The flexible laryngeal mask airway (FLMA) was

GENERAL ARTICLES The Influence of the Tonsillar Gag on Efficacy of Seal, Anatomic Position, Airway Patency, and Airway Protection with the Flexible Laryngeal Mask Airway: A Randomized, Cross-Over Study of Fresh Adult Cadavers J. R. Brimacombe, MB, ChB, FRCA MD*, C. Keller, MD, A. R. Gunkel, MD, and F. Pühringer, MD *Department of Anaesthesia and Intensive Care, University of Queensland and Cairns Base Hospital, Australia; and Departments of Anaesthesia and Intensive Care Medicine and Ear, Nose, and Throat Surgery, Leopold-Franzens University, Innsbruck, Austria We conducted a randomized, controlled, cross-over cadaver study to test the hypothesis that the efficacy of seal for ventilation and airway protection, anatomic position, and airway patency with the flexible laryngeal mask airway (FLMA) are altered by the application of a Boyle Davis (B-D) gag. We also determined the airway sealing pressure (ASP) at which the FLMA prevents aspiration when large volumes of fluid are placed above the cuff. We studied 20 adult cadavers (6 24 h postmortem). Efficacy of seal for ventilation and airway protection, anatomic position, and airway patency were determined with and without a B-D gag (two blade sizes: 8 and 10 cm) for the size 3, 4, and 5 FLMA in random order. Efficacy of seal for ventilation was determined by measuring the ASP at an intracuff pressure of 60 cm H 2 O. Efficacy of seal for airway protection was determined by flooding the mouth with 55 135 ml of water, reducing intracuff pressure until aspiration was detected fiberoptically and measuring ASP at this intracuff pressure. Anatomic position and airway patency were determined with a fiberoptic scope at an intracuff pressure of 60 cm H 2 O. In addition, in vivo compliance and ASP for the FLMA were measured in 10 cadavers and 10 paralyzed, anesthetized patients. Efficacy of seal for ventilation and airway protection, anatomic position, and airway patency did not change with the application of a gag for any mask size. The mean (range) ASP at which aspiration occurred when large volumes of fluid were placed above the cuff was 11 (7 15) cm H 2 O. The ASP for ventilation was always higher than the ASP for airway protection (P 0.0001). The FLMA had similar in vivo compliance and ASP in cadavers and anesthetized patients. We conclude that efficacy of seal for ventilation and airway protection, anatomic position and airway patency for the FLMA are unaffected by the application of a B-D gag in adults. ASP should be 15 cm H 2 O if there is a maximal risk of aspiration from above the cuff. Implications: The flexible laryngeal mask airway forms an effective seal for ventilation and protection of the airway that is unaffected by the application of a mouth gag that provides surgical access to the oropharynx. The efficacy of the seal should be 15 cm H 2 O if there is a maximal risk of aspiration from above the cuff. (Anesth Analg 1999;89:181 6) The flexible laryngeal mask airway (FLMA) was designed for use in surgery in which the standard laryngeal mask airway (LMA) tube would either interfere with the surgical field, be occluded, or be displaced by the surgeon (1). It is identical to the standard device in all respects except that the airway tube has been replaced by a narrower, floppy flexometallic tube Accepted for publication April 8, 1999. Address correspondence to Dr. J. Brimacombe, Department of Anaesthesia and Intensive Care, Cairns Base Hospital, The Esplanade, Cairns 4870, Australia. Address e-mail to 100236,2343 @compuserve.com. that gives it flexibility and compression resistance and makes dislodgment unlikely if the tube or head and neck are moved (2). The FLMA has been widely used for oropharyngeal (3 8) and nasal (9,10) surgery, in which it is offers advantages over tracheal intubation including avoidance of muscle relaxants, less hemodynamic stress response to placement and removal, and less coughing during emergence, while maintaining and protecting the airway (11). The effectiveness of laryngeal mask devices as protective barriers across the laryngeal inlet has been demonstrated using small volumes of fluid (12 14), but larger volumes can accumulate above the cuff, and lifethreatening aspiration has been reported (15,16). Perhaps 1999 by the International Anesthesia Research Society 0003-2999/99 Anesth Analg 1999;89:181 6 181

