Diagnostic Accuracy of Ultrasound Measurements of Anterior Neck Soft Tissue in Determining a Difficult Airway.

CONTEXT: Airway management in the emergency department is challenging because conventional screening tools cannot be applied. Therefore, a rapid noninvasive means of identifying a difficult airway will be advantageous for emergency physicians. AIMS: The aim of this study is to examine the association between ultrasound measurements of anterior neck soft tissue and difficult airway as judged by the Cormack Lehane grading. SETTINGS AND
DESIGN: A prospective study was done for 18 months on patients requiring intubation presenting to the emergency medicine department. SUBJECTS AND METHODS: Ultrasound measurements of anterior neck soft tissue were obtained in 60 cases at the levels of thyrohyoid membrane, hyoid bone, and vocal cords. Another examiner who was blinded to the ultrasound measurements performed endotracheal intubation and noted Cormack-Lehane grading. STATISTICAL ANALYSIS USED: Descriptive statistics such as mean, standard deviation, frequency, and percentage were used. Inferential statistics such as Student's t-test and receiver operating characteristic (ROC) curve analysis were done using the SPSS software version 22.
RESULTS: The thickness of anterior neck soft tissues at the level of hyoid bone in difficult patients was 0.73 cm (95% confidence interval = 0.65-0.80) compared to easy patients 0.47 cm (95% confidence interval = 0.44-0.51) with a P = 0.001 and at the level of thyrohyoid membrane in difficult patients it was 1.83 cm (95% confidence interval = 1.7-1.89) compared to easy patients 1.46 cm (95% confidence interval = 1.41-1.51) with a P = 0.001. Area under the ROC curve was significant at all the three levels with the highest at the level of thyrohyoid membrane 0.99 and least at the level of vocal cords 0.79, the area under the curve was 0.92 at the level of hyoid bone.
CONCLUSIONS: Sonographic measurements of the anterior neck soft tissue can be used as a screening tool by an emergency physician to detect difficult intubation. Copyright:

INTRODUCTION

Emergency physicians are often faced with a situation of managing a difficult airway. Failure to secure the airway in an emergency can cause life threatening morbidity and mortality. The clinical tests performed before the intubation for the identification of a difficult airway (such as inter-incisor gap, thyromental distance, Mallampatti scale, and assessing for neck mobility) are often not possible to do in the emergency rooms because patients usually cannot cooperate, are in altered sensorium, are not able to follow commands, or are lethargic and unresponsive. This poses a challenge to the emergency physician for securing the patient's airway.

Ultrasound has become a popular and viable tool for the assessment of the patient in the critical care and emergency medicine setting because it is portable, noninvasive, can be done at the bedside, and can be repeated multiple times with minimal side effects and discomfort to the patient.

There has been some research about the application of ultrasound in the assessment of the airway. The rationale is that while performing direct laryngoscopy, soft tissues of the anterior neck at the level of the thyrohyoid membrane and hyoid bone are displaced to view the vocal cords. The ultrasound measurements of soft tissue of the anterior neck at the level of the thyrohyoid membrane and hyoid bone may allow the prediction of difficult intubation.

A few studies examining the association between the ultrasound measurements of the anterior neck soft tissue and difficult intubation (as judged by Cormack–Lehane grading) do exist. These studies, conducted on different sets of the population, have shown varied results.

While it is evident that a quick and noninvasive method of assessing the airway for the prediction of difficult endotracheal intubation would be highly desirable, especially in an emergency setting, we found that there were very few studies on the usefulness of ultrasound measurements of the anterior neck tissues in the prediction of difficult airway and no studies on the same conducted in an emergency department.

The present study was conducted with the objective of defining the role of point-of-care ultrasound measurements of the anterior neck soft tissue in the identification of difficult airway as judged by Cormack–Lehane grading and its usefulness in our population, especially from the standpoint of an emergency physician.

SUBJECTS AND METHODS

Study design

This was an observational study performed over a period of 18 months on patients presenting to the department of emergency medicine at a tertiary care center. Adult patients presenting to the emergency department requiring endotracheal intubation satisfying the inclusion and exclusion criteria were included in the study.

A total of sixty patients who met the criteria were included in the study by convenience sampling. This number was chosen similar to a few previous studies as there is little information in the published literature to perform a formal sample-size calculation. Approval was obtained from the Institutional Ethical Committee. Informed written consent from patients, their legally authorized representatives, or close families was obtained after thorough explanation of the study. The point-of-care ultrasound examination was conducted parallel to the standard-of-care management of the patient, and its patient safety was approved by the clinical governance committee of the institution (JSSMC/PG/4700/2017-18 dated November 04, 2017).

