Increased risk of Bell palsy in patients

with migraine
A nationwide cohort study
ABSTRACT
Objective: To evaluate the association between migraine and Bell palsy and to examine
the effects of age, sex, migraine subtype, and comorbid risk factors for Bell palsy.
Methods: This nationwide cohort study was conducted using data from the Taiwan
National Health Insurance Research Database. Subjects aged 18 years or older with
neurologist- diagnosed migraine from 2005 to 2009 were included. A nonheadache ageand propensity scorematched control cohort was selected for comparison. All subjects
were followed until the end of 2010, death, or the occurrence of a Bell palsy event. Cox
proportional hazards regres- sion was used to calculate the adjusted hazard ratios and
95% confidence intervals to compare the risk of Bell palsy between groups.
Results: Both cohorts (n 5 136,704 each) were followed for a mean of 3.2 years. During
the follow-up period, 671 patients (424,372 person-years) in the migraine cohort and 365
matched control subjects (438,677 person-years) were newly diagnosed with Bell palsy
(incidence rates, 158.1 and 83.2/100,000 person-years, respectively). The adjusted hazard
ratio for Bell palsy was 1.91 (95% confidence interval, 1.682.17; p , 0.001). The
association between migraine and Bell palsy remained significant in sensitivity analyses,
and tests of interaction failed to reach significance in all subgroup analyses.
Conclusion: Migraine is a previously unidentified risk factor for Bell palsy. The
association between these 2 conditions suggests a linked disease mechanism, which is
worthy of further exploration.

Bell palsy is an acute, idiopathic, unilateral peripheral facial palsy, although

bilateral involvement has been reported rarely.1,2 This condition affects patients of all
ages; the incidence is the lowest among children and increases gradually with age. 3 The
annual incidence rate ranges from 11 to 40 per 100,000 persons,4 and both sexes are
affected equally.5 The pathophysiology of Bell palsy is largely unknown. Common
hypotheses include acute or reactivation of latent viral infection, vascular ischemia, and
inflammatory responses.69 Several predisposing factors, such as diabetes mellitus,
hypertension, pregnancy, and the use of inactivated intranasal influenza vaccine, 3 have

been recognized. Other conditions can also present with unilateral facial palsy, such as
Ramsay Hunt syndrome, HIV infection, Lyme disease, autoimmune diseases,
amyloidosis, and space- occupying lesions.3
Migraine is a common and disabling disorder with an annual global prevalence of
approxi- mately 10% and a female/male ratio of 23:1.10,11 It is associated with vascular
disorders of the CNS, i.e., ischemic and hemorrhagic strokes. 12 Cranial nerve symptoms,
such as ophthalmoplegia, have also been reported in some patients. 13 We recently
reported an asso- ciation between migraine and sudden sensori- neural hearing loss.14 An
association with isolated peripheral facial palsy has been re- ported rarely, 15 although
facial and limb weak- ness is common in patients with hemiplegic migraine. 16 Vascular
ischemia remains a possi- ble etiology of Bell palsy, 9 and a recent report identified
asymmetrical facial blood perfusion in patients with migraine. 17 Thus, this study was
conducted to determine whether migraine is associated with Bell palsy.
DISCUSSION The results of this study indicated that the risk of subsequent
development of Bell palsy is approximately doubled in patients with migraine compared
with age- and propensity scorematched con- trol subjects. This association was not
affected by sex, migraine subtype, or other risk factors for Bell palsy. To our knowledge,
this epidemiologic study found a novel association between migraine and Bell palsy.
The annual incidence rate of Bell palsy in our con- trol cohort (83.2/100,000
person-years) was higher than previously reported (13.153.3/100,000 person- years). 5,25
27
This discrepancy might be accounted for by the accessibility and global coverage of the
NHI in Taiwan; an average of 15.6 clinical visits per person were recorded in 2010. 28 In
line with the results of previous studies,5,26,27 the occurrence of Bell palsy showed no
obvious sex-based predilection in our cohort. Furthermore, the finding of higher adjusted
absolute event rates of Bell palsy in patients with hypertension or diabetes mellitus,
compared with those without these conditions, is also in agreement with the
documentation of these conditions as risk factors for Bell palsy.29,30

