Journal of Oral & Facial Pain and Headache, 04/2013

During an overseas flight I had the pleasure to read the book How Doctors Think by Jerome Groopman.1 In this book, the author explores clinical decision-making with a particular emphasis on cognitive errors that often lead to misdiagnosis and inappropriate treatment. This book, and the numerous case reports submitted for publication to the journal pointing to diagnostic mistakes, inspired this editorial. Indeed, the case reports almost invariably describe patients with a wrong initial diagnosis of temporomandibular disorders (TMD) that is questioned and reconsidered only after several treatment failures. Diagnostic mistakes typically involve system-related and cognitive factors.2 The latter may be caused by inadequate knowledge (knowledge gap), faulty data gathering, inaccurate clinical reasoning, and erroneous verification of diagnostic hypotheses.2 Thus, cognitive errors may involve faulty thought processes and subconscious bias,3 and are intrinsic to the pitfalls of using heuristic approaches, eg, experience-based reasoning shortcuts, unconsciously used by clinicians to arrive at a quick diagnosis. In medicine, diagnostic mistakes are more often caused by cognitive errors than knowledge gaps. However, it is evident from reading the case reports that knowledge gaps contribute in a number of cases to the diagnostic errors made when evaluating patients with orofacial pain. This is not surprising given the large number of pathologies and disorders leading to orofacial pain and the lack of a medical education of the vast majority of dentists. In the following I am highlighting those cognitive errors I consider most likely to occur due to the nature of orofacial pain and the limited training of dental practitioners in making a differential diagnosis. These are anchoring, availability bias, premature closure, confirmation bias, and framing effect (for a full list see Croskerry,4 Nendaz and Perrier,5 and Stiegler et al3). Anchoring refers to the tendency to hang on to the first diagnosis and failing to consider the full spectrum of differential diagnoses, and availability bias refers to considering a diagnosis more likely because it readily comes to mind. Premature closure is the propensity to accept prematurely a diagnosis without considering other possible causes, and confirmation bias refers to interpreting clinical findings only to support the first diagnostic hypothesis without looking for, or even disregarding, disconfirming evidence. Framing effect refers to the fact that ensuing thinking is influenced by leading aspects of the initial presentation. Because TMD is the most frequent cause of chronic orofacial pain, it is most often seen by dentists. Therefore, it is easy for dentists to consider this diagnosis more likely than other ones because it readily comes to mind (anchoring and availability biases). Moreover, the presence in the clinical examination of the "classical" signs of TMD is often considered sufficient to confirm the diagnosis of TMD (confirmation bias). This, in turn, often prevents widening the differential diagnosis (premature closure). The International Research Diagnostic Criteria for TMD Consortium Network and Orofacial Pain Special Interest Group of the International Association for the Study of Pain worked together to improve the validity of the diagnostic criteria for the most common TMD, and they now include criteria for modification of pain by function, movement, or parafunction, and for replication of the patient's pain complaint by provocation tests (familiar pain).6 The new criteria have increased sensitivity and specificity. The clinician must, however, acknowledge that this may lead to increasing the risk of a confirmation bias. Thus, it is necessary during the diagnostic process to keep in mind that the presence of "familiar pain" during jaw movement or palpation of the associated structures is not associated exclusively with a myalgia, a myofascial pain with referral, or an arthralgia, as it is also present with other diseases or disorders affecting these structures, such as a temporomandibular joint inflammation, a myositis, and a metabolic muscle disease. In order for the criterion to warrant a diagnosis, the signs must explain the symptoms (signs plausibility vs confirmation bias) and the history, or additional assessment, must effectively rule out other competing diagnoses.7 Different strategies have been proposed to decrease the likelihood of cognitive errors. A thorough history, the foundation of a reliable diagnosis, must be initiated keeping in mind all diseases and disorders that can elicit orofacial pain. These will be ruled in or out during the history-taking process. This approach reduces the likelihood of making a premature closure error and avoids anchoring and availability biases. Other strategies imply debiasing and metacognition. The latter refers to the analysis of one's own thinking, eg, the process by which the clinician reflects upon, and has the option of regulating, what he/she is thinking.8 Thus, during the diagnostic process the clinician should systematically ask himself/herself questions that force him/her to explore other diagnostic possibilities (reflective thinking or reasoning): Are all the patient's findings accounted for by my working diagnosis? Do my working diagnosis and/or the clinical signs explain the patient's symptoms (plausibility)? Which alternative diagnoses should be considered? In addition, the clinician should list the findings that either support or do not support his/her working diagnosis and rank the different working hypotheses in order of likelihood.9,10 It will always be impossible to rule out diagnostic errors. However, to reduce their likelihood, the astute clinician must engage in reflective thinking when diagnosing an orofacial pain patient. In particular, he/she must reconsider the correctness of his/her diagnosis and not of the therapy, in case this fails within a reasonable time period of a maximum of 2 to 3 months.