182 BRIMACOMBE ET AL. ANESTH ANALG FLEXIBLE LARYNGEAL MASK AND TONSILLAR GAG 1999;89:181 6 airway sealing pressure (ASP) should be 10 cm H 2 Oto ensure that aspiration does not occur (1), but this has not been verified experimentally. The Boyle-Davis (B-D) gag was designed for use with tracheal tubes to improve surgical access to the oropharynx. It comprises a gag for opening the mouth, a blade with a central groove that allows oropharyngeal structures to be elevated from the surgical field without occluding the tube, and a suspension system to maintain the position of the gag and blade (Figure 1). Although the FLMA was designed for use with a B-D gag, airway obstruction and loss of seal have been reported during oropharyngeal procedures in children and adults (3 7). We postulated that the change in pharyngeal geometry would reduce the efficacy of seal by distorting the interface between the cuff and periglottic tissues and that the blade would occlude the FLMA tube. In our randomized, controlled, cross-over cadaver study, we tested the hypothesis that efficacy of seal for ventilation and airway protection, anatomic position, and airway patency with the FLMA are altered by the application of a B-D gag. We also determine the ASP at which the FLMA prevents aspiration when large volumes of fluid are placed above the cuff. Methods Ten male and 10 female adult cadavers (6 24 h postmortem) were included in this randomized, controlled, cross-over study. Ethical committee approval was obtained, and all patients, or their relatives, gave their written, informed consent to research. Cadavers with oropharyngeal anatomical abnormalities were excluded from the trial. For each cadaver, three FLMA devices (size 3, 4, and 5) and two sizes of the B-D gag were used. All FLMAs were in routine clinical use, had been through at least 20 autoclave cycles, and had passed the preuse check tests. A single experienced LMA user ( 1000 uses) inserted/fixed the FLMA according to the manufacturer s instructions (17). The insertion technique included full deflation of the cuff, flattening it against the hard palate and pushing it along the posterior palatopharyngeal curve using the index finger to provide centrifugal force. The pilot balloon was attached via a three-way tap to a 20-mL syringe and a calibrated pressure transducer with an accuracy of 5%. The position of the head and neck was adjusted to the classical tonsillar position (head and neck extension) with the head stabilized by a shallow head ring and a rolled drape 5 cm thick under the shoulder blades (Figure 2). When the head and neck was in the tonsillar position, the intracuff pressure was adjusted to 60 cm H 2 O by withdrawing or adding air. The B-D type gag (Karl Storz, Innsbruck, Austria) comprised a Figure 1. The flexible laryngeal mask airway with the Boyle-Davies gag. McIvor self-retaining mouth gag, Wuerzburg blades (8 and 10 cm), and a hand-cranked suspension system. The B-D gag was inserted, assembled, positioned, and suspended by an oral surgeon who was instructed to obtain a view of the oropharynx adequate for dissection tonsillectomy (Figure 3). If an adequate view could not be obtained, it was documented. The blade was lubricated before insertion to assist passage over the silicone tube of the FLMA. Care was taken to avoid displacement of the FLMA during head and neck movement and gag changes, and to position the FLMA and pilot tubing in the midline over the lower lip to avoid looping of the FLMA tube that could potentially become trapped by the gag. Efficacy of seal for ventilation and airway protection, anatomic position, and airway patency were determined with and without a B-D gag (two blades sizes: 8 and 10 cm) for the size 3, 4, and 5 FLMA in random order. Efficacy of seal for ventilation