灼口综合征是神经病理性痛吗？
说明：文章选自Jääskeläinen S K. Is burning mouth syndrome a neuropathic pain condition?[J]. PAIN,2017:1-4.
原文链接：
https://insights.ovid.com/pubmed?pmid=29257770
1.前言
根据国际头痛协会（IHS）最新分类，灼口综合征（burning mouth syndrome, BMS）归属于痛性颅神经病，定义为：口腔内有烧灼疼痛感，每天反复发作时间超过2小时，持续3个月以上，没有明确的临床病因。疼痛程度为中到重度，与牙痛相似，具有明显黏膜烧灼感，常常伴有味觉改变和口干（50%-70%的患者出现），最常见于舌尖，但口腔内任何一处黏膜都有可能受影响。据估算，灼口综合征人群发病率差异相当大，从0.02%到40%，但在最近一份具有较好对照设置的临床研究中，运用目前的诊断标准，估计灼口综合征发病率小于1%-3.7%。灼口综合征在绝经后妇女中发病率最高（18%），女性与男性比例范围从3：1到20：1。许多灼口综合征患者在夜间缓解，疼痛不影响睡眠，早晨症状轻，白天症状加重。根据定义，无量化心理物理学的临床调查和临床感官检查结果是正常的，因此BMS与目前国际疼痛研究协会（IASP）的神经性疼痛的定义不相符合。因此IHS所持的BMS为痛性颅神经病和IASP的神经病理性疼痛定义出现了分歧。尽管如此，在近二十年的对BMS进行的研究中采用了严格诊断标准，以及与临床神经生理学，心理物理学，神经病理学工具以及功能性大脑成像有关的精密诊断，这些研究提示神经病理参与大多数患者不同层面的神经轴索。
在下文中，总结原发性BMS的神经病原学证据。
2.神经病理参与灼口综合征的证据
2.1灼口综合征与周围神经系统
由大髓鞘的Aβ纤维介导的脑干反射记录表明约20%临床典型的原发性BMS患者在三叉神经或脑干回路已经有损伤的迹象。BMS亚组患者为没有明显临床征象的亚临床三叉神经痛，这与临床感官检查诊断敏感性差相当，尤其是神经损伤后慢性阶段。
热定量感官测试（QST）和舌头黏膜活检进一步阐明了周围神经参与BMS痛。BMS大部分患者（76%）的热定量感官测试表现出感觉减退，特别是在无损伤的冷热变化范围内出现较小的程度痛觉减退。BMS患者定量感官测试图与其诉说的神经性病理疼痛症状非常相似。BMS患者热量感官测试功能丧失迹象，后来被证明是由于舌上皮纤维小神经的局灶性损伤造成的。在个别BMS患者，大小纤维病理可能重叠发生。此外，电味觉测量异常表明原发BMS病理生理过程也同样涉及小Aδ味觉纤维，这可以解释这些病人味觉频繁改变。有趣的是在BMS中，Aδ冷传入较C纤维发生损伤更常见，表明小纤维向中枢神经系统传输的不平衡。在正常情况下，Aδ纤维对多型C感受器信号产生抑制作用，从而对C纤维有相对保护的作用，而在BMS患者体内当Aδ纤维受到损害时，对C纤维抑制解除，将导致持续烧灼疼痛感，这与中枢疼痛相类似。
除了热模式功能丧失外，QST还显示一小部分BMS患者的功能获益，热痛耐受性降低，热痛感觉过敏，异常性疼痛。同样脑干反射记录客观显示了解除对三叉神经脑干复合体的抑制，如在三分之一BMS患者眨眼反射R2成分缺失。因为眨眼反射受下行纹状体多巴胺能控制[3]，神经递质-正电子发射断层扫描（PET）研究阐释了BMS的中枢多巴胺系统病理生理学机制。这些实验旨在验证纹状体多巴胺能下行抑制缺陷作为神经性头面疼痛的主要触发因素的学说。这些结果，以及其他收集到的有关中枢神经系统(CNS)参与原发性BMS的证据，将在下一部分讨论。
2.2灼口综合征与中枢神经系统
典型的神经性或中枢性疼痛变化可通过功能性脑成像方法显示出来。在fMRI上，BMS患者对疼痛热刺激较对照组脑活动量下降，尤其是双侧丘脑，这与其他神经性疼痛的功能性脑成像结果相似，都是因为躯体感觉神经通路传入受阻。使用氟代-DOPA和11C雷氯必利扫描的纹状体多巴胺系统神经递质研究显示BMS患者突触多巴胺水平较对照组下降。BMS患者PET检查结果与帕金森病早期相似，在帕金森病患者人群中，中枢型神经性痛非常常见，其BMS发生率也被推测较正常人群增加。关于BMS，人类基底节尤其是大脑多巴胺能系统在处理和调节临床疼痛的重要作用，我们研究小组的神经生理和神经递质PET研究成果给出了第一个直接的证据，这一点在后来几份综述中得到进一步的证实。试验证据证实纹状体多巴胺系统和三叉神经痛之间的联系，例如纹状体多巴胺通路损害已经被证实可以诱导三叉神经分布区域异常性疼痛。
