Ⅰ. NEUROPATHIC PAIN DISORDER
1. Neuropathic pain: What we know
Neuropathic pain is defined as pain caused by a disorder of the somatosensory system, according to the International Association for the Study of Pain (1). The characteristic features of neuropathic pain include dysesthesia or abnormal pain induced by innocuous stimuli. These clinical signs and symptoms often persist without prominent triggering events or stimulation of the peripheral nervous system.
Neuropathic pain has been addressed as a functional consequence of central sensitization that is defined as amplified central pain perception in response to both noxious and innoxious stimuli (2). Such sensitized pain responses, termed hyperalgesia and allodynia, are translated as prolonged duration of pain and its widespread distribution (3). Patients experiencing consecutive and widespread pain lasting over three months, along with the absence of a peripheral cause, can be diagnosed with central pain syndrome (3).
Central sensitization can be induced by various processes, including inflammation, injury, or trauma. For example, deafferentation, or permanent loss of afferent fibers, following trauma or peripheral nerve injury may cause pain that has the potential to progress centrally. This is clinically significant to dentists or oral surgeons as patients may experience centralized pain after surgical procedures such as extractions or endodontic treatments. In addition, radical neck dissections due to head and neck malignancies pose a risk of inducing deafferentation pain caused by the damaged trigeminal nerve.
2. Trigeminal neuralgia and traumatic neuropathy
The representative cases of neuropathic pain in the orofacial area include trigeminal neuralgia (TN) (4) and post-traumatic trigeminal neuropathy (PTTN) (5). TN, with its prevalence rising in the Korean population (6), is characterized by episodic attacks of sudden electrical shock-like pain (4). A recent study indicates the incidence of TN as 100.21 per 100,000 persons in 2018, with a slight female predilection (6). TN involves both the facial skin and the intraoral mucosa and is triggered by innocuous daily habits or activities such as tooth brushing or mastication. The primary case of TN features spontaneous elicitation of pain in combination with compression of the trigeminal nerve root (7). In the case of PTTN, pain caused by traumatic injury to the trigeminal nerve branches may persist even after tissue healing, characterized by a spectrum of abnormal sensations from a tingling feeling to severe disability (5, 8, 9). According to the International Classification of Headache Disorders diagnostic criteria, PTTN (10) is defined as “unilateral or bilateral facial or oral pain following and caused by trauma to the trigeminal nerve(s), with other symptoms and/or clinical signs of trigeminal nerve dysfunction” (10).
Recent animal model-based studies on trigeminal neuropathic pain, including both TN and PTTN, have indicated the role of primary afferents in the development of neuropathic pain. For instance, a direct injury to the infraorbital branch of the trigeminal nerve or physical compression of the trigeminal nerve roots has phenocopied the characteristics of PTTN and TN, respectively (11, 12). Dysfunctional primary afferent signaling then propagates to the trigeminal subnucleus caudalis and the rostral ventromedial medulla, eliciting sensitization of the central terminals of the trigeminal nerve projection (7).
Ⅱ. BURNING MOUTH SYNDROME AS A NEUROPATHIC DISORDER
1. Clinical features of burning mouth syndrome
While TN and PTTN are classified as neuropathic pain disorders often associated with traumatic injuries to peripheral afferent nerve fibers, burning mouth syndrome (BMS) is defined as an “intraoral burning or dysesthetic sensation, recurring daily for more than 2 hours per day over more than 3 months, without evident causative lesions on clinical examination and investigation” (13). Similarly, the International Association for the Study of Pain and the International Headache Society describe BMS as chronic oral mucosal pain with no identifiable causative lesions or disease, lasting 4 to 6 months (1, 14). With this respect, the primary BMS can be distinguished from the secondary burning disorders associated with other local or systemic diseases.
A recent meta-analysis study indicated the worldwide incidence of BMS in the general population and clinical patients as 1.73% and 7.72%, respectively (15), but this is believed to be significantly underestimated due to the challenges in accurately diagnosing the disease. Unlike TN and PTTN, a higher incidence of BMS in women has been consistently reported, with the ratio between women and men varying from 3:1 to 16:1 (15). The incidence of BMS appears to increase significantly with age, especially in postmenopausal women over 50 years old (15). A higher incidence near or after menopause in women may indicate the contribution of hormonal factors to BMS.
The most commonly affected site of BMS is the anterior two-thirds of the tongue, followed by the anterior hard palate, lip mucosa, and mandibular alveolar regions. The signs and symptoms of BMS almost always indicate bilateral and symmetric involvements, with no correlation to the anatomical distribution of sensory nerves (15, 16). Patients suffering from BMS often experience xerostomia, dysgeusia, halitosis, or dysphagia, often without any evident triggering factor (16, 17).
2. Etiology of burning mouth syndrome
Despite persistent efforts over the last few decades, the etiology of BMS remains elusive. Multiple factors have been proposed to contribute to the pathophysiology of BMS, including peripheral and central neuropathy and altered mucosal epithelial function (18). As stated above, a higher incidence of BMS among menopausal or postmenopausal women suggests transient ovarian steroid imbalance, including follicle-stimulating hormone and estradiol, as a contributing endocrinological factor in developing BMS (19). In line with this idea, hormone replacement therapy was effective in subsets of BMS patients, with upregulation of nuclear estrogen receptors (20). Reduction in ovarian steroids following menopause may abolish the neuroprotective role of steroids, thus leading to the deterioration of thin nerve fibers that mediate pain signaling (21).