was determined by measuring the ASP at an intracuff pressure of 60 cm H 2 O. ASP was measured by closing the expiratory valve of the circle system at a fixed gas flow of 3 L/min and noting the airway pressure at which the dial on a calibrated aneroid manometer (accurate to 0.5 cm H 2 O) reached equilibrium. This is the airway pressure at which the leak is in equilibrium with fresh gas flow. The interobserver reliability and accuracy of this measuring system has recently been validated (18). Efficacy of seal for airway protection was determined by increasing intracuff pressure to 120 cm H 2 O and filling the oropharynx with water to the level of the gums using a 50-mL syringe and recording the volume. A fiberoptic scope was positioned distal to the mask aperture bars to provide a view of the glottic inlet. The intracuff pressure was reduced in 5-cm H 2 O increments every 15 s until water was seen at the glottic inlet. The intracuff pressure was recorded, and water was carefully removed using the suction catheter of the fiberoptic scope. The

ANESTH ANALG BRIMACOMBE ET AL. 183 1999;89:181 6 FLEXIBLE LARYNGEAL MASK AND TONSILLAR GAG Figure 2. The flexible laryngeal mask airway with the Boyle-Davies gag in a patient undergoing oropharyngeal surgery. male patients (matched for height and weight) who were suitable for FLMA usage. Anesthesia was induced with propofol 2.5 mg/kg and was maintained with 100% O 2 and sevoflurane 1% 2%. Muscle relaxation was accomplished with atracurium 0.5 mg/kg. In vivo intracuff pressure, ASP, and fiberoptic position were documented at zero volume with the cuff fully evacuated and after each additional 10 ml up to 40 ml with the head and neck in the neutral position. The investigators and measurement techniques were identical for cadavers and anesthetized patients. ASP was determined by an investigator blinded to the FLMA and gag size by a head drape. An investigator who was not blinded to the FLMA and gag size conducted all the fiberoptic observations. Sample size was selected to detect a projected difference of 25% among the groups with respect to ASP for a type I error of 0.05 and a power of 0.9. The power analysis was based on data from a cross-over pilot study of 10 cadavers in which airway protection was determined for the size 3, 4, and 5 FLMA with and without the two gags. The distribution of data was determined using Kolmogorov-Smirnov analysis (21). Statistical analysis was performed by using a paired t-test (normally distributed data) and 2 test (non-normally distributed data). Unless otherwise stated, data are presented as mean (range). Significance was taken as P 0.05. Figure 3. The flexible laryngeal mask airway with the Boyle-Davies gag fully open to provide a view of the oropharynx. accuracy of the fiberoptic detection of water was confirmed by noting a decrease in the water level in the mouth. The ASP was then determined at this intracuff pressure. Anatomic position was determined with a fiberoptic scope (3.2 mm external diameter) at an intracuff pressure of 60 cm H 2 O using the following scoring system: 4 only vocal cords visible, 3 vocal cords plus posterior epiglottis visible, 2 vocal cords plus anterior epiglottis visible, and 1 vocal cords not seen (19,20). The degree of rotation in the sagittal plane was judged fiberoptically by noting the angle between the mask aperture bars and vocal cords: nil ( 10 ), mild (10 45 ), moderate (46 90 ), and severe ( 90 ). Airway patency was determined by passing a fiberoptic scope along the FLMA tube and noting any occlusion of the tube or cuff by the gag. If any occlusion was present, the position of the gag was readjusted and the measurements were repeated. The FLMAs were inspected for damage after each use. To provide information about pharyngeal rigidity, in vivo compliance for the size 5 FLMA was measured in 10 male cadavers and 10 paralyzed, anesthetized Results The mean (range) age, height, and weight for B-D gag cadavers were 73 (57 92) yr, 169 (151 188) cm, and 72 (52 88) kg, respectively. The mean (range) volume of water required to fill