SAN FRANCISCO – Half of migraineurs suspect chocolate can trigger their migraine attacks, but new evidence from a large prospective study suggests almost 99% of them are mistaken.
Additional analyses of this same dataset of 774 individuals with migraine threw cold water on two other widely accepted putative migraine triggers: neck pain/tension and dietary nitrate intake, Stephen Donoghue, PhD, reported at the annual meeting of the American Headache Society.
He presented a series of statistical analyses of often-cited migraine attack triggers conducted in 774 migraineurs who registered to use, a sophisticated proprietary digital headache diary. At the outset, participants rated on a 0-10 scale how strongly they suspected chocolate, neck pain or tension, and other factors acted as triggers for migraine attacks in their own personal experience. They then spent 2-3 minutes daily tracking more than 70 migraine-related elements for at least 90 days, recording the data using an iPhone or iPad rather than using a traditional, more hit-or-miss conventional paper diary.
The N1–Headache software, developed by Curelator of Cambridge, Mass., then tackled the key issue of what proportion of individuals who suspected a given trigger actually showed a statistically significant association prospectively between an individual’s day-to-day variations in experiencing that trigger and their headache risk, explained Dr. Donoghue, vice president for clinical development for Curelator.
The Curelator concept involves bringing personalized medicine to headache patients by identifying their true migraine attack triggers, which can enable individuals to deal with those triggers without the disruption involved in unnecessarily avoiding numerous nontriggers or missing real triggers.
The study results punctured some widely held beliefs. For example, at baseline 51% of migraineurs indicated they suspected chocolate served to some degree as a trigger for their own migraine attacks; the majority of them rated their suspicion as moderate or strong. However, in individually determined correlations, chocolate was indeed associated with migraine attacks in only 1.3% of those who suspected it to be a trigger. Moreover, in an another 3.9% of chocolate suspecters, chocolate consumption was actually associated with decreased risk; in other words, for them, chocolate appeared to serve as a protector against migraine, despite their preconceptions.
Also, among the 49% of participants who didn’t consider chocolate to be a personal migraine attack trigger, the Curelator analysis demonstrated that chocolate consumption was associated with a significantly increased risk of migraine attack in 2.2%, and a significantly lower risk of migraine in another 1.5%.
Eighty percent of migraineurs who registered to use N1–Headache via the company’s website or theindicated they believed neck pain/tension to be a migraine trigger for them; 46% rated it as a strong trigger. The detailed analysis of 90 days’ worth of data showed that 32% of participants with adequate data showed a statistically significant association between neck pain/tension and migraine headache.
The strength of an individual’s suspicion of neck paint/tension as a trigger was associated with the frequency of a statistically confirmed association. However, unlike for chocolate, there were zero instances where neck pain/tension was associated with protection against migraine.
In a twist, when the investigators reanalyzed their data after eliminating those instances where neck pain/tension occurred 1 day before the start of the headache, the association disappeared.
“The temporal association that we find using the lag-day analysis strongly suggests that neck pain/tension is part of the symptomatology of migraine attacks rather than acting as a trigger,” Dr. Donoghue observed.
This concept is strongly supported by a recent German study that used neck muscle electromyography to establish the same point (), he added.
Dr. Donoghue offered one caveat regarding the Curelator analyses: “We can’t say whether or not neck pain/tension and these other factors are actually triggers. What we’re looking at is associations. We’re not showing causality.”