a b s t r a c t

Study objective: Endotracheal intubation is frequently performed in emergency departments (EDs). First- pass success is important because repeated attempts are associated with poor outcomes. We sought to identify factors associated with first-pass success in emergency endotracheal intubation.

Methods: We analyzed emergency orotracheal intubations on adult patients in an ED located in South Korea from Jan. 2013 to Dec. 2016. Various operator-, procedure- and patient-related factors were screened with univariable logistic regression. Using variables with P-values less than 0.2, a multiple logis- tic regression model was constructed to identify independent predictors.

Results: There were 1154 eligible cases. First-pass success was achieved in 974 (84.4%) cases. Among operator-related factors, clinical experience (OR: 2.93, 5.26, 3.80 and 5.71; 95% CI: 1.62-5.26, 2.80-9.84, 1.81-8.13 and 2.07-18.67 for PGY 3, 4 and 5 residents and EM specialists, respectively, relative to PGY 2 residents) and physician based outside the ED (OR: 0.10; 95% CI: 0.04-0.25) were independently associated with first-pass success. There was no statistically or clinically significant difference for first- pass success rate as determined by operator’s gender (83.6% for female vs. 84.8% for male; 95% CI for difference: -3.1% to 5.8%). Among patient-related factors, restricted mouth opening (OR: 0.47; 95% CI: 0.31-0.72), restricted neck extension (OR: 0.57; 95% CI: 0.39-0.85) and swollen tongue (OR: 0.46; 95% CI: 0.28-0.77) were independent predictors of first-pass success.

Conclusions: Operator characteristics, including clinical experience and working department, and patient characteristics, including restricted mouth opening, restricted neck extension and swollen tongue, were independent predictors of first-pass success in emergency endotracheal intubation.

(C) 2019

Introduction

Endotracheal intubation is one of the most important critical interventions performed in the emergency department. While it is essential for securing the airway and providing respiratory sup- port in many clinical conditions, it is also associated with various adverse events ranging from transient desaturation to the death of a patient.

It has been emphasized that successful intubation at the first attempt (first-pass success) should be sought especially in the emergency department where most of the patients being intubated are already critically ill. This has been supported by previous stud- ies reporting the link between repeated intubation attempts and various adverse outcomes [1-4].

* Corresponding author.

E-mail address: [email protected] (J. Kim).

First-pass success can be influenced by various factors [5,6]. Further knowledge about these influences can be useful in predict- ing high risk airway procedures as well as in improving overall quality of endotracheal intubation in emergency departments. We sought to assess the association between first-pass success rate and various characteristics of operators, procedures and patients in emergency endotracheal intubation.

Materials and methods

Study design

This is a retrospective observational study analyzing emergency airway records of a single emergency department in South Korea from Jan. 2013 to Dec. 2016. The institutional review boards of the facility approved the study and provided a waiver of informed consent.

https://doi.org/10.1016/j.ajem.2019.09.001 0735-6757/(C) 2019

Study setting, data source and study population

The study facility was a tertiary academic hospital in South Korea with an annual emergency department visit of greater than 80,000 patients per year. Emergency endotracheal intubations were primarily performed by emergency medicine (EM) residents (postgraduate year [PGY] 2-5 in South Korea; PGY 1 residents, called ”interns”, rotate through different disciplines every month for a year after which they determine their specialty) and EM spe- cialists. All intubations were performed using direct laryngoscopy as the study emergency department did not have video laryn- goscopy during the study period.

The data source comprised a single-center records of emer- gency airway procedures in the ED from Jan. 1, 2013 to Dec. 31, 2016. Information about each attempt was recorded by the opera- tors, which were EM residents in most cases. If the first operator was not an EM resident (i.e. EM specialists or physicians based out- side the ED) and was unable to enter the data, it was done by EM residents on duty, in which case clinical findings verbally expressed by the operator were recorded by the residents. Data review, management and refinements were performed by an EM specialist.

