BOSTON—Occipital nerve stimulation may be highly effective when other treatments fail to relieve the intractable pain of chronic posterior headache, according to David Dodick, MD. Small trials in migraine, cluster headache, and hemicrania continua have shown its potential. In addition, results from one sham-controlled trial have been reported, and two more trials are under way.
Candidates for occipital nerve stimulation are “patients that we see in practice weekly,” said Dr. Dodick, Assistant Professor of Neurology at the Mayo Graduate School of Medicine in Scottsdale, Arizona. These patients have gone through “years or decades” of unsuccessful treatment with many types of medications, but they continue to have chronic pain. Occipital pain is common among such patients, affecting almost half of those with chronic headache, and up to 90% of those with migraine. Dr. Dodick addressed the 50th Annual Scientific Meeting of the American Headache Society.
The Therapeutic Effect
The rationale for occipital nerve stimulation is based partly on the success of occipital nerve blocks in this patient population, according to Dr. Dodick. Studies have shown that short-acting local anesthetics to the occipital nerve can effectively diminish pain in the majority of patients. The onset of the blockade’s therapeutic effect—typically two to three days—is much longer than the duration of action of the anesthetics used. In addition, the duration of the therapeutic effect is even longer—in controlled trials, pain is relieved for several weeks or longer. “We think there is some central neural modulation occurring,” commented Dr. Dodick.
“Nerve blocks are effective over a range of primary headache disorders, including migraine, cluster headache, and hemicrania continua,” he said, but they do not work for all patients. Based on the success of pharmacologic blocks and the lack of other options, Dr. Dodick and colleagues explored the use of neurostimulation of the occipital nerve