此外，基因决定的多巴胺D2受体（DRD2）通过DRD2957C>T单核苷酸多态性起作用的功能已经被证明与包括口面疼痛等神经性疼痛症状严重程度风险有关。与神经突触有多巴胺水平低有关的纯合子957TT基因型发生率在神经性疼痛患者人群（50%）较正常人群（27%）增加，这种基因型患者NRS评分具有最高疼痛程度。
除了心理学和神经生理学方法，舌神经外周利多卡因阻滞可将BMS患者分为2个不同的亚型组。外周型组对局部麻醉表现出良好的镇痛反应，而中枢型组在外周神经阻滞后无反应或甚至出现痛觉过敏。这个简单方法似乎也可以用来预测氯硝西泮局部外用治疗仅对外周型组BMS患者有益。此外中枢型组在焦虑抑郁评分上得分较高。
BMS患者的精神共病得到认识，目前认为是次要的、非特异性现象，在慢性疼痛患者中频繁出现。然而有关BMS心理因素的大多数研究却运用了不能用来诊断精神病的调查问卷。BMS精神共病的一种解释已经在一项关于BMS患者的深入研究中提出，利用结构化精神病访问，提供目前状态精神疾病（轴I）和终生品质人格障碍（轴II）。根据这项研究，BMS患者遭受精神疾病的折磨是因为大脑多巴胺水平低下所致，55%的BMS患者有焦虑和社交恐惧，16%BMS患者是C型人格（恐惧或神经质型）。在一个运用调查问卷的BMS研究中，也发现了相似的精神病状况。因此脑多巴胺系统内在或诱发的衰弱可能是慢性神经性疼痛和共病精神障碍两者的共同路径。BMS疼痛的多巴胺假说的进一步支持来自非侵入性脑刺激研究，该项显示重复经颅磁刺激，初始释放纹状体多巴胺能激活内源性阿片物质系统，进而也可效缓解BMS疼痛。
3.讨论
从几方面研究搜集到证据，包括了从上皮神经纤维到大脑的神经通路，表明临床大多数典型BMS病例是慢性神经性疼痛，包含2个亚组（周围型组和中枢型组）。这些情形是亚临床的，没有表现出明显的神经病临床体征，因此只能依赖神经生理学、心理物理学和神经病理学调查才能正确识别诊断和分类。把BMS患者分类周围型和中枢型，使用外周舌神经利多卡因阻滞方法进行分类有一定的帮助，但仍然需要再通过神经生理学、物理心理学和神经病理学诊断确认。第一个BMS亚组涉及亚临床周围神经病性疼痛，是由更广泛亚临床三叉神经病变和三叉神经脑干损伤引起，或是由口腔黏膜单纯小纤维神经病变，小纤维神经病变在相关确认性试验中显示功能丧失。局部小纤维性BMS也可适用于非长依赖性小纤维神经病，在女性中更常见，表现出症状分布混乱，例如可以覆盖面部或躯干。另一个BMS亚组是由伴见神经生理学和神经递质异常的中枢型BMS组成，表现出脑多巴胺水平低下和神经病共病发生率增加，在QST时有功能获益征象。个别患者，周围型和中枢型相互重合，在功能丧失或增益方面有者不同组合。
BMS患者共病精神障碍的调查分析是有趣的，尤其是关于纹状体低多巴胺水平的PET表现。内源性多巴胺-阿片轴和基底节循环调控下行抑制功能减弱可能代表容易遭遇抑郁焦虑以及C型人格障碍，以及慢性神经性疼痛。因此，神经性疼和抑郁不是彼此间因果关系，而是因神经病理外周或中枢神经系统参与下，先天或获得性脑低多巴胺水平的结果。
最近，一种新的假设被提出来，来概括导致BMS的各种病理过程和解释相关问题，如BMS患者性别差异、年龄分布等。将BMS的四个特点考虑进去，（1）绝经后妇女易发；（2）精神疾病类型共病；（3）症状的口腔定位；（4）在确证性试验中，周围神经和中枢神经系统异常。总之，慢性焦虑抑郁伴随肾上腺类固醇生理的压力性改变或性激素水平的更年期剧变导致神经保护性类固醇下降如脱氢表雄酮。在神经毒影响存在下，这种神经保护和再生的能力下降将诱导小Aδ纤维（依赖依赖脱氢表雄酮）和黑质纹状体多巴胺神经元早期损伤，Aδ和黑质纹状体多巴胺神经元特别容易受到内外毒素的影响。这些因素单独或与多巴胺阿片肽轴先天基因薄弱，最终导致临床BMS表型（外周型、中枢型，或伴见神经病特征）。
此外值得注意的是，在神经生理学研究中，大约20%的临床典型BMS患者显示亚临床三叉神经系统病理学；例如，舌部下颌神经亚临床损伤和脑干水平病变。从现有的文献来看，很明显的是感官检查不能提供敏感的工具来足够精确诊断和分类神经性口面部疼痛。目前神经性病理疼痛定义需要解剖学症状和体征以及确证性试验检测，这是不恰当的，因为没有认识到亚临床神经病，因此这个定义引起了争论。此外，它目前不独立接受确证性试验结果作为神经性痛证据。然而，例如纯小纤维神经病变和面部区域内神经病变，临床感官检查特别困难，因此只运用传统工具来诊断是具有挑战性的。在神经病理性疼痛定义中，从症状直接到神经生理、心理物理学和神经病理学，和或功能脑成像结果，越过临床体征，然后可以建立神经性疼痛定义（如BMS）。此外，非长依赖性小纤维神经病变和基底节区中枢型疼痛的症状和体征分布不符合经典的外周或中枢神经解剖分布，因此在确定神经病理性疼痛标准时，应该考虑这一点。因此考虑到现有的神经病病因学证据，BMS的IHS分类中的疼痛性神经病似乎是最合适的。
