- Email: [email protected]
Comparison of airway intubation devices when using a biohazard suit: a feasibility study
American Journal of Emergency Medicine 33 (2015) 810–814
Contents lists available at ScienceDirect
American Journal of Emergency Medicine journal homepage: www.elsevier.com/locate/ajem
Brief Report
Comparison of airway intubation devices when using a biohazard suit: a feasibility study☆,☆☆,★ Kevin R. Weaver, DO, Gavin C. Barr Jr., MD, Kayla R. Long, DO, Leonel Diaz Jr., DO, Aaron S. Ratner, DO, Jeffery P. Reboul, DO, Douglas A. Sturm, DO, Marna Rayl Greenberg, DO, MPH ⁎, Stephen W. Dusza, DrPH, Bernadette Glenn-Porter, BS, Bryan G. Kane, MD Department of Emergency Medicine, Lehigh Valley Hospital and Health Network/USF MCOM, CC & I-78, Allentown, PA 18103
a r t i c l e
i n f o
Article history: Received 4 February 2015 Received in revised form 25 February 2015 Accepted 26 February 2015
a b s t r a c t Objectives: We set out to compare emergency medicine residents’ intubating times and success rates for direct laryngoscopy (DL), GlideScope-assisted intubation (GS), and the Supraglottic Airway Laryngopharyngeal Tube (SALT) airway with and without biohazard gear. Methods: Each resident passed through 2 sets of 3 testing stations (DL, GS, SALT) in succession, intubating Laerdal mannequin heads with the 3 modalities after randomization to start with or without biohazard gear. Results: Thirty-seven residents participated, and 27 were male (73%); 14 (37.8%) had prior experience intubating in biohazard suits. There was a statistically significant difference in those who had prior intubation experience between DL (37, 100%), GS (32, 86.5%), and SALT (12, 32.4%) (P b .001) and in median time to intubation (48 seconds, no suit; 57 seconds, with suits) (P = .03). There was no statistically significant difference between the overall times to intubate for the 3 devices. First-pass success was highest for DL (91.2%, no suit; 83.7%, suit) followed by GS (89%, no suit; 78.3%, suit) and SALT (51%, no suit; 67.6%, suit). Conclusion: A minority of participants had prior experience intubating in biohazard suits. Use of biohazard suits extends time to successful intubation. There was no difference in time to intubation for the 3 devices, but firstpass success was highest for DL (with or without biohazard gear). © 2015 Elsevier Inc. All rights reserved.
1. Introduction 1.1. Background and importance With cases of Ebola recently being reported in the United States [1], the reality of patients presenting to the emergency department (ED) after exposure to novel infectious pathogens is suddenly a very real and complex scenario. Providing clinical care while still ensuring
☆ The authors have no outside support information, conflicts, or financial interest to disclose; and this work is not under consideration elsewhere. This study was funded in part by a PCOM MedNet Grant; funding was received as institutional support in the form of an unrestricted resident grant. The study identifier is NCT01924559 and is registered on ClinicalTrials.gov. ☆☆ A poster of the study results was presented on October 27, 2014, at the ACEP Research Forum. ★ Author delegations: KW, KL, GB, MRG, BGP, SD, and BK conceived the study, designed the trial, and obtained research funding. KW, KR, GB, JR, AR, LD, DS, KL, MRG, BGP, and BK supervised the conduct of the trial and participated in data collection. BGP enrolled participants and managed the data, including quality control. SD provided statistical advice and performed the analysis. JR, AR, LD, DS, BK, MRG, and SD drafted the manuscript; and all authors contributed substantially to its revision. MRG takes responsibility for the article as a whole. ⁎ Corresponding author at: 1909 Earls Court, Allentown, PA 18103. Tel.: +1 610 967 3140 (home), +1 610 360 4849 (mobile). E-mail address: [email protected] (M.R. Greenberg). http://dx.doi.org/10.1016/j.ajem.2015.02.051 0735-6757/© 2015 Elsevier Inc. All rights reserved.
the safety of the medical team is fundamental to being prepared for these types of scenarios. The Chemical, Biological, Radiation, NuclearPersonal Protective Equipment (CBRN-PPE) used while treating patients with biohazardous exposure can pose a great challenge and potentially hinder patient care (eg, securing the airway in a critically ill patient with airway compromise while wearing CBRN-PPE) [2]. Furthermore, with the wide variety of commercial intubation products available, determining which product provides an advantage during a biohazard situation is important. Castle et al [2], in a randomized crossover study involving 66 paramedic students, tested 6 different intubation aids on a mannequin while wearing both their standard uniform and CBRN-PPE. These aids included bougie, stylet, McCoy laryngoscope, Airtraq, intubating laryngeal mask airway (ILMA), and standard intubation. Out of these 6, ILMA was found to be the easiest intubating aid [3]. They found that wearing CBRN-PPE had a negative impact on supraglottic airway insertion time [3]. Another study evaluating the effects of CBRN-PPE on high-dexterity airway and vascular skills demonstrated that complex skills such as endotracheal intubation and peripheral intravenous access were more difficult and time consuming while wearing PPE [4]. These high-dexterity skills were compared to the lower-dexterity skills of laryngeal mask airway insertion and intraosseous access [4]. Interestingly, a contradictory report from another group indicated that wearing CBRN-PPE does not alter clinical staff’s performance [5]. The supraglottic airway laryngopharyngeal tube
K.R. Weaver et al. / American Journal of Emergency Medicine 33 (2015) 810–814
(SALT) is a disposable supraglottic device alternative to the ILMA; we investigated this airway device within these parameters.
