Bag-mask ventilation is a fundamental airway skill that can be lifesaving in many emergency situations. There are concerns that the art of bag-mask ventilation is being lost due to reduced teaching of the technique and the increased use of the laryngeal mask airway. Patients continue to be at risk as a result of failed intubation and inadequate bag-mask ventilation .
Considerable variation in face mask design exists and many new single-use devices are becoming available. Manikins have previously been used to evaluate bag-mask ventilation but there have been very few attempts to assess the influence of face mask design [2-5].
The aim of this study was to compare the performance of four face mask designs in the bag-mask ventilation of a manikin.
We received confirmation, from the hospital research and development department and the local research ethics committee, that no formal ethics or research review was required for this study because the main objective of the study was to determine the performance of the equipment not the performance of anaesthetists.
This was a randomized crossover study in which 48 anaesthetists (of all grades) were asked to ventilate a ‘basic airway model’ with four different face masks (Fig. 1), each for 3 min. The face masks were all size 4 and the order in which they were used was determined by computer-generated randomization (to eliminate bias due to fatigue and learning). These four face masks were routinely used in the department and were therefore familiar to all anaesthetists. Anaesthetists were told to use jaw thrust in the same way that they would treat a patient (although they were limited to holding the mask with one hand) and to aim for tidal volumes of 500 mL at a rate of 15 breaths per minute. The lung simulator was clearly visible to allow a visual indicator of successful ventilation (comparable to observing chest wall movement clinically).
The airway model used  was an adapted Laerdal manikin attached to a Medishield lung simulator (Fig. 2). The manikin has been adapted to limit head extension and the lung simulator is adjusted to its lowest compliance setting to simulate a patient that is difficult to ventilate. Flow meters, attached to the self-inflating bag and the lung simulator, allow the volume delivered to the mask and bellows to be measured. The flow meters used were Ohmeda 5420s. The same two meters were used throughout the study and were calibrated prior to initial use and regularly thereafter. A software program (Lab View 5; National Instruments Inc., Austin, Texas, USA) was then used to calculate the percentage leak for each tidal volume delivered and the results were displayed graphically on a computer screen. The average leak for each 3 min ventilation period was determined by a blinded observer drawing a line of best fit, taking into account only the leak data for the most representative tidal volumes over the 3 min and ignoring the periods when masks became disconnected or the flow meters transiently malfunctioned.
After using each face mask, volunteers were given a rest period until they felt able to continue with the next mask in the study. During this rest period, visual analogue scores were completed. Masks were rated on a 100 mm line (from 0 to 100 mm) for each of three questions: ‘How comfortable did your hand feel at the end of 3 min?’; ‘How easy did you find it to achieve and maintain a seal?’; ‘How happy would you be to use this mask in an emergency situation?’.
ANOVA with post hoc testing with Bonferoni correction for multiple comparisons was used to analyse the results (SPSS version 11, Chicago, IL, USA). Correlations were also obtained using SPSS version 11.
Forty-eight anaesthetists took part in the study: 8 senior house officers, 25 registrars and 15 consultants. The mean anaesthetic experience of the volunteers was 10 yr (range 2-37 yr). The median glove size was 7.5 (range 5.5-9).
The study demonstrated statistically significant differences in the performance of the face masks (Table 1). Overall, the two disposable face mask designs both performed better than the reusable masks. The reusable Laerdal mask performed relatively poorly in comparison with the other masks. There was a large variation in performance and satisfaction with each face mask (Figs 3 and 4).
We analysed the data to determine the effect of anaesthetic experience and glove size on performance. There was no significant correlation between the number of years experience in anaesthesia and leak (mean leak for the four face masks) (P = 0.83).
There was a mild but significant (negative) correlation between glove size and leak (Pearson correlation −0.34, P = 0.019).
To determine the effects of fatigue and learning, we subdivided the leak data for each of the 3 min of ventilation. We analysed the results using a double repeated measures ANOVA test and found no significant difference in leak over time for the four masks as a group (P = 0.121) or for the masks individually (P = 0.071).