the mouth was 85 (55 135) ml. An adequate view of the tonsillar bed was obtained with all gags and mask sizes. No tube or cuff occlusion was seen. ASP for ventilation and airway protection, anatomic position, and airway patency did not alter with the application of a gag for any mask size (Table 1). The mean (range) ASP at which aspiration occurred when large volumes of fluid were placed above the cuff was 11 (7 15) cm H 2 O. The ASP for ventilation was always higher than the ASP for airway protection (P 0.0001). ASP for ventilation and fiberoptic scores was lower for the size 3 FLMA compared with the size 4 or size 5 FLMA (P 0.0001). Intracuff pressure was higher for the size 3 FLMA compared with the size 4 and size 5 FLMAs at the ASP for airway protection (P 0.0001). Rotation of the FLMA was not seen. There was no damage to the FLMAs. In the comparison between cadavers and paralyzed, anesthetized patients, cadavers were significantly older (79 [55 89] vs 36 [24 59] yr; P 0.0001) but had identical mean height (176 cm) and weight (78 kg). The FLMA had similar in vivo compliance, ASP, and

184 BRIMACOMBE ET AL. ANESTH ANALG FLEXIBLE LARYNGEAL MASK AND TONSILLAR GAG 1999;89:181 6 Table 1. Airway Sealing Pressure (ASP), Fiberoptic Score (FOS), and Intracuff Pressure (ICP) ASP for ventilation FOS 4/3/2/1 a ASP for airway protection ICP Size 3 Control 15 (13 17) 1/5/10/4 11 (10 11) 17 (13 to 21) Small blade 15 (13 18) 0/5/9/6 11 (10 11) 17 (13 to 20) Large blade 15 (13 17) 0/4/10/6 10 (9 11) 17 (14 to 20) Size 4 Control 22 (20 23) 3/9/8/0 10 (9 11) 4 ( 7 to1) Small blade 22 (20 24) 2/8/10/0 11 (11 12) 6 ( 9 to2) Large blade 22 (20 24) 2/7/11/0 11 (11 12) 6 ( 9 to3) Size 5 Control 23 (21 25) 4/8/8/0 10 (9 11) 5 ( 9 to2) Small blade 22 (20 24) 2/7/11/0 11 (11 12) 7 ( 9 to4) Large blade 23 (21 25) 2/7/11/0 11 (10 12) 7 ( 9 to2) Data are mean (95% CI). a 4 only vocal cords visible; 3 vocal cords plus posterior epiglottis visible; 2 vocal cords plus anterior epiglottis visible; 1 vocal cords not seen (19). fiberoptic position in cadavers and anesthetized patients (Table 2). Discussion John et al. (12) and Cork et al. (13) showed that the LMA protected the airway from 10-ml methylene blue dye and 15-ml barium sulphate respectively. Webster et al. (14) showed that the FLMA protected the airway from 5 ml of barium sulfate. Williams and Bailey (3) found that aspiration of blood was more common with the tracheal tube than the FLMA after adenotonsillectomy, but this was not confirmed by Webster et al. (4). Our data show that, when adequately inflated, the FLMA protects the airway when the mouth is filled with 55 135 ml of water. This situation may occur during oral or nasal surgery when there is unrecognized or uncontrolled bleeding, suction failure, or during irrigation. The mean ASP at which aspiration occurred was 11 cm H 2 O, which approximates to the height of the water above the lowest part of the FLMA cuff. An ASP 12 cm H 2 O would protect 95% of patients at maximal risk. We recommend an ASP 15 cm H 2 O if there is a risk of aspiration from above the cuff. During the application of a B-D gag, there is an increase in the antero-posterior diameter of the pharynx and a change in the tension of the pharyngeal muscles. The blade of the gag fixes the flexometallic tube of the FLMA, but the cuff is free to rotate out of place if the change in pharyngeal shape is sufficient. Our data show that airway protection, efficacy of the seal, and anatomic position are unaffected by the application of the B-D gag. We postulate that the cuff is able to adapt to the changes in pharyngeal shape without loss of seal. The efficacy of the seal for the LMA depends on the degree of conformity with pharyngeal tissues, rather than pressure against the mucosa (22). In our study, the intracuff pressure was 60 cm H 2 O, and the cuff was semi-inflated. This value was chosen because ASP is higher at low, rather than high, intracuff pressures (23). FLMA tube compression or loss of seal occurs in 2% 20% of patients during the application of the B-D gag (3 7). Although most