Bag-valve-mask (BVM) ventilation is mainly used in prehospital settings to ventilate patients in respiratory failure and/or cardiac arrest. This difficult procedure has attracted the attention of scientists since the eighties. Indeed, many studies have investigated factors such as ventilation techniques [
All these factors raise the question of the reliability of manikin-based studies. This research consists of an evaluation of the impact of these factors on manual ventilation performance, in terms of analysis methodology, system-based simulations, or operator characteristics. We also propose a new analysis method, able to provide a chronological view of ventilation variability in order to accurately assess manual ventilation performance in bench conditions.
We conducted this experimental trial at the Department of Emergency Medicine and Critical Care at the University of Franche-Comté Medical and Trauma Centre, Doubs Fire Department, and Jussieu Ambulance Services. The study protocol was submitted to the French data protection authority (Commission Nationale de l’Informatique et des Libertés, CNIL, registration number 1645179). The need for ethical approval was waived by the institutional ethics committee (Comité de Protection des Personnes CPP Est II). We enrolled healthcare professionals who were still in service, aged over 18 years, coming from different emergency structures (University of Franche-Comté Medical and Trauma Centre, Doubs Fire Department, and Jussieu Ambulance Services). Data from 140 volunteers were collected.
Prior to start, participants signed informed consent and fulfilled a questionnaire. We determined the size of the hand squeezing the bag and the grip strength of both hands using a tape measure and a Takei® dynamometer, respectively. Participants were instructed to ventilate with a bag-valve-mask a manikin simulating a 75 kg adult patient in respiratory arrest as they are used to do in their everyday practice. They were blinded regarding their ventilation performance. Because participants were not able to see the manikin chest rise, they were asked to ventilate the manikin for one minute before the tests in order to accustom themselves to the bench model. Then, participants manually ventilated the manikin with two adult self-inflating bags (Laerdal Bag® II and Ambu Spur® II with a reservoir volume of 2900 and 2600 mL, resp.) for five minutes each in a random order. Ventilation was performed in a standing position using a standard technique without chest compression, that is, one hand keeping the mask on the manikin’s face and the other hand squeezing the bag.
A Laerdal Airway Management Trainer manikin (Laerdal Medical, Stavanger, Norway) was installed on a stretcher. The manikin’s lungs were bypassed and directly connected to an ASL 5000® breathing simulator (IngMar Medical, Ltd., Pittsburgh, PA, USA) with a short respiratory hose (ID = 2.2 cm,
We also used two VTT® flowmeters (Jeulin, MediaScience, Haute-Normandie, France) to measure gastric and BVM insufflation flows.
VTT sensor signals were triggered using MATLAB® (version R2008b, MathWorks, Natick, MA, USA) in order to detect each ventilation phase of each ventilatory cycle. Gastric tidal volumes
In a previous literature review, we showed a wide heterogeneity in the definition of successful ventilation, in terms of both analysis methods and judgment criteria [ Method Method Method
Although the main judgment criteria described in our review were tidal volume (
Different definitions and judgment criteria identified in the literature and ILCOR guidelines.
Judgment criteria | Definition |
Definition |
Definition |
Definition |
Definition |
---|---|---|---|---|---|
Tidal volume | 450–525 mL |
400–600 mL | — | — | 450–525 mL |
Ventilation rate | — | — | 8–10 bpm |
10–15 bpm | 8–10 bpm |
Finally, considering that none of these approaches is able to provide a clear understanding of ventilation variability [
This algorithm evaluates the performance of a one-minute window depending on two judgment parameters: tidal volume ( 1st situation: 2nd situation: 3rd situation:
In order to explain how to implement this new analysis method, we tried to illustrate its operating process in Figure
Operating process of the new analysis algorithm. This figure shows the evaluation of one-minute sliding windows with a shift of three ventilation cycles. The 1st window is considered insufficient as there are only 5 adequate ventilation cycles. The 2nd window is excessive as mean
The program appraises the global performance of the five-minute ventilation test by considering every sliding window performance and giving the general trend of the whole ventilation sequence. This novel method will be used to evaluate which factors influence manual ventilation performance of our healthcare professionals.
