Study on Tidal Volume According to the 4-Point Sealing Forces of Bag-Valve-Mask : An Adult Respiratory Arrest Simulator Based Prospective, Descriptive Study

Background For adequate ventilation during bag-valve-mask ventilation, rescuers should ensure a proper mask seal using the one-handed or two-handed technique. Little is known how much sealing forces of bag-valve-mask are needed for adequate ventilation. This study aimed to explored the effect of the 4-point sealing forces of a bag-valve-mask on tidal volume while using the one-handed technique, focusing on the moderating effect of C length ( C length is the distance from the thumb to index nger in the C shape of in the one-hand EC grip). This was a prospective, descriptive simulation study design. Convenience sample of 125 undergraduate paramedic students from two universities was participated. A self-reported questionnaire was used to collect subject data. Tidal volumes, 4-point sealing forces of the mask, peak pressure, and C length of the C shape in the one-hand EC grip, were measured using the mechanical lung model under a simulated adult respiratory arrest. Hierarchical regression analysis was used to determine the moderating effect of C length on tidal volume in bag-valve-mask ventilation. (± and (± respectively. The range of the were The apex sealing force was the weakest point among the 4-point sealing forces. Hierarchical regression analysis for of the by peak and during (F = effect of and and


Abstract Background
For adequate ventilation during bag-valve-mask ventilation, rescuers should ensure a proper mask seal using the one-handed or two-handed technique. Little is known how much sealing forces of bag-valve-mask are needed for adequate ventilation. This study aimed to explored the effect of the 4-point sealing forces of a bag-valve-mask on tidal volume while using the one-handed technique, focusing on the moderating effect of C length ( C length is the distance from the thumb to index nger in the C shape of in the one-hand EC grip).

Methods
This was a prospective, descriptive simulation study design. Convenience sample of 125 undergraduate paramedic students from two universities was participated. A self-reported questionnaire was used to collect subject data. Tidal volumes, 4-point sealing forces of the mask, peak pressure, and C length of the C shape in the one-hand EC grip, were measured using the mechanical lung model under a simulated adult respiratory arrest. Hierarchical regression analysis was used to determine the moderating effect of C length on tidal volume in bag-valve-mask ventilation.

Results
The average C length, peak pressure and tidal volume were 7.54 (± 1.85) cm, 11.62 (± 5.40) cmH2O, and 321.66 (± 135.18) mL, respectively. The average range of the 4-point sealing forces were 0.03-0.69 newton. The apex sealing force was the weakest point among the 4-point sealing forces. Hierarchical regression analysis demonstrated that tidal volume accounted for 62.7% of the variance by C length, peak pressure, and apex sealing force during bag-valve-mask ventilation (F = 9.676, p < .001). C length moderated the effect of apex sealing force and peak pressure on tidal volume. That is, the more peak pressure and apex sealing force, the more the tidal volume, and the longer the C length, the more the tidal volume.

Conclusion
We developed effective advice that can be adopted in clinical practice without side effects as the rst simulation study measuring the 4-point sealing forces during bag-valve-mask ventilation, as well as underpin continuous retraining and assessment that focus on individual physical characteristics, such as C length and bag-valve-mask sealing force.

Background
Bag-valve-mask (BVM) ventilation is standard for cardiopulmonary arrest, as well as emergency airway management in cases where spontaneous breathing is failing or has ceased, regardless of place or time. The ability of rescuers to deliver adequate BVM ventilation is considered relatively simple task, but it is an important and challenging skill that requires considerable practice. The adequate ventilation is to deliver each rescue breathing over 1 second to give a su cient tidal volume, using the BVM, to produce a visible chest rise (1). This is deeply associated with the survival rate as well as oxygenation of patients with cardiac or respiratory arrest (2). Excessive ventilation, however, results in signi cantly increased intrathoracic pressure and decreased coronary perfusion pressures and survival rates (3,4). Rescuers seal the mask against the face using either one-or two-handed techniques, with the thumb and index nger wrapped in a " C "shape around the mask apex, and the remaining ngers in an "E" shape, lifting the jaw (5); this is called the "EC grip". Successful ventilation via the EC grip can be achieved by creating a tight seal between the mask and face and squeezing the bag with reasonable force (6). Several previous studies (7)(8)(9)(10)(11)(12)(13) have reported that the BVM tidal volume delivered is affected by rescuers' squeezing methods, as well as individual characteristics such as gender, height, weight, hand width and length, grip power, education, practice, experience etc.; however, reported results are inconsistent.
The most important aspect of air leaking prevention during BVM ventilation is the perfect seal between the mask and patient's face, generated by the sealing force of the mask toward the patient's face during the EC grip. Few studies have explored the relationship between BVM ventilation and sealing force in South Korea; therefore, the purpose of this study was to identify the factors affecting the tidal volumes with respect to the 4-point sealing forces of BVMs, with a focus on the moderating effects of C length, using a RespiTrainer® Advance manikin and Quick Lung®.

Study design
This was a prospective, descriptive study that simulated adult respiratory arrest using a mechanical lung model.