Inclusion criteria

Patients aged >18 years presenting to the emergency department requiring endotracheal intubation for various indications.

Exclusion criteria

Patients requiring crash intubation

Patients having open injuries on the neck

Patients having facial fractures, maxillofacial tumors, and tumors of the face and neck

Patients with known airway pathology

Uncooperative patients

Pregnant patients

Patients with abnormal dentition.

Method of collection of data

Patients' data were collected on a pro forma which included:

Demographic details

Vital signs

Indication for intubation

The anterior neck soft-tissue ultrasound measurements

The Cormack–Lehane grading

The time taken for ultrasound measurements.

Prior to the collection of data, all coinvestigators underwent training to perform airway ultrasound. All the demographic data and vital signs were collected. The ultrasound measurements were taken using Fujifilm Sonosite (Bothell, WA, US) M-turbo using HFL 38, 13–6 MHz linear transducer. The ultrasound examination was performed with the patient in the supine position. The head and neck were in a neutral position without a pillow. The ultrasound measurements of the anterior neck soft tissue in the central axis at the level of the hyoid bone, thyrohyoid membrane, and at the level of the vocal cords were recorded by one of the investigators. The ultrasound measurements were done before giving a paralytic agent. The time taken for measurements was also noted [Figures 1-3].

Figure 1

(a and b) The hyoid bone level. (1) The arrows indicate hyoid bone (2) Yellow-dotted line denote the distance from skin to hyoid bone

Figure 3

(a and b) Anterior commissure level. (1) VL: Vocal ligament. (2) AC: Anterior commissure. (3) Yellow-dotted line denote the distance from skin to anterior commissure

(a and b) The thyrohyoid membrane level. (1) A-M air mucosal interface (2) PRE-pre epiglottis space (3) Yellow dotted line denote the distance from skin to epiglottis midway between hyoid bone and thyroid cartilage

Another investigator blinded to the ultrasound measurements performed the endotracheal intubation and noted down the Cormack–Lehane grading by direct laryngoscopy using Macintosh blade 3 or 4 as appropriate. All the patients in the study received induction and paralytic agents before the laryngoscopy. Cormack–Lehane grading 1 and 2a were taken as “easy” and grades 2b, 3, and 4 were considered as “difficult” airway (the criteria used in multiple studies before).[1]

The data were entered into Microsoft excel followed by the analysis using the IBM SPSS Statistics for Windows, Version 23.0. (Armonk, NY: IBM Corp). Arithmetic mean, standard deviation, and percentages were used to represent the gender distribution, age distribution, and time taken for ultrasound measurements.

The comparison between easy and difficult groups was done by using the Student's t-test. Sensitivity and specificity calculations were performed. Receiver operating characteristic (ROC) curves were plotted for the data obtained at the three levels (hyoid bone level, thyrohyoid membrane level, and level of vocal cords).

RESULTS

In the present study, low Glasgow Coma Scale was the main indication for intubation and road traffic accidents (58.3%) being the most common reason for it. The majority of the population were males (78.3%). In the present study, males had a higher proportion of difficult intubation. Difficult airway was encountered in 40.42% of male patients and in 30.76% of female patients.

The mean time taken for the ultrasound measurements was 103.68 s with a 95% confidence interval 99.65–107.71s. The average time taken for the measurements at these above-mentioned levels is <2 min.

Ultrasound measurement of the soft tissues of the anterior neck was the index test and the Cormack–Lehane grading of “easy” (Grade 1 and 2a) and “difficult” (Grade 2b, 3 and 4), was taken as the gold standard for the evaluation of the index test.

At hyoid bone level, the thickness of soft tissue anterior to the neck varied from 0.39 cm to 1.05 cm. The anterior neck soft tissue at the level of hyoid bone in difficult patients was 0.73 cm (95% confidence interval = 0.65–0.80) compared to easy patients 0.47 cm (95% confidence interval = 0.44–0.51) with a P = 0.001 [Tables 1 and 2].

