Abstract
Objective:
To characterize the quality of bag-mask ventilation (BMV) before tracheal intubation in children in the PICU, and to evaluate the association between poor BMV quality and adverse airway outcomes.
Design:
Single center, pilot observational study, 2019-2022.
Setting:
Large, urban quaternary care PICU.
Patients:
Pediatric patients requiring BMV before tracheal intubation.
Interventions:
None.
Measurements and Main Results:
Using a respiratory function monitor, we collected flow and pressure data from 8,446 BMV breaths prior to tracheal intubation in 85 children in the PICU (median age 3.3 [interquartile range 1.4, 8.3] years). Adverse airway outcomes (i.e., tracheal intubation-associated event and/or pulse oximetry desaturation <80%) occurred in 14/85 (16.5%) patients. Low-quality BMV breaths were defined as: 1) inadequate or excessive exhaled tidal volume (VTe <4 or >12mL/kg); 2) excessive peak inspiratory pressure (PIP) and excessive VTe; 3) excessive facemask leak (>40%); or 4) failure to relieve upper airway obstruction. Overall, 78.0% of BMV breaths met at least one low-quality criterion; most frequently inadequate or excessive VTe (55.5%), followed by excessive leak (46.2%). Infants (<1 year) and young children (1-7 years), compared to older children (8-17 years), had a higher proportion of low-quality BMV breaths overall (86.0%, 85.5% vs. 57.9%, p<0.001 for both), with inadequate or excessive VTe (57.7%, 61.1% vs. 43.7%, p<0.001 for both), excessive leak (50.6%, 49.2% vs. 37.0%, p<0.001 for both), and excessive PIP with excessive VTe (17.5%, 19.4% vs. 6.4%, p<0.001). After controlling for respiratory pathology, low-quality BMV was associated with 2.8-times greater odds of adverse airway outcome (adjusted odds ratio 2.8 [95% confidence interval 1.2, 6.2], p=0.01).
Conclusions:
The majority of BMV breaths delivered to children before tracheal intubation in the PICU were of low-quality. And, such breaths, were more frequent in younger children and were associated with greater odds of adverse airway outcomes.
Keywords: mask ventilation, tracheal intubation, hypoxemia, adverse events, pediatric critical care, ICU
Airway management is a fundamental skill for pediatric intensive care unit (PICU) providers. Bag-mask ventilation (BMV) is performed in almost all intubations in children admitted to the PICU. However, objective BMV quality is rarely measured, especially in the clinical environment. Components of poor-quality BMV, such as inadequate or excessive pressures or volumes, excessive leak, and upper airway obstruction, can lead to adverse mechanical effects (e.g., atelectasis, hyperinflation, pneumothorax, gastric insufflation) that may lead to adverse physiologic effects (e.g., oxygen desaturation, hypercarbia, acidosis and aspiration). In turn, these physiologic effects limit laryngoscopy time (e.g., desaturation requiring aborting the attempt at intubation) and further compromise BMV quality. These physiologic effects may also result in adverse short-term effects (e.g., severe oxygen desaturation and tracheal intubation-associated events [TIAEs]), which may lead to poor long-term outcomes (e.g., longer mechanical ventilation, longer PICU stay, and risk of mortality) (1). Historically, the clinical tools used to evaluate the quality and proficiency of BMV (e.g., visual chest rise and/or end-tidal carbon dioxide [ETCO2] level) were not reliable and did not differentiate between causes of low-quality BMV (2-5).
Therefore, in 2019-2022 we carried out a pilot observational study with the objective of characterizing the quality of BMV prior to tracheal intubation in critically ill children, and to evaluate the association of potentially modifiable quality of BMV with clinical outcomes. Our hypothesis was that low-quality BMV would be associated with adverse airway outcomes, and that low-quality BMV characteristics would differ between infants, young children, and older children.
METHODS
This single-center pilot observational study called “Evaluation of Bag Mask Ventilation in Pediatric Patients during Tracheal Intubation using a Respiratory Function Monitor” was approved on August 28, 2019, by the Children’s Hospital of Philadelphia’s Institutional Review Board (IRB 18-015880). A parent or guardian provided consent for their child’s participation. For non-urgent intubations, consent was obtained prior to data collection. However, given the non-invasive nature of the data collection, when pre-intubation consent was not feasible (e.g., urgent or emergent intubations), investigators were permitted to collect data during the intubation and request consent after data collection. All clinical research procedures were fully compliant with our local IRB ethical standards, and in accordance with the Helsinki Declaration of 1975.
The BMV quality data were collected from a convenience sample of critically ill children requiring BMV prior to tracheal intubation in our large, urban quaternary care children’s hospital between December 2019 and July 2022. Any child aged <18 years undergoing primary tracheal intubation in the PICU was eligible. We excluded patients with a tracheostomy or those undergoing tracheal tube exchange.
