Summary We conducted a two-part study to assess the practice of withholding neuromuscular blockade until the ability to ventilate the lungs using a bag and face mask (mask ventilation) has been established following induction of anaesthesia. The first part of the study consisted of a postal survey (71% response rate) of 188 anaesthetists in the Oxford region to assess their current practice. Thirty per cent of respondents always checked mask ventilation before administering a neuromuscular blocking drug, whereas 39% of respondents (all them consultants) never did this. A further 31% only did so in the case of known or anticipated difficulty with the airway. In the second part of the study, we measured inspired (VTI) and expired (VTE) tidal volumes before and after neuromuscular blockade in 30 patients undergoing general anaesthesia. The ratio VTE/VTI was used as a measure of the efficiency of ventilation. There was no difference in VTE/VTI before [mean (SD) 0.47 (0.13)] and after [0.45 (0.13)] neuromuscular blockade. We conclude that neuromuscular blockade does not affect the efficiency of mask ventilation in patients with normal airways.
It is the practice of some anaesthetists during routine induction of anaesthesia for elective surgery to check that they are able to ventilate the patient's lungs using a face mask (mask ventilation) before administering a long-acting neuromuscular blocking drug. We observed that this practice appeared to be more common among trainees. Consultants tended to administer neuromuscular blocking drugs immediately following the induction agent, claiming that mask ventilation is easier if neuromuscular drugs have been given. Full neuromuscular blockade might facilitate mask ventilation by increasing chest wall compliance or by reducing upper airway tone ; alternatively, it could make mask ventilation more difficult by inducing upper airway collapse [2, 3]. The value of withholding neuromuscular blocking drugs following induction of anaesthesia is therefore unknown.
We aimed first, to estimate the prevalence of the practice of mask ventilation before instituting neuromuscular blockade. Second, we wished to study the effects of neuromuscular blockade on the efficiency of mask ventilation during induction of general anaesthesia in patients with normal airways.
In the first part of this study, a questionnaire was sent to all consultant anaesthetists and specialist registrars (years 2–4) working in the Oxford region, asking whether it was their practice, in patients at low risk of regurgitation and aspiration, to ventilate the lungs by bag and face mask following induction of anaesthesia but prior to administration of a neuromuscular blocking drug. Respondents could either answer ‘yes’ or ‘no’ to the question, and those responding ‘yes’ were asked whether it was always their practice. Reasons for respondent's practice were also sought.
In the second part of the study, we recruited 30 ASA I–II patients in whom intubation was planned for elective oral surgery procedures, following approval from the Central Oxford Research Ethics Committee and obtaining written informed consent. Patients with a known or anticipated difficult airway and those at risk of regurgitation and aspiration of gastric contents were excluded. Height, weight and body mass index (BMI) were also recorded.
Standard monitoring (ECG, pulse oximetry, noninvasive blood pressure) was instituted, a 20 G intravenous cannula was inserted, and anaesthesia was induced using 2 mg midazolam, 0.1 mg fentanyl and a sleep dose of propofol. The patient's lungs were ventilated via a face mask connected to a Bain system, with 2% isoflurane in 10 l.min−1 oxygen. One of three anaesthetists performed mask ventilation while an assistant collected the data and administered the drugs. The anaesthetist ventilating the lungs aimed to achieve optimal chest movements using mask ventilation based on personal clinical assessment and was blinded to the data collected, to the nerve stimulator and to the drugs given. A Datex–Engstrom spirometry monitor (Helsinki, Finland) incorporating a D-lite reusable sensor in series with the breathing system was used to measure inspired (VTI) and expired (VTE) tidal volumes and mean (Pmean) and peak (Ppeak) airway pressures breath-by-breath. After six manual inflations of the lungs, 1 mg.kg−1 vecuronium was administered, and ventilation continued until complete loss of train-of-four as measured by peripheral nerve stimulation of the ulnar nerve. Following loss of the train-of-four a further six breaths were given, after which time the patient's trachea was intubated and anaesthesia and surgery continued.
To assess the possibility that the anaesthetist ventilating the lungs may tire during the course of the study, in two additional patients 5 ml normal saline was administered instead of vecuronium following the initial six breaths in a blinded manner. Mask ventilation was continued for 2 min following saline administration. These two subjects were incorporated randomly into our study group of 30 patients.
Data analysis and statistics
We used the ratio VTE/VTI as our primary measure of the efficiency of mask ventilation. This ratio was calculated for each of the six breaths before the neuromuscular blocking drugs were given, and the values averaged to obtain a mean VTE/VTI ratio for each subject. These values were then averaged to obtain the mean for the group, and this was used as a measure of the efficiency of mask ventilation before neuromuscular blockade. A similar process was followed for ventilation following the administration of neuromuscular blocking drugs, and for Pmean and Ppeak before and after these drugs were given. In the control patients, data from the final six breaths were taken as neuromuscular blocking drugs were not given to these patients.
We assumed that a mean difference of ≈ 100 ml between VTI and VTE would be clinically important, and from pilot studies the standard deviation of the tidal volume was 125 ml. Thus, to achieve a power of 0.8 at the 5% significance level we would need to study at least 25 subjects . Student's paired t-test was used to compare mean VTE/VTI ratio and mean Ppeak and mean Pmean before and following neuromuscular blockade. A p-value of < 0.05 was taken as statistical significance.
