Temporomandibular Disorders – the Evolution of Controversy

The origin of the notion of the possibility of temporomandibular disorders (TMDs) dates back to the time of the ancient Egyptians (McNeill, 1997). What we refer to as TMDs today was referred to in the past by a vast array of terms such as: temporomandibular joint (TMJ) pain dysfunction syndrome, myofascial pain dysfunction syndrome, craniomandibular articulation disorders, mandibular dysfunction, TMJ disorders, and TM disorders (Rinchuse, 1987). The modern history of TMDs appears to have started in 1934. Based on the analysis of 11 cases, an otolaryngologist, Dr James Costen (1934), described a syndrome (Costen’s syndrome) related to the TMJs and ears. The etiology was believed to be related to overclosure of the mandible due to loss of dental vertical dimension subsequent to tooth loss. Symptoms of Costen’s syndrome include pain in and around the jaw, TMJ sounds, limited mandibular opening, and myofascial tenderness/pain, as well as ear symptoms such as dizziness, tinnitus, pain and impaired hearing. The ear/hearing effect was assumed to be due to the close anatomical proximity of the TMJ to the external auditory meatus. A decade later, a famous anatomist, Dr Harry Sicher (1948), proved Costen’s syndrome as fallacious from an anatomical viewpoint. Even though Costen’s proposals were disproved, it certainly initiated great interest and awareness of the TMJs among dentists. Sved (1944), Block (1947), Christensen (1970), and others, however, still supported aspects of Costen’s concepts and advocated the use of bite-raising appliances to restore the vertical dimension to alleviate any TMJ/ear symptoms. Over time modifications of these appliances eventually led to the use of oral occlusal appliances or splints for TMD management, and other bite- and joint-related issues.

In 1927, McCollum founded the Gnathological Society of California, and together with colleagues Stuart and Stallard, developed and taught the principles of gnathology, which involved harmonizing occlusal and jaw relationships for ideal dental and TMD treatment. Originally, gnathology relied on the principles of science and much useful information was gleaned from studies involving chewing kinematics (Guttetman, 1961; Sheppard and Sheppard, 1965; Ahlgren, 1967; Gillings et al., 1973; Gibbs et al., 1977; Wickwire et al., 1981; Alexander et al., 1984) and intra-oral telemetry (Graf and Zander, 1963; Pameijer et al., 1969, 1970; Glickman et al., 1970). Nonetheless, gnathology has moved away from the tenets of science to those emphasizing mechanistic, and often instrument-driven procedures.

Many well-regarded clinicians (Schuyler, 1929; Posselt, 1952; Mann and Pankey, 1963; Ramfjord and Ash, 1971) endorsed the principles of gnathology, and/or modified versions of it with their own improved instrumentation. This development undoubtedly fed the growing belief in the importance of occlusion and occlusal adjustment to the overall health of the stomatognathic system. Others such as Dawson (1974), Guichet (1979), Williamson et al. (1980) and Roth (1981a, b) popularized various techniques and recommended occlusal equilibration as an important adjunctive treatment to occlusal therapy as well as the prophylactic use of equilibration to prevent TMD.

When current evidence discredited existing notions and/or beliefs, occlusionists were quick to put new spins on old ideas. In the 1970s and early 1980s, many originally in the occlusionist camp abandoned the belief that occlusion was the primary cause of TMD. Their mantra now became that TMJ dysfunction, or craniomandibular dysfunction, was caused by eccentric (or displaced) TMJ condyles rather than discrepancies in occlusal relationships. Dentists were now orthopedists of the stomatognathic system and instead of calling the appliances used in TMJ treatments occlusal splints, they were now referred to as dental orthotics. The focus became trying to determine the most optimal condyle (or centric relation [CR]) position and working out the best method of recording it and then directing dental treatments toward it. Gelb (1977), a famous New York dentist, even went so far as to postulate that his mandibular orthopedic repositioning appliances (MORA splints)/orthotics, or any device that discluded the teeth such as an athletic mouthguard, could increase overall athletic performance. Early on, several anecdotal reports and uncontrolled and unblinded studies were used to verify this tenet. However, when blinded, controlled studies were eventually conducted, which neutralized many of the placebo/psychological effects due to subject/investigator/appliance interaction, no benefit of MORA on athletic strength/performance was found (Friedman and Weisberg, 1981; Greenberg et al., 1981; Hart et al., 1981). Nonetheless, aspects of this notion seem to resurface time and again.

