Secondary Logo

Journal Logo

Airway management

Comparison of laryngeal mask airway insertion methods, including the external larynx lift with pre-inflated cuff, on postoperative pharyngolaryngeal complications

A randomised clinical trial

Hu, Ling-Qun; Leavitt, Olga S.; Malwitz, Christine; Kim, Haisook; Doty, Robert A. Jr; McCarthy, Robert J.

Author Information
European Journal of Anaesthesiology: July 2017 - Volume 34 - Issue 7 - p 448-455
doi: 10.1097/EJA.0000000000000650
  • Free



Since its introduction into clinical anaesthesiology practice, the laryngeal mask airway (LMA) has been used routinely as an alternative to tracheal intubation during general anaesthesia, or as a rescue device for difficult airways.1 The LMA has been shown to provide a reliable airway with a lower incidence of airway complications compared with tracheal intubation.2 The most common LMA-associated postoperative complication is sore throat, with a reported incidence of up to 42%.3,4 Other complications include pharyngeal mucosal abrasions, cranial nerve damage secondary to pressure neuropraxia,5 dysphonia and dysphagia.3

Despite studies demonstrating that insertion with the LMA cuff inflated reduces the incidence of airway trauma,6 the ‘standard’ description of LMA insertion entails deflating the cuff and lubricating it prior to placement. With the mouth open as wide as possible, the LMA is pressed along the hard palate until the back of the oropharynx is reached. At this point, the LMA is then directed posterior to the tongue with either the index finger or the thumb until the device can no longer be advanced forward. After placement, the introducing finger is removed and the cuff is inflated. Even when using the standard method, LMA insertion can be difficult and traumatic to the tissues, resulting in pharyngolaryngeal morbidity. It is well known that the airway structures, including the jaw, tongue and larynx, fall backwards under the influence of gravity when oropharyngeal muscle tone is lost with the induction of anaesthesia. To overcome these obstacles, often a manual jaw thrust or tongue pull is used, with variable success. Insertion of the LMA against increased resistance from the collapsed soft tissues may be traumatic to airway structures, as evidenced by blood visible in oral secretions or on the LMA surfaces after device removal. This phenomenon can be problematic particularly with LMA's with an oesophageal access port.7 Difficulty in placement may also occur with orogastric or nasogastric tubes and transoesophageal echocardiography probes.8

In an effort to overcome problems encountered when advancing the LMA in the oropharynx, we propose a technique for LMA placement that involves manually lifting the larynx when resistance to advancement of an LMA with a pre-inflated cuff is experienced. The external larynx lift with pre-inflated cuff (ELL-PIC) method is designed to align the pharyngeal and laryngeal axes, thus avoiding traumatic impact of the LMA on laryngeal structures on entry of the device tip into the laryngeal inlet (Fig. 1). Also, using a semi-inflated LMA with 0 cuff pressure has been described for easier insertion due to a smoother surface against the palate.9 We hypothesised that this method would decrease airway trauma, thereby reducing postoperative complications. The purpose of this study was to validate the safety of LMA insertion using the ELL-PIC technique by assessing the postoperative pharyngolaryngeal complications of sore throat, dysphagia or dysphonia, as well as blood on the LMA at removal and to compare the incidence of these with standard insertion methods.

Fig. 1
Fig. 1:
(a) Shows that following the induction of anaesthesia, the entrance to oesophagus (indicated by the arrow) is obstructed by the laryngeal structures. (b) shows the entrance to the oesophagus (indicated by the arrow) during the external laryngeal lift.


Ethical approval for this study was provided by the Institutional Review Board of Northwestern University, Arthur Rubloff Building, 7th Floor, 750 N. Lake Shore Dr, Chicago, IL 80611, USA. The protocol (STU00046964) was reviewed and approval granted by Biomedical Research Panel D (Chairperson Frank Palella, MD) on 29 September 2011. The study was registered with registration number NCT01749033, principal investigator Ling-Qun Hu. The study was started on 25 April 2012.

