index: known difficult intubation during previous anaesthesia, pathologies associated with difficult laryngoscopy or intubation (facial malformation, acromegaly, cervical spondylosis with limited neck movements, occipito-atlanto-axial disease, tumours of the airway and long-term diabetes mellitus with ‘stiff joint syndrome’), clinical symptoms of airway disorder (dyspnoea related to airway compression, dysphonia, dysphagia and sleep apnoea syndrome), Mallampati's modified test, inter-incisor gap and mandible luxation, thyromental distance measured with the head fully extended and maximum range of head and neck movement, as described by Wilson et al.14 The exclusion criteria were: pregnancy, impossible evaluation of the Arné risk index (non-contributive examination), predicted difficult mask ventilation (four or more risk factors, according to the Kheterpal score15) or mouth opening less than 2 cm (requiring a fibreoptic awake nasal intubation). Patients were recruited only if two of the three investigators (O.B., E.L. or Y.B.) were available on the day of surgery.
One of three anaesthesiologists (O.B., E.L. and Y.B.) performed all intubations. The Airtraq was our institutional device for difficult intubation with a single-lumen tube before the trial. Before beginning the study, we trained in the use of the Glidescope and Airtraq DL (the yellow one, dedicated to double-lumen tubes) for intubation, using a double-lumen tube on manikins. Before starting the study, each of the three anaesthesiologists had inserted at least 10 DLTs with each device on a manikin devoted to normal intubation training.16
Patient monitoring included ECG, non-invasive blood pressure measurement, pulse oximetry, Bispectral Index (BIS) and neuromuscular response to train-of-four (TOF) stimulation at the orbicularis oculi (contraction of eyebrow). Patients were assigned randomly to the Glidescope group or the Airtraq group by opening a sealed envelope in the operating theatre. The random allocation sequence was generated by a computer in blocks of four (http://biostat.med.univ-tours.fr/mediawiki/index.php/Cat%C3%A9gorie:Randomisations. Criteria used: strata = 1, groups = 2, number of patients per strata = 72 and balancing = 2) All envelopes were prepared, sealed and sequentially numbered before the beginning of the study by FE.
The Airtraq is an indirect laryngoscope; we used its specific Airtraq clippable camera system (ref 0VCA 301, Prodol Meditec S.A., Vizcaya, Spain), allowing video transmission to a 14-in Sony trinitron colour video monitor (model no. PVM-14NSMDE, Sony Corp., Pencoed, UK). Therefore, all tracheal intubations performed with the Airtraq DL were managed with an external screen that was positioned exactly at the same place as the screen on the Glidescope Digital, DVD-quality 6.4-in colour monitor. We managed to achieve the same visualisation conditions for both devices, allowing coordination of eyes and hands to be similar.
During induction of anaesthesia, patients were in a supine neutral position with a 4-cm high gelatin headrest pillow. After preoxygenation (end-tidal O2 > 90%), anaesthesia was induced by intravenous injection of sufentanil 0.15 μg kg−1 and propofol 2 to 3 mg kg−1. Once mask ventilation was established, rocuronium 1.2 mg kg−1 was injected. When the orbicular TOF ratio showed no response and BIS was less than 50, tracheal intubation was performed with the allocated device, using a left Robertshaw DLT (Bronchopart RUSCH (Bronchopart®, Teleflex, Research Triangle Park, NC), 35 or 37 FG for women, 39 or 41 FG for men) lubricated with silicon spray. When using the Glidescope, the malleable stilette was used to bend the DLT, so that the curve of the DLT followed that of the blade. We performed intubations with plastic single-use blades nos. 3 and 4, the latter dedicated to obese patients (BMI > 31 kg m−2). When using the Airtraq DL, the stilette was removed from the DLT before it was positioned inside the lateral conduit, as recommended by the manufacturer. The tip of the DLT was introduced through the vocal cords and advanced into the trachea; the stilette (Glidescope group) was removed and the tube was rotated 90° counterclockwise to catheterise the main left bronchus. At any moment, the anaesthesiologist could ask for the use of the ‘Backwards Upwards Rightwards Pressure’ (BURP) manoeuvre.
The correct position of the DLT was determined by the presence of three capnograms, auscultation of both lungs before and after selective clamping of the tracheal and bronchial lumens, and fibreoptic bronchoscopy. An attempt was considered to be a success if the DLT was correctly positioned and if there was no desaturation (SpO2 < 92%). If it was impossible to introduce the bronchial cuff through the vocal cords or to rotate the DLT, or if SpO2 less than 92% occurred, the attempt was considered to be a failure and mask ventilation was re-established; once SpO2 was more than 98%, a new tracheal intubation, using the same device, was attempted by a second anaesthesiologist. In the event of a second failure, the second anaesthesiologist switched to the other device. If DLT insertion failed using both videolaryngoscopes, tracheal intubation was performed using a single-lumen tube with a blocker (Univent Tube; Fuji Systems, Tokyo, Japan).