Table 2 Cross-classification of the number of attempts at intubation by the number of tubes used, stratified by the device type Device
1.2. Goals In this prospective, randomized, feasibility study, we investigated the impact that biohazard gear has on emergency medicine (EM) resident time to successful intubation and, in particular, determined how the SALT airway device compares to traditional direct laryngoscopy (DL) and fiberoptic (video) GlideScope (GS) intubation techniques based on their first-pass and overall success rates.
Attempts
Number of tubes used 2
3+
Missing
Total
DL
1 2 3 4 5 Total
31 (91.2) 2 (5.9) 0 (0) 1 (2.9) 0 (0) 34 (100)
0 (0) 1 (100) 0 (0) 0 (0) 0 (0) 1 (100)
0 (0) 0 (0) 1 (50) 0 (0) 1 (50) 2 (100)
0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0)
31 (83.8) 3 (8.1) 1 (2.7) 1 (2.7) 1 (2.7) 37 (100)
GS
1 2 3 4 Total
29 (82.9) 6 (17.1) 0 (0) 0 (0) 35 (100)
0 (0) 0 (0) 0 (0) 1 (100) 1 (100)
0 (0) 0 (0) 0 (0) 0 (0) 0 (0)
1 (50) 0 (0) 0 (0) 0 (0) 1 (100)
30 (79) 6 (15.8) 0 (0) 1 (2.6) 37 (100)
SALT
1 2 3 4 Total
25 (67.6) 7 (18.9) 4 (10.8) 1 (2.7) 37 (100)
0 (0) 0 (0) 0 (0) 0 (0) 0 (0)
0 (0) 0 (0) 0 (0) 0 (0) 0 (0)
0 (0) 0 (0) 0 (0) 0 (0) 0 (0)
25 (67.6) 7 (18.9) 4 (10.8) 1 (2.7) 37 (100)
2. Methods and materials 2.1. Design and setting This prospective, randomized, feasibility was reviewed and approved by the hospital’s Institutional Review Board. The study took place in the fall of 2013 at a large suburban academic dually accredited residency program in the United States.
811
1
2.2. Selection of participants Postgraduate year I to IV EM residents were offered study participation. As part of their standard training, they participated in a daylong airway training course, where each of the study devices was used by trainees both in and out of CBRN-PPE. Before the training, the principal investigator explained the study procedures to the residents and offered them the opportunity to participate. The subjects who volunteered to join the study were consented and enrolled. No compensation was offered for their participation. Intubation performance data were collected only on consented participants. Residents were assigned chronologic numbers as they arrived and signed in for the educational session. Enrolled subjects were then randomized by odd or even identification numbers as to whether they would make their initial attempt in standard clothing or biohazard gear.
Airway Management Trainer heads for a total of 6 measurements per resident. Stations were prepared with a drawn schematic to maintain equipment layout consistency. Intubations were timed using each of the 3 intubation devices, both in conventional clothing and in CBRN-PPE. 2.4. Methods and measurement Research study assistants were trained and evaluated on proper techniques for timing intubation procedures. Metrics for intrarater reliability were collected and study assistants approved. Each evaluation station had a research assistant who measured the elapsed time from start to when 2 successful ventilations were delivered to the mannequin. The order of devices used was the same for each participant: (1) DL, (2) the GS video laryngoscope, and (3) the SALT airway device.
2.3. Intervention Using this randomization, study participants passed through 2 sets of 3 testing stations, in prearranged succession, intubating Laerdal
Table 1 Demographic characteristics of study participants n (%) Sex Male Female Program year 1 2 3 4 Prior biohazard suit experience No Yes Experience intubating in biohazard suit No Yes Experience using DL No Yes Experience using GS No Yes Experience using SALT No Yes
27 (73) 10 (27) 11 (29.7) 11 (29.7) 9 (24.3) 6 (16.2) 18 (48.7) 19 (51.4) 23 (62.2) 14 (37.8) 0 (0) 37 (100) 5 (13.5) 32 (86.5) 25 (67.6) 12 (32.4)
2.5. Outcomes Data were collected, per device, on elapsed time to successful intubation, first-attempt success, number of endotracheal tubes used for each attempt, and unrecognized esophageal intubations performed wearing standard clothing vs CBRN-PPE. Subjects were given data sheets with their assigned numbers to have their times recorded by the timer at each station. Only study participants returned these completed data forms to a research staff member after all 6 stations. Limited demographic information was also collected from participants such as sex, resident year, prior intubation experience, prior exposure to each of the tested devices, and experience wearing a biohazard suit. The resulting data were entered into a password-protected database and deidentified before analysis. 2.6. Data analysis The primary outcome measure of the study was mean time required to successfully establish a definitive airway in the study mannequin. The distribution of the primary outcome measure was nonnormally distributed. A transformation using the natural log was used to stabilize the variance estimates. The goal of the analysis was to determine if wearing CRBN-PPE affected a clinician’s performance in performing intubation. A summary variable was created reflecting the level of prior experience with intubations (based primarily on self-reported training records). The variable was categorized to reflect differing levels of resident experience and then used to stratify all initial analyses to
812
K.R. Weaver et al. / American Journal of Emergency Medicine 33 (2015) 810–814
used to assess the differences in time to successful intubation between wearing and not wearing CBRN-PPE. Because 3 different intubation devices were used, an indicator variable was included in the model for type of device; and an interaction term was used to assess the type of device, CBRN-PPE effect modification. Because we observed differences in time to successful intubation in our univariate analyses for participant sex and prior experience with biohazard suits, these variables were included in the regression model. A random effect for participant was included in the model. To visualize the observed effect modification, predictive values of log time to successful intubation while wearing CBRN-PPE, stratified by intubation device type, were created. All statistical analyses were performed using Stata v.12.1 (Stata Corporation, College Station, TX). 3. Results 3.1. Characteristics of study subjects Fig. 1. Participant in CBRN-PPE gear using DL device.