Questionnaires were sent to 147 consultants and 41 specialist registrars: 134 (71%) were completed and returned. Fifty-three respondents (39%) did not routinely ventilate the lungs by face mask before giving neuromuscular blocking drugs, whereas 81 (61%) did. Of the latter group, 40 (30% of the total) always practised this, whereas 41 (31%) only did so if a difficult airway was predicted. One respondent only employed this practice in infants. (Other individual comments made by respondents concerning reasons for mask ventilating prior to neuromuscular blockade included: ‘to oxygenate the patient during the apnoea seen after propofol target controlled infusion induction, but before the patient is adequately anaesthetised’; ‘to maintain oxygenation from the earliest opportunity’; and ‘for teaching purposes’.)
In total, 25 (83%) of 30 respondents with < 10 years' experience in anaesthesia routinely mask ventilated patients' lungs before muscle relaxation, compared with 55 (53%) of the 104 respondents with > 10 years experience. This last group was made up entirely of consultants.
In part 2 of the study, 13 males and 17 females were included. Mean (range) age 27 (16–51) years; height 1.71 (1.55–1.87) m; weight 75 (55–120) kg; and BMI 25.9 (20–38) kg.m2.
Figure 1 shows the VTE / VTI ratio for each subject before and after neuromuscular blockade. Values of VTE / VTI varied considerably but there appeared to be no consistent trend in the ratio. For the group as a whole, the mean VTE / VTI ratio was 0.47 (SD 0.13) before neuromuscular blocking drugs were given and 0.45 (0.13) following loss of the train-of-four (p < 0.45). Airway pressures were less variable than VTE / VTI (Figs 2 and 3), and like VTE/VTI, Ppeak and Pmean did not change consistently with neuromuscular blockade. For the group as a whole, Pmean was 15.4 (3.1) cmH2O before neuromuscular blocking drugs were given and 14.6 (3.1) cmH2O after loss of train-of-four (p < 0.31). Values for Ppeak were 18.3 (3.0) cmH2O before neuromuscular blocking drugs were given and 17.5 (3.7) cmH2O after loss of train-of-four (p < 0.12).
For the two control patients, VTE/VTI ratio was very variable for each breath and did not show any consistent change over time, as for the study patients who received neuromuscular blocking drugs. Mean VTE/VTI ratio was 0.42 (0.05) for the first six breaths and 0.45 (0.08) for the last six breaths (p < 0.20). Ppeak and Pmean values for the control patients also did not change over time: Ppeak was 18.3 (3.0) for the first six breaths and 19.0 (0.9) for the last six breaths (p < 0.63); and Pmean was 16.3 (4.1) for the first six breaths and 15.3 (0.8) for the last six breaths (p < 0.32). Anaesthetists did not report feeling tired during the study period.
Our postal survey suggested that although the majority of anaesthetists do consider mask ventilation before administering neuromuscular blocking drugs, only a minority do so routinely. The result that most anaesthetists who practice this are less experienced might suggest either that this is a recent trend, or that the technique is taught and practised early in one's career, but a significant number of anaesthetists choose to restrict or abandon it as they become more experienced.
The VTE/VTI ratio might not necessarily be the best measure of the efficiency of mask ventilation. However, no single measurement is accepted as a gold standard for assessing the balance between upper airway patency and chest wall/lung compliance in this situation. Abrams et al.  used noninvasive dynamic compliance from expired air analysis to estimate ease of mask ventilation, but they conceded that this measurement does not distinguish between upper airway and chest wall/lung factors. We felt that, if Pmean and Ppeak remained constant, the ratio VTE/VTI would be a reasonable measurement. This ratio has been used previously by others to assess airway patency during cricoid pressure [5, 6], as has the change in VTE against a background of fixed VTI.
We chose to ventilate the patients' lungs by hand. Previous investigators  have used a ventilator for this purpose while the anaesthetist held a face mask tightly over the patient's face. This enabled a fixed tidal volume to be delivered with each inspiration. However, this is not routine practice, and it is more common to ventilate by hand. The individual assessment of what constituted ‘suitable’ mask ventilation may have differed between operators in our study. However, the overall variability of our results was low, which does not suggest that this was a problem in practice. Furthermore, our mean values for airway pressure during mask ventilation are similar to those reported by others .
The values for VTE/VTI that we obtained (0.45–0.47) seemed unusually low. Following tracheal intubation, the mean value was only 0.67 (0.18), also surprisingly low. This was because we chose a relatively high fresh gas flow of 10 l.min−1. During some pilot measurements in which we used lower fresh gas flows of 1.0 l.min−1, we recorded VTE/VTI ratios of ≈ 0.7 with mask ventilation and ≈ 0.9 following tracheal intubation. We chose a fresh gas flow of 10 l.min−1 because this is often used during pre-oxygenation [9, 10], and hence subsequent induction of anaesthesia. This high fresh gas flow presumably leads to some of the excess ‘flow-by’ of gas past the spirometry monitor being detected as excess inspired volume. This effect is unlikely to have influenced our results, as the degree to which neuromuscular blockade affects VTE/VTI should be constant regardless of the initial value of VTE/VTI.
Our conclusions are necessarily limited. We studied patients whose lungs were all easy to ventilate and cannot comment on patients whose lungs might be difficult to ventilate. This is a relatively rare group , but one in which it would be important to investigate further the effect of neuromuscular blockade. Interestingly, Abrams et al.  reported that in patients made difficult to ventilate by face mask following the administration of sufentanil (which probably increases upper airway muscle tone), neuromuscular blockade restored the ease of mask ventilation.
For those anaesthetists who choose to ventilate a patient's lungs by face mask and find this easy, our results suggest that giving neuromuscular blocking drugs will make little difference to the subsequent ease of mask ventilation. However, if they find ventilation difficult, our study provides no data to help answer the question whether neuromuscular blocking drugs should be given. Larger studies, or those focusing on patients in whom mask ventilation is more likely to be difficult , might shed more light on this last question.