Atypical Odontalgia: A Review of the Literature

Marcello Melis, DDS, RPharm; Silvia Lobo Lobo, DDS, MS; Caroline Ceneviz, DDS; Khalid Zawawi, BDS; Emad Al-Badawi, BDS, MS; George Maloney, DMD; Noshir Mehta, DMD, MDS, MS


Headache. 2003;43(10) 

In This Article


The mechanism through which pain is generated in AO is far from being clear. Some authors suggest an idiopathic origin of this condition because no other hypotheses have proven responsible for precipitating and perpetuating the pain,[20,27,28] while others propose a psychogenic origin of the disease highlighting the evidence of association between AO and several psychological conditions.[7,8,10,11,18,24,26,29,30,31,32] Most of the reports, however, identify AO as a neuropathic pathology characterized by deafferentation.

A possible alternative hypothesis could be based on the Melzack theory of the "neuromatrix"–a neural network whose composition and connections are determined genetically and are later influenced by multiple inputs coming from different parts of the body.[33]

Woda and Pionchon stress that many facial pain conditions occur without any clear explanation or evidence of organic lesions, and this is why they grouped these pathologies under the term idiopathic orofacial pain.[20,27,28] Their approach underscores the unknown origin of the problem, although different hypotheses have been suggested.

One hypothesis suggests a major role of psychological diseases in the development of AO. Many studies indeed support a strong correlation between AO and different psychological conditions such as depression,[7,8,11,18,29,30,32] somatoform pain disorder,[24,26,30,31] anxiety,[18] demoralization,[18,30] introversion,[10] or hypochondriacal psychosis.[17] Rees and Harris examined 44 patients with AO and found that 29 (66%) had a history of depression or depressive symptoms and others had personality disorders.[11] Brooke and Schnurr also reported depression in 41% of 22 patients they examined with AO.[7]

Thus, as in other chronic pain conditions, we do not know if such conditions are the cause or the result of the pain.[19,23,28,34,35] Other authors criticize this hypothesis and question the relevance of the psychological component on pain.[13,19,23,26,27,34,35,36,37] Graff-Radford and Solberg evaluated 19 patients with AO using the Minnesota Multiphasic Personality Inventory (MMPI), and compared them with 19 patients with headache.[37] The profiles of the patients in the study were essentially unelevated, and there was no difference between the patients with AO and those with headache. This may suggest that psychological factors are not significant in the genesis of AO.

The most accredited theory currently is the hypothesis that trauma to the orofacial structures (traumatic injury, periodontal surgery, pulp extirpation, endodontic therapy, apicoectomy, tooth extraction, implant insertion), or even minor trauma (crown preparation, inferior alveolar nerve block) might alter the neural continuity of the tissues creating deafferentation.[21] This falls into the category of neuropathic pain, in that after the wound has healed the neural tissue is responsible for the pain and other related symptoms (paresthesia, dysesthesia).[23,24,38,39]

Multiple mechanisms are involved in the pathogenesis of pain: (1) sensitization of nociceptive fibers,[27,39,40,41,42,43] (2) sprouting of somatic afferent fibers from adjacent intact nerves,[39,40,42,44,45] (3) activation of afferent fibers by sympathetic efferents,[2,14,19,27,39,40,42,44,45,46,47,48,49] (4) cross-activation between injured afferent fibers (ephaptic crosstalk),[27,40,45,50,51] (5) phenotypic switching of afferent neurons,[27,39,40,42,52,53] (6) neuroma formation,[2,9,40,44] (7) changes induced in the central nervous system (CNS),[9,24,39,40,42,45,50,54,55,56,57,58] and (8) loss of inhibitory mechanisms.[9,27,39] Following a nerve injury, afferent fibers may become sensitized showing a lower activation threshold and developing spontaneous ectopic activity as a result of increased expression or redistribution of sodium channels.[39,41] This could explain some of the clinical manifestations of AO such as mechanical or thermal allodynia and persistent spontaneous pain.

Trauma to a nerve also has been associated with the formation of nerve collaterals from other non-injured nerves or from other parts of the same nerve.[45] When this occurs in the dorsal horn of the spinal cord, second-order neurons within lamina II, which usually receive nociceptive inputs, will also receive collaterals from non-nociceptive Aβ-fibers,[39,42] that might bring information misinterpreted as pain even when non-noxious stimuli such as hot, cold, or pressure are applied to the tooth (allodynia).[17,39,40]

Nerve collaterals have been observed sprouting from sympathetic fibers and reaching sensory afferents and sensory neurons of the dorsal root ganglia.[39,40,41,48,49] In addition, α-adrenergic or β-adrenergic receptors become expressed in those cells which develop sensitivity to catecholamines.[27,40,41,42] The role of the sympathetic system in the perception of pain in AO seems to be confirmed by the fact that sympatholytic procedures (stellate ganglion block, phentolamine infusion) usually significantly reduce the pain.[14,19,21]