正确诊断和分类BMS的2个亚型（外周和中枢型）是非常重要的，因为在将来将指导病人选择最佳治疗方案。外周型BMS患者可能受益于局部治疗，例如氯硝西泮和辣椒素，而中枢型BMS通过非侵入性神经调节技术（如经颅重复磁刺激或多巴胺能药物）作用于多巴胺-阿片轴来缓解疼痛。此外，BMS患者的单纯小纤维神经病变体征提示系统研究小纤维神经病变可能的病因，其中一些可以治愈。这些选择方案值得进一步的研究，通过适当和灵敏的诊断方法加以控制。
原文：
Is burning mouth syndrome a neuropathic pain condition?
1. Introduction
In the most recent classification by International Headache Society,21 burning mouth syndrome (BMS) is classified under the header “Painful cranial neuropathies” and defined as an “intraoral burning or dysaesthetic sensation, recurring daily for more than 2hours per day over more than 3 months, without clinically evident causative lesions.” The pain is moderate to severe, similar to tooth ache in intensity but has a distinct superficial, burning character and is often accompanied by taste alterations and xerostomia (in 50% to 70% of the patients).31 The tip of the tongue is most frequently affected, but any part of the intraoral mucosa may be involved. The estimates of population prevalence of BMS range considerably, from 0.01% to 40%, but in more recent well-controlled studies using current diagnostic criteria, it has been estimated to be less than 1% to 3.7%.31 Prevalence of BMS is highest in postmenopausal women (18%), and reported female-to-male ratios range from 3:1 to 20:1.31 Many patients with BMS report night benefit; the pain does not disturb sleep and is better in the morning, gettingworse during the day.31 By definition,21 clinical investigations and clinical sensory examination, without quantitative psychophysical measures, are normal and thus, BMS does not seem to fit the current IASP definition of neuropathic pain, which arises as a direct consequence of a lesion or disease affecting the somatosensory system, with both symptoms and clinical signs within a neuroanatomically plausible distribution.12 Thus, there seems to be a discrepancy between the IHS classification of BMS as a painful cranial neuropathy and the IASP definition of neuropathic pain.