The primary outcome was the success rate of tracheal intubation after two attempts.
Secondary outcomes were: success rate of tracheal intubation on the first attempt; duration of the intubation procedure (defined as the time from when the videolaryngoscope entered between the patient's lips until confirmation of successful tracheal intubation by the three capnograms); number of attempts; need for BURP; Cormack and Lehane grade seen on monitors during the successful attempt (or the best one in case of intubation failure); need for the alternative device; need for a bronchial blocker; dental trauma; or presence of blood on the videolaryngoscope after intubation.
On the first postoperative day, patients were asked to score the degree of sore throat using a visual analogue scale from 0 (no pain) to 10 (unbearable pain) and to report postoperative hoarseness.
At the time of study design, no comparable study was available for DLT insertion using Glidescope and Airtraq DL. The sample size was calculated using online BiostaTGV epiR package (https://cran.r-project.org/web/packages/epiR/) 0.9 to 30, based on a study that compared the Glidescope and Airtraq for tracheal intubation in obese patients with a single-lumen tube.17 A 100% success rate of tracheal intubation had been reported for the Glidescope versus 80% for the Airtraq. An a-priori power analysis revealed that 35 patients per group were needed to detect such a difference with a power of 0.8, at an α-level of 0.05 (two-tailed). Data are presented as number (proportion), mean ± SD or median inter-quartile range. Data were analysed using R v2.12.1 (Vienna, Austria) (Free Software Foundation's GNU General Public License. R Development Core Team. A language and environment for statistical computing. R Foundation for Statistical Computing: Vienna, Austria, 2006). Groups were compared using Student's t test, χ2 test or Fisher's exact test as required. All the variables were checked for normal distribution prior to use of parametric tests. The result was considered to be statistically significant if P was less than 0.05.
Between March 2014 and March 2015, 277 patients were scheduled for elective thoracic surgery requiring DLT insertion. Among the 277 scheduled patients, 78 were predicted to have a difficult airway. Finally, 72 patients were enrolled (Fig. 1). The baseline characteristics of the patients (Table 1) were similar. However, there were fewer patients with an Arné risk index of at least 11 (high risk of difficult intubation) in the Glidescope group than in the Airtraq group (P = 0.03) (Table 2).
The success rate for tracheal intubation after two attempts was 86% (31/36) in the Glidescope group and 94% (34/36) in the Airtraq group (P = 0.43). The success rate for tracheal intubation at the first attempt was 80% in the Glidescope group and 77% in the Airtraq group (P = 0.70). Insertion of the DLT, using the first device, was unachievable in seven patients, five using the Glidescope and two using the Airtraq. These seven patients had an Arné risk index less than 11. Two of the seven were successfully managed by using the alternative device (one in each group). The remaining five patients were intubated using a Univent tube (Fig. 2); four of these five were successfully intubated using a Macintosh laryngoscope, the insertion of an Eschmann bougie (15 FG × 70 cm) allowing subsequent Univent tube placement. We failed in performing the same procedure for the fifth patient, who suffered from a supraglottic tumour; in this case, a size 5 laryngeal mask airway Fastrach was inserted, then an Eschmann bougie, followed by its exchange for a Univent tube. Three patients had a higher Cormack and Lehane grade. Two out of three had a moderate risk of difficult intubation (scores 7 and 9). The third one had a high risk (score 11).
All nine patients who had a history of difficult intubation were successfully intubated.
None of the other secondary outcomes differed between groups (Table 3).
In this randomised study of patients with a predicted risk of difficult intubation and needing a tracheal intubation with a DLT, Glidescope and Airtraq DL had the same rate of successful tracheal intubation.
Five previous clinical studies have compared single-lumen tube insertion using the Glidescope and Airtraq in adults.17–21 One additional clinical study compared these devices for the insertion of DLTs in patients without a predicted difficult airway.10 In five out of six of these reports, as in our study, the success rate for tracheal intubation was high and did not differ between the devices.10,12,21 In the sixth study,17 the Glidescope had a higher success rate than the Airtraq for single-lumen tube insertion in obese patients (100 versus 80%). The authors of this study attributed this difference to the shorter length of the handle of the Glidescope compared with the Airtraq. The short handle eased intubation in several cases. In non-obese patients, this advantage may disappear. The ‘handle length’ effect in obese patients may explain why a difference between the devices was not observed in the five other studies or in ours, which included only a minimal number of obese patients. In the only clinical study using DLTs,10 the success rate of tracheal intubation was 100% for both devices. However, the studied population had no pre-anaesthetic risk of difficult airway, in contrast to our study, which may explain the intubation failures which we observed with both devices.