Participants and setting
To ensure an adequate sample size, we performed a power analysis using G Power 3.1.5 (SOFTMEDIA) program; the desired sample size for a computation of test power (1-β) of 0.80 with 17 predictors was 146, with an effect size of 0.15 and alpha of 0.05. After contacting the directors of two universities from a big city in South Korea to obtain permission for recruitment, a convenience sample of 160 undergraduate paramedic students from two universities was recruited; 35 students dropped out due to personal reasons, such as part time jobs or family events. Finally, 125 questionnaires were used in the analysis; participation was both voluntary and anonymous. The inclusion criterion was right-handed individuals who had completed the cardiopulmonary resuscitation (CPR) theory and practical course in accordance with the 2015 American Heart Association (AHA) guidelines.
Measurements in this study consisted of both subjective and objective variables. Subjective variables were measured using the self-reported questionnaire, which included general characteristics, knowledge, and con dence performing the BVM ventilation technique for respiratory arrest patients; this took approximately 10 minutes to complete. Data collection of objective variables, such as tidal volume (TV), peak pressure (PP), C length, and 4-point sealing forces of the mask, took about 5 minutes per participant.

Subjective variables
General characteristics General characteristics of interest included gender, school year, academic grade, clinical practice satisfaction, major satisfaction, experience in the BVM ventilation of respiratory arrest patients, acquired certi cates, and body mass index (BMI, kg/m 2 ).

Knowledge
Knowledge regarding BVM ventilation was assessed using a previously published scale (14). The questionnaire comprised ten items; a correct response was scored 1. The total score ranged from 0 to 10; high scores indicated more knowledge regarding BVM ventilation. The Kuder-Richardson Formula 20 was 0.69 in a previous study (14), and 0.68 in the present study.

Performance con dence
The scale used to assess performance con dence was developed in a previous study (14). The questionnaire comprises ten items, each rated on a ve-point Likert scale ranging from "strongly disagree" (1), to "strongly agree" (5). High scores indicated a higher level of performance con dence. Cronbach's alpha in the previous study (14) was 0.90, and 0.93 in the present study.

Objective variables
The objective variables in this study included TV, PP, C length, and 4-point sealing forces of the mask. All participants were asked to hold the mask with their left hand, and the bag with their right hand; data collection took about 5 minutes per participant.

Experimental environment TV and PP
A RespiTrainer® Advance manikin (IngMar Medical, Ltd., Pittsburgh, PA, USA) was placed at the same height as the participants' middle femur line (15,16), using the foothold to adjust the height. A Quick Lung® (IngMar Medical) and personal digital assistance were connected to the manikin, and data were sent to the personal digital assistance through its embedded device. The Quick Lung® was set with compliance of 50 mL⋅cmH 2 O − 1 and resistance of 5 cmH 2 O⋅L − 1 ⋅s based on previous studies (17,18). Participants were instructed to use a BVM (1600 mL, Ambu Mark -Reusable Resuscitator with silicon face mask size 5; Ambu, Copenhagen, Denmark) to ventilate the manikin, simulating an adult with respiratory arrest, for two minutes (measured using a metronome (10 times/min)). BVM ventilation was performed while standing without chest compression and oral or nasal airway; i.e., the left hand secures the mask on the manikin's face using the "EC" grip, while the right hand holds the bag comfortably. Before data collection, participants were given the opportunity to practice ventilation for two minutes to familiarize themselves with the study setting. Study participants were blinded for the study purpose. The average TV (mL) and PP (cmH2O) were collected and calculated for each participant.
Apex, bottom, left, and right sealing forces of the mask in C shape of EC grip The 4-point forces of the mask in the EC grip are the apex, bottom, left, and right sealing forces. To measure the sealing forces in the C shape of the EC grip, sensors were attached to four points where the mask touches the manikin's face (Fig. 1). Four 0.5" Circles (Part No. 402) of Interlink Force Sensing Registors® (Interlink Electronics, Inc, Camarillo, California, United States) were installed for force measurement; the force data (N) collected from the sensor were coded and calculated through ARDUINO Software (https://www.arduino.cc/en/main/software).

C length
The C length (cm), de ned as the distance from the thumb to index nger in the C shape of the "EC" grip, was measured using a distance measuring application from AGUMON LAB (20). First, a picture was taken of the EC shaped hand that held the BVM. A reference object was then selected for length measurement (the reference object for this study was a 1,000 won Korean bill; Fig. 2), and the start and end points of the reference object were selected. Finally, the length was measured after the start and end points of the reference object were tted to the thumb and index ngers in the picture.

