The results of this study present different symptomatic profiles of patients seeking treatment in a TMD clinic, who were classified by cluster analysis into four to six homogenous groups based on their clinical presentation.
It is important to note that the cluster names were intuitively defined considering the time of onset, the location, the presence of pain and extent of the symptoms. Patients’ profiles may help clinicians to quickly recognize patient characteristics and provide a reasonable treatment approach. After clustering procedures, data also illustrate the prevalence of symptomatic profiles, which is helpful for identification of symptomatic profiles, would help clinicians to estimate how common a disorder is within a population of TMD patients and understand the probability of certain pattern of clinical complaints.
Although internal validity of the diagnostic criteria and the efficacy of the examiners’ performance were not tested, standardized criteria of the AAOP were used and both examiners were trained and certified in the same Orofacial Pain Center (University of Kentucky, USA).
Masticatory muscle pain is recognized as the most common complaint in patients in general practice, and perhaps only a few patients will present for treatment in private practice with a chronic stage. This could be explained by the fact that many patients might have previously searched for treatments that turned out to be ineffective and then began to give up on seeking new alternatives, leading to an even more chronic problem , or it could be because many muscle or joint disorders do not only go unnoticed by patients and professionals in general practice, but also because most professionals do not know how to treat them properly. All those situations may worsen the problem [5–7].
In particular, this could occur with some internal derangement of the TMJ, which will eventually become a chronic condition quite easily. Kalaykova et al.  studied the natural course of patients with anterior disc displacement with reduction after two years and observed that clicking commonly disappear completely without symptoms of permanent locking even with the disc displacement still present, but with no, or only a partial, reduction. They also observed that intermittent locking may be indicative of the development of a disc displacement without reduction which is only rarely accompanied by symptoms of permanent locking .
These could be some of the reasons why patients with acute muscle pain and chronic joint impairments were more frequent in this sample. Many patients do not know what to do in these situations or which professionals they should look for: an orthodontist, a maxillofacial surgeon, a prosthodontist or an otolaryngologist, for example. Moreover, many professionals are unsure about what to say to patients with this distressing disorder, when (or whether) to refer them to specialists and even which experts to refer them to .
Another explanation for the high frequency of acute problems could be the fact that most of the TMDs are cyclical. Patients with a chronic condition that they had already been adapted, for some reason or event, may have worsened and make them seek for treatment [1, 9].
A strong relationship between sore muscles (myalgia or localized masticatory muscle pain) and bruxism (patients from cluster 2), myofascial masticatory muscle pain and bruxism (patients from cluster 1) and the need to seek for treatment was revealed by our results. A similar finding was found by Lobbezzo-Scholte et al. , whose results showed relatively more patients of the mainly myogenous component group who reported clenching and grinding than the other patient groups. Although a relationship is not clear from the literature, one could suppose that bruxism may be highly associated with pain referred from masticatory muscles , initiating or perpetuating masticatory muscle soreness or pain [12–14].
The significance of dealing with bruxism and masticatory muscle pain early on and the importance of recognizing these conditions in everyday practice are highlighted, since myalgia, being a deep pain stimulus, can yield central excitatory effects and referred pain. Therefore, this statement may explain the fact that when a muscle pain is maintained, more muscle pain may be restarted, keeping the patient in a cycle of pain . This entire situation might be sustained by bruxism (clenching or grinding) as an initiating or perpetuating factor for muscle disorders [13, 14].
A treatment plan could be outlined for patients with similar characteristics as for those comprising clusters 1 and 2 with muscle dysfunctions and bruxism. A more conservative management should be carried out since such disorders are related to multiple factors, and often, if not always, there is a strong role of emotional factors [2, 9, 12, 15–18]. This might include pharmacological therapy, cognitive-behavioural therapies, psychotherapy, self-care (resting, relaxation techniques, massage, hot and/or cold packs, stretching or exercise), physiotherapy (jaw exercises, postural training), low-level laser therapy and wearing occlusal appliances [19–27].