In the 1970s, Jankelson (1979, 1984) refuted the gnathological ideals of the time and developed his own method of determining and recording CR or myocentric position with an instrument termed the ‘Myomonitor’. The myocentric position was believed to be the optimum neuromuscular jaw relationship to serve as the basis for occlusal and TMD treatment. This myocentric position was typically found somewhere between centric relation occlusion (CRO, as per the older definition) and centric occlusion (CO). When natural dentitions did not adhere to specific myocentric or gnathological ideals, the followers of the associated philosophy would consider the dentition as not ideal and potentially pathological. Dao et al. (1988) found that the ‘Myomonitor’ only stimulates the distal axons of the motor neurons peripherally and not centrally, therefore only producing a simple superficial masseter muscle twitch as opposed to producing a reflex activation of the jaw closing muscles through stimulation of the motor root of the trigeminal nerve as originally purported (Jankelson et al., 1975). As a result of their findings, they questioned whether a physiological occlusal position could be established with this method.

With the development of better TMJ imaging techniques in the mid to late 1970s, Farrar and McCarty (Farrar, 1972, 1979, 1982–1983; McCarty, 1979) focused on TMJ internal derangements as the primary cause of TMJ dysfunction. Most of the TMJ internal derangements were found to be anterior-medial displacements of the TMJ disks rather than posterior disk displacements. They advocated recapturing displaced disks with the use of anterior repositioning appliances and/or TMJ surgery. Interestingly, the internal derangement camp was not able to explain the original cause of the TMJ internal derangements, although there was a focus on TMJ trauma.

The introduction of computed tomography (CT) and magnetic resonance imaging (MRI) in the mid-1980s significantly improved imaging of the TMJs. With better imaging techniques and devices, many TMJ surgical procedures were carried out during that era, with the focus on structural repair of the joint and disk and eventually replacement of the disk with autogenous or alloplastic materials. For the most part, these aggressive surgical treatments failed miserably. Around this time functional orthodontists, mostly not university trained specialists, such as Witzig and Spahl (1991), became strong advocates of advancing the mandible with the use of functional appliances with a view to encouraging mandibular growth and establishing specific occlusal and jaw relationships for the management or prevention of TMD. They also believed that ‘traditional orthodontics’ involving premolar extractions, headgear, and incisor retraction predisposed patients to present and future TMD.

By the 1980s there were many diverse views of TMJ dysfunction as related to cause and treatments, i.e. occlusion, condyle (CR) position, and/or TMJ internal derangements. With so many different views and controversies related to TMJ dysfunction, the American Dental Association (ADA) held a conference in June 1982. In 1983, the ADA published guidelines derived from the conference for the examination, diagnosis, and management of dysfunctions of the TMJ (Griffiths, 1983), which were now defined as temporomandibular disorders. TMDs were, and are now, considered to be a collection of musculoskeletal disorders/dysfunctions affecting the TMJ complex and neighboring structures of the head and neck, embracing many clinical conditions with many different and diverse causes. Incidentally, the ADA then held a second conference in 1989 (McNeill et al., 1990) and the National Institutes of Health held a conference in 1996 (National Institutes of Health Technology Assessment Conference Statement, 1996) to further address the issues surrounding TMD. Much of what was determined in the first ADA TMD Conference held up well and was reinforced in the subsequent conferences. There are several TMD classifications, but for the purpose of an overview, a summarized and simplified version is provided in Box 24.1 (Griffiths, 1983; Dworkin and LeResche, 1992; Okeson, 2005).

Although the prevalence of minor disturbances of the TMJ area is common in the healthy population (popping and clicking), epidemiological studies suggest that approximately 5–12% of the population may be significantly affected by TMDs or require some form of treatment (Solberg et al., 1979; Dworkin et al., 1990). Today, TMD has moved away from the historic, mechanical, dental-based model, which primarily focused on occlusal modification or jaw repositioning (Greene and Laskin, 2000; Greene, 2001; Klasser and Greene, 2007a) to a biopsychosocial model which integrates a host of biological, clinical and behavioral factors that account for the onset, maintenance and remission of TMD (Fernandez and Turk, 1989; Dworkin and LeResche, 1992; Flor and Birbaumer, 1993; Turk et al., 1993, 1996; Dworkin and Massoth, 1994; Dworkin et al., 1994; Rudy et al., 1995; Greco et al., 1997; Mishra et al., 2000; Gardea et al., 2001; Dworkin et al., 2002a,b; Gatchel et al., 2006). Cognitive behavioral therapies (CBT) and biofeedback (BFB) are becoming recognized as the standard in the early therapy of TMD (Gardea et al., 2001; Gatchel et al., 2006). See Table 24.1 for a general summary of TMD controversies.