The study was a randomised controlled study. Informed written consent was obtained from all patients or a legal surrogate. Inclusion criteria were patients receiving general anaesthesia for which LMA placement was not contraindicated, American Society of Anaesthesiologists class 1 to 3 and age 18 to 90 years. Patients were interviewed, consent obtained, clinical characteristics recorded and airway examination performed in the preoperative holding area by study personnel. Exclusion criteria were limited mouth opening (inter incisor gap <3 cm), preoperative sore throat, dysphagia or dysphonia, patients at increased risk of aspiration, morbid obesity (BMI > 45 kg m−2), poorly controlled gastroesophageal reflux disease, pregnancy, suspected oropharyngeal abnormalities, the planned use of nitrous oxide, anticipation of need for oral or pharyngeal suctioning, patients undergoing oral or nasal surgery, planned intubation or any oral instrumental manipulations (including the use of an oral airway) intraoperatively or postoperatively.

Prior to commencing the study, one of the study investigators (R.J.M.), using a random number generator, created group assignments to allocate patients into one of three groups. These group assignments were placed in opaque-sealed envelopes that were opened in sequence by the primary attending anaesthesiologist just before induction of anaesthesia. The study groups were: Group 1, insertion using the technique recommended by the LMA manufacture with a completely deflated cuff; Group 2, as for group 1, but without removal of the pre-inflated volume that is present in the LMA when received from the manufacturer; Group 3, as for Group 2, but with an added external laryngeal lift.

The external laryngeal lift was performed by placing the thumb and middle finger of one hand on either side of the larynx. The larynx was lifted in an upward direction, against gravity, to the neutral position while the LMA was inserted above the tongue with continued lifting of the larynx until there was loss of resistance to the advancement of the airway (Fig. 2). As the larynx is lifted, correct alignment of the tip of the airway with the oesophagus can be felt by the fingers of the lifting hand as the airway is advanced. Attending anaesthesiologists who participated in this study had practiced the manoeuvre on an airway mannequin prior to participation.

Fig. 2
Fig. 2:
Illustration of the external laryngeal lifting-inflating air technique.

A single use LMA Unique (uLMA) was used in all patients. The size of uLMA was selected based on the recommendation of the LMA manual: size 3 for adult 30 to 50 kg, size 4 for adult 50 to 70 kg and size 5 for adult 70 to 100 kg ( All the uLMAs were lubricated on the convex surface with a water-soluble lubricant. The primary attending anaesthesiologist assigned to the case inserted the uLMA.

After pre-oxygenation to an end tidal oxygen partial pressure at least 10.7 kPa, induction of anaesthesia was achieved by administration of intravenous propofol (200 mg) while simultaneously the patient breathed sevoflurane 8% in oxygen at a flow of 10 l m−1.10 Mask ventilation was not attempted, nor was a Berman-type oral airway used. The uLMA insertion was initiated when the patient was apnoeic and the depth of anaesthesia was judged appropriate (jaw relaxed, loss of eyelash reflex). Following insertion of the LMA, its cuff was inflated using a 20-ml syringe and the volume adjusted to a minimal oropharyngeal leak pressure (OLP) of 20 cmH2O (1.96 kPa). The OLP was assessed by presetting the positive end expiratory valve on the anaesthesia machine to 20 cmH2O. The total air volume in the cuff complied with the manufacturer recommendation ( Assessment of ventilation was confirmed by observing the chest rise and fall and the end tidal carbon dioxide (ETCO2) curve on the capnograph. A heat and moisture exchange device was used in all cases. Once regular spontaneous ventilation was achieved, the intra-cuff pressure of the uLMA was measured with a hand-held airway pressure manometer (King System Cuff Pressure Gauge; Ambu, Inc., King Systems Corporation, Noblesville, IN USA).

An anaesthetist, other than the primary attending anaesthesiologist, recorded the procedure data. Procedure data included the number of attempts for uLMA insertion and total uLMA insertion time. An insertion attempt was recorded with each passage of the uLMA device into the mouth and insertion time was the time from mouth entry to the appearance of an end tidal CO2 trace on the capnograph. In the case of multiple insertion attempts, this time was cumulative. While performing the uLMA placement the primary attending anaesthesiologist placed their non-dominant hand above the patient's larynx and, with a surgical drape, concealed the hand and larynx from the assistant.

