So is there no relationship between familial hemiplegic migraine and migraine with typical aura? “Yes, there is,” Dr. Olesen said. “Our old studies showed that spreading oligemia looks very much like spreading depression—and this has been reproduced by many people using fMRI, using PET—and everybody now believes that the migraine aura is caused by spreading depression, but that is also true for familial hemiplegic migraine. And we could conclude that from our big clinical series by simply analyzing how the symptoms were progressing—from visual to sensory to speech and then to paresis. Timing that spread, it was so obvious that in familial hemiplegic migraine it had to be a spreading depression. In migraine with aura, it is also a spreading depression.”
But, of course, there is a difference. Patients with familial hemiplegic migraine have rare attacks; they don’t have frequent attacks. But when they get an attack, it moves all over the whole hemisphere. In migraine with typical aura, it stays in the visual cortex in most patients and sometimes spreads to the sensory and sometimes to the speech area but never to the motor area. They have more frequent attacks, but they don’t spread as much as in the patients with familial hemiplegic migraine.
“You would think that patients with familial hemiplegic migraine would be more severely affected and therefore would have more attacks—not only more widely spreading attacks but also more frequent attacks—but that is not so,” said Dr. Olesen. That difference in clinical phenomenology, he said, suggests that the mechanism underlying cortical spreading depression in familial hemiplegic migraine and more common migraines must be different.
The Genetics of Common Migraine
There are many ways to investigate the genetics of complex disorders. There are genetic epidemiology studies, twin studies, large family studies, sib pair analysis, candidate gene analysis, and genome-wide scans with hundreds of thousands of SNPs. At one point or another, Dr. Olesen and his research colleagues have pursued each of these methods.
In genetic epidemiology, the goal is to assess what causes familial relative risk, and that is the risk of migraine in first-degree relatives compared with the risk in the general population. “So if your migraine is genetic, you’ll expect to have more close relatives affected than you would see if you take random persons in the population and ask, ‘How many of your first-degree relatives have migraine?’ That’s the reasoning in genetic epidemiologic studies,” Dr. Olesen said. This ends up being more complex than it seems, “because it turns out people don’t know.”
When probands were asked if their first-degree relatives had migraines, they would answer one way, but then when the researchers interviewed those first-degree relatives, they often got completely different results. “So one has to interview all the first-degree relatives.” When Dr. Olesen and colleagues did this in the mid-1990s using a large sample from the Danish population, they found a fourfold increase in the risk of migraine with aura in the first-degree relatives of a proband with migraine with aura. Similar studies with probands who had migraine without aura found a twofold increase in first-degree relatives. Complex segregation analysis, which aims at determining the most likely mode of inheritance for any particular disease, pointed toward complex inheritance for migraine with and without aura. And this was confirmed in many other studies. “We are back to the threshold idea that everybody has a migraine threshold, and most have it so high that they never get a migraine attack. Others have [a lower threshold] and they get some attacks, and others have an even lower threshold and they get attacks all the time. It’s probably a combination of several genes, good and bad,” Dr. Olesen said.
But environment could still be a factor. “To really prove the genetics, you need a twin study,” Dr. Olesen said. Again drawing a study cohort from the Danish population, the researchers identified 77 monozygotic twin pairs and 134 dizygotic twin pairs. “There are two ways of showing concordance—pairwise and proband-wise,” Dr. Olesen said. In migraine with aura, both methods revealed higher concordance in monozygotic than in dizygotic twins. In migraine without aura, both methods also revealed higher concordance in monozygotic than in dizygotic twins, although the differences were not as great as those for migraine with aura. So the concordance data from twin studies corroborated the epidemiologic studies, showing an increased relative risk in migraine without aura and a greater increased risk in migraine with aura.
To assess the fraction of the risk that’s caused by genetic factors as compared to the fraction of risk that’s caused by environmental factors, the researchers next turned to heritability studies. In migraine with aura, the genetic component was assessed at 50%; in migraine without aura, inheritance was estimated at 65%.