Volume 72, Issue 12 p. 1484-1490
Original Article
Free Access

The effect of neuromuscular blockade on the efficiency of facemask ventilation in patients difficult to facemask ventilate: a prospective trial

S. Soltész

Corresponding Author

S. Soltész


Department of Anaesthesia and Intensive Care, Kreiskrankenhaus Dormagen, Germany

Correspondence to: S. Soltesz

Email: [email protected]

Search for more papers by this author
P. Alm

P. Alm


Department of Anaesthesia and Intensive Care, Kreiskrankenhaus Dormagen, Germany

Search for more papers by this author
A. Mathes

A. Mathes


Department of Anaesthesia and Intensive Care Medicine, University Hospital of Cologne, Germany

Search for more papers by this author
M. Hellmich

M. Hellmich


Institute of Medical Statistics and Computational Biology, University of Cologne, Germany

Search for more papers by this author
J. Hinkelbein

J. Hinkelbein


Department of Anaesthesia and Intensive Care Medicine, University Hospital of Cologne, Germany

Search for more papers by this author
First published: 14 September 2017
Citations: 37

You can respond to this article at http://www.anaesthesiacorrespondence.com


Facemask ventilation of the lungs can be an important rescue intervention in a ‘cannot intubate’ scenario. We assessed the effect of neuromuscular blockade on expiratory tidal volumes in patients with expected difficulty in mask ventilation. The lungs of patients with at least three predictors of difficulty in mask ventilation were ventilated using a facemask held with two hands, with mechanical ventilation set in a pressure-controlled mode. Tidal volumes were recorded before and after the establishment of complete neuromuscular block. In 113 patients, median (IQR [range]) tidal volume increased from 350 (260–492 [80–850]) ml initially, by 48% to 517 (373–667 [100–1250]) ml 30 s after rocuronium administration, (p < 0.001). After the onset of the complete neuromuscular block, a median tidal volume of 600 (433–750 [250–1303]) ml was observed, corresponding to an increase of 71% from baseline values (p < 0.001), and 16% from values obtained 30 s after rocuronium administration, respectively; p = 0.003). No decrease in the tidal volume during the measurements was observed. We conclude that the administration of rocuronium at a dose of 0.6 mg.kg−1 was able to improve facemask ventilation in all cases with a potentially clinically relevant increase in tidal volume. The early use of a neuromuscular blocking agent can be considered as a therapeutic option in case of difficulty with mask ventilation.


Facemask ventilation is one of the fundamental skills required during induction of anaesthesia. Especially in cases of difficult tracheal intubation, it is an important alternative to ensure adequate oxygenation until the patient′s airway can be secured. Fortunately, the combination of tracheal intubation and facemask ventilation difficulties is a rare event 1, 2. Additionally, a difficult airway situation cannot be well predicted 3, 4, notwithstanding a recent commentary proposing that a binary approach leads to better prediction 5.

The role of neuromuscular blocking agents in this scenario is debated 6. They usually facilitate tracheal intubation and might improve facemask ventilation 7-9. However, on the other hand, they might impair ventilation by inducing upper airway collapse 10-12. Furthermore, return to spontaneous ventilation (a potential rescue technique) is achieved earlier if neuromuscular blocking agents are not used. Many investigations addressing the role of neuromuscular blockade in mask ventilation were usually performed in patients with normal airways. The more relevant data would be from patients with expected difficult facemask ventilation.

Therefore, the aim of the present study was to assess the influence of complete neuromuscular block in patients presenting with three or more risk factors for difficult facemask ventilation 2, 13. The primary outcome was the difference in tidal volume before vs. after onset of complete neuromuscular block.


The study design was prospective and took place at a single centre. After obtaining approval from the local ethics committee (Ethikkommission der Ärztekammer Nordrhein, Düsseldorf, Germany) and with written informed consent, we recruited adult patients, ASA physical status 1–3, undergoing elective surgery with general anaesthesia and presenting with three or more risk factors for difficult facemask ventilation 2, 13.

Exclusion criteria were: planned awake tracheal intubation; increased risk of pulmonary aspiration (gastro-oesophageal reflux, full stomach and intestinal obstruction); known allergy to the study drugs; pregnancy; neuromuscular disorders; drugs affecting neuromuscular block (e.g. furosemide, magnesium or cephalosporins); and hepatic or renal insufficiency.

