Can Mandibular Condylar Mobility Sonography Measurements Predict Difficult Laryngoscopy?

Abstract

BACKGROUND: 

Limited mandibular condylar mobility plays an important role in difficult laryngoscopy. Indirect assessment methods, such as mouth opening, have been proven to be useful predictors of difficult laryngoscopy. Sonography is a new direct assessment method for the limited mandibular condylar mobility. However, whether this method could be used in predicting difficult laryngoscopy still remains unknown. This study aimed to observe its ability to predict difficult laryngoscopy.

METHODS: 

Adult patients who were administered tracheal intubations for elective surgery under general anesthesia were enrolled in the study. Mandibular condylar mobility was assessed by sonography through condylar translation measurements. Beside mouth opening, other indirect variables that correlated with temporomandibular joint mobility, such as mandibular protrusion distance, upper lip bite test, and whether the condyle-tragus distance was <1 finger breadth, were also evaluated before anesthesia. The primary outcome was difficult laryngoscopy defined as the Cormack-Lehane level 3 or 4.

RESULTS: 

A total of 484 patients were prospectively included, and difficult laryngoscopy was reported in 41 patients. The condylar translation prediction criterion for difficult laryngoscopy was ≤10 mm. The condylar translation was correlated with Cormack-Lehane level (Spearman correlation coefficient, −0.46; 99% confidence interval [CI], −0.55 to −0.36) and owned the highest area under the receiver operating characteristic curve (0.93; 99% CI, 0.90 to 0.96, compared with that of the other predictors, P < .001) with difficult laryngoscopy. The condylar translation ≤10 mm was with a considerable κ value (κ = 0.52; 99% CI, 0.37 to 0.67) to difficult laryngoscopy and proved to be an independent predictor by a multivariate logistic regression.

CONCLUSIONS: 

Compared with indirect assessments, such as mouth opening and other parameters, mandibular condylar mobility, as assessed directly using sonography, was correlated with difficult laryngoscopy and demonstrated an independent and notably predictive property.

Difficult airways can cause ventilation failure, oxygen desaturation, arrhythmia, bronchoconstriction, airway injury, cardiac arrest, brain damage, or even death in anesthetic practice.¹ Hence, preoperative evaluations of difficult airways are very important. Limited temporomandibular joint (TMJ) mobility is a meaningful indicator for predicting difficult airways.^2,3 When performing laryngoscopy, it is well-known that the jaw needs to be lifted anteriorly and inferiorly (relative to the patient), and the mandibular condyle will be translated anteriorly and inferiorly beneath the articular eminence and articular disk. With decreased TMJ mobility, especially with decreased mandibular condyle mobility, there is also reduced mobility in the mandibular, which will affect the laryngoscopy and glottis views, causing difficulty in performing laryngoscopy. The problem is how to evaluate TMJ mobility. There are certain indirect or related TMJ mobility evaluations. The interincisor distance (IID) is the most commonly used method for assessing the TMJ mobility, regardless of the fact that it only has a poor correlation with TMJ mobility.^4,5 The upper lip bite test,^6,7 mandibular protrusion distance,³ and maximum interval of the condyle to the tragus (condyle-tragus distance, described by Pat Sullivan in Anaesthesia for Medical Students,1999 Edition, published by Department of Anesthesia, Ottawa Civic Hospital, Ottawa, Ontario, Canada) were also considered related assessments for TMJ mobility. However, all these methods assess TMJ mobility in an indirect or a vague manner, but these methods have become important and valuable in performing difficult airway evaluations.^3,6–8 Sonography is a new direct and reliable method for the assessment of TMJ mobility.^9–11 The translated condyle distance can be measured directly and accurately with sonography, which can reflect TMJ mobility.^10,11 However, compared with other indirect assessment methods, it is unclear whether a limited condylar translation can or cannot predict difficult laryngoscopy. We assumed that the limited condylar translational distance directly measured by sonography was equal or superior to indirect assessment methods in predicting a difficult laryngoscopy. The purpose of this study was to observe the correlation between mandibular condylar mobility, as measured by sonography, and other related assessment results, and their capacity for predicting difficult laryngoscopy.

METHODS

The Institutional Review Board of Yijishan Hospital, Wannan Medical College, certificated the study protocol. It was also registered into chictr.org.cn on November 17, 2013 (one of the primary registries in the WHO Registry Network, ChiCTR-RNC-13003898; W.Y., principal investigator), before patient enrollment. Patients were provided with a written document describing the study. Written consents were obtained from the participants.

We selected elective surgery patients who were administered tracheal intubations under general anesthesia, ASA physical status I to III, and who were 18 to 90 years old. Considering some maxillofacial diseases would affect the TMJ mobility in asymmetric or uncertain way, the exclusion criteria included (1) patients with maxillofacial trauma or malformation; (2) patients with maxillofacial, oral, or tongue cancer; (3) laryngeal and tracheal patients with neck tumor compressions; and (4) patients who needed to be awake during tracheal intubations because of a difficult airway history. Because of the limited 2 ultrasound equipments and the limited 2 sonographers, only the patients who were assigned to the operating rooms in specified area should be enrolled.

Airway Assessments

In a transitional waiting hall before transport to the operating room, all the selected patients had a mandibular condylar mobility sonographic measurement performed on them with a high-frequency linear array ultrasound probe (S8; SonoScape Corp. LP, Shenzhen, China). Sonographic measurements were performed by the first author and fifth author, both of whom are experienced sonographers and performed mandibular condylar mobility sonographic measurements in at least dozens of volunteers before the case enrollment. The utilized ultrasonic mandibular condylar mobility imaging measurement methods are as follows: The probe was placed on one side of the joint site transversely at the level of the zygomatic arch. One end of the transducer was close to the ear canal and tragus (left of the screen) and the other end pointed to the patient’s philtrum (right of the screen). The transducer was kept perpendicular on the ear’s skin (see Figure 1A), which allowed the condyle imaging to be shown on the ultrasound screen (a high point in hyperechoic arc, see Figure 1B). The probe was placed so that it did not cause any relative motion on the ear and skin, and the condylar process images were captured separately when the mouth was opened and closed. When the 2 images were (see Figure 1, B and D) compared, the mandibular condyle position could be found shifted. By identifying the condylar vertex in 1 picture and its corresponding location in another picture with the corresponding coordinates allows for the condylar translation to be accurately measured with ultrasonic instrument measurement tools (see Figure 1, B and D). A limited condylar translation was defined as the numerical value of condylar translation that was lower than its optimum prediction criteria for difficult laryngoscopy.

Figure 1.: The temporomandibular joint mobility sonographic methods and condylar translation measurement. Transducer position (A and C) and images (B and D) captured separately when the mouth was opened and closed. When the 2 images were compared, the mandibular condyle position could be found shifted from one point (crosshair marked “1” in B and D) to the other (the other crosshair in D). The condylar translation distance of 1 author is 15.15 mm (D).

Two residents recruited the patients and performed other classical difficult airway evaluations that are associated with TMJ mobility on an earlier preoperative day. The IID was measured when the mouth was opened as widely as possible. An IID <4 cm was deemed as a difficult airway predictor.⁸

The upper lip bite test⁷ was conducted when the patients were in the upright sitting position with their jaw protruded. The ability of their lower teeth to bite their upper lip was recorded in 3 grades. Grade 1 was obtained when the lower teeth were able to bite the upper vermilion line and completely cover the upper lip membrane. Grade 2 was obtained when the lower teeth bit below the upper vermilion line so that only some of the membrane was covered. Grade 3 was obtained when the lower teeth could not bite the upper lip. Grades 2 and 3 were deemed as difficult airway predictors.

When patients were instructed to maximize the protraction of their lower incisors without their mouth open, the moving distance of lower incisors was measured with a scale ruler. If the lower incisors could not be anterior to the upper incisors, then this was considered a difficult airway predictor.⁸

Whether the condyle-tragus distance was <1 finger breadth was assessed when the patients were asked to open their mouth to the greatest degree possible. Thyromental distance was measured when the patient’s head is fully extended. A thyromental distance of <6.5 cm was considered predictive of difficult airway.¹²

Also, the evaluations of the modified Mallampati test were performed. A grade 3 or 4 was deemed as a predictor for difficult airway.^13,14

The primary outcome of present study was the results of laryngoscopy evaluated as Cormack-Lehane level.¹⁵ A standardized agent (vecuronium), doses (0.1 mg/kg), and time (3 minutes after muscle relaxants treatment) were used to standardize the intubation conditions. The general anesthesia induction was performed in a standard protocol with a combination of 0.05 mg/kg midazolam, 0.003 to 0.004 mg/kg fentanyl, 0.1 mg/kg vecuronium, and 1 to 2 mg/kg propofol. Laryngoscopy and tracheal intubations were carried by clinical attending anesthesiologists who were assigned responsible for the patients as routine work and had >5 years of experience. The patients’ heads were placed in a sniff position,¹⁶ with a 10-cm-height soft pillow. A Macintosh 3 or 4 blade, which was determined by the attending anesthesiologists’ judgments, was used for laryngeal exposure, and the Cormack-Lehane levels were assessed by the attending anesthesiologists after tracheal intubation. A total of 15 attending anesthesiologists were involved in the present study. Levels 1 and 2 are classified as “easy,” whereas levels 3 and 4 are designated as “difficult.” If a difficult tracheal intubation (defined as an insertion of the endotracheal tube with conventional laryngoscopy requiring >2 attempts or lasting >10 minutes, or requiring an alternate technique) occurred in a patient, the event would be recorded too. In the clinical anesthesia routine procedure of our institution, no more than 4 intubation attempts via the application of conventional laryngoscope blades were permitted to ensure patient safety, and the operating time for each attempt was no longer than 1 minute. Before next intubation attempts, mask ventilation was used for at least 1 minute to ensure that the Spo₂ was 98% or higher.

Concealing the Results of Airway Assessments

Classically difficult airway evaluations were conducted by the 2 residents, the practitioners of the present study, and did not replace or interfere with the work of the responsible anesthesiologists. The sonography operator and responsible anesthesiologists were blinded to the results of these evaluations until the surgery was complete. Correspondingly, the consequences of sonography were also concealed from the responsible anesthesiologists and classical difficult airway evaluations operator.

Reliability Test

To decrease the variations between the 2 sonographers, the 2 sonographers were trained to examine dozens of volunteers. Finally, reliability tests between the 2 sonographers and between the left side and right side sonographic assessments were performed. Twenty volunteers were examined by the 2 sonographers independently. The volunteers were required to open their mouths maximally when performing condylar translation sonographic assessments and to not relay any further information to the other sonographer. Twenty volunteers were also examined by 1 sonographer in both sides.

Statistical Analysis

The SPSS software version 16.0 (SPSS, Chicago, IL) and the MedCalc version 12.7 (MedCalc Software, Mariakerke, Belgium) were used for the statistical analysis. The measurement data are displayed as the means ± standard deviation, and the categorical variable data are presented as numbers. Correlations between continuous variable data or ordinal category data were analyzed with the Spearman correlation analysis. The paired χ² (McNemar test) and internal agreement tests were performed between the condylar translation assessed by sonography and other predictive indicators that were associated with TMJ mobility. In addition, κ values were calculated. The same correlation analysis was conducted for all the predictors and Cormack-Lehane levels. The Youden index was used to determine criteria for condylar translation to predict difficult laryngoscopy. The odds ratios, sensitivity, specificity, positive predictive value, negative predictive value, and their 99% confidence intervals (99% CIs) for various indicators were calculated. The receiver operating characteristic curve and the area under the curve (AUC) were utilized to analyze the abilities to predict difficult laryngoscopy. The comparison of 2 AUCs was performed using the nonparametric test.¹⁷ The predictors significant in the univariate analysis and meted the collinearity diagnostics test were included in the multivariate logistic regression. Predictors that were significant according to multivariate logistic regression (P < .05) and without significant interaction effect (P > .05) were considered independent predictors. The AUC of the multivariable mode was calculated. A Krippendorff αvalue was utilized to evaluate the reliability between the 2 sonographers’ measurements and between both sides’ measurements, and a paired t test was used to analyze their difference. For the statistical analyses with 2-tailed tests, P < .01 was defined as statistical significance. In consideration of an estimated 9% to 10% incidence of difficult laryngoscopy in a previous investigation of our institution,¹⁸ a 0.2 difference value of AUC in the preliminary experiment, and a 90% statistic power, a sample size of at least 300, was calculated to detect the difference of receiver operating characteristic curves between condylar translation and other predictors.

RESULTS

Patients’ Outcome in Airway Management

Table 1.: Comparison of Patients With and Without Difficult Laryngoscopy

Figure 2.: Flow chart of the study.

During the observation period (January 2014 to June 2014), 594 patients were consecutively enrolled. Important missing values occurred in 87 patients, and 14 patients were excluded because a laryngeal mask was used, and 9 patients were excluded because of cancellation of the operation. Thus, 484 patients (219 men) were successfully included in this study in the final analysis. There were 41 patients with a Cormack-Lehane level 3 or 4, 5 patients with difficult tracheal intubation (see Figure 2, Table 1), and no case of impossible tracheal intubation. Among these difficult tracheal intubation patients, 3 cases were determined because of >2 attempts and 3 cases determined because of >10-minute intubation duration (because 2 criteria occurred in the same patient). All 5 patients were successfully intubated with video laryngoscope. Descriptive data of the patients and the airway assessment results are shown in Table 1.

Comparison Between Condylar Translation and the Other Predictors

Table 2.: Predictor Agreement With Limited Condylar Translation and Difficult Laryngoscopy (n = 484)

The condylar translation r (Spearman correlation coefficient) values to the other related TMJ mobility assessments, such as IID, upper lip bite test, mandibular protrusion distance, and whether the condyle-tragus distance was <1 finger was 0.41 (99% CI, 0.31 to 0.50), −0.32 (99% CI, −0.42 to −0.21), 0.30 (99% CI, 0.19 to 0.40), and 0.62 (99% CI, −0.68 to −0.54), respectively. The paired χ² and the agreement test results are shown in Table 2.

Comparison Between Cormack-Lehane Levels and Preoperative Predictors

The r values for all the observed variables, such as the condylar translation, IID, upper lip bite test, mandibular protrusion distance, whether the condyle-tragus distance was <1 finger, thyromental distance, and modified Mallampati test in comparison with the Cormack-Lehane levels were −0.46 (99% CI, −0.55 to −0.36), −0.45 (99% CI, −0.54 to −0.35), 0.21 (99% CI, 0.10 to 0.32), −0.17 (99% CI, −0.28 to −0.05), 0.24 (99% CI, 0.13 to 0.35), −0.36 (99% CI, −0.46 to −0.25), and 0.16 (99% CI, 0.04 to 0.27), respectively. The paired χ² and the agreement tests showed that the limited condylar translation had a significant κ value when compared with the difficult laryngoscopy (see Table 2).

Statistical Results for Variables to Predict Difficult Laryngoscopy

Table 3.: Variable Values to Predict Difficult Laryngoscopy (n = 484)

Figure 3.: Receiver operating characteristic curve analysis of airway assessment tests and their areas under curve (AUC) for predicting difficult laryngoscopy. *Comparison with the AUC of the other 3 airway assessment tests, P < .001. Calculated by nonparametric test. IID indicates interincisor distance; ULBT, upper lip bite test.

The receiver operating characteristic curve analysis showed that the AUC of the condylar translation, IID, upper lip bite test, mandibular protrusion distance, thyromental distance, and modified Mallampati test (excluding the whether the condyle-tragus distance was <1 finger because of the dichotomy characteristic) for predicting difficult laryngoscopy were 0.93 (99% CI, 0.90 to 0.96), 0.80 (99% CI, 0.75 to 0.84), 0.72 (99% CI, 0.66 to 0.77), 0.70 (99% CI, 0.64 to 0.75), 0.70 (99% CI, 0.64 to 0.75), and 0.72 (99% CI, 0.66 to 0.77), respectively (compared with AUC = 0.5, P < .001 for all of them). Significant differences were observed between AUCs of the condylar translation to the other predictors in predicting difficult laryngoscopy (Figure 3, P < .001 for all the comparisons). The condylar translation prediction criterion for difficult laryngoscopy determined by Youden index was the condylar translation ≤10 mm. The odds ratios, sensitivity, specificity, positive predictive value, negative predictive value, and their 99% CIs were calculated and are presented in Table 3. Among the variables that were significant in the univariate analysis and meted the collinearity diagnostics test, such as sex, age, condylar translation, IID, upper lip bite test, mandibular protrusion distance, thyromental distance, modified Mallampati test, and the whether the condyle-tragus distance was <1 finger, a multivariate logistic regression and interaction effect analysis showed that the sex (male), the modified Mallampati test (>2 grade), and the limited condylar translation (≤10 mm) were considered independent predictors and without significant interaction effect between them (P > .05 for all pairwise comparison of them). The AUC of this mode was 0.97 (99% CI, 0.94 to 0.99).

