Secondary Logo

Journal Logo

Original Article

A comparison of three facemasks used during the induction of general anaesthesia

Nandalan, S. P.*; Waters, J. H.

Author Information
European Journal of Anaesthesiology: October 2006 - Volume 23 - Issue 10 - p 869-873
doi: 10.1017/S0265021506000627



Facemasks are a vital component of airway management in anaesthesia and emergency resuscitation. In the field of resuscitation, bag-valve-mask ventilation is the first line approach at managing the airway. There are several types of facemasks available, which vary in their fit to the patient's face and in their efficiency for gas delivery to the patient. We hypothesized that different designs of facemasks would provide different results in the same patient. A prospective crossover randomized study was designed to compare three facemasks used during the induction of general anaesthesia.


Approval was obtained from the Gloucestershire Hospital Research Ethics Committee. After written consent, 60 adult patients of ASA status 1 or 2 and undergoing elective ENT surgical procedures requiring anaesthesia with neuromuscular relaxation were studied. Patients at risk for gastric regurgitation and vomiting were excluded from the study.

The following three facemasks were compared:

‘PLASTIC-1’: Clear plastic ‘Anaesthetic Facemask’ with pre-inflated air cushion, no inflation valve, single use (Intersurgical Limited, Wokingham, Berkshire, RG41 2RZ, UK).

‘PLASTIC-2’: Clear plastic ‘Air Cushion Mask’ with inflatable air cushion and inflation valve, single use (Proact Medical Limited, Kettering, Northants, NN16 8ER, UK).

‘RUBBER’: Black ‘Antistatic Rubber Facemask’ with pre-inflated air cushion, no inflation valve, reusable (Datex Ohmeda Limited, Hatfield, Hertfordshire, AL9 5EN, UK).

Following preoxygenation for 3 min, anaesthesia was induced with intravenous (i.v.) fentanyl 1–2 μg kg−1 and propofol 2–4 mg kg−1. Then an intubating dose of a non-depolarising muscle relaxant was administered (atracurium 0.5 mg kg−1, rocuronium 0.6 mg kg−1 or vecuronium 0.1 mg kg−1). Anaesthesia was maintained using isoflurane and nitrous oxide in oxygen delivered through a facemask, Bain co-axial circuit and a 2-L reservoir bag. Standard monitoring was instituted using electrocardiogram, non-invasive blood pressure, pulse oximetry, capnography and inspired oxygen analysis.

After anaesthetic induction, adequate time (3 min for atracurium and vecuronium, or 90 s for rocuronium) was allowed to achieve intubating condition of muscle relaxation. Then an appropriately sized Guedel's oropharyngeal airway was inserted. Each of the facemasks was used in sequence to ventilate the patient's lungs using the Bain circuit. The total fresh gas flow was set to 6L min−1. A size 3 or 4 facemask was used for female patients and a size 4 or 5 for male patients. Six combinations were possible for the sequence in which the facemasks could be tested on any single patient. Each patient was allocated to one of these sequences by using a block randomization table, each block consisting of the six possible sequences. Thus, each sequence was tried in a total of 10 patients. All three facemasks were tested in each patient in a crossover fashion. Blinding was not performed.

The anaesthetist normally allocated to the patient performed the insertion of oropharyngeal airway, and facemask ventilation. A one-handed mask ventilation technique was employed. We selected a range of anaesthetists to participate in the study, many of whom participated more than once. All three facemasks had been in use in our institution for more than 12 months. All the consultants were familiar with the three facemasks, whereas the trainees had varying levels of expertise.

After five successful lung inflations with each facemask, the following data (grade) were recorded: the degree of chest expansion (Good–Moderate–Poor), the feel on the reservoir bag (Good–Moderate–Poor), the degree of leak around the mask (Minimal– Moderate–Excessive) and the overall satisfaction of the anaesthetist (Good–Moderate–Poor). One of the authors ensured that the methods were correctly followed, and recorded the data. Laryngeal mask airways (LMA), and cricothyrotomy sets were immediately available as airway rescue option for any patient with impossible bag-valve-mask ventilation. Thereafter, the anaesthesia continued as planned.

Any decreases in pulse oximetry reading below 95% were recorded. We also recorded the patient's age, body weight, the level of anticipated difficulty in mask ventilation and the seniority of the anaesthetist. To evaluate the anticipated difficulty in mask ventilation, we considered the following: a modified Mallampati score greater than 3, a thyromental distance less than 6.5 cm, body mass index (BMI) greater than 35, the absence of teeth and the presence of a beard. Presence of one of these factors was recorded as ‘moderate’ (difficulty) and the presence of two or more factors was noted as ‘difficult’ mask ventilation.

Statistical methods

Due to the absence of objective end-points, a power analysis was not performed to calculate the sample size. The three facemasks were compared using the Friedmann test for multiple treatment of a series of subjects. Where a significant difference was found, the facemasks were compared in pairs using Wilcoxon signed rank sum test.


The patients had a median age of 35.5 yr (range 18– 76). There were 35 male and 25 female patients and the mean (SD) body weight was 82 (18.5) kilograms. The anticipated difficulty in mask ventilation was noted as ‘moderate’ in 20 cases and ‘difficult’ in two cases. The anaesthetists were trainees in 31 cases and consultants in 29 cases. None of our patients had a pulse oximetry reading of less than 95%. The airway rescue option was not required in any of the cases. Table 1 shows the scores achieved by each facemask, and the statistical comparison (P value) between the three facemasks for all patients. Tables 2–5 show the scores achieved by each facemask and the statistical comparison between the facemasks in subgroups by the anticipated difficulty in mask ventilation, and by the grade of the anaesthetist.

Table 1
Table 1:
All patients (n = 60); scores achieved by each facemask, and comparison between the three facemasks using the Friedmann test (P value).
Table 2
Table 2:
Patients with anticipated moderate, or difficult mask ventilation (n = 22); scores achieved by each facemask, and comparison between the three facemasks using the Friedmann test (P value).
Table 3
Table 3:
Patients with anticipated easy mask ventilation (n = 38); scores achieved by each facemask, and comparison between the three facemasks using the Friedmann test (P value).
Table 4
Table 4:
Patients anaesthetized by trainee anaesthetists (n = 31); scores achieved by each facemask, and comparison between the three facemasks using the Friedmann test (P value).
Table 5
Table 5:
Patients anaesthetized by consultant anaesthetists (n = 29); scores achieved by each facemask, and comparison between the three facemasks using the Friedmann test (P value).

All patients

The Friedmann test demonstrated that the facemasks differed significantly (P < 0.05) in all the four categories. Wilcoxon signed rank sum test showed that the two ‘PLASTIC’ masks were similar in their performance (P > 0.05), and the ‘RUBBER’ mask was significantly better than the two ‘PLASTIC’ masks.


The above difference was maintained when the cases with any anticipated difficulty in mask ventilation (moderate, and difficult cases; n = 22) were analysed. With the other three subgroups, a significant difference was observed in at least two categories. Where there was a significant difference, the Wilcoxon signed rank sum test showed that the ‘RUBBER’ mask was significantly better than the two ‘PLASTIC’ masks, which were similar.


We tested three facemasks with unique features and were in routine use in our institution. The ‘PLASTIC-1’ mask was made of clear plastic, had a pre-inflated air cushion. ‘PLASTIC-2’ mask was also made of clear plastic but had a valve that was used to inflate the air cushion to fit the patient's mouth and nose. The ‘RUBBER’ mask was made of black antistatic rubber and had a pre-inflated air cushion. The former two were cheap and for single use whereas the ‘RUBBER’ mask was expensive and reusable. We could not identify a disposable facemask in clinical use, which mimicked the features of the ‘RUBBER’ mask. Overall, the ‘RUBBER’ mask performed better than the other two. It was heavier than the others, and the rubber material may have conformed better to the mouth and nose providing better mask-fit and more ease to the operator.