Table 24.1 The judge and jury of temporomandibular disorder (TMD) controversies^a

^aThe above is based on population data and may not apply to all clinical situations at the individual patient level.

Recent information indicates that TMD patients exhibit greater sensitivity to experimental pain than control subjects, suggesting that nociceptive processing may be upregulated in TMD patients. The generalized hyper-excitability of their central nociceptive pathways may contribute to the development and/or maintenance of chronic TMD pain and possibly increase the likelihood of other chronic pain disorders. Further, it appears that women exhibit greater excitability of their central nociceptive neurons than men, making the female central nervous system more easily upregulated to a pathologically hyperexcitable state, contributing to the greater prevalence of various chronic pain conditions among women (Sarlani and Greenspan, 2005). Moreover, there appears to be a genetic predisposition among individuals who experience chronic pain.

In addition, it has been suggested that variations of the gene encoding for catechol-O-methyl-transferase (COMT) is associated with influencing pain regulatory mechanisms in the central nervous system. COMT is one of several enzymes that degrade catecholamines (i.e. neurotransmitters) such as dopamine, epinephrine, norepinephrine, and enkephalins. Three common COMT haplotypes have been found to account for some variability in experimental pain perception in females. These markers may in the future allow us to identify a subgroup of subjects who are at a higher risk of developing TMD (Diatchenko et al., 2005; Slade et al., 2008). These data and findings reinforce the idea that there is a paradigm shift regarding the etiology of TMD; moving away from a focus on ‘chasing’ occlusal contacts to ‘chasing’ vulnerability alleles (Stohler, 2004).

Orthodontics and TMD

It was not until the early 1970s that a serious consideration was made of a possible relationship between orthodontics and TMD. Arguing an orthodontic gnathological view, Roth (1973, 1976) maintained that certain functional occlusion and centric relation standards needed to be achieved in orthodontic treatments. His view was taken from the classic gnathological–prosthodontic philosophy which fostered the concepts of canine protected occlusion, retruded CR, anterior guidance, and the use of articulators in orthodontic diagnostics. Roth believed that orthodontic treatments that did not foster traditional gnathological functional goals would cause and/or predispose orthodontic patients to TMD. Purveyors of this thinking reasoned that orthodontic treatment is analogous to doing full-mouth occlusal rehabilitations, with the difference being that orthodontists did not ‘cut’ or modify the natural tooth structure.

Orthodontics was viewed as a cause of TMD from two perspectives. First, orthodontists who did not achieve a gnathological occlusal finish were believed to produce iatrogenic functional occlusions (i.e. functional balancing interferences) and/or eccentric condyle (or CR) positions. This then led to TMD. Second, certain orthodontic appliances or techniques (e.g. Class II or III mechanics, headgears, extractions, chin cups, certain retainers, and so on) were believed to directly cause TMD (Witzig and Spahl, 1991; Wyatt, 1987).

In 1987, a landmark US court case took place, which resulted in an unfavorable judgment against a Michigan orthodontist for purportedly causing TMD in a 16-year-old girl with a Class II Division 1 malocclusion. The orthodontic treatment involved maxillary first premolar extractions and the use of a headgear. The patient filed a complaint against the orthodontist alleging that following the removal of the appliances she developed severe clicking in the TMJ with pain and crepitus. The theory was that certain orthodontic appliances and procedures could cause distal displacement of the mandible leading to TMJ internal derangements, i.e. anterior displaced TMJ disks. In this case it was alleged that the distal displacement of the mandible was due to the over-retraction and lingual inclination of the maxillary incisors. An expert witness for the plaintiff, a prominent and vocal functional orthodontist, Dr John W Witzig testified in part as follows:

In Susan’s case with her type of malocclusion or her type of problem when she went to see the orthodontist, no way should headgear and retraction of the upper front teeth back toward the tongue have occurred. This left Susan with a bite that her lower jaw now bites in a displaced position, and no way should any patient be left in that condition – this is dental negligence.

JW Witzig, cited in Luecke and Johnston (1992)

Unfortunately, the jury sided with the plaintiff, awarding her US$850 000 (Pollack, 1988).

Mostly due to this famous Michigan TMD case (Brimm v Malloy), the debate over orthodontics causing TMD was renewed but at a more serious level. This prompted the American Association of Orthodontists (AAO) and the orthodontic community to re-examine the relationship between orthodontic treatment and TMD. Nonetheless, many subsequent investigations that were performed discredited the allegations in the Brimm case. These studies have demonstrated that orthodontic treatments performed with and without extractions and/or headgears, resulting in the retraction of maxillary incisors do not cause distally positioned mandibles and anterior displaced TMJ disks (Gianelly et al., 1988, 1991a,b; Gianelly, 1989a,b).