Nevertheless, research done on BMS during the last 2 decades,using strict clinical diagnostic criteria for primary BMS and sophisticated diagnostics with clinical neurophysiologic, psychophysical,and neuropathological tools as well as functional brainimagi ng, has revealed neuropathic involvement at various levels ofthe neuraxis in the majority of the patients with BMS [reviewed in
detail in: Ref. 26,31].
In the following section, the existing evidence for neuropathicetiology of primary BMS will be summarized
2. Evidence for neuropathic involvement in burning mouth syndrome
2.1. Peripheral nervous system in burning mouth syndrome
Brainstem reflex recordings, mediated via large myelinated Ab afferents have shown signs of damage in the trigeminal nerve or its brainstem circuits in approximately 20% of clinically typical primary patients with BMS.13,26,27 This subgroup of patients with BMS represents subclinical trigeminal neuropathic pain without clear clinical signs, which is compatible with the poor diagnostic sensitivity of clinical sensory examination, especially at chronic stage after nerve injury.25,30,44,45 Thermal quantitative sensory testing (QST) and tongue mucosal biopsies have further elucidated the peripheral nervous system involvement in BMS pain. The majority (76%13) of patients with BMS show hypoesthesia in thermal QST, especially to innocuous cooling and warming and, to a lesser extent,
hypoalgesia.13,15,22,38,42 Quantitative sensory testing profiles in patients with BMS thus very much resemble those reported in a large cohort of different neuropathic pain conditions.35 These
loss-of-function signs in thermal QST of patients with BMS have later been shown to be due to focal damage of the small nerve fibers of the tongue epithelium.4,33,38,47 In individual patients with
BMS, overlap between large and small fiber pathology may occur.13,31,33 Furthermore, abnormalities in electrogustatometry suggest that pathophysiological process in primary BMS involves the small Ad taste afferents as well,11,16,24,37 giving explanation to frequent taste alterations in these patients. Of interest is that in BMS, the Ad cool afferents seem to be more often impaired than C fibers,13,38 indicating an imbalance within small fiber input to the central nervous system. As the cool Ad fibers exert a tonic inhibition of the polymodal C nociceptor signaling in normal conditions,9 this kind of more severe damage to the Ad fiber system with relative preservation of C fiber function could lead to ongoing burning pain sensation because of unmasking or disinhibition of the system, in BMS 26 similarly as in central pain.9 In addition to loss-of-function in thermal modalities, QST has shown gain-of-function in a small proportion of patients with BMS
either in the form of decreased heat pain tolerance15,17 or heat pain hyperalgesia and allodynia.13,16 Likewise, brainstem reflex recordings have objectively shown disinhibition of the trigeminal
brainstem complex in the form of deficient habituation of the blink reflex R2 component in approximately 1/3 of the patients.13,27,36 As blink reflex habituation is under descending nigrostriatal dopaminergic control,3 neurotransmitter–positron emission tomography (PET) studies have been conducted to elucidate possible central dopaminergic system pathophysiology in BMS.
These were intended to test the hypothesis of deficient striatal dopaminergic top–down inhibition as a major trigger of neuropathic orofacial pain.18,19,26,28,29 The results, together with other converging evidence for central nervous system (CNS) involvement in primary BMS will be dealt with in the next part.
