Comparison of the LMA Supreme vs the i-gel™ in paralysed patients undergoing gynaecological laparoscopic surgery with controlled ventilation


  • Presented in part at the 13th Asian Australasian Congress of Anaesthesiologists in Fukuoka, Japan, June, 2010.

Correspondence to: Dr Wendy H.L. Teoh


We compared the efficacy of the inflatable cuff of the LMA Supreme™ against the non-inflatable i-gel™ cuff in providing an adequate seal for laparoscopic surgery in the Trendelenburg position in 100 female patients. There was no difference in our primary outcome, oropharyngeal leak pressure, between the LMA Supreme and the i-gel (mean (SD) 26.4 (5.1) vs 25.0 (5.7) cmH2O, respectively; p = 0.18). Forty-seven (94%) LMA Supremes and 48 (96%) i-gels were successfully inserted on the first attempt, with similar ease, and comparable times to the first capnograph trace (mean (SD) 14.3 (4.7) s for the LMA Supreme vs 15.4 (8.2) s for the i-gel; p = 0.4). Gastric tube insertion was easier and achieved more quickly with the LMA Supreme vs the i-gel (9.0 (2.5) s vs 15.1 (7.3) s, respectively; p < 0.001). After creation of the pneumoperitoneum, there was a smaller difference between expired and inspired tidal volumes with the LMA Supreme (21.5 (15.2) ml) than with the i-gel (31.2 (23.5) ml; p = 0.009). There was blood on removal of two LMA Supremes and one i-gel. Four patients in the LMA Supreme group and one patient in the i-gel group experienced mild postoperative sore throat.

Supraglottic airway devices with gastric access tubes are increasingly being used in surgery requiring general anaesthesia and positive pressure ventilation. The i-gel™ (Intersurgical Ltd, Wokingham, UK) is a unique disposable supraglottic airway device introduced clinically in January 2007. It comprises a soft, gel-like, non-inflatable cuff made of thermoplastic elastomer, a widened, flattened stem with a rigid bite-block that acts as a buccal stabiliser to reduce axial rotation and malpositioning, and an oesophageal vent through which a gastric tube can be passed. Preliminary studies have demonstrated its easy reliable insertion, providing an adequate seal with a low morbidity rate [1–3]. It is a reasonable alternative to tracheal intubation during pressure-controlled ventilation [4], and can be used as a conduit for tracheal intubation and rescue airway management [5–7].

The LMA Supreme™ (LMA-S™; Laryngeal Mask Company, Singapore), on the other hand, is another supraglottic device introduced in 2007 [8, 9], with many similar characteristics to the i-gel: single use; presence of a drain tube to separate the gastrointestinal tract from the respiratory tract; and built-in bite block. It differs from the i-gel in that it is constructed of medical grade silicone, and has an inflatable cuff, a reinforced tip, and an elliptical, anatomically shaped, semi-rigid airway tube.

We aimed to compare the efficacy of the non-inflatable cuff of the i-gel with the inflatable cuff of the LMA Supreme in providing an adequate seal for laparoscopic surgery in the Trendelenburg position in paralysed patients with apparent normal airways, receiving controlled ventilation. Our primary outcome measure was oropharyngeal leak pressure. We also studied their relative ease of insertion, haemodynamic response and time to insertion, efficacy in controlled positive pressure ventilation and complications of use as secondary outcome measures.


This study was approved by the domain-specific Centralised Institutional Review Board of Singapore Health Services Private Limited (Ref: 2009/771/D), and was conducted in accordance with the declaration of Helsinki. Written, informed consent was obtained from all patients. A total of 100 patients scheduled for elective gynaecological laparoscopic procedures in our tertiary maternity and women’s hospital were recruited from October to December 2009. We did not study patients of ASA physical status 3 or 4 or those aged < 21 or > 80 years, those weighing < 30 kg or with BMI > 40 kg.m−2, or those with a predicted difficult airway (based on a history of difficult airway, inter-incisor distance < 20 mm, cervical spine pathology, modified Mallampati class 4, or thyromental distance < 65 mm), a high risk of regurgitation or aspiration, respiratory tract pathology, pre-operative sore throat, or a planned operation time > 4 h.

