a b s t r a c t

Objective: Difficult airway situations, such as trismus and neck rigidity, may prohibit standard midline orotracheal intubation. An alternative route of intubation from the retromolar space using a fiberoptic scope or rigid intuba- tion stylet has been reported. There is no study investigating the applicability of retromolar intubation using a video intubating stylet. This study comparatively analyzed difficult airway management using a video intubating stylet in the retromolar and standard midline approaches.

Methods: A randomized crossover manikin study was conducted between January 2021 and June 2021 at a ter- tiary teaching hospital. Thirty-six emergency medicine residents and attending physicians were enrolled, and all participated in an educational course regarding video intubating stylet in standard midline and retromolar ap- proaches. Then, they performed both intubation approaches in a randomized order on a manikin seven times with different airway settings each time. The duration of successful intubation, first attempt success rate, overall success rate, number of attempts, and self-reported difficulty were recorded and compared.

Results: Thirty-six emergency physicians were included in the study. Compared with the standard midline ap- proach, the use of the retromolar approach significantly reduced the duration of successful intubation in difficult airway scenarios such as limited mouth opening and neck rigidity with (44.77 [28.58-63.65] vs. 120 [93.86-120] s, p < 0.001) and without tongue edema (31.5 [22.57-57.74] vs. 44.72 [36.23-65.34] s, p = 0.012). Furthermore, the retromolar approach increased the first attempt success rate in scenarios of limited mouth opening and neck rigidity with (91.67% vs. 16.67%, p < 0.001) and without (97.22% vs. 72.22%, p = 0.012) tongue edema. The self- reported difficulty was also significantly lower with the retromolar approach than with the standard approach in the above two scenarios.

Conclusions: The retromolar approach for intubation using a video intubating stylet may be a promising choice for selected patients with a combination of difficult airway features such as limited mouth opening, neck rigidity, and edematous tongue.

(C) 2022

1. Introduction

Difficult airway intubation in emergency settings is not as well eval- uated as intubation in elective conditions, making tracheal intubation in these settings one of the most substantial challenges for emergency physicians. Only one-third of non-cardiac arrest patients in the emer- gency department (ED) undergo proper physiognomic assessments such as the Mallampati score assessment, neck mobility test, and thyromental distance measurement before intubation [1,2]. The inci- dence of failed intubation is higher in the ED, intensive care units, and

* Corresponding authors at: Department of Emergency Medicine, Ditmanson Medical Foundation Chia-Yi Christian Hospital, No. 539, Zhongxiao Road, East District, Chiayi City 600, Taiwan.

E-mail addresses: tshi33@gmail.com (M.-J. Tsai), cimbatw@gmail.com (K.-Y. Cheng).

¹ Ming-Jen Tsai and Kai-Yuan Cheng contributed equally to this work.

pre-hospital settings (approximately 1 in 50-100) than in elective set- tings (approximately 1 in 1000-2000) [3].

Difficult airway problems can be classified into anterior, middle, and posterior columns based on anatomy [4]. The anterior column refers to the space surrounded by the teeth, Hyoid bone, and bilateral tempo- romandibular joints. Any problems involving decreased volume or compliance of the submandibular space and the range of temporoman- dibular joint movement may interfere with oral intubation. Common anterior column problems include large tongue, short inter-incisor dis- tance (limited mouth opening), oral-facial infection, trauma, tumor, and micrognathia. The middle column represents the airway passage route. Problems in the middle column include obstruction by foreign bodies, tumors, lingual tonsils, and epiglottitis. The posterior column represents the range of motion of the neck. Common pathologic problems of this column include Ankylosing spondylitis, cervical spine injury requiring in-line stabilization, and rheumatoid arthritis [4].

https://doi.org/10.1016/j.ajem.2022.02.008

0735-6757/(C) 2022

Unanticipated difficult airway can be encountered at any time in the ED. For patients with trismus or limited mouth opening, nasal endotra- cheal intubation under flexible fiberoptic endoscopy should be consid- ered [5]. However, a fiberoptic endoscope is not always immediately available in the ED settings. Accurate use of a flexible fiberoptic endo- scope also depends on the experience of the practitioner. Improper op- eration can easily damage the equipment. Further, nasal stenosis, history of basilar Skull fractures, coagulopathy, and proposed surgery in- volving the nasal passage are contraindications concomitant with se- vere trismus [6]. Hence, emergency cricothyroidotomy is usually chosen as the last resort under the abovementioned conditions.

Since 2012, flexible fiber-optic orotracheal intubation through the retromolar space has been reported to be an alternative to nasal intuba- tion or tracheostomy for patients with conditions such as trismus, max- illofacial trauma, nasal stenosis, and oncological diseases require oral surgery [6-12]. The retromolar space is present in all individuals and can be visualized by retracting the skin at the angle of the mouth. It is bound posteriorly by the ascending ramus of the mandible, superiorly by the maxillary tuberosity, and anteriorly by the molars (Fig. 1) [7].

