What does migraine aura look like? And how do these visual manifestations differ between patients and in the same patient?
These questions have intrigued neurologists for years, but getting clear-cut information has been difficult because most reports of migraine characteristics are retrospective — when asked, patients report details of their attack and how they felt after the fact.
But now, researchers are delving deeper into the issue by using prospectively gathered reports and are turning up some surprising findings. For example, only about half of patients with migraine actually experience nausea, which is a main diagnostic symptom of migraine.
This could affect accurate diagnosis of migraine during clinical trials, according to Jakob Møller Hansen, MD, PhD, research fellow, Danish Headache Center, Department of Neurology, University of Copenhagen, Denmark.
Dr Hansen shared some of his research findings with colleagues attending the Congress of the European Academy of Neurology (EAN) 2016.
Unlike many other conditions, migraine is diagnosed according to what the patient tells a clinician, said Dr Hansen. “We don’t have any biomarkers, and you can’t scan people like you can in stroke.”
After listening to a patient’s story, the clinician determines whether that patient fits the “very strict and very valid” criteria laid out in the International Classification of Headache Disorders.
About a third of patients with migraine experience attacks with aura. Although these auras are similar in some respects, they are, for the most part, “highly variable,” said Dr Hansen.
Large Prospective Trial
To learn more about this variability, Dr Hansen and his colleges accessed data from a large prospective clinical trial. Their aim was to describe the clinical features of migraine aura in this large group and to compare retrospective (self-reported) migraine symptoms with prospective recordings of attacks in the same patients.
The trial included 267 patients with migraine with typical aura, including visual symptoms, from 16 centers across the United States who were enrolled in a clinical trial (to assess the effects of transcranial magnetic stimulation).
Upon entering the trial, patients described their migraine attack symptoms in detail. They often had more than one visual aura symptom; the median number was two, but the number ranged from one to five.
Then, during the prospective phase of the trial, when a migraine with aura occurred, patients had an hour to record details of the attack.
“Whenever they experienced aura symptoms, they entered a description in an electronic device, so it was in real time, which adds to the strength of the study,” said Dr Hansen.
From the 267 patient reports, the researchers collected data on 861 migraine attacks.
The researchers found that the four most prevalent visual aura symptoms were:
- Dots or flashing lights (recorded by 70% of patients);
- Wavy or jagged lines (recorded by 47% of patients);
- Blind spots (recorded by 42% of patients); and
- Tunnel vision (recorded by 27% of patients).
Many of these symptoms overlapped, as described in a diagram shown by Dr Hansen.
“From this graph, it’s clear that these symptoms rarely occur in isolation; rather, they seem to come in all kinds of different combinations of these symptoms, adding to this notion that migraine aura is very heterogeneous, very variable,” he said.
Study patients were also asked about nonvisual aura symptoms. About half (52%) reported these symptoms.
Here again, many patients had more than one symptom; the median was one, but the range was one to four.
Among the most prevalent nonvisual aura symptoms were:
- Numbness or pins and needles (recorded by 29.5% of patients);
- Difficulty recalling or speaking (26%);
- Changes in smell (olfaction) (19%); and
- Changes in taste or touch (14%).
“When we look at this data, it’s clear that these symptoms also rarely show up in isolation, but rather come in these clusters and in different combinations,” said Dr Hansen.
Dr Hansen noted that most of the prospectively recorded migraine attacks occurred during the “wake hours.” A graph illustrated that the number of attacks peaked at about 10 to 11 am, but there was another smaller peak between 7 and 8 pm.
“This is different from what you see, for instance, in cluster headaches, where people wake up at night with an attack,” commented Dr Hansen.
He pointed out what he considered some surprising findings when they compared self-reported information from prospectively collected data. One was that 95.9% of patients included nausea as a symptom in their self-reports, but in the prospective data, almost half of the attacks (49%) had no nausea associated with them.
Dr Hansen also noted that the severity of nausea seemed to be reduced with advancing age.
Also, while almost all patients self-reported phonophobia and photophobia, the incidence of these characteristics was lower in the prospective study. In 16% of attacks occurring in the prospective phase, photophobia occurred without phonophobia.
“Nausea as well as photo- and phonophobia are part of the diagnostic criteria — they are something you have to have in order to have a migraine,” said Dr Hansen. “So I think one of the most interesting things we found looking at these data is that these defining features of migraine actually in many cases did not occur during these prospectively recorded attacks.”
This might make diagnosing migraine with aura difficult In a clinical trial setting where each attack has to be diagnosed separately, he said. If patients don’t have nausea or phonophobia and photophobia at the same time, patients wouldn’t fulfill the diagnostic criteria.
“That is actually a source of confounding for clinical studies and is something to be aware of,” said Dr Hansen.
That the features of migraine with aura are “highly variable” might be an important factor to consider in clinical trials “where you have to diagnose whether this is a migraine attack or not a migraine attack,” he added.
Results of the study might help guide patient care, according to Dr Hansen. For example, he said, “if it turns out that most attacks are not associated with nausea, antiemetics shouldn’t really be at the center of your treatment.”
Among the strengths of the study was use of an electronic device to capture prospective data. Once entered, it could not be changed. “I think that adds to the internal validity of these data,” said Dr Hansen.
Electronic diaries provide reliable prospective recordings, and this also adds to the validity of the study results, he said.
And because patients were typical of the migraine aura population, having been diagnosed at 16 specialized headache centers, external validity should be strong, too, he added.
A delegate attending Dr Hansen’s presentation commented that changes in smell during a migraine attack might be considered “osmophobia” — a hypersensitivity to odors. He questioned whether Dr Hansen had noted whether migraine patients have an “enhanced smell for flowers and roses and nature, et cetera.”
Dr Hansen responded that the questions in the study were posed “in a very general way” and did not include such detail. “But you’re right; we should ask is the smell unpleasant, is it something like a déjà vu?”
Another delegate wondered why some patients have huge variability in their aura symptoms while others “are completely consistent every time” they have an attack and whether Dr Hansen’s study categorized patients by these differences.
Although his study didn’t stratify patients in this way, Dr Hansen said he believes other research is doing so.
Congress of the European Academy of Neurology (EAN) 2016. Abstract #O2209. Presented May 29, 2016.