Patients were randomly allocated into two groups, ‘LMA Supreme’ or ‘i-gel’, using a computer-generated random number table, by the principal investigator. After recruitment, the enrolling investigators opened sealed, opaque envelopes that concealed the group allocation. Participants were blinded to their group allocation. The size of the airway was chosen in accordance to manufacturers’ recommendations. For the Supreme LMA, a size 3 was used if < 50 kg, a size 4 if 50–70 kg and a size 5 if 70–100 kg for the i-gel, a size 3 was used if 30–60 kg, a size 4 for 50–90 kg and a size 5 for > 90 kg.

Patients were not premedicated. They were positioned supine on the operating table, with the head resting on a jelly doughnut. Standard monitoring was instituted before induction of anaesthesia, i.e. pulse oximetry, electrocardiograph and non-invasive blood pressure. Pre-oxygenation was carried out with high flow oxygen for 3 min before induction of anaesthesia with intravenous fentanyl 1–2 μ.kg−1 and propofol 2.0–3.0−1. Atracurium 0.5−1 was administered for neuromuscular blockade after confirmation of successful manual bag-mask ventilation.

The LMA Supreme and i-gel were lubricated and prepared according to the manufacturers’ recommendations. Three minutes after administration of the neuromuscular blocking drug, the airway device was inserted when the jaw was sufficiently slack. The cuff of the LMA Supreme was inflated with air to attain a cuff pressure of 60 cmH2O as measured with a handheld aneroid manometer (Portex® Pressure Gauge; Smiths Medical Intl Ltd, Kent, UK). The appearance of the first square end-tidal carbon dioxide trace denoted successful establishment of effective ventilation. Otherwise, the device was completely removed for another insertion attempt. Three insertion attempts were allowed. Each ‘attempt’ was defined as re-insertion of the airway device into the mouth. We defined ‘insertion failure’ of the device as one comprising > 3 unsuccessful attempts or if the entire process of insertion exceeded 120 s. This included the time the airway device was removed from the mouth and any bag-mask ventilation in between. In case of failure of both devices, the airway was secured according to the decision of the attending anaesthetist.

Once the airway device was in place, the tube was fixed by taping over the patient’s cheek. A gel plug was placed in the proximal 1 cm of the gastric drain outlet, and the suprasternal notch test was done to confirm placement (gently tapping the suprasternal notch causes the gel to pulsate, confirming the tip’s location behind the cricoid cartilage) [10]. For both the airway devices, a gastric tube was inserted through the gastric drain outlet (size 14 FG for the LMA Supreme and 12 FG for the i-gel). These gastric tubes were prelubricated with a water-soluble lubricant. Ease of insertion was graded 1–3 (1 = easy, 2 = difficult, 3 = impossible). Time to insertion of the gastric catheter was also noted. Confirmation of correct placement of the gastric catheter was through detection of injected air by auscultation of the epigastrium and aspiration of gastric contents. Gastric decompression was performed and the amount of gastric fluid aspirated was noted.

The oropharyngeal leak pressure was measured after closing the adjustable pressure limiting valve with a fresh gas flow of 3 l.min−1, noting the airway pressure at equilibrium or when there was an audible air leak from the throat. The maximum pressure allowed was 40 cmH2O. The epigastrium was also auscultated when measuring the oropharyngeal leak pressure to detect any air entrainment in the stomach.

We recorded the number of insertion attempts and the time to establish effective ventilation (interval from when the LMA Supreme or i-gel entered the mouth to first CO2 trace), the ease of insertion of the airway, subjectively assessed on a 5-point scale (1 = easy, 2 = not so easy, 3 = difficult, 4 = very difficult, 5 = impossible), blood pressure and heart rate every minute for the first 5 min from induction of anaesthesia, and any manoeuvres required to optimise positioning or ventilation (adjusting head and neck position or depth of insertion, applying jaw lift, and changing the size of device).

Maintenance of anaesthesia was achieved with an oxygen:air mixture in 1–2 MAC sevoflurane. Initial ventilator tidal volumes were set at 8−1. Volume controlled, positive pressure ventilation was used to achieve O2 saturation ≥ 95% and end-tidal CO2 4.7–6.6 kPa through tidal volumes of 8–10−1 and respiratory rate of 10–16 per minute. All patients were positioned head-down (Trendelenburg position) after the surgeon inserted the laparoscopic ports and created a pneumoperitoneum using carbon dioxide insufflation. The magnitude of this Trendelenburg position was left to the surgeon’s discretion. Thirty minutes after commencement of the laparoscopic procedure, we recorded the difference between the inspired and expired tidal volumes achieved and calculated the expired-inspired tidal volume difference. Airway pressure (before and after creation of pneumoperitoneum) and intra-abdominal pressure during pneumoperitoneum were noted. The LMA Supreme cuff pressure was measured 30 and 60 min into surgery.