2.2. Central nervous system in burning mouth syndrome
Changes typical for neuropathic or central pain have been shown in BMS with functional brain imaging methods (for detailed review, c.f.31). Patients with BMS show less volumetric brain activation in fMRI to painful hot stimuli than control subjects, especially in bilateral thalamus,1 which is similar to functional brain imaging findings in other neuropathic pain conditions because of deafferentation of the somatosensory pathways.2
Neurotransmitter PET studies on the striatal dopamine system with fluoro-DOPA and 11C-raclopride scans indicate a decrease in synaptic dopamine levels in patients with BMS compared to controls.18,28 The PET findings in BMS are similar to those in early Parkinson disease,20 in which central type neuropathic pain is rather common, and the incidence of BMS has been suggested to be increased compared to general opulation.7,8,36,39 Our neurophysiologic and neurotransmitter PET findings in BMS gave the first direct evidence in humans for the important role of basal ganglia, and especially the brain dopaminergic network in processing and modulation of clinical pain,18,19,28 which has later been further corroborated in several reviews.5,23,26,31 Experimental evidence supports the connection between the striatal DA system and trigeminal pain, as lesioning of the
nigrostriatal DA pathway has been shown to induce allodynia within the trigeminal distribution.10
Furthermore, the genetically determined function of the dopamine D2 receptors (DRD2) via the DRD2 957C . T singlenucleotide polymorphism has been shown to be related to the risk for and symptom severity in neuropathic orofacial pain, including BMS.29 The prevalence of the homozygous 957 TT genotype, associated with low synaptic dopamine levels in the striatum, is increased (50%) in patients with neuropathic orofacial pain compared to general population (27%), and patients with this
genotype report the highest pain intensities in NRS scores.29
In addition to psychophysical and neurophysiological methods, peripheral lidocaine block of the lingual nerve can be used to cluster patients with BMS into 2 distinct subgroups. The peripheral subgroup demonstrates good analgesic response to local anesthesia, whereas the central subgroup shows no response or even hyperalgesia after peripheral nerve block.16 This easy procedure also seems to be able to predict the response to topical clonazepam treatment that was beneficial
only in the peripheral subgroup of patients with BMS.16 Furthermore, the central subgroup showed higher scores in hospital anxiety and depression scores.
Psychiatric comorbidity in patients with BMS is recognized and currently often considered a secondary, nonspecific phenomenon frequently encountered in patients with chronic pain.14 However,
most studies on psychological factors in BMS have applied questionnaires that cannot be used for proper psychiatric diagnostics. A novel explanation for psychiatric morbidity in BMS
has been presented in a thorough study on patients with BMS utilizing structured psychiatric interviews, giving current (state) and lifetime (trait) diagnoses for both psychiatric (axis I) and personality (axis II) disorders.43 According to this study, patients with BMS seemto suffer only of psychiatric conditions attributed to low-brain dopamine tone: major depression and social phobia were found in 55% of the patients with BMS, and type C (fearful/neurotic) personality in 16% of the patients with BMS.43 Similar psychiatric profile has also been found in a BMS study using questionnaires.16 Thus, inherent or induced weakness of brain dopamine system may offer a common pathway to both chronic neuropathic painand comorbid psychiatric disorders.26,31,43
Further support to the dopamine hypothesis of BMS paincomes from noninvasive brain stimulation studies showing that repetitive transcranial magnetic stimulation, by initially releasing dopamine in the striatum40,41 thereby activating the endogenous opioid system,32 also effectively relieves BMS pain.34,46
3. Discussion
Current converging evidence from several lines of investigations, covering neural pathways from the epithelial nerve fibers to the brain, indicates that clinically typical BMS, in the majority of cases,is a chronic neuropathic pain condition, consisting of 2 main subgroups, peripheral and central. These conditions are subclinical, as they do not show evident clinical signs of neuropathy and thus, they can be correctly identified and classified only by means of neurophysiologic, psychophysical, and neuropathological investigations. Peripheral lingual nerve lidocaine blocks may also help in classifying the patients with BMS into peripheral and central subgroups, but this method should still be validated against thorough neurophysiologic, psychophysical, and neuropathological diagnostics. The first BMS subgroup involves subclinical peripheral neuropathic pain, caused either by more
extensive yet subclinical trigeminal neuropathies and trigeminal brainstem lesions, or by pure small fiber neuropathy of the intraoral mucosa with loss-of-function signs in confirmatory tests.