Another important factor contributing to the development and progression of BMS is psychological distress. BMS patients often address symptoms of somatization characterized by anxiety- and depression-associated behaviors. They frequently experience chronic fatigue, disturbed sleep patterns, depression, anger, and irritation, all of which lead to higher levels of neuroticism (22). However, it remains challenging to identify whether psychological disturbances such as anxiety and depression serve as either a cause or an effect of BMS. Patients with BMS often complain of psychological discomfort, which further worsens the symptoms of BMS. While the causative relationship between BMS and psychological distress remains to be further investigated, pharmacological approaches targeting these aspects have been documented as effective in improving the symptoms of BMS in subsets of patients (23).
Recent studies on BMS biology have also revealed a potential immunological contribution to its development and progression. For instance, the levels of BDNF, IL-1β, and IL-8 were significantly elevated in BMS patients, while those with anxiety or depression exhibited further association with the elevated saliva level of TNF-α (24). Such dysregulated inflammatory mechanisms, characterized by altered profiles of pro-inflammatory cytokines and chemokines, may be translated into clinical signs and symptoms of BMS (25).
3. Neuropathic nature of BMS
Despite the diverse etiological explanations suggested over the last decades, the most promising hypothesis to support the pathophysiology of BMS has come from the neuropathic pain theories. With this respect, both central and peripheral neuropathy have been proposed as culprits of burning sensation in BMS patients. The peripheral hypothesis describes BMS as a small fiber neuropathy affecting cranial nerves, specifically the trigeminal nerve and the chorda tympani branch of the facial nerve (26, 27). In line with this idea of peripheral neuropathy, the reduced density of thin fibers innervating the tongue epithelium may underlie abnormal sensation termed “phantom taste”, often described as bitter or metallic (27). Further molecular analyses have also revealed an increase in TRPV1-positive, NGF-pos itive, and P2X3-positive fibers, along with overexpression of Nav1.7 channels in BMS patients (27).
The alternative central theory of pain in BMS patients involves reduced activities of the dopaminergic and GABAergic inhibitory pathways in the central nervous system as well as enhanced activities of postsynaptic glutamatergic circuits, all of which would lead to central sensitization and subsequent temporal summation of pain signaling (17, 28). In line with this central role, the structural changes in the cortical grey matter processing pain signals, including the limbic system, have also been documented in BMS patients (26). These central changes would eventually decrease the pain threshold to noxious stimuli, exacerbating responses in the trigeminal-facial brainstem circuits.
Another neuropathic feature underlying BMS is its circadian dependence, as circadian disturbances are closely associated with abnormal pain sensation, psychological discomfort, and altered sleep patterns in BMS patients (29). Considering the critical role of the dopaminergic and hypothalamic- pituitary-adrenal (HPA) axis activities in circadian rhythms, corresponding alterations in the dopaminergic and HPA signals may also be evident in BMS patients, raising a possibility of therapeutic targeting of these activities.
Ⅲ. ANIMAL STUDIES OF BURNING MOUTH SYNDROME
1. Rodent models of neuropathic pain as a potential platform to study BMS
As stated above, the complex nature of the etiological factors of BMS hampers identifying the causative relationships between individual abnormalities and the development of BMS symptoms in patients. For the experimental validation of each contributing factor, animal models have been developed to provide insights into their mode of action. Representative animal studies of BMS include rodent models of neuropathic pain, with evident strengths and pitfalls. To date, there is no experimentally validated rodent model of BMS, primarily due to its failure to reproduce the multifaceted nature of BMS and to quantify changes in pain perception. Together with a lack of specific biomarkers that allow monitoring of disease progression, it should be noted that these rodent models of neuropathic pain only exhibit a limited representation of BMS.
Orofacial pain in rodent models can depict some features of BMS pathology, assessed by either evoked or non-evoked measures. Evoked measures center around reflex- like behaviors in response to mechanical or thermal stimuli delivered using von Frey filaments in the skin, thermal probes, or radiant heat stimulation to the pad or tongue. These stimuli often induce aversive behaviors, including head withdrawal and persistent facewash strokes (30, 31, 32). On the other hand, non-evoked measures of orofacial pain include the evaluation of changes in nocifensive behaviors, such as facial grooming, head flinching, or chewing-like mandibular movements. Changes in these measures are often considered altered pain perception in animals (33). In addition, food preference assay in rodents may reflect clinical features of BMS patients (32), as they often avoid spicy foods and seek hydration following intraoral burning and dry mouth symptoms (34). Taken together, rodent models of neuropathic pain capture some aspects of mechanical and thermal sensitivity relevant to BMS, with measures in head withdrawal reflex, drinking preference, and licking and capsaicin avoidance behaviors (35, 36, 37, 38, 39).