As alternatives to flexible fiberoptic scopes, a variety of rigid video intubating stylets, such as Bonfils, Trachway, or Tuoren Kingtaek video intubating stylets, which incorporate fiberoptic imaging elements in an intubation stylet, are now available for use in EDs [13]. Manikin stud- ies have proven that a rigid video intubating stylet provided faster, eas- ier, and less traumatic intubation than the traditional Macintosh laryngoscope in difficult intubation scenarios [14]. There are several case reports on the use of the Bonfils rigid intubation stylet to intubate patients with trismus via the retromolar space [15,16]. However, no study has investigated or compared the standard approach from the midline and retromolar approaches using a video rigid intubating stylet in difficult airway scenarios. In daily practice, we find that a video intubating stylet from the midline approach cannot overcome a difficult airway due to anterior column problems [4], especially in cases of tris- mus. A laryngoscope is required in some cases to overcome the enlarged tongue and facilitate the passage of the video intubating stylet. How- ever, a video intubation stylet can be inserted from the retromolar space through the posterolateral pharyngeal wall to access the larynx and glottis. Theoretically, this maneuver can overcome most cases of difficult airway caused by anterior column problems.

To our knowledge, there is no study investigating the applicability

and difficulty of retromolar intubation. We hypothesize that retromolar intubation will have a higher first attempt success rate and shorter intu- bation time in some specific difficult airway scenarios compared with midline intubation using a video intubation stylet. Hence, we conducted

a manikin study to evaluate and compare retromolar and standard mid- line intubation using a video intubation stylet in different scenarios of difficult airway.

1. Methods
   1. Study design and setting

A prospective, randomized, crossover manikin study was conducted in the Ditmanson Medical Foundation Chia-Yi Christian Hospital, a tertiary referral and teaching hospital. The study was approved by the Institutional Review Board of the Ditmanson Medical Foundation Chia-Yi Christian Hospital (CYCH-IRB 2020145).

• 1. Study participants

After receiving approval from the Institutional Review Board, the residents and attending physicians working in the ED were requested to participate in this study.

• 1. Intubation and simulation devices

All intubations were performed using a video intubating stylet (Tuoren Kingtaek Video Intubating stylet; Henan Tuoren Medical De- vice Co., China) (Fig. 2). The video intubating stylet is a reusable, rigid, straight device with a malleable tip, and an adjustable monitor attached to the rechargeable handle. The stylet on which the tracheal tube was mounted was approximately 32 cm in length, with an external diameter of 5.0 mm. At the distal end of the stylet, there is a light source for illu- mination with an angle of view of 45?. The simulation manikin corresponded to the Difficult Airway Management (DAM) simulator (MW13, KYOTOKAGAKU, Japan). This manikin can be used to simulate and combine different conditions of difficult airway, including neck ri- gidity, trismus, tongue edema, and laryngospasm. Trismus was further classified into normal, moderate, and difficult levels, representing different mouth-opening limitations, and a specific level (maximal inter-incisor distance: 2.5 cm) was applied in this study. A cuffed 7.0-mm endotracheal tube was used for all intubations.

• 1. Study protocol

This study was divided into two parts. The first part was an educa- tional course, and the second part was the experimental course. We ad- ministered a questionnaire before the educational course which

Fig. 1. Retromolar space is an alternative space to be used for orotracheal intubation.

Fig. 2. Tuoren Kingtaek Video Intubating stylet.

showed that most of the participants had performed intubation using the video intubation stylet less than five times in the past six months.

• 1. The educational course

In the educational course, all participants received 3 h of learning to familiarize themselves with the Tuoren Kingtaek video intubating

stylet. In this course, all participants intubated a normal manikin with- out a difficult airway. We demonstrated and trained all participants using the following two intubation methods and instructions:

1. Video intubating stylet with standard approach from the midline of the mouth

We inserted a video intubating stylet with a preloaded endotracheal tube along the midline of the tongue towards the pharynx. The oropha- ryngeal structure was visualized on the monitor, and the epiglottis was seen while we advanced the intubating stylet along the midline of the tongue. The tip of the stylet was moved inferiorly to cross the epiglottis, and the vocal cords were observed. The stylet was advanced into the trachea until it was located at a short distance below the vocal cord, fol- lowing which the tracheal tube was slid into the trachea to the proper depth [17].

1. Video intubating stylet with retromolar approach

After lifting the corner of the mouth, we inserted a video intubating stylet with a preloaded endotracheal tube into the oral cavity through the right retromolar space. The tip of the stylet was moved through the posterolateral pharyngeal wall to visualize the oropharyngeal struc- ture and epiglottis. By rotating clockwise and tilting the tip of the video intubation stylet, we could access the glottis. After the tip of the stylet was placed at the glottic entrance, the endotracheal tube was advanced over the rigid stylet and smoothly inserted into the trachea [18].

After the demonstration and training, each participant was asked to perform tracheal intubation using the two techniques mentioned above on a Laerdal(R) airway simulator manikin (Laerdal Medical, Wappingers Falls, NY) with a normal airway in the supine position. Participants who continuously and successfully performed intubation five times using each technique were included for participation in the experimental course.