Aims: To explore the impact of trigeminal nerve injuries on quality of life, including the effect of pain on psychological and affective function. Methods: An observational, cross-sectional survey design was employed. Fifty-six patients with inferior alveolar nerve injury (IANI) and 33 patients with lingual nerve injury (LNI) completed standardized self-report measures of pain intensity, pain catastrophizing, self-efficacy to cope with pain, and mood, in addition to generic and oral health-related quality of life (HRQoL) indicators. The impact of pain severity on these aspects of psychosocial function was examined. Summary statistics were calculated for all measures and compared with norms or values of other relevant studies, when available, using t tests. The impact of pain severity on these aspects of psychosocial function was examined using analysis of variance and hierarchical multivariate regression models. Results: The majority of patients reported pain associated with their nerve injury (86%). Nerve injury had a significant impact on all investigated domains, and this was closely linked with reported pain levels. Patients with severe pain showed particularly elevated levels of depression and pain catastrophizing, as well as substantially reduced HRQoL and coping efficacy levels. Pain intensity level was a significant predictor in all models except anxiety, uniquely contributing between 17% and 26% of variance to the prediction of pain catastrophizing, depression, coping efficacy, and generic and oral HRQoL. Conclusion: Traumatic injury to the trigeminal nerve is associated with a substantial patient burden, particularly in patients who experience severe neuropathic pain as part of their condition. These findings highlight the need to identify, develop, and evaluate more effective treatments for neuropathic pain in trigeminal nerve injury that will not only provide clinically meaningful reductions in pain but also improve patients' quality of life

Aims: To examine sleep complaints in patients with burning mouth syndrome (BMS) and the relationships between these disturbances, negative mood, and pain. Methods: Fifty BMS patients were compared with an equal number of healthy controls matched for age, sex, and educational level. The Pittsburgh Sleep Quality Index (PSQI), the Epworth Sleepiness Scale (ESS), the Hamilton Rating Scales for Depression (HAM-D) and Anxiety (HAM-A) were administered. Descriptive statistics, including the Mann-Whitney U test and hierarchical multiple linear regression analyses were used. Results: BMS patients had higher scores in all items of the PSQI and ESS than the healthy controls (P .001). In the BMS patients, a depressed mood and anxiety correlated positively with sleep disturbances. The Pearson correlations were 0.68 for PSQI vs HAM-D (P .001) and 0.63 for PSQI vs HAM-A (P .001). Conclusion: BMS patients reported a greater degree of sleep disorders, anxiety, and depression as compared with controls. Sleep disorders could influence quality of life of BMS patients and could be a possible treatment target.

