Abstract
Background
Migraine with unilateral cranial autonomic symptoms (UAS) is a putative migraine endophenotype with convincing response to trigeminal-targeted treatments that still needs a thorough characterization.
Objective
The objective of this article is to carefully investigate the clinical phenotype of migraine with UAS in a large group of patients for more accurate migraine diagnoses, improved clinical management, and better outcome prediction.
Methods
We studied 757 consecutive episodic and chronic migraineurs in a tertiary headache clinic with face-to-face interviews, detailing in depth their lifestyle, sociodemographic and headache characteristics.
Results
Migraineurs with UAS (37.4%) differed from the general migraine population with respect to longer attack duration (OR = 2.47, p < 0.02, having >72-hour long attacks), more strictly unilateral (OR = 3.18, p < 0.001) and severe headache (OR = 1.72, p = 0.011), more frequent allodynia (OR = 3.03, p < 0.001) and photophobia (OR = 1.87, p = 0.019).
Conclusions
Migraine patients with UAS are characterized not only by symptoms due to intense peripheral trigeminal activation but also to central sensitization. Our study broadens the knowledge on the clinical and phenotypic characteristics of migraine with UAS, suggests pathophysiological implications, and supports the need for future prospective clinical studies.
Introduction
Unilateral cranial autonomic symptoms (UAS), i.e. lacrimation, conjunctival injection, eye redness, eyelid edema, ptosis, nasal congestion/rhinorrhea, forehead/facial sweating, ear fullness, the hallmark of trigeminal autonomic cephalgias, may also be present during typical migraine attacks. Following the first anecdotal report by Blau and Davis in 1970 (1), we described the occurrence of UAS in 45.8% of migraine outpatients in a headache center (2). Afterwards, in a population-based study, Obermann et al. reported a prevalence of 26.9% among migraineurs (3). These studies, designed to assess UAS prevalence in a migraine population, also suggested that migraine patients with UAS (UAS+) could be characterized by a more strictly unilateral and more severe pain than those without (UAS–) (2).
The use of unambiguous criteria for the identification of migraine with UAS is critical to avoid misdiagnosis (e.g. cluster headache, sinus headache) (4) and may have therapeutic implications (5). An open study in migraineurs treated with sumatriptan 50 mg (6), subsequently confirmed by a randomized, double-blind, placebo-controlled parallel-group trial using rizatriptan 10 mg (7), suggested that UAS+ are characterized by a very good response to triptans. This clinical hypothesis is supported by the biochemical evidence that in UAS+, but not in UAS–, increased levels of calcitonin gene-related peptide (CGRP) and neurokinin A (NKA) (markers of trigeminal activation) and detectable vasoactive intestinal peptide (VIP) levels (markers of parasympathetic activation) during the attack are reversed by rizatriptan administration (8). High triptan responsiveness in UAS+ is probably due to an increased trigeminal activation, responsible for a large-scale recruitment of peripheral neurovascular 5-HT1B/1D receptors, which leads to the excitation also of the efferent parasympathetic arm of the trigemino-autonomic reflex (5–8). The recent applications of multidimensional approaches in patients affected by chronic diseases has demonstrated that complex clinical conditions, such as migraine, may be better interpreted taking into consideration the full set of parameters that may influence disease development and progression (9). The present study was specifically designed to investigate the full clinical phenotype of UAS+ in a large population of migraine outpatients, detailing in depth their lifestyle, sociodemographic and clinical characteristics. The availability of such a comprehensive database will help to shed light on this migraine endophenotype, eventually leading to more accurate diagnoses, improved clinical management, and better outcome prediction.
Methods
All patients affected by episodic migraine (with or without aura) (10) and chronic migraine (with or without medication overuse) (11) consecutively seen at the Headache and Pain Unit of the IRCCS San Raffaele Pisana Hospital in Rome from July 1, 2012 to December 31, 2012 were investigated.
Following a careful physical and neurological examination, specifically trained neurologists (PB, CA, GE, LF) screened all patients with face-to face interviews using a semi-structured questionnaire addressing three main issues: 1) lifestyle, behavioral and sociodemographic factors, including age, sex, civil status, occupation, body mass index (BMI), blood pressure, sports activity, use of coffee, alcohol, smoking, illicit drugs, sleep disturbances, menopause, contraceptive use; 2) comorbidities (including head or neck trauma) and concomitant medications; and 3) clinical migraine features including family history, disease duration, site, quality and intensity of pain, attack duration and frequency, presence, type and duration of aura, accompanying symptoms, dopaminergic symptoms (yawning, nausea, vomiting), triggers and alleviating factors, prodromes, postdromes, allodynia, UAS and their characteristics, patients’ satisfaction with triptans, previous and current acute or preventive treatments. The presence of UAS was determined by asking the following question: “During the migraine attack do you also have at least one of the following symptoms: unilateral conjunctival injection, lacrimation, nasal congestion/rhinorrhea, ptosis, eyelid swelling or forehead/facial sweating?” Patients reporting at least one of the above symptoms during the attack were considered UAS+. Patients were asked to rate their satisfaction with triptans as “dissatisfied,” “moderately satisfied” or “satisfied/very satisfied.” Allodynia was assessed using the Allodynia Symptom Checklist (12).