Fig. 3. Randomization of participants to different approach and scenarios.

• 1. The experimental course

Once all participants completed the educational course, they were asked to perform difficult airway intubation on the DAM simulator using the two intubation methods mentioned above in the following scenarios:

Scenario A: Normal airway. Scenario B: Limited mouth opening. Scenario C: Tongue edema.

Scenario D: Limited mouth opening with tongue edema. Scenario E: Neck rigidity.

Scenario F: Limited mouth opening and neck rigidity.

Scenario G: Tongue edema with limited mouth opening and neck rigidity.

All participants performed intubations for the seven different airway scenarios mentioned above using the retromolar and standard midline in- tubation techniques. The participants were asked to complete the intuba- tion by insufflating the cuff of the tube, attaching a bag valve mask, and providing one breath to ventilate the lungs of the simulator. Up to three at- tempts were made to perform intubation using each intubation method in each scenario. The intubation procedure was closely monitored by one of the investigators to ensure that all techniques were performed in an ade- quate manner, and we measured both the intubation time and success rate.

• 1. The randomization

The sequence of difficult airway scenarios and type of intubation method were assigned randomly throughout the study (Fig. 3). The

Table 1

Data from intubation in scenarios A to D.

Parameter	Standard	approach (n = 36)	Retromolar	approach (n = 36)	P value
Scenario A: normal airway First attempt success (n, %)	35	(97.22)	35	(97.22)	1
Overall success (n, %)	36	(100)	36	(100)	1
Duration of successful intubation attempt (s) Median (IQR)	22.43	(17.95-30.28)	25.06	(20.22-34.23)	0.138
Mean (SD)	26.83	(13.4)	30.62	(17.12)	0.164
Number of intubation attempts (n, %) 1	35	(97.22)	35	(97.22)	1
2	1	(2.78)	1	(2.78)
3	0	(0)	0	(0)
Mallampati score	1.5	(1-2)	1	(1-2)	0.523
Cormack-Lehane classification	1	(1-1)	1	(1-2)	0.206
Difficulty	2	(1-2.25)	2	(1-3)	0.173
Scenario B: limited mouth opening First attempt success (n, %)	35	(97.22)	36	(100)	1
Overall success (n, %)	36	(100)	36	(100)	1
Duration of successful intubation attempt (s) Median (IQR)	25.83	(20.72-34.18)	29.22	(19.95-33.98)	0.85
Mean (SD)	30.3	(14.23)	29.23	(10.85)	0.661
Number of intubation attempts (n, %) 1	35	(97.22)	36	(100)	0.314
2	0	(0)	0	(0)
3	1	(2.78)	0	(0)
Mallampati score	3	(2-3)	3	(2-4)	0.166
Cormack-Lehane classification	2	(1-2)	1.5	(1-2)	0.564
Difficulty	3	(2-4)	3	(2-4)	0.075
Scenario C: tongue edema First attempt success (n, %)	35	(97.22)	36	(100)	1
Overall success (n, %)	36	(100)	36	(100)	1
Duration of successful intubation attempt (s) Median (IQR)	25.96	(19.44-30.78)	24.21	(20.11-34.22)	0.414
Mean (SD)	27.96	(13.82)	29.93	(16.99)	0.321
Number of intubation attempts (n, %) 1	35	(97.22)	36	(100)	0.314
2	1	(2.78)	0	(0)
3	0	(0)	0	(0)
Mallampati score	2	(1.75-3)	2	(1-2)	0.066
Cormack-Lehane classification	1	(1-2)	1	(1-2)	0.248
Difficulty	3	(2-3)	3	(2-4)	0.15
Scenario D: limited mouth opening and tongue edema First attempt success (n, %)	34	(94.44)	34	(94.44)	1
Overall success (n, %)	36	(100)	36	(100)	1
Duration of successful intubation attempt (s) Median (IQR)	31.09	(21.81-39.74)	29.63	(19.36-38.01)	0.084
Mean (SD)	35.35	(18.04)	30.19	(12.12)	0.081
Number of intubation attempts (n, %) 1	34	(94.44)	34	(94.44)	1
2	2	(5.56)	2	(5.56)
3	0	(0)	0	(0)
Mallampati score	3	(2-4)	3	(2-4)	0.583
Cormack-Lehane classification	2	(1-2)	2	(1-2)	0.405
Difficulty	3.5	(2-5)	3	(2-4)	0.762

Data are presented as median (interquartile range [IQR]), mean (standard deviation [SD]), or number (percentage).

participants drew shuffled cards labeled with the type of intubation method to determine the sequence of intubation methods. Once the se- quence of intubation methods was assigned, the participants drew cards to determine the sequence of airway scenarios. After completing the in- tubations of all the airway scenarios using the first intubation method assigned to them, participants drew cards again to determine the se- quence of scenarios for the second intubation method. The participants were required to perform all the intubations (14 times in total) consec- utively. The participants and recorder were blinded to the airway sce- narios; however, they could not be blinded to the intubation methods. We only included participants who completed the experimental course into the final analysis. The participants performed intubation separately and were not allowed to discuss the intubation obstacles or successes during or after the experiment.