A combination of capsulitis/synovitis and disc dislocation without reduction (DDWOR) was found in patients from cluster 3. Regarding the concept that internal TMJ derangement is significantly involved in the production of TMJ pain and dysfunction, these results corroborate findings from other authors . These findings may also strengthen the idea that patients with a primary TMJ pathology, such as a DDWOR, frequently develop an inflammatory response to the dysfunctional disc-condyle relationship, clinically represented by a diagnosis of capsulitis/synovitis [3, 9, 29]. While DDWOR can be seen as a chronic disorder, is quite evident here that acute pain caused by capsulitis/synovitis may be the main reason why patients seek specialized treatment. The first efforts in treating patients with similar characteristics should be directed towards the complaint of pain: the reason why the patient initially sought treatment. In the case of an acute DDWOR, an attempt should be made to unlock the patient . While most therapies are not evidence-based, we strongly recommend a conservative approach, which may include patient orientation, pharmacotherapy, physical therapy for pain (thermotherapy, low-level laser therapy) and then physical therapy to improve function (passive TMJ manipulation) and the use of occlusal appliances [9, 26, 30].
Patients presenting with non-painful articular impairment from cluster 4 were the second most frequent group of patients who were referred for or who sought treatment for TMD. Despite the fact that this cluster was named TMJ impairments, due to the main disorders clustered, perhaps we can infer that treatment seeking was mainly guided by dysfunctions associated with muscle disorders or DDWR, as shown in Table 4.
As discussed above for clusters 1 to 3, a combination of treatment modalities can also be outlined for the management of patients who have a similar presentation to those of cluster 4, due to the many associated muscle and joint disorders. Each particular case should be evaluated in order to define appropriate treatment planning. As there are many therapeutic options, if the selected treatment does not eliminate the patient’s pain complaint, the next more complex condition or chronic muscle condition should be considered by the clinician . Due to the chronicity of the clinical conditions presented by these patients, a psychosocial component (such as anxiety and depression) may be present and thus a psychological approach should be included in the treatment plan [9, 31, 32].
Velly et al.  clustered 162 patients with TMD based on their clinical condition and degree of severity, and they also studied some related factors such as psychosocial aspects. Although the authors also obtained four clusters of TMD patients (two clusters with more than one TMD condition and two with only one), the profiles of patients in the clusters obtained were different from the clusters found in this sample; one of the reasons for this was the diagnostic criteria adopted by these authors—the Research Diagnostic Criteria for Temporomandibular Disorders (RDC/TMD) . In our study, emotional and psychosocial aspects would be either cause or effect of the TM disorders. They were only addressed as part of the overall treatment approach and were not included in this study. We consider that our study used routinely collected data that did not describe in full detail other variables that are associated with the cause or the effect of TMDs. Cross-sectional studies using data originally collected for other purposes are often unable to include data on confounding factors or other variables that affect the relationship between the presumed cause and effect. This may be viewed as a weakness of the study design and, certainly, a consequence of the limited scope of our study since we decided to focus on the description of the distribution of patients’ common symptomatic profiles.
Despite some limitations regarding the generalization of results (better related to patients seeking treatment, cases derived from only one private clinic and the evaluation and classification system adopted did not contemplate the RDC/TMD), we believe that the most interesting aspects that deserve particular attention from the results of our study are the large sample size, the large number of patients presenting with more than one TMD condition simultaneously diagnosed and the strong homogeneous relationship between patients in the formation of clusters. The clustering method may depend on the choice of classification variables and how they were collected . However, bias was avoided in the cluster formation by use of the two-step cluster analysis, because it does not involve hypothesis testing or a pre-determined number of clusters, perfectly acceptable for cluster data that may not meet the assumptions for best performance. As a non-hierarchical method of clustering, the two-step cluster method has two advantages over the hierarchical methods used in similar studies: it is less impacted by outlier elements and the final solution optimizes within-cluster homogeneity and between-cluster heterogeneity . Finally, identifying the characteristics of each group of patients seeking treatment may direct future research in the pursuit of outcomes at a higher level of efficacy and allow a more appropriate clinical decision-making process for patients with multiple, simultaneous TMDs.