Patients were premedicated with midazolam 7.5 mg orally. Anaesthesia was induced with a continuous infusion of remifentanil 0.1 μg.kg.min−1, a single dose of fentanyl 2 μg.kg−1 and propofol 2–3 mg.kg−1 until loss of eyelash reflex. Haemodynamic parameters were maintained within the normal range (±20% of baseline values) by additional administration of propofol and remifentanil in case of tachycardia or arterial hypertension; ephedrine or atropine were administered if hypotension or bradycardia occurred. All drug dosages (including neuromuscular blocking agents) were administered corrected to ideal body weight. For men, we calculated the ideal body weight (kg) to be: height (cm)−100−10%. For women, we calculated: height (cm)−100−15%.

After loss of eyelash reflex, the patients’ heads were slightly extended and an oropharyngeal airway was placed for standardisation. Anaesthesia for all patients was provided by the same anaesthetist (SS) with > 20 years clinical experience. A single-use facemask (Ambu™, Ambu GmbH, Bad Nauheim, Germany) was positioned on the mouth/nose with both hands simultaneously performing a jaw thrust in order to minimise airway resistance. The ventilator (Primus™, Draeger, Lübeck, Germany) was set to deliver pressure-controlled ventilation with initial peak inspiratory pressure (Pinsp) of 13 cmH2O, positive expiratory pressure (PEEP) of 3 cmH2O, 12 breaths.min−1, inspiratory–expiratory ratio of 1:1, and a fresh gas flow of 6 l.min−1.

If expiratory tidal volume was < 4 ml.kg−1 ideal body weight, Pinsp was adjusted in incremental steps of 5 cm H2O until the target tidal volume of 4 ml.kg−1 ideal body weight could be delivered. However, if pressure levels > 20 cm H2O had to be provided, a lower expiratory tidal volume of > 2 ml.kg−1 ideal body weight or 150 ml, respectively, was accepted (in order to reduce the risk of gastric distension and/or pulmonary aspiration).

Facemask ventilation was defined as impossible mask ventilation if during ventilation with elevated pressures > 30 cmH2O, delivered tidal volume was < 2 ml.kg−1 ideal body weight, no adequate chest rise was observed, no end-tidal CO2 was observed, or oxygen saturation measured by pulse oximetry decreased < 90%. In such instances, the patient was excluded from the study and a rescue plan was followed according to the standards of airway management in the department.

Further secondary exclusion criteria after induction of anaesthesia were potentially life-threatening emergency situations, such as: pulmonary aspiration; bronchospasm; severe hemodynamic instability; cardiac arrest; or allergic reaction.

Additionally, difficulty of facemask ventilation was assessed by the Warters grading scale for mask ventilation (Table 1) 9.

Table 1. Warters grading scale for mask ventilation. The point system is based on the ability to achieve a target volume of 5 ml.kg−1 9
Description/Definition Points
Oral or nasal airway 1
PIP 20–25 cmH2O 1
PIP 26–30 cmH2O 2
PIP > 30 cmH2O 3
Unable to generate PIP > 30 cmH2O 3
Two-person ventilation 2
Tidal volume 2–5 ml.kg−1 2
Unable to ventilate 4
  • PIP, peak inspiratory pressure.

After adequate ventilation had been established, respiratory minute volume and expiratory tidal volumes were measured over a period of 30 s (six breaths) by checking measurements on the ventilation machine. This was followed by administering rocuronium 0.6 mg.kg−1 (ideal body weight). Measurements of respiratory variables were repeated over a period of 30 s (six breaths), starting immediately after rocuronium injection and directly after establishment of complete neuromuscular block (loss of train-of-four, see below).

Neuromuscular transmission was measured by assessment of the train-of-four ratio using acceleromyography (TOF Watch SX, Essex Pharma GmbH, Munich, Germany) at the right adductor pollicis muscle with transcutaneous Ag/AgCl electrodes (electrocardiogram electrodes; Ambu Inc., MD 21060, USA); measurements started after induction of anaesthesia and were performed in intervals of 15 s. To minimise movement-induced changes in twitch response, the right hand was carefully fixed with tape on an arm board. Block onset was defined as the complete absence of contractions during train-of-four stimulation.