Emergency medicine residents (37/38 who attended the session) participated in the study, and their characteristics are presented in Table 1. Of these, a majority were male (73%, n = 27) and had previous experience with wearing CBRN-PPE (51.4%, n = 19). However, only 14 (37.8%) had ever had the opportunity to practice intubation while wearing CBRN-PPE. Experience using the 3 different intubating devices in this study varied, with all participants reporting experience using DL (100%, n = 37), a majority using the GS (86.5%, n = 32), and a minority (32.4%, n = 12) using the SALT device (P b .001). 3.2. Main results The cross-classification of number of attempts at successful intubation and the number of tubes used is presented in Table 2. A majority
100
100
assess the impact of intubation-specific experience on changes to success and time to intubation under biohazard conditions. Previous experience with CBRN-PPE was quantified as a dichotomous variable (“Yes/No”) based on participant response. Descriptive statistics were used to characterize the study participants and describe the time required to successfully intubate the study mannequin. Median and geometric mean time required to successful intubation with each device was compared between standard conditions and CBRNPPE conditions using the Wilcoxon rank sum test and the Student t test. Pearson χ2 was used to assess the distribution of participant responses by wearing CBRN-PPE. A linear random-effects regression model was
Percent
60
80
B
0
0
20
40
60 40 20
Percent
80
A
0
100
200
300
400
500
0
100
200
D 60
Percent
0
0
20
20
40
40
60
80
80
C Percent
400
300
Time (seconds)
100
100
Time (seconds)
0
100
200
Time (seconds)
300
2
3
4
5
6
Log Time (seconds)
Fig. 2. Histograms of the time required to successfully compete intubation using (A) DL, (B) GS, and (C) SALT. Because these distributions were right-skewed, a log transformation using the natural log was performed. The resulting distribution of time to complete intubation for all devices combined is presented (D).
813
3
4
5
6
4. Discussion
DL
Bi oh az ar d
Bi oh az ar d N
N
N
o
o
Bi oh az ar d
Bi oh az ar d
Bi oh az ar d
o
Bi oh az ar d
2
log time to successful intubation
K.R. Weaver et al. / American Journal of Emergency Medicine 33 (2015) 810–814
Glide Scope
SALT
Fig. 3. Boxplots of log time to successful intubation by device and whether the biohazard suit was being worn.
of participants successfully intubated the training mannequin at the first attempt for each device (91.2%, 82.9%, and 67.6% for the DL, GS, and SALT devices, respectively). Although DL had the highest success rate on the first attempt at intubation (91%, n = 31), several participants required multiple attempts to intubate; and 50% of these attempts required 2 or more tubes to successfully complete the task (Fig. 1). For the SALT device, only 67.6% of the participants successfully intubated the mannequin in the first attempt; but only 1 tube was required in all cases. The distribution of the time, in seconds, to successfully complete intubation is presented in Fig. 2 (A through C) for each device. The distribution of time for each device is right-skewed, with several intubation attempts, for each device, taking much longer than the bulk of the attempts. After taking a natural log transformation, the resulting distribution of time for all devices combined is presented in Fig. 2D. For all of the study observations (combining attempts with or without CBRNPPE), no differences were observed between devices, with geometric mean time to completing intubations being approximately 60.6 (95% confidence interval [CI], 52.4-70.1) seconds for DL, 59.8 (95% CI, 51.968.9) seconds for GS(P = .89), and 52 (95% CI, 44.1-61.5) seconds for SALT (P = .16, DL vs SALT; P = .29, GS vs SALT). When exploring time to successful intubation for each device and for wearing CBRN-PPE, a significant difference in time to successful intubation was observed for the DL device, with attempts in CBRN-PPE taking on average 27 seconds longer (95% CI, 17.8-39.8; P = .001). No significant difference was observed for GS (Δ = 8.7 seconds; 95% CI, − 3.5 to 19.1; P = .49) or SALT (Δ = 0.25 seconds; 95% CI, − 11.7 to 12.1; P = .79) (Fig. 3). Since previous biohazard suit experience and sex showed significant associations with the outcome, a multivariate model combining these factors is presented in Table 3.