Respiratory Function Monitor Measurements
We used a respiratory function monitor (RFM; Nihon Kohden Inc., Tokyo, Japan) to collect breath-to-breath BMV quality data (6). The RFM utilizes a chamber with a flow and pressure sensor, which was placed between the face mask and flow-inflating bag, and continuously recorded flow (mL/s) and pressure (cmH2O) with a sampling frequency of one data point every 16 milliseconds. The RFM chamber adds minimal deadspace and resistance to the system (6). As is a standard setup in our unit, the flow-inflating bag included a positive end-expiratory pressure (PEEP) valve. Respiratory therapists were trained in operation of the RFM device and clinicians were blinded to the RFM data. Each BMV breath was detected using an algorithm with the following criteria: an inspiratory tidal volume (VTi) of ≥10 mL and <2,000 mL and a driving pressure (i.e., peak inspiratory pressure [PIP] minus PEEP) of ≥5 cmH2O for children <5 years or ≥10 cmH2O for children ≥5 years. For each detected breath, the VTi and expiratory tidal volume (VTe) were calculated using the flow and time data. Mask leak (proportion of the breath leaked between the mask and the face) was calculated as (VTi-VTe)/VTi.
Definitions and Outcome Measures
Low-quality BMV breaths were defined a priori as meeting at least one of the following four criteria: 1) inappropriate VTe, meaning inadequate (<4mL/kg of ideal body weight [IBW]) or excessive (>12 mL/kg) VTe (7-10); 2) excessive PIP, meaning PIP >25 cmH2O with VTe >12 mL/kg of IBW; 3) excessive leak, which was >40% (9, 10); and 4) upper airway obstruction, with VTe <4 mL/kg, PIP >30 cmH2O and leak >80%. The overall low-quality of BMV was defined a priori as ≥70% of an individual patient’s BMV breaths being low-quality.
The primary outcome was adverse airway outcome, defined as a TIAE using the standard NEAR4KIDS (National Emergency Airway Registry for Children) definitions (see Supplemental Digital Content [SDC] Table S1) (11, 12) and/or pulse oximetry oxygen desaturation during intubation (SpO2 <80% when SpO2 was >90% after preoxygenation).
Patient characteristics, including indication for intubation and airway provider role were abstracted from our local NEAR4KIDS database (12). Respiratory pathology (binary categorization defined as respiratory disease or anatomy that would affect pulmonary resistance or compliance) was determined based on review of respiratory diagnoses in the NEAR4KIDS database and agreed upon by two PICU experts (LNS, AN). Age was categorized as infant (<1 year), young child (1-7 years), or older child (8-17 years). Weights and heights were measured by bedside nurses. Ideal body weight (IBW) was determined using the median body mass index (Centers for Disease Control and Prevention) or weight-for-length (World Health Organization) for age and gender and height/length. (The NEAR4KIDS methodology has been described in a 2024 report (13))
Statistical Analysis
In considering project sample size, we estimated that at least 30% (95% confidence interval [CI] 25%, 35%) of BMV breaths would meet low-quality criteria, that ≥100 breaths would be captured per patient, and that BMV quality characteristics would be clustered within each subject with an intraclass coefficient of 0.25. Therefore, we planned to enroll 84 patients.
The patient, provider, and BMV quality characteristics are described as frequencies and percentages or medians and interquartile ranges (IQR). We analyzed categorical data using Fisher’s exact test and continuous data using Wilcoxon rank-sum test. Pearson’s chi-squared test was used to assess whether the proportion of low-quality BMV breaths differed by age group, provider type, and respiratory pathology, with Bonferroni adjustment for post-hoc pairwise comparisons. Logistic regression was used to evaluate the association between low-quality BMV breaths and adverse airway outcome, controlling for respiratory pathology, with generalized estimating equations (GEE) to account for clustering by patient. Given the number of patients in the study, and therefore the limited number of outcomes, we only controlled for respiratory pathology to avoid overfitting the model. Data are reported as odds ratio (OR), or adjusted OR (aOR), with 95%CI.
We performed a sensitivity analysis using actual body weight, rather than IBW. We performed an additional post-hoc sensitivity analysis using specific adverse airway outcomes including cardiac arrest, emesis (with or without aspiration), hypotension, pneumothorax or oxygen desaturation. All analyses were performed using Stata 17/18 BE (StataCorp LLC, College Station, TX).
RESULTS
Over the period 2019-2022, we collected data from 85 children in the PICU, with 8,446 BMV breaths prior to tracheal intubation. There was a median of 82 [IQR 61, 111] captured BMV breaths per patient. Patient and provider characteristics are shown in Table 1. The median age was 3.3 [IQR 1.4, 8.3] years; with 20/85 (23.5%) infants, 43/85 (49.4%) young children, and 23/85 (27.1%) older children. Respiratory pathology was present in 30/85 (35.3%), and 10/85 (11.8%) had a known history of difficult airway. The indication for intubation included procedural in 50/85 (58.8%), ventilation failure in 24/85 (28.2%), oxygenation failure in 14/85 (16.5%), absent airway protective reflexes in 8/85 (9.4%), and unstable hemodynamics in 7/85 (8.2%). Prior to BMV, 55/85 (64.7%) children were in room air (i.e., no supplemental oxygen or respiratory support), with 23/85 (27.1%) requiring non-invasive bi-level positive airway pressure and 3/85 (3.5%) requiring continuous positive airway pressure.