The success rate of tracheal intubation was chosen as the primary outcome. Indeed, we considered, like other authors,20,22 that it was more relevant than mean intubation time. In our study, the difference of 17 s in average intubation time between the Glidescope and Airtraq DL, favouring the Glidescope, is not statistically significant and not clinically relevant. In addition, we found no difference in complications between the devices, as in previous studies. Of the six clinical studies that compared these devices, the complication rate was not different in two studies, as in ours.17,20 The Airtraq was less traumatic in three studies10,19,21 but more traumatic in the study which included patients with airway tumours.18 To summarise, both devices seem to be efficient for orotracheal intubation, with a possible better performance of the Glidescope in obese patients, and a potentially less traumatic profile for the Airtraq except when an airway tumour is suspected.
One original feature of the current study is the selected population: patients with documented high risk of difficult airway. Five out of six clinical studies that compared the Glidescope and Airtraq included patients with a predicted normal airway,10,21 simulated difficult airway,20 obese patients without a predicted difficult airway17 or face-to-face intubation.19 Only one study included patients with a suspected tumour of the upper airway18 but it was restricted to single-lumen tube insertion. As in our study, tracheal intubation was successful in most cases in both groups (100% with a Glidescope versus 93% with an Airtraq).
In our study, despite randomisation, more patients had a high risk of difficult intubation according to the Arné score (score ≤11) in the Airtraq group than in the Glidescope group. None of these patients had an intubation failure. Hence, the difference in the Arné score between both groups may have had no impact on the success rate of tracheal intubation.
There are several limitations to our study. First, 10 manikin training intubations with each device may be considered insufficient; however, all patients with a history of difficult intubation were intubated successfully. Second, a total number of 72 patients may be insufficient to detect clinically relevant differences. We calculated the number of patients to be included using a previous study which focused on tracheal intubation in obese patients. Two major points are different in our study; our patients had criteria predicting difficult intubation but were not all obese, and we used a DLT. However, the small difference that was observed could be used to calculate a hypothetical number of patients to be included to observe a significant difference on the principal criteria of judgement. Such calculations made from our results show that it would be necessary to include around 200 patients per group to find a significant difference in the overall success rate of tracheal intubation after two attempts between both groups. Third, using the Arné index may be debatable. Numerous scores are available to predict difficult intubation but none has gained global agreement. The Arné score was one of the most exhaustive clinical and multivariate risk indexes at our disposal for preoperative assessment of difficult intubation13; we used it because it is employed widely in our institution. No score has 100% specificity and the Arné score has a specificity of 96% in general surgery. However, it has not been validated for DLT insertion; therefore, it is possible that its specificity is lower in such cases, which could explain why we encountered patients with predictive signs of difficult intubation but whose intubation was easy. Fourth, regarding monitor sizes, the difference between devices (14 versus 6.4 in) was not clinically relevant as both systems were high-quality colour monitors.
There were two oesophageal intubations when using the Airtraq (Cormack and Lehane grades 1 and 2) and some intubations which failed with both devices despite correct visualisation of the glottis or because of impossible rotation of the tube. Our few failures can be explained by the inherent constraints related to the way the videolaryngoscope was used and by the characteristics of the DLTs themselves. Although bending the stilette of the DLT is very important to success with the Glidescope, it may not be sufficient especially in patients in whom directing the DLT into the glottis is very challenging (small mouth opening, large tongue, Cormack and Lehane grade 3 or 4). Furthermore, this specific and exaggerated angulation may be responsible for a conflict between the DLT's bronchial cuff and the ventral tracheal wall or arytenoids. The Airtraq has a lateral conduit through which the DLT can be passed and guided to the glottis. Although this conduit helps to lead the tube to the glottis, it makes rotation of the DLT almost impossible without separating the DLT from the Airtraq, as it firmly grips it in its conduit, risking involuntary extubation followed by oesophageal intubation. Furthermore, having to separate the DLT from the Airtraq often results in losing the alignment of oral, pharyngeal and tracheal axes, making progression of the DLT into the trachea very difficult. To avert this difficulty, we propose softly lifting the jaw with the non-dominant hand with the aim of re-establishing the lost alignment and directing the DLT into the trachea.
We had one failure at the first attempt due to a Glidescope ‘shutdown’; the backup battery was low because it was stored unplugged after its last use. Because the patient was then intubated without difficulty on the second attempt, we did not exclude the patient from further analysis as our primary outcome was success rate after two attempts.
In conclusion, the current study compared two widely used devices.23 Both the Glidescope and Airtraq DL are suitable for DLT insertion when facing a risk of difficult intubation.