Aims: To assess Research Diagnostic Criteria for Temporomandibular Disorders (RDC/TMD) Axis II variables in an initial psychosocial screening and as a part of biopsychosocial subtyping of Finnish referral patients with TMD pain for adjunct multidisciplinary assessment. Methods: Consecutive Finnish referral patients with TMD pain (n = 135) participated in this questionnaire-based survey. Psychosocial screening was based on Graded Chronic Pain Scale (GCPS) and culturally adjusted Symptom Checklist 90-revised (SCL-90R) depression scale scores and subtyping on GCPS pain-related interference in accordance with previous treatment tailoring studies. Biopsychosocial subtyping variables included symptoms of depression and somatization, general health, pain-related worry, sleep dysfunction, and coping ability. Subtype comparisons were analyzed with Bonferroni adjusted P values and multivariable logistic regression (SAS 9.3). Results: Based on psychosocial screening, 44% of the patients were psychosocially uncompromised (TMD subtype 1), 33% moderately, and 23% severely compromised (TMD subtypes 2 and 3). Compared to TMD subtype 1, TMD subtype 2 patients reported intermediate scores, and the most vulnerable TMD subtype 3 had the poorest general health, most elevated depression, somatization, worry and sleep dysfunction, and poor coping ability (P .05). According to multivariable logistic regression, depression and worry levels were significantly higher in TMD subtype 3 compared to TMD subtype 1, whilst patients in TMD subtypes 1 and 2 reported significantly better coping ability compared to TMD subtype 3 (P .05). Conclusion: The Finnish RDC/TMD Axis II was found reliable in initial TMD pain patient screening and with further biopsychosocial assessment identified three main TMD subtypes, two with compromised psychosocial profiles for adjunct multidisciplinary assessment.

Aims: To investigate the effectiveness of single and concomitant treatment of migraine and temporomandibular disorders (TMD) in women with the comorbidity. Methods: Eligible female patients met International Classification of Headache Disorders, second edition (ICHD-2) criteria for migraine with or without aura and the Research Diagnostic Criteria for myofascial TMD (Grade ll or lll). After a run-in period (30 days), women with both migraine and TMD were enrolled into a four-arm, double-blind, placebocontrolled, factorial study testing the separate and joint effects of a migraine treatment (propranolol 90 mg) and a TMD treatment (stabilization splint [SS]) in four groups of patients. The four treatment groups were propranolol and SS (n = 22); propranolol placebo and SS (n = 23); propranolol and non-occlusal splint (NOS) (n = 23); and propranolol placebo and NOS (n = 21). The primary endpoint for migraine was change in headache days from baseline to the third month, and the secondary endpoint was change in days with at least moderate headache in the same period. The TMD endpoints included pain threshold and mandibular vertical range of motion. Data were analyzed using analysis of variance (ANOVA, Dunn's post-hoc test) or Kruskal-Wallis test. Results: For the primary endpoint, in intention-to-treat (ITT) analyses (n = 94), propranolol and SS were associated with a nonsignificant reduction in the number of headache days, relative to all other groups. For per-protocol (PP) Completer analyses (n = 89), differences in the number of headache days reached significance (P .05). The propranolol and SS group was significantly superior to the other groups on all other headache endpoints and in disability, in both ITT and PP analyses. No significant differences among groups were seen for the TMD parameters. Conclusion: In women with TMD and migraine, migraine significantly improved only when both conditions were treated. The best treatment choice for TMD pain in women with migraine is yet to be defined.

Aims: To investigate the reliability and magnitude of intraoral mechanical pain sensitivity by using a palpometer with add-on devices with different physical properties. Methods: Sixteen healthy volunteers participated. Three palpometers (0.5, 1.0, and 2.0 kg) were used. Add-on devices were put on the circular metal stamp of the palpometer. Four diameters (3, 4, 5, and 10 mm) and two shapes of the rubber-top (flat and round) of the add-on devices were tested at each force level, ie, a total of 24 combinations. Participants were stimulated at the gingival mucosa around the maxillary central incisors and first molars on both sides by using the palpometers in randomized order. Participants rated perceived stimulus intensity on a 0-50-100 numerical rating scale (NRS). Ten volunteers were examined twice on the same day and recalled for a second session for assessment of within- and between-session reliability. Intraclass correlation coefficients were calculated for reliability measures, and NRS scores were analyzed with analysis of variance. Results: Reliability of NRS scores was excellent (interclass correlation coefficients 0.76 to 0.99). Analysis of NRS values corrected for pressure level revealed that there were main effects of site (P = .006), force (P .001), size (P .001), and shape (P .001) but not side (P = .051). Conclusion: Reliability of intraoral novel palpometer measures of pressure sensitivity was excellent, and sensitivity to pressure stimulation was dependent on the applied force and physical properties of the add-on device. The study indicated that semi-quantitative assessment of intraoral mechanical sensitivity is feasible and could be applied in further studies on different intraoral pain conditions.