Table 1.
Patient and provider characteristics.
| Patient and Provider Characteristics | Overall (N=85) | Adverse Airway Outcome | ||
|---|---|---|---|---|
| Yes (N=14) | No (N=71) | |||
| Age (years) | 3.3 [1.4, 8.3] | 1.6 [0.5, 7.8] | 4.0 [1.6, 8.4] | |
| Age group | Infant (<1 year) | 20 (27.1%) | 6/14 | 14 (19.7%) |
| Young Child (1-7 years) | 42 (49.4%) | 5/14 | 37 (52.1%) | |
| Older Child (8-17 years) | 23 (27.1%) | 3/14 | 20 (27.2%) | |
| Female sex | 41 (48.2%) | 8/14 | 33 (46.5%) | |
| Actual body weight (kg) | 15.9 [9.4, 25.8] | 12.4 [7.8, 17.9] | 16.5 [10.5, 26.5] | |
| Ideal body weight (kg) | 14.2 [8.9, 24.2] | 9.6 [7.6, 17.3] | 15.3 [9.4, 24.5] | |
| Difficult airway features | Known history of difficult airway | 10 (11.8%) | 4/14 | 6 (8.5%) |
| Limited neck extension | 7 (8.2%) | 3/14 | 4 (5.6%) | |
| Limited mouth opening | 5 (5.9%) | 2/14 | 3 (4.2%) | |
| Short thyromental space | 2 (2.4%) | 0 | 2 (2.8%) | |
| Upper airway obstruction | 3 (3.5%) | 0 | 3 (4.2%) | |
| Midface hypoplasia | 1 (1.2%) | 0 | 1 (1.2%) | |
| Respiratory pathology1 | 30 (35.3%) | 12/14 | 18 (25.4%) | |
| Night or weekend | 20 (23.5%) | 3/14 | 17 (23.9%) | |
| All indications for intubation | Procedural | 50 (58.8%) | 3/14 | 47 (66.2%) |
| Ventilation failure | 24 (28.2%) | 11/14 | 13 (18.3%) | |
| Oxygenation failure | 14 (16.5%) | 6/14 | 8 (11.3%) | |
| Absent airway protective reflexes | 8 (9.4%) | 1/14 | 7 (9.9%) | |
| Unstable hemodynamics | 7 (8.2%) | 1/14 | 6 (8.5%) | |
| Upper airway obstruction | 5 (5.9%) | 0 | 5 (7.0%) | |
| Neuromuscular weakness | 4 (4.7%) | 1/14 | 3 (4.2%) | |
| Airway clearance | 2 (2.4%) | 0 | 2 (2.8%) | |
| Level of respiratory support prior to bag-mask ventilation | Room Air | 55 (64.7%) | 4/14 | 51 (71.8%) |
| Simple Nasal Cannula | 2 (2.4%) | 0 | 2 (2.8%) | |
| Humidified High Flow Nasal Cannula | 2 (2.4%) | 0 | 2 (2.8%) | |
| CPAP | 3 (3.5%) | 2/14 | 1 (1.4%) | |
| BiPAP | 23 (27.1%) | 8/14 | 15 (21.1%) | |
| Airway provider: mask2 | Fellow | 51 (60.0%) | 8/14 | 43 (60.6%) |
| Nurse Practitioner | 17 (20.0%) | 2/14 | 15 (21.1%) | |
| Hospitalist | 8 (9.4%) | 1/14 | 7 (9.9%) | |
| Resident | 6 (7.1%) | 1/14 | 5 (7.0%) | |
| Attending | 3 (3.5%) | 2/14 | 1 (1.4%) | |
| Airway provider: bag2 | Respiratory Therapist | 78 (91.8%) | 14/14 | 64 (90.1%) |
| Fellow | 6 (7.1%) | 0 | 6 (8.5%) | |
| Resident | 1 (1.2%) | 0 | 1 (1.4%) | |
CPAP = Continuous Positive Airway Pressure; BiPAP = Bi-level Positive Airway Pressure
Respiratory tract disease or anatomy that would affect pulmonary resistance or compliance
At our institution, typical practice involves one airway provider holding the mask and a separate airway provider using the bag to provide ventilation. Although airway providers may change during the course of an intubation, here we report the first airway provider for each task.