The local small fiber type of BMS might also fit in the entity of nonlength-dependent small fiber neuropathy that is more prevalent in women than men and shows a patchy distribution that may cover, for example, face or trunk.6 The other subgroup consists of central BMS with neurophysiologic and neurotransmitter PET signs, indicating low-brain dopamine tone and increased prevalence of psychiatric comorbidity and, sometimes, additional gain-of-function signs in QST. In individual patients, these subtypes, peripheral and central, may overlap with different combinations of loss- or gain-of-function signs.26,31
Profile of comorbid psychiatric disorders in patients with BMS is interesting, especially regarding the PET findings of low striatal dopamine tone in these patients. Weak top–down inhibitory control via endogenous dopamine–opioid axis and basal ganglia circuits could represent shared vulnerability both to depression, anxiety as well as type C personality disorders, and to chronic neuropathic pain. Accordingly, for example, neuropathic pain and depression would not be causal to each other but, instead, result from common predisposition because of low-brain dopamine tone, genetic or acquired, in the presence of neuropathic involvement of the peripheral and/or CNS.26,31,43
Recently, a novel hypothesis has been proposed to encompass various pathophysiological processes leading to BMS and to explain open questions regarding, for example, gender differences and age distribution of patients with BMS.31 Four key features of BMS have to be taken into account: (1)
preponderance of postmenopausal women, (2) type of psychiatric comorbidity, (3) oral location of the symptoms, and (4) peripheral and CNS abnormalities found in confirmatory tests. In short, chronic anxiety/depression with stress-related alterations in adrenal steroid physiology and/or drastic menopausal changes in gonadal hormone levels lead to significant decrease in neuroprotective steroids (eg, dehydroepiandrosterone).48 In the presence of neurotoxic influences, this reduction in neuroprotective and neuroregenerative capacity induces preferential damage of small Ad fibers (dependent on dehydroepiandrosterone in maintenance), and nigrostriatal DA neurons, both of which
are especially vulnerable to internal and external toxic agents. These factors alone or in combination with genetically weak top–down inhibition via DA–opioid axis can then eventually cause
clinical BMS phenotype with peripheral, central or combined neuropathic features (for detailed discussion, c.f.31,48).
In addition, it is noteworthy that in neurophysiologic investigations, approximately 20% of clinically typical patients with BMS show subclinical trigeminal system pathology; for example, subclinical lingual of mandibular nerve injuries or brainstem level lesions. From the existing literature, it is obvious that clinical sensory examination does not provide tools sensitive and accurate enough for the diagnostics and classification of neuropathic orofacial pain.25,26,30,31,44,45 It may thus be argued
that the current definition of neuropathic pain, requiring neuroanatomically plausible symptoms and clinical signs in addition to confirmatory tests for definite diagnosis of neuropathic pain,12 is not appropriate, as it does not recognize subclinical neuropathic pain conditions. Furthermore, it currently does not accept results of confirmatory tests alone as evidence for neuropathic nature of the pain.12 Nevertheless, for example, pure small fiber neuropathies and neuropathies within the orofacial region, where clinical sensory examination may be especially difficult, are challenging to diagnose with traditional clinical tools only. In the definition of neuropathic pain, a shortcut from symptoms to
neurophysiologic, psychophysical, neuropathological, and/or functional brain imaging findings, bypassing the requirement of neuroanatomically plausible clinical signs, could thus provide
a reasonable way to establish definite diagnosis of neuropathic pain in conditions such as BMS. In addition, the distribution of symptoms and signs in nonlength-dependent small fiber neuropathies6 and central pain conditions due to basal ganglia disorders7,9 does not comply with classic peripheral or central neuroanatomical distributions, which should be taken into account when defining criteria for neuropathic pain. Accordingly, taking into account the existing evidence for neuropathic etiology
of BMS, the current IHS classification ofBMSunder painful cranial neuropathies21 seems most appropriate.
Correct diagnostics and classification of the 2 neuropathic subgroups of BMS, peripheral and central, is important, as this may in the future guide the choice of the best treatment on individual patient level. Patients with signs of peripheral neuropathic BMS pain could benefit from topical treatments, for
example, with clonazepam or capsaicin, while pain in central BMS may best be alleviated by boosting the endogenous top–down control via brain dopamine–opioid axis with noninvasive neuromodulation techniques such as repetitive transcranial magnetic stimulation or with dopaminergic medications.
Furthermore, signs of pure small fiber neuropathy in patients with BMS imply systematic nvestigation of possible etiologies for small fiber neuropathy, some of which may be curable. These options warrant further studies, controlled with appropriate and sensitive diagnostic methods.
Conflict of interest statement
The author reports no conflicts of interest.
联系客服