Continuous data are expressed as mean ± SD. Results are presented as percentages for nominal variables. Chi2 and Fisher exact test were used to compare professional categories and experience, BVM type, hand size, and hand grip strengths between the three different performance levels. Odds ratio, estimated by logistic regression, was used to analyse performance level for the multivariate model. A
Forty-five physicians (29 emergency medicine physicians and 16 anaesthesia/critical care physicians), forty-five nurses (27 anaesthesia nurses and 18 emergency medicine nurses), and fifty rescuers (31 firefighters, 17 emergency medical technicians, and 2 Red Cross first-aid rescuers) were enrolled in the study. Their professional experience ranged from less than one year to greater than 20 years. The mean population age was
Characteristics of study population (
|
37.28 ± 8.97 |
|
|
Female | 47 (33.57%) |
Male | 93 (66.43%) |
|
|
Physicians | 45 (32.14%) |
Nurses | 45 (32.14%) |
First-aid workers | 50 (35.71%) |
|
|
High (≥10 years) | 63 (45.00%) |
Medium (5 ≤ |
36 (25.70%) |
Little (<5 years) | 41 (29.30%) |
|
|
Right-handed | 118 (84.29%) |
Left-handed | 14 (10.00%) |
Ambidextrous | 8 (5.71%) |
|
|
Large (≥23 cm) | 21 (15.00%) |
Medium (19 ≤ |
101 (72.14%) |
Small (15 ≤ |
18 (12.86%) |
|
|
Hand squeezing the bag | |
High (≥40 kgF) | 53 (37.90%) |
Medium (20 ≤ |
65 (46.40%) |
Weak (0 ≤ |
22 (15.70%) |
Hand keeping the mask |
|
High (≥40 kgF) | 38 (27.70%) |
Medium (20 ≤ |
68 (49.60%) |
Weak (0 ≤ |
31 (22.60%) |
|
|
Ambu® | 76 (54.29%) |
Laerdal® | 3 (2.14%) |
Both | 39 (27.86%) |
Neither | 22 (15.71%) |
|
|
Good | 55 (39.57%) |
Medium | 83 (59.71%) |
Bad | 1 (0.72%) |
SD: standard deviation;
Two hundred eighty ventilation tests have been recorded as each volunteer performed twice the five-minute ventilation test with two different BVM devices. Three different analysis methods and five definitions of ventilation efficiency were applied to this database to evaluate the manual ventilation performance of healthcare professionals. More than 54,000 ventilation cycles have been analysed. Ventilatory parameter values and manual ventilation performance results are reported in Tables
Ventilation parameter values measured during all the 5-minute ventilation tests (
Variable | Mean (SD) | Lower quartile | Upper quartile |
---|---|---|---|
Instantaneous ventilation rate ( |
24.09 (9.47) | 17.20 | 29.09 |
Tidal volume ( |
333.94 (124.19) | 245.60 | 419.95 |
BVM insufflation volume ( |
590.20 (193.31) | 458.11 | 723.40 |
Gastric tidal volume ( |
37.58 (25.13) | 18.92 | 52.43 |
Lung peak flow ( |
39.99 (16.53) | 28.40 | 50.16 |
BVM peak flow ( |
69.26 (28.07) | 49.16 | 85.92 |
Gastric peak flow ( |
5.35 (4.33) | 2.53 | 7.34 |
Manual ventilation efficiency (
Analysis methods | Definition |
Definition |
Definition |
Definition |
Definition |
---|---|---|---|---|---|
Method 1 (overall mean value analysis) | 0 (0.00%) | 0 (0.00%) | 0 (0.00%) | 0 (0.00%) | 0 (0.00%) |
Method 2 (individual mean value analysis) | 27 (19.29%) | 64 (45.71%) | 0 (0.00%) | 36 (25.71%) | 0 (0.00%) |
Method 3 (breath-by-breath analysis) | 9232 (16.86%) | 21913 (40.01%) | 1883 (3.44%) | 7860 (14.35%) | 222 (0.41%) |
Using the overall mean value analysis (Method
Whatever the definition used, mean values of
Unlike Method
The results show that the use of different BVM models (Laerdal Bag II and Ambu Spur II) does not significantly affect manual ventilation performance (
Our new method showed only 21 (7.50%) efficient ventilation tests while 37 (13.21%) were insufficient and 222 (79.29%) were excessive.
