Radiological versus Traditional Parameters for Airway Assessment: A Comparison.

Background: The incidence of difficult tracheal intubation includes not only difficult and failed intubation but also difficult laryngoscopy (DL) and difficult mask ventilation. Aim: The study was done to compare the sonographic assessment of tongue thickness (TT) and condylar mobility with traditional airway assessment scores for prediction of difficult intubation (DI). Settings and Design: This experimental, randomized prospective study was conducted in the Department of Anaesthesia, Himalayan Institute of Medical Sciences, Dehradun. Materials and
Methods: After obtaining approval from the Institutional Ethical and Research Committee, this study was conducted in the Department of Anaesthesiology, Himalayan Institute of Medical Sciences, Swami Rama Himalayan University, Swami Ram Nagar, Dehradun. After obtaining written and informed consent, 385 subjects of either sex between 18 and 60 years of age, undergoing various elective surgeries under general anesthesia with endotracheal intubation, were selected. The number of attempts at intubation, the need for alternative DI approaches, and invasive airway access or cancellation of the procedure due to inability to secure the airway were noted. Statistical Analysis Used: Data analysis was done using SPSS 20.0. Categorical data were assessed by Chi-square test, while independent t-test/Mann-Whitney test was used to determine the association between continuous data.
Results: Age, inter-incisor distance (IID), body mass index (BMI), and Mallampati score were positively correlated with TT. BMI (kg.m-2) >26.0, modified Mallampati scoring >2, IID ≤5 cm, TT >6 cm, and condylar mobility >10 mm were significantly more among subjects with DL and difficult tracheal intubation.
Conclusion: Ultrasonography can be used in elective settings to rule out any possibility of difficult airway and to prevent airway-related complications. Copyright:

INTRODUCTION

The American Society of Anesthesiologists (ASA) defines “a difficult tracheal intubation as a tracheal intubation requiring 3 or more attempts, in the presence or absence of tracheal pathology.”[1] Benumof defined difficult endotracheal intubation as “Cormack and Lehane (C-L) Grade III with several attempts made and defined intubation failure as C-L grade III or IV with failure.”[2] The incidence of difficult tracheal intubation includes not only difficult and failed intubation but also difficult laryngoscopy (DL) and difficult mask ventilation.

The predictors of DL and difficult intubation (DI) have been grouped as individual indices and group indices. Individual predictors of the difficult airway (DA) include physical examination indices such as the modified Mallampati score (MMS), the distance between mentum and hyoid bone, i.e., hyomental distance (HMD), and distance between thyroid notch and mentum, i.e., thyromental distance, neck movements, inter-incisor distance (IID), and neck circumference,[345] radiological indices, and advanced indices including flow–volume loops, ultrasonography (USG), magnetic resonance imaging (MRI), esophagogram, and fluoroscopy [Figure 1]. Group indices include the Wilson scoring system, Benumof's 11-parameter analysis, Arne's simplified scoring model, and the EGRI test.[2]

Figure 1

Variable values to predict difficult tracheal intubation

Imaging techniques, such as three-dimensional computed tomography (CT), X-ray, and MRI, display anatomical features of upper airways and have been recommended for the evaluation of DAs.[678] However, it is difficult to use these techniques in airway research and clinical applications due to their high cost and potential harm to the body.

With the advancements in ultrasound technology, descriptive and accurate evaluation of the anatomical features of the upper airway also provides real-time visualization of airway interventions. In addition, the easy availability of the ultrasound in the clinical setting, as well as the increasing familiarity of practitioners with the bedside use of ultrasound, puts it in position for consideration in the management of the DA.[910]

Various parameters can be used ultrasonographically as a predictor of DL such as mandibular condylar mobility, tongue thickness (TT), skin-to-epiglottis distance, HMD ratio, and anterior soft neck tissue thickness.[11] Restricted temporomandibular joint (TMJ) mobility is a relevant indicator for DA prediction.[12] A tongue which is large is associated with DI and is well known by anesthetists that increased TT directly affects the laryngoscopy and tracheal intubation performance.[5] Previous observations and practice have found that USG imaging of a patient's tongue can accurately measure TT in the “sniffing” position.