This study confirms that intubating in CBRN-PPE takes more time than in conventional clothing and is consistent with some prior citations [2,3]. Although clinically significant time may be longer than statistically significant, it does lead the authors to believe that clinician training could be advantageous in narrowing this gap. With the minority of participants having practiced intubation in CBRN-PPE previously, most hospitals need time and training to get ready for Ebola or similar threats [6]. Making airway simulation training in CBRN-PPE a standard educational offering seems prudent. This small study done at 1 institution with a convenience sample of residents puts into question the generalizability of the results. The choice of order of devices was simply one of convenience and easily could have impacted device comparison results. Specifically, the lack of randomization of airway devices may have allowed for learning bias. It is unclear why multiple tubes were required with DL and not with other devices. The authors might hypothesize that because there were differences in visualization and in resident training level, participants chose a different tube (whether an accurate decision or not) when trouble-shooting in a missed attempt. Further, in regard to DL, it is not surprising that intubation with DL took longer than GS or SALT. It is a skill that requires direct visualization, which is clearly impaired while wearing CBRN-PPE. Although prior experience with intubating in CBRN-PPE was controlled for, it was noted that during the study, the biohazard suits did not fit a couple of the residents properly (not tall or large enough). This is likely to occur in a real-life scenario, and our duct taping may not have resulted in consistent results for participants while in gear. Because our sample size was small, it seems unlikely that we can make strong statements about the SALT device. The method for assuring placement of the endotracheal tube while in CBRN-PPE while using the SALT device in comparison to the others must be considered. Listening with a stethoscope has its limitations; and although capnography may be reassuring, visualizing the tube passing through the cords is an advantage in favor of DL and GS. Those concerns being noted, although SALT had low first-pass attempt rates in standard clothing, it was the only device of the 3 to improve in CBRN-PPE. Additionally, the cleaning and sterilization of the devices in a real encounter must be considered. If the time to intubation is similar between all 3 devices, then certainly the most advantageous for a clinician to consider would be the easiest to clean or cheapest to replace if disposed. In real-life biohazard emergencies, the cost and time of cleaning and replacement would favor the SALT. In summary, a minority of participants had prior experience intubating in CBRN-PPE. Use of CBRN-PPE extended time to successful intubation. There was no difference in time to intubate for the 3 devices studied, but first-pass success was highest for DL (with or without CBRN-PPE). Increasing opportunities for residents and EM clinicians to practice intubation in CBRN-PPE is ideal especially considering the recent Ebola threat.
Table 3 Results from multivariate random-effects regression model β coefficient
GS SALT Biohazard suit (yes) GS × biohazard interaction SALT × biohazard interaction Female Experience (yes) _cons
0.135457 0.072779 0.445129 −0.29869 −0.45005 0.247833 −0.31536 3.977407
The dependent variable is log time to intubation.
P value
.311 .586 .001 .114 .017 .036 .003 0
95% CI
Interpretation
Lower
Upper
−0.12676 −0.18944 0.182908 −0.66953 −0.82088 0.0156 −0.52171 3.740576
0.397678 0.335 0.70735 0.072143 −0.07921 0.480066 −0.10902 4.214239
Not really interpretable because of interaction term Not really interpretable because of interaction term Difference in time between DL by biohazard The effect of biohazard suit differs between DL and GS. The effect of biohazard suit differs between DL and SALT. Women have significantly longer times than men. Participants with biohazard experience have significantly shorter intubation times. Log time when all other model coefficients are set to zero
814
K.R. Weaver et al. / American Journal of Emergency Medicine 33 (2015) 810–814
Acknowledgement The authors would like to acknowledge the efforts of Matthew Meyers who helped participate in data collection and enrollment.
References [1] Centers for Disease Control and Prevention. Cases of Ebola diagnosed in the United States. http://www.cdc.gov/vhf/ebola/outbreaks/2014-west-africa/united-statesimported-case.html; 2014. [Accessed December 3, 2014].
[2] Castle N, Pillay Y, Spencer N. Comparison of six different intubation aids for use while wearing CBRN-PPE: a manikin study. Resuscitation 2011;82:1548–52. [3] Castle N, Pillay Y, Spencer N. Insertion of six different supraglottic airway devices whilst wearing chemical, biological, radiation, nuclear-personal protective equipment: a manikin study. Anaesthesia 2011;66:983–98. [4] Castle N, Owen R, Hann M, Reeves D, Gurney I. Impact of Chemical, Biological, Radiation, and Nuclear Personal Protective Equipment on the performance of low- and high-dexterity airway and vascular access skills. Resuscitation 2009;80(11):1290–5. [5] Udayasiri R, Knott J, Taylor D, et al. Emergency department staff can effectively resuscitate in level C personal protective equipment. EMA 2007;18:113–21. [6] Courge KH. Hospitals need time, training, to get ready for Ebola. Scientific American online release 10/25/2014. http://www.scientificamerican.com/article/hospitalsneed-time-training-to-get-ready-for-ebola/; 2014. [Accessed December 3, 2014].
