Abstract
We present a series of seven migraine patients with typical features of a migraine attack without aura, but atypical pain localization in the face in one or both of the lower two distributions of the trigeminal nerve (V2 and V3). All of them responded well to triptans. Three patients responded to preventive treatment for migraine with β-blockers (n = 2) or valproic acid (n = 1). These cases underline the heterogenic clinical presentation of migraine, which is sometimes difficult to diagnose even for headache specialists, and broaden the pathophysiological understanding of trigeminal nociceptive processing in migraine in the light of neuronal plasticity.
Introduction
Migraine is a common, sometimes severe, episodic disorder of the central nervous system affecting an estimated 10–15% of the population and is associated with mostly unilateral headache, sensitivity to movement and autonomic symptoms such as nausea and vomiting, as well as photo- and phonophobia (1, 2). Migraine can present in multiple subtypes with different phases and accompanying complaints. Most migraine patients locate their hemicranial pain at the front, temporal and/or parietal part of the head, but in a considerable number of patients pain exceeds this classic trigeminal territory and spreads to the back of the head, the neck and face. The underlying pathophysiology of these pain referral mechanisms and concomitant changes in sensory perception (e.g. allodynia) in affected painful structures is just beginning to be understood. A small number of single case descriptions in the recent literature have described migraine presenting as pain in the mouth and teeth (3, 4). Often, these patients are first seen by their dentist and the underlying migraine symptoms remain unrecognized or are frequently misdiagnosed as trigeminal neuralgia or ‘sinus headache’ (5).
We report a series of seven migraine patients with typical unilateral features of a common migraine attack, but atypical pain localization in the face in one or both of the two usually unaffected distributions of the trigeminal nerve (maxillary nerve = V2, or mandibular nerve = V3). The majority of these patients have been misdiagnosed as trigeminal neuralgia. Our description provides further evidence for the heterogenic nature and clinical presentation of migraine, and supports the hypothesis of convergence within the trigeminal pain processing system.
Case reports
The seven patients had been repeatedly examined at the University Duisburg-Essen headache clinic between July 2002 and December 2006. They all fulfilled the International Headache Society criteria for migraine, but presented with atypical pain localization in maxillar or mandibular regions of the face (Table 1). This cohort consisted of one man and six women, mean age 55.4 ± 3.2 years (range 46–68), with a mean duration of illness of 12.4 ± 3.9 years (range 3–29) and a mean attack frequency of 2.9 ± 0.46 attacks (range 1–4) per month. Reported pain intensity was moderate to severe, reflecting a mean pain intensity on a verbal rating scale (VRS) of 8 (range 6–10). The most common symptomatic site of pain was the maxillary nerve (V2) and, to a lesser extent, the mandibular nerve (V3) division of the trigeminal nerve. Pain was always reported to be unilateral. All patients but one reported having accompanying autonomic symptoms such as nausea and vomiting as well as photo- and phonophobia. One patient reported having osmophobia and one observed ipsilateral eylid oedema during the migraine attack. All patients received diagnostic magnetic resonance imaging (MRI) with magnetic resonance angiography to exclude secondary headaches. All MRIs were rated normal by an experienced neuroradiologist. Extensive laboratory blood work was obtained from all patients, including haematology, chemistries, toxicology, endocrinology and urine. Cerebrospinal fluid was obtained from three patients, but revealed no additional information. Three of the seven patients reported frequently having moderately elevated blood pressure treated with β-blockers (patients 5 and 7) in two patients, which remained untreated due to non-compliance in one patient (patient 4). No other concomitant somatic or psychiatric illnesses were reported. Five patients had a positive family history of migraine. All were scheduled to receive a dental examination to exclude pain due to musculoligamental (soft tissue) or dentoalveolar disease. Dental examinationss were normal in all patients except one (patient 3), who refused it because of anxiety. Five of these seven patients had been previously diagnosed with trigeminal neuralgia and unsuccessfully treated with carbamazepine (patients 4–7) and gabapentin (patient 3), one was diagnosed as atypical migraine (patient 2) and one as paroxysmal hemicrania (patient 1). Patient 1 received indomethacin up to 50 mg three times a day with no pain relief. All patients were given an adequate trial medication with a 5-hydroxytriptamine (5-HT1B/1D) agonist (rizatriptan, zolmitriptan or eletriptan) for at least three consecutive migraine attacks. All patients reported significant pain relief after triptan medication. Two patients received preventive medication with β-blockers (metoprolol, patients 5 and 7) and one with valproic acid (patient 1). All of them reported a considerable reduction in the frequency of attacks under treatment, which was documented in the headache diaries all patients were asked to fill out for at least 3 months (see Table 1).
Patient characteristics according to IHS diagnostic criteria
VRS, Verbal rating scale; MRI, magnetic resonance imaging; HIS, International Headache Society; NA, not applicable.