The advent of disposable equipment has resulted in many different styles of facemasks becoming commercially available. The mask design should allow the anaesthetist to achieve effective seal and easy ventilation for the many facial types of patients. This study has demonstrated that the black antistatic rubber facemask outperformed the plastic facemasks in these respects. Whilst we fully support the idea of disposable facemasks to improve infection control, it must not be at the cost of patient safety. Today's trainee anaesthetists have a shorter training and less clinical exposure to ‘bag and mask’ anaesthetics than their predecessors. Walsh and colleagues [1] commented that spending less time manually ventilating the lungs, since the introduction of the laryngeal mask, might have reduced proficiency among anaesthetists. In their study, the emergency medical technicians had outperformed all grades of anaesthetists in single operator bag-valve-mask ventilation.

It is notable that the ‘RUBBER’ mask performed significantly better in all the four categories when used for patients with any anticipated difficulty in facemask ventilation. The difference was less evident in those with anticipated easy facemask ventilation. The trainee anaesthetists in our study encountered significant degree of leak with the ‘PLASTIC’ masks and were more satisfied with the ‘RUBBER’ mask. On the other hand, the consultant anaesthetists, who were more familiar with the given facemasks, did not experience a significant difference in leak. Garcia-Guasch and colleagues demonstrated that inexperienced operators took 29% more time to correctly place the facemask, 13% more time with the laryngeal mask and 12% more time with the cuffed oropharyngeal airway when compared with the experienced operators [2]. Therefore the trainees would benefit from more training with the ‘PLASTIC’ masks to achieve leak free placement.

Actual scores achieved by each facemask, as shown in the Tables 1–5, show that the ‘RUBBER’ mask mostly improved the ‘moderate’ scores to ‘good’ scores. The ‘poor’ scores remained numerically similar between the three facemasks. It is possible that the ‘RUBBER’ mask may not perform any better than the ‘PLASTIC’ masks in some really difficult scenarios.

This study has the following limitations. Firstly, our results may have been influenced by the fact that many of the anaesthetists participated more than once, and the participants were at different parts of the learning curve with the facemasks. Secondly, we did not evaluate the facemasks in awake patients. The clear plastic masks would have the advantage of patient preference in situations such as preoxygenation and inhalational induction. Thirdly, all our patients were anaesthetized, paralysed and fully monitored, which offered ideal conditions for airway management. However, the patients who need emergency airway control often have intact reflexes, and the prevailing conditions would be far from ideal. Therefore, our results are not directly applicable to the emergency situations. Finally, we used subjective criteria to score the ease and quality of ventilation. Subjective assessment of chest expansion is the method of evaluation used clinically and during training. In the presence of any obvious leaks with bag-valve-mask ventilation, there is a natural tendency for the anaesthetist to increase the grip on the facemask in order to achieve a better seal. The difference in performance between the facemasks would be only apparent to the person using them, hence the use of subjective criteria.

In conclusion, this study has shown that during routine induction of general anaesthesia, the antistatic rubber facemask performed significantly better than the two plastic disposable facemasks. We hope this study will encourage the manufacturers to improve the design of clear plastic disposable facemasks to simulate the features of the antistatic rubber facemask.


1. Walsh K, Loveday K, O'Rathaille M. A comparison of the effectiveness of pre-hospital bag-valve-mask ventilation performed by Irish emergency medical technicians and anaesthetists working in a tertiary referral teaching hospital. Ir Med J 2003; 96: 77–79.
2. Garcia-Guasch R, Ferra M, Benito P, Oltra J, Roca J. Ease of ventilation through the cuffed oropharyngeal airway (COPA), the laryngeal mask airway and the face mask in a cardiopulmonary resuscitation training manikin. Resuscitation 2001; 50: 173–177.


© 2006 European Society of Anaesthesiology