Table 1

The mean of Ultrasound measurements in patients with easy and difficult laryngoscopy

	Mean±SD	95% CI for mean
		Lower bound	Upper bound
Hyoid bone
Easy	0.4757±0.10548	0.4405	0.5108
Difficult	0.7317±0.16946	0.6585	0.8050
Thyrohyoid membrane
Easy	1.4673±0.15591	1.4153	1.5193
Difficult	1.8317±0.14086	1.7708	1.8927
Vocal cords
Easy	0.7468±0.90022	0.4466	1.0469
Difficult	0.7770±0.16027	0.7077	0.8463

The mean±SD, and CI in easy and difficult patients at the levels of hyoid bone, thyrohyoid membrane, and vocal cords. CI: Confidence interval, SD: Standard deviation

Table 2

Distribution of Cormack Lehane grade according to ultrasound thickess at hyoid level

Ultrasound measurements at hyoid membrane level (mean thickness in cm)	Cormack-Lehane grading (gold standard)
	Difficult (Grade 2b,3 and 4)	Easy (Grade 1 and 2a)
>0.57	20	6
<0.57	3	31
Total	23	37

At the thyrohyoid membrane level, the thickness of soft tissue anterior to the neck varied from 1.19 to 2.0 cm. The anterior neck soft tissue at the level of thyrohyoid membrane in difficult patients was 1.83 cm (95% confidence interval = 1.7–1.89) compared to easy patients 1.46 cm (95% confidence interval = 1.41–1.51) with a P = 0.001 [Tables 1 and 3].

Table 3

Distribution of Cormack Lehane grade according to ultrasound thickess at thyrohyoid membrane level

Ultrasound measurements at Thyrohyoid membrane level (mean thickness in cm)	Cormack-Lehane grading (gold standard)
	Difficult (Grade 2b,3 and 4)	Easy (Grade 1 and 2a)
>1.6	23	3
<1.6	0	34
Total	23	37

The values at the vocal cords level in our study ranged from 0.52 to 1.04 and had a mean of 0.75 cm. The anterior neck soft tissue at the level of vocal cords in difficult patients was 0.77 cm (95% confidence interval = 0.70–0.84] compared to easy patients 0.74 cm (95% confidence interval = 0.44–1.04) with a P = 0.874 [Tables 1 and 4].

Table 4

Distribution of Cormack Lehane grade according to ultrasound thickess at hyoid level

Ultrasound measurements at vocal cord level (mean thickness in cm)	Cormack-Lehane grading (gold standard)
	Difficult (Grade 2b, 3, and 4)	Easy (Grade 1 and 2a)
>0.76	12	5
<0.76	11	32
Total	23	37

A ROC curve was plotted at all three levels. Area under the curve was significant at all the three levels with the highest at the level of thyrohyoid membrane 0.99 and least at the level of vocal cords 0.77, the area under the curve was 0.92 at the level of hyoid bone [Figure 4].

Figure 4

Receiver operating characteristic curve. ROC curve was plotted which was significant at all the three levels with highest area under the curve for thyrohyoid membrane 0.99 and the least at vocal cord level 0.77, and was 0.92 at the level of hyoid bone. (VC: Vocal cord level, HYB: Hyoid bone level, THM: Thyrohyoid membrane level)

At the hyoid bone, an optimal cutoff point of 0.58 cm was derived from the ROC curve-yielding sensitivity and specificity of 87% and 87.5%, respectively. Similarly, at the thyrohyoid membrane level, optimal cutoff value was found to be 1.59 cm with sensitivity and specificity of 100% and 91.9%, respectively. Finally, at the level of vocal cords, the optimal cutoff value was found to be 0.61 cm with sensitivity and specificity of 87% and 64.9%, respectively.

DISCUSSION

Previous studies which explored ultrasound of the airway to predict the difficult intubation by Cormack-Lehane grading have used multiple parameters at various levels.[12345] Measurements at all levels would be time-consuming and delay the appropriate care in an emergency setting. In our study, we obtained central axis measurements at the levels of thyrohyoid membrane, hyoid bone and vocal cords as recommended by the previous study to obtain the minimum number of measurements and gain the required information.[2]

The mean time taken for the measurements was 103.68 s, with a 95% confidence interval 99.65–107.71. The mean time taken for ultrasound measurements in the central axis at these above-mentioned levels is <2 min, which makes it an ideal screening test in the emergency department and critical care setting where time is a constraint. Adhikari et al. reported a mean time of 9.6 min to obtain three measurements each at all levels of the tongue and anterior neck soft tissue.[2]

In the present study, anterior neck soft tissue at the level of hyoid bone in difficult patients was 0.73 cm (95% confidence interval = 0.65–0.80] compared to easy patients 0.47 cm (95% confidence interval = 0.44–0.51) with a P = 0.001 [Table 1].

