| Sign In to gain access to subscriptions and/or personal tools. |
Risk Factors for Diagnostic Subgroups of Painful Temporomandibular Disorders (TMD)
1 Department of Orthodontics, Box 357446, Correspondence: *corresponding author, ghuang{at}u.washington.edu
Temporomandibular Disorders (TMD) encompass several entities, which may have differing etiologies. To test this hypothesis, we investigated risk factors for three diagnostic subgroups of painful TMD. Ninety-seven subjects with myofascial pain only, 20 with arthralgia only, 157 with both myofascial pain and arthralgia, and 195 controls without TMD pain met criteria for study eligibility. Investigated risk factors included both physical and psychological variables. Adjusted odds ratios were calculated by multiple logistic regression analyses. Myofascial pain occurring alone was significantly associated with trauma (Odds Ratio {OR} = 2.0), clenching (OR = 4.8), third molar removal (OR = 3.2), somatization (OR = 3.7), and female gender (OR = 4.2). Myofascial pain with arthralgia was significantly associated with trauma (OR = 2.1), clenching (OR = 3.3), third molar removal (OR = 4.0), somatization (OR = 5.1), and female gender (OR = 4.7). No significant associations were found for the small-arthralgia-only group.
Key Words: TMD • risk factors • temporomandibular disorders • subgroups
In 1996, the NIH held a Technology Assessment Conference to provide an "assessment of management approaches to temporomandibular disorders (TMD)" (NIH, 1996). Conference participants concluded that TMD is a collection of conditions affecting the temporomandibular joint and contiguous structures, and that further investigation of risk factors associated with TMD was needed. A systematic review of potential risk factors was recently performed, with two criteria being identified as important in allowing for comparison between studies (Drangsholt and LeResche, 1999). First, pain should be used as a major aspect of case definition, and second, the effects of both age and gender should be taken into account. Of the several hundred articles that were reviewed, only ten studies fulfilled these two criteria (Von Korff et al., 1988, 1993; Hackney et al., 1993; Bibb et al., 1995; Lee et al., 1995; McGregor et al., 1996; Sato et al., 1996; Kitai et al., 1997; LeResche et al., 1997; Molina et al., 1997). Based upon these ten studies, depression, number of pre-existing pain conditions, and factors associated with female gender were identified as possible risk factors. However, none of these studies reported analyses based on TMD subgroups. If TMD represents a collection of conditions, as concluded by the NIH conference, then categorization of these different conditions may be important for the more accurate identification of risk factors. Thus, the purpose of this study was to investigate risk factors for TMD after the classification of subjects into diagnostic subgroups of painful TMD, while controlling for age, gender, and other potentially confounding variables.
Subjects in this study were recruited from among enrollees of the Group Health Cooperative of Puget Sound, a large HMO in the greater Seattle area, during the years 1985 and 1986. At that time, the number of enrollees was approximately 320,000, representing a population that was demographically similar to that of the greater Seattle area (Von Korff et al., 1988). Enrollees who were consecutively referred to one of two Group Health TMD clinics were recruited as Clinic Cases (n = 289), of which 261 (90%) agreed to participate. Community Cases and Community Controls were identified through a screening questionnaire sent to an age-stratified probability sample of 1265 adult Group Health enrollees, of which 1016 (80%) responded. The question which distinguished Community Cases from Controls was, "During the past six months, have you had a problem with facial ache or pain in the jaw muscles, the joint in front of the ear, or inside the ear (other than infection)?" (Dworkin et al., 1990). One hundred twenty-three subjects answered positively and were classified as Community Cases. Of 893 who answered negatively to the screening question, 264 (30%) were randomly chosen to serve as Community Controls. Informed consent, which was reviewed and approved by the University of Washington Institutional Review Board, was obtained from all subjects who participated in the study. Of the participants, 261 Clinic Cases, 121 Community Cases, and 210 Community Controls underwent an interview and clinical TMD examination, conducted in the field by trained and calibrated registered dental hygienist examiners. Based on data collected during the interview and clinical examination, cases were classified into three groups of painful TMD, based on the Research Diagnostic Criteria for TMD (RDC/TMD) (Dworkin and LeResche 1992): one group with myofascial pain only, one with arthralgia only, and one with both myofascial pain and arthralgia. These entities are operationally defined in the RDC/TMD by the presence of specific combinations of signs and symptoms gathered through standardized examination procedures. (It should be noted that these data were collected prior to the development of the RDC/TMD. The examination specifications for this study were identical to those of the RDC/TMD with the following exceptions: (1) The middle temporalis muscle was not palpated, and (2) a click was defined as present if it was detected on a single mandibular movement, instead of on two out of three movements as specified in the RDC/TMD.) Specific risk factors of interest included self-report of facial trauma ("Have you ever been hit, had a car accident, a sports injury or other accident where you received a hard blow or bang to your jaw or face?"), clenching ("Do you hold your teeth together, even lightly, or clench your jaw?"), oral habits ("Do you chew your fingernails, bite pencils, chew gum, or hold the telephone receiver between your shoulder and jaw?"), orthodontic treatment, third molar removal, recent dental treatment, and stress ("Would you describe yourself as being under stress all of the time, most of the time, sometimes, rarely, or never?"). Somatization (pain items excluded), anxiety, and depression were measured with scales of the SCL-90 as described in the RDC/TMD. The Clinic and Community Cases were combined in analyses of the three TMD subgroups, since subjects from both groups were required to meet the more stringent criteria described by the RDC/TMD. The chi-squared test was used for the assessment of differences between the pain and control groups in distributions of exposures and potentially confounding variables. Statistical significance was assessed at the 0.05 level, and then adjusted for multiple comparisons by the Bonferroni method. Adjusted odds ratios (OR) were calculated from multiple logistic regression models that included the relevant risk factors and covariates, in addition to age. Temporal relationships between risk factors and age at first onset of TMD pain were explored.