Luecke and Johnston evaluated pre- and post-treatment records of 42 ‘edgewise’ patients with Class II Division 1 malocclusions treated in conjunction with the extraction of the upper two first premolars. About 70% of the patients showed forward displacement of the mandible and the remaining 30% exhibited distal displacement. The changes in condylar position were not correlated with incisor retraction but rather with the displacement of the buccal segments. The authors suggested that given the pronounced overjet in a Class II malocclusion, the incisors would normally go through marked retraction without touching or ‘trapping’ the lower incisors. It would seem that a patient’s centric occlusion position is determined by the occluding cusp-fossa relationships. With combined forward growth of the midface and anchorage loss from the reciprocal closure of the maxillary extraction spaces, marked mesial movement of the buccal occlusion will occur, producing an anterior shift of the mandible rather than a distal displacement (Luecke and Johnston, 1992). Further studies by Johnston and co-workers have also shown no differences between extraction and non-extraction groups with regards to TMD (Luppanapornlap and Johnston, 1993; Beattie et al., 1994).

Chin-cup therapy in Class III malocclusions has also been implicated in causing TMD. It is believed that the applied upward and backward directed forces to the mandible will predispose patients to anterior disk displacements. However, studies on the long-term follow-up of patients and MR scans of the condyle–disk relationships in patients following chin-cup therapy have concluded that chin-cup therapy when applied with appropriate forces is not a risk factor for TMD (Gökalp et al., 2000; Arat et al., 2003).

In 1990, Reynders completed the first comprehensive literature review to evaluate whether there was a relationship between orthodontics and TMD. He reviewed all articles published from 1966 to 1988 and concluded that ‘orthodontic treatment should not be considered responsible for causing TMD regardless of orthodontic technique. The data also rejects the assumption that orthodontic treatment is specific or necessary to cure signs and symptoms of TMD.’ Similarly, in a review article by McNamara et al. in 1995, it was concluded that orthodontic treatment performed during adolescence generally does not increase or decrease the odds of developing TMD later in life. In addition, in 2002, Kim et al. published a meta-analysis further supporting the premise that orthodontics does not cause or predispose to TMD. A recent long-term, prospective cohort study by Macfarlene et al. (2009) conducted in South Wales, UK, initially involving over 1000 subjects, concluded that orthodontic treatment neither causes nor prevents TMD. This has been the largest longitudinal epidemiological study in orthodontics to investigate the relationship between orthodontic treatment and TMD. The study demonstrated that TMD prevalence in young adults and adolescents is the highest at age 19–20 and higher in females than in males. Female gender and TMD in adolescence were the only predictors for TMD in young adults. In addition, a large recent cross-sectional study involving randomly selected 1011 children and adolescents between the ages of 10 and 18 found that the group undergoing orthodontic treatment were not at an increased risk of TMD and bruxism (Hirsch, 2009). This author also found that the orthodontic group exhibited fewer wear facets and reported less parafunctional behavior. These studies, as well as others, over time have helped to define (and redefine) and clarify the possible relationship between orthodontics (and orthodontists) and TMD.

Centric Relation Controversy

CR is the position of the condyles independent of tooth contact, whereas CO is an interocclusal dental position of the maxillary teeth relative to the mandibular teeth. Other terms that have been used synonymously with CO are MI (maximum intercuspation) and ICP (intercuspal position). The definition of CR has evolved considerably over the past half-century from being a posterior position of the condyle in relation to the glenoid fossa to a posterior-superior position to eventually an anterior and superior position (Academy of Prosthodontics, 2005). Before 1968, CR was considered as the retruded most posterior condylar position. The latest edition of the Glossary of prosthodontic terms (GPT) (Academy of Prosthodontics, 2005) defines CR as ‘a maxillomandibular relationship in which the condyles articulate with the thinnest avascular portion of their respective disks with the complex in the anterior-superior position against the slopes of the articular eminences.’ This edition of the GPT also includes six historical definitions of CR. Unfortunately changes in the definition and concept of CR have been determined for the most part arbitrarily, and not based on science and evidence.

The current gnathological view dictates that CO should ideally be coincident with an anterior-superior CR (Klar et al., 2003; Schmitt et al., 2003; Cordray, 2006). This hypothesis was adopted despite the absence of evidentiary support. In the 1970s, Roth advocated establishing a retruded, posterior-superior ‘seated’ CR position when the interdigitating occlusion was in CO; that is, CR (CRO) equals CO, or CR is coincident with MI or ICP (Roth, 1973, 1976). He believed that if orthodontists failed to reach this goal of a posterior-superior seated CR position coincident with CO as part of their orthodontic treatment, patients would be prone to develop TMD symptoms. Furthermore, the attainment of a retruded, posterior-superior CR position would mitigate the development of TMD. Roth later recanted his view of retruded CR and adopted the contemporary view of antero-superior CR.

