Refractory burning mouth syndrome: clinical and paraclinical evaluation, comorbidities, treatment and outcome

BMS is a chronic pain condition characterized by persistent intraoral burning without related objective findings and unknown etiology that predominantly affects elderly females. Previously used terms include stomatodynia, or glossodynia when confined to the tongue (based on the Greek words stoma that means mouth and glossa that means tongue). The condition has been recently recognized in the current International Classification for Headache Disorders (ICHD-III beta) as an intraoral burning or dysaesthetic sensation, recurring daily for more than 2 h per day over more than 3 months, without clinically evident causative lesions. Oral mucosa should appear normal with normal sensory testing [1]. How prevalent the condition is remains debatable but recent general population based studies showed an incidence of 11.4 per 100,000 person-years [2] and a point prevalence of 0.11%, or 105.6 (95% CI, 88.6–122.6) per 100,000 persons [3]. Age-adjusted prevalence in women was significantly higher than men: 168.6 (95% CI, 139.0–198.2) vs. 35.9 (95% CI, 21.4–50.3) per 100,000 persons (P < 0.001). The highest prevalence was in women aged 70 through 79 years (527.9 per 100,000 persons). Mean (SD) age at BMS diagnosis was 59.4 (15.1) years (range, 25–90 years) [3]. In other reports the BMS prevalence varies from 3.7% up to 40% in elderly [4]. Little is known regarding the aetiopathology of the condition. There is evidence that BMS is related to mild sensory and autonomic small fiber neuropathy with concomitant central disorders particular within the brainstem [5, 6], as well as to impaired endogenous dopamine system in the putamen [7]. In other reports MBS is often comorbid with mood and somatoform disorders with prolong immunoendocrine changes [8]. In some refractory BMS cases comorbided mood disorders are severe [9, 10]. Altered structure and function in the hippocampus and medial prefrontal cortex in BMS patients explain the concomitant affective conditions [11].

There is no satisfactory treatment for BMS. The most widely accepted treatment options that show variable results include tricyclic anti-depressants, benzodiazepines, antipsychotic drugs [12], and SSRIs [13]. Because there is enough evidence that in BMS patients TRPV1 receptors are over-expressed [14] local application of capsaicin was used efficiently [15]. The best evidence for BMS treatment includes only topical capsaicin, alpha-lipoic acid (ALA), and clonazepam [16, 17]. Interestingly, there are reports of BMS comorbidity with Parkinson disease (PD) or Restless Syndrome [18‐20] and dopamine agonists showed potential efficacy in reducing BMS symptoms [21, 22]. Non-pharmaceutical treatments may help in addition, like cognitive psychotherapy [23]. Nevertheless, outcome for BMS is poor. In a study of 91 BMS patients the symptoms remitted spontaneously within the five years of treatment. Only 42% of the study population had improved the symptomatology significantly, and this improvement would reach 60% if clonazepam were associated to psychotherapy [24].

In this study we aimed to observe the long-term efficacy of high velanfaxine doses combined with clonazepam in a particular subgroup of BMS patients who do not respond to current clinical management.

Patients with burning mouth symptoms that were resistant to current available treatments were selected at presentations for long follow-up, since the beginning of 2010. After the publication of ICHD-III beta all patients fulfilled the BMS criteria were identified and among them those they had no improvement after at least three consecutive drug treatments were followed. Pharmacotherapy might include any agent for chronic pain alone, or in combination with psychotherapy, counting anti-depressants and anti-epileptics in various doses and treatment durations, without any specific pattern. After three consecutive treatment attempts without a 50% reduction in daily pain, as defined by patients them selves, the selection criterion was fulfilled. No exclusion criteria were applied. Once a patient had recruited, a daily pain diary was delivered to follow pain duration and severity during follow-up period. A relevant physician treated any concomitant non-neurological condition according to current clinical practice but treatments with local anesthetics, anti-epileptics and anti-depressants were not allowed. Antiarrhythmic agents that may have analgesic effects could be used only after approval by the research team. All participants were told that they will get two treatments simultaneously that have been trialed alone, but not in combination before. The potential treatment adverse effects were announced as well. A specific investigation protocol was designed for diagnostic reasons including brain and facial MRI, trigeminal blink reflex and peripheral nerve conduction studies apart from other tests necessary because of patients’ particular comorbidity and medical history. Participants allowed to performed neurophysiological studies of facial nerve in private laboratories, thus no common methodology used. All patients had a specific oral and orofacial clinical examination to exclude any oral condition or temporomandibular joint and occlusion dysfunction from an expert. In addition all patients had a neuropsychological evaluation with a battery of tests including the Hamilton Rating Scales for Anxiety and Depression, as well as clinical psychiatric consultation. In all participants cognitive psychotherapy treatment was suggested, in addition to pharmacotherapy. Participants used the analogue visual scale (VAS) to assess mean daily pain.

All participants singed inform consent. The Ethical and Scientific Committee of the Athens Naval Hospital approved the study.