The magnitude of the maximal mouth opening is regarded as the TMJ mobility.[6] Several studies suggested that different skeletons and dental morphology have a different impact on maximum mouth opening.[67] That's why the maximum opening of mouth could not provide precise values of mobility of TMJ.[6]

TMJ mobility is assessed in multiple ways.[13] Restricted TMJ mobility is a relevant indicator for DA prediction.[14] Consequently, mandibular condyle displacement may indicate the magnitude of mouth opening.

The study was done to compare the sonographic assessment of TT and condylar mobility with traditional airway assessment scores for the prediction of DI.

MATERIALS AND METHODS

After obtaining approval from the Institutional Ethical and Research Committee, this study was conducted in the Department of Anaesthesiology, Himalayan Institute of Medical Sciences, Swami Rama Himalayan University, Swami Ram Nagar, Dehradun, over a period of 12 months.

Study sample

The subjects were recruited from patients presenting in preanesthetic checkup clinics. After obtaining written and informed consent, 385 subjects sample size was calculated using the formula:

n = Z2α/2/PQ/d2

where n = required sample size, z = 1.96 at 0.05 level of significance, P = unknown efficacy we assume it as 50%, q = 1 − p, and d = 05%

385 patients, of either sex between 18 and 60 years of age, undergoing various elective surgeries under general anesthesia with endotracheal intubation were selected.

Method of study

The patients were kept nil per oral 6 h prior to the surgery for solid food and clear liquids, and plain water was allowed until 2 h prior to the surgery. Classically, DA evaluations were conducted by a resident not involved in the study. Two anesthesiologists who were trained did all the USG-guided assessments independently.

The eligible subject was asked to assume a supine position with his mouth closed and neck stretched. The subject was then asked to slightly touch his tongue tip on his incisors without phonation. Following this, a high-frequency curvilinear array ultrasound probe 5–2 mHz was placed under the subject's chin in the median sagittal plane and further adjusted till the clear outline of the tongue was obtained on the screen completely. The clear image was then immediately frozen. On the frozen image, the maximal vertical distance from the tongue surface to the submental skin was measured, labeling as TT.

Following laryngoscopy, an appropriate-sized ETT was inserted and anesthesia was maintained. The number of attempts at intubation, the need for alternative DI approaches, and invasive airway access or cancellation of the procedure due to inability to secure the airway were also noted. In the case of a DA, the event was recorded.

Statistical analysis

Data were analyzed by statistical software SPSS 20.0. Quantitative variables expressed as mean ± standard deviation and median (min–max) followed normal and skewed distribution, respectively. Categorical data were expressed as frequency and percentage. Chi-square/Fisher's exact test was used to check the association between categorical variables. The independent t-test/Mann–Whitney test was used to compare quantitative variables between two groups (easy and DI). The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated to analyze the diagnostic accuracy of various parameters correlating with DI. Further univariate and multivariate logistic regression was used to estimate odds and independent predictors for difficulty in intubation. P < 0.05 was considered statistically significant.

RESULTS

Table 1 describes the sociodemographic characteristics of the recruited subjects. The majority (31.9%) of the subjects belonged to 50–60 years followed by 30–40 years (23.6%). The mean age of the subjects was 41.85 ± 13.86 years which included 54.3% of males and 45.7% of females.

Table 1

Demographic characteristics (n=385)

Variables	n (%)
Age (year)
≤20	28 (7.3)
20-30	69 (17.9)
30-40	91 (23.6)
40-50	74 (19.2)
≥50	123 (31.9)
Mean±SD	41.85±13.86
Gender
Male	209 (54.3)
Female	176 (45.7)
BMI (kg.m−2)
Underweight	26 (6.8)
Normal	244 (63.4)
Overweight	97 (25.2)
Obese	18 (4.7)
Mean±SD	61.94±9.43
ASA
I	235 (61.0)
II	118 (30.6)
III	32 (8.3)

SD=Standard deviation, BMI=Body mass index, ASA=American Society of Anesthesiologists

A significantly higher number of subjects had <6 cm TT and condylar mobility ≥10 mm with Mallampati scores I and II, IID ≥5, and Cormack-Lehane grades III and IV. A higher number of subjects had ≥6 cm TT among subjects with a Wilson score of ≥4 [Table 2].