In the study conducted by Adhikari et al., they found that the increase in anterior neck soft-tissue thickness at the hyoid bone level was associated with difficult intubation with values 1.69 cm (95% confidence interval = 1.19–2.19) and in easy intubations 1.37 cm (95% confidence interval = 1.27–1.46].[2] Wu et al. conducted a study in China and found soft tissue from the hyoid bone to skin was an independent predictor of difficult laryngoscopy with values of 1.51 ± 0.27 cm in the difficult patients and 0.98 ± 0.26 cm in the easy patient group.[6] In a study conducted by Reddy et al., anterior neck soft tissue did not prove to be a significant predictor of difficult intubation at the level of the hyoid bone.[4]

At the thyrohyoid membrane level, the thickness of soft tissue anterior to the neck varied from 1.19 to 2.0 cm. The anterior neck soft tissue at the level of thyrohyoid membrane in difficult patients was 1.83 cm (95% confidence interval = 1.7–1.89) compared to 1.46 cm in easy patients (95% confidence interval = 1.41–1.51) [Table 1].

A pilot study done by Adhikari et al. found that at the thyrohyoid membrane level, increase in the thickness was observed in difficult patients 3.47 cm (95% confidence interval = 2.88–4.07) cm versus in easy patients 2.37 cm (95% confidence interval = 2.29–2.44). Furthermore, a thickness of 2.8 cm by ultrasound measurement differentiated the patients with easy and difficult laryngoscopies at the thyrohyoid membrane level.[2] Parameswari et al., in their study, found that a value of 1.8 cm at the thyrohyoid membrane level was able to differentiate between easy and difficult laryngoscopies.[5] Wu et al., in their study, found that the thyrohyoid membrane distance by ultrasound was higher in the difficult group 2.39 ± 0.34 cm when compared with easy group 1.49 ± 0.3 cm.[6]

The values at the vocal cords level in the present study were ranging from 0.52 to 1.04 and had a mean of 0.75 cm. The anterior neck soft-tissue thickness at the level of vocal cords in difficult patients was 0.77 cm (95% confidence interval = 0.70–0.84) compared to easy patients 0.74 cm (95% confidence interval = 0.44–1.04) [Table 1].

An anterior neck soft-tissue thickness at the level of vocal cords >0.23 cm had a sensitivity of 85.7% in predicting a difficult airway in the study done by Reddy et al.[4] In a study to predict difficult laryngoscopy by ultrasound quantification of the anterior neck soft tissue in obese patients, Ezri et al. found that increased thickness of soft tissue at the vocal cord level was associated with difficult intubation with values of 2.7 cm in the difficult group versus 1.8 cm in the easy group.[7] Jinhong Wu also observed similar results with values of 1.30 ± 0.31 in the difficult group versus 0.92 ± 0.20 in the easy group.[6] In a study by Komatsu et al., it was found that sonographic measurements of soft tissue anterior to the neck failed to identify difficult laryngoscopy in obese patients in the United States.[8] In the present study, we found poor association between ultrasound measurements at the level of vocal cords and Cormack–Lehane grading as adjudged by direct laryngoscopy.

The variation of the values in the above-mentioned studies can be attributed to the differences in methods such as laryngoscopic techniques, application of Sellick's maneuver, differences in ultrasound techniques, and ethnicity of the patient populations. The differences were recognized in various ethnic groups in terms of fat distribution.[910]

A ROC curve was plotted at all three levels. Area under the ROC (AUROC) curve was the highest at the level of thyrohyoid membrane (0.99) followed by the hyoid bone level (0.92), suggesting that these measurements are excellent at predicting a difficult airway. AUROC was least among the three but still significant at the level of vocal cords (0.77), suggesting good probability of identifying a difficult airway with measurements at this level.

At the hyoid bone, an optimal cutoff point of 0.58 cm was derived from the ROC curve yielding sensitivity and specificity of 87% and 87.5%, respectively. Similarly, at the thyrohyoid membrane level, optimal cutoff value was found to be 1.59 cm with sensitivity and specificity of 100% and 91.9%, respectively. Finally, at the level of vocal cords, the optimal cutoff value was found to be 0.61 cm with sensitivity and specificity of 87% and 64.9%, respectively.

Thus, while it is clear that there is no consensus regarding cutoff values for the anterior neck soft tissue at various levels which indicates difficult laryngoscopy, it is also certain that increasing thickness of anterior neck soft tissues correlates well with higher laryngoscopic grades of difficulty of airway.

In the present study, we found that measuring anterior neck soft tissue at predefined levels by point-of-care ultrasound in the emergency department was feasible as the measurements can be done within 2 min and were useful in the prediction of difficult Cormack–Lehane grading by direct laryngoscopy.

Hence, it can serve as a screening modality by the emergency physicians and critical care physicians to detect difficult intubation and thus aid decision-making in patients undergoing intubations.

Limitations

This was a single-center study. Larger studies involving populations from different geographical regions will be required to shed light on the appropriate cutoff values of anterior neck soft-tissue thickness by point-of-care ultrasound for the prediction of difficult airway.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.