Eight females out of 14 patients with refractory BMS cases were included in the protocol. All were above 58 years old (mean ± SD: 66.1 ± 6.2) with the mean duration of the condition lasting 4.3 ± 1.4 years. All were followed-up in our department for more than 2 years (mean follow-up duration 35.4 ± 12.1 months).

BMS is rather a rare chronic idiopathic pain condition affecting 1‰ of people, predominantly elderly women that often resist to common treatment with anti-epileptics and anti-depressants. We describe here eight cases of female patients with refractory BMS and depression who followed for more than two years. Pain was severe, daily and exaggerated during the night in all cases. Repetitive trials with anti-depressants and anti-epileptics agents commonly used in chronic pain syndromes had failed, but all responded to a combination of high dose venlafaxine and oral solution of clonazepam. Participants experienced several adverse events in the treatment initiation but after a while they tolerated it. BMS treatment remains controversial [25] yet there is little evidence that topical clonazepam helps [16]. Indeed all patients reported immediate pain improvement after clonazepam oral solution. Furthermore, increased clonazepam dose was used to treat all relapses observed successfully. As in previous reports clonazepam was used in combination of oral topical and systematic administration [26]. We used clonazepam mouthwash as reported by others [27]. Clonazepam’s mechanism of action in BMS stays unspecified. The drug binds to the peripheral GABA_A receptors that have been identified within the tongue nerve fibers of rats [28] and enhances GABA inhibitory effects that may control pain pathways. Additionally, clonazepam promotes brain stem serotonergic descending pain inhibition by binding to the central GABA_A receptors located within the brainstem [29]. Because of this dual-site action of clonazepam (peripheral and central) the drug was given both systematically and locally in our patients. Clonazepam is a long acting benzodiazepine agonist with low risk of abuse, often used for in short-acting benzodiazepine withdrawal. However, there are observations suggesting the existence of its abuse [30]. No case of clonazepam abuse was observed in our study.

Venlafaxine, a serotonin-norepinephrine reuptake inhibitor (NSRI) that controls equally depression and chronic pain [31, 32] was selected to combine clonazepam. Venlafaxine inhibits serotonin and norepinephrine reuptake leading to enhanced descending inhibition of centrally sensitized pain and has been used to treat pain conditions comorbid with depression [33]. Whether BMS, a chronic pain condition, leads to mood disturbances or in the contrary, established mental conditions might predispose an individual to symptoms related with BMS remains unanswered. Repeated observations confirm however that chronic exposure to either pain or stress can guide to maladaptive hormonal and neuronal modulations that can result in chronic pain and a wide spectrum of stress-related disorders including anxiety and depression [34], by affecting neuropeptide neurotransmission and signaling within several brain structures including the mesolimbic dopamine system [35].

Other studies showed that the majority of BMS patients present with several additional unexplained extra-oral comorbidities, indicating that various medical disciplines should be involved in the BMS diagnostic process and that BMS may be classified as a complex somatoform disorder rather than a neuropathic pain entity [36].

This study carries several limitations. It is an uncontrolled observational study with a small number of patients, thus placebo biases cannot be excluded. In a meta-analysis of placebo-controlled trials for BMS, treatment with placebos produced a response that was 72% as large as the response to active drugs [37]. These high placebo response rates documented in this review pose a significant challenge for the design of future controlled studies evaluating therapies for BMS. The aim of our study however, was to explore the potential efficacy of a combination of two pharmaceutical agents in a very selective group of BMS patients. Future controlled large scale studies are needed to provide good evidence of the suggested treatment benefits. Some patients included in this study may have not the typical clinical characteristics of BMS, e.g. pain during the night and unilateral burning, but this was the patients’ clinical picture we observed. Patients did not undergo an intraoral quantitative sensory test that has been shown to be abnormal in recent reports indicating significant loss of thermal function but not mechanical function in BMS patients, supporting the hypothesis that BMS may be a probable neuropathic pain condition [38].

BMS is often misdiagnosed and pain physicians or headache specialists should be aware of the condition. Notably, brain and facial MRI and facial nerve conduction studies showed no significant findings. Secondary burning sensation due to drugs (e.g. angiotensin converting enzyme inhibitors, dopaminergic agonists) or thyroid hormone dysregulation should be ruled. BMS has high psychiatric comorbidity but can occur in the absence of psychiatric diagnosis. Patients with atypical chronic pain syndromes must be considered as potential candidates for under-diagnosed depression (major) and suicidal thoughts. Other common BMS comorbidities include PD that should be treated accordantly. Thus, BMS management requires multidisciplinary management.

Refractory BMS deserves bottomless psychiatric evaluation and management when current available treatments fail. Paraclinical investigation including brain imaging and peripheral facial nerve conduction evaluation may be needed. Controlled, large-scale trials with high dose NSRIs in combination with mouthwash and systemic clonazepam are required to establish this treatment that might help to improve pain control in this rare and severely disable pain condition.