Table 2

Comparison among study parameters

Variables	Tongue thickness		χ2; P	Condylar mobility		χ2; P
	<6 cm	≥6 cm		<10 mm	≥10 mm
Modified Mallampati score
I and II	322	22	45.9; 0.001	20	322	20.091; 0.001
III and IV	26	17		11	32
Wilson score
<4	324	30	13.23; 0.001	27	327	1.07; 0.301
≥4	22	9		4	27
Inter-incisor distance
<5	46	15	17.56; 00.001	10	51	7.41; 0.006
≥5	300	23		20	303
Cormack-Lehane grade
I and II	333	21	29.67; 00.001	21	333	24.56; 0.001
III and IV	21	18		10	21

Age, IID, body mass index (BMI), and Mallampati score were positively correlated with TT. IID and Mallampati score were positively correlated with condylar mobility [Table 3].

Table 3

Correlation among classical parameters and ultrasonography parameters

Variables	Tongue thickness		Condylar mobility
	Correlation coefficient	P	Correlation coefficient	P
Age	0.171	0.001*	0.001	0.991
Inter-incisor distance	0.102	0.045*	0.166	0.001*
Wilson score	0.095	0.064	0.035	0.499
Condylar mobility	0.006	0.059
BMI	0.205	0.001*	0.054	0.290
Mallampati score	0.356	0.001*	0.145	0.004*
IDS	0.013	0.793	0.035	0.497

*Spearman’s correlation coefficient. BMI=Body mass index, IDS: Intubation difficulty scale

BMI (kg.m−2) > 26.0, modified Mallampati scoring >2, IID (in cm) ≤5, TT (in cm) >6, and condylar mobility <10 mm were significantly more among subjects with DL and difficult tracheal intubation [Table 4].

Table 4

Comparisons of preoperative characteristics between patients with and without difficult laryngoscopy or difficult tracheal intubation

Variables	Difficult laryngoscopy				Difficult intubation
	Yes (n=35), n (%)	No (n=349), n (%)	Total (n=385), n (%)	χ2; P	Yes (n=30), n (%)	No (n=354), n (%)	Total (n=385), n (%)	χ2; P
BMI (kg/m2)
≤25.9	20 (57.1)	285 (81.7)	306 (79.5)	11.985;	18 (60)	287 (81.1)	306 (79.5)	7.790; 0.001*
>26.0	15 (42.9)	64 (18.3)	79 (20.5)	0.002*	12 (40)	67 (18.9)	79 (20.5)
Wilson score
≤4	16 (45.7)	162 (46.4)	179 (46.5)	1.160; 0.560	13 (43.3)	165 (46.6)	179 (46.5)	1.273; 0.529
>4	19 (54.3)	187 (58.6)	206 (53.5)		17 (56.7)	189 (8.7)	206 (53.5)
Modified Mallampati scoring
≤2	20 (57.1)	321 (92.0)	342 (88.8)	39.031; <0.001*	14 (46.7)	327 (92.4)	342 (88.8)	58.359; <0.001*
>2	15 (42.9)	28 (8.0)	43 (11.2)		16 (53.3)	27 (7.6)	43 (11.2)
Inter-incisor distance (cm)
≤5	12 (34.3)	59 (16.9)	71 (18.4)	8.475; <0.001*	15 (50)	51 (14.1)	66 (17.1)	25.052; <0.001*
>5	23 (65.7)	290 (83.1)	314 (81.6)		15 (50)	303 (84.1)	318 (82.5)
Tongue thickness (cm)
≤6	17 (48.6)	332 (95.1)	350 (90.9)	83.531; 0.001*	13 (43.3)	332 (93.8)	346 (89.9)	77.440; <0.001*
>6	18 (51.4)	17 (4.9)	35 (9.1)		17 (56.7)	22 (6.2)	39 (10.1)
Condylar mobility (mm)
≤10	17 (48.6)	24 (6.9)	41 (10.6)	58.236; <0.001*	18 (60)	23 (6.5)	41 (10.6)	83.320; <0.001*
>10	18 (51.4)	325 (93.1)	344 (89.4)		12 (40)	331 (93.5)	344 (89.4)