At the end of surgery, the effects of neuromuscular blocking drug were reversed with neostigmine 0.04−1 and atropine 0.02−1. The airway device was removed upon return of spontaneous breathing and eye opening of the patient. The airway device was then inspected for the presence of visible blood. Forty-five minutes later, patients were assessed by a blinded independent observer for postoperative sore throat, dysphonia or dysphagia.

All airway insertions were supervised by senior anaesthetists (WHLT and ATHS, each with > 10 years’ experience with supraglottic airway management), and performed by anaesthetic trainees and fellows (LKM, TS, ZY, MMT) with 1–5 years’ experience with the classic laryngeal mask airway, and at least 10 LMA Supreme and i-gel insertions before the trial’s commencement. Contemporaneous data collection of airway insertion times, ventilatory parameters and complications of placement (desaturation < 95%, gross regurgitation or aspiration (defined as fluid in the ventilation tube), bronchospasm, mucosal, lip, tongue or dental injury) was done by an unblinded observer who was not involved in the study.

Our primary comparison measure was oropharyngeal leak pressure. Sample size was based on a pilot study we conducted involving 20 LMA Supreme insertions that demonstrated a mean (SD) oropharyngeal leak pressure of 28 (5) cmH2O. To detect a difference of 10%, prospective power analysis at 80% power and the 0.05 level of significance (two-tailed) showed that a sample size of 47 patients would be required. Therefore, we recruited 50 patients per group to account for dropouts and protocol breaches. We used Student’s t-test to analyse insertion times and other continuous data, the Mann–Whitney U-test for Mallampati scores and other non-continuous data, Fisher’s exact test for comparison of side effects, and a general linear model for repeated measures for haemodynamic variables. All statistical analyses were performed using spss 13.0TM (SPSS Inc., Chicago, IL, USA) software. A p value of < 0.05 was deemed statistically significant.


One hundred patients were recruited and there were no dropouts. Their baseline characteristics and pre-operative airway morphology are summarised in Table 1.

Table 1.   Patients’ characteristics. Data are mean (SD) or number (proportion).
 LMA Supreme (n = 50)i-gel (n = 50)
Age; years37.9 (7.8)34.4 (8.5)
Height; cm159.2 (6.2)158.0 (5.7)
Weight; kg60.5 (9.9)60.9 (15.9)
Body mass index; kg.m−224.2 (4.8)24.4 (6.0)
Fasting time; h12.1 (2.9)13.3 (2.7)
ASA class 1/235/15 (70.0/30.0)37/13 (74.0/26.0)
Mallampati class 1/2/322/21/7 (44.0/42.0/14.0)21/17/12 (42.0/34.0/24.0)
Thyromental distance
 < 6.5 cm2 (4%)5 (10%)
 > 6.5 cm48 (96%)45 (9%)
Sternomental distance
 < 12.5 cm1 (2%)4 (8%)
 > 12.5 cm49 (98%)46 (92%)
Interincisor distance
 < 4 cm5 (10%)9 (18%)
 > 4 cm45 (90%)41 (82%)
 Ability to protrude lower jaw; yes/no47 (94%):3 (6%)48 (96%):2 (4%)
Head/neck movement;
 normal > 90°:Abnormal < 90°50 (100%):049 (98%):1 (2%)
Duration of surgery; min69.8 (48.5%)61.2 (38.2%)
Type of surgery; n (%)
 Total laparoscopic hysterectomy4 (8%)2 (4%)
 Laparoscopic cystectomy12 (24%)9 (18%)
 Laparoscopic myomectomy7 (14%)3 (6%)
 Therapeutic laparoscopy6 (12%)1 (2%)
 Laparoscopic tubal ligation14 (28%)29 (58%)
 Others7 (14%)6 (12%)

There was no difference in oropharyngeal leak pressure in the two airway devices (Table 2). Forty-seven (94%) LMAs Supreme and 48 (96%) i-gels were successfully placed on the first attempt, with similar ease and comparable times to achieve an effective airway successfully (Table 2). Three LMAs Supreme and two i-gels were successfully placed on the second attempt. There were no cases of failed insertion of either airway device. Of the subjective five-point scale used, 88% of both LMA Supreme and i-gel insertions scored 1 (easy), and 12% scored 2 (not so easy). None was rated 3 (difficult to insert).

