Physicians provide perspective on the unique challenges in accurately diagnosing migraine and the prevalence among headache sufferers coming into the clinic.
Stephen Silberstein, MD; Jessica Ailani, MD; David Dodick, MD; Peter Goadsby, MBBS; and Stewart Tepper, MD
PUBLISHED December 20, 2018
Stephen Silberstein, MD: Jessica Ailani runs the MedStar Georgetown Headache Center. I’ve had the opportunity of working with her for a long time. So I’m going to ask you, Jessica: What are some of the challenges in diagnosing migraine?
Jessica Ailani, MD: There are many challenges that are unique to diagnosing migraine. I think the first is getting the patient to come into the clinic and be seen by a provider who wants to listen to the symptoms that they’re having. How often are we in clinic and we ask a patient about their symptoms and they say we’re the first one who’s asking these questions, or we’re the only ones who are looking at their calendars or listening? I think one of the barriers is having providers who listen to the patient, who care, and then who get the patient to move forward in telling us what brings them there, how often they’re having headaches, and how they were feeling.
Stephen Silberstein, MD: David?
David Dodick, MD: The other issue, too, is that headache is ubiquitous; it’s a clinical diagnosis. There are no objective biomarkers. You can’t take a blood test—it’s a challenge. In medical school, I received no education in this area, even in residency; in my senior year of neurology residency I knew just about nothing regarding migraine. Every time I saw someone on my calendar who was coming in with a chief complaint of headache, I would shudder because I didn’t have a systematic approach to diagnosing and managing these cases. Clinicians and doctors don’t like feeling incompetent, right?
Peter Goadsby, MBBS: We spend a lot of time, maybe too much time, emphasizing all the different possibilities when teaching students. The landmark studies, and Stewart could comment better than me, show that you immediately put migraine and headache together. More than 90% of those who turn up to see primary care physicians complaining of headache have migraine. Maybe at some level we’ve made it difficult for people because we give them at least 50 different things to remember, when actually what’s most important is discussing migraine. So it seems to me we’ve done that kind of neurological clever thing where we know that someone could have a number of headaches so we obsess about it a little bit, whereas we’d do a lot of good in the world if we just followed the LANDMARK study.
Stewart Tepper, MD: What was interesting in that study was that we used the International Classification of Headache Disorders with which we evaluated attacks on patients who came in with a chief or secondary complaint, to either general neurology or primary care, with a headache; 94% of the time it was migraine. But what was challenging, getting back to Steve’s original question, is that, if a patient says, “I have tension headache,” then the doctor diagnoses tension headache regardless of what the actual diagnosis was, when it turned out to be general migraine. My old mentor Charlie Barlow MD, who wrote a textbook on pediatric headache at Boston Children’s Hospital, would say, ”It’s always migraine.” At the time I didn’t realize how wise he was.
David Dodick, MD: Well I think you’re right, Peter. We’ve done the neurological community a disservice by overly complicating this. When you think about primary headache disorders and classification, it’s either tension type headache, which neurologists almost never see in clinical practice, or it’s an attack—a trigeminal autonomic cephalalgia—or it’s migraine in the primary headache sphere. If you’ve got a patient with a side-locked headache and it never moves from one side, ask about cranial parasympathetic features, and maybe you’ve got attack. Otherwise, invariably, it’s migraine.
The problem is—and I think what makes neurologists nervous—is that the diagnostic criteria for migraine is very sensitive but not specific, so that you could have carbon dioxide poisoning, meningitis, subarachnoid hemorrhage, arterial dissection, venous sinus thrombosis—and the list goes on—that will present with a phenotype of migraine. This is what makes neurologists nervous. You know you can be nauseated and have sensitivity to light, and you can have a pounding headache. That doesn’t necessarily mean it’s migraine.
Stephen Silberstein, MD: David, what you said is brilliant. I think the classification has to change where we could have the phenotype of migraine as a phenotype. When does it become a disorder? What is the difference between a migraine attack and a migraine headache?
David Dodick, MD: Exactly.
Stephen Silberstein, MD: It requires 4 attacks. Why? We should be diagnosing phenotypes and more often the clinical disorder that leads to the phenotype. We’re not doing that.
David Dodick, MD: You’re absolutely right, Stephen. One of the other problems of course is that you know 12% to 15% of the population has migraine. There is high likelihood that someone is going to come in with a secondary headache that recapitulates the migraine phenotype, right? If I’ve already had migraine and I come into the emergency department now with a severe headache, and it resembles my prior migraine— but this time it’s a dissection—you can see how that could be overlooked.
Stewart Tepper, MD: Usually there’s a change in pattern, which you have talked about. Tell us a little bit about that.
David Dodick, MD: It’s all about the change in the pattern and not necessarily the phenotype. You worry about a patient who comes into the emergency department because they’re coming in for a good reason. They would rather not be there. The majority of patients are not seeking drugs or medication: They’re either in migrainosus status and looking for pain relief, or they’ve got another problem. You really have to ask about onset and you have to ask about the pattern. Has there been an increase in frequency? Has the pattern changed? There is a list of very simple questions that can be effective in eliminating secondary headaches. But I think that’s what makes neurologists nervous.
Jessica Ailani, MD: Often in clinic we’ll see patients who mention this change in pattern. It’ll be subtle sometimes and sometimes they’ll be using more acute medications, which may not be working well. While that can still be the underlying migraine issue, I think it’s important that we recognize that these might be the patients you need to stop and ask a few more questions [of] to see if anything has changed.
David Dodick, MD: Yes.
Jessica Ailani, MD: That will keep us in a safe zone to see whether this person is now developing a different type of headache that might have the phenotype of migraine. Yet, there’s been a change significant enough that we might need to image them at some point, and do a more extensive exam to pick up on other issues.
David Dodick, MD: Exactly.
Stephen Silberstein, MD: At the time sumatriptan was launched, we were saying that it was specific for migraines. I published a paper regarding a patient who came in with a phenotypic migraine headache. They gave sumatriptan and the headache went away. It turned out later that she had another subarachnoid hemorrhage. We published this paper on the headache of subarachnoid hemorrhage response to sumatriptan. That was based on the concept that the drug was not only good treatment, but diagnostic. One of the things they’ve learned over the years is just because something responds to drug X doesn’t mean that’s an indication of diagnosis.
Peter Goadsby, MBBS: It’s worthwhile finding that history taking is called history taking because you have to take it. Does light bother you? What do you mean by not really? Do you mean a little bit? If you don’t take past cases into account, you will have trouble diagnosing. If you engage with history on the first visit, you will save a lot of time going forward because you get the diagnosis correct, as well as the person’s confidence. History taking is a therapeutic bonding exercise in medicine, which also has incredible value from a diagnostic perspective. There’s no easy way to get around that.
Stewart Tepper, MD: I would just like to add another adage that I think is useful. Say you have a patient with subarachnoid hemorrhage. This was an abomination for Fred Sheftel MD, who taught me an awful lot about this. He said that when a patient visits a neurologist and the chief complaint is headache, one should be nervous. If the patient comes in and complains about headaches, one can relax a bit because the presence of headaches suggests a pattern of recurrence—a generally stable pattern. Whereas a patient complaining about a new headache should be carefully considered.
Stephen Silberstein, MD: A very good point. And you were saying that people get misdiagnosed where they say they have a tension headache or a sinus headache. When I hear that, my reaction is what do you mean by that? And then you get to the truth.