A randomised crossover trial comparing a single-use polyvinyl chloride laryngeal mask airway with a single-use silicone laryngeal mask airway


Correspondence to: S. Bell
Email: cerebellum100@hotmail.com


We compared insertion rates of single-use polyvinyl chloride laryngeal mask airways (LMAs) vs single-use silicone LMAs in 72 anaesthetised patients. Both airways were produced by Flexicare Medical. Laryngeal mask airway insertion was successful on the first attempt in 68/72 (94%) polyvinyl chloride LMAs vs 64/72 (89%) silicone LMAs (p = 0.39). Overall insertion rates were 72/72 (100%) for the polyvinyl chloride LMAs and 71/72 (99%) for the silicone LMAs (p = 1.0). Mean (SD) insertion times were similar for polyvinyl chloride and silicone LMAs: 24.3 (5.1)s vs 24.8 (7.8)s (p = 0.64). Laryngeal mask airway position, as assessed using a fibrescope, was not different (p = 0.077). The median (IQR [range]) leak pressure was 16 (12–20 [6–30]) cmH2O for the polyvinyl LMA and 18 (13–22 [6–30]) cmH2O or the silicone LMA (p = 0.037). In conclusion, we did not find any important differences between polyvinyl chloride and silicone laryngeal mask airways.

The Classic™ laryngeal mask airway (LMA) was introduced in 1988 and manufactured as a reusable device made from silicone. Single-use polyvinyl chloride (PVC) LMAs became available in 1997. The use of PVC LMAs has increased, partly due to concerns that the reuse of silicone LMAs risks disease transmission [1, 2]. Silicone is softer and less rigid than PVC, but costs more [3].

Single-use silicone LMAs are now available. These do not contain phthalates, which increase the plasticity of PVC. Phthalates leach into the environment where their accumulation exerts anti-androgen effects that disrupt animal embryogenesis, which led the United States National Research Council to recommend an investigation in 2008 [4]. Human research is limited, but exposure to phthalates in the general population may approach the maximum daily intake [5]. In selected high-risk groups, such as neonates or pregnant women, the US Food and Drug Administration recommend the use of phthalate-free devices whenever possible [6]. In Europe, devices containing phthalates must be clearly labelled as such and the use of phthalates must be justified by the manufacturer when the devices are intended for use with children or pregnant or nursing women [7].

Laryngeal mask airways made from PVC and from silicone have similar characteristics, such as first-time insertion rates and times [8–11]. However, Hanning et al. found that the oropharyngeal leak pressure was significantly higher in the single-use PVC LMA compared with the reusable (silicone) Classic LMA [12] Williams et al. found no differences for any outcome between single-use LMAs made of PVC or silicone (seal pressure, insertion times or success, ease of insertion, cuff pressures and volumes) [13].

No study has yet compared disposable PVC vs disposable silicone LMAs produced by the same manufacturer.

This randomised crossover study compared the success rate of insertion of a single-use PVC LMA (LarySeal Clear) with a single-use silicone LMA (LarySeal Blue), both produced by Flexicare Medical (Mountain Ash, UK). The shape of each type of LMA is identical (Fig. 1). The primary outcome was success rate for insertion of the devices at first attempt. Secondary outcomes included ease of insertion, ease of ventilation, cuff leak pressure and fibreoptic laryngeal view through the aperture of the LMA.

Figure 1.

 The single-use PVC LMA (LarySeal Clear) and single-use silicone LMA (LarySeal Blue) produced by Flexicare Medical (Mountain Ash, UK).


Following South Wales Local Research Ethics Committee approval and written informed consent, we enrolled 72 patients of ASA physical status 1-2 undergoing general anaesthesia, in whom an LMA was considered appropriate (January 2011 to August 2011). We did not study patients: younger than 18 years; having emergency surgery; with a BMI > 35 kg.m−2; with gastro-oesophageal reflux; with potentially difficult airways (Mallampati grade 3 or 4, mouth-opening less than two finger–breadths, thyromental distance < 6 cm); or otherwise unsuitable for a LMA.