Aims: To assess the influence of myofascial temporomandibular disorder (TMD) pain on the pressure pain threshold (PPT) of masticatory muscles in women during a migraine attack. Methods: The sample comprised 34 women, 18 to 60 years of age, with a diagnosis of episodic migraine previously confirmed by a neurologist. All subjects were evaluated using the Research Diagnostic Criteria for TMD (RDC/TMD) to determine the presence of myofascial pain. They were divided into two groups: group 1 (n = 18) included women with migraine; group 2 (n = 16) included women with migraine and myofascial TMD pain. Participants were evaluated by measuring PPT values of the masseter and anterior temporalis muscles and Achilles tendon with a pressure algometer at two moments: pain free and during a migraine attack. A three-way analysis of variance with a 5% significance level was used for statistical purposes. Results: Significantly lower PPT values were found during the migraine attack, especially for women with concomitant myofascial pain, regardless of the side of the reported pain. Conclusion: Migraine attack is associated with a significant reduction in PPT values of masticatory muscles, which appears to be influenced by the presence of myofascial TMD pain.

Aims: To investigate whether experimental tooth clenching leads to a release of algesic substances in the masseter muscle. Methods: Thirty healthy subjects (16 females, 14 males) participated. During two sessions, separated by at least 1 week, intramuscular microdialysis was performed to collect masseter muscle 5-hydroxytryptamine (5-HT) and glutamate as well as the metabolic markers pyruvate and lactate. Two hours after the start of microdialysis, participants were randomized to a 20-min repetitive experimental tooth-clenching task (50% of maximal voluntary contraction) or a control session (no clenching). Pain and fatigue were measured throughout. The Friedman and Wilcoxon tests were used for statistical analyses. Results: No alterations were observed in the concentrations of 5-HT, glutamate, pyruvate, and lactate over time in the clenching or control session, or between sessions at various time points. Pain (P .01) and fatigue (P .01) increased significantly over time in the clenching session and were significantly higher after clenching than in the control session (P .01). Conclusion: Low levels of pain and fatigue developed with this experimental tooth-clenching model, but they were not associated with an altered release of 5-HT, glutamate, lactate, or pyruvate. More research is required to elucidate the peripheral release of algesic substances in response to tooth clenching.

Aims: To evaluate the nature of the antinociceptive interaction of systemic administration of a combination of the anticonvulsant gabapentin with the antidepressant nortriptyline, by isobolographic analysis in the formalin orofacial pain test of mice. Methods: The study was carried out in 168 male CF-1 mice weighing 30 g, and the protocol was to test each drug (at dosages of 1, 3, 10, 30, and 100 mg/kg of gabapentin and 0.1, 1, 3, 10, and 30 mg/kg of nortriptyline; ip) alone and in combination. The isobolographic assay has two phases: phase 1 corresponds to the 5-minute period starting immediately after the formalin injection and reflects a tonic acute pain due to peripheral nociceptor sensitization; phase 2 is recorded as the 10-minute period starting 20 minutes after the formalin injection and reflects an inflammatory pain state. Results were analyzed by Student t test for independent means. Results: Gabapentin was 1.61 times more potent in phase 2 than in phase 1, and nortriptyline 1.37 times more potent in phase 2 than in phase 1. The combination of both drugs was synergic, with an index of interaction of 0.134 and 0.148 for phase 1 and phase 2, respectively. Differences in the pharmacological profiles of gabapentin and nortriptyline could underlie the synergism of the two drugs. Conclusion: The findings of this study are important, because they are concordant with some clinical studies and also raise the possibility of potential clinical advantages of combining gabapentin and nortriptyline in pain management, since the low doses of the components may potentially have a lower incidence of adverse reactions.

Cutaneous mastocytosis (CM) has been associated with urticaria, itching, and pain of the affected regions. Although the occurrence of CM in the facial skin is rare, it may be a cause of chronic facial pain, and pain characteristics may mistakenly be interpreted as trigeminal nerve pathology. However, the dermatological appearance of the different variants of cutaneous mastocytosis is distinct and should be considered as an uncommon differential diagnosis in an orofacial pain diagnostic algorithm. This article presents a case of telangiectasia macularis eruptiva perstans, a rare type of cutaneous mastocytosis, as the underlying cause of chronic facial pain, erythema, and swelling.