The protocol was approved by the institutional review board at IRCCS San Raffaele Pisana and was developed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki. All participants gave their informed consent prior to their inclusion in the study.
Statistical methods
Sample characteristics were resumed by means of summary statistics (n, %, means, standard deviations). Individuals with and without UAS were compared on sociodemographic and lifestyle factors, comorbidities and clinical migraine features using the unpaired t test for continuous variables (age, BMI, cigarettes/day, attack frequency, and attack duration) after checking assumptions, and the Pearson’s chi-square test for categorical variables.
A multiple logistic regression analysis was performed to identify those characteristics associated with the UAS+ phenotype taking into account the influence of actual confounders. Results are expressed as odds ratios (ORs), relative 95% confidence intervals (CIs), and p values. A p value <0.05 was the threshold for statistical significance. STATA/SE V12 was used for all statistical analyses.
Results
Symptoms prevalence in migraineurs with unilateral cranial autonomic symptoms (UAS) (n = 283).
Lifestyle, behavioral and sociodemographic factors
UAS+ were older (three years on average, p < 0.001), more frequently married/living common-law (61% vs 50%, p = 0.046), had more frequent sleep disturbances (19% vs 13%, p = 0.029), used fewer antihypertensive drugs as concomitant medications (3% vs 7%, p = 0.025), and had a longer illness duration compared to UAS– (20.8 vs 18.2 years, p = 0.009).
Headache characteristics
UAS+ reported more frequently severe head pain (83% vs 72%, p < 0.005), stabbing pain (6% vs 3%, p = 0.044), strictly unilateral (57% vs 35%) and less bilateral pain (19% vs 37%) (p < 0.001), prolonged head pain (attack duration >24 hours: 72% vs 56%, p = 0.001) and more frequent prodromes (17% vs 11%, p = 0.042), photophobia (91% vs 81%, p < 0.001), phonophobia (83% vs 73%, p = 0.003) and allodynia (57% vs 29%, p < 0.001).
Acute and preventive treatments
Sociodemographic and clinical characteristics of migraine patients with or without unilateral cranial autonomic symptoms (UAS). All values are reported as frequencies n (%) or mean ± SD.
BMI: body mass index; NSAIDs: nonsteroidal anti-inflammatory drugs.
Multiple logistic regression analysis resulted in a model that selected pain location, allodynia, photophobia, attack duration and pain intensity as the clinical features that significantly characterized migraine with UAS after adjusting for confounding. In particular, it was estimated that the presence of a strictly unilateral headache was three times more likely to be present in UAS+ than bilateral pain (OR = 3.18, p < 0.001). Pain intensity was also associated with UAS (when compared to mild/moderate, migraineurs with severe headache suffered more frequently from UAS (OR: 1.72, p = 0.011). A dose-related association with attack duration was observed, i.e. when compared to a ≤12-hour duration, participants with a 48- to 72-hour attack showed an OR = 1.87 of showing UAS (p = 0.048), while patients with the longest attacks (≥72 hours) reported 2.47 times more frequently the presence of UAS (p = 0.018).
Multivariable logistic regression analysis on the clinical features most closely associated to the presence of unilateral cranial autonomic symptoms (UAS) in a sample of migraine patients (n = 283).
Other variables, including osmophobia, phonophobia, sex, age, illness duration, and pain quality (pulsating, pressing, continuous, stabbing), were fitted in the multivariable model; however, not enough evidence emerged to establish a statistical association with the presence of UAS.
Discussion
Migraine is a very common disorder (13) characterized by a substantial phenotypic heterogeneity. The extent of this variability requires a meticulous evaluation of migraine features in an attempt to identify homogeneous subgroups of patients with the aim of improving clinical management, tailored treatments, and outcome prediction (5). We suggest that the presence of UAS during the attack could identify one such migraine endophenotype.
Emerging evidence over the last decade indicates that UAS+ are frequent both in headache centers (45.8%) (2) and, at a lesser extent, in the general population (26.9%) (3). These patients are characterized by a convincing response to triptans (6–8) and may be characterized by specific genetic polymorphisms (14).
The main results of our study, specifically designed to highlight the clinical phenotype of UAS+ in a large sample of migraine outpatients, may be summarized in three new substantial categories of clinical information: Firstly, the frequency of UAS+ among migraineurs referred to headache centers is lower (37.4%) than previously described (2) and closer to that of the general migraine population (3); secondly, UAS are associated with longer attack duration in a dose-related manner; thirdly, the probability of showing UAS during migraine is higher in the presence of allodynia and photophobia. Furthermore, we confirm in a large case series that pain in UAS+ is more frequently strictly unilateral and severe than in UAS–.
How could these clinical findings be translated into pathophysiological terms? UAS express activation on the trigemino-autonomic reflex (15), a defensive response physiologically aimed to protect the eye and nose through lacrimation and rhinorrhea following harmful stimuli (16,17).