Of the 592 individuals interviewed and examined, the RDC/TMD criteria classified 469 subjects into one of the following four groups: pain-free controls (n = 195), myofascial pain only (n = 97), arthralgia only (n = 20), and myofascial pain with arthralgia (n = 157). Of the 123 excluded subjects, 93 cases did not meet diagnostic criteria for myofascial pain or arthralgia, 15 cases had missing data for myofascial pain or arthralgia, 13 controls did not report facial pain but met diagnostic criteria for arthralgia, and two controls had missing data for myofascial pain or arthralgia. Disc displacement diagnoses were not an exclusion criterion, and 31% of the subjects in the three pain groups had diagnoses of disc displacement, as did 12% of the controls.
The TMD groups and controls were similar with respect to age, race, median income bracket, and marital status (Table 1
A high proportion of subjects with myofascial pain (with or without arthralgia) reported clenching, facial trauma, third molar removal, stress, somatization, depression, and talking with the telephone resting on the shoulder (Table 2  ). No significant associations were found for the arthralgia-only group.
In multivariate analyses with simultaneous adjustment for the presence of each risk factor, increased risk of myofascial pain (with or without arthralgia) was associated with facial trauma, clenching, third molar removal, somatization, and female gender (Table 3 ). These relationships were not altered by adjustment for race, marital status, income, or education. Among cases reporting previous facial trauma, 82% of subjects reported the trauma to have occurred prior to the first onset of TMD pain. Similarly, 76% of cases with third molars removed stated that the extractions occurred prior to the first onset of TMD pain. In subanalyses comparing controls and cases with trauma or third molar extraction prior to first onset of TMD pain, odds ratios were decreased but remained significant for third molar removal in both myofascial pain groups (Table 4).
In this study, we investigated risk factors associated with subgroups of painful TMD, distinguishing between muscular (myofascial) pain and joint pain (arthralgia). However, a considerable number of subjects experienced both types of pain. Risk factors associated with myofascial pain and myofascial pain with arthralgia appear to be similar in type and magnitude. Moreover, to our knowledge, this is the first study to demonstrate an association between third molar removal and increased risk of TMD, after adjustment for potentially confounding variables. We combined the Clinic Cases and Community Cases in our final analyses based on the following factors. First, all subjects in both groups met diagnostic criteria specified by the RDC/TMD. Second, minimal differences in demographic and clinical characteristics between the groups were observed (Dworkin et al., 1990). Third, associations between risk factors and TMD were similar in direction and magnitude in separate analyses of the Clinic Cases and Community Cases. Thus, pooling these groups increased our study power and the precision of our risk estimates. Several studies have categorized TMD into subgroups similar to those in this study. Studies that examined psychological differences between subgroups (Marbach and Lund, 1981; Eversole et al., 1985; Lundeen et al., 1987; Dworkin et al., 1989; McCreary et al., 1991) generally indicated that patients with myogenic diagnoses had more pain and distress than those with joint-related diagnoses. In a study comparing subjects with disc displacement and those with myogenic disorders (Isacsson et al. 1989), patients with permanent disc displacement reported more pain in the temporomandibular joint in association with jaw function, while those with myogenic pain reported more bruxism and dental discomfort. Pullinger investigated trauma history in association with six diagnostic categories of TMD (Pullinger and Seligman, 1991), and found that a history of trauma was reported by more than 50% of subjects with disc displacement with reduction, disc displacement without reduction, or myalgia only. He suggested that the high incidence of trauma among patients with myalgia only provides evidence that myalgia is not always stress-related. Last, in a study dividing TMD into internal derangements with and those without reduction, a myogenous category, and osteoarthrosis, Lobbezoo-Scholte et al. (1995) found that these four groups were similar for trauma history, occlusal factors, and clinical evidence of parafunction. However, subjects in the myogenous group more often reported parafunction, depression, and worrying. Overall, these studies suggest that subjects with muscular diagnoses have more severe pain and psychological distress than those with joint diagnoses. If this is true, then muscular pain may overshadow joint