All patients were fasted as per hospital and Royal College of Nursing guidelines [14]. The size of LMA was based on the patients’ body weight and the manufacturer’s recommendations. A size-3 LMA was used for patients weighing < 50 kg, size-4 for patients weighing 50–70 kg and size-5 for patients over 70 kg. The standard pre-use tests were performed on all devices and LMAs were lubricated with AquagelTM (Adams Healthcare, Leeds, UK) on the tip and posterior surface as recommended by the manufacturer. The LMA was deflated before insertion. The order of insertion of the LMAs was random and generated by Excel (Microsoft, Redmond, WA, USA).The study design is summarised by Fig. 2.

Figure 2.

 CONSORT flow diagram to illustrate the study design.

No pre-medications were given. The patient was positioned supine with the head resting on a soft pillow. Standard monitoring was established, intravenous access secured and baseline variables (blood pressure, oxygen saturation and heart rate) were recorded throughout. Whilst the patient was pre-oxygenated for 3 min using 100% oxygen, fentanyl 1 μg.kg−1 was administered intravenously. Anaesthesia was induced intravenously using 2–5 mg.kg−1 propofol 1% titrated by the anaesthetist until loss of the eyelash reflex was observed. Sevoflurane 2–8% in 100% oxygen was then given and the patient’s lungs manually ventilated until the anaesthetist determined that the jaw was relaxed.

Once adequate depth of anaesthesia had been achieved the first LMA was inserted, according to the manufacturer’s instructions [15]. The insertion time was defined from reception of the LMA by the anaesthetist to one complete successful breath seen on the capnographic trace. An insertion attempt was deemed unsuccessful if the device had to be taken out because of the absence of a capnographic trace or an audible leak. Two insertion attempts were allowed for each device. A failure was defined as two unsuccessful insertion attempts. If the oxygen saturation fell below 94% or the patient required intubation they were removed from the study.

Sevoflurane was then titrated to maintain general anaesthesia as assessed by clinical variables. The pressure inside the cuff of the LMA was measured using a calibrated aneroid cuff pressure gauge (VBM Medizintechnik GmbH, Germany) to ensure a pressure of below 60 cmH2O. At a fresh gas flow of 5 l.min−1, the airway pressure limiting valve was set to 30 cmH2O and the minimum airway pressure at which gas leaked around the device was obtained by listening for an audible leak. The subjective ease of insertion and ventilation were recorded using rating scores (easy, moderately difficult, very difficult, impossible). A fibreoptic endoscope was used to assess the position of the LMA according to the Brimacombe score [16].

The first device was removed after 3 min of use and replaced by the second LMA. Identical procedures were carried out and measurements obtained. The second LMA was then left in place for the surgical procedure. On removal both LMAs were examined for blood.

The same anaesthetists (SB) and (AR), each with a personal experience of more than 500 LMA placements, inserted all the devices.

The primary outcome measure was first-time insertion success rate. No prior data were available regarding success rates for insertion of either device. Brimacombe et al. reported the first-time success rate for insertion of the Portex Soft SealTM Laryngeal Mask to be 80% and that of the LMA Unique® to be 89% [17] (both examples of PVC LMAs). A sample size of 72 was calculated to detect a projected difference in 15% between the groups with respect to the first-time insertion success, to demonstrate superiority with a significance level of 0.05 and a power of 90%. Data were entered on to an Excel spreadsheet and statistical analysis was performed using SPSS (version 16; SPSS Inc., Chicago, IL, USA). First-time insertion rate was analysed using McNemar’s test. Ease of insertion, ease of ventilation, grade of view, seal pressure and insertion time were analysed using the Wilcoxon signed-rank test.


We recruited 72 patients to the study, all of whom were included in the analysis. Patients’ characteristics are shown in Table 1.

Table 1. Characteristics of 72 patients recruited to the study. Values are mean (SD) or number.
Patient characteristicValues
Age;years39 (17)
BMI; kg.m−225.9 (4.9)
ASA grade 1:252:20

There was no difference between PVC and silicone LMAs in: rate of first time insertion; overall success rate; insertion time; ease of insertion; ease of ventilation; or fibreoptic position (Table 2). The leak pressure was less with PVC than silicone LMAs. Laryngospasm precipitated removal of a silicone LMA in one patient.