Cross-activation of nerve fibers also has been reported by the formation of ephapses between injured sensory neurons.[27,40,51] This anatomical and functional connection might be responsible for enhanced ectopic and natural firing of the involved neurons,[27,40,50] leading to increased pain perception.[40]

Nerve fiber lesion, as well as other factors (hormonal, degenerative, traumatic, or psychological), has even been theorized to induce phenotypic changes in sensory afferent neurons causing altered expression of sodium/potassium channels, receptors, and neurotransmitters.[27,39,40] Non-nociceptive first-order neurons start releasing substance P and calcitonin gene-related peptide, activating second-order neurons in the dorsal horn of the spinal cord.[40,42,52,53]

Another phenomenon that may occur after a nerve has been damaged is the formation of a neuroma–a heterogeneous formation containing axoplasmatic elements, myelin, Schwann cells, and connective tissue elements that originates from the injured nerve growing in a disorganized fashion.[9,23,40,50,59] Neuromas are extremely sensitive to mechanical stimulation (pressure and tension) and norepinephrine, and produce continuous or episodic pain that can be spontaneous or triggered by external stimuli.[9,23,40,60] Similar characteristics appear in demyelinated axons, even in absence of a true neuroma.[40,50,61,62,63]

The CNS is also involved in the perception of pain. There is evidence that nerve lesions occurring peripherally lead to central changes at the cellular level and to functional changes in the CNS.[40,64,65,66,67] Similar to first-order neurons, the neurons in the receptive zones of the brain stem, within the CNS, demonstrate ectopic activity after trauma.[40,57] Some second-order neurons are normally intrinsically rhythmogenic, but the magnitude of discharge increases after nerve injury.[40,68,69,70,71,72]

A major role in the perpetuation of chronic pain by the CNS is attributed to N-methyl-D-aspartate (NMDA) receptors located on central nociceptive neurons.[27,39,40,42,73,74] Their activation is subordinated to activation of amino-3-hydroxy-5-methylisoxazole-4-propionic acid (AMPA) receptors or metabotropic receptors. These receptors cause the removal of the magnesium block from NMDA receptors and render them available for binding the neurotransmitter, glutamate, which has high affinity for this type of receptor.[39,40,73] The effect of this interaction is increased sensitivity to external stimuli and chronic pain that becomes independent of peripheral inputs, and these are the characteristics of AO.[40,73,74]

As a significant part of the CNS, we think inhibitory control of afferent stimuli deserves to be discussed separately. Since the amount of impulses that reaches the brain from the periphery is the result of excitatory and inhibitory modulation existing within the CNS,[74] we can achieve the same effect of increasing pain perception by either increasing the excitatory component or decreasing the inhibitory component.

A major role in the modulation of nociceptive stimuli in the CNS is performed by the reticular formation of the brain stem, which is part of the paleospinothalamic tract that carries "slow pain" impulses.[75] Marbach hypothesized that deafferentation by, for example, dental pulp denervation or tooth extraction, reduces the amount of inputs into the reticular formation, decreasing its inhibitory influence.[9] Yet, disinhibition can occur in other sites of the CNS such as the dorsal horn of the spinal cord. This effect might be due to inhibitory interneurons' death following deafferentation,[27,39] down-regulation of inhibitory neurotransmitters within the same interneurons, or down-regulation of presynaptic inhibitory receptors on primary sensory neurons. These changes would enhance primary afferent excitability.[39]

According to the Melzack theory, a matrix of neurons exists that is genetically predisposed and is successively modified in response to inputs coming from several sources. These include somatic receptors, visual and other sensory inputs that influence the cognitive interpretation of the situation, phasic and tonic cognitive and emotional inputs from other areas of the brain, intrinsic neural inhibitory modulation inherent in all brain function, and the activity of the body's stress regulation system, including cytokines as well as the endocrine, autonomic, immune, and opioid systems.[33] The summation of these factors leads to the output of the neuromatrix that is called neurosignature, and is the processing and synthesis of all the nerve impulses in the brain. Body sensations that are normally elicited and modulated by sensory stimuli can be felt also in absence of these stimuli, because of processing occurring in the brain.[33] This is why previous pain experience could lead to change in existing neuromatrix and, in turn, in neurosignature; persistent input from altered neuromatrix can make the change persist after the pain event and become chronic, even without true deafferentation.

These changes in the peripheral and CNSs are common to many neuropathic diseases as it is evident reading the original articles. This is why AO probably shares the same pathophysiological mechanisms of other pathologies such as complex regional pain syndrome (CRPS) and traumatic neuralgia. It is also assumable that inferior alveolar or lingual nerve lesions, that are moderately common during dental procedures, can start a symptomatology resembling symptoms of either AO, CRPS, or traumatic neuralgia, even though CRPS seems to be very rare in the head and neck region.[40]

Nevertheless, all the mechanisms reviewed in this section could possibly play a role in the precipitation and perpetuation of pain in AO. For this reason, both diagnosis and management of this condition are still a challenge even for expert clinicians.