The anaesthetic agents used for maintenance during the surgery were not protocol driven, although nitrous oxide was not allowed because of its effect on cuff pressure. Intraoperatively, following induction and the commencement of spontaneous respiration, fentanyl was titrated in 25 μg increments to maintain a respiratory rate of 11 to 13 breaths per minute. The uLMA was replaced if any of the following conditions were met: inability to achieve a 20-cmH2O seal pressure with the current uLMA placement, possible airway obstruction (noise with respiration, lack of chest movement, SpO2 less than 12.8 kPa with FiO2 100%), no ETCO2.

At the end of surgery, the anaesthetist removed the uLMA when the patients were able to open their mouths in response to a verbal command. To avoid the need for intraoral suction, all the uLMAs were removed without deflating the cuff. The uLMA cuff pressure, any use of pharyngeal suctioning, or the presence of blood on uLMA were noted by the anaesthetist managing the patient during the procedure. This anaesthetist was unaware of study group allocation. Patients were transferred to the post anaesthesia care unit and received routine care. Recovery time was divided into two phases. Phase 1 recovery was the time from the arrival of the patient in the recovery room unit until an Aldrete score of 8 to 10 was achieved. Phase 2 recovery was defined as the time from the end of Phase 1 recovery until the patient achieved a score of 9/10 on the post-anaesthesia discharge scoring system scale and fulfilled the clinical readiness discharge criteria.11

Using a standardised questionnaire as previously described,3,4 postoperative data collection was undertaken by study personnel who were unaware of study group allocation. Patients were interviewed in the post anaesthesia care unit and again at 1 and 2 h postoperatively. Patients were also contacted by telephone 24-h post discharge. Patients were asked to repeat standardised phrases and questioned as to the presence and degree of symptoms of sore throat (continuous pain in the throat independent of swallowing), dysphonia (difficulty or pain on speaking) and dysphagia (difficulty or pain evoked by swallowing) using a 0 to 100 scale. Patient satisfaction with the method of anaesthetic care was assessed at 2 h using a simple yes/no response.

The primary outcome was the incidence of a pharyngolaryngeal complications or blood on the uLMA at removal. A pharyngolaryngeal complication was defined as the occurrence of one or more of either a sore throat, dysphonia or dysphagia at any of the time points (1, 2 or 24 h). A score more than 0 for any of the individual pharyngolaryngeal complications was considered a positive response. Secondary outcomes included the individual component outcomes of the composite complications, uLMA insertion time and number of insertion attempts. Exploratory analysis included evaluation of independent predictors of a pharyngolaryngeal complication.

The sample size determined for this study was based on the following assumptions. A clinically important reduction in postoperative pharyngolaryngeal morbidity would be 50%, compared with the 21.4% reported in a study using a pre-inflated LMA in England and 33% based on a study of LMA intra-cuff pressure from Toronto.3,9 A sample size of 429 would attain an 80% power to detect an effect size difference with Cramér's ω of 0.15 using the chi-square test with 2 degrees of freedom and a significance level (alpha) of 0.05. We planned to recruit seven additional patients per group to account for failures to follow protocol or patients lost to follow-up, resulting in a total sample size of 450 patients. Sample size analysis was performed using PASS version 11 (NCSS, LLC, Kaysville, Utah, USA).

The primary outcome, the incidence of any one of the three primary airway complications or blood on the uLMA at removal, was analysed using a chi-square statistic. A post-hoc comparison was performed for the difference in the incidence of blood on the uLMA, using the external airway lift compared with the combined incidence in the groups without a laryngeal lift, and corrected for 2 comparisons and 97.5% confidence intervals (CIs) of the difference reported. Interval data were compared among groups using the Kruskal–Wallis H test, with post-hoc comparisons made using the Dunn test with Bonferroni correction for six comparisons. Nominal data were analysed using a chi-squared test. Exploratory analyses of risk factors for pharyngolaryngeal complications in patients with and without a pharyngolaryngeal complication were analysed using a chi-squared statistic and the Mann–Whitney U test. The severity of sore throat in patients who had blood present on the uLMA at removal compared with those who did not have blood present was analysed using then Mann–Whitney U test. Differences in the incidences of binomial outcomes were calculated using the Clopper–Pearson method. A value of P less than 0.05 was used to indicate statistical significance. Data analysis was performed using R-studio, version 0.99.473 (Foundation for Open Access Statistics, Boston, Massachusetts, USA) and R version 3.2.2 (release date 14 August 2015, The R Foundation of Statistical Computing, Vienna, Austria).


A total of 450 patients gave consent to participate in the trial, and 441 patients completed the study (Fig. 3). Patient characteristics, type of surgical procedure, opioid use and discharge times among study groups are shown in Table 1. There were no differences in intraoperative, postoperative or total opioid administration in the operating or recovery room. Median phase 1 recovery times were shorter in group G3 compared with group G1, but phase 2 recovery and discharge times did not differ among the groups.

Fig. 3
Fig. 3:
Consort diagram of participant flow in the study.
Table 1
Table 1:
Patient and airway characteristics, surgical procedures, opioid use and discharge times among study groups

There were no differences in uLMA insertion times, number of insertion attempts or duration of time the LMA was in the larynx among the study groups (Table 2). Cuff pressure and incidence of airway suctioning were also similar among groups. The incidence of pharyngolaryngeal complications in the groups were: Group 1, 57% (95% CI 49 to 65%); Group 2, 55% (95% CI 47 to 63%); Group 3, 52% (95% CI 44 to 60%); P = 0.77. Sore throat was the most frequently reported pharyngolaryngeal complications, reported by almost half of the patients. There was no difference in the severity of reported sore throat among groups. The incidence of blood on the uLMA was lower in patients in Group 3 (8%) compared with Groups 1 and 2 combined (17%), difference −8% (95% CI of the difference −0.8 to −17%, P = 0.01).

Table 2
Table 2:
Oropharyngeal airway characteristics and pharyngolaryngeal complications in the study groups

A greater number uLMA insertion attempts, a greater incidence of blood on the uLMA, higher cuff pressures and longer duration of the uLMA in the larynx were observed in patients who had a pharyngolaryngeal complication (Table 3). There was a difference in the incidence of sore throat between those who did or did not have blood on the uLMA: 68.3% of those patients who had blood present on the uLMA reported a sore throat compared with 43.9% who did not have blood present on the uLMA: the difference between the means was 24.2% (95% CI, 11 to 38%), P = 0.002. In patients who had blood present on the LMA, the median (interquartile range) of the severity of sore throat as assessed by the area under the throat pain score X time curve was greater than for those patients with no blood on the LMA: score × h, 213 (40 to 506) vs. 116 (30 to 330), P = 0.03.

Table 3
Table 3:
Comparison of risks factors associated with increased pharyngolaryngeal complications


The important finding of this study was that the time to placement of the uLMA using the ELL-PIC method was equivalent to the classic methods. Similarly, no difference in the composite pharyngolaryngeal complications was observed among insertion methods, confirming the validity of ELL-PIC as an alternative LMA insertion technique without increasing airway morbidity. The use of a laryngeal lift was associated with a lower incidence of blood on the airway at removal, suggesting that the method may decrease traumatic impact with the oropharyngeal structures during insertion. Given that the presence of blood on LMAs has been shown to be a predictor for both the occurrence and the severity of sore throat, our findings suggest that the ELL-PIC technique may be a useful alternative method for placement of a LMA, especially a LMA device that provides oesophageal access.

Anaesthesiologists strive to minimise iatrogenic injury to patients and to improve safety in the operating room. The LMA is a reliable airway device with relatively low morbidity associated with its use. The use of a LMA insertion method designed to align the airway structures may improve patient and practitioner satisfaction by minimising adverse events without increasing the time required for placement as well as minimising the number of times alternative airway management methods are required. The basic LMA insertion technique endorsed by the device manufacturers recommends the use of either the index finger or the thumb as a guide, but this method does not align the pharyngeal and laryngeal axes, especially when oropharyngeal muscle tone is lost with the induction of anaesthesia. Although LMA device design changes have been made no newer methods of insertion have been endorsed.

Previous studies have evaluated LMA insertion methods designed to make insertion easier and reduce pharyngolaryngeal morbidity. Pre-inflation of the balloon was one of the first methods examined for facilitating LMA placement and decreasing sore throat.9 Methods suggested to facilitate placement of the ProSeal LMA12 or the LMA Supreme13 have included the use of stylets or other types of tracheal tube introducer, with or without laryngoscopy. In addition, a rotating 90-degree technique for inserting the ProSeal LMA has been suggested to facilitate placement and reduce postoperative sore throat.14 Most of the proposed methods for insertion have been designed to overcome oropharyngeal impediments to LMA advancement, but have not been designed to align airway structures.

The incidence of pharyngolaryngeal complications (approximately 55%) observed in this study is greater than that commonly reported following LMA use (5 to 34%).15 Reasons for this difference between our study and others may include: the use of any airway complication outcome including sore throat, dysphagia and dysphonia; direct questioning of the patients at multiple time points 1, 2 and 24 h following LMA removal;3 and the acceptance of any reported pain intensity more than 0 as a positive response. However, the incidence of sore throat in this study was similar to that reported in the studies of Brimacombe et al.4 and in the control group of the study of Seet et al.,3 both studies that used methods of assessment similar to those used in the current study. Other factors that may have contributed to the high incidence of pharyngolaryngeal complications in our study include studying patients who were spontaneously breathing, use of intra-cuff volumes that created high intra-cuff pressures, and longer durations of surgery.

Previous studies have found that cuff deflation during insertion, using an inappropriate sized device, less experienced practitioners, pharyngeal suctioning, higher intra-cuff inflation pressure, and longer duration of surgery were associated with difficult placement or increased pharyngolaryngeal complications.3 In a retrospective study, female sex or a large neck circumference were also noted to be independent risk factors for difficulty on insertion.16 Although, in the current study, we did not find any difference in insertion times among our study groups, we did find a greater number of LMA insertion attempts, a greater number of patients with blood on the LMA, a longer duration of the LMA in the oropharynx and higher cuff pressures in patients who had pharyngolaryngeal complications.

The results of our study must be considered in light of our study's limitations. We did not limit or actively manage the intra-cuff pressure, but rather used the manufactures recommended volume in the cuff. In a randomised control trial of the LMA classic when intra-cuff pressure was limited to 3.9 to 4.3 kPa, as opposed to allowing practitioners to inflate the LMA cuffs according to their usual practice (measured pressures up to 15.2 kPa), Seet et al.3 reported a decrease in postoperative pharyngolaryngeal adverse outcomes, including sore throat, dysphagia and dysphonia. We chose a minimal OLP of 20 cmH2O (1.9 kPa) that should have been achieved with the LMA volumes used in this study.17 Nonetheless, even though the intra-cuff pressure reached 26.7 kPa, the limit of the Cuff Pressure Gauge, the desired OLP still could not be achieved in all cases. This may be the result of a poor fit between the uLMA and the oropharyngeal cavity making it impossible to achieve the desired OLP. Low OLP's may cause problems such as gas leak into the stomach with subsequent gastric distention and regurgitation. There is currently no standard of care guidelines for the ideal OLP. However, in one study, no differences in pharyngolaryngeal complications were found between groups with ∼4 and ∼24 kPa LMA intra-cuff pressures when LMA removal was performed in sleeping patients.18 We cannot be certain that all anaesthesiologists applied the same force when performing the ELL-PIC manoeuvre, but there were no differences observed among anaesthesiologists in terms of outcomes in the study. Neither can we determine if the ELL-PIC manoeuvre itself results in sore throat, but it is unlikely as the incidence of sore throat was not different among groups.

In conclusion, the observation of fewer cases with blood on the LMA device upon removal suggests that the ELL-PIC insertion method may have the potential to reduce morbidity associated with LMA devices that allow gastric access. This may be a result of lifting the relaxed musculature of the larynx that follows the induction of anaesthesia and so enabling non-obstructed access to the entrance of oesophagus.

Acknowledgements relating to this article

Assistance with the study: none.

Financial support and sponsorship: this study was supported by the Department of Anesthesiology, Northwestern University Feinberg School of Medicine, Chicago, Illinois, USA.

Conflicts of interest: none.

Presentation: preliminary data for this study were presented at the 2015 Annual Meeting of the International Anesthesia Research Society Meeting, 21 to 24 March 2015, Honolulu, Hawaii, USA.


1. Apfelbaum JL, Hagberg CA, Caplan RA, et al. Practice guidelines for management of the difficult airway: an updated report by the American Society of Anesthesiologists Task Force on Management of the Difficult Airway. Anesthesiology 2013; 118:251–270.
2. McHardy FE, Chung F. Postoperative sore throat: cause, prevention and treatment. Anaesthesia 1999; 54:444–453.
3. Seet E, Yousaf F, Gupta S, et al. Use of manometry for laryngeal mask airway reduces postoperative pharyngolaryngeal adverse events: a prospective, randomized trial. Anesthesiology 2010; 112:652–657.
4. Brimacombe J, Holyoake L, Keller C, et al. Pharyngolaryngeal, neck, and jaw discomfort after anesthesia with face mask and laryngeal mask airway at high and low cuff volumes in male and females. Anesthesiology 2000; 93:26–31.
5. Thiruvenkatarajan V, Van Wijk RM, Rajbhoj A. Cranial nerve injuries with supraglottic airway devices: a systematic review of published case reports and series. Anaesthesia 2015; 70:344–359.
6. Wakeling HG, Butler PJ, Baxter PJ. The laryngeal mask airway: a comparison between two insertion techniques. Anesth Analg 1997; 85:687–690.
7. Vezina MC, Trepanier CA, Nicole PC, et al. Complications associated with the Esophageal-Tracheal Combitube in the pre-hospital setting. Can J Anaesth 2007; 54:124–128.
8. Kallmeyer IJ, Collard CD, Fox JA, et al. The safety of intraoperative transesophageal echocardiography: a case series of 7200 cardiac surgical patients. Anesth Analg 2001; 92:1126–1130.
9. Matta BF, Marsh DS, Nevin M, et al. Laryngeal mask airway: a more successful method of insertion. J Clin Anesth 1995; 7:132–135.
10. Brimacombe J, Berry A. Insertion of the laryngeal mask airway – a prospective study of four techniques. Anaesth Inten Care 1993; 21:89–92.
11. Chung F, Chan VWS, Ong D. A postanesthetic discharge scoring system for home readiness after ambulatory surgery. J Clin Anesth 1995; 7:5600–6506.
12. El Beheiry H, Wong J, Nair G, et al. Improved esophageal patency when inserting the ProSeal laryngeal mask airway with an Eschmann tracheal tube introducer. Can J Anaesth 2009; 56:725–732.
13. Cook TM, Gatward JJ, Handel J, et al. Evaluation of the LMA Supreme™ in 100 non-paralyzed patients. Anaesthesia 2009; 64:555–562.
14. Jeon YT, Na HS, Park SH, et al. Insertion of the ProSeal laryngeal mask airway is more successful with the 90 rotation technique. Can J Anaesth 2010; 57:211–215.
15. Figueredo E, Vivar-Diago M, Muñoz-Blanco F. Laryngo-pharyngeal complaints after use of the laryngeal mask airway. Can J Anaesth 1999; 46:220–225.
16. Katsiampoura AD, Killoran PV, Corso RM, et al. Laryngeal mask placement in a teaching institution: analysis of difficult placements. F1000Res 2015; 4:102.
17. Keller C, Brimacombe J. Pharyngeal muscosal pressures, airway sealing pressures, and fiberoptic position with the intubating versus the standard laryngeal mask airway. Anesthesiology 1999; 90:1001–1006.
18. Rieger A, Brunne B, Striebel HW. Intracuff pressures do not predict laryngopharyngeal discomfort after use of the laryngeal mask airway. Anesthesiology 1997; 87:63–67.
© 2017 European Society of Anaesthesiology