pain. The similarity in risk factors among subjects with myofascial pain, with and without arthralgia, observed in our study lends support to this hypothesis. Unfortunately, the small number of subjects with arthralgia only prevented meaningful comparison with subjects with myofascial pain only. Our findings regarding specific TMD risk factors are generally consistent with those from previous studies, and support the multifactorial theory of TMD etiology currently proposed by several TMD researchers (Clark, 1991; Greene, 1995). Facial trauma, clenching, somatization, and female gender are commonly reported risk factors. However, a potential role for third molar surgery in the etiology of TMD has been infrequently reported (Greene et al., 1969; Butler et al., 1975; Pullinger and Monteiro, 1988; Raustia and Oikarinen, 1991). Furthermore, the only study to include a comparison group to assess this association was that by Pullinger and Monteiro. Biologic plausibility is one of several factors that should be considered in the assessment of evidence for causation. It is well-known that trauma results in injury and pain. Clenching can be considered a form of microtrauma that results in pain (Okeson, 1998). Third molar removal may involve wide opening of the mouth, application of considerable forces to the mandible, and, if performed under general anesthesia, a reduction in a subject's protective mechanisms. Any of these could result in trauma to the TMJ or the muscles of mastication. The association of TMD with third molar removal merits further study, since this procedure is commonly performed in the United States. Some researchers have advocated a more conservative stance on the removal of asymptomatic third molars, and if third molar extraction does contribute to an increased incidence of TMD, their argument would be further strengthened (Tulloch et al., 1990). High somatization scores suggest the tendency to report distress related to numerous physical symptoms, including pain, and as such, may indicate a generalized heightened perceptual sensitivity. Biological, cultural, hormonal, or environmental factors acting alone or in combination may be responsible for the observed association between TMD and female gender. We attempted to assess the temporal relationship between trauma and third molar removal by performing secondary analyses restricting cases to those reporting trauma prior to the first onset of TMD. However, because many subjects experienced recurrent TMD pain for which temporal data were unavailable, we were unable to fully assess the impact of trauma or third molar removal. A prospective design would allow for a more definitive assessment of these risk factors. Our study had several strengths, including the population-based sample of cases and controls; reliable, criterion-based examination of all subjects; and adjustment for potentially confounding variables. We believe that the results of this study are generalizable to other populations with similar demographic characteristics. Limitations include the potential for recall bias, in that cases might be more likely to remember specific exposures related to TMD. In an effort to minimize this, we first questioned subjects about exposures and then about facial pain. Thus, exposure history was ascertained prior to the recall of TMD history. We were also limited by the small number of subjects with arthralgia only. While arthralgia was a common diagnosis, it almost always occurred in combination with myofascial pain. In summary, trauma, clenching, third molar removal, somatization, and female gender were identified as risk factors for subjects with myofascial pain, as well as for subjects with concurrent myofascial pain and arthralgia. Although this study helps to clarify our understanding of risk factors for diagnostic subgroups of painful TMD, further studies are warranted to clarify the temporal sequence of risk factors, as well as the mechanisms accounting for the association between TMD pain and female gender. In addition, studies with larger numbers of subjects with only arthralgia are needed. A more detailed understanding of specific risk factors may be helpful in the development of targeted approaches to TMD therapy and prevention.
We thank Dr. Samuel Dworkin, Principal Investigator of the epidemiologic study through which these data were collected. This paper is based on a thesis submitted to the graduate faculty, University of Washington, in partial fulfillment of the requirements for the MPH degree. This investigation was supported in part by NIDCR Grants DE07227 and DE08773. Received for publication June 11, 2001. Revision received November 19, 2001. Accepted for publication February 12, 2002.
Journal of Dental Research, Vol. 81, No. 4,
284-288 (2002) This article has been cited by other articles:
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