Results of Reliability Test

Twenty volunteers (11 males, 9 females) were examined by the 2 sonographers. The result of condylar translation was 14.0 ± 2.9 mm for 1 of the 2 sonographers’ measurements and 13.8 ± 2.6 mm for the other sonographer; 13.6 ± 2.5 mm for the left side and 13.9 ± 2.6 mm for the right side. Between the 2 sonographers’ measurements, the Krippendorff αvalue was 0.92 (99% CI, 0.82 to 0.98); between both sides’ measurements, the Krippendorff αvalue was 0.95 (99% CI, 0.88 to 0.99). There was no significant difference between the 2 sonographers with regard to their measurements (P = .204) and between both sides’ measurements (P = .665).

DISCUSSION

The present data showed that limited condylar translation is a meaningful TMJ mobility evaluation for predicting difficult laryngoscopy. The AUC comparisons demonstrated that, in predicting difficult laryngoscopy, limited condylar translation has the highest AUC. Limited condylar translation also presented considerable values in sensitivity and specificity. Because of the low incidence rate of the difficult airway, all the negative predictions of difficult airway remained high. However, for a positive predictive value, which is more important for anesthesiologists, all these values were low.^19,20 Interestingly, condylar translation has a considerable positive predictive value of 0.45, and it is a notable point.

Similar to a study by Chen et al¹⁰ and Landes and Sader,¹¹ our study also confirmed that condylar translation is a reliable TMJ mobility evaluation. Our study showed that sonography can determine accurate mandibular condylar translation measurements. The Krippendorff αvalues >0.9 showed that, no matter between the 2 sonographers or between both sides of a volunteer, the results of their examinations presented good repetitive stability.

There are multiple factors that contribute to difficult laryngoscopy, such as TMJ mobility, thyromental distance, modified Mallampati test, and body mass index. An AUC of 0.93 demonstrated that TMJ mobility plays an important role in determining a difficult laryngoscopy, because for the other factors, most of their AUCs were <0.8. There are several evaluation methods for TMJ mobility, such as IID, the upper lip bite test, and the qualitative or quantitative estimation of the condylar translation distance. In difficult airway predictions, our research showed that condylar translation might be a new independent predictor. Correlation analysis showed that only a low correlation coefficient was observed between the condylar translation and other variables. We analyzed the correlations between variables and difficult laryngoscopy to determine their ability to predict difficult airways. Our results showed that condylar translation had a considerable correlation with the Cormack-Lehane level among all our observed predictors. Previous research showed that the mandibular protrusion distance and IID were sensitive difficult airway predictors, and IID was significantly better than the mandibular protraction degree.³ This finding is also consistent with our observed data. However, our results further showed that the condylar translation’s predictive values might be better compared with IID. Considering the low correlation and little interaction between condylar translation and the other indicators, and the results of the multivariate logistic regression analysis, we determined that condylar translation may be a new valuable independent predictive variable. The specific and complete details of its predictive value still need further systematic research.

The assessing devices for TMJ mobility involved include rulers, radiography equipment, computerized tomography, magnetic resonance imaging, 3-dimensional optoelectronic jaw-tracking system, and so on.^4,21–23 Compared with the aforementioned methods, ultrasonic inspection may have the advantages of convenience and preciseness.^10,11,24,25 For condylar translation evaluations, the ultrasound method possesses good sensitivity, reliability, and accuracy.^10,11

The TMJ is a complex joint. There are 5 TMJ movement types such as anterior sliding, inferior sliding, lateral sliding, medial sliding, and rotating.¹⁰ In the mouth opening, movements involved mainly are the condylar rotating and anterior-inferior sliding movements.^4,10 Nevertheless, in the mandibular protrusion test and upper lip bite test movements, the condylar sliding movement is mainly involved, but not the rotating movement. In our data, the maximum condylar translation distance was less correlated with the IID, while Travers et al⁴ and Fukui et al⁵ showed that there was no significant correlation between condylar translation and IID. In addition, the mouth-opening angle and mandibular length were more highly correlated with the IID.²² As we know, the mouth-opening angle is mainly involved in rotary motion. These results showed that condylar rotation but not condylar translation plays an important role in the IID decision effect.⁴ These also suggest that in predicting difficult airways, the IID and condylar translation mechanisms are not exactly the same. To the best of our knowledge, when laryngoscopy is performed, the condyle should shift anterior-inferior. Does this translation distance equal or have a high correlation with the condylar translation distance upon mouth opening? To verify this, further investigation may be required.