Airway Processes and Outcomes
The most frequently used induction agents included fentanyl in 58/85 (68.2%), ketamine in 55/85 (64.7%), and propofol in 38/85 (41.2%) patients. Neuromuscular blockade was used in all 85 events. Apneic oxygenation was used in 81/85 (95.3%) and airway adjuncts (e.g., oral airways) in 15/85 (17.7%) children. Intubation was successful on the first laryngoscopy attempt in 68/85 (80.0%) children, with 10/85 (11.8%) requiring two attempts, and the other 7/85 (8.2%) requiring ≥3 attempts. The median apneic time for the first laryngoscopy attempt was 58.4 [IQR 47.3, 92.2] seconds. Adverse airway outcomes occurred in 14/85 (16.5%) patients (6 with SpO2 desaturation, 4 with TIAE, and 4 with combined SpO2 desaturation and TIAE).
Quality of Mask Ventilation
We collected a total of 8,446 BMV breaths. Overall, 78.0% (6,584/8,446) of these breaths met ≥1 low-quality criterion (Table 2). Inappropriate VTe was the most frequent criterion of low-quality BMV (4,690/8,446, 55.5%), followed by excessive leak (3,898/8,446, 46.2%). Overall event-level low-quality BMV (i.e., at least 70% of captured BMV breaths meeting ≥1 low-quality criteria) occurred in 57/85 (67.1%) patients.
Table 2.
Criteria for low-quality bag-mask ventilation by adverse airway outcome.
| Adverse Airway Outcome | ||||
|---|---|---|---|---|
| Low-Quality Bag-Mask Ventilation Criteria |
Overall (N=8,446 BMV breaths from 85 patients) |
Yes (N=2,256 BMV breaths from 14 patients) |
No (N=6,190 BMV breaths from 71 patients) |
p-value |
| Low quality (any criteria) | 6,584 (78.0%) | 1,947 (86.3%) | 4, 637 (74.9%) | <0.001 |
| Inappropriate VTe1 | 4,690 (55.5%) | 1,331 (59.0%) | 3,359 (54.3%) | <0.001 |
| Inadequate VTe | 1,926 (22.8%) | 554 (24.6%) | 1,372 (22.2%) | 0.022 |
| Excessive VTe | 2,764 (32.7%) | 777 (34.4%) | 1,987 (32.1%) | 0.044 |
| Excessive leak2 | 3,898 (46.2%) | 1,331 (59.0%) | 2,567 (41.5%) | <0.001 |
| Excessive PIP3 | 1,302 (15.4%) | 491 (21.8%) | 811 (13.1%) | <0.001 |
| Failure to relieve upper airway obstruction4 | 269 (3.2%) | 71 (3.1%) | 198 (3.2%) | 0.94 |
VTe = exhaled tidal volume; PIP = peak inspiratory pressure
Inappropriate VTe: inadequate VTe (<4 mL/kg of ideal body weight [IBW]) or excessive VTe (>12 mL/kg of IBW)
Excessive PIP: PIP >25 cm H2O with VTe >12 mL/kg of IBW
Excessive leak: >40%
Upper airway obstruction: VTe <4 mL/kg, PIP >30 cm H2O and leak >80%
On comparing the 14 patients with adverse airway outcomes with the 71 patients without such outcomes, there was a higher proportion of breaths that were low-quality by any criteria in those with adverse airway outcomes (86.3% vs 74.9%, p<0.001). Patients with adverse airway outcomes also had a higher proportion of breaths that were low-quality by inappropriate VTe (59.0% vs. 54.3%, p<0.001), excessive leak (59.0% vs. 41.5%, p<0.001), and excessive PIP criteria (21.8% vs. 13.1%, p<0.001).
Patient and Provider Factors Associated with Quality of BMV
Figure 1 and SDC Table S2 show the criteria for low-quality BMV by age. Compared to older children, the infant and young child groups had higher proportions of low-quality BMV breaths (57.9% vs. 86.0% and 85.5%, respectively, p<0.001 for both). The older child group, compared to the infant and young child groups had lower proportions of low-quality BMV breaths with inappropriate VTe (43.7% vs. 57.7% and 61.1%, respectively, p<0.001 for both), excessive leak (37.0% vs. 50.6%, and 49.2%, respectively, p<0.001 for both), and excessive PIP (6.4% vs. 17.5% and 19.4%, respectively, p<0.001). Figure 2 shows the proportion of breaths meeting excessive, appropriate, and inadequate VTe criteria by age. Among BMV breaths that were inappropriate VTe, the infant and young child groups had a higher proportion of low-quality VTe breaths due to excessive VTe compared to the older child group (26.5% and 30.6% vs. 16.8%, p<0.001 for both).
Figure 1. Criteria for low-quality bag-mask ventilation by age category.
Bar graph showing the percentage of bag-mask ventilation (BMV) breaths meeting criteria for low-quality BMV across different age groups: infants (<1 year), young children (1-7 years) and older children (8-17 years). Low-quality criteria included: 1) excessive (>12 mL/kg) or inadequate (<4 mL/kg) exhaled tidal volume (VTe); 2) excessive leak (>40%); 3) excessive peak inspiratory pressure (PIP) (>25 cmH2O) with excessive VTe (>12 mL/kg); or 4) upper airway obstruction. Infants and young children more frequently had low-quality BMV by any criteria, and by VTe, excessive leak, and excessive PIP criteria compared to older children.
