Role of Interventional Therapies for Migraine and Headache: Onabotulinum Toxin A, Stimulators, and Nerve Blocks

Please subscribe to use our print features or to download PDF files.

DR. AURORA: Whenever anybody asks me that question, I always ask him/her whether we have this question answered for any other therapies or, for that matter, any other clinical indications. What I tell clinicians is that anybody who meets the criteria for chronic migraine should probably have a trial, and, as Dr. Goadsby said, only those patients who respond should continue. One subset of patients who did well in the trial included those who were overusing acute medications for migraine, and these were randomized investigator sites.

This finding surprisingly contradicted the clinical dogma that if patients overuse acute medications for migraine, they don‘t respond to preventive therapy.

DR. DODICK: So, there are no powerful predictors of response or nonresponse.

DR. AURORA: Not according to the trials so far.

DR. DODICK: Okay, and in your clinical practice?

DR. AURORA: In my clinical practice, I feel that among patients who perhaps are at their end-stage, who have tried everything under the sun, and who have several psychological comorbidities, some of those patients may not respond as well. But, contrary to my aforementioned belief, I have sometimes been surprised. So, I don‘t write off anybody anymore, because I‘m shocked at who has responded.

DR. DODICK: Final question here, Dr. Goadsby, we‘re often asked this: What is the potential mechanism of action by which onabotulinum toxin works in chronic migraine; do we know that yet or can you speculate?

DR. GOADSBY: That‘s the “easiest” question: physiology. I don‘t really think we have a good idea certainly in comparison to other therapies that we have, but I‘d like to think that its effect is on large-fiber sensory nerves or even cranial parasympathetic nerves. I think it‘s unlikely that it‘s a muscle effect or an exclusively peripheral effect at the end of the day, but I don‘t think we‘ve spent nearly enough time even starting to ask the questions, and I don‘t really think that what‘s being done in the general pain area is going to be terribly informative because it hasn‘t been informative about any of the other migraine treatments. It‘s an exciting work in progress.

DR. DODICK: Dr. Goadsby, I‘m going to switch topics now and ask you a question about neurostimulation techniques that have been evaluated for the treatment of a wide variety of primary headache disorders. Speaking specifically about extracranial neurostimulation, do you actually see neurostimulation as a viable treatment option? And if you do, which modalities do you think look most promising?

DR. GOADSBY: Yes, I think extracranial stimulation is a viable modality. I don‘t think that it‘s been adequately proven certainly in the migraine sphere. The need for better-designed studies is obvious, and the difficulties with the perception and stimulation potentially influencing the placebo response are extremely important. In general, the less-invasive treatments are, the more excited I am, because the simpler they are to deploy, the more likely it is that they‘ll be used more widely.

I try to discourage open-label random use of peripheral stimulation in just a few patients in any center. I don‘t think that helps anybody, although I do look forward to the day we will have a better idea of how it might work and how to conduct the control trials that‘ll convince regulatory authorities and us that this is the right way forward. It‘s like the Mars lander; it‘s almost inconceivable that we are driving around Mars with a 1-ton vehicle. Five years ago, that would have been unbelievable, but there‘s someone driving around Mars with a 1-ton vehicle today.

DR. DODICK: Yes, Dr. Silberstein, for example, the use of an implantable stimulator that stimulates the occipital nerves is fairly invasive and costly. So, in which patients would you consider neurostimulation, and are there any clinical features, for example, a response to an occipital nerve block, that might identify patients who are likely to respond?

DR. SILBERSTEIN: Let me start with the second half of the question first, Dr. Dodick. In the first trial, based on open-label data,3 patients were administered an occipital nerve block. Those patients who achieved a response from an occipital nerve block were then administered an occipital neurostimulator. In our trial, we strongly argued that this relationship had never been proven.4 We created an auxiliary group of patients who showed no response to the occipital nerve block. The occipital nerve block was completely unpredictive and in fact, if anything, neurostimulation was numerically superior in those who did not respond.

I think first, we cannot predict, based on the response to occipital nerve block, the type of patients who will do well. Second, at the current time, my recommendations are that patients who have experienced a failure with all other treatments, including oral medications and botulinum toxin—and I even go to the point of inpatient infusion with intravenous lidocaine—are the patients I would then use a stimulator on: the most intractable and the failures.

DR. DODICK: Would any of you consider performing deep brain stimulation in any patient population, and in particular, in a patients with cluster headache, prior to at least a trial of occipital nerve stimulation?

DR. SILBERSTEIN: I would say yes. If you have a patient with a severe intractable cluster headache, in whom all treatment has failed, everything including invasive procedures, I might consider deep brain stimulation if the patient‘s life is so impaired that their alternative is what we call these patients‘ suicide headaches. I realize there‘s a lot of controversy, but again, although I‘ve never done it, if I had a patient like that, in whom all treatment has failed including peripheral stimulation, even using the new device for sphenopalatine ganglion stimulation, I might consider deep brain stimulation.