(INDRA)
Several mechanisms may underlie the association between migraine and Bell
palsy. Neuritis or demyeli- nation after viral infection/reactivation remains the most
frequently accepted hypothesized pathogenesis of Bell palsy.69 During migraine attacks,

trigeminovas- cular system activation triggers regional neuropeptide release. 31 Recurrent

migraine attacks result in neuro- genic inflammation of nearby cranial nerves, 32,33 and
may predispose the facial nerve to demyelination after viral infection. In addition, Bell
palsy can be attributed to ischemia in some cases, which is reflected in its increased
incidence in elderly subjects and those with vascular risk factors, such as hypertension
and diabe- tes.3 Migraine is associated with various vascular dis- orders34 and
asymmetrical facial blood perfusion,17 which raises the suspicion of ischemic
mononeuropa- thy of the facial nerve. Of note, migraine with aura is associated with
hypertension, unfavorable cholesterol profiles, and higher Framingham risk scores 35;
how- ever, the HRs remained similar in all migraine sub- groups in the current study.
Thus, migraine per se, rather than certain comorbidities, might be associated with
subsequent development of Bell palsy. In addition, migraine patients with more clinical
visits for migraine were more likely to develop Bell palsy. A possible explanation is that
higher disease activities might link to the development of Bell palsy. Nonethe- less, a
higher clinical awareness of Bell palsy among patients or physicians, or drug-related
effects in this subset of patients remains possible. Overall, these ex- planations remain
provisional, and further studies exploring this association are needed.
Our findings have several pathologic and clinical implications. First, migraine is
an unidentified risk fac- tor for Bell palsy. However, whether migraine per se or the
frequency of migraine attacks contributes more to the development of Bell palsy remains
uncertain. Thus, the exploration of migraine history in patients receiving treatment for
Bell palsy could be a first step of future investigation. In addition, the link between
migraine and Bell palsy suggests that a common mechanism underlies both diseases.
Patients with diabetes tend to have worse functional recovery after Bell palsy than
nondiabetic patients,36 but whether the outcome of Bell palsy differs in patients with
migraine warrants further study. At present, the best treatment regimen is early steroid 37;
the optimal treatment strategy in this subgroup of patients requires further investigation.
The major strengths of this study include the large sample, given that the migraine
cohort was taken from a nationwide population-based database. In addition, we enrolled
only patients with neurologist-diagnosed migraine to increase the validity of migraine
diagnoses. A previous report confirmed that neurologists diagnoses of migraine
according to the International Classification of Headache Disorders, 2nd Edition, criteria
in Taiwan were 91.6% accurate.38 Third, an age- and propensity.scorematched control
cohort was used to minimize selection bias and any imbalance in medical care seeking
behavior. Fourth, patients diagnosed with her- pes zoster infection and Bell palsy within a
30-day period were excluded from our cohort because Ramsay Hunt syndrome is a major
consideration in the differ- ential diagnosis of Bell palsy.
Several limitations of this study should also be considered. First, because the
diagnosis of Bell palsy is primarily clinical and limited clinical information could be
obtained from the database, the proper exclusion of other less-common causes of
acquired facial weakness could not be assured, which may have led to overestimation of

the occurrence of Bell palsy in both cohorts. Second, although most patients with Bell
palsy achieve complete recovery,39 the NHIRD contains neither the prognostic
information nor the duration and frequency of migraine, preventing the exploration of
potential differences in the clinical course of Bell palsy in patients with migraine. Third,
all patients enrolled in the migraine cohort had active migraine, leading to the
underrepresentation of sub- jects with previous or nonactive migraine. Fourth, in the
current study, we used a neurologist-diagnosed migraine cohort. Thus, this cohort might
be subject to Berkson bias, i.e., a migraine patient with neuro- logic consultation might be
more likely to be diag- nosed with Bell palsy; however, the HRs of Bell palsy determined
by neurologists and nonneurologists were similar (table 3). Finally, the control cohort
may have included patients with migraine who did not seek medical assistance,
potentially leading to underesti- mation of the risk of Bell palsy in the migraine cohort.
Migraine is a previously unidentified risk factor for Bell palsy with at least equal
importance as hypertension and diabetes. The association between migraine and Bell
palsy suggests that a common mechanism underlies these diseases, which is worthy of
further exploration.