*P value less than or equal to .05

In the present study, TT and IID show the highest sensitivity of 65.7% and 46.15%, respectively, and the lowest specificity of 16.9%. In addition, IID has the lowest negative predictive value (38.3%) as well. Thus, it serves as a potentially valid tool for screening DAs. However, TT represents the highest specificity (95.1%) with positive (48.6%) and negative predictive values (96.24%) compared to others. Thus, it serves as a valid diagnostic tool. The remaining BMI and Wilson score have low sensitivity values and low positive predictive values [Table 5].

Table 5

Variable values to predict difficult tracheal intubation

Variables	OR (99% CI)	Sensitivity	Specificity	NPV	PPV
Modified Mallampati scoring
>2	10.91 (3.97-18.62)	34.85	95.32	92.09	48.39
BMI
>26.0	3.34 (1.62-6.88)	42.85	81.66	91.11	18.98
Wilson score
>4	3.1 (0.51-2.07)	19.35	92.94	92.94	19.35
Inter-incisor distance (cm)
>5	0.39 (0.18-0.83)	65.72	16.9	38.3	83.8
Tongue thickness (cm)
>6	21.9 (9.08-47.08)	46.15	96.24	94.07	58.06
Condylar mobility (mm)
≤10	0.13 (0.04-0.17)	32.26	94.07	94.07	32.26

Variables were determined by CI, NPV, and PPV. CI=Confidence interval, NPV=Negative predictive value, PPV=Positive predictive value, OR=Odds ratio, BMI=Body mass index

DISCUSSION

Unanticipated DI always remained a challenge to anesthesiologists,[7] and various studies and investigations had mentioned methods to predict DI. Identification of a DA prior to intubation is important as it gives time for optimal patient preparation, an equipment selection, and involvement of the experienced personnel.[7] Thus, airway examination prior to the laryngoscopy might be benefitted from better methods of airway assessment.

Our study explored the potential utility of ultrasound to measure condylar mobility and TT and its comparison with classical methods for assessing airways.[15] Literature suggests that approximately 10%–15% of adults with temporomandibular disorders remain undiagnosed till the time of surgery.[16] Although the IID is the most common method to assess TMJ mobility, its correlation with TMJ mobility remains very poor.[17] Scholars suggested that maximal mouth opening does not provide reliable information of TMJ mobility.[18]

Increased TT influences the performance of laryngoscopy and tracheal intubation and hence increases the risk of DA. The MMT reflects tongue volume to some extent but poses limited predictive power[19] and also difficult application in unconscious, uncooperative, and emergency patients. The size of the tongue, in relation to the oropharyngeal space, is an important determinant of the ease of introduction of the laryngoscope blade.

We included patients in the age group of 18–60 years in our study. With age, the larynx descends, and in females, it stops at an earlier age but continues in males even after puberty. As the age advances, there is calcification of cartilages of the larynx contributing to difficulty during airway management. Hence, in order to avoid differences in airway measurements due to age discrepancies, patients younger than 18 years and more than 60 years were not included. The total number of females was 45.7%, and males were 54.3% in our study. Uribe et al. found more chances of DI in males with higher BMI.[20] Similar studies done by Yao and Wang and Wojtczak to evaluate DA were also done on adult subjects.[617] All the patients who required GA for surgery belonged to ASA PS Classes I (61%), II (30.6%), and III (8.3%).

In our study, we had a maximum number of patients with MMS I and II in 58.4% and 30.1% of cases, while MMS III and MMS IV were 10.9% and 0.6% of cases, respectively. When ease of intubation in the form of Cormack and Lehane (C-L) grade was correlated with MMS, the score was able to predict the DI. Lee et al. found poor-to-good accuracy of this test in a systemic review of 42 studies on 34,513 patients.[21] Mallampati determined the size of the tongue in relation to oral cavity and pharynx, while its relation with glottic view on laryngoscopy is determined by C-L.