Discussion
This observational study was initiated by one of the authors (H.C.D.), who had himself experienced unilateral pain in the third division of the trigeminal nerve (V3) accompanied by typical autonomic features of migraine such as phono- and photophobia as well as nausea. Disregarding the autonomic symptoms, he first thought of odontalgia and consulted a dentist, who assured him of an unremarkable dental status. After repetition of similar attacks with photophobia and nausea coming into focus, he resorted to a triptan, which successfully relieved the pain. Since then seven similar cases have been collected from 2002 to 2006, underlining the heterotopic nature of migraine and its diverse clinical presentations.
There are previous reports in the literature of migraine patients with pain location in the lower two distributions of the trigeminal nerve (maxillary nerve = V2 and mandibular nerve = V3). Early studies from 1969 reported migraine cases with features comparable to those described here (6). Few case reports have focused on migraine presenting as odontalgia (3, 4), and most studies on migraine representation in the face have reported concomitant pain in the mouth and face distribution of the trigeminal nerve with additional typical unilateral headache (7). Benoliel et al. reported eight patients suffering from orofacial pain with vascular-type features similar to those in migraine without aura (8). Daudia et al. identified 24 patients with facial pain in the second distribution of the trigeminal nerve and concomitant migraineous features in a cohort of 973 consecutively recruited patients in a rhinology clinic (9). Penarrocha et al. have recently described 11 patients with facial pain and migraineous features, with nine patients responding to antimigraine treatment with ergotamine. They referred to this condition as ‘lower-half facial migraine’ (10). Unfortunately, this condition usually remains misdiagnosed and inappropriately treated as trigeminal neuralgia or ‘sinus headache’ due to rhinosinusitis. The Sinus, Allergy and Migraine Study reviewed 100 patients with physician- or self-diagnosed ‘sinus headache’ and found that up to 85% of those patients had migraine or probable migraine and 1.6% reported their pain confined to the second division of the trigeminal nerve (5). The importance of identifying these patients in clinical practice is obvious, as they respond very well to triptan medication and migraine-specific prophylactic treatment. Most of the cases documented here had been previously treated unsuccessfully for long periods of time with carbamazepine, gabapentin or indomethacin, because reported concomitant autonomic symptoms were not considered adequately. Although prompt identification of these patients will remain very difficult, trial triptan treatment should be considered in complicated facial pain patients with unilateral autonomic symptoms or prolonged periods of pain. Successful treatment with triptans might subsequently lead to implementation of migraine prophylaxis.
The clinical phenomenon of spread and referred pain has been intensively investigated, but the underlying pathophysiological mechanisms are not fully understood. The phenomenon of clinical referral, in which pain originates from an affected tissue but is perceived as coming from a distant receptive field, can be explained by a convergence mechanism of trigeminal (dura, skin) and cervical (muscle, joints, skin) afferents onto the same nociceptive second-order ascending projection neuron in the trigeminocervical complex in the brainstem (11–13) and receives input from many afferent sources, including facial and corneal receptive fields representing dural-sensitive neurons with high convergence potential (14–16). The direct coupling between meningeal afferents and cervical afferents in the cervical spinal dorsal horn has recently been shown electrophysiologically as evidence for the suspected anatomical overlap of trigeminal and cervical afferents throughout the trigeminocervical complex (17). This convergence of a visceral nerve (dura) with a somatic spinal nerve (greater occipital nerve) may explain the atypical localization of pain in our migraine patients, where the somatic part of the trigeminal nerve (V2 and V3) may also converge with its visceral part (V1) in the trigeminal nucleus caudalis, leading to a referral of pain with otherwise typical clinical symptoms of a migraine attack.
It has been hypothesized that the level of excitability in second-order neurons in the trigeminocervical complex could be increased by dysfunctional descending brainstem pain-modulatory antinociceptive structures such as the periaqueductal grey (PAG), nucleus raphe magnus and rostroventral medulla (18). Pharmacological blockage of P/Q-type calcium channels in the ventrolateral division of the PAG facilitates nociceptive input resulting in increased responses to dural stimulation and increased spontaneous activity of neurons in the trigeminocervical complex (19). These dysfunctional pain-modulating brainstem structures may provide an explanation for the periodic (triggerable) pain and may contribute substantially to central sensitization in patients with migraine. Malfunctioning sensorimotor integration, reflecting not purely sensory regulatory mechanisms, may also have modulatory effects on central sensitization in migraine (20).
In summary, we have presented seven migraine patients with atypical pain localization in the second and third division of the trigeminal nerve. These observations support the hypothesis of a convergence mechanism in the trigeminal nucleus caudalis associated with atypical pain referral, improving the current understanding of the pathophysiological mechanisms of trigeminal pain transmission and neural plasticity in general and the heterogenic nature and different clinical facets of migraine in particular.