Contents lists available at ScienceDirect
American Journal of Emergency Medicine journal homepage: www.elsevier.com/locate/ajem
Brief Report
Comparison of airway intubation devices when using a biohazard suit: a feasibility study☆,☆☆,★ Kevin R. Weaver, DO, Gavin C. Barr Jr., MD, Kayla R. Long, DO, Leonel Diaz Jr., DO, Aaron S. Ratner, DO, Jeffery P. Reboul, DO, Douglas A. Sturm, DO, Marna Rayl Greenberg, DO, MPH ⁎, Stephen W. Dusza, DrPH, Bernadette Glenn-Porter, BS, Bryan G. Kane, MD Department of Emergency Medicine, Lehigh Valley Hospital and Health Network/USF MCOM, CC & I-78, Allentown, PA 18103
a r t i c l e
i n f o
Article history: Received 4 February 2015 Received in revised form 25 February 2015 Accepted 26 February 2015
a b s t r a c t Objectives: We set out to compare emergency medicine residents’ intubating times and success rates for direct laryngoscopy (DL), GlideScope-assisted intubation (GS), and the Supraglottic Airway Laryngopharyngeal Tube (SALT) airway with and without biohazard gear. Methods: Each resident passed through 2 sets of 3 testing stations (DL, GS, SALT) in succession, intubating Laerdal mannequin heads with the 3 modalities after randomization to start with or without biohazard gear. Results: Thirty-seven residents participated, and 27 were male (73%); 14 (37.8%) had prior experience intubating in biohazard suits. There was a statistically significant difference in those who had prior intubation experience between DL (37, 100%), GS (32, 86.5%), and SALT (12, 32.4%) (P b .001) and in median time to intubation (48 seconds, no suit; 57 seconds, with suits) (P = .03). There was no statistically significant difference between the overall times to intubate for the 3 devices. First-pass success was highest for DL (91.2%, no suit; 83.7%, suit) followed by GS (89%, no suit; 78.3%, suit) and SALT (51%, no suit; 67.6%, suit). Conclusion: A minority of participants had prior experience intubating in biohazard suits. Use of biohazard suits extends time to successful intubation. There was no difference in time to intubation for the 3 devices, but firstpass success was highest for DL (with or without biohazard gear). © 2015 Elsevier Inc. All rights reserved.
1. Introduction 1.1. Background and importance With cases of Ebola recently being reported in the United States [1], the reality of patients presenting to the emergency department (ED) after exposure to novel infectious pathogens is suddenly a very real and complex scenario. Providing clinical care while still ensuring
☆ The authors have no outside support information, conflicts, or financial interest to disclose; and this work is not under consideration elsewhere. This study was funded in part by a PCOM MedNet Grant; funding was received as institutional support in the form of an unrestricted resident grant. The study identifier is NCT01924559 and is registered on ClinicalTrials.gov. ☆☆ A poster of the study results was presented on October 27, 2014, at the ACEP Research Forum. ★ Author delegations: KW, KL, GB, MRG, BGP, SD, and BK conceived the study, designed the trial, and obtained research funding. KW, KR, GB, JR, AR, LD, DS, KL, MRG, BGP, and BK supervised the conduct of the trial and participated in data collection. BGP enrolled participants and managed the data, including quality control. SD provided statistical advice and performed the analysis. JR, AR, LD, DS, BK, MRG, and SD drafted the manuscript; and all authors contributed substantially to its revision. MRG takes responsibility for the article as a whole. ⁎ Corresponding author at: 1909 Earls Court, Allentown, PA 18103. Tel.: +1 610 967 3140 (home), +1 610 360 4849 (mobile). E-mail address: [email protected] (M.R. Greenberg). http://dx.doi.org/10.1016/j.ajem.2015.02.051 0735-6757/© 2015 Elsevier Inc. All rights reserved.
the safety of the medical team is fundamental to being prepared for these types of scenarios. The Chemical, Biological, Radiation, NuclearPersonal Protective Equipment (CBRN-PPE) used while treating patients with biohazardous exposure can pose a great challenge and potentially hinder patient care (eg, securing the airway in a critically ill patient with airway compromise while wearing CBRN-PPE) [2]. Furthermore, with the wide variety of commercial intubation products available, determining which product provides an advantage during a biohazard situation is important. Castle et al [2], in a randomized crossover study involving 66 paramedic students, tested 6 different intubation aids on a mannequin while wearing both their standard uniform and CBRN-PPE. These aids included bougie, stylet, McCoy laryngoscope, Airtraq, intubating laryngeal mask airway (ILMA), and standard intubation. Out of these 6, ILMA was found to be the easiest intubating aid [3]. They found that wearing CBRN-PPE had a negative impact on supraglottic airway insertion time [3]. Another study evaluating the effects of CBRN-PPE on high-dexterity airway and vascular skills demonstrated that complex skills such as endotracheal intubation and peripheral intravenous access were more difficult and time consuming while wearing PPE [4]. These high-dexterity skills were compared to the lower-dexterity skills of laryngeal mask airway insertion and intraosseous access [4]. Interestingly, a contradictory report from another group indicated that wearing CBRN-PPE does not alter clinical staff’s performance [5]. The supraglottic airway laryngopharyngeal tube
K.R. Weaver et al. / American Journal of Emergency Medicine 33 (2015) 810–814
(SALT) is a disposable supraglottic device alternative to the ILMA; we investigated this airway device within these parameters.