Pain characteristics in UAS+ are in keeping with a very pronounced sensitization of peripheral trigeminovascular neurons during attack. The strictly unilateral (anterior) pain topography in UAS + traces the activation of the ophthalmic branch of the trigeminal nerve, a key pathophysiological migraine event that is not always clinically evident in migraineurs because, in spite of its etymology, pain is bilateral or median in one-third of patients (18). Similarly, severe intensity and long pain duration in UAS+, which significantly outlast those of the general migraine population (19), is conceivably the clinical consequence of such an intense trigeminal activation. Thus, it is likely that UAS+ differ from UAS– for a more marked and prolonged trigeminal peripheral sensitization that reaches the threshold for triggering the trigemino-autonomic reflex. Once activated, the parasympathetic outflow, in turn, further activates the perivascular sensory trigeminal fibers, amplifying and perpetuating peripheral pain mechanisms in a vicious circle via the release of acetylcholine, VIP and nitric oxide, which induce mast cell degranulation, intracranial vasodilation, protein plasma extravasation and release of inflammatory molecules (20–30).
We hypothesize that such an intense peripheral input from highly sensitized trigeminal neurons in UAS+ would impinge more heavily on brainstem second-order neurons in the trigeminal nucleus than in UAS−, giving rise more frequently to a central sensitization, clinically expressed by cutaneous allodynia and photophobia (31). Cephalic cutaneous allodynia is known to reflect the sensitization of second-order neurons in the trigeminal nucleus caudalis receiving convergent inputs from intracranial (blood vessels and meninges) and extracranial structures (facial skin and scalp) (32,33). Photophobia is believed to be due to light signals converging from the retina onto sensitized thalamic neurons in the pulvinar projecting to the visual cortex (34). It’s worth noting that in our study UAS+ had a longer headache history than UAS−: This in accordance with the notion that allodynia increases in parallel with illness duration (31).
It could be argued that the significant occurrence of cephalic allodynia in UAS+ is in contrast with their preferential use and greater satisfaction with triptans. However, bearing in mind that our study was not designed to assess triptan responsiveness, we speculate that this could be due, at least in part, to the fact that UAS+ have a higher prevalence of prodromes (originating from the hypothalamus, which interestingly is connected to the parasympathetic superior salivatory nucleus, involved in UAS) (35), which could allow patients an early treatment of the attack, a recognized effective acute migraine therapeutic strategy that circumvents the occurrence of central sensitization (36,37).
Among the strengths of the study—the first specifically designed to achieve a full clinical characterization of UAS+—there is the large size of the study group, represented by consecutive unselected patients. In addition, they were directly interviewed using a detailed semistructured questionnaire that integrated accurate information from different boundaries such as lifestyle, behavioral and sociodemographic factors, comorbidities and clinical migraine features. Limitations are the fact that we did not assess ear fullness, now listed among UAS in the new International Classification of Headache Disorders (10)—being that our study was carried out before its publication—and recruited patients in a tertiary-referral center. Nor can we rule out a potential recall bias given the study design.
In conclusion, the present study broadens our previous knowledge on the UAS+ phenotype, demonstrating that these patients are characterized not only by symptoms related to an intense peripheral trigeminal sensitization—i.e. strictly unilateral, severe and long-lasting pain—but also to central sensitization—namely allodynia and photophobia. The activation of the cranial parasympathetic system (an emerging target for headache treatment (38,39), is likely to contribute to inducing, amplifying and maintaining trigeminal pain in UAS+, through a circuitry sensitization occurring both peripherally and centrally (31).
Our clinical findings, coupled with recent biochemical, therapeutic, and genetic achievements (5–8,14), outline UAS+ as a characteristic, recognizable migraine endophenotype and support the need for future prospective clinical studies.
Clinical implications
Migraine with unilateral cranial autonomic symptoms (UAS) is characterized by longer attack duration, more strictly unilateral and severe headache and more frequent allodynia and photophobia than migraine without. The clinical phenotype of migraine with UAS reveals not only an intense peripheral trigeminal activation, as suggested by pain features, but also by a central sensitization, as indicated by its accompanying symptoms. Migraine with UAS is an easily recognizable putative migraine endophenotype, characterized by a convincing response to trigeminal-targeted treatments.
Footnotes
Declaration of conflicting interests
The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Dr Piero Barbanti has served as a consultant or scientific advisor and has received travel grants or research support and honoraria or both for lecturing from Merck, Janssen Pharmaceuticals, Lusofarmaco, Abbott, Allergan, Allmiral and ElectroCore. CA, GE and LF have received travel grants from Lusofarmaco. Dr Stefano Bonassi has served as consultant and/or scientific advisor for Novartis and Indena. Valentina Dall'Armi is currently working at Quintiles Ltd, Real-World and Late Phase Research, UK. No conflict of interest was present at the time of manuscript preparation.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by the Italian Ministry of Health for Institutional Research. The work of SB was supported by grants funded by the Associazione Italiana per la Ricerca sul Cancro (AIRC) (IG 2015 Id.17564).