Table 2.   Airway insertion characteristics and performance of airway devices. Values are mean (SD) or number (proportion).
 LMA Supreme (n = 50)i-gel (n = 50)p value
  1. *Defined as time from insertion of airway device into patient’s mouth to the first end-tidal carbon dioxide trace.

  2. †Defined as time from the gastric tube’s first entering the gastric drainage channel to 64 cm depth of insertion.

Size of airway used : 3/4/510/40/022/28/00.071
Number of insertion attempts
 147 (94%)48 (96%)0.648
 23 (6%)2 (4%) 
Reported ease of insertion
 1 = easy44 (88%)44 (88%)1.000
 2 = not so easy6 (12%)6 (12%) 
 3 = difficult00 
 4 = very difficult00 
 5 = impossible00 
 Time to successful airway insertion; s*14.3 (4.7)15.4 (8.2)0.404
 Cuff volume at 60 cmH2O; ml25.2 (5.0)NA 
 Time to gastric tube insertion; s†9.0 (2.5)15.1 (7.3)< 0.001
Ease of gastric tube insertion
 1 = easy50 (100%)39 (78%)< 0.001
 2 = difficult011 (22%) 
 3 = impossible00 
 Gastric volume on insertion; ml7.5 (12.9)5.8 (12.3)0.500
 Leak pressure < 20 cmH2O3 (6%)6 (12%)0.487
 Oropharyngeal leak pressure; cmH2O26.4 (5.1)25.0 (5.7)0.180
LMA Supreme cuff pressure; cmH2O
 at 30 min51.6 (7.0)NA 
 at 60 min49.2 (4.8)NA 

The mean times from insertion of the airway device to the first capnograph trace were similar for both LMA supreme and i-gel however, it was more difficult and took significantly longer to insert the gastric tube in the i-gel group (Table 2). The volume of gastric aspirate obtained upon insertion of the gastric tubes was comparable.

The haemodynamic consequences of airway placement did not differ significantly between the groups using the general linear model analysis for repeated measures. Systolic blood pressure (p = 0.64), diastolic blood pressure (p = 0.70), mean arterial pressure (p = 0.69) and heart rate (p = 0.63) were comparable between groups.

None of the patients in our study tested positive for air leakage into the stomach by auscultation over the epigastrium at the registered oropharyngeal leak pressure. Forty-seven LMA Supreme and 47 i-gel insertions did not require any optimisation manoeuvres and had no difficulty in oxygenation or ventilation. The depth of insertion of the i-gel had to be changed in two patients with one patient requiring removal and reinsertion of the device, whereas three LMA Supremes had to be reinserted to optimise ventilation. Intra-operative ventilatory parameters are summarised in Table 3. The mean airway pressure before and after creation of a pneumoperitoneum, the ventilator rate and intra-abdominal pressures were similar in both groups. However, there was significantly more leakage with the i-gel, with a greater difference in mean expired and inspired tidal volume with the LMA Supreme (Table 3).

Table 3.   Ventilatory parameters of the LMA Supreme and I-Gel. Values are mean (SD).
 LMA Supreme (n = 50)i-gel (n = 50)p value
Airway pressure; cmH2O
 Before pneumoperitoneum16.5 (4.1)15.8 (3.5)0.359
 After pneumoperitoneum23.8 (5.0)22.4 (4.7)0.167
Intra-abdominal pressure; cmH2O14.3 (2.1)15.0 (2.3)0.104
Ventilator rate; min12.0 (1.6)12.1 (1.7)0.764
Initial inspired tidal volume; ml471.7 (55.8)482.2 (71.7)0.417
Expired tidal volume; ml454.5 (56.9)461.7 (73.6)0.585
Expired-inspired tidal volume difference; ml21.5 (15.2)31.2 (23.5)0.009

The mean (SD) volume of air needed to achieve a cuff pressure of 60 cmH2O was 25.2 (5.0) ml in the LMA Supreme. This cuff pressure decreased to 51.6 (7.0) cmH2O after 30 min, and was 49.2 (4.8) cmH2O at 60 min in 22 out of 50 patients. Data on LMA Supreme cuff pressures at 60 min were not available in the remaining 28 patients as two laparoscopic procedures were abandoned 45 min into surgery and converted to an open surgical technique, due to technical difficulty and bowel perforation respectively. The other 26 laparoscopic procedures were completed before 60 min in the LMA Supreme group.

There were no significant differences in complications associated with the use of either airway devices: 45 (90%) LMA Supreme vs 48 (96%) i-gel placements were uneventful. One patient in the LMA Supreme group experienced transient desaturation (O2 saturation < 95%). No patient in the study experienced gross regurgitation, bronchospasm, lip injury, tongue trauma, dysphonia, dysphagia or dental injury. On removal, visible blood indicative of mucosal injury was noticed on one i-gel and two LMAs Supreme. Four patients in the LMA Supreme group and one patient with the i-gel experienced mild postoperative sore throat.


In our study, we found both the LMA Supreme and i-gel easy and quick to insert, with an equally high successful insertion rate on the first attempt. Both airway devices achieved comparable oropharyngeal leak pressures and proved to be effective ventilatory devices for gynaecological laparoscopic procedures. All gastric tubes were successfully inserted on the first attempt. It was, however, more difficult to insert a well-lubricated 12-FG gastric tube into the i-gel due to the smaller aperture of the gastric access port, and therefore this took longer.

We also found a statistically significant air leak with the i-gel, with a mean difference in expired and inspired tidal volume of 10 ml between both devices. It may be argued that the non-inflatable cuff and gel-like material of the i-gel theoretically renders it more susceptible to airway leaks if the wrong size is chosen and the anatomical fit is not correct. In our study, we chose the size of the i-gel airway according to the weight criteria of the manufacturer’s recommendations: size 3 for 30–60 kg and size 4 for 50–90 kg [11], but some degree of overlap exists. If the patient weighed between 50 and 60 kg and the size of the i-gel was chosen at the discretion of the consultant anaesthetist depending on the patient’s mouth opening, thyromental distance and neck width [12]. We inserted 22 size-3 and 28 size-4 i-gels. Indeed, Gatward et al. [2] have previously demonstrated the suitability of the size-4 i-gel in 100 non-paralysed patients weighing 42–113 kg, with findings comparable to ours: a median leak pressure of 24 cmH2O, insertion times of 15 s, and 7−1 expired tidal volume in 96% of patients.

We measured oropharyngeal leak pressure using two methods: auscultation and manometer stability [13]. The values obtained have been found to be similar using either method. There was no air leak into the stomach or gastric insufflations in any of our patients at the equilibrium leak pressure.

Uppal et al. [4] showed that the i-gel had no significant difference in gas leak compared with tracheal tubes when ventilating at moderate pressures up to 15–20 cmH2O, but did not study pressures higher than 25 cmH2O. All our patients were placed in the head-down position as required in most gynaecological laparoscopic procedures lasting approximately an hour. However, we found the mean airway pressures in our study did not exceed 25 cmH2O despite the Trendelenburg position and pneumoperitoneum. This could be due to the relatively low BMI in our female population and our surgeons’ care in limiting intra-abdominal pressures throughout the operation (mean 15 cmH2O).

Theiler et al. [14] recently compared the use of both the LMA Supreme and the i-gel in a randomised, crossover setting in simulated difficult airways and found equal insertion success rates (95% for the LMA Supreme vs 93% for the i-gel) and clinical performance (initial tidal volumes and airway leak pressure), with less epiglottic downfolding and a better fibreoptic view, but a longer insertion time with the i-gel: 34 (12) s vs 42 (23) s; p = 0.024.

Unlike our findings, when Fernandez et al. [15] evaluated the ease of insertion of the LMA Supreme and the i-gel in 85 patients requiring mechanical ventilation, they found the i-gel had longer insertion times (LMA Supreme 27.1 s vs i-gel 32.5 s), lower first attempt placement rates (LMA Supreme 95.2% vs i-gel 86%), and harder insertion of the nasogastric tube (lower success on first attempt (97.6% vs 85.7%), and longer insertion (9.5 s vs 22.1 s, respectively). However, as in our study, they found both devices provided an effective seal and efficacy of mechanical ventilation (peak pressure, mean pressure, compliance, and the ratio of tidal volume to respiratory frequency at 10, 30 and 60 min into surgery) with similar minor incidences of complications.