• 1. Measurement

We compared the difference between the standard and retromolar approaches using a video intubation stylet in one normal and six diffi- cult airway scenarios. We set the primary endpoints as the duration re- quired for successful intubation and the first-attempt success rate. The secondary endpoints were overall success rate, number of attempts re- quired to achieve successful intubation, and degree of difficulty. The du- ration of an intubation attempt was defined as the time when the

operator picks up the assigned device until inflation of the lungs is con- firmed after connecting the endotracheal tube to a bag valve mask. The intubation attempt was deemed as failed in the following cases: mani- kin teeth dislodged, any attempt of more than 120 s, esophageal intuba- tion, or when the operator gave up [19]. A maximum of three intubation attempts were allowed. We measured and presented the duration and number of attempts required for successful intubation. If the participant failed to intubate in three attempts, we recorded the duration of intuba- tion as 120 s and the number of attempts as four. Before performing in- tubation in each scenario, each participant was asked to evaluate and record the Mallampati score. At the end of each scenario, each partici- pant had to report the best glottic view visualized by the instrument using the Cormack-Lehane classification and score the degree of diffi- culty of each intubation method on a rating scale with scores ranging from 0 (extremely easy) to 10 (extremely difficult) [14,20].

• 1. Statistical analysis

We performed a pilot study (n = 6) and chose the mean intubation time corresponding to the difficult airway scenario F to determine the sample size. The mean intubation times using the video intubation stylet with the standard and retromolar approaches were 64.2 s +- 35.8 s and

28.5 s +- 14.8 s, respectively. A two-sided test size of 5% and a power of 95% were applied (calculated effect size, dz. = 1.145762). We found

Table 2

Data from intubation in scenarios E to G.

Parameter Standard approach (n = 36) Retromolar approach (n = 36) P value

Scenario E: neck rigidity
First attempt success (n, %)	33	(91.67)	34	(94.44)	1
Overall success (n, %)	35	(97.22)	35	(97.22)	1
Duration of successful intubation attempt (s)
Median (IQR)	33.89	(27.58-49.63)	29.89	(23.04-45.62)	0.057
Mean (SD)	43.13	(24.52)	38.59	(25.20)	0.243
Number of intubation attempts (n, %)					0.84
1	33	(91.67)	34	(94.44)
2	2	(5.56)	1	(2.78)
3	1	(2.78)	1	(2.78)
Mallampati score	3	(2-4)	3	(2-4)	0.932
Cormack-Lehane classification	2	(1-2)	2	(2-2)	0.806
Difficulty	4	(3-6)	3	(2-5)	0.164
Scenario F: limited mouth opening and neck rigidity
First attempt success (n, %)	26	(72.22)	35	(97.22)	0.012
Overall success (n, %)	35	(97.22)	36	(100)	1
Duration of successful intubation attempt (s)
Median (IQR)	44.72	(36.23-65.34)	31.5	(22.57-57.74)	0.012
Mean (SD)	53.84	(26.11)	41.83	(25.00)	0.025
Number of intubation attempts (n, %)					0.047
1	25	(69.44)	35	(97.22)
2	8	(22.22)	1	(2.78)
3	2	(5.56)	0	(0)
4 (unsuccessful intubation for 3 times)	1	(2.78)	0	(0)
Mallampati score	4	(4-4)	4	(4-4)	0.317
Cormack-Lehane classification	2	(2-2)	2	(2-2)	0.096
Difficulty	6	(5-7)	4	(3-5)	<0.001
Scenario G: limited mouth opening, neck rigidity and tongue edema
First attempt success (n, %)	6	(16.67)	33	(91.67)	<0.001
Overall success (n, %)	11	(30.56)	36	(100)	<0.001
Duration of successful intubation attempt (s)
Median (IQR)	120	(93.86-120)	44.77	(28.58-63.65)	<0.001
Mean (SD)	102.6	(31.11)	47.71	(23.71)	<0.001
Number of intubation attempts (n, %)					<0.001
1	6	(16.67)	33	(91.67)
2	3	(8.33)	2	(5.56)
3	2	(5.56)	1	(2.78)
4 (unsuccessful intubation for 3 times)	25	(69.44)	0	(0)
Mallampati score	4	(4-4)	4	(4-4)	0.317
Cormack-Lehane classification	2.5	(2-4)	2	(2-2)	<0.001
Difficulty	9	(8-10)	5	(3-7)	<0.001

Data are presented as median (interquartile range [IQR]), mean (standard deviation [SD]), or number (percentage).

that a total of 13 participants would be required to find a significant dif- ference. To compensate for dropouts, we included at least 14 participants under different grades of emergency physicians in this study. The data obtained from the pilot study was also included in the final results.

All statistical analyses were performed using SPSS (version 20.0; SPSS, Chicago, IL, USA). For continuous data, the distribution was tested using the Kolmogorov-Smirnov test. For dependent samples (standard approach vs. retromolar approach), data of the duration of the successful intubation and the degree of difficulty were analyzed using paired t-test (normal dis- tribution) or Wilcoxon test (non-normal distribution) and presented as mean +- standard deviation or medians (interquartile range), as appropri- ate. Data regarding the rate of successful intubations were analyzed using McNemar’s test. Regarding independent samples (residents vs. attending physicians), Mann-Whitney U test (presented as medians [interquartile range]) and Fisher’s exact test were used for continuous and categorical variables, respectively. Moreover, Kruskal-Wallis test was used to analyze the continuous data between different scenarios. For all statistical analyses, a two-tailed p value of <0.05 was considered significant.

1. Results

Between January 2021 and June 2021, a total of 36 emergency phy- sicians including 14 residents and 22 attending physicians participated

in the study (Fig. 3). All participants completed the full course of the study, and none of them were excluded from the final analysis. On com- paring the retromolar and standard approaches using a video intubation stylet, the first attempt success rate and overall success rate did not show significant differences in scenarios A, B, C, D, and E (Tables 1 and 2). In scenario F, which combined limited mouth opening and neck ri- gidity, a significantly higher first attempt success rate was found in the retromolar approach than in the standard approach (97.22% vs. 72.22%, p = 0.012). However, the overall success rate was not different between the two intubation methods (100% vs. 97.22%, p = 1) (Table 2). In scenario G, which combined limited mouth opening, neck rigidity, and tongue edema, the retromolar approach showed sig- nificantly higher first attempt success (91.67% vs. 16.67%, p < 0.001) and overall success rates (100% vs. 30.56%, p < 0.001) than the standard approach (Table 2).

Regarding the duration required for successful intubation, there was no significant difference in the intubation time between the two intubation methods in scenarios A, B, C, D, and E (Tables 1 and 2, Fig. 4a). However, the intubation time was significantly shorter in sce- narios F (median: 31.5 vs. 44.72 s, p = 0.012) and G (median: 44.77 vs. 120 s, p < 0.001) with the retromolar approach than with the stan- dard approach (Table 2, Fig. 4a). There was a significant difference in the intubation time between different scenarios, regardless of the

Fig. 4. Box plot of (a) intubation time and (b) difficulty between standard and retromolar approaches in different difficult airway scenarios. *p < 0.05, **p < 0.001 compared to scenario A using the standard approach. #p < 0.05 compared to scenario A using the retromolar approach. +p < 0.05 compared to scenario A to D using the retromolar approach. ?p < 0.001 compared to scenario A to D using the standard approach.

Table 3

The first attempt success rate and intubation time between emergency residents and attending physicians

Method Parameter Scenario Resident (n = 14) Attending physician (n = 22) P value

Standard approach	First attempt success	A	14	(100)	21	(95.46)	1
	(n, %)	B	14	(100)	21	(95.46)	1
		C	14	(100)	21	(95.46)	1
		D	14	(100)	20	(90.91)	0.511
		E	14	(100)	19	(86.36)	0.267
		F	9	(64.29)	17	(72.27)	0.462
		G	1	(7.14)	5	(22.73)	0.37
	Intubation time (s)	A	23.37	(19.45-29.83)	21.86	(17.82-34.18)	0.911
		B	24.92	(19.61-31.75)	26.27	(21.09-35.39)	0.413
		C	26.53	(20.53-27.72)	24.78	(18.54-32.45)	0.697
		D	27.82	(20.07-35.89)	35.59	(22.97-48.41)	0.267
		E	33.21	(27.51-48.28)	34.28	(30.71-51.00)	0.721
		F	46.91	(38.37-67.52)	42.94	(34.77-60.65)	0.413
		G	120	(87.19-120)	120	(118.97-120)	0.735
Retromolar approach	First attempt success	A	14	(100)	21	(95.46)	1
	(n, %)	B	14	(100)	22	(100)	1
		C	14	(100)	22	(100)	1
		D	13	(92.86)	21	(94.46)	1
		E	14	(100)	20	(90.91)	0.511
		F	13	(92.86)	22	(100)	0.389
		G	14	(100)	19	(86.36)	0.267
	Intubation time (s)	A	24.51	(20.85-36.62)	25.2	(19.49-33.67)	0.835
		B	26.95	(18.06-33.53)	29.67	(22.09-33.57)	0.381
		C	24.3	(19.04-34.33)	24.21	(20.81-33.96)	0.689
		D	30.23	(19.73-43.95)	27.34	(19.97-37.79)	0.49
		E	26.3	(20.50-43.34)	30.2	(25.67-44.29)	0.307
		F	28.42	(22.22-59.16)	35.69	(25.16-49.84)	0.833
		G	42.58	(25.51-66.96)	44.77	(31.30-60.76)	0.371

Data are presented as number (percentage) or median (IQR).

approach (p < 0.001). With the standard approach, the intubation time in scenarios F and G was significantly higher than that in sce- nario A (both p < 0.001). With the retromolar approach, only the in- tubation time in scenario G was significantly higher than that in scenario A (p < 0.05).

The number of intubation attempts was not significantly different between the two intubation methods in scenarios A-E. However, there were significantly fewer attempts with the retromolar approach than with the standard approach intubation in scenarios F (p = 0.047) and G (p < 0.001) (Table 2).

The Mallampati score, which was evaluated by the participants be- fore intubation in each scenario, was not significantly different between the two intubation methods in all scenarios. This indicates that regard- less of which intubation method was performed in each scenario, the condition of the manikin was consistent. A significantly higher Mallampati score was found in scenarios F and G (both medians: 4 [4-4]) than in scenarios A to E (p < 0.001) (data not shown).

Regarding the best glottic view evaluated by the Cormack-Lehane classification, the results showed a significantly lower Cormack- Lehane classification in the retromolar approach group than in the standard approach group in scenario G (p < 0.001) (Table 2). There was no significant difference in the Cormack-Lehane classification be- tween the retromolar and standard approaches in scenarios A-F (Table 1 and 2).

The degree of difficulty reported by the participants was signifi- cantly lower with the retromolar approach than with the standard approach in scenarios F (median: 4 [3-5] vs. 6 [5-7], p < 0.001)

and G (median: 5 [3-7] vs. 9 [8-10], p < 0.001; Table 2, Fig. 4b). A significantly higher degree of difficulty was reported in scenarios F and G than in scenarios A-D, especially with the standard approach (p < 0.001).

To understand the difference between residents and attending physicians in terms of performing the two methods of intubation, we compared the first-attempt success rate and intubation time between residents and attending physicians. The results showed no difference both for the first attempt success rate and intubation time in all scenarios (Table 3).

1. Discussion

A previous study demonstrated that a video intubating stylet en- sured faster and easier intubation with less dental trauma in a manikin with reduced mouth opening compared with the traditional Macintosh laryngoscope [14]. A previous case report also demonstrated the feasi- bility of a video intubation stylet via the standard midline approach in patients with posterior column problems of ankylosing spondylitis [17]. In our study, we simulated difficult airway problems involving the anterior and posterior columns to further evaluate the applicability of the video intubation stylet with the retromolar approach compared to the standard midline approach. In scenarios B (limited mouth open- ing), C (tongue edema), and D (limited mouth opening and tongue edema), which are related to purely anterior column problems, both midline and retromolar approaches demonstrated similar glottic views, intubation success rates, and difficulty. When we added poste- rior column problems to the difficult airway scenario (e.g., E [neck ri- gidity], F [limited mouth opening and neck rigidity], and G [limited mouth opening, tongue edema, and neck rigidity]), significant differ- ences could be found between the two approaches. Significantly higher first attempt success rate, shorter intubation time, and lower difficulty were found with the retromolar intubation, especially in scenarios with combined anterior and posterior column problems (scenarios F and G, Table 2).

In emergency settings, difficult airway combined with different column problems may be encountered, such as facial trauma with cervi- cal spine injury, complex maxillofacial trauma with the need for intermaxillary fixation, and severe trismus [21,22]. It may be difficult to achieve successful intubation using a standard oral route with the midline approach. Hence, several case reports have demonstrated that non-invasive oral intubation using a flexible fiberoptic scope through the retromolar space may be an alternative to emergency cricothyroidotomy for managing difficult airway [6,8-12,22]. However, considering the high cost and expertise required for using a fiberoptic scope, using a video intubation stylet through retromolar intubation may be a better solution for managing these patients in an emergency setting, as demonstrated in this study.

Fig. 5. Angle of tilting in scenario G (tongue edema with limited mouth opening and neck rigidity) via the (a) midline approach and (b) retromolar approach.

An important finding in this study is that in scenario G, which in- volves a combination of limited mouth opening, neck rigidity, and tongue edema, the retromolar approach could provide a better glottic view for performing intubation than the midline approach (Table 2). Ac- cording to our observation, the difficulty associated with intubation in this scenario is due to the tilted angle of the video intubating style; the angle was insufficient and limited. In the midline approach, in order to move the tip of stylet close to the glottis, the stylet would have to press against the incisor teeth. This made it difficult to pass the endotracheal tube into the trachea. Furthermore, the insufficient, tilted angle also contributed to failed intubation. If the participants tilted the stylet too much, the incisor teeth would dislodge. However, this ob- stacle could be overcome by the retromolar approach because the angle of tilting can be increased and is not limited through the retromolar space. The angle between the intubating stylet and the horizon was ap- proximately 75? in scenario G, while using the standard midline ap- proach (Fig. 5a). The angle could be augmented to 90? with retromolar intubation (Fig. 5b), resulting in the intubating stylet being positioned horizontal to the laryngeal axis, thus facilitating intubation. Applicability, difficulty, and ease of learning are important factors for teaching and learning new procedures. The viewing angle of the anat- omy of the oropharynx from the route of the retromolar space is oblique and different from the view obtained in the midline approach. After learning, experienced (attending physician) and inexperienced partici- pants (residents) achieved similar first attempt intubation success rates and successful intubation durations (Table 3). This indicates that the use of a video intubation stylet via the retromolar approach can be learned through adequate training. After this training and experimental course, we successfully intubated a patient with out-of-hospital cardiac arrest with trismus and severe neck rigidity using the retromolar intu- bation with a video intubation stylet (Supplementary video). The course was performed without interruption of the cardiopulmonary resuscita- tion. This also prevented the need for emergency cricothyroidotomy in

this patient.