In our study, we found the odds ratio of MMS to be 10.91 and specificity and sensitivity of 0.95 and 0.34, respectively. Lundstorm et al. found the odds ratio for a DI with MMS of III or IV to be 5.89 by a meta-analysis, and specificity and sensitivity came out to be 0.91 and 0.3, respectively.[7] Our results were in accordance with the above. Similarly, the sensitivity and specificity of MMS following C-L laryngoscopic grading as assessed by Patel et al. were 28.6% and 75.8%, respectively.[13]

Wilson score was previously proposed by Wilson et al. in 1988 as a suitable useful predictive test.[22] In the current study, 350 patients had scores <4 which predicts easy intubation, while the rest of the patients had a score ≥4. Our study depicts that Wilson score was able to predict DI. In our study, Wilson score had a sensitivity and specificity of 19.35% and 92.94%, respectively.

We took TT and condylar mobility as USG parameters to predict DI. We found TT of >6 cm to have DL in 4.6%, while DI was seen in 4.4% of subjects. The sensitivity and specificity of TT were 0.46 and 0.96, respectively, and NPV and PPV were 0.94 and 0.58, respectively.

The condylar translation had a correlation with the C-L level among subjects and was correlated with C-L grade. We found DL in 4.4% of subjects, while 4.6% of subjects had DI. The sensitivity and specificity were 0.32 and 0.94, respectively, while NPV and PPV were 0.94 and 0.382, respectively. In a study done by Yao and Wang, they concluded that sonography can easily determine mandibular condylar translation and found DL in 9.2% of subjects with specificity, sensitivity, NPV, and PPV of 0.81, 0.91, 0.98 and 0.45 respectively, which was nearly precise to our study, except sensitivity.[17]

During the mouth opening, condylar rotation and anterior-inferior sliding movements are mainly involved.[23] While performing mandibular protrusion and ULBT, condylar sliding but not rotating movements are usually involved. During laryngoscopy, jaw is lifted anteriorly and inferiorly, and the mandibular condyle is translated anteriorly and inferiorly under articular eminence and the articular disk. With decreased condylar mobility, there is difficulty while performing laryngoscopy with poor glottic views.[24]

It is generally considered that a high BMI is a predictor of DA while some studies made a different conclusion, indicating no association of BMI with DI. In our study, the mean BMI was 61.94 ± 9.43, with 58 patients having BMI >26 kg.m−2 with anticipated DA but only 30 patients possible to have DL. The sensitivity of BMI in prediction of DI comes to be 92%, while the specificity is 81.6% in our study. Higher BMI was associated with a thick tongue >6 cm. It has been documented in the literature that high BMI is associated with a thick tongue, thus suggesting that higher BMI has a correlation with large tongue which makes laryngoscopy difficult.[14]

In our study, when TT was compared with IID, BMI, MMS, and Wilson score, a positive correlation was found, and also, the positive association was seen between DA and the TT. In a study conducted by Yao and Wang,[17] similar results were found where the above parameters showed a positive correlation. Higher BMI causes a thicker tongue and poses difficulty in tracheal intubation.

We assessed the correlation between condylar mobility with parameters such as age, IID, MMS, WS, and BMI. A positive correlation was found between IID and MMS, while no correlation was found between other measured parameters. In a study by Yao and Wang,[17] low correlation was found between IID and condylar translation.

There were a few limitations in our study. However, our study was a single-centered study that showed a confirmed relation of sonographic prediction of DAs; however, any validation except for the subjective assessment was not done, which could have been done by other methods such as CT and MRI. However, in our setting with patients having limited resources and the risk of radiation exposure, we have not included these parameters. Since the cutoff values have not been determined, we have taken them from previous studies.

CONCLUSION

We have found that sonographic measurement of TT and condylar mobility can be a useful prediction tool for the assessment of a DA. It is comparable, although, found to be better than the traditional assessment indices of a DA. Since it is an upcoming modality, with our area of interest, requiring a short learning curve, this can prove to be very useful in quick and accurate prediction.

Financial support and sponsorship

This study was financially supported by the Swami Rama Himalayan University.

Conflicts of interest

There are no conflicts of interest.