Differences in mean tidal volume (ml) were evaluated over time (baseline/rocuronium injection/complete neuromuscular block) using a linear mixed model for repeated measures with fixed effects group (easy/difficulty in mask ventilation), time and the interaction of group and time 14. An unstructured variance–covariance matrix was specified for the repeated effect time. Pairwise post-hoc contrasts were derived from estimated marginal means. A Bonferroni correction was applied to guard against type-1 error inflation due to multiple testing. Bivariate correlation was calculated according to Pearson. Values of p < 0.05 were considered to indicate statistical significance. Calculations were done with SPSS Statistics 24 (IBM Corp., Armonk, NY, USA) and SigmaPlot 12.3 (Systat Software, Inc., San Jose, CA, USA).

Based on results from previous studies 15, 16, we expected a mean (SD) change in tidal volume over time of 100–150 (300) ml. To detect such differences with a power of at least 95% at two-sided signficance level 5%, the paired t-test requires about 120 data pairs 17. Thus, a repeated measures ANOVA based on 120 patients with one between-subject factor (i.e. group) and one within-subject factor (i.e. time) should have ample power to detect relevant differences, even in the presence of some drop-outs 14.


Of the 120 patients who were included in the study, seven patients were not studied because the study protocol was not followed (incorrect drug doses). Baseline characteristics and the prevalence of risk factors for difficulty in mask ventilation of the remaining 113 patients are displayed in Tables 2 and 3.

Table 2. Risk factors for difficult mask ventilation [2, 13]. Values are number (proportion)
n =113
Neck radiation changes 0
Sex; male 67 (59%)
Snoring/sleep apnea 90 (80%)
Mallampati 3 or 4 90 (80%)
Presence of beard 33 (29%)
BMI ≥ 30 kg.m−2 62 (55%)
Age ≥ 57 years 83 (73%)
Jaw protrusion limited 38 (34%)
Cumulative number of risk factors for difficult mask ventilation
3 37 (33%)
4 50 (44%)
5 16 (14%)
6 8 (7%)
7 2 (1.8%)
8 0
Sum of risk factors 4 (3-4 [3-7])
Table 3. Baseline characteristics and cumulative risk factors for difficulty in mask ventilation (see Table 1 for risk factors) 2, 13. Values are presented as mean (SD) or number (proportion)
Age; years 62.2 (13.7)
Weight; kg 94.3 (19.1)
Height; cm 172.4 (9.2)
Sex; men 67 (59%)
BMI; kg.m−2 31.7 (5.9)
  • BMI, body mass index.

In all, 27 patients (23.8%) reached a score of 4 or more on the Warters scale (Table 1) and were considered as difficult to mask ventilate due to one or a combination of the following reasons. In 13 patients (11.5%), the Pinsp reached > 20 cm H2O (Fig. 1). In two cases (2%), leakage impeded adequate ventilation. In another 17 patients (15%), the desired tidal volume of 4 ml.kg−1 ideal body weight was not achieved initially with pressure levels < 20 cm H2O. In all these cases, the delivered expiratory tidal volume exceeded 2 ml.kg−1, or 150 ml, respectively; therefore the patients received rocuronium and measurements were performed as scheduled.

Details are in the caption following the image
Mean tidal volume (ml) over time, that is, estimated marginal means from linear mixed model with 95% confidence intervals (error bars); type-3 tests of fixed effects time (F2,111 = 77.4, p < 0.001), group (F1,111 = 5.9, p = 0.017), time × group (F2,111 = 8.2, p < 0.001). Black line: easy mask ventilation. Dotted line: difficulty in mask ventilation.

With regard to the primary outcome measure, median (IQR [range]) tidal volume of all patients increased from 350 (260–492 [80–850]) ml at baseline, by 48% to 517 (373–667 [100–1250]) ml 30 s after rocuronium administration; p < 0.001). After onset of complete neuromuscular block, a tidal volume of 600 (433–750 [250–1303]) ml was observed, corresponding to an increase of 71% from baseline values (p < 0.001), and 16% from values obtained 30 s after rocuronium administration, respectively; p = 0.003. Of note, no cases with a decrease in the tidal volume during the measurements were observed. Initial tidal volumes were lower in the sub-group of patients who were difficult to mask ventilate as compared with patients who were easy to ventilate. However, after rocuronium injection, tidal volumes increased in both groups compared with initial values. Therefore, 30 s after injection of rocuronium and after the onset of neuromuscular block, differences between groups were lost (Fig. 1).