Figure 2. Proportion of breaths meeting VTe criteria by age category.
Bar graph showing the proportion of bag-mask ventilation (BMV) breaths meeting exhaled tidal volume (VTe) criteria by age group: infants (<1 year), young children (1-7 years) and older children (8-17 years). Inadequate VTe was defined as <4 mL/kg, appropriate VTe was defined as 4-12 mL/kg and excessive VTe was defined as >12 mL/kg. The infant and young child groups had a higher proportion of BMV breaths with excessive VTe and a lower proportion of breaths with inadequate VTe compared to the older child group.
Figure S1 and Table S3 (see SDC) show the criteria for low-quality BMV by provider holding the mask. Fellows and attendings had a lower proportion of BMV breaths that were low-quality overall (76.3% vs. 81.4%, p<0.001) compared to other providers (i.e., resident, nurse practitioner, hospitalist). Fellows and attendings also had a lower proportion of BMV breaths that were low quality due to excessive VTe (31.4% vs. 35.5%, p<0.001) and excessive leak (44.3% vs. 50.0%, p<0.001) compared to other providers. The provider giving manual ventilations via the bag was a respiratory therapist in most events (78/85, 91.8%).
Figure S2 and Table S4 (see SDC) show the criteria for low-quality BMV by patients with respiratory pathology (i.e., respiratory disease or anatomy that would affect pulmonary resistance or compliance). There was no difference in the proportion of low-quality BMV breaths in patients with respiratory pathology compared to those without (77.1% vs. 78.5%, p=0.13). Patients with respiratory pathology more frequently received BMV with excessive VTe (40.0% vs. 27.6%, p<0.001) and excessive PIP (23.7% vs. 9.5%, p<0.001) and less frequently received BMV with excessive leak (41.1% vs. 49.7%, p<0.001) compared to those without respiratory pathology.
BMV Quality with Airway Outcomes
In unadjusted analysis, we failed to identify an association between low-quality BMV by any criteria and greater odds of adverse airway outcome (OR 2.1 [95%CI 1.0, 4.5], p=0.05). After controlling for respiratory pathology, low-quality BMV by any criteria was associated with 2.8-fold greater odds of adverse airway outcome (aOR 2.8 [95%CI 1.2, 6.2], p=0.01).
In a sensitivity analysis using actual body weight instead of IBW, 79.8% (6,737/8,446) of BMV breaths met at least one low-quality criteria. After controlling for respiratory pathology, low-quality BMV by any criteria using actual body weight was associated with 3.2-fold greater odds of adverse airway outcome (aOR 3.2 [95%CI 1.5, 6.6], p<0.001). Specific adverse airway outcomes (i.e., cardiac arrest, emesis [with or without aspiration], hypotension, pneumothorax or SpO2 desaturation) occurred in 12/85 (14.1%) patients. After controlling for respiratory pathology, low-quality BMV by any criteria was associated with 3.1-fold greater odds of specific adverse airway outcome (aOR 3.1 [95%CI 1.4, 6.8], p=0.006).
DISCUSSION
In this pilot study carried out in our PICU (2019-2022) we used RFM to evaluate the quality of BMV prior to intubation in children and found that low-quality BMV was common (>75% of BMV breaths) and was associated with greater odds of adverse airway outcomes. The etiology of low-quality BMV differed across age groups. Our description of BMV quality in the PICU demonstrates feasibility of collecting such data, and identifies potential targets to remediate low-quality BMV in the future.
In our population, we found that overall low-quality BMV (≥70% of an individual’s captured BMV breaths meeting ≥1 low-quality criteria) occurred in two-thirds (67.1%) of BMV episodes prior to tracheal intubation events. This is especially notable when the most common indication for intubation was procedural, and most patients were not receiving any respiratory support or supplemental oxygen prior to intubation. This novel quantitative data shows that problem is substantially more prevalent than the 9.5% self-reported rate of BMV difficulty previously described in a multicenter 2013-2018 NEAR4KIDS study in the PICU (14). There are multiple possible explanations for this discrepancy. First, it is likely that low-quality BMV is underrecognized and self-report is therefore lower. At the time of the 2013-2018 NEAR4KIDS study, low-quality BMV lacked a standard definition. Hence for our 2019-2022 study we pre-specified our criteria. Now, the 2022 American Society of Anesthesiologists practice guidelines for the management of the difficult airway defines difficult facemask ventilation as that for which “it is not possible to provide adequate ventilation (e.g., confirmed by ETCO2 detection), because of one or more of the following problems: inadequate mask seal, excessive gas leak, or excessive resistance to the ingress or egress of gas” (15). The 2022 definition was an update to the 2013 guidelines, which required an anesthesiologist to be unable to provide adequate ventilation (16). Notably, the 2013 and 2022 definitions are subjective, rely on the skillset of the provider, and their ability to recognize inadequate ventilation. Our a priori definition of low-quality BMV also includes identification of overventilation (i.e., excessive VTe and excessive PIP) because hyperventilation can also be harmful (17). Interestingly, since 2021, simulation studies have also found that providers frequently hyperventilate mannequins (18-20). Taken together, we conclude that non-invasive clinical measures of BMV (i.e., chest rise and ETCO2) cannot reliably identify overventilation during BMV, and thus it is likely underrecognized by clinicians.
