Abstract
Dear Sir We read with interest the case reports by Gaul et al. (1) and Obermann et al. (2). As persistent supporters of the existence of a primary orofacial pain with neurovascular features (3–5), we were pleased to see these appear in the ‘headache’ literature. Before proceeding with our discussion on the subject, we point out that Oberman et al. (2) have mistakenly quoted our report as including eight cases of ‘facial migraine’. In our prospective study published in 1997, we collected a total of 55 patients with features that classified them as neurovascular pain, features including episodic, unilateral, throbbing pain that often woke patients from sleep (3). Many had local autonomic signs or complaints of nausea and photo/phonophobia. Applying the then current International Headache Society's criteria to the 55 cases, we established that 12 were cluster headaches, eight migraine without aura and six were paroxysmal hemicranias. We were then left with 29 cases that were ‘atypical’ largely due to the fact that the main focus of pain was intra- or perioral; these 29 cases were our study group for facial neurovascular pain. Two-thirds of these 29 cases also described radiation of pain to the cheek or ear. Similar case series of isolated facial or orofacial pain with neurovascular features have been reported by Daudia and Jones (6), who presented 24 patients with maxillary pain, the 11 cases by Penarrocha et al. (7) and the current case reports (1, 2). What clinical entity do these cases represent?
In tune with the authors' way of thinking, we suspect that these cases may indeed be based on migraine-like mechanisms (8). However, great care must be exercised when approaching classification; the use of ‘orofacial or facial migraine’ establishes an ostensibly clear pathophysiology and subsequently a therapeutic strategy. The lack of supportive evidence suggests that such classification may be premature. This is the reason we have preferred to use the term ‘vascular orofacial pain’, or, more recently, neurovascular orofacial pain (NVOP) (3, 9). This terminology allows for the ambiguity as to the underlying pathophysiology until further data accumulate. Indeed, patients with NVOP described in the literature may be atypical migraines, atypical trigeminal autonomic cephalgias (TACs) or a separate, distinct entity (3, 6, 7).
Obermann et al. (2) have suggested that the pain experienced by their patients probably arose from dural structures, but perceived in the orofacial tissues via referral mechanisms involving convergence. If we examine the clinical profile of NVOP vs. other neurovascular-type headaches, the results are inconclusive. In terms of location, NVOP fits the referral pattern of the TACs more than it does migraine. TAC with pain originating in the jaws or radiating to the teeth is a very common occurrence widely described in the literature, and has caused extensive misdiagnosis (10–14). Little has been investigated concerning intraoral referral or location of migraine headaches. Furthermore mild and less frequent migrainous symptoms such as nausea and photophobia are persistent features of NVOP (1, 3, 7). This is similar to some TACs that are accompanied by mild migraine-like symptoms (e.g. nausea) (15–17). The alternative explanation is that migraine mechanisms may be present at sites other than the dural cerebrovascular junction. We have extensively discussed the possibility that NVOP patients may be ‘facial migraines’, but have also presented anatomical and neurophysiological data that may support NVOP as a primary phenomenon occurring in orofacial structures (3, 9). Activation of such mechanisms may then induce a primary neurovascular pain located at orofacial tissues.
Demographically, most NVOP patients are female, as in migraine patients (1–3, 7). An interesting finding is that the reported age of onset in NVOP is relatively higher than would be expected in migraine: from 35 (7) to about 43 years old (2, 3). The significance of this finding is unclear; some of our cases have reported that in the past they suffered typically located and symptomatic migraines, and the appearance of NVOP may simply signify a change in location. Treatment response is a complex issue; some TACs (e.g. cluster headache) respond to abortive triptan therapy (18, 19), and divalproex has been successful in the prophylaxis of cluster headaches, particularly with migrainous features (17). It is true that most NVOP cases are in terms of attack duration more similar to migraines than TACs; however, long-lasting cluster headaches are well documented (14) and may be the case in these a priori atypical cases. Most cases of NVOP are indeed episodic (1, 3, 7), but we have recent unpublished data suggesting that some NVOP patients present with a chronic-type temporal pattern. These may be the equivalent of chronic migraine, or other chronic TACs.
The question remains whether NVOP is a separate diagnostic entity within the neurovascular headaches. Classifications develop as data gather; in the past cluster was thought a migraine variant, paroxysmal hemicrania a cluster variant and SUNCT an atypical form of trigeminal neuralgia. Multicentre prospective trials with careful data collection regarding clinical features and treatment response are desperately needed. In the absence of these and evidence of specific patterns of brain activation from imaging studies comparable to those performed in TACs and migraines, we are at present unable confidently to term these ‘orofacial migraine’.