|
|
Emily B Schroeder,
Kathryn M Rose, Duanping Liao, Gerardo Heiss University of North Carolina
at Chapel Hill
There has been extensive debate about the possible role of autonomic
nervous system dysfunction in migraine. Cardiovascular tests, vasomotor
reactions to temperature changes, responses to pharmacological tests,
and changes in biochemical parameters have suggested hypo- and hyperfunctioning
of both the sympathetic and parasympathetic nervous system. Despite
numerous small case-control studies, these contradictory results have
not been resolved. We investigated the putative association of parasympathetic
dysfunction with migraine in the Heart Rate Variability (HRV) ancillary
study of the Atherosclerosis Risk in Communities (ARIC) study, a population-based
cohort of 15,792 men and women aged 45-64 at baseline. We analyzed
high frequency power (HF), a marker of parasympathetic tone; low frequency
power (LF), a marker of parasympathetic and sympathetic tone; and
standard deviation of all R-to-R intervals (SDNN), a marker of overall
heart rate variability. HRV was measured at Visit 1 (1987-1989), and
migraine status was assessed at Visit 3 (1993-1995) using a modified
version of the International Headache Society criteria. Associations
were examined using unconditional logistic regression, with all models
adjusted for age, study center, and resting mean heart rate. The HRV
measures were specified in four manners: continuously, log-transformed,
in quartiles, or dichotomously (the lowest quartile compared to the
upper three quartiles). There were large differences in migraine prevalence
by race and gender (14% in white women, 4% in white men, 5% in black
women, and 1% in black men). Analyses were stratified to account for
these differences and the association of HRV with both race and gender.
The small number of black males with migraine prevented analysis of
that subgroup. No consistent associations between HRV and migraine
were found. White women with migraine were slightly more likely to
be in the lowest quartile of SDNN compared to the highest quartile
(OR 1.30, 95% CI 1.00-1.71). White men with migraine were less likely
to be in the lowest quartile of LF compared to the rest (OR 0.48,
95% CI 0.28 - 0.81). No other statistically significant differences
in HRV by migraine status were detected. We conclude that the ARIC
data does not provide support for an association between migraine
and autonomic nervous system dysfunction.
View Slides |