We also found in our dataset that low-quality BMV was not only frequent, but it was associated with greater odds of adverse airway outcomes. Tracheal intubation results in hypoxia and other life-threatening complications 10-20% of the time (12, 21-24). The physiologic instability (e.g., SpO2 desaturation, hypercarbia, acidosis, and aspiration) resulting from the poor mechanical effects of low-quality BMV (e.g., atelectasis, hyperinflation, pneumothorax, gastric insufflation) may at least in part explain this relationship. Thus, our finding of an extremely high frequency of low-quality BMV has important clinical relevance. For example, our study shows that the etiology of low-quality BMV differs by age groups. Contrary to historical (2013-2018) self-reported difficulty with BMV, which was more commonly reported with increasing patient age (14), we found that the infant and young child groups had a higher proportion of low-quality BMV breaths compared to the older child group. Given the known association between younger age and increased frequency of TIAE and desaturation events (1), our data further highlights the need for using more objective measures of BMV quality. Of note, we found that the older child group, compared to the infant and young child groups, had a higher proportion of low-quality breaths with excessive VTe, excessive leak, and excessive PIP. Here, the ability to identify these discrete types of low-quality BMV, which require different modification techniques, far surpasses what is currently achievable when using only some observed visible chest rise and change in ETCO2 monitoring.
Identifying that low-quality BMV occurs frequently, and the etiology differs across pediatric age groups may lead to more personalized training and real-time adjustment of BMV technique. In neonates, RFM has been used in the delivery room to guide resuscitation. In a single center feasibility study carried out 2008-2010, use of RFM in the delivery room was associated with more adjustments to facemask position and airway pressure, less mask leak, and a lower rate of excessive VTe (10). However, in a 7-center randomized trial of preterm infants carried out 2013-2019, use of RFM did not result in an increased proportion of ventilations in a predefined target range (25). In fact, in the 4-months after this 2013-2019 trial ended, the trialists carried out a survey of providers’ opinions on the use of an RFM during neonatal resuscitation and found that they considered the RFM helpful, but additional training on real-time use was necessary (26). Taken together, we consider that further work is required to better understand the practicalities of implementing RFM at the bedside. We also need to refine the exact criteria of low-quality BMV that should be used in critically ill children.
This study has a number of important limitations. First, we recruited a convenience sample of patients, which may lead to selection bias. Second, our cohort may have been overrepresented by patients requiring procedural intubation without significant lung pathology. However, it is notable that despite this feature, the majority of pre-intubation events had overall poor-quality BMV. Additionally, these findings may reflect clinical practice at our center and may not be generalizable. Third, our typical practice is two-person BMV technique, with one airway provider holding the mask and a separate airway provider providing ventilation with the bag. Although two-person BMV may allow for better mask seal and lower leak (27, 28), this technique may lead to decreased tactile response in providers managing the airway, and thus increased incidence of low-quality BMV. Verification of our findings warrants further investigation in a multicenter study with variable airway management practices. Fourth, nearly half of breaths delivered met the excessive leak criterion. Leak has been shown to result in an overestimation of the delivered tidal volume in simulated neonatal positive-pressure ventilation with a set pressure or inspiratory tidal volume (29). However, our study used the measured expiratory flow to calculate VTe, which eliminates the effect of leak on overestimating delivered tidal volume. Fifth, there were few captured breaths that met low-quality by the upper airway obstruction criterion. It is possible that some attempted BMV breaths with upper airway obstruction may not have been detected. Sixth, all patients in our cohort received neuromuscular blockade, but the timing of drug administration relative to captured BMV breaths was not determined, thus the association between neuromuscular blockade and quality of BMV in our study is unknown. Seventh, our work was a pilot study, and due to the number of patients enrolled, and therefore number of outcomes, we were limited in the number of confounders we could control for in our logistic regression model. Finally, it is plausible that we have the causality of the association reversed, and that patients with sicker lungs and worse lung compliance may be more difficult to provide BMV and thus be more likely to have adverse events. Our definition of low-quality BMV incorporated poor lung compliance into the excessive PIP criterion. However, it is possible that patients with poor lung compliance may have more leak, meeting the excessive leak criterion. Unfortunately, although we attempted to address this by adjusting for respiratory pathology in our multivariable analysis, this question is not answerable in an observational study.