Acknowledgements relating to this article
Assistance with the study: the authors thank John Gage, MD and Polly Gobin for editorial assistance with French-English translation.
Financial support and sponsorship: the Glidescope device was lent for the duration of the study and blades were donated for research purposes free of charge by Verathon. The Airtraq DL devices were donated for research purposes free of charge by Vygon.
Conflicts of interest: none.
Appendix 1 Arné et al.13 clinical multivariate risk index for preoperative assessment of difficult intubation
1. Hsu HT, Chou SH, Wu PJ, et al. Comparison of the GlideScope videolaryngoscope and the Macintosh laryngoscope for double-lumen tube intubation. Anaesthesia
2. Frerk C, Mitchell VS, McNarry AF, et al. Difficult Airway Society intubation guidelines working group. Difficult Airway Society 2015 guidelines for management of unanticipated difficult intubation in adults. Br J Anaesth
3. Bensghir M, Alaoui H, Azendour H, et al. Faster double-lumen tube insertion with the videolaryngoscope than with a standard laryngoscope. Can J Anaesth
4. Lin W, Li H, Liu W, et al. A randomised trial comparing the CEL-100 videolaryngoscope with the Macintosh laryngoscope blade for insertion of double-lumen tubes. Anaesthesia
5. Purugganan RV, Jackson TA, Heir JS, et al. Videolaryngoscopy versus direct laryngoscopy for double-lumen endotracheal tube intubation: a retrospective analysis. J Cardiothorac Vasc Anesth
6. Suppan L, Tramèr MR, Niquille M, et al. Alternative intubation techniques vs Macintosh laryngoscopy in patients with cervical spine immobilization: systematic review and meta-analysis of randomized controlled trials. Br J Anaesth
7. Gill RL, Jeffrey AS, McNarry AF, et al. The availability of advanced airway equipment and experience with videolaryngoscopy in the UK: two UK surveys. Anesthesiol Res Pract
8. Russell T, Slinger P, Roscoe A, et al. A randomised controlled trial comparing the Glidescope and the Macintosh laryngoscope for double-lumen endobronchial intubation. Anaesthesia
9. Wasem S, Lazarus M, Hain J, et al. Comparison of the Airtraq and the Macintosh laryngoscope for double-lumen tube intubation. Eur J Anaesthesiol
10. Yi J, Gong Y, Quan X, et al. Comparison of the Airtraq and the GlideScope for double-lumen tube intubation in patients with predicted normal airways: a prospective randomized trial. BMC Anesthesiol
11. Chastel B, Perrier V, Germain A, et al. Usefulness of the Airtraq DL™
videolaryngoscope for placing a double-lumen tube. Anaesth Crit Care Pain Med
12. El Tahan MR, Al’ghamdi AA, Khidr A, et al. A comparison of three videolaryngoscopes for double-lumen tubes intubation in simulated easy and difficult airways: a randomised trial. Minerva Anestesiol
13. Arné J, Descoins P, Fusciardi J, et al. Preoperative assessment for difficult intubation in general and ENT surgery: predictive value of a clinical multivariate risk index. Br J Anaesth
14. Wilson ME, Spiegelhalter D, Robertson JA, et al. Predicting difficult intubation. Br J Anaesth
15. Kheterpal S, Han R, Tremper KK, et al. Incidence and predictors of difficult and impossible mask ventilation. Anesthesiology
16. 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
17. Putz L, Dangelser G, Constant B, et al. Prospective trial comparing Airtraq and Glidescope techniques for intubation of obese patients. Ann Fr Anesth Reanim
18. Lange M, Frommer M, Redel A, et al. Comparison of the Glidescope and Airtraq optical laryngoscopes in patients undergoing direct microlaryngoscopy. Anaesthesia
19. Arslan ZI, Alparslan V, Ozdal P, et al. Face-to-face tracheal intubation in adult patients: a comparison of the Airtraq™, Glidescope™ and Fastrach™ devices. J Anesth
20. Kleine-Brueggeney M, Greif R, Schoettker P, et al. Evaluation of six videolaryngoscopes in 720 patients with a simulated difficult airway: a multicentre randomized controlled trial. Br J Anaesth
21. Al-Ghamdi AA, El Tahan MR, Khidr AM. Comparison of the Macintosh, GlideScope®
, and King Vision™
laryngoscopes in routine airway management. Minerva Anestesiol
22. Serocki G, Neumann T, Scharf E, et al. Indirect videolaryngoscopy with C-MAC D-Blade and GlideScope: a randomized, controlled comparison in patients with suspected difficult airways. Minerva Anestesiol
23. Asai T. Videolaryngoscopes: do they truly have roles in difficult airways? Anesthesiology