Burning mouth syndrome (BMS) is a chronic pain disorder the etiology of which has remained unclear. Symptoms associated with BMS vary and may have a negative impact on oral health-related quality of life in patients with BMS (1). McMillan et al. (2) reported that, so far, no therapy has been proven to be effective or ineffective in patients with BMS. To date, there is no clear evidence that there is dysfunction of central and/or peripheral nervous system in patients with BMS (3, 4). There are 3 subgroups of BMS patients: subgroup with peripheral small diameter fiber neuropathy of intraoral mucosa (50-65%), subgroup with subclinical lingual, mandibular or trigeminal system pathology (20-25%), subgroup with presentation of central pain pattern which can be attributed to dysfunction of central dopaminergic or serotoninergic pathways (20-40%) (5). Functional studies using 6-(18F) fluorodopa (FDOPA) positron emission tomography in BMS patients have shown a dysfunction of striatal dopaminergic system characterized by reduced presynaptic activity of striatal dopamine neurons and changes in dopamine D2 receptors availability (6). Recently, Sinding et al. (3) have found grey matter changes in 12 subjects with BMS using voxel-based morphometry. The anterior and posterior cingulate gyrus, lobules of the cerebellum, insula/frontal operculum, inferior temporal area, primary motor cortex, and dorsolateral pre-frontal cortex were affected. Furthermore, Liu et al. (7) found lower regional cerebral blood flow in the left temporal and parietal lobes, which might correlate with depression in patients with BMS.
The aim of this study was to analyze patients with burning mouth syndrome using transcranial sonography.
Material and methods
This study was approved by the Ethics Committee of the School of Dental Medicine, University of Zagreb, Croatia. All participants gave their informed consent in writing prior to inclusion in the study, according to Helsinki II. An evaluation of brain parenchyma using transcranial ultrasonography was performed in 20 patients with BMS (64.7±12.3 years) and 20 controls with chronic lumbosacral pain (61.5±15) (Figure1).
The evaluation of brain parenchyma was performed by 3,5 MHz probe in a standardized manner (Aloka Prosound SSD-5500, Hitachi Medical Corporation, Japan) at acoustic temporal bone windows (the transducer placed at the pre-auricular site) in supine position with head elevation of up 45° and side tilt of 30° to the right and to the left. The evaluation was performed at midbrain level (axial section). A distinct brain area was described as hyperechogenic if the planimetrically measured area of the echogenic signal was larger in the patient than in the general population (substantia nigra -SN) or when the increased intensity of the ultrasound signal was visually seen in that area compared to the surrounding brain tissue (thalamus, lenticular nucleus, caudate nucleus were isoechoic to surrounding tissue; normal, moderate or pronounced hyperechogenicity - grade 1, 2, 3). Certain areas of brain were noticed as hypoechogenic when the planimetrically measured echogenic region was smaller in the patient than in the general population. Furthermore, if the area known to have high echogenicity (e.g. midbrain raphe) had a low echogenicity, this was also measured as hypoechogenic due to discontinuity or “invisibility” of echo signal. The echogenic sizes on one side, which were smaller than 0.20 cm˛, were classiﬁed as normally echogenic. The sizes ranging between 0.20 cm˛ and 0.25 cm˛ were categorized as moderately hyperechogenic, whereas sizes of 0.25 cm˛ and larger (10% of the general population) were classiﬁed as markedly hyperechogenic. A normal finding for substantia nigra was weakly echogenic, echogenic area <0.20 cm2, for red nucleus and nuclei raphe weakly echogenic, markedly echogenic (grade 2).
The t-test with significance set at p<0.05 was used for statistical analysis.
Clinical findings of the transcranial sonography of the brain parenchyma have shown that there was a significant difference in echogenicity of substantia nigra, midbrain raphe and brain nucleus in patients with BMS. Hypoechogenicity of the substantia nigra and midbrain nuclei as well as hyperechogenicity of the brain nucleus were found in BMS patients (p<0.05) (Table 1) (Figure 1) when compared to controls (Figure 2).
Results of pain scales (VAS, LANSS), depression scales (Back&Hamilton) and the quality evaluation (SF36) showed high intensity of pain with dominant neuropathic component, anxio-depressive disorders and lower life quality in patern with BMS when sompared the controls (p<0.05), (Table 2).
Transcranial sonography findings of the brain parenchyma have shown that there was a difference in echogenicity of substantia nigra, midbrain raphe and brain nucleus in patients with BMS. Most of the dopaminergic neurons of the brain originate in the midbrain and are found in the substantia nigra and they control descending pain inhibition pathways. It is a well-known fact that abnormalities in dopaminergic neurotransmission within these pathways are connected with different painful syndromes as well as other conditions associated with BMS. In previous studies, those abnormalities were confirmed by EMNG blink reflex hyperactivity which is under inhibitory control of mesencephalic dopamine (8). Positron emission tomography in BMS patients is characterized by reduced presynaptic activity of striatal dopamine neurons and changes in dopamine D2 receptors availability (9).
Studies of the neuroendocrine role of serotonin have shown that nuclei raphe play an important role in the regulation of secretion of the corticotrophin releasing hormone (CRH) and adrenocorticotropic hormone (ACTH), both at the hypothalamic, pituitary portal and pituitary gland level and possibly also at the adrenal level. These mechanisms are involved in descending serotonin and norepinephrine inhibition of descending pain pathways. These studies were performed in vivo on experimental animals (rat model) with stereotactic lesions of target pathways (10). These neurotransmitters are also involved in humans and they affect depression due to close association of the brainstem raphe nuclei with dorso-caudal limbic system. In some previous studies, the authors (11, 12) used transcranial ultrasonography of the brain parenchyma, thus revealing reduced raphe echogenicity in patients with major depressive disorder and, also, with suicidal tendencies. In the present study, hypoechogenicity of the midbrain raphe was found in BMS patients.
Dieb et al. (13) reported that nigrostriatal dopaminergic depletion produces trigeminal neuropathic pain that at least involves a segmental mechanism in a rat model. Khan et al. (14) showed multiple lines of evidence of disturbed structure and function in the medial prefrontal cortex and hippocampus in BMS patients, which might also affect mood.
Three sonographic abnormalities were found in patients with typical restless legs syndrome (13-15): substantia nigra hypoechogenicity (17%), nuclei raphe hypoechogenicity (26%) and red nucleus hyperechogenicity (26%). These results suggest similar underlying mechanism of both conditions. It seems that BMS could be analyzed as an oral variant of restless legs syndrome.
In conclusion, transcranial ultrasonography is a non-invasive, reproducible method which enables visualization and evaluation of stem structures and their environment of patients with BMS. The only disadvantage of this method is a narrow or thickened acoustic window.