patients in these studies were young children, these problems were also noted in older children and adults with the larger LMAs (4,6,7). There have been three reports of the pilot tube being severed (24 26). Bailey et al. (1) suggest that these problems can be avoided if the airway tube and pilot tube are maintained together in the midline during the application of the gag. Our data support this suggestion because we detected no compression of the tube or loss of seal in 120 gag applications. Perhaps the length of blade chosen for the gag is important (1). A blade that is too long may result in airway obstruction when the gag is opened as it lies posterior to the mask and pushes the mask against the laryngeal inlet. Too short a blade will not support the tube of the FLMA that may be indented by the tip of the blade and partially obstructed. We could not demonstrate any difference in performance between the 8- and 10-cm blades for all adult sizes of FLMA. We conducted this study in cadavers because it would have been unethical to put patients at maximal risk of aspiration or to perform multiple applications of the B-D gag. Compared with the anesthetized human, cadavers have a lower temperature and more rigid pharyngeal musculature, and they do not undergo spontaneous or positive pressure ventilation. However, we consider our model a reasonable representation of the in vivo situation because we showed that the in vivo compliance and ASP for the FLMA is similar in fresh male cadavers and paralyzed, anesthetized male patients. This may be related to the freshness of the cadavers and age-related differences in the rigidity of the pharyngeal tissues. We speculate that

ANESTH ANALG BRIMACOMBE ET AL. 185 1999;89:181 6 FLEXIBLE LARYNGEAL MASK AND TONSILLAR GAG Table 2. In Vivo Intracuff Pressure, Airway Sealing Pressure, and Fiberoptic Score In vivo intracuff pressure Airway sealing pressure Fiberoptic score 4/3/2/1 a Cuff volume (ml) Anesthetized patients Cadavers Anesthetized patients Cadavers Anesthetized patients Cadavers 0 29 ( 33 to 25) 30 ( 35 to 25) 13 (10 16) 12 (10 13) 0/7/3/0 3/4/3/1 10 36 (18 54) 35 (26 44) 18 (14 22) 20 (16 23) 1/6/3/0 3/4/3/1 20 72 (51 94) 87 (71 147) 23 (19 26) 23 (21 24) 2/6/2/0 3/4/3/1 30 119 (91 147) 133 (119 147) 23 (21 26) 23 (22 24) 3/4/3/0 3/4/3/1 40 200 (181 220) 236 (212 260) 24 (18 29) 23 (22 24) 3/4/3/0 3/4/3/1 Data are mean (95% CI). Results were achieved by using a size 5 flexible laryngeal mask airway. a 4 only vocal cords visible; 3 vocal cords plus posterior epiglottis visible; 2 vocal cords plus anterior epiglottis visible; 1 vocal cords not seen (19). the effects of rigor mortis on increasing pharyngeal rigidity are offset by lower pharyngeal rigidity in the elderly population. Nonetheless, some caution should be applied when extrapolating from the present study to the in vivo situation. A further limitation of our study is that the head and neck were only in one position during testing. However, the B-D gag is only used in the head and neck extended position. Furthermore, a study has shown that the seal for the FLMA is least effective in this position; therefore, the risk of aspiration from above the cuff is highest in the position we tested (2). Our finding that ASP and fiberoptic score is lower for the size 3 mask compared with larger sizes also matches data from anesthetized patients (27). Finally, this study was conducted in adults, and our findings may not apply to the pediatric population, in which the caliber of the FLMA tube is smaller and the pharyngeal tissues are probably more complaint. We conclude that efficacy of seal for ventilation and airway protection, anatomic position, and airway patency for the FLMA are unaffected by the application of a B-D gag in adults. ASP should be 15 cm H 2 Oif there is a maximal risk of aspiration from above the cuff. References 1. Bailey P, Brimacombe JR, Keller C. The flexible LMA: literature considerations and practical guide. Int Anesthesiol Clin 1998;36: 111 22. 