DR. DODICK: But you wouldn‘t consider deep brain stimulation in advance of or prior to peripheral nerve stimulation.

DR. SILBERSTEIN: That is absolutely correct.

DR. DODICK: I gather my colleagues would agree with that?


DR. GOADSBY: Yes, I certainly would. I think that the risk of the brain stimulation and the benefit have to be considered. It‘s not that everyone responds to deep brain stimulation. The best experience comes from the study by the Milan group, Leone, Bussoni, and Franzini,5 who have the most experience in the world. They would not say that 100% of their patients respond; their published results are about two-thirds.

Two-thirds is reasonable, but I don‘t think, in the context of having a less-invasive procedure, that is it is the first choice. There are new procedures underway and new devices, such as the sphenopalatine stimulator approach, that are now in trials in Europe, reportedly with success.6 So, there are so many solutions being developed that it‘s not time to think about deep-brain stimulation until all the peripheral options have been tried.

DR. DODICK: One of those peripheral procedures is transcranial magnetic stimulation, for which a trial has been conducted and completed in the United States,7 and I know that it‘s been approved in the United Kingdom. Do any of you have any experience with transcranial magnetic stimulation in patients with migraine? What‘s your take on this treatment modality? Since you used to practice in the United Kingdom, Dr. Goadsby, let‘s start with you.

DR. GOADSBY: Yes, I‘ve spoken to a number of patients who have used this device open label in the United Kingdom. There‘s a cohort, not all, who think it is useful to avoid attacks and reduce the severity of attacks. It almost seems too simple to be true, but as you know, we‘ve done some laboratory work, and indeed, it seems a very good basis for what‘s happening on the experimental side. We did some laboratory studies to examine the ability of a scaled-down version of the human stimulator that would produce the same Tesla field, the magnetic field, to affect cortical spreading depression. Indeed, in anesthetized rats and cat, this device could block the passage of induced cortical spreading depression. We have done additional work to determine the mechanism by which this device can interrupt thalamo-cortical relay circuits that are activated by trigeminal pain. So, there seems to be at least a few potential mechanisms.

DR. DODICK: Does that mean it might be effective in patients who have migraine without aura, in addition to those who have aura?

DR. GOADSBY: Yes, and I think that was the driving factor to start examining thalamic-cortical mechanisms, because there are patients who, both in the clinical trials and certainly in the open-label work, report, not an exclusive effect on their aura, but certainly an effect on pain, and indeed, patients without an aura at all report an effect on pain,which needs to be explained. One way of explaining that effect would be to think about the involvement of the cortical relays in the action of this modality.

DR. DODICK: Okay, now the third and final topic: extracranial nerve blocks. Dr. Silberstein, I‘ll start with you this time. Off-label treatment is common in clinical practice for most physicians by necessity. But there are few treatments that clinicians feel so passionately in favor of, despite the absence of evidence as extracranial nerve blocks and, in particular, occipital nerve blockade. So, if you can tell us, in which primary headache disorders do we actually have any evidence to support the use of occipital nerve block?

DR. SILBERSTEIN: I think the best scientific evidence for occipital nerve block with a local anesthetic in combination with corticosteroids is for cluster headache. Other people have conducted clinical trials to show that the combination works in comparison to placebo injection. If you look at the data for occipital nerve block, it‘s very difficult to control for an anesthetic. One can argue that perhaps, you‘ll use a short-acting versus a long-acting anesthetic, but the problem is that the duration of response often exceeds the duration of anesthesia.

We did a trial asking whether the addition of corticosteroids makes any difference in migraine as opposed to cluster headache, and we found no difference.8 I think we can say that steroids don‘t make any difference, but it‘s very difficult to control for a local anesthetic if people have had it for other simple procedures, like having teeth removal or a laceration on the skin. I cannot yet envision how to set up a clinical trial where the duration of the response exceeds the duration of action of the drug.

Bogduk, in a series of sophisticated experiments aimed at determining the cause of pain in the neck, used differential blocks with local anesthetics of different duration. But, in his clinical trials,9 the duration of response was equal to the duration of action of the local anesthetic, and when we perform occipital nerve blocks in humans with problems in that area, we do not find that correlation.

Psychologists would use waiting-room controls, but I think we‘ve all seen patients who have come into the office with severe significant pain and exacerbation of headaches, and when we perform a nerve block, their pain disappears for a month or may never come back. Others last for the duration of the anesthesia. So, I believe that it works, but I can‘t conceive, yet, of how to do a trial to prove it works.