In conclusion, in our 2019-2022 prospective observational study of PICU children requiring tracheal intubation we found that the majority of BMV breaths delivered immediately prior to tracheal intubation met at least one low-quality BMV criteria. Identified low-quality BMV criteria (i.e., inappropriate VTe, excessive leak, excessive PIP and upper airway obstruction) differed across pediatric age groups. Low-quality BMV was associated with adverse airway outcomes. Further investigation in a multicenter study is warranted to evaluate discrepancies between quantitative BMV quality assessment with provider self-assessment, and to determine whether personalized training and real-time adjustment of BMV technique can improve the quality of BMV breaths and patient outcomes.
Supplementary Material
Research In Context
Bag-mask ventilation (BMV) is commonly performed prior to intubation in the PICU and is a critical skill for pediatric intensivists
Self-reported BMV difficulty has been associated with adverse airway outcomes (i.e., tracheal intubation-associated events and/or severe desaturation on pulse oximetry), however objective BMV quality is rarely measured
We aimed to characterize the quality of BMV prior to tracheal intubation in children admitted to the PICU, and to evaluate the association of BMV quality with clinical outcomes
At The Bedside
Among 85 children and ~8,500 BMV breaths delivered prior to intubation, low-quality bag-mask ventilation was common (78% of all BMV breaths)
Low-quality BMV was associated with greater odds of adverse airway outcomes (i.e., tracheal intubation-associated events and/or severe desaturation on pulse oximetry)
Identification of causes of low-quality bag-mask ventilation is feasible, and could lead to more personalized training and real-time adjustment of BMV technique
Financial support:
This project was supported by the Agency for Healthcare Research and Quality (R03HS026939).
Dr. Lindsay Shepard’s participation in this project was supported by the Pediatric Hospital Epidemiology and Outcomes Research Training (PHEOT) Program, an NICHD-funded postdoctoral fellowship (T32 HD060550).
Nihon Kohden Inc. provided the research device for data collection and technological support.
The contents of this manuscript are solely the responsibility of the authors and do not necessarily represent the official views of the AHRQ, NICHD or NIH.
References:
- 1.Parker MM, Nuthall G, Brown C 3rd, Biagas K, et al. : Relationship Between Adverse Tracheal Intubation Associated Events and PICU Outcomes. Pediatr Crit Care Med 2017; 18(4):310–318. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Pearlman SA, Zern SC, Blackson T, Ciarlo JA, et al. : Use of neonatal simulation models to assess competency in bag-mask ventilation. J Perinatol 2016; 36(3):242–246 [DOI] [PubMed] [Google Scholar]
- 3.Poulton DA, Schmolzer GM, Morley CJ, Davis PG: Assessment of chest rise during mask ventilation of preterm infants in the delivery room. Resuscitation 2011; 82:175–179 [DOI] [PubMed] [Google Scholar]
- 4.Nassar BS, Schmidt GA: Capnography During Critical Illness. Chest 2016; 149(2):576–585. [DOI] [PubMed] [Google Scholar]
- 5.Jabre P, Jacob L, Auger H, Jaulin C, et al. : Capnography monitoring in nonintubated patients with respiratory distress. Am J Emerg Med 2009; 27:1056–1059. [DOI] [PubMed] [Google Scholar]
- 6.Takatori F, Inoue S, Togo S, Yamamori S: Development of respiratory function monitor for neonates. Annu Int Conf IEEE Eng Med Biol Soc 2017; 2017:3719–3722. [DOI] [PubMed] [Google Scholar]
- 7.Wood FE, Morley CJ, Dawson JA, Kamlin CO, et al. : Assessing the effectiveness of two round neonatal resuscitation masks: study 1. Arch Dis Child Fetal Neonatal Ed 2008; 93:F235–237. [DOI] [PubMed] [Google Scholar]
- 8.van Vonderen JJ, Kamlin CO, Dawson JA, Walther FJ, et al. : Mask versus Nasal Tube for Stabilization of Preterm Infants at Birth: Respiratory Function Measurements. J Pediatr 2015; 167:81–85 e81. [DOI] [PubMed] [Google Scholar]
- 9.Schmolzer GM, Dawson JA, Kamlin CO, O'Donnell CP, et al. : Airway obstruction and gas leak during mask ventilation of preterm infants in the delivery room. Arch Dis Child Fetal Neonatal Ed 2011; 96:F254–257. [DOI] [PubMed] [Google Scholar]
- 10.Schmolzer GM, Morley CJ, Wong C, Dawson JA, et al. : Respiratory function monitor guidance of mask ventilation in the delivery room: a feasibility study. J Pediatr 2012; 160:377–381 e372. [DOI] [PubMed] [Google Scholar]
