The risk of upper airway collapse increases in unconscious or anaesthetised patients, because a low central drive decreases the activity of the pharyngeal dilator muscles.1,2 An oropharyngeal airway is helpful in relieving such upper airway obstruction as it moves the tongue and hypopharyngeal structures forward, improving airway patency.3 Although the airway is simple to use, it is important to select an appropriate size. If the airway is too small, its distal end will be obstructed by the tongue, resulting in inadequate ventilation4 and, if too big, there is a risk of traumatic injury to the surrounding laryngeal structures.
To estimate an appropriate airway size, the distance between the maxillary incisors and the angle of the mandible has been recommended by the European Resuscitation Council guidelines.5 In contrast, the American Heart Association guidelines recommend the use of the distance between the corner of the mouth and the angle of the mandible.6 Despite these guidelines, no study has evaluated the utility of these external facial measurements for the selection of an appropriate airway size. The present study, using airway sizes based upon the two guidelines, aimed to determine which of the two guidelines resulted in an airway providing optimal ventilation parameters. Both mask and mechanical ventilations as well as the endoscopic view of the airway tip via a fibreoptic bronchoscope were used in the assessment.
The prospective crossover study was approved by the Institutional Review Board of Severance Hospital (ref: 4–2013–0440), Seoul, Korea (Chairperson Professor Seung Min Kim) on 21 August 2013. Written informed consent was obtained from all patients. Adults aged 20 to 70 years with an American Society of Anesthesiologists physical status of 1 or 2 who were scheduled for elective ear, nose or neck surgery, requiring tracheal intubation for general anaesthesia, were included. Patients with known anatomical abnormalities of their airways, a history of difficult intubation, cervical spine injury or lesion, dental problems or a mouth opening of less than 2 cm were excluded. Airway assessments included the modified Mallampati classification,7 thyromental distance and inter-incisor gap, which were evaluated before the induction of anaesthesia.
On arrival in the operating room, the patient was laid supine with the head in a neutral position resting on a ring support. Two facial distances were measured by a physician (S.H.K.) using a commercially available set of dividers: the distance between the corner of the mouth and the angle of the mandible and the distance from the maxillary incisors to the angle of the mandible (Fig. 1). To measure the distances, the tip of one leg of the divider was set at either the corner of the mouth or the tip of the upper central maxillary incisors, whereas the tip of the other leg was set at the mandibular angle. The length between the two tips was then measured using a ruler.
Following these measurements, standard monitoring, including electrocardiography, pulse oximetry and non-invasive blood pressure, was commenced. Anaesthesia was induced with propofol 1.5 mg kg−1, remifentanil 1.0 μg kg−1 and atracurium 0.5 mg kg−1. Once the patient had lost consciousness, a facemask was placed over the mouth and nose and with one hand this was held in place while, at the same time, the head was held in an extended position by gentle traction on the symphysis menti (in the anterocephalic direction). The other hand was used to ventilate the lungs with 100% oxygen set at a flow rate of 6 l min−1 with 4 to 5% sevoflurane. After the muscles were completely relaxed, as assessed at the adductor pollicis muscle by supramaximal train-of-four stimuli applied to the ulnar nerve using a peripheral nerve stimulator (Innervator 252; Fisher & Paykel Healthcare, Auckland, New Zealand), control measurements without an airway were obtained. Manual ventilation without an airway was classified into three grades: clear, partial obstruction or complete obstruction. Partial obstruction was defined by the presence of adventitious sounds (snoring) and insufficient chest wall movements. Complete obstruction was defined by the absence of chest expansion despite the application of positive pressure ventilation. Because we could not deliver a consistent inspired tidal volume and pressure during manual ventilation, after evaluating the adequacy of manual ventilation we commenced pressure-controlled mechanical ventilation (Aisys; GE Datex-Ohmeda, Műnich, Germany) to provide a consistent driving pressure. Both manual and mechanical ventilations were administered by one physician (W.K.P.). To evaluate the adequacy of mechanical ventilation, the anaesthetist held the facemask with both hands with the head in extension and the mandible in a forward position. For mechanical ventilation, the respiratory parameters were: oxygen flow, 2 l min−1; peak inspiratory pressure, 15 cmH2O; respiratory rate, 15 breaths min−1 and inspiratory expiratory ratio, 1 : 2. Mechanical ventilation was classified as either adequate or inadequate. Inadequate ventilation was defined when the mean expired tidal volume from the sixth to 10th breath was less than 5 ml kg−1. One physician (N.H.M.), blinded to the airway size, assessed the adequacy of ventilation and it was assumed that inspiratory and expiratory volume were equal.