Shin et al. [16] found airway leak pressures of the i-gel and Proseal LMA (27.1 (6.4) and 29.8 (5.7) cmH2O respectively) to be significantly higher than that of the classic LMA 24.7 (6.2) cmH2O, with similar first-attempt success insertion rates, but a larger incidence of sore throat in the classic LMA group.

We found oropharyngeal leak pressures of < 20 cmH2O in three LMA Supreme and six i-gel cases after uneventful insertions. Each case passed the positive suprasternal notch test and had gastric tubes successfully inserted. Their leak pressures were obtained using the manometer stability technique. No audible throat noise or leak was heard at the airway pressure when equilibrium was achieved on the manometer. Ventilation was not compromised in any of these cases, with delivery of adequate tidal volumes and anaesthetic agents for the duration of surgery. This finding challenges the widely perceived notion that a leak pressure of < 20 cmH2O in supraglottic airway devices necessitates removal of the ‘malpositioned’ device due to potential to compromise its ventilatory efficacy. Indeed, a recent randomised crossover study that compared the disposable LMA Supreme with the Proseal concluded that ventilation was not affected despite lower leak pressures (mean (SD) 19.6 (5.8) vs. 20.9 (6.7) cmH2O) and inferior fibreoptic positioning of the LMA Supreme [17].

The i-gel was designed to create a perilaryngeal anatomical seal without an inflatable cuff, decreasing the risk of compression trauma in cadaveric studies [18]. The use of the i-gel has now been shown clinically to result in fewer postoperative sore throat and neck complaints compared with disposable LMAs [19]. Although our study was not powered to examine airway complications or morbidity, bloodstaining on removal was found in one i-gel (2%) and two (4%) LMA Supremes. Four patients (8%) with the LMA Supreme and one patient (2%) with the i-gel experienced mild postoperative sore throat.

Paradoxically, patients’ tongues can get caught inside the i-gel’s bowl during insertion and any undue resistance encountered or force used on insertion can result in tongue trauma, although this is quite rare, reported in 3 out of 1100 i-gel insertions [20]. Nerve damage to the mental nerve has also been described resulting in lower lip numbness in 1 out of 800 insertions [21]. We did not encounter any of these in our study. The wide buccal stabiliser aimed at preventing rotation and the rigid integral bite block renders the design of the i-gel bulkier and harder around the lips compared with the design of the LMA Supreme. This can make the lower lip of the patient less visible to the anaesthetist standing at the head of the patient, and care should be exercised when securing the i-gel with tape as the lower lip can get caught between the i-gel’s bite block and the lower teeth, with this going unnoticed throughout the procedure.

Our study has a few limitations. We did not measure the leak pressure at the end of surgery for either device. It could have perhaps added important information as reports have emerged that the seal of the i-gel seems to improve over time due to the thermoplastic cuff’s warming to body temperature [1]. We did not use fibreoptic bronchoscopy to assess the anatomical position of the i-gel and LMA Supreme in relation to the vocal cords for two reasons. First, we wanted this study to reflect our clinical practice and replicate our daily workflow and high surgical turnover. It was deemed not clinically and logistically feasible to perform endoscopy in all cases. Second, there is evidence that anatomical findings do not necessarily correlate with clinical consequences [22]. It is also impossible to blind the airway operator to the device used, hence there is a potential for bias. We only studied non-obese women and the results cannot directly be extrapolated to other types of patients.

In conclusion, we have demonstrated in this study that the airway seal pressure offered by the i-gel is comparable to that of the LMA Supreme. We found a mean difference in expired and inspired tidal volume of 10 ml between both devices; and a longer time to insert the gastric tube in the i-gel even though the clinical relevance is questionable. Both supraglottic airway devices are comparable in terms of ease of insertion, success rates on the first attempt, time to insertion and oropharyngeal leak pressure, proving to be equally effective ventilatory devices for gynaecological laparoscopic procedures in our study.


We thank our anaesthetic colleagues at the Department of Women’s Anaesthesia, KK Hospital and nursing staff for their help in the trial, and TE Medicare Pte. Ltd, the local distributor of Intersurgical (Wolkingham, Berkshire, UK) for providing free samples of i-gel. No external funding or competing interests declared.