Statistical analysis
Data was analyzed using IBM SPSS Statistics for Windows, Version 26.0 (SPSS Inc., Chicago, IL, USA). To con rm normal distribution of the data, the kurtosis and skewness were con rmed. Logarithmic and squared transformation was applied on apex sealing force data; those kurtosis and skewness were less than the standard of normality (skewness < 3, kurtosis < 10). Statistical analyses were performed via descriptive statistics, Pearson's correlation analysis, and hierarchical regression analysis. To verify the moderating effects of C length on tidal volumes, general characteristics and subjective variables (knowledge, performance con dence) were adjusted (controlled), and hierarchical regression analysis was performed to con rm the in uence of the predictors. Mean centering was used to remove multicollinearity among predictor variables (PP, apex sealing force, bottom sealing force, right sealing force, and left sealing C length). Additionally, to resolve the problem of multicollinearity, the independent and moderating variables were standardized. Control variables including gender, school year, academic grade, experience with BVM ventilation, and acquired certi cate of general characteristics, were transformed into dummy code; continuous variables such as clinical practice satisfaction, major satisfaction, BMI, knowledge, and con dence performing BVM ventilation were used without transformation.

Ethical considerations and data collection
This study was conducted in accordance with the principles of the Declaration of Helsinki, and was approved by the relevant institutional review board.
Written, informed consent was obtained from all participants prior to conducting this study; participants were informed that they could withdraw their consent at any point during this study without consequences. Data were collected between November 4, 2019 and December 13, 2019.

General characteristics
General characteristics are presented Table 1. Half of participants (53.6 %) were female, while 43.2 % and 8.0 % were juniors and seniors during the school year, respectively; 67.2 % had an academic grade between 3.0 and 3.99. The mean scores regarding clinical practice and major satisfaction were 3.76 (± 0.78), 4.07 (± 0.76), respectively.     The average C length in this study was 7.54 cm, with a longer C length being associated with a greater TV; however, a previous study (22) reported that the OE grip, where the distance from the thumb to the index nger is 0 cm, provided more TV than the EC grip. Traditionally, the EC grip has been recommended for effective BVM ventilation by the AHA guidelines. The C length measured as a variable affecting TV in this study, but that was considerably different from previous studies (7,8,10,11,24) that measured the hand length (distance from the middle edge of the palm to the distal thumb) and width (distance from the middle ngertip to the distal skin crease at the wrist) as TV-related variables. It is di cult to compare the results of these previous studies with those of the current study, as the measurement criteria for hand size differed. Nevertheless, BVM ventilation is traditionally performed by EC grip; in addition to the other variables, C length may be a signi cant determinant of TV and hence, successful bagging.
PP is the highest level of pressure applied to the lungs during inhalation. A previous study (25) reported that PP > 20 cmH2O resulted in adverse effects such as gastric insu ation. In this study, the average PP was 11.62 cmH2O, lower than that from a previous study (13); however, this was higher than the 8.71 cmH2O previously reported (17). PP is directly proportional to TV; low PP was associated with suboptimal TV (321.6 mL) in this study.
The average range of the 4-point sealing forces was 0.03-0.69 N.
There are no previous studies measuring mask sealing force in BVM ventilation; this is the rst simulation study measuring the 4-point sealing forces in BVM ventilation as a variable affecting TV. We believe that the mask sealing force results provide a hypothesis for future studies, as the apex sealing force was the weakest among the 4-point sealing forces in this study. It may be that the upward pressure of the three ngers, forming the letter "E" on the inferior of the mandible to hold the mask to the face, was stronger than the downward pressure of the thumb and fore nger, forming "C" shape in the 1-handed technique, on the mask. In-depth future studies are needed to con rm whether the apex sealing force is 10 times lower than the sealing force of the other three points.
Previous studies (8,11) on the effect of hand grip strength on BVM ventilation were inconsistent. The results of this study suggest more technicalities than those of previous studies. For the development of effective BVM ventilation, continuous research on sealing force in various clinical situations is needed.
The average TV (321.6 mL) in this study was comparable with that of a previous study (24); participants in this study did not achieve the adequate TV, even though they had completed the CPR theory and practical courses in accordance with the AHA guidelines. We are not certain whether this result is related to the ability to perform the procedure, or due to lack of proper education; however, the evaluation of whether students have done well is as important as the education, and the AHA BLS-HP course does not provide feedback regarding BVM ventilation. Although simple, BVM ventilation is an important technique of emergency airway management that requires skill; therefore, ongoing education and training is necessary to enhance the effective sealing technique, creating a tighter seal between the mask and face, of paramedic students who will work in the prehospital stage.

Limitations
First, a manikin model was used instead of humans; although the manikin model allowed us to test our study in a highly controlled and consistent environment, it is unclear whether the same results would be achieved in patients. Second, the results of this study may not extend to all cases of BVM ventilation, as we only selected participants through convenience sampling. Additionally, this study did not use nasal or oral airways and did not consider the patient's facial characteristics.

Conclusions
As the rst simulation study measuring the 4-point sealing forces in BVM ventilation, we believe that the results of this study suggest a new criteria for effective BVM ventilation in clinical practice and future research. The continuous retraining and assessment that focus on individual physical characteristics, such as C length and bag-valve-mask sealing force are also needed.

List Of Abbreviations
American Heart Association, AHA

Declarations
Ethics approval and consent to participate: All participating students received written and verbal information about the purpose and process of the study before launching the study. Participation was voluntary. All participating students provided written consent. This study was approved by the institutional review board of Daejeon University (104067-201611-HR-008-02).

Consent for publication:
Not applicable. (All data was recorded anonymously and does not allow inference to the participating students.)