Table 2 Cross-classification of the number of attempts at intubation by the number of tubes used, stratified by the device type Device
1.2. Goals In this prospective, randomized, feasibility study, we investigated the impact that biohazard gear has on emergency medicine (EM) resident time to successful intubation and, in particular, determined how the SALT airway device compares to traditional direct laryngoscopy (DL) and fiberoptic (video) GlideScope (GS) intubation techniques based on their first-pass and overall success rates.
Attempts
Number of tubes used 2
3+
Missing
Total
DL
1 2 3 4 5 Total
31 (91.2) 2 (5.9) 0 (0) 1 (2.9) 0 (0) 34 (100)
0 (0) 1 (100) 0 (0) 0 (0) 0 (0) 1 (100)
0 (0) 0 (0) 1 (50) 0 (0) 1 (50) 2 (100)
0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0)
31 (83.8) 3 (8.1) 1 (2.7) 1 (2.7) 1 (2.7) 37 (100)
GS
1 2 3 4 Total
29 (82.9) 6 (17.1) 0 (0) 0 (0) 35 (100)
0 (0) 0 (0) 0 (0) 1 (100) 1 (100)
0 (0) 0 (0) 0 (0) 0 (0) 0 (0)
1 (50) 0 (0) 0 (0) 0 (0) 1 (100)
30 (79) 6 (15.8) 0 (0) 1 (2.6) 37 (100)
SALT
1 2 3 4 Total
25 (67.6) 7 (18.9) 4 (10.8) 1 (2.7) 37 (100)
0 (0) 0 (0) 0 (0) 0 (0) 0 (0)
0 (0) 0 (0) 0 (0) 0 (0) 0 (0)
0 (0) 0 (0) 0 (0) 0 (0) 0 (0)
25 (67.6) 7 (18.9) 4 (10.8) 1 (2.7) 37 (100)
2. Methods and materials 2.1. Design and setting This prospective, randomized, feasibility was reviewed and approved by the hospital’s Institutional Review Board. The study took place in the fall of 2013 at a large suburban academic dually accredited residency program in the United States.
811
1
2.2. Selection of participants Postgraduate year I to IV EM residents were offered study participation. As part of their standard training, they participated in a daylong airway training course, where each of the study devices was used by trainees both in and out of CBRN-PPE. Before the training, the principal investigator explained the study procedures to the residents and offered them the opportunity to participate. The subjects who volunteered to join the study were consented and enrolled. No compensation was offered for their participation. Intubation performance data were collected only on consented participants. Residents were assigned chronologic numbers as they arrived and signed in for the educational session. Enrolled subjects were then randomized by odd or even identification numbers as to whether they would make their initial attempt in standard clothing or biohazard gear.
Airway Management Trainer heads for a total of 6 measurements per resident. Stations were prepared with a drawn schematic to maintain equipment layout consistency. Intubations were timed using each of the 3 intubation devices, both in conventional clothing and in CBRN-PPE. 2.4. Methods and measurement Research study assistants were trained and evaluated on proper techniques for timing intubation procedures. Metrics for intrarater reliability were collected and study assistants approved. Each evaluation station had a research assistant who measured the elapsed time from start to when 2 successful ventilations were delivered to the mannequin. The order of devices used was the same for each participant: (1) DL, (2) the GS video laryngoscope, and (3) the SALT airway device.
2.3. Intervention Using this randomization, study participants passed through 2 sets of 3 testing stations, in prearranged succession, intubating Laerdal
Table 1 Demographic characteristics of study participants n (%) Sex Male Female Program year 1 2 3 4 Prior biohazard suit experience No Yes Experience intubating in biohazard suit No Yes Experience using DL No Yes Experience using GS No Yes Experience using SALT No Yes
27 (73) 10 (27) 11 (29.7) 11 (29.7) 9 (24.3) 6 (16.2) 18 (48.7) 19 (51.4) 23 (62.2) 14 (37.8) 0 (0) 37 (100) 5 (13.5) 32 (86.5) 25 (67.6) 12 (32.4)
2.5. Outcomes Data were collected, per device, on elapsed time to successful intubation, first-attempt success, number of endotracheal tubes used for each attempt, and unrecognized esophageal intubations performed wearing standard clothing vs CBRN-PPE. Subjects were given data sheets with their assigned numbers to have their times recorded by the timer at each station. Only study participants returned these completed data forms to a research staff member after all 6 stations. Limited demographic information was also collected from participants such as sex, resident year, prior intubation experience, prior exposure to each of the tested devices, and experience wearing a biohazard suit. The resulting data were entered into a password-protected database and deidentified before analysis. 2.6. Data analysis The primary outcome measure of the study was mean time required to successfully establish a definitive airway in the study mannequin. The distribution of the primary outcome measure was nonnormally distributed. A transformation using the natural log was used to stabilize the variance estimates. The goal of the analysis was to determine if wearing CRBN-PPE affected a clinician’s performance in performing intubation. A summary variable was created reflecting the level of prior experience with intubations (based primarily on self-reported training records). The variable was categorized to reflect differing levels of resident experience and then used to stratify all initial analyses to
812
K.R. Weaver et al. / American Journal of Emergency Medicine 33 (2015) 810–814
used to assess the differences in time to successful intubation between wearing and not wearing CBRN-PPE. Because 3 different intubation devices were used, an indicator variable was included in the model for type of device; and an interaction term was used to assess the type of device, CBRN-PPE effect modification. Because we observed differences in time to successful intubation in our univariate analyses for participant sex and prior experience with biohazard suits, these variables were included in the regression model. A random effect for participant was included in the model. To visualize the observed effect modification, predictive values of log time to successful intubation while wearing CBRN-PPE, stratified by intubation device type, were created. All statistical analyses were performed using Stata v.12.1 (Stata Corporation, College Station, TX). 3. Results 3.1. Characteristics of study subjects Fig. 1. Participant in CBRN-PPE gear using DL device.