- 11.Nishisaki A, Ferry S, Colborn S, DeFalco C, et al. : Characterization of tracheal intubation process of care and safety outcomes in a tertiary pediatric intensive care unit. Pediatr Crit Care Med 2012; 13:e5–10. [DOI] [PubMed] [Google Scholar]
- 12.Nishisaki A, Turner DA, Brown CA 3rd, Walls RM, et al. : A National Emergency Airway Registry for children: landscape of tracheal intubation in 15 PICUs. Crit Care Med 2013; 41:874–885. [DOI] [PubMed] [Google Scholar]
- 13.Gladen KM, Tellez D, Napolitano N, Edwards LR, et al. : Adverse Tracheal Intubation Events in Critically Ill Underweight and Obese Children: Retrospective Study of the National Emergency Airway for Children Registry (2013-2020). Pediatr Crit Care Med 2024; 25:147–158. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Daigle CH, Fiadjoe JE, Laverriere EK, Bruins BB, et al. : Difficult Bag-Mask Ventilation in Critically Ill Children Is Independently Associated With Adverse Events. Crit Care Med 2020; 48:e744–e752. [DOI] [PubMed] [Google Scholar]
- 15.Apfelbaum JL, Hagberg CA, Connis RT, Abdelmalak BB, et al. : 2022 American Society of Anesthesiologists Practice Guidelines for Management of the Difficult Airway. Anesthesiology 2022; 136:31–81. [DOI] [PubMed] [Google Scholar]
- 16.Apfelbaum JL, Hagberg CA, Caplan RA, Blitt CD, et al. : Practice guidelines for management of the difficult airway: an updated report by the American Society of Anesthesiologists Task Force on Management of the Difficult Airway. Anesthesiology 2013; 118:251–270. [DOI] [PubMed] [Google Scholar]
- 17.Aufderheide TP, Lurie KG: Death by hyperventilation: a common and life-threatening problem during cardiopulmonary resuscitation. Crit Care Med 2004; 32(9 Suppl):S345–351. [DOI] [PubMed] [Google Scholar]
- 18.Garcia-Martinez M, Espinosa-Pereiro CE, Rodriguez-Nunez A: Bag and mask ventilation, an outstanding challenge that we should not leave behind. Resuscitation 2024; 199:110230. [DOI] [PubMed] [Google Scholar]
- 19.Culbreth RE, Gardenhire DS: Manual bag valve mask ventilation performance among respiratory therapists. Heart Lung 2021; 50:471–475. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Scott JB, Schneider JM, Schneider K, Li J: An evaluation of manual tidal volume and respiratory rate delivery during simulated resuscitation. Am J Emerg Med 2021; 45:446–450. [DOI] [PubMed] [Google Scholar]
- 21.Shiima Y, Berg RA, Bogner HR, Morales KH, et al. : Cardiac Arrests Associated With Tracheal Intubations in PICUs: A Multicenter Cohort Study. Crit Care Med 2016; 44:1675–1682. [DOI] [PubMed] [Google Scholar]
- 22.Li S, Hsieh TC, Rehder KJ, Nett S, et al. : Frequency of Desaturation and Association With Hemodynamic Adverse Events During Tracheal Intubations in PICUs. Pediatr Crit Care Med 2018; 19:e41–e50. [DOI] [PubMed] [Google Scholar]
- 23.Schmidt SK, Brou L, Deakyne SJ, Mistry RD, et al. : Frequency and Characterization of Tracheal Intubation Adverse Events in Pediatric Sepsis. Pediatr Crit Care Med 2018; 19:e80–e87. [DOI] [PubMed] [Google Scholar]
- 24.Capone CA, Emerson B, Sweberg T, Polikoff L, et al. : Intubation practice and outcomes among pediatric emergency departments: A report from National Emergency Airway Registry for Children (NEAR4KIDS). Acad Emerg Med 2022; 29:406–414. [DOI] [PubMed] [Google Scholar]
- 25.van Zanten HA, Kuypers K, van Zwet EW, van Vonderen JJ, et al. : A multi-centre randomised controlled trial of respiratory function monitoring during stabilisation of very preterm infants at birth. Resuscitation 2021; 167:317–325. [DOI] [PubMed] [Google Scholar]
- 26.Kuypers K, van Zanten HA, Heesters V, Kamlin O, et al. : Resuscitators' opinions on using a respiratory function monitor during neonatal resuscitation. Acta Paediatr 2023; 112:63–68. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Tracy MB, Klimek J, Coughtrey H, Shingde V, et al. : Mask leak in one-person mask ventilation compared to two-person in newborn infant manikin study. Arch Dis Child Fetal Neonatal Ed 2011; 96:F195–200. [DOI] [PubMed] [Google Scholar]
- 28.Bibl K, Wagner M, Dvorsky R, Haderer M, et al. : Impact of a Two-Person Mask Ventilation Technique During Neonatal Resuscitation: A Simulation-Based Randomized Controlled Trial. J Pediatr 2025; 282:114568. [DOI] [PubMed] [Google Scholar]
- 29.Takatori F, Isayama T, Hosono S, Iwasaki K: Comparison of tidal volume and airway pressure during neonatal resuscitation in mask leaks. Pediatr Int 2025; 67:e70139. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.