A Comparative Trial of the GlideScope® Video Laryngoscope to Direct Laryngoscope in Children with Difficult Direct Laryngoscopy and an Evaluation of the Effect of Blade Size

Abstract

BACKGROUND: 

GlideScope® video laryngoscope (GVL) has been proposed to be useful for airway management, but its efficacy for difficult airways has not been confirmed in pediatric patients. In this study, we evaluated the usefulness of the GVL for improving the laryngoscopic view in patients whose Cormack and Lehane grade (C&L grade) was ≥3 under direct laryngoscopy. We also assessed the effect of GVL blade size on the laryngoscopic view.

METHODS: 

This randomized open trial was conducted in a tertiary pediatric center. Patients whose previous C&L grade was ≥3, or who were anticipated to have a difficult airway, were enrolled. The initial modified C&L grade was scored using a direct laryngoscope (DL). If the patient’s C&L grade was ≥3, the laryngoscopic view was scored again using GVLw (selected based on weight) and GVLs (1 size smaller than GVLw) in random order by a single experienced anesthesiologist. All laryngoscopic views were graded both with and without the backward, upward, and right lateral displacement of the thyroid cartilage (BURP) maneuver. The primary outcome was the difference in the C&L grade between DL and GVLw, and the secondary outcome was that between GVLw and GVLs. For statistical analysis, the modified C&L grade was converted to an ordinal scale.

RESULTS: 

Data from 23 pediatric patients were analyzed. When compared with DL, improvement of laryngoscopic view with the GVLw wa not obvious (98.3% confidence interval [CI] for differences of ordinal scale, 0–1 without BURP, P = 0.15 and 0–1 with BURP, P = 0.11). However, GVLs improved the laryngoscopic view in comparison with both DL (98.3% CI for differences, 3.5–5.0 without BURP, P = 0.00007 and 3.5–4.5 with BURP, P = 0.0001) and GVLw (98.3% CI for differences, 3.0–4.5 without BURP, P = 0.00007 and 2.5–4.0 with BURP, P = 0.0001). There was no adverse outcome during this study.

CONCLUSIONS: 

In patients with C&L grade ≥3 under direct laryngoscopy, GVLs significantly improved the laryngoscopic view when compared with DL or GVLw. The GVLs is recommended for improving the laryngoscopic view in patients with a difficult airway.

In Brief

Published ahead of print May 17, 2013

GlideScope^® video laryngoscope (GVL; Verathon Medical, Bothell, WA) is a device for intubation that has a digital camera located at the distal end of blade. The usefulness of GVL has been shown in several studies.^1–7 Improvement in the glottis view of GVL compared with direct laryngoscopy is considered to be due to its 60° blade angulation. The camera on the blade provides a more anterior laryngeal view that may be invisible using a direct laryngoscope (DL).

Previous studies have demonstrated that GVL has potential advantage for tracheal intubation in adults with difficult airways.^2,6–9 However, there is limited evidence concerning the superiority of a GVL over a DL in children because of an insufficient number of studied patients whose C&L grades ≥3 with backward, upward, and right lateral displacement of the thyroid cartilage (BURP).^1,5 Selection of adequately sized GVL blades is important especially in children with difficult airways due to their variable airway sizes. However, there has been no evaluation of the effect of size of the GVL blade on the laryngoscopic view.

We assumed that a 1-size smaller GVL blade (GVLs) could provide a better laryngoscopic view than a GVL selected by weight (GVLw) because GVLs could be inserted deeper over the tongue during rotation. The camera angle would be optimized by inserting the blade further with the blade tip directed toward the larynx. Furthermore, the location and the angle of the camera differ according to GVL blade size. Considering that patients with difficult airways have hypognathia, a more cephalad larynx, and limited neck extension, it may be more helpful to use a GVLs rather than a GVLw to improve the laryngoscopic view in these patients.

Visualized intubation is likely to be more successful and possibly safer than when it is performed blindly. For this reason, improvement of C&L grades is important and definitely required in patients especially in those with a difficult airway. The purpose of this study was to evaluate whether a GVL could improve the laryngoscopic view compared with a DL and to investigate whether the GVL blade size could influence the laryngoscopic view in pediatric patients with C&L grades ≥3.

METHODS

This study protocol was approved by our IRB and registered at ClinicalTrials.gov (NCT01616771). Written informed consent was obtained from all patients or their guardians. Patients were excluded if they had a risk of pulmonary aspiration, increased intracranial pressure, or severe cardiovascular disease that could deteriorate their hemodynamics by repeated laryngoscopy or if the attending anesthesiologist considered use of laryngoscopy to be contraindicated. Patients who were scheduled for surgery under general anesthesia were eligible for recruitment if their C&L grade was ≥3 on previous anesthetic records. We also recruited patients with congenital syndromes associated with a difficult airway. Patients were finally included in this study if their modified C&L grades were ≥3a with BURP using DL after induction of anesthesia.

Patients were not premedicated. In the operating room, appropriate monitoring for each patient was applied and 100% oxygen was administered. Anesthesia was induced with 5 mg/kg IV thiopental sodium. After adequate mask ventilation was assured, rocuronium (0.6 mg/kg) was administered to facilitate airway instrumentation. Manual ventilation was performed with 4 to 8 vol% sevoflurane in 100% oxygen using a facemask before the laryngoscopy. For a possible cannot ventilate–cannot intubate situation, an emergency cricothyrotomy kit and fiberoptic bronchoscopy were prepared near the patients. After no response to train-of-four stimulations with a nerve stimulator (TOF-Watch; Organon Ireland Ltd., Dublin, Ireland), the initial laryngoscopic view was scored using a Macintosh blade (size 1 for infants and small children, size 2 for older children, and size 3 for adolescents). The scoring system was based on modified C&L grade:^10–12 grade 1, all or most of the glottic aperture was visible; grade 2a, posterior cords and cartilage visible; grade 2b, only posterior cartilage visible; grade 3a, epiglottis visible and can be lifted; grade 3b, epiglottis adherent to the posterior pharynx; and grade 4, the epiglottis could not be visualized (Fig. 1).

Figure 1: Modified Cormack and Lehane grade.

If the modified C&L grade was ≤2b under direct laryngoscopy with BURP, the patient was excluded from this study and endotracheal intubation was performed. If the modified C&L grade was ≥3a under initial direct laryngoscopy with BURP, the patient was included in this study. The second and third laryngoscopic view was then scored using the GVLw and GVLs. The order of application of GVLw and GVLs was randomized by a random-number table. All laryngoscopic views were graded both with and without the BURP. All laryngoscopic attempts and evaluations were performed by the same anesthesiologist (J-TK), who has conducted >300 GVL procedures. After evaluation of the laryngoscopic view, either endotracheal intubation or laryngeal mask airway insertion was performed according to clinical requirement regardless of modified C&L grade. For endotracheal intubation, we used the blade that provided the best laryngoscopic view.

Four sizes of reusable GVL are currently available (GVL 2 is for 1.8–10.0 kg, GVL 3 for 10–40 kg, and GVL 4 and 5 for >40 kg patients). In this study, GVL 2 to 4 were used (Fig. 2). GVLw was selected depending on the patient’s weight based on the manufacturer’s guideline. For patients who weighed >40 kg, size 4 GVL was chosen as GVLw and a size 3 was used as GVLs. The primary outcome was the difference in the C&L grade between DL and GVLw, and the secondary outcome was that between GVLw and GVLs. For the statistical analysis, the modified C&L grade was converted to an ordinal scale; grade 1 to 1, grade 2a to 2, grade 2b to 3, grade 3a to 4, grade 3b to 5, and grade 4 to 6.

Figure 2: Three sizes of GlideScope® video laryngoscope (GVL) used in this study.

An improvement of C&L grade ordinal scale by 2 was considered a clinically significant change. The mean difference of C&L grade ordinal scale between DL and GVLw (primary outcome) was 1.3, and its standard deviation was 2.0 in our previous study.⁵ The difference of C&L grade ordinal scale between GVLw and GVLs (secondary outcome) was 2.6, and its standard deviation was 2.2 in our pilot study. The required sample size for the Wilcoxon signed rank test was 21 for primary outcome and 13 for secondary outcome with an α error of 0.05 and 80% power.

To verify the difference in modified C&L grades using each blade, the ordinal scales of each blade were compared (DL–GVLw, GVLw–GVLs and DL–GVLs) using the Wilcoxon signed rank test. P values and confidence intervals (CIs) have been corrected for the 3 comparisons. Hodges–Lehmann method was used to calculate 98.3% CIs of paired differences of 3 comparisons. The effect of BURP for each blade was evaluated by 95% CI and the Wilcoxon signed rank test without any correction of P values. Symmetry for the distribution of the pairwise difference was confirmed by skewness of the data. Continuous variables are presented as means ± SD (range). All data were analyzed using the SPSS statistics version 19 software (SPSS, Inc., Chicago, IL). A P value of <0.05 was considered to indicate statistical significance.

RESULTS

Patient enrollment and the experimental process are summarized in Figure 3. Thirty-eight patients were recruited, and 15 of them were excluded because their modified C&L grade was ≤2b with BURP. Therefore, 23 patients were studied. Patients’ characteristics are shown in Table 1. The initial size of the GVLw was size 2 in 4 patients who weighed <10 kg. In those 4 patients, we could only compare the laryngoscopic view between DL and GVLw because smaller reusable GVLs (GVL size 1) were not available in our institution. Therefore, the primary outcome was evaluated in 23 patients and the secondary outcome in 19 patients.

Figure 3: Consolidated Standards of Reporting Trials (CONSORT) diagram. C&L = Cormack and Lehane; GVL = GlideScope® video laryngoscope; GVLs = 1-size smaller GVL than GVLw; GVLw = GVL selected by patient’s weight.*C&L grade under GVLs cannot be evaluated in 4 patients <10 kg due to lack of GVL size 1.

Table 1: Patients’ Characteristics

Seventeen of 23 patients were diagnosed with syndromes that might be associated with difficult airway management: 4 with Treacher–Collins syndrome, 3 with Goldenhar syndrome, 2 with CATCH 22 syndrome (cardiac defects, abnormal facial features, thymic hypoplasia, cleft palate, and hypocalcemia due to microdeletion of 22q11), 3 with a VATER anomaly (vertebral, anal, tracheoesophageal fistula, and renal defects), 1 with Denys–Drash syndrome, 1 with DiGeorge syndrome, 1 with Noonan syndrome, 1 with Poland syndrome, and 1 with Ullrich disease. Two patients had arthrogryposis, which affects jaw and neck movement; thus, neck extension was restricted. Another 3 patients had no genetic disease but had asymmetric faces.

Median values of modified C&L grade under DL, GVLw, and GVLs were 4, 3b, and 1 without BURP, respectively. (Fig. 4) When compared with DL, improvement of laryngoscopic view with GVLw was not obvious: 98.3% CI for differences of C&L grade ordinal scale was 0 to 1 with and without BURP (P = 0.11 and P = 0.15, respectively). However, GVLs improved the laryngoscopic view in comparison with both DL (98.3% CI for differences, 3.5–5.0 without BURP, P = 0.00007 and 3.5–4.5 with BURP, P = 0.0001) and GVLw (98.3% CI for differences, 3.0–4.5 without BURP, P = 0.00007 and 2.5–4.0 with BURP, P = 0.0001). Using GVLw, there were only 2 of 23 patients (9%) whose modified C&L grades were decreased to ≤2b without BURP and 3 patients (13%) with BURP. However, using the GVLs, the modified C&L grade of all patients was decreased to ≤2b, regardless of BURP. Overall individual changes of modified C&L grade and CIs for differences in C&L grade ordinal scale are illustrated in Figure 5.

Figure 4: Box and whisker plots illustrating the distribution of modified Cormack and Lehane (C&L) grade at each blade, without backward, upward, and right lateral displacement of the thyroid cartilage (BURP (grey) and with BURP (white). The heavy horizontal line within the box represents the median value. Improvement of modified C&L grade using GlideScope® video laryngoscope (GVL) was statistically significant, when compared with a direct laryngoscope (DL) or GVL selected by weight (GVLw), regardless of BURP. GVLs = 1-size smaller GVL than GVLw. *P < 0.001.

Figure 5: Individual changes of modified Cormack and Lehane (C&L) grade and 98.3% confidence intervals (CIs) of the difference in C&L grade ordinal scale. Blue circle, triangle, and square with error bars represent the median difference in C&L grade ordinal scale and 98.3% CI of the difference between blades. Without backward, upward, and right lateral displacement of the thyroid cartilage (BURP) (A), median differences between direct laryngoscope (DL) and GVL selected by weight (GVLw), GVLw and 1-size smaller GVL than GVLw (GVLs), and DL and GVLs were 0.5, 3.5, and 4.0, respectively. Those were same in with BURP (B). (B) = with BURP. *P < 0.01.

The best laryngoscopic views were obtained by GVLs in all but 4 patients (<10 kg) who could not be evaluated with GVLs. In 17 patients, the trachea was intubated using GVL and the other 6 patients were ventilated through laryngeal mask airways because there was no need for intubation. Although some manipulation of the stylet curvature was needed after the first trial, all intubations were successfully performed within 3 attempts in 17 patients.

BURP improved the laryngoscopic view in 7 of 23 patients (30%) with DL and GVLw and 6 of 19 patients (32%) with GVLs. The effectiveness of BURP was statistically significant in all groups (P values of DL, GVLw, and GVLs: 0.008, 0.011, and 0.014, respectively). Median changes in ordinal scale of C&L grade induced by BURP were 0.5, and 95% CI were 0.0 to 0.5 with all blades (Fig. 6). There was no adverse outcome during this study.

Figure 6: Overall changes of individual modified Cormack and Lehane (C&L) grade with backward, upward and right lateral displacement of the thyroid cartilage (BURP) and 95% confidence interval (CI) of the difference in C&L grade ordinal scale. BURP had additional effect for improving laryngoscopic view using a direct laryngoscope (DL), GlideScope® video laryngoscope selected by weight (GVLw), and 1-size smaller GVL than GVLw (GVLs). Each line represents the individual change of modified C&L grade. Median changes in C&L grade ordinal scale induced by BURP were 0.5 (95% CI, 0–0.5) in all 3 comparisons. (B) = with BURP. *P < 0.05.

DISCUSSION

We evaluated whether a GVL improved the laryngoscopic view in patients with C&L ≥3 and whether the size of the GVL blade influenced the degree of improvement. The main finding was that GVLs, rather than GVLw, facilitated obtaining a better laryngoscopic view in patients with C&L grade ≥3.

The GVL was developed to improve the laryngoscopic view without alignment of the oral, pharyngeal, and tracheal axes. According to a recent meta-analysis, the GVL is associated with improved glottis visualization, particularly in adult patients with potential or simulated difficult airways.¹³ However, there are limited data on pediatric patients. One pilot study suggested that the GVL improved laryngoscopic view when compared with a DL in 18 pediatric patients with difficult airways. However, one third of patients had their C&L grade <3 under direct laryngoscopy with BURP and their vocal cords were still invisible, even with GVL, in 9 patients in that study.¹ Furthermore, there was no evaluation on the effect of blade size. The clinically relevant and important C&L grade should be >3a, to conclude the superior effect of the GVL as a rescue method for difficult intubations with a DL. For this reason, we studied only patients whose modified C&L grade was ≥3a with BURP. Indeed, 61% of the patients in this study had a C&L grade 4 without BURP using a DL.

The blade size of the GVL appears to be important for improving the laryngoscopic view. The laryngoscopic view was not significantly improved with the GVLw in patients with C&L grades ≥3 in our study. Nine of 23 patients had an improved laryngoscopic view using the GVLw, but a C&L grade <3 was observed in only 3 patients with BURP. However, GVLs improved modified C&L grade of all patients <3a regardless of BURP. Furthermore, GVLs improved C&L grades by 4.0 and 3.5 points on an ordinal scale when compared with a DL and GVLw. On the contrary, the median difference of C&L grade ordinal scale between DL and GVLw was 0.5. In other words, changing to a smaller blade size clinically relevantly ameliorated difficult airway management in patients with C&L grades ≥3.

Hirabayashi and Otsuka¹⁴ provided a reason for the weak performance of the GVLw. They found a blind spot in the small and midsized GVL, just below the blade tip. It was thought that this occurred because the tip of the GVL was not located in the glottis but in the vallecula of the epiglottis. The height of the blind area is only 2 mm for the small-size GVL, but it was 13 mm for the midsize GVL; thus a high and anteriorly positioned glottis of a relatively small child may be concealed entirely by this area. A midsized GVL should be placed anteriorly to obtain a better laryngoscopic view, but that could be difficult, particularly in patients with small oropharyngeal space.

Figure 7 shows the possible reason why the GVLs is better than the GVLw. A small blade can be introduced into the mouth deeper, to the end of the proximal part of the blade and the distal handle. By sliding above the tongue, the blade tip can be rotated more anteriorly and cephalad in the vallecula. Because the epiglottis is lifted anteriorly more by the GVLw, the camera view can be extended to the anterior larynx structures. On the contrary, the GVLw is not easy to insert deeper with rotation to optimize the angle of camera. Although the view through the camera at the tip is more angled to the vocal cords than the DL, the epiglottis still cannot be lifted, which interrupts the glottis view. Therefore, in patients with a retrognathia, micrognathia, or limited neck extension, the GVLs can improve the laryngoscopic view. However, it may be difficult even with GVLs to improve the laryngoscopic view in patients with very limited mouth opening or small oropharyngeal space. In fact, the majority of our study patients had retrognathia or micrognathia. Additionally, there was no case of very limited mouth opening or small oropharyngeal space because of a high C&L grade in our study population.

Figure 7: The influence of the size of GlideScope® video laryngoscope (GVL on glottis view (A) when standard-sized GVL based on weight (GVLw) is used and (B) when smaller-sized GVL (GVLs) is used. In panel A, camera view is still blocked by the epiglottis which cannot be lifted enough by the GVLw (a). In panel B, the glottis opening is visible by rotating the smaller blade over the tongue and matching the angle of the camera to the vocal cord. The whole blade and the distal part of the handle can be inserted into the mouth (b). One-size smaller blade can provide better glottis view in patients with hypognathia, the larynx positioned more cephalad or limited neck extension.

Previous studies demonstrated that BURP improves the laryngoscopic view using the GVL.^1,5 This result is comparable with that of our study, although BURP improved median modified C&L grade by only 0.5. In theory, the BURP maneuver is not needed to expose the glottis using the GVL. However, as Hirabayashi and Otsuka¹⁴ commented, laryngoscopic viewing using the GVL causes a blind area and sometimes there is a need for the BURP to obtain more glottis exposure with the GVL and DL.

Some limitations of this study should be mentioned. First, we evaluated mainly modified C&L grade, not success rate or time to tracheal intubation. However, we assumed that improvement of laryngoscopic view could increase success rate or decrease time to intubation because intubation using a DL was almost impossible or traumatic especially in patients whose C&L grades were 4. Second, there was a wide range of patients’ weights (5.6–54.5 kg), which caused an uneven distribution of GVLw sizes. A size 3 GVL was selected as GVLw for 16 patients weighing 10 to 40 kg, whereas a GVL 4 for 3 patients weighing >40 kg and GVL 2 for 4 patients weighing <10 kg. Third, the laryngoscopic view using GVLs could not be evaluated in 4 patients who weighed <10 kg due to the lack of a reusable GVL smaller than size 2. Therefore, the superiority of GVLs to GVLw in terms of improvement of C&L grade in small children weighing <10 kg requires further study. Finally, the participating anesthesiologist was not blinded to the laryngoscopy equipment used. Because all evaluations of glottis view were performed by a single anesthesiologist, unintentional biases could have influenced the results.

In conclusion, the laryngoscopic view improved significantly with the GVLs compared with GVLw when the C&L grade is ≥3 using a DL. The GVLs is recommended for improving the laryngoscopic view especially in patients with difficult airways.

DISCLOSURES

Name: Ji-Hyun Lee, MD.

Contribution: This author collected and analyzed the data, and prepared the manuscript.

Attestation: Ji-Hyun Lee attests to the integrity of the original data and the analysis reported in this manuscript.

Name: Yong-Hee Park, MD.

Contribution: This author helped design the study and collect data.

Attestation: Yong-Hee Park is the archival author.

Name: Hyo-Jin Byon, MD.

Contribution: This author helped design the study and collect data.

Attestation: Hyo-Jin Byon approved the final manuscript.

Name: Woong-Ki Han, MD.

Contribution: This author helped collect the data and prepare the manuscript.

Attestation: Woong-Ki Han approved the final manuscript.

Name: Hee-Soo Kim, PhD.

Contribution: This author helped design the study and collect data.

Attestation: Hee-Soo Kim approved the final manuscript.

Name: Chong-Sung Kim, PhD.

Contribution: This author helped design the study and collect data.

Attestation: Chong-Sung Kim approved the final manuscript.

Name: Jin-Tae Kim, PhD.

Contribution: This author designed the study, collected the data, and helped prepare the manuscript.

Attestation: Jin-Tae Kim attests to the integrity of the original data and the analysis reported in this manuscript.

This manuscript was handled by: Peter J. Davis, MD.