A total of 51 patients (45.1%) could be mask ventilated sufficiently with the initial settings of the ventilator. However, the majority required higher pressures in order to reach desired tidal volumes, the mean (SD) pressure for all patients was 16.5 (3.9) cmH2O. Pressures in the sub-group of patients who were difficult to mask ventilate were 19.9 (5.6) cmH2O and therefore higher than in patients who were easy to ventilate (15.5 (2.5) cmH2O). Additionally, patients with BMI > 40 kg.m−2 required higher inspiratory pressures (19.2 (5.1) cm H2O) than patients with BMI <30 kg.m−2 (15.5 (3.1) cm H2O); p = 0.01, Fig. 2).

Details are in the caption following the image
Scatterplot of body mass index (BMI; kg.m−2) and peak inspiratory pressure (cmH2O) with linear regression line overlaid (Pearson's r = 0.344, p < 0.001). The bubble area is proportional to the number of (binned) observations.

Mean (SD) oxygen saturation increased during measurements from 97.5 (4.3)% initially to 99.2 (1.8)% 30 s after rocuronium administration (p < 0.01); and 99.5 (1.5)% at the onset of neuromuscular blockade (p < 0.01). However, in five patients, a temporary decrease in oxygen saturations below 90% occurred during mask ventilation (before administration of rocuronium), which was treated by adjusting peak pressure limits in all cases. The lowest SaO2 value measured was 80%.

Two patients were difficult to intubate tracheally (more than two attempts). Fortunately, mask ventilation could be performed without difficulty in both.


Our main findings were that neuromuscular block improved tidal volume during pressure-controlled ventilation via a facemask by a median of ~250 ml (70%) in patients presenting with three or more risk factors for difficult facemask ventilation and in a sub-group of patients whose lungs were actually difficult to ventilate. Of note, we did not observe a single case with deteriorating mask ventilation after the onset of neuromuscular block.

Definitions of difficult facemask ventilation are variable 18. Available scales (Han et al., Warters et al.) rely on individual evaluation rather than exact measurements 9, 19. Thus, the reported incidence varies from 0.08% to 15% depending on the criteria applied, thereby making direct comparisons of studies dealing with this topic difficult 1. This might contribute to the fact that results vary substantially between studies. The Han scale is a simple scoring system using four grades from 1 (ventilation without adjuncts) to 4 (unable to ventilate). The Han scale was not applicable for our investigation: due to the scheduled use of an oral airway and a two-handed technique, all of our patients would have been automatically classified as grade 3.

The Warters scale is a more appropriate score, adding points if measures such as oral airway, two-handed technique, high pressures etc. are required to achieve the desired tidal volume. Therefore, it reflects more accurately the measures performed in our study. A disadvantage of the score is that the target tidal volume was originally defined as > 5 ml.kg−1, which is higher than the minimum tidal volume recommended by other authors 16, 20. The decision to use the Warters scale probably influenced the proportion of patients whose lungs were considered to be difficult to ventilate in our study: If we had used 5 ml.kg−1 as a target tidal volume, either the required pressures would have been substantially higher in many patients, or the target tidal volume would not have been achieved in a higher proportion of cases. Thus, both situations would have increased the number of patients with a score of 4 or higher on the Warters scale.

An advantage of the present study was the use of pressure-controlled ventilation, both hands and an oropharyngeal airway device in all cases. With this ‘two-handed-jaw-thrust technique’ adopted from Joffe et al., greater tidal volumes can be achieved than with one-handed manual ventilation alone 16. Additionally, we facilitated the comparison of our data by providing clearly defined, identical and optimal settings for facemask ventilation without any changes during measurements. Furthermore, the documentation of the onset of the neuromuscular block with accelerometry contributed to the reliability of our data.

Another advantage of the present investigation is the assessment of patients not only at risk for difficult facemask ventilation, but being in fact actually difficult to ventilate via facemask. Several studies have assessed the influence of neuromuscular blocking agents, but most of them did not include many patients who were difficult to ventilate 8, 9, 15, 20, 21. However, this is the sub-group of real interest. If a patient's lungs are easy to ventilate, a slight deterioration does not matter. Nevertheless, in the scenario of difficult facemask ventilation, even small changes in tidal volume may be crucial.

The present investigation has several limitations. We did not observe any cases with impossible mask ventilation, most likely due to the fact that this situation, fortunately, is a very rare event. Therefore, the important question whether to administer neuromuscular blocking agents in a ‘cannot ventilate situation’ cannot be answered by our data.

Some of the underlying causes for difficulty in mask ventilation are not influenced by neuromuscular blocking agents, such as obesity, presence of a beard etc. On the other hand, several problems are caused by supraglottic obstruction, laryngospasm or by the administration of high doses of opioids 10. In these situations, a neuromuscular blocking agent might be beneficial in order to establish adequate ventilation. However, whether such causes can be discerned in an emergency situation, remains speculative.

Another methodological problem is that changes in the depth of anaesthesia over time might also contribute to changes in ventilation variables, even in the absence of neuromuscular blocking agents. In this regard, the second measurement directly after rocuronium administration in our investigation might represent a state of deeper anaesthesia than compared with baseline, yet without significant neuromuscular block, because even a fast acting drug such as rocuronium does not produce a complete neuromuscular block within a few seconds when administered at the dose used in our study. Therefore, the increase in tidal volumes might have been caused by deeper anaesthesia. This interpretation might explain why other authors have not observed improved ventilation after onset of neuromuscular block 15. Thus, a control group without neuromuscular blocking drugs would probably have been advantageous to exclude these effects.

Several studies investigating the effect of neuromuscular blocking agents on facemask ventilation confirm our results. Sachdeva et al. observed improved ventilation in 125 patients with normal airway anatomy 21. They administered pressure-controlled ventilation similar to our method. No case of a deterioration in ventilation after injection of rocuronium was documented. However, difficulty in mask ventilation was described in only eight patients classified by the Han scale 21. Thus, the authors recommended further research in patients with difficult airway anatomy. Warters et al. investigated 90 patients during rocuronium-induced neuromuscular block in a placebo-controlled study. They also observed a significant improvement in mask ventilation following the onset of neuromuscular block in all cases. Most of the patients in their cohort were easy to ventilate, which is reflected by mean Warters scale ventilation scores of 2.3 initially and 1.2 after rocuronium. In a review of paediatric patients, the use of neuromuscular blocking agents in order to improve ventilation was recommended by Engelhardt et al. 7.

In contrast, several authors came to different conclusions. Goodwin et al. did not find any differences before and after neuromuscular blockade in 30 patients with normal airways 15. Ikeda et al. compared the effects of rocuronium or succinylcholine in 31 patients whose lungs were ventilated with a pressure-controlled mode. Although rocuronium did not influence ventilation, succinylcholine improved tidal volume 8.

Recently, Joffe et al. studied 210 patients before and after neuromuscular blockade. Ventilation was performed using a manual single-hand technique; drug dosages were left to the discretion of the attending anaesthetist and 20% of the patients presented with three or more risk factors for difficulty in mask ventilation. Overall, tidal volume improved after onset of the block, by about 50 ml. However, they also observed deterioration in ventilation parameters in 40 patients after onset of neuromuscular blockade. No patient became impossible to ventilate. The authors concluded that facemask ventilation was not worse after administration of neuromuscular blocking agents 20.

Since we found that neuromuscular blockade improved mask ventilation, early use of a neuromuscular blocking agent should be considered as a therapeutic option. Nevertheless, checking the ability to mask ventilate before administration of neuromuscular blocking agents remains important, because it provides useful information about the non-paralysed condition of the patient's airway 22. If difficulty in mask ventilation is observed before paralysis, a short-acting neuromuscular blocking drug should always be preferred over a long-acting one to maximise the chance of return of spontaneous ventilation 22, 23.


Registered at ClinicalTrial.gov (NCT01849211). No external funding or competing interests declared.