Emergency medicine residents (37/38 who attended the session) participated in the study, and their characteristics are presented in Table 1. Of these, a majority were male (73%, n = 27) and had previous experience with wearing CBRN-PPE (51.4%, n = 19). However, only 14 (37.8%) had ever had the opportunity to practice intubation while wearing CBRN-PPE. Experience using the 3 different intubating devices in this study varied, with all participants reporting experience using DL (100%, n = 37), a majority using the GS (86.5%, n = 32), and a minority (32.4%, n = 12) using the SALT device (P b .001). 3.2. Main results The cross-classification of number of attempts at successful intubation and the number of tubes used is presented in Table 2. A majority
100
100
assess the impact of intubation-specific experience on changes to success and time to intubation under biohazard conditions. Previous experience with CBRN-PPE was quantified as a dichotomous variable (“Yes/No”) based on participant response. Descriptive statistics were used to characterize the study participants and describe the time required to successfully intubate the study mannequin. Median and geometric mean time required to successful intubation with each device was compared between standard conditions and CBRNPPE conditions using the Wilcoxon rank sum test and the Student t test. Pearson χ2 was used to assess the distribution of participant responses by wearing CBRN-PPE. A linear random-effects regression model was
Percent
60
80
B
0
0
20
40
60 40 20
Percent
80
A
0
100
200
300
400
500
0
100
200
D 60
Percent
0
0
20
20
40
40
60
80
80
C Percent
400
300
Time (seconds)
100
100
Time (seconds)
0
100
200
Time (seconds)
300
2
3
4
5
6
Log Time (seconds)
Fig. 2. Histograms of the time required to successfully compete intubation using (A) DL, (B) GS, and (C) SALT. Because these distributions were right-skewed, a log transformation using the natural log was performed. The resulting distribution of time to complete intubation for all devices combined is presented (D).
813
3
4
5
6
4. Discussion
DL
Bi oh az ar d
Bi oh az ar d N
N
N
o
o
Bi oh az ar d
Bi oh az ar d
Bi oh az ar d
o
Bi oh az ar d
2
log time to successful intubation
K.R. Weaver et al. / American Journal of Emergency Medicine 33 (2015) 810–814
Glide Scope
SALT
Fig. 3. Boxplots of log time to successful intubation by device and whether the biohazard suit was being worn.
of participants successfully intubated the training mannequin at the first attempt for each device (91.2%, 82.9%, and 67.6% for the DL, GS, and SALT devices, respectively). Although DL had the highest success rate on the first attempt at intubation (91%, n = 31), several participants required multiple attempts to intubate; and 50% of these attempts required 2 or more tubes to successfully complete the task (Fig. 1). For the SALT device, only 67.6% of the participants successfully intubated the mannequin in the first attempt; but only 1 tube was required in all cases. The distribution of the time, in seconds, to successfully complete intubation is presented in Fig. 2 (A through C) for each device. The distribution of time for each device is right-skewed, with several intubation attempts, for each device, taking much longer than the bulk of the attempts. After taking a natural log transformation, the resulting distribution of time for all devices combined is presented in Fig. 2D. For all of the study observations (combining attempts with or without CBRNPPE), no differences were observed between devices, with geometric mean time to completing intubations being approximately 60.6 (95% confidence interval [CI], 52.4-70.1) seconds for DL, 59.8 (95% CI, 51.968.9) seconds for GS(P = .89), and 52 (95% CI, 44.1-61.5) seconds for SALT (P = .16, DL vs SALT; P = .29, GS vs SALT). When exploring time to successful intubation for each device and for wearing CBRN-PPE, a significant difference in time to successful intubation was observed for the DL device, with attempts in CBRN-PPE taking on average 27 seconds longer (95% CI, 17.8-39.8; P = .001). No significant difference was observed for GS (Δ = 8.7 seconds; 95% CI, − 3.5 to 19.1; P = .49) or SALT (Δ = 0.25 seconds; 95% CI, − 11.7 to 12.1; P = .79) (Fig. 3). Since previous biohazard suit experience and sex showed significant associations with the outcome, a multivariate model combining these factors is presented in Table 3.
