By Dan Hurley
July 21, 2016
ARTICLE IN BRIEF
In new analysis, diary-validated migraine triggers were found to be unique to each patient in 85 percent of cases, providing a pretext for personalized migraine management, researchers said.
SAN DIEGO — Patients' beliefs about the triggers that precede their headaches are often not supported by the results of their diary entries, according to new research that was presented here in June at the annual meeting of the American Headache Society.
By far the largest proportion of patients were found to have individual triggers, which could not be identified in the overall analysis of all patients, the researchers said. Among their findings, the constellation of diary-validated triggers was unique to each patient in 85 percent of cases.
The new analysis provides information about the correlation between migraine attacks and a broad spectrum of possible trigger factors for each individual patient, and is therefore a step towards personalized migraine management,” said lead study author Christian Wöber, MD, associate professor head of the headache group in the department of neurology at the Medical University of Vienna, Austria.
DR. ANDREW CHARLES said he was uneasy with interpreting the symptoms preceding migraine as “triggers,” when in fact the brain changes associated with the premonitory phase of a migraine might make it appear that whatever they were doing before the attack “caused” what was already underway.
DR. CHRISTIAN WöBER: “We found more factors important for individual patients than we previously found in the population analysis.”
The findings were based on an innovative new analysis of data previously published in a 2007 study and updated in May in Cephalagia.
For that earlier study, Dr. Wöber and colleagues recruited via newspapers 327 migraineurs who kept a comprehensive paper-based diary for three months. That study confirmed many of the usual triggers cited by patients, including menstruation, tiredness, mental tension, and noise.
STUDY METHODOLOGY
For the new analysis, rather than look for triggers that were significant for the entire group on average, Dr. Wöber applied a novel statistical analysis for 327 patients from the earlier study, examining which of 33 potential triggers or premonitory symptoms preceded migraines in each individual patient.
A factor was considered as a potential trigger if the univariate hazard ratio was greater than one and the p-value was equal to or less than 0.05. Individual “potential trigger” profiles were then generated for each patient, including whichever factors were significantly associated with the occurrence of migraine.
Individual profiles were generated for all but 13 percent of patients, for whom too few attacks occurred in the 90-day period on which to base a profile. The average number of factors associated with triggers was four per patient (95% CI: 3.6-4.2; range 0 to 13).
“We found more factors important for individual patients than we previously found in the population analysis,” Dr. Wöber said, adding that the research involved a collaborative effort with scientists at Curelator Inc., the provider of the digital application and analytical engine; a biostatistical group in Barcelona confirmed the findings.
“I have offered the diary to several of my patients,” said Dr. Wöber, who is on the company's advisory board. “Some are very enthusiastic, and some say ‘I've had migraine for 20 years, I know everything, I don't think it's useful for me.’ It's an important approach, but it requires a patient to keep the diary for three months, doing it every night for two or three minutes.”
Another paper, conducted solely by scientists on the staff of Curelator, examined data on 254 individuals who used the app for at least 90 days to determine whether factors the individuals believed to be triggers were confirmed by the digital diary entries. Out of 3,278 suspected triggers named by the participants, only 592 (18 percent) were statistically associated with migraines.
“Patients cannot reliably assess their own trigger factors,” said Dr. Wöber, who reached similar findings in another study he previously published using paper-based diaries. The only way to find it out is if patients keep a prospective diary.”
EXPERT COMMENTARY
Andrew Charles, MD, professor of neurology and director of the University of California Los Angeles Goldberg Migraine Program at the David Geffen School of Medicine, said he was not surprised that Dr. Wöber's study found individual factors unique to each patient.
“I certainly think there's a heterogeneity of mechanisms and symptoms that are quite broad, both in patient to patient, and from attack to attack,” Dr. Charles, who was not involved with the studies, said.
He emphasized, however, that he was uneasy with interpreting the symptoms preceding migraine as “triggers,” when in fact the brain changes associated with the premonitory phase of a migraine might make it appear that whatever they were doing before the attack “caused” what was already underway.
“Perhaps, as a premonitory symptom, a person feels worse and wants to assuage that feeling by eating chocolate,” Dr. Charles said. “So they eat chocolate, and when the full migraine attack follows, they now interpret the chocolate as the ‘trigger.’”
Yet he acknowledged that many migraine patients are obsessed with identifying and avoiding what they consider to be triggers.
“That's one of the primary things patients are looking for: guidance on how to lead their lives to help with their headaches,” Dr. Charles said. “I don't want to suggest that people shouldn't avoid the things that they believe to be triggers, as long as it's in reason. To the extent that it helps them and it doesn't interfere with their quality of life, I have no problem with that.”