Different from Chen et al¹⁰ and Landes and Sader¹¹ regarding their sonographic methods, the direction of the ultrasound probe was pointed toward the philtrum and slightly to the forward and downward direction in our sonographic methods. In the previous studies, the ultrasound probe was parallel to the zygomatic arch, which is similar to the horizontal direction. But in the condylar translation, the movement directions include the forward and downward directions.⁵ During sonographic detection, to obtain a condylar image during mouth closing and opening, we had to adjust the direction and height of the probe, ensuring a high tip arc for echo imaging instead of a round blunt arc echo. In our experience, a round blunt arc echo means the probe is lower than the condyle. Therefore, in our study, the probe pointing direction was situated slightly forward and downward, which is more in line with the trajectory and direction of the condylar translation.⁵ We believe that the forward and downward condylar translation motions reflect the physiologic motion characteristics, and this can also more directly influence the laryngoscopy. Unfortunately, we have not compared the efficiency of these tiny measuring direction changes with the measurement results.

There were some limitations in the current study. First, for enrollment, the patients needed to provide informed consent, the limited 2 ultrasound equipments and the limited 2 sonographers, making it difficult for us to enroll all the patients who met the inclusion criteria. Only the patients who were assigned to the operating rooms in specified area should be enrolled. Therefore, it is difficult for the subjects to be randomly selected. Although we did our best with blinding, there might remain some bias. Second, one of the current study aims was to primarily contrast the values of the different TMJ mobility assessments in predicting difficult laryngoscopy and to reveal the ability to predict difficult airways. Nevertheless, a difficult laryngoscopy is just 1 of the 5 aspects of difficult airways,⁸ and with the current limited observational data, it was hard to reveal the relationship of other difficult airway characteristics such as difficult intubations and difficult mask ventilations. To do this, a large systematic research study is needed. In addition, because of the lack of a stomatologist participating in this study, it was difficult to identify how many cases presented with clinical TMJ disorders and to reveal the correlation between TMJ disorders and difficult laryngoscopy. There also might remain bias in results because the patients who were with the baseline decreased TMJ function (eg, TMJ arthritis and osteoarthrosis, myofascial pain, postfacial radiation therapy, or known TMJ disorders) were not excluded from this study. Moreover, we lacked data to compare this sonographic examination with an airway aggregate scoring system to predict difficult laryngoscopy.

In summary, compared with indirect mandibular condylar mobility assessments, thyromental distance, and modified Mallampati test, condylar translation sonographic measurements were correlated with difficult laryngoscopy and exhibited an independent and better predictive property. These results may provide further prediction and evaluation guidance for difficult airways using ultrasound measurements. Moreover, a systematic investigation of this method in predicting difficult airway is required.

ACKNOWLEDGMENTS

We are thankful to our colleagues in the Department of Anesthesiology and Intensive Care Unit, Wannan Medical College First Affiliated Hospital, Yijishan Hospital, for help with samples.

DISCLOSURES

Name: Weidong Yao, MD.

Contribution: This author helped design the study, conduct the study, collect the data, analyze the data, and prepare the manuscript.

Name: Yumei Zhou, MD.

Contribution: This author helped design the study, analyze the data, and prepare the manuscript.

Name: Bin Wang, MD.

Contribution: This author helped conduct the study, collect the data, analyze the data, and prepare the manuscript.

Name: Tao Yu, MD, PhD.

Contribution: This author helped conduct the study and collect the data.

Name: Zhongbing Shen, MD.

Contribution: This author helped conduct the study and collect the data.

Name: Hao Wu, MD.

Contribution: This author helped conduct the study and collect the data.

Name: Xiaoju Jin, MD.

Contribution: This author helped design the study.

Name: Yuanhai Li, MD, PhD.

Contribution: This author helped design the study and prepare the manuscript.

This manuscript was handled by: Maxime Cannesson, MD, PhD.