To our knowledge, this is the first study to investigate and compare the retromolar approach and standard midline approaches of intubation of a video stylet in a simulator with multiple difficult airway features. The retromolar approach was found to be better in difficult airway sce- narios which combined anterior and posterior airway column problems. The participants also reported that the retromolar approach had a lower difficulty level than the standard approach.

This study has several limitations. First, as with any manikin study, airway scenarios in a manikin cannot simply be translated into a clinical setting. Further studies are needed to evaluate the feasibility of our re- sults in human beings, who are obviously more complicated than man- ikins. Second, Difficult airway mannequins are designed to increase the difficulty of traditional midline orotracheal intubation performed with a

laryngoscope. The retromolar anatomy on the difficult airway manne- quins may not correspond well with actual human anatomy. Third, re- garding the trismUS setting in the DAM simulator, we were allowed to select approximately 2.5 cm of inter-incisor distance in the difficult level. There was no further severe trismus setting that could be applied. Hence, this simulator may not have mimicked real trismus (inter-inci- sor space less than 1 cm) [6,7]. Fourth, patients with middle column problems such as tumors, lingual tonsillar hypertrophy, or massive vomitus could not be simulated in our study. Hence, the feasibility of the retromolar approach in cases of middle column problems is yet to be investigated. In clinical practice, a few patients may experience mas- sive vomitus during intubation; hence, suction-assisted laryngoscopy airway decontamination should be adopted to facilitate glottic view and intubation [23]. While performing retromolar intubation using a video intubating stylet, massive emesis is a critical problem with respect to navigating the glottic view. Theoretically, we can insert a large bore suction catheter such as a DuCanto catheter in the opposite retromolar space to assist in suction before inserting the video intubating stylet. Fi- nally, a computerized tomography study with 3D reconstruction of 80 retromolar spaces showed no difference between the left and right retromolar spaces, and the space was suitable for an endotracheal tube size of <=7.0 mm when the teeth were closed [24]. In this study, we did not investigate the applicability of retromolar intubation using a large endotracheal tube.

1. Conclusions

Previously, the retromolar intubation method has been attempted successfully for difficult airway conditions such as limited mouth open- ing, trismus, maxillofacial trauma, and oral surgery [8,9,15,16,21,22,25]. Although there is difficulty in performing intubation via this route, it could be overcome by using a video intubating stylet. In this manikin study, we found that in selected patients with limited mobility of the mouth and neck with or without edematous tongue, intubation from the retromolar space using a video intubation stylet was better than standard midline intubation. Further prospective studies including ac- tual patients are required to document the applicability of retromolar intubation with a video intubation stylet either in the setting of the ED or in that of the operation theatre.

Supplementary data to this article can be found online at https://doi. org/10.1016/j.ajem.2022.02.008.

Source(s) of support

This research was funded by Ditmanson Medical Foundation Chia-Yi Christian Hospital Research Program, grant number R110-013.

Credit authorship contribution statement

Yen-Yu Chen: Writing – original draft, Investigation. Jih-Chun Lin: Investigation. Ming-Jen Tsai: Writing – review & editing, Supervision, Methodology, Investigation, Conceptualization. Kai-Yuan Cheng: Writ- ing – review & editing, Writing – original draft, Methodology, Investiga- tion, Conceptualization.

Declaration of Competing Interest

The authors declare no conflicts of interest directly or indirectly con- nected with this manuscript.

Acknowledgments

We thank Mr. Andy P. Tsai (Indiana University School of Medicine) for assisting in the production of Fig. 1 created with BioRender.com.

References

1. Levitan RM, Everett WW, Ochroch EA. Limitations of difficult airway prediction in patients intubated in the emergency department. Ann Emerg Med. 2004;44: 307-13. https://doi.org/10.1016/j.annemergmed.2004.05.006.
2. Mallampati SR, Gatt SP, Gugino LD, 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.
3. Cook TM, MacDougall-Davis SR. Complications and failure of airway management. Br J Anaesth. 2012;109(Suppl. 1):i68-85.
4. Greenland KB. Airway assessment based on a three column model of direct laryn- goscopy. Anaesth Intensive Care. 2010;38:14-9. https://doi.org/10.1177/ 0310057X1003800104.
5. Akasapu KR, Wuduru S, Padhy N, Durga P. Unanticipated cannot intubate situation due to difficult mouth opening. J Anaesthesiol Clin Pharmacol. 2015;31:123-4. https://doi.org/10.1093/bja/aes393.
6. Truong AT, Truong DT. Retromolar flexible fiber-optic orotracheal intubation: a novel alternative to nasal intubation and tracheostomy in severe trismus. Head Neck. 2015;37:423-4. https://doi.org/10.1002/hed.23732.
7. Togioka BM, Pinkert V, Geltzeiler M, Chen EM. Retromolar intubation for severe tris- mus: a less invasive alternative to tracheostomy and submental intubation. BMJ Case Rep. 2021;14. https://doi.org/10.1136/bcr-2020-239817.
8. Habib A, Zanaty OM. Use of retromolar intubation in paediatric maxillofacial trauma. Br J Anaesth. 2012;109:650-1. https://doi.org/10.1093/bja/aes336.
9. Boker AM. A novel approach to manage patients with compromised airway. Videoscopic assisted retromolar intubation. Saudi Med J. 2013;34:308-12.
10. Rao M, Shetty D, Goyal KA, Nanda KP. Retromolar intubation: a better alternative to submental intubation or tracheostomy for dental occlusion by intermaxillary fixa- tion. Saudi J Anaesth. 2015;9:222-4. https://doi.org/10.4103/1658-354X.152896.
11. Truong AT, Ahmad B, Cata JP, Martin JW, Truong DT, Rahlfs TF. Measurement of retromolar space dimensions using dental pantomograms for intubation feasibility. Head Neck. 2016;38(Suppl. 1):E638-42. https://doi.org/10.1002/hed.24062.
12. Jain D, Samra T, Sabarwal P, Rai S. Fiberoptic-guided retromolar intubation in an in- fant with intraoral tumor. J Anaesthesiol Clin Pharmacol. 2018;34:416-7. https://doi. org/10.4103/joacp.JOACP_119_17.
13. Liem EB, Bjoraker DG, Gravenstein D. New options for airway management: intubating fibreoptic stylets. Br J Anaesth. 2003;91:408-18. https://doi.org/10. 1093/bja/aeg011.
14. Hung KC, Tan PH, Lin VC, Wang HK, Chen HS. A comparison of the Trachway intubating stylet and the Macintosh laryngoscope in tracheal intubation: a manikin study. J Anesth. 2013;27:205-10. https://doi.org/10.1007/s00540-012-1491-6.
15. Shollik NA, Ibrahim SM, Ismael A, Agnoletti V, Piraccini E, Corso RM. Use of the Bonfils intubation fiberscope in patients with limited mouth opening. Case Rep Anesthesiol. 2012;2012:297306. https://doi.org/10.1155/2012/297306.
16. Thong SY, Wong TG. Clinical uses of the bonfils retromolar intubation fiberscope: a review. Anesth Analg. 2012;115:855-66. ANE.0b013e318265bae2.
17. Tsay PJ, Hsu SW, Peng HC, Wang CH, Lee SW, Lai HY. Trachway intubating stylet for tracheal intubation in an ankylosing spondylitis patient undergoing total hip re- placement under general anesthesia. Acta Anaesthesiol Taiwan. 2011;49:159-61. https://doi.org/10.1016/j.aat.2011.11.005.
18. Chen YT, Ho CN, Hung KC. Successful endotracheal intubation with Trachway after failed fiber-optic manipulations in a patient with retropharyngeal cervical chordoma. Ci Ji Yi Xue Za Zhi. 2017;29:174-6. https://doi.org/10.4103/tcmj.tcmj_ 63_17.
19. Ruetzler K, Roessler B, Potura L, et al. Performance and skill retention of intubation by paramedics using seven different airway devices-a manikin study. Resuscitation. 2011;82:593-7. https://doi.org/10.1016/j.resuscitation.2011.01.008.
20. Kariya T, Inagawa G, Nakamura K, et al. Evaluation of the Pentax-AWS((R)) and the Macintosh laryngoscope in difficult intubation: a manikin study. Acta Anaesthesiol Scand. 2011;55:223-7. https://doi.org/10.1213/ANE.0b013e3181d2aad7.
21. Lokesh U, Sudarshan, Jannu AA, Bhattacharya D. Retromolar Intubation: An Alterna- tive Non Invasive Technique for Airway Management in Maxillofacial Trauma; 2013.
22. Nguyen LT, Thakar SD, Truong AT, Truong DT. Orotracheal intubation using the retromolar space: a reliable alternative intubation approach to prevent dental in- jury. Case Rep Anesthesiol. 2016;2016:3529415. https://doi.org/10.1155/2016/ 3529415.
23. DuCanto J, Serrano KD, Thompson RJ. Novel airway training tool that simulates vomiting: suction-assisted laryngoscopy assisted decontamination (SALAD) system. West J Emerg Med. 2017;18:117-20.
24. Sittitavornwong S, Mostofi P, DiLuzio K, Kukreja P, Deatherage H, Kukreja P. Does the retromolar area provide adequate space for an oral endotracheal tube without inter- fering with intermaxillary fixation? J Oral Maxillofac Surg. 2021;79(12):2455-61. https://doi.org/10.1016/j.joms.2021.06.011.
25. Truong A, Truong DT. Retromolar tracheal intubation in a pediatric patient with a difficult airway and bilateral nasal stenoses. Can J Anaesth. 2012;59:809-10. https://doi.org/10.1007/s12630-012-9722-8.