Although contemporary orthodontic gnathologists believe in attaining an anterior-superior condyle position at the same time that the teeth are in CO (CR = CO), there is little or no scientific evidence to support this view (Rinchuse, 1987). Alexander et al. (1993), using TMJ MRI, revealed that condyles are not located in the assumed positions as advocated and provided by several gnathological centric bite registrations. Several popular centric bite registrations attempting to locate retruded (posterior-superior) CR and contemporary anterior-superior CR did not correspond to the condyle positions of individuals who were TMD asymptomatic. It would appear that the attempted doctor positioning of the condyles into specific locations within the glenoid fossae through chin guidance, manipulation, and/or bite registration is essentially a blind procedure. Dentists, who believe in establishing a coincidence of CR with CO, unnecessarily subject their patients to procedures that may lead to irreversible bite alterations and increased financial costs. The location and position of the condyles in the glenoid fossa, irrespective of where that may be, has not been demonstrated to be consequential to the presence or absence of TMD symptoms (Griffiths, 1983; McNeill et al., 1990; Dixon, 1991; Mohl and Dixon, 1994).

Centric Slides and TMD

Centric slides typically refer to an anterior-posterior shift, present between CO and CRO (the point of occlusal contact when the TMJs are in centric relation). In the 1960s, centric slides were believed to cause TMD (Ramfjord and Ash, 1971; Mohl, 1991). These findings, however, were based on descriptive studies that lacked control/comparison groups. Interestingly, when control/comparison groups that included subjects without TMD were included in subsequent studies, it was found that the exact same centric slides were also present and observed in subjects who did not have TMD. So the early studies possessed many false-positive findings for TMJ pain dysfunction, i.e. high sensitivity with poor specificity, which further fueled the TMD debate (Mohl, 1991).

On average and ‘normally,’ most centric slides are in the range of 1–2 mm. Furthermore, it has been shown that large slides (4–7 mm) are associated with degenerative changes within the TMJ. However, it appears that these large centric slides are more the result or consequence of the TMD or joint changes rather than the cause. As a result, the removal of such centric slides even in the presence of a TMJ articular disorder is not advisable (Seligman and Pullinger, 1991; McNamara et al., 1995).

Deprogramming to Obtain an Ideal CR

The use of ‘deprogramming splints’ for either the short or long term prior to taking centric bite registrations in order to obtain a more accurate record of centric relation is a controversial topic. It is believed that mandibular movements are governed by preprogrammed muscular engrams, or habitual muscular patterns. Muscle engrams are the memorized patterns of muscle activity developed from the habitual repetition of proprioceptive sensory information. Gnathologists hypothesize that these memorized patterns of the muscles of mastication may adversely change the position of the mandible in the presence of occlusal interferences. They therefore recommended the use of a deprogramming splint or other type of apparatus before obtaining centric bite registrations. It is believed that using various deprogramming splints to disclude the posterior teeth would remove any occlusal interferences or proprioceptive errors and permit the muscles of mastication to establish a more physiologic engram (Roth, 1973, 1976, 1981a, b, 1985; Roth and Rolfs, 1981; Cordray, 1996, 2006).

Orthodontic gnathologists maintain that patients need to be deprogrammed from their pre-existing occlusion prior to obtaining CR records even in patients without TMD. They contend that without the use of deprogramming splints prior to articulator mountings (especially in patients with Class II skeletal malocclusions), clinicians may miss the diagnosis of large centric sagittal slides (‘Sunday bites’ – greater than 2 mm) as well as slight transverse and vertical discrepancies. Finally, it is argued that the use of hand-held dental casts made from CO-generated pretreatment records (including lateral cephalograms), vis-à-vis CR records preceded by deprogramming splints, may compromise the initial orthodontic diagnosis because the true skeletal discrepancy may not be elucidated (Roth, 1973, 1976, 1981a, b; Williamson et al., 1978; Roth and Rolfs, 1981; Shildkraut et al., 1994; Cordray, 2006).

However, the evidence for using deprogrammers is equivocal and lacks a true physiologic basis. Following the use of a ‘Lucia-type anterior deprogramming jig’ (i.e. anterior tooth contact without posterior tooth contact) for 6 hours in TMD subjects, Karl and Foley in 1999 found small differences in before and after articulator condyle position indicator (/>