The airway size was determined using the measured distance, rounded down to the nearest whole number. For example, if the measured corner of the mouth to the angle of the mandible distance was 9.6 cm, a size 9 airway was chosen. The airway sizes were chosen according to the International Organization for Standardization standard: the size marked on the airway indicates the shortest (straight line) distance from the lower outside edge of the flange to the outside of the distal end. Five different airway sizes, namely 7 (7 cm, white), 8 (8 cm, green), 9 (9 cm, yellow), 10 (10 cm, red) and 11 (11 cm, orange), were used. The airway was initially inserted into the mouth with the tip positioned upward, then rotated 180° and advanced until the inner side of the flange rested on the upper incisors. Following control assessments without an oropharyngeal airway, the two chosen oropharyngeal airways (Guedel-type colour-coded oropharyngeal airway, Hudson RCI, Teleflex Medical, Research Triangle Park, North Carolina, USA) were then evaluated in a random order. The adequacy of manual and then mechanical ventilation was assessed for each airway as described above.
Following evaluation of mechanical ventilation with an airway, another physician (H.J.K.), blinded to the airway size, performed an endoscopic examination through the lumen of the airway using a fibreoptic bronchoscope (Olympus LF-GP; Olympus Optical Co., Tokyo, Japan) with the patient's head extended. The endoscopic view at the distal end of the airway was classified as follows: clear, partial obstruction by the tongue (more than 50% or less than 50% of the view), complete obstruction by the tongue and passing beyond the epiglottis tip (impaction in the vallecula, contact between the tip of the airway and the tip of the epiglottis, entry of the epiglottis into the lumen of the airway or close proximity to the vocal cord; Fig. 2).
As the oropharyngeal airway is curved, the length of the lumen was measured, using the markings on the fibrescope while it was within the lumen of the airway. The lumen lengths for the different airways were: size 7, 8.3 cm; size 8, 9.4 cm; size 9, 10.4 cm; size 10, 11.7 cm and size 11, 12.8 cm. When the length of the airway was shorter than the distance from the maxillary incisors to the tip of the epiglottis, after the distal tip of the fibrescope was placed adjacent to the epiglottis, a piece of tape was applied as a reference point on the fibrescope at the position of the flange. The distance from the distal end of the airway to the tip of the epiglottis was then calculated by subtracting the measured length of the lumen of the airway from the length of the reference point on the fibrescope. When the airway was longer than the distance from the incisors to the tip of the epiglottis, the airway was partly withdrawn from the mouth until the distal end was positioned adjacent to the tip of the epiglottis. The point of the airway that contacted the upper incisors was marked using a marker, and the length from the flange to the marking site was measured with a ruler after airway removal. This latter distance was then subtracted from the length of the lumen. After the endoscopic examination the airway was removed from the mouth and the presence or absence of visible blood stains on the airway was noted. Orotracheal intubation was then performed using an appropriately sized endotracheal tube, and anaesthesia was maintained with sevoflurane and remifentanil. During laryngoscopy with a Macintosh laryngoscope, the glottic view was graded according to the Cormack and Lehane grading system.8
The primary outcome was the proportion of patients who met three combined criteria: the lungs were adequately ventilated in the mechanical pressure-controlled mode, the distal end of the airway was not completely obstructed by the tongue and the distal end of the airway did not pass beyond the tip of the epiglottis. On the basis of the results of a previous pilot study, during which it was observed that the primary outcomes for the corner of the mouth to the angle of the mandible and the maxillary incisors to the angle of the mandible airways were 50 and 70%, respectively, we estimated that the difference in the primary outcome between the corner of the mouth to the angle of the mandible and the maxillary incisors to the angle of the mandible airways was 20%. Assuming a 10% dropout rate, a minimum of 113 patients were required to detect a difference of 20%, with a type I error of 0.05 and power of 0.8. The data on expired tidal volumes were compared using a paired t-test. The data pairs for the adequacy of manual and mechanical ventilation were compared using McNemar's test. Values are expressed as mean (SD) or number (percentage). SAS software (version 9.2, SAS Institute Inc., Cary, North Carolina, USA) was used for all statistical analyses. A value of P < 0.05 was considered statistically significant.
The patient characteristics and airway assessment findings are shown in Table 1. The average (±SD) corner of the mouth to the angle of the mandible and maxillary incisors to the angle of the mandible distances were 8.1 ± 0.6 and 9.9 ± 0.7 cm, respectively. The sizes of the airways selected based on these measurements are shown in Table 1.
All patients in the maxillary incisors to the angle of the mandible group had clear manual ventilation, whereas partially obstructed ventilation occurred in a small number of patients in the control and corner of the mouth to the angle of the mandible airway groups (8 and 6%, respectively; Table 2). There were no patients with complete obstruction during manual ventilation in either of the groups (Table 2).
With regard to mechanical ventilation, compared with the control group both airway groups demonstrated more adequate ventilation [corner of the mouth to the angle of the mandible airway (P = 0.009) and maxillary incisors to the angle of the mandible airway (P = 0.003), Table 2]. Although mechanical ventilation was adequate in all patients with the maxillary incisors to the angle of the mandible airway, inadequate ventilation was observed in 18 and 7% of patients in the control and corner of the mouth to the angle of the mandible airway groups, respectively (corner of the mouth to the angle of the mandible airway vs. maxillary incisors to the angle of the mandible airway, P < 0.001). An expired tidal volume of less than 2 ml kg−1, which indicated almost complete obstruction, was observed in 30 and 38% of the inadequately ventilated patients in the control and corner of the mouth to the angle of the mandible airway groups, respectively. Compared with the control group, the expired tidal volumes during mechanical ventilation were significantly larger in the corner of the mouth to the angle of the mandible group (P = 0.003) and maxillary incisors to the angle of the mandible group (P = 0.003). The expired tidal volume was greater in the maxillary incisors to the angle of the mandible airway group than in the corner of the mouth to the angle of the mandible airway group (P = 0.003).
The fibreoptic views at the distal end of the airways are shown in Table 3 and Fig. 2. Complete obstruction by the tongue was observed in a large number of patients in the corner of the mouth to the angle of the mandible airway group (40%) but not in any of the patients in the maxillary incisors to the angle of the mandible airway group. Some of the maxillary incisors to the angle of the mandible airways (22%) passed beyond the tip of the epiglottis, and in 10% of patients, the epiglottis had entered into the lumen of the airway; in two patients this was more than 1 cm. In addition, close proximity to the vocal cords was observed in one patient in the maxillary incisors to the angle of the mandible group (Table 3). In contrast, no airways in the corner of the mouth to the angle of the mandible group passed beyond the tip of the epiglottis. The distances from the distal ends of the corner of the mouth to the angle of the mandible and maxillary incisors to the angle of the mandible airways to the tip of the epiglottis were 2.4 ± 0.7 and 0.9 ± 0.9 cm, respectively. After removal, no blood staining was observed on any of the airways from either group. We did not observe any complications in the post-anaesthesia care unit.
The incidence of adequate mechanical ventilation (the distal end of the airway did not pass beyond the tip of the epiglottis and, on fibreoptic view the lumen was not completely obstructed by the tongue) was significantly higher in the maxillary incisors to the angle of the mandible airway group than in the corner of the mouth to the angle of the mandible airway group [88 (78%) vs. 68 (60%), P = 0.005].
We found that airway size, selected on the basis of the maxillary incisors to the angle of the mandible rather than the corner of the mouth to the angle of the mandible distance, was more appropriate for achieving adequate ventilation and an acceptable fibreoptic view.
The main objective of an oropharyngeal airway is to improve ventilation.9 In the present study, mechanical ventilation appeared to be inadequate in eight patients (7%) in the corner of the mouth to the angle of the mandible airway group. These eight patients also had complete obstruction of the distal end of their airway by the tongue. In these patients, when the maxillary incisors to the angle of the mandible airway was used, the expired tidal volume improved significantly from 175 ± 112 to 667 ± 288 ml and the fibreoptic view was clear in seven, and partially obstructed in one patient. These findings indicate that if the distal end of the airway is completely obstructed by the tongue, there is an increased risk of inadequate ventilation. Therefore, considering the high prevalence (40%) of complete obstruction by the tongue at the distal end of the corner of the mouth to the angle of the mandible airways, it is unlikely that the corner of the mouth to the angle of the mandible airway is appropriate for use.
However, when the corner of the mouth to the angle of the mandible airway was used, interestingly, none of the patients showed complete obstruction during manual ventilation, despite inadequate mechanical ventilation in the eight patients. This was probably as a result of much higher inflation pressures during manual ventilation. To investigate the effects of forced inspiration on ventilation, we used a small airway in a few patients. After observing complete obstruction of the distal end of the airway by the tongue [viewed with a fibreoptic bronchoscope connected to a video monitoring system inserted via the small hole of an endoscopy mask (VBM endoscopy mask; Medizintechnik GmbH, Sulz, Germany)], we ventilated the lungs manually with an inspiratory pressure of more than 15 cmH2O and observed that the tongue was pushed backward. The small space created between the tongue and the distal end of the airway during inspiration (Fig. 3) allowed lateral channelling of oxygenated airflow around the obstruction.4 Because of this tongue displacement mechanism, manual ventilation would be possible in most patients despite the high incidence of complete obstruction of the airway by the tongue when an corner of the mouth to the angle of the mandible airway is used. However, although such manual ventilation is possible with the mouth to the angle of the mandible airway, it is not desirable: the greater airway pressure (>15 to 20 cmH2O) could result in gastroesophageal insufflation and gastric distension.10,11
On the other hand, when a maxillary incisors to the angle of the mandible airway was used, mechanical ventilation appeared to be adequate in all patients, even if the distal end of the airway passed beyond the tip of the epiglottis. This suggests that the efficacy of ventilation is barely affected when the epiglottis is within the lumen of the airway or the airway is impacted in the vallecula. Marsh et al.4 demonstrated clear manual ventilation of the lungs, even though on radiographs the distal end of the airway was positioned in the vallecula or obstructed by the epiglottis. Other studies have demonstrated adequate ventilation despite the aperture of a laryngeal mask airway being covered by the epiglottis.12,13 In the maxillary incisors to the angle of the mandible airway group, in 12 of the 113 patients (approximately 10%), the epiglottis was within the airway lumen, and in two cases more than 1 cm of epiglottis was within the airway lumen. Although this number of patients was small, the possibility of injury to the epiglottis and its base should not be ignored in such cases.
When the ‘best-fit airway’ for an individual is defined as positioning of the distal end of the airway as close as possible to the tip of the epiglottis without any obstruction by the tongue, and which also allows effective mask ventilation, the maxillary incisors to the angle of the mandible airway is not entirely appropriate: even though it facilitates effective mask ventilation, there is a risk of injury to the surrounding laryngeal structures. Further studies, using other external facial measurements, are required to determine which facial measurement best predicts a ‘best-fit’ airway.
To compare our results with those for other commonly used Guedel-type airways, such as the Portex (Smiths Medical, Ashford, UK) and Berman (Vital signs; GE Healthcare, Totowa, New Jersey, USA) airways, we measured the dimensions of each airway. Even though the manufactures were different, these airways exhibited the same horizontally measured length from the flange to the distal end and a similar curvilinear length of the internal lumen, with only a small discrepancy (0 to 3 mm) when compared with the airway used in our study.
The limitations of our study were as follows: first, we performed manual ventilation in the head extended position, but mechanical ventilation was performed with the head in extension and with forward displacement of the mandible. The position of head extension with or without forward displacement of the mandible (jaw thrust) could influence the patency of the oropharyngeal airway.3,14 If we had used the head-extended position only during mechanical ventilation, a fairer comparison of manual and mechanical ventilations would have been possible. Second, in this study, 31 (67%) men were taller than 170 cm with eight (17%) taller than 180 cm, whereas 20 (30%) women were taller than 160 cm. Because of the relatively narrow height range in our studied population, the results may not be applicable to patients outside this range.
In conclusion, the results of this study suggest that an oropharyngeal airway based on the maxillary incisors to the angle of the mandible distance is more advantageous than an airway based on the corner of the mouth to the angle of the mandible distance.
Acknowledgements relating to this article
Assistance with the study: the authors are grateful to Hye Sun Lee and Ha Yan Kim (Biostatistics Collaboration Unit, Yonsei University College of Medicine, Seoul, South Korea) for statistical consultation and analysis of data. The authors would also like to thank Dong-Su Jang, MFA (Medical Illustrator, Medical Research Support Section, Yonsei University College of Medicine, Seoul, South Korea), for his help with the illustrations.
Financial support and sponsorship: none.
Conflicts of interest: none.