This study confirms that intubating in CBRN-PPE takes more time than in conventional clothing and is consistent with some prior citations [2,3]. Although clinically significant time may be longer than statistically significant, it does lead the authors to believe that clinician training could be advantageous in narrowing this gap. With the minority of participants having practiced intubation in CBRN-PPE previously, most hospitals need time and training to get ready for Ebola or similar threats [6]. Making airway simulation training in CBRN-PPE a standard educational offering seems prudent. This small study done at 1 institution with a convenience sample of residents puts into question the generalizability of the results. The choice of order of devices was simply one of convenience and easily could have impacted device comparison results. Specifically, the lack of randomization of airway devices may have allowed for learning bias. It is unclear why multiple tubes were required with DL and not with other devices. The authors might hypothesize that because there were differences in visualization and in resident training level, participants chose a different tube (whether an accurate decision or not) when trouble-shooting in a missed attempt. Further, in regard to DL, it is not surprising that intubation with DL took longer than GS or SALT. It is a skill that requires direct visualization, which is clearly impaired while wearing CBRN-PPE. Although prior experience with intubating in CBRN-PPE was controlled for, it was noted that during the study, the biohazard suits did not fit a couple of the residents properly (not tall or large enough). This is likely to occur in a real-life scenario, and our duct taping may not have resulted in consistent results for participants while in gear. Because our sample size was small, it seems unlikely that we can make strong statements about the SALT device. The method for assuring placement of the endotracheal tube while in CBRN-PPE while using the SALT device in comparison to the others must be considered. Listening with a stethoscope has its limitations; and although capnography may be reassuring, visualizing the tube passing through the cords is an advantage in favor of DL and GS. Those concerns being noted, although SALT had low first-pass attempt rates in standard clothing, it was the only device of the 3 to improve in CBRN-PPE. Additionally, the cleaning and sterilization of the devices in a real encounter must be considered. If the time to intubation is similar between all 3 devices, then certainly the most advantageous for a clinician to consider would be the easiest to clean or cheapest to replace if disposed. In real-life biohazard emergencies, the cost and time of cleaning and replacement would favor the SALT. In summary, a minority of participants had prior experience intubating in CBRN-PPE. Use of CBRN-PPE extended time to successful intubation. There was no difference in time to intubate for the 3 devices studied, but first-pass success was highest for DL (with or without CBRN-PPE). Increasing opportunities for residents and EM clinicians to practice intubation in CBRN-PPE is ideal especially considering the recent Ebola threat.
Table 3 Results from multivariate random-effects regression model β coefficient
GS SALT Biohazard suit (yes) GS × biohazard interaction SALT × biohazard interaction Female Experience (yes) _cons
0.135457 0.072779 0.445129 −0.29869 −0.45005 0.247833 −0.31536 3.977407
The dependent variable is log time to intubation.
P value
.311 .586 .001 .114 .017 .036 .003 0
95% CI
Interpretation
Lower
Upper
−0.12676 −0.18944 0.182908 −0.66953 −0.82088 0.0156 −0.52171 3.740576
0.397678 0.335 0.70735 0.072143 −0.07921 0.480066 −0.10902 4.214239
Not really interpretable because of interaction term Not really interpretable because of interaction term Difference in time between DL by biohazard The effect of biohazard suit differs between DL and GS. The effect of biohazard suit differs between DL and SALT. Women have significantly longer times than men. Participants with biohazard experience have significantly shorter intubation times. Log time when all other model coefficients are set to zero
814
K.R. Weaver et al. / American Journal of Emergency Medicine 33 (2015) 810–814
Acknowledgement The authors would like to acknowledge the efforts of Matthew Meyers who helped participate in data collection and enrollment.
References [1] Centers for Disease Control and Prevention. Cases of Ebola diagnosed in the United States. http://www.cdc.gov/vhf/ebola/outbreaks/2014-west-africa/united-statesimported-case.html; 2014. [Accessed December 3, 2014].
[2] Castle N, Pillay Y, Spencer N. Comparison of six different intubation aids for use while wearing CBRN-PPE: a manikin study. Resuscitation 2011;82:1548–52. [3] Castle N, Pillay Y, Spencer N. Insertion of six different supraglottic airway devices whilst wearing chemical, biological, radiation, nuclear-personal protective equipment: a manikin study. Anaesthesia 2011;66:983–98. [4] Castle N, Owen R, Hann M, Reeves D, Gurney I. Impact of Chemical, Biological, Radiation, and Nuclear Personal Protective Equipment on the performance of low- and high-dexterity airway and vascular access skills. Resuscitation 2009;80(11):1290–5. [5] Udayasiri R, Knott J, Taylor D, et al. Emergency department staff can effectively resuscitate in level C personal protective equipment. EMA 2007;18:113–21. [6] Courge KH. Hospitals need time, training, to get ready for Ebola. Scientific American online release 10/25/2014. http://www.scientificamerican.com/article/hospitalsneed-time-training-to-get-ready-for-ebola/; 2014. [Accessed December 3, 2014].