Moreover, statistical analyses have revealed no effect of hand size (
Percentage of excessive, efficient, and insufficient ventilation tests for professional categories (
Univariate analysis has revealed that the grip strength of the hand keeping the mask has an impact on ventilation performance (Figure
Percentage of excessive, efficient, and insufficient ventilation tests for grip strength categories of the hand keeping the mask (
In a previous manuscript, we identified three different analysis methods and five definitions of adequate ventilation used in several reviewed studies [
For these reasons, we defined a new analysis method that allows a chronological evaluation of ventilatory parameters by segmenting a whole ventilation test into one-minute sliding windows. Figure
With our novel method, 280 ventilation tests realized by 140 healthcare professionals were analysed and we found that only 21 (7.50%) were efficient; the remaining 259 tests (92.50%) can be considered potentially deleterious. However, using the overall mean value analysis method, the BVM ventilation is ineffective with 0% of efficient ventilation tests according to different definitions reviewed.
Among the 92.50% of inadequate ventilation tests, 79.29% were excessive which may impair haemodynamics and induce pulmonary barotrauma, and 13.21% of them were insufficient which may cause hypoxia despite the fact that most of the participants thought their ventilation was adequate. This confirms on a larger sample size the results reported in previous studies [
Manikin-based simulations could influence manual ventilation performance as the anatomical design of facemasks is not particularly adapted to the manikin’s face shape [
Furthermore, our study showed that the use of different BVM models has no impact on ventilation performance, although many healthcare professionals perform ventilation more with Ambu bags than with Laerdal ones (115 versus 42). A similar result was obtained by Augustine et al. who demonstrated that there was no obvious relationship between various BVM models used and the average tidal volume delivered [
In order to determine which human factors can affect manual ventilation performance, operator characteristics such as professional categories and experience, hand grip strength, and hand size have been evaluated.
There was no influence of professional experience and hand size on manual ventilation performance. These results were similar to those obtained by Otten et al. [
We proposed a novel analysis method that is more relevant as it enables scientists to observe variability within and between persons, but it still has some limitations. Actually, only the correlation with clinical data, which remain the only real indicators of ventilation performance, will help to validate this new method. Therefore, we still cannot confirm that bag-valve-mask ventilation performance described in this study is a reliable representation of what could be obtained in clinical trials. Moreover, we only conducted this study in three different emergency structures. Even if a low ventilation performance rate has also been reported by other studies, diverse findings could be obtained in other centres where Basic Life Support training courses could differ.
Many factors may have an impact on manual ventilation performance. In order to accurately assess human factors, it is important to use adequate analysis methods to avoid biases related to methodology. The important variability within and between persons proves the relevance of our new analysis method, which allows observing ventilatory parameter variability on an entire ventilation period. We showed, with this novel method, that professional category and grip strength of the hand keeping the mask have a significant impact on manual ventilation performance. This study confirms that the evaluation of bag-mask ventilation performance is complex and cannot be fully determined on a manikin model. Extrapolations to humans have to be taken with caution. However, we can argue that healthcare professionals perform hyperventilation in most cases and have difficulties in performing adequate manual ventilation with current devices. We believe this problem could be prevented by implementing monitoring tools in order to give direct feedback to healthcare professionals regarding ventilation efficiency and ventilatory parameter values.
The authors declare that there are no competing interests regarding the publication of this paper.
The authors would like to gratefully acknowledge the staff of the Departments of Emergency Medicine and of Anaesthesia and Critical Care of the University of Franche-Comté Medical and Trauma Centre. They also thank Colonel René Cellier, Colonel François-Xavier Lagré, and the staff of the Doubs Fire Department, Mr. Alexis Coche, and the Besancon Jussieu Ambulances team who voluntarily participated in this research. They also thank Pr. Alain Neidhardt and Dr. Christophe Lambert of the University of Franche-Comté Medical and Trauma Centre. This work is supported by unrestricted grant from the European Commission (FEDER), the Bpifrance of Besançon, the Greater Besançon Urban Area Community (CAGB), the Regional Council of Franche-Comté, and the General Council of Doubs Department.