Table 2. Success rate for first-time insertion, overall insertion successes, insertion time, leak pressure and Brimacombe score for PVC and silicone LMAs in 72 patients. Values are number (proportion), median (IQR [range]) or mean (SD).
 PVCSiliconep value
  1. *n = 71.

  2. †4, only vocal cords visible; 3, vocal cords plus posterior epiglottis visible; 2, vocal cords plus anterior epiglottis visible; 1, vocal cords not seen.

First-time insertion success68 (94%)64 (89%)0.39
Overall insertion success72 (100%)71 (99%)1.0
Insertion time; s24.3 (5.1)24.8 (7.8)*0.64
Leak pressure; cmH2O16 (12–20 [6–30])18 (13–22 [6–30])*0.037
Brimacombe score†

As this was a crossover trial, no assessment of postoperative pharyngeal morbidity was made; however, blood was seen on seven PVC LMAs and 10 silicone LMAs on their removal. Three patients had blood present on removal of both devices. One patient described a sore throat in recovery and one patient coughed during insertion. There were no episodes of desaturation or displacement.


Our results indicate that the material of manufacture may have less impact on the performance of a LMA than the shape, except for the slightly higher leak pressure provided by silicone.

The LarySeal Clear is made from PVC and is cheaper than silicone. If the performance is similar to devices made from silicone, then the cheaper cost confers an advantage to the PVC product. However, the question of whether devices made from PVC containing phthalates are safe is still under investigation. Phthalates pose a risk to particular patient groups, such as children and pregnant and nursing women. In the European Union, phthalates are banned from use in children’s toys, where the toys are intended to be placed in the mouths of children aged 3 and under [18]. Obviously, LMAs and tracheal tubes are also intended to be placed in patients’ mouths, including those of children; however, phthalates are still permitted for use in medical devices. Morton and colleagues demonstrated that phthalates can be released from tracheal tubes manufactured from PVC [19]. The Medical Devices Directive requires devices containing phthalates to be clearly labelled and a justification to be provided for the use of phthalates in medical devices intended for use with children and nursing and pregnant women [7]. We are not aware of any device manufactured from PVC to have this justification at present.

Studies comparing the performance of LMAs were reviewed in the Buyers’ Guide for LMAs published by the Centre for Evidence-based Purchasing, part of the NHS Purchasing and Supply Agency [3]. The first-time insertion success rates of both LMAs included in this study compare favourably with those of other devices available on the market.

The ‘ADEPT’ initiative developed by the Difficult Airway Society [20] recommends that only airway devices for which there is at least some evidence of efficacy should be purchased. This study provides evidence on two LMAs that have not previously been investigated.

The current list price for the silicone LarySeal Blue is £6.16 (€7.69; $9.91) per unit vs £4.48 (€5.59; $7.21) for the PVC LarySeal Clear, a difference of £1.68 (€2.10; $2.70) per unit. However, a recent amendment to EN ISO 14971, the standard for risk management of medical devices [21], states that economic considerations cannot be used when reducing risks: any risk has to be reduced as far as possible whatever the economic cost. Previously, risks have been reduced using the ALARP principle, as low as reasonably practical, and this included cost. The impact of this change of interpretation on reducing risk on the use of PVC, particularly when containing phthalates, in medical devices that have contact with patients, remains unknown.

Limitations to our study include the fact that both of the anaesthetists had more experience with PVC than silicone LMAs. This study was not blinded because we could not conceal the colour differences of the two LMAs. Our results may not be extrapolated to patients with different characteristics to our population.

In conclusion, the results of our study showed no differences between LMAs made of PVC or silicone, except that the seal pressure was higher for silicone. The phthalate content of PVC LMAs is a concern; additional research is required to rule out potential effects detrimental to human health, or an alternative plasticiser should be used.


The LarySeal Blue LMAs used during the study were provided by Flexicare Medical.

Competing interests

No external funding and no competing interests declared.