This study has shown that for any given face mask there was a large variation in the performance and satisfaction of anaesthetists. No design was rated highly by all the volunteers. We feel that this highlights the importance of having a variety of different face masks available for emergency use.
An area of great concern currently is the potential risk for the transmission of prion diseases such as variant Creutzfelt-Jacob disease through the use of reusable surgical and anaesthetic equipment. Prions are highly resistant to standard methods of cleaning and sterilization . Therefore there is a demand for disposable equipment of an adequate quality to replace reusable items safely. It is reassuring that the single-use devices tested in this study outperformed the reusable face masks.
A modified manikin was used to assess the four face masks. The manikin has its limitations and is not a perfect simulation of the human face. The rubber surface of the manikin is much firmer than human skin. This probably reduced the ability of the masks to form a seal (particularly the masks without inflatable seals). The other limitation of using a manikin model to simulate a difficult airway is that it does not reflect the variations in face shape, dentition and facial hair that commonly cause problems with bag-mask ventilation. There are other factors that influence the choice of face mask such as dead space, patient comfort and ability of the anaesthetist to see what is happening inside the mask (vomit, secretions, misting and lip colour). A clinical study could provide a more representative evaluation of face mask performance.
A Royal College of Anaesthetists working party discussed the problem of declining standards in airway management . It was suggested that less time was being spent teaching good face mask techniques to trainees. This is as a result of less overall time spent in theatres (reduction in doctors' working hours), the routine use of the laryngeal mask airway and less emphasis on airway skills from consultant supervisors.
The problem is highlighted by the large variation in performance between anaesthetists in this study although surprisingly there was no correlation between experience and performance. Several senior anaesthetists commented that the task would have been a lot easier 10-15 yr ago when they regularly performed face mask anaesthetics. The use of simulators or airway models could provide a way to improve the training and maintenance of bag-mask ventilation skills [8-10].
This study was designed to test face masks and therefore leak was chosen as the primary outcome. However, some anaesthetists in this study delivered perfect tidal volumes for the entire 3 min period but had substantial air leak from around the mask. Therefore we feel that the ability to deliver good tidal volumes is a better outcome (of more clinical relevance) when the airway model is used to test anaesthetists' performance.
In conclusion, we believe that bag-mask ventilation is fundamental airway technique that is often poorly performed. We recommend that more importance be given to the choice and range of face masks available for use in airway emergencies.
1. Cooper GM. Is the art of airway management being lost? Roy Coll Anaesth (UK) Bull
2. Stewart RD, Kaplan R, Pennock B, Thompson F. Influence of mask design on bag-mask ventilation
. Ann Emerg Med
3. Finer NN, Rich W, Craft A, Henderson C. Comparison of methods of bag and mask ventilation for neonatal resuscitation. Resuscitation
4. Devitt JH, Brookes DA, Oakley PA, Webster PM. Mask lung ventilation by ambulance personnel: a performance assessment. Can J Anaesth
5. Osterwalder JJ, Schuhwerk W. Effectiveness of mask ventilation in a training manikin. A comparison between the Oxylator EM100 and the bag-valve device. Resuscitation
6. Stacey MRW, Doherty J, Mecklenburgh JS. Basic airway model (poster presentation). World Congress of Anaesthesiologists 2004, Paris.
7. Blunt MC, Burchett KR. Variant Creutzfeldt-Jakob disease and disposable anaesthetic equipment
- balancing the risks. Br J Anaesth
8. Schaefer JJ. Simulators and difficult airway management. Paediatr Anaesth
9. Stringer KR, Bajenor S, Yentis SM. Training in airway management. Anaesthesia
10. Mason RA. Education and training in airway management. Br J Anaesth
1Presented in part at the AAGBI Annual Congress Meeting, 21-24 September 2004, Cardiff, UK.