2. Keller C, Brimacombe J. The influence of head and neck position on oropharyngeal leak pressure and cuff position with the flexible and the standard laryngeal mask airway. Anesth Analg. In press. 3. Williams PJ, Bailey PM. Comparison of the reinforced laryngeal mask airway and tracheal intubation for adenotonsillectomy. Br J Anaesth 1993;70:30 3. 4. Webster AC, Morley-Forster PK, Dain S, et al. Anaesthesia for adenotonsillectomy: a comparison between tracheal intubation and the armoured laryngeal mask airway. Can J Anaesth 1993; 40:1171 7. 5. Wehrle HJ, Gottstein P. Experiences with use of the laryngeal mask with flexible, wire reinforced tube for ENT interventions in childhood. Anaesthesiol Intensivmed Notfalmed Schmerzther 1997;32: 151 4. 6. Heath ML, Sinnathamby SW. The reinforced laryngeal mask airway for adenotonsillectomy. Br J Anaesth 1994;72:728 9. 7. Sher M, Brimacombe J, Laing D. Anaesthesia for laser pharyngoplasty: a comparison of the tracheal tube verses reinforced laryngeal mask airway. Anaesth Intensive Care 1995;23: 149 54. 8. Quinn AC, Samaan A, McAteer EM, et al. The reinforced laryngeal mask airway for dento-alveolar surgery. Br J Anaesth 1996; 77:185 8. 9. Williams PJ, Thompsett C, Bailey PM. Comparison of the reinforced laryngeal mask airway and tracheal intubation for nasal surgery. Anaesthesia 1995;50:987 9. 10. Webster AC, Morley-Forster PK, Janzen V, et al. Anesthesia for intranasal surgery: a comparison between tracheal intubation and the flexible reinforced laryngeal mask airway. Anesth Analg 1999;88:421 5. 11. Brimacombe J. The advantages of the LMA over the tracheal tube or facemask: a meta-analysis. Can J Anaesth 1995;42: 1017 23. 12. John RE, Hill S, Hughes TJ. Airway protection by the laryngeal mask: a barrier to dye placed in the pharynx. Anaesthesia 1991;46:366 7. 13. Cork RC, Depa RM, Standen JR. Prospective comparison of use of the laryngeal mask and endotracheal tube for ambulatory surgery. Anesth Analg 1994;79:719 27. 14. Webster AC, Morley-Forster PK, Watson J, et al. Evaluation of the flexible reinforced laryngeal mask airway (FRLMA) for intranasal surgery (abstract). Anesthesiology 1997;87:A30. 15. Norton A, Germonpre J, Semple T. Pulmonary aspiration of blood following traumatic laryngeal mask airway insertion. Anaesth Intensive Care 1998;26:213 5. 16. Terada H. Airway obstruction due to pus from sinusitis during the use of the LM. Rinsho Masui 1993;17:673 4. 17. Brimacombe J, Brain AIJ, Berry A. The laryngeal mask airway instruction manual. Henley-on-Thames: Intavent Research Limited, 1996. 18. Keller C, Brimacombe J, Keller K, Morris R. A comparison of four methods for assessing airway sealing pressure with the laryngeal mask airway in adult patients. Br J Anaesth 1999;82: 286 7. 19. Brimacombe J, Berry A. A proposed fiber-optic scoring system to standardize the assessment of laryngeal mask airway position. Anesth Analg 1993;76:457. 20. Joshi S, Sciacca RR, Solanki DR, et al. A prospective evaluation of clinical tests for placement of laryngeal mask airways. Anesthesiology 1998;89:1141 6. 21. Sachs L. Der Kolmogoroff-Smirnov-Test fuer die Guete der Anpassung. In: Angewandte Statistik. Berlin: Springer Verlag, 1992;426 30.

186 BRIMACOMBE ET AL. ANESTH ANALG FLEXIBLE LARYNGEAL MASK AND TONSILLAR GAG 1999;89:181 6 22. Brimacombe J, Keller C. A comparison of pharyngeal mucosal pressure and airway sealing pressure with the laryngeal mask airway in anesthetized adult patients. Anesth Analg 1998;87: 1379 82. 23. Brimacombe J, Keller C, Morris R, Mecklem D. A comparison of the disposable versus the reusable laryngeal mask airway in paralyzed adult patients. Anesth Analg 1998;87:921 4. 24. Wat LI, Brimacombe J, Gee S. Laryngeal mask airway longevity and pilot-balloon failure. J Clin Anesth 1997;9:432. 25. Short JA, Melillo EP. Damage to a laryngeal mask during tonsillectomy. Anaesthesia 1997;52:501. 26. Venn PJH. Use of the laryngeal mask during tonsillectomy. Br J Anaesth 1998;81:298 9. 27. Berry AM, Brimacombe J, McManus KF, Goldblatt M. An evaluation of the factors influencing selection of the optimal size of laryngeal mask airway in normal adults. Anaesthesia 1998;53: 565 70.