2. Rodent models of BMS-related burning sensation
Based on the aforementioned neuropathic pain models, recent efforts in BMS study have led to the development of three preclinical models with induced BMS-like symptoms in rodents. They include 1) experimental induction of dryness in the tongue (35, 36), 2) transection of the chorda tympani branch of the facial nerve (40), and 3) overexpression of artemin in the tongue epithelium (37, 38). These models aim to recapitulate a dry mouth condition, altered taste sensation, and thermal hypersensitivity, respectively.
The rodent model of a dry-tongue condition is established by exposing the tongue to dry room air for 2 hours over a week, resembling a dry mouth sensation in BMS patients (35, 36). This model is based on the idea that dryness in the oral cavity, or xerostomia, is closely associated with tongue pain (35, 36). In this model, dry-tongue enforcement has been related to mechanical hypersensitivity, presumably via increased p38 activity and upregulated TRPV4 in trigeminal ganglion neurons (35). Additional changes in the central pain signaling pathway have also been documented in the trigeminal spinal subnucleus caudalis neurons, as evidenced by increased glutamate receptors and ERK phosphorylation (36). However, it should be noted that these dry-tongue models have only demonstrated enhanced mechanical hypersensitivity due to the reduced mechanical reflex threshold but failed to recapitulate the altered heat sensitivity that most BMS patients experience.
While the dry-tongue model focuses on enhanced mechanical hypersensitivity in male rats, an additional model of BMS addresses the effect of the transected chorda tympani branch of the facial nerve in female rodents. This experimental intervention aims to recapitulate altered taste sensation in BMS patients (41), as the chorda tympani branch is responsible for taste perception (40). However, recent reports indicate that this nerve transection model cannot encompass the full spectrum of BMS signs and symptoms. Importantly, transection of the chorda tympani nerve alone is not likely to induce capsaicin-evoked pain hypersensitivity, one of the key characteristics of BMS (39, 40). Since altered taste perception and direct damage to the facial nerve branch are not commonly seen in BMS patients, this nerve transection model is less likely to be directly translated into clinical courses of BMS.
Finally, tongue pain in rodents can be elicited by transgenic or pharmacological upregulation of artemin, a vascular- derived neurotropic factor for developing sympathetic neurons (37, 38, 42). A similar manipulation of artemin expression can induce thermal hypersensitivity that lasts up to a week, presumably via upregulation of TRPV1 and TRPA1 channels and artemin receptors in trigeminal neurons (37). Therefore, this model may be more representative of human BMS biology, as the key characteristics of BMS can be monitored without tissue inflammation, a dry-tongue condition, or nerve damage required in other rodent models. Furthermore, it recapitulates a 3-fold increase in artemin expression and pain flare-ups in BMS patients (37, 43). In line with these findings, injection of neutralizing anti-artemin antibody in the tongue was sufficient to suppress enhanced capsaicin-triggered current in trigeminal neurons (37, 38). However, short-lived thermal hypersensitivity induced by artemin overexpression may not fully represent the chronic nature of BMS-related pain, along with a lack of taste perception change or hormone-specific effects.
In addition to the three models stated above, tongue pain can be elicited by intra-tongue injection of chemical irritants, including complete Freund’s adjuvant, capsaicin, formalin, and glutamate, most of which would result in tissue inflammation (39, 44, 45). Another experimental approach to simulate BMS also includes chronic constriction of the lingual nerve. Such manipulation has been reported to induce tongue hypersensitivity and neuropathic states via increased secretion of pain-associated substances such as calcitonin gene-related peptide from trigeminal ganglion neurons and subsequent ERK-mediated activation of satellite glial cells (39, 46, 47). In addition to these chemical and surgical manipulations, nutritional imbalance in some BMS patients can be simulated in rodents by exposing them to a zinc-deficient diet, which would lead to thickening and enhanced mitotic activity of the tongue epithelium (48). While these additional models may capture some facets of BMS phenotypes, it should be noted that BMS patients manifest oral burning sensation often without apparent pathological changes, including tissue inflammation and physical nerve damage, or nutritional disturbances. Thus, whether these experimental manipulations provide valuable insights into the mechanistic understanding of BMS remains unclear.
Ⅳ. CONCLUSION
BMS, a disorder in complex natures with multiple contributing factors, features moderate to severe burning pain in conjunction with xerostomia, dysgeusia, halitosis, and psychological distress, all of which would significantly impair the life quality of those affected by the disorder. However, the lack of knowledge on causative relationships between each etiological factor and the expression of BMS symptoms hampers development of universal diagnostic criteria and effective treatment remedies. As a result, our current efforts are limited to symptomatic relief in patients. To overcome this obstacle, recent studies have focused on dissecting the molecular actions each factor might take, especially in their neuropathic nature. As various animal models were established as a part of this effort, they have demonstrated some promise in deciphering molecular mechanisms underlying the pathophysiology of BMS. However, the failure of these alternative models to replicate the full spectrum of BMS phenotypes needs to be overcome to extend their applications in mechanistic studies. Nevertheless, future studies on BMS can significantly broaden our understanding of BMS pathology by strategically refining multi-faceted animal models, which may lead to groundbreaking advancements in research and treatment strategies.