We included information from first attempt orotracheal intuba- tions of adults aged >=20 years (N = 1154) from Jan. 1, 2013 to Dec. 31, 2016 and subsequent intubations were excluded. We excluded cases of patients directly transferred from the emergency depart- ment to other hospitals without admission and cases of first intu- bations attempted by interns (PGY1). The reason for excluding intern operators was because the study facility does not allow interns to perform the procedure in general, making such case anomalous (number of cases = 8/1162, 0.7%).

Study outcome and the potential predictors of first-pass success

Study outcome was first-pass success rate. The primary objec- tive was to identify factors predictive of the outcome. The sec- ondary objective was to identify independent predictors of the outcome using multiple logistic regression. Factors considered include (1) operator characteristics: Experience level, clinical department (whether based inside or outside ED) and gender, (2) procedure characteristics: reason for intubation (medical vs. trauma), crash airway (cardiac arrest or near cardiac arrest) and use of rapid sequence intubation (RSI, defined by use of paralytics) and (3) patient characteristics: age, sex and various difficult airway indicators including hyoid mental distance, hyothyroidal distance, glottic exposure, facial injury, mouth opening, possible neck exten- sion, obesity, drooling, and tongue swelling.

Statistical analyses

Categorical variables were reported using frequencies and pro- portions. Continuous variables were reported using median and inter quartile range (IQR). T-test, Wilcoxon’s rank-sum test, chi- square test, and Fisher’s exact test were performed as appropriate for comparisons between groups.

Factors predictive of the outcome were identified using univari- able logistic regression. Because multiple associations were tested, Bonferroni-corrected P-values were also reported. A multiple logis- tic regression model of first-pass success was constructed to iden- tify independent predictors. Glottic exposure grade was not considered as an independent variable as we saw it as an outcome rather than an initial condition. Variables with P-value less than 0.2 were included in the model and a stepwise variable selection scheme using Akaike’s information criterion (AIC) was applied to construct a parsimonious model.

The goodness of fit of the model was tested using the Hosmer- Lemeshow test. Interactions were tested up to first order, none of which were significant. Multicollinearity was evaluated using the variance inflation factor and was not found to be problematic. The results of the logistic regression analyses were presented as odds ratios (ORs) with CIs. P-values < 0.05 were considered

Table 1

Baseline characteristics of the study population.

N (%)

Operator-related variables Operator department

ED 1120 (97.4%)

Non-ED* 30 (2.6%)

Operator rank

First-year residency (PGY2) 83 (7.2%)

Second-year residency (PGY3) 482 (41.8%)

Third-year residency (PGY4) 398 (34.5%)

Fourth-year residency (PGY5) 130 (11.3%)

EM specialist 61 (5.3%)

Case-related variables Patient age

<=40 71 (6.2%)

41-65 350 (30.3%)

66-80 461 (39.9%)

>80 272 (23.6%)

Patient sex

Male 708 (61.4%)

Female 446 (38.6%)

Mode of ED arrival

Public ambulance 795 (68.9%)

Private ambulance 229 (19.8%)

Other methods 130 (11.3%)

Reason for emergency intubation

Medical cause 1025 (88.8%)

Traumatic cause 119 (10.3%)

Undetermined 10 (0.9%)

Crash airway 553 (47.9%)

Rapid sequence intubation 644 (55.8%)

Cardiopulmonary resuscitation 328 (28.4%)

Assessed as difficult (subjective assessment) 330 (29.3%) Objective measures of difficulty

Hyomental distance

1 FB	32 (2.8%)
2 FB	468 (41.5%)
3 FB	522 (46.2%)
4 FB Hyothyroidal distance	107 (9.5%)
1 FB	89 (7.9%)
2 FB	524 (46.4%)
3 FB	422 (37.4%)
4 FB	94 (8.3%)
Glottic exposure grade
Entire vocal cords	463 (41.1%)
Part of vocal cords	446 (39.6%)
Epiglottis only	169 (15.0%)
Nonvisualized epiglottis	49 (4.3%)
Facial injury	46 (4.1%)
Restricted neck extension	271 (23.7%)
Restricted mouth opening	512 (44.9%)
Obese with short/thick neck	179 (15.9%)
Stridor, muffled sound or drooling	56 (5.0%)
Swollen tongue	95 (8.4%)

Outcomes

First pass success 974 (84.4%)

Number of trials

1 974 (84.4%)

2 154 (13.3%)

3 23 (2.0%)

4 or more 3 (0.3%)

overall success 1151 (99.7%)

RSI, rapid sequence intubation; EM, emergency medicine; ED, emergency depart- ment; FB, finger breadth.

^* Physicians based outside emergency department; 16 Internal medicine resi- dents (PGY2:3, PGY3:7, PGY4:6), 8 Neurology department residents (PGY2:2, PGY3:6) and 6 Neurosurgery department residents (PGY2:2, PGY3:3, PGY4:1).

significant. All data handling and statistical analyses were per- formed using R-packages version 3.5.1 (R Foundation for Statistical Computing, Vienna, Austria).

Results

Table 1 describes the whole study population (N = 1154). Over- all, more patients were men than women (61.4% vs. 38.6%) with a median age of 72 years (IQR, 58-80 years). Almost half of the cases were crash airway situation (N = 553, 47.9%). First orotracheal intubation was attempted by physicians with a variety of clinical experience (PGY2, PGY3, PGY4, PGY5 and EM specialist). 409 cases (35.4%) were performed by female physicians and only 30 (2.6%) cases were attempted by non-ED physicians. First-pass success was achieved in 974 (84.4%) cases.

The results of univariable logistic regression are reported in Table 2. First-pass success rate was associated with the amount of clinical experience (OR: 2.57, 5.07, 3.52 and 6.34; 95% CI:

1.54-4.24, 2.90-8.82, 1.82-6.99 and 2.47-19.69 for PGY 3, 4 and

5 residents and EM specialist, respectively, relative to PGY 2) and clinical department (non-ED) of operators (OR: 0.13; 95% CI: 0.06-0.27; Table 2, Fig. 1). However, physician gender was not a significant predictor with 83.6% success rate for male vs. 84.8% suc- cess rate for female physicians, with an absolute difference of 1.2% (95% CI, -3.1% to 5.8%, supplementary Table 2). Procedure charac- teristics including cause of emergency intubation (OR: 0.80 and

1.63; 95% CI: 0.50-1.33 and 0.30-30.13 for traumatic cause and other cause, respectively, relative to medical cause), crash airway

(OR: 0.93; 95% CI: 0.68-1.28) and RSI (OR: 1.10; 95% CI: 0.80-

1.51) were not significantly associated with the outcome. As for patient characteristics, most difficult airway indicators including hyomental distance (OR: 0.87, 0.55 and 0.18; 95% CI: 0.43-1.62,

0.28-1.02 and 0.07-0.46 for lengths of 3 FB[finger breadth], 2 FB and 1 FB, respectively, relative to 4 FB) and hyothyroidal distance (OR: 0.71, 0.45 and 0.17; 95% CI: 0.30-1.47, 0.20-0.92 and 0.07-

0.37 for lengths of 3 FB, 2 FB and 1 FB, respectively, relative to 4

FB), glottic exposure grade 0.27, 0.08 and 0.07; 95% CI: 0.16- 0.44, 0.05-0.13 and 0.03-0.13 for exposure level of ‘part of vocal cords’, ‘epiglottis only’ and ‘nonvisualized epiglottis’, respectively, relative to ‘entire vocal cords’), facial injury (OR:0.30; 95% CI: 0.16-0.57), restricted mouth opening (OR:0.32; 95% CI: 0.23-

0.45), restricted neck extension (OR:0.39; 95% CI: 0.28-0.54), obe- sity with short/thick neck (OR:0.47; 95% CI: 0.32-0.70) and swol- len tongue (OR:0.29; 95% CI: 0.19-0.46) were significantly associated with the outcome.

We constructed a multiple logistic regression model to identify independent predictors of first-pass success. Variables with P- value less than 0.2 were included as initial independent variable set and stepwise variable selection scheme using AIC was applied. The results are reported in Table 3. Experience level (OR: 2.93, 5.26,

3.80 and 5.71; 95% CI: 1.62-5.26, 2.80-9.84, 1.81-8.13 and 2.07-

18.67 for PGY 3, 4 and 5 residents and EM specialist, respectively, relative to PGY 2), Non-ED clinical department (OR: 0.10; 95% CI: 0.04-0.25), restricted mouth opening (OR: 0.47; 95% CI: 0.31-

0.72), restricted neck extension (OR: 0.57; 95% CI: 0.39-0.85) and swollen tongue (OR: 0.46; 95% CI: 0.28-0.77) were identified as independent predictors of first-pass success in the final model.

Table 2

Univariable logistic regression models for first-pass success.

	Levels	OR (95% CI)	P	P-trend	P (corrected)	P-trend (corrected)
Operator gender, male Operator rank	Male First year (PGY2) Second year (PGY3)	1.10 (0.79-1.52) Baseline 2.57 (1.54-4.24)	0.587 <0.001	<0.001	1.000 0.007	<0.001
	Third year (PGY4)	5.07 (2.90-8.82)	<0.001		<0.001
	Fourth year (PGY5)	3.52 (1.82-6.99)	<0.001		0.007
Operator department	EM specialist Non-ED*	6.34 (2.47-19.69) 0.13 (0.06-0.27)	<0.001 <0.001		0.011 <0.001
Patient age	<=40	Baseline		0.142		1.000
	41-65	0.84 (0.40-1.62)	0.613		1.000
	66-80	1.01 (0.48-1.94)	0.984		1.000
	>80	1.24 (0.57-2.52)	0.564		1.000
Patient sex Cause of emergency intubation	Male Medical cause Traumatic cause	0.88 (0.63-1.22) Baseline 0.80 (0.50-1.33)	0.447 0.368		1.000 1.000
	Others	1.63 (0.30-30.13)	0.645		1.000
Crash airway	Yes	0.93 (0.68-1.28)	0.656		1.000
Rapid sequence intubation Hyomental distance	Yes 4 FB 3 FB	1.10 (0.80-1.51) Baseline 0.87 (0.43-1.62)	0.573 0.683	<0.001	1.000 1.000	0.001
	2 FB	0.55 (0.28-1.02)	0.072		1.000
Hyothyroidal distance	1 FB 4 FB 3 FB	0.18 (0.07-0.46) Baseline 0.71 (0.30-1.47)	<0.001 0.388	<0.001	0.011 1.000	<0.001
	2 FB	0.45 (0.20-0.92)	0.042		1.000
Glottic exposure grade	1 FB Entire vocal cords Part of vocal cords	0.17 (0.07-0.37) Baseline 0.27 (0.16-0.44)	<0.001 <0.001	<0.001	0.001 <0.001	<0.001
Facial injury	Epiglottis only Nonvisualized epiglottis Yes	0.08 (0.05-0.13) 0.07 (0.03-0.13) 0.30 (0.16-0.57)	<0.001 <0.001 <0.001		<0.001 <0.001 0.004
Restricted mouth opening	Yes	0.32 (0.23-0.45)	<0.001		<0.001
Restricted neck extension	Yes	0.39 (0.28-0.54)	<0.001		<0.001
Obese with short/thick neck	Yes	0.47 (0.32-0.70)	<0.001		0.003
Stridor, muffled sound or drooling Swollen tongue	Yes Yes	1.13 (0.56-2.62) 0.29 (0.19-0.46)	0.751 <0.001		1.000 <0.001

OR, odds ratio; CI, confidence interval; RSI, rapid sequence intubation; EM, emergency medicine; FB, finger breadth.

^* Physicians based outside emergency department; 16 Internal medicine residents (PGY2:3, PGY3:7, PGY4:6), 8 Neurology department residents (PGY2:2, PGY3:6) and 6 Neurosurgery department residents (PGY2:2, PGY3:3, PGY4:1).

Fig. 1. The first-pass success rate of emergency endotracheal intubation with 95% confidence interval by operator characteristics.

Table 3

Multivariable logistic regression model for first-pass success.

Variable	Level	OR (95% CI)	P
Operator rank	First year (PGY2) Second year (PGY3)	Baseline 2.93 (1.62-5.26)	<0.001
	Third year (PGY4) Fourth year (PGY5) EM specialist	5.26 (2.80-9.84) 3.80 (1.81-8.13) 5.71 (2.07-18.67)	<0.001 <0.001 0.002
Operator department Hyothyroidal distance	Non-ED 4 FB 3 FB	0.10 (0.04-0.25) Baseline 0.99 (0.40-2.18)	<0.001 0.977
	2 FB	0.90 (0.36-2.00)	0.802
	1 FB	0.43 (0.15-1.09)	0.087
Facial injury		0.55 (0.27-1.15)	0.101
Restricted mouth opening		0.47 (0.31-0.72)	0.001
Restricted neck extension		0.57 (0.39-0.85)	0.005
Swollen tongue		0.46 (0.28-0.77)	0.003

OR, odds ratio; CI, confidence interval; EM, emergency medicine; FB, finger breadth.

Variables with P-values less than 0.2 in the univariable logistic regression analyses were included as initial variable set. Variable selection scheme using Akaike’s information criterion was applied. The variable Selection process did not induce significant decrease in model performance (AUC 0.745 in full model, AUC 0.745 in the parsimonious model, p for difference = 0.981).

Discussion

In this study, various factors predictive of first-pass success of emergency endotracheal intubation were identified. However, only several operator-related factors (experience level and clinical department) and patient-related factors (restricted mouth open- ing, restricted neck extension and swollen tongue) were identified as independent predictors.

The results of the study support previous findings. The associa- tion between operators’ clinical experience and first-pass success has been reported previously [7,8], and suggests the importance of clinical education and practice in improving the skill. The higher success rate of physicians based in the ED compared to those based outside the ED was expected as the ED-based physicians had more exposure to the procedure compared to the others. However, the effect size of current study seems to be too large. It is possible that cases with unsuccessful initial attempt were more likely to be included in the study if the initial operator was based outside the ED, because successful first-pass success by a non-ED physician could be more likely to be unnoticed by EM residents who were in charge of making a record of the procedure.

As for the patient characteristics, while most of known indica- tors of difficult airway were associated with first-pass success rate, only restricted mouth opening, restricted neck extension and swol- len tongue were found to be independently associated with the outcome. There has been few study reporting the association between individual patient characteristics of difficult airway and first-pass success in emergency endotracheal intubation [9] because of most the previous studies used a binary indicator of dif- ficulty based on original or modified LEMON or DACs (difficult air- way characteristics) criteria [2,10,11].

In a previous Mannequin study, Waddington et al. assessed the influence of gender and experience on the intubation ability of anesthesiology trainees and experts [12]. In the study, the gender difference of success rate (90% for female vs. 97% and male, P = 0.29) and mean time to intubate (24 s for female vs 19 s for male, P = 0.057) in novice trainees were not significant. In addition, the forces generated during intubation reduced with clinical expe- rience. One of the major educational point in the paper was that novice females need not be concerned by the suggestion that they are less able at intubation than their male counterparts. In our cur- rent study, the first-pass success rate of both genders was equiva- lent because average and confidence interval of the difference was small and narrow, respectively. Considering almost half of the intu- bation cases were crash airway situation, we think our study pro- vides a robust clinical evidence of gender equivalence in emergency endotracheal intubation.

This study has several limitations. First, this is a single center study with limited generalizability. Second, we could not keep track of all the intubation attempts in the ED. Especially, intuba- tions with first-pass success by non-ED physicians could be more likely to be missed. Third, all the operators in this study had at least 1 year of Clinical training (PGY1, internship in Korea).

Conclusion

Various operator- and patient-related factors were predictive of first-pass success of emergency endotracheal intubation. Opera-

tors’ experience level and clinical department and restricted mouth opening, restricted neck extension and swollen tongue were iden- tified as independent predictors.

Declaration of Competing Interest

The authors declare no conflicts of interest.

Acknowledgements

We sincerely thank Sarah Hearne and all the others for encour- aging us to revise our original paper.

Appendix A. Supplementary material

Supplementary data to this article can be found online at https://doi.org/10.1016/j.ajem.2019.09.001.

References

1. Mort TC. The incidence and risk factors for cardiac arrest during emergency tracheal intubation: a justification for incorporating the ASA Guidelines in the Remote location. J Clin Anesth 2004;16:508-16. https://doi.org/10.1016/j. jclinane.2004.01.007.
2. Sakles JC, Chiu S, Mosier J, Walker C, Stolz U. The importance of first pass success when performing orotracheal intubation in the emergency department. Acad Emerg Med 2013;20:71-8. https://doi.org/10.1111/ acem.12055.
3. Hasegawa K, Shigemitsu K, Hagiwara Y, Chiba T, Watase H, Brown CA, et al. Association between repeated intubation attempts and adverse events in emergency departments: an analysis of a Multicenter prospective observational study 749-754.e2. Ann Emerg Med 2012;60. https://doi.org/ 10.1016/j.annemergmed.2012.04.005.
4. Kim J, Kim K, Kim T, Rhee J, Jo Y, Lee J, et al. The clinical significance of a failed initial intubation attempt during emergency department resuscitation of out- of-hospital cardiac arrest patients. Resuscitation 2014;85:623-7. https://doi. org/10.1016/j.resuscitation.2014.01.017.
5. Adnet F, Borron SW, Racine SX, Clemessy J, Fournier J, Plaisance P, et al. The intubation difficulty scale (IDS). Anesthesiology 1997;87:1290-7. https://doi. org/10.1097/00000542-199712000-00005.
6. Mallampati RS, Gatt SP, Gugino LD, Desai SP, Waraksa B, Freiberger D, et al. A clinical sign to predict difficult tracheal intubation; a prospective study. Can Anaesth Soc J 1985;32:429-34. https://doi.org/10.1007/bf03011357.
7. Sagarin MJ, Barton ED, Chng Y-M, Walls RM. on behalf of the Investigators N. Airway management by US and Canadian emergency medicine residents: a multicenter analysis of more than 6,000 endotracheal intubation attempts. Ann Emerg Med 2005;46:328-36. https://doi.org/10.1016/j. annemergmed.2005.01.009.
8. Platts-Mills TF, Campagne D, Chinnock B, Snowden B, Glickman LT, Hendey GW. A comparison of glidescope video laryngoscopy versus direct laryngoscopy intubation in the emergency department. Acad Emerg Med 2009;16:866-71. https://doi.org/10.1111/j.1553-2712.2009.00492.x.
9. Soyuncu S, Eken C, Cete Y, Bektas F, Akcimen M. Determination of difficult intubation in the ED. Am J Emerg Med 2009;27:905-10. https://doi.org/ 10.1016/j.ajem.2008.07.003.
10. Hagiwara Y, Watase H, Okamoto H, Goto T, Hasegawa K, Investigators J. Prospective validation of the modified LEMON criteria to predict difficult intubation in the ED. Am J Emerg Med 2015;33:1492-6. https://doi.org/ 10.1016/j.ajem.2015.06.038.
11. Kim C, Kang H, Lim T, Choi B, Shin Y, Choi H. What factors affect the success rate of the first attempt at endotracheal intubation in emergency departments?. Emerg Med J 2013;30:888. https://doi.org/10.1136/emermed- 2012-201708.
12. Waddington M, Paech M, Kurowski I, Reed C, Nicholls G, Guy D, et al. The influence of gender and experience on intubation ability and technique: a manikin study. Anaesth Intens Care 2009;37:791-801. https://doi.org/ 10.1177/0310057×0903700502.