Printer Friendly

Craniofacial and shoulder asymmetry and its relationship to temporomandibular disorder: a study of 183 patients.


We conducted a prospective study to ascertain the prevalence of features of craniofacial and shoulder asymmetry and to determine if they are related to temporomandibular disorder (TMD). Our study population was made up of 183 patients--105 females and 78 males, aged 8 to 92 years (mean: 53.5; median: 60) who presented to a private otolaryngologic practice in a rural retirement community in Arizona. These patients completed a questionnaire to determine their self-assessment of the their body asymmetry. Next, all patients underwent a nurse evaluation for the presence or absence of asymmetry, which was based on the relative position of the ears, lateral canthi, hemimandibles, and shoulders. All patients were then evaluated independently by a physician for the presence or absence of TMD. The most important finding of this study was that asymmetry as determined by the nurse evaluation was associated with a relative risk of TMD of 5.89 (p = 0.0001); the perception of asymmetry on the patient self-assessments was associated with a relative risk of only 1.86 (p = 0.0026). We conclude that the recognition and diagnosis of TMD is significantly enhanced by a brief evaluation by a health professional who has been trained in recognizing the signs of facial and shoulder asymmetry.


Although there is no simple or standard definition of temporomandibular disorder (TMD), it has been reported that 20 to 25% of the population is affected by the diverse signs and symptoms of this malady. (1) Signs of TMD appear in about 60 to 70% of the general population, but only 5% have symptoms severe enough to impel them to seek treatment. (2) The cardinal features of TMD include otalgia, cephalalgia, and restricted or irregular mandibular motion; others include tinnitus, vertigo, aural fullness, and hearing loss. (3) Even more specific symptoms--such as hyperacusis; a warm, wet, or stuffed-cotton sensation; and sensitivity to cold air--have also been described as a part of the aural manifestations of TMD. (4)

Craniofacial asymmetry (CFA) and internal derangement are interrelated to TMD, and the medicodental literature is replete with cephalometric radiographic image analyses documenting these relationships. (5,6) Typically, cephalometrics are not readily available to the clinician for determination of CFA as a possible prognosticator of TMD. However, even without the use of cephalometric studies, it appears that reliance on direct visual inspection of clinical patients can identify a large number of patients with TMD who have CFA and other body asymmetries.

Asymmetry can be detected in the frontal view of a patient by noting the relative positional differences of the ears, lateral canthi, and hemimandibles (figure 1). Frequently, patients with a CFA are observed to have or report that they have asymmetry elsewhere, such as in their shoulder height (figure 2), breast size (female), leg length, foot size, and the position of their iliac crests. Not all patients are aware of these variations in their bodies, but they may comment that their glasses are crooked, or one shoulder drops, or their beltline is crooked, or their shoe sizes are different.

In this article, we describe our study to determine if a useful relationship exists between the presence of observed asymmetries and the presence of TMD. We also investigate patients' self-perception of asymmetries.

Patients and methods

This study was conducted at a private otolaryngologic practice in a rural retirement community in Arizona. The study population was made up of 183 patients--105 females and 78 males, aged 8 to 92 years (mean: 53.5; median: 60)--who visited this practice during a 3-month period beginning in May of 2009. The demographic data reflected the fact this study was performed in a retirement town with an older population and in a clinic that has become a TMD referral center. All patients were told that their participation was voluntary, and all received an explanation of the nature of the study. Only a few patients who were asked declined to participate.

The study contained three components: a written patient self-assessment of body asymmetry in questionnaire form, a nurse evaluation of asymmetry based on a brief inspection, and a physician evaluation for the presence or absence of TMD.

Patient self-assessment. Patients were asked to fill out a 9-item questionnaire regarding their perceptions of any asymmetric features of their own bodies:

1. Do your glasses/sunglasses appear crooked on your face?

2. Is one ear higher on the side of your head than the other?

3. Is one corner of the eye higher than the other?

4. Is one shoulder higher?

5. Is one breast larger than the other? (females only) 6. Is your beltline higher on one hip than the other?

7. Is one leg longer?

8. Is one foot larger?

9. Do you have/have you been told you have scoliosis?

Patients were asked to choose the response that was most applicable to their particular situation; the choices were Do not know, Same, Right is higher (larger, longer) or Left is higher (larger, longer), and Yes or No. Also, a general question regarding the presence of back and/or hip pain was included to determine if increased skeletal pain had any association with body asymmetry or TMD.

Nurse evaluation. After the questionnaire was completed, an office RN examined each patient in an examination room. The nurse was blinded to the patient's responses to the self-assessment questionnaire. The nurse looked for asymmetry by evaluating six features and then recording the findings as either Same, Right, or Left for each item.

1. Elevation of the ear

2. Elevation of the lateral canthus

3. Asymmetric mandible with increased vertical height

4. Asymmetric mandible with increased horizontal distance at the corner of the mouth

5. Deviation of the mandible on opening

6. Shoulder elevation

Physician evaluation. Patients were then seen and evaluated independently by one of the authors (K.W.C.) to determine the presence or absence of TMD. The physician was also blinded to the results of both the patient self-assessments and the nurse evaluations.

The diagnosis of TMD was based on the patient's history and physical examination. Symptoms in the history that were significant in the support of a diagnosis of TMD included cephalalgia that increased during dental procedures or while chewing gum, jerky, bagels, etc.

Signs of TMD may include deviation of the mandible on excursion, restricted jaw motion, and temporomandibular joint (TMJ) crepitus. Myalgia and palpable dystonia of the temporalis, trapezius, and masseter muscles, as well as the upper half of the sternocleidomastoid muscle, are important signs indicating TMD. Pain on pressure in proximity to the lateral pterygoid muscles in an intraoral examination and evidence of bruxism are also indicative of TMD.

In addition, TMD patients typically describe ear symptoms, including a sensation of warmth, fluid, fullness, or "stuffed cotton" in their ear. Also, female patients frequently report the need to protect their ears from cold air because it is very painful. Despite these ear complaints (many of these patients have a history of unsuccessful treatment for "ear infections"), findings on ear examination are typically normal.

For our study, the presence of a strong history of TMD symptoms and at least 2 TMD signs was diagnostic. Based on these criteria, 103 of the 183 patients (56%) were found to have TMD. Their foci of pain were located in the jaw and back of the head with radiation into the neck and shoulder and into the temple. Overall, the pain was located in multiple sites among the entire group. Patients described it in different combinations of locations, which were generally lateral to the pupil line and not in the sagittal zone of the head between the eyes. Ear examinations were normal in all 103 patients.

Analysis of findings. For both the patient self-assessments and the nurse evaluations, the answers Right, Left, and Yes were recorded as a positive response. If a patient self-assessment yielded positive responses to 4 or more of the first 8 questions (50% positive response), then the entire self-assessment was considered to be positive (while any single positive response in itself is strongly suggestive of asymmetry, for this study we arbitrarily decided on a minimum of 4 positives). More stringency was applied to the nurse's findings, as positivity required positive responses to at least 4 of the o evaluated areas (67% positive response).

The data were grouped according to sex, and the male and female groups were further divided on the basis of the presence or absence of TMD. The Fisher exact test was used to statistically analyze the relationship between the presence or absence of asymmetry and the presence or absence of TMD. This test is similar to the chi-square ([chi square]) analysis, but it is more appropriate for the present-versus-absent data in this study. All calculations were computed on statistical software (JMP v. 7.0.2; SAS; Cary, N.C.).

The prevalence of each of the separate features of asymmetry evaluated by the nurse was tabulated for TMD and non-TMD patients, as was the associated relative risk of TMD for each feature. A relative risk greater than 1.0 indicated an increased risk compared with the overall population. The risks were determined by dividing the percentage of a feature of asymmetry in the TMD group by the percentage of the same feature in the non-TMD group.


Patient self-assessment. Of the 103 patients with TMD, 10 (9.7%) indicated that they had at least 4 features of facial and/or shoulder asymmetry. None of the 80 patients without TMD self-reported any asymmetry (table 1).

In the entire group of 183 patients, a perception of asymmetry was associated with a relative risk of TMD of 1.86 (p = 0.0026; 95% confidence interval [CI]: 1.62 to 2.14).

Nurse evaluation. The prevalence of asymmetry as determined by the nurse evaluation was much higher than the prevalence based on the patients' self-assessments. The nurse found asymmetry in 94 of the TMD patients (91%) and in 23 of the 80 non-TMD patients (29%) (tab]e 1).

The most significant finding of this study was that the presence of at least 4 asymmetric features as determined by the nurse evaluation was associated with a relative risk of TMD of 5.89 (p = 0.0001; 95% CI: 3.19 to 10.88). The nurse was significantly better than the patients in identifying asymmetry, as the two 95% CIs did not overlap.

In terms of sensitivity and specificity, the nurse evaluation was 91% sensitive and 71% specific, indicating that there were few false negatives and more false positives. The patients' self-assessments were only 9.7% sensitive but 100% specific. In other words, there were a fair number of false negatives, but no false positives.

Other factors. Female sex increased the relative risk of TMD to 3.51 (95% CI: 2.31 to 5.31). The ratio of female to male patients with TMD in our study (85 to 18) was 4.7 to 1, which is very consistent with data reported in the TMD literature.

The data on the prevalence of various specific asymmetric features observed by the nurse indicate the highest risk of TMD was associated with asymmetry of the hemimandibles (~2.5). The easiest feature to evaluate is ear asymmetry, but it is also common in non-TMD patients, and in our study it was associated with a relative risk of only 1.4. The risk of TMD in patients with asymmetries of the lateral canthi, mandible openings, and shoulder heights ranged from 1.7 to 1.9 (table 2).

Clearly, a single asymmetry is not as useful as the aggregate in determining TMD risk. A composite score of 4 or more on the nurse evaluation was at least 2.4 times (5.89/2.5) more predictive of TMD than was a score of only 1.

Finally, females with TMD had twice as much back pain and/or hip pain as did the females in the non-TMD group (59 of 85 [69%] vs. 7 of 20 [35%]); there was no association in males. Also, there was no association between TMD and scoliosis in either females or males.


Sensitivity to the features of asymmetry that can be detected in patients is an acquired attribute, if one can simply remember to look for them. These features also provide a simplified model of the pathogenesis of TMD. The TMJ is located immediately anterior to and at the same level as the external auditory meatus. For the mandibles to compensate when the position of one ear is different from that of the other, the hemimandibles must be asymmetric and/or the mentum must be displaced from the midline. The hemimandible of the higher ear is frequently noted to be longer and flatter, while the hemimandible of the lower ear is more angulated and shorter.

The amount of asymmetry noted both in the mandible and in the position of the TMJ should be directly related to (1) the course and degree of deviation on excursion of the mandible, (2) the variation of joint pressure-loading during mastication, and (3) the potential for TMJ internal derangement, synovitis, and dysfunction.

Some of the chemical, neuromuscular, vascular, and direct inflammatory pathways that possibly lead to secondary myalgia, dystonia, and aural symptoms in contiguous structures have been studied. (7) It should also be mentioned that only one apparent relationship between locations of asymmetry and the presentation of TMD symptoms was noted in our study. Anecdotally, the side associated with the smaller hemimandible was usually the side that had the preponderance of pain and aural TMD symptoms.

A significant study of CFA in humans was reported by Rossi et al in Brazil. (8) Their cephalometric analysis of the skulls of fetuses, infants, children, and adults demonstrated that each of these four age groups had the same incidence of CFA. Their work is relevant to our study because they reported that statistically significant CFAs occurred in predentition fetuses and infants to the same extent as they did in children and adults. If a CFA occurs before the development of masticatory function and malocclusion, it refutes the hypothesis that CFAs originate as dental disorders. Based on the findings of Rossi et al, (8) it appears that CFAs precede TMDs rather than TMDs being the cause of CFAs.

Radiologic cephalometric data have demonstrated that asymmetry is present to some degree in all anatomic landmarks in all patients. (9) However, if it is reasonable that the origin of a TMD is related to the extent of a CFA, then we can assume that TMDs can arise from genetic and/or environmental anomalies during embryonic development. A lack of structural symmetry in the cranium and mandible would result in a CFA, and in a similar manner, it would account for asymmetry of the shoulders, iliac crests, legs, and feet.

Absolute symmetry has been the goal of aesthetic orthodontic procedures and orthopedic procedures. Perhaps the initial and ultimate treatment for TMD, and even other orthopedic disorders, will entail closed remodeling of craniofacial and skeletal asymmetries in the malleable framework of the neonate.

The primary goal of our study was to raise the index of suspicion for TMD as a way to aid clinicians in recognizing it when it is present. The recognition and treatment of TMD is important not only to the patient's well-being, but also for reducing the high costs associated with treating its manifestations.

In one analysis of the database of a large managed care organization, the difference in utilization of inpatient and outpatient services between TMD patients and non-TMD patients was as high as a factor of 3 to 1. (10) Part of the reason for the increased cost is the failure of physicians to initially recognize TMD and instead begin a workup for possible sinusitis, ear infections, neoplasms, etc., with magnetic resonance imaging, computed tomography, and antibiotic prescriptions on a best-guess basis.

As demonstrated in our study, it is significant that an office nurse can detect a six-fold increase in risk for TMD by simply spending several seconds analyzing a patient's asymmetric features. It is our hope that when a patient presents with lateral cephalalgia, aural symptoms of TMD, and a normal ear examination, TMD will be a part of the differential diagnosis. And if a few of the signs of patient asymmetry described here were to be recognized, the diagnosis of TMD by healthcare providers would become even more apparent.


(1.) Detamore MS, Athanasiou KA, Mao J. A call to action for bioengineers and dental professionals: Directives for the future of TMJ bioengineering. Ann Biomed Eng 2007;35(8): 1301-11.

(2.) Dimitroulis G, Dolwick MF, Gremillion HA. Temporomandibular disorders. 1. Clinical evaluation. Aust Dent J 1995;40(5):301-5.

(3.) Kuttila S, Kuttila M, Le Bell Y, et al. Aural symptoms and signs of tempormandibular disorder in association with treatment need and visits to a physician. Laryngoscope 1999;109(10):1669-73.

(4.) Cox KW. Temporomandibular disorder and new aural symptoms. Arch Otolaryngol Head Neck Surg 2008;134(4):389-93.

(5.) Buranastidporn B, Hisano M, Soma K. Effect of biomechanical disturbance of the temporomandibular joint on the prevalence of internal derangement in mandibular asymmetry. Eur J Orthod 2006;28(3):199-205.

(6.) Trpkova B, Major P, Nebbe B, Prasad N. Craniofacial asymmetry and temporomandibular joint internal derangement in female adolescents: A posteroanterior cephalometric study. Angle Orthod 2000;70(1):81-8.

(7.) Ramirez Aristeguieta LM, Sandoval Ortiz GP, Ballesteros LE. Theories on otic symptoms in temporomandibular disorders: Past and present. International Journal of Morphology 2005;23(2):141-56.

(8.) Rossi M, Ribeiro E, Smith R. Craniofacial asymmetry in development: An anatomical study. Angle Orthod 2003;73(4):381-5.

(9.) Arnold TG, Anderson GC, Liljemark WE Cephalometric norms for craniofacial asymmetry using submental-vertical radiographs. Am J Orthod Dentofacial Orthop 1994;106(3):250-6.

(10.) Shimshak DG, DeFuria MC. Healthcare utilization by patients with temporomandibular joint disorders. Cranio 1998;16(3):185-93.

Kent W. Cox, MD, PhD; Rob Klein, MD

From the Department of Surgery, Summit Health Care Regional Medical Center, Show Low, Ariz. (Dr. Cox); and the Department of Pathology, University of Arizona Medical College, Tucson (Dr. Klein). The study described in this article was conducted at a private otolaryngologic practice in a rural retirement community in Arizona.

Corresponding author: Kent W. Cox, MD, PhD, Summit Health Care Regional Medical Center, 2200 Show Low Lake Rd., Show Low, AZ 85901. Email:

Previous presentation: This study was originally presented at the TMJ Bioengineering Conference; Nov. 6, 2009; Broomfield, Colo.

Table 1. Distribution of asymmetry based on patient self-assessments
and nurse evaluations according to TMD status


                       Present   Absent

    Females (n = 85)      9        76
    Males (n = 18)        1        17
    Females (n = 20)      0        20
    Males (n = 60)        0        60

Nurse evaluations
    Females (n = 85)     78        7
    Males (n = 18)       16        2
    Females (n = 20)      8        12
    Males (n = 60)       15        45

Table 2. Specific types of asymmetry based on nurse evaluations
according to TMD status

                          TMD status, n (%)

                         Positive    Negative   Relative
                         (n = 103)   (n = 80)     risk

Elevation of the ear        92          50        1.4

Elevation of the            91          42        1.7
lateral canthus

Deviation of the            79          35        1.8
mandible on opening

Elevation of the            84          34        1.9

Increased vertical          60          19        2.4
height in hemimandible

Increased horizontal        60          15        3.1
width in hemimandible
COPYRIGHT 2013 Vendome Group LLC
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2013 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Cox, Kent W.; Klein, Rob
Publication:Ear, Nose and Throat Journal
Article Type:Report
Geographic Code:1USA
Date:Mar 1, 2013
Previous Article:A prospective study of parents' compliance with their child's prescribed analgesia following tonsillectomy.
Next Article:Necrotizing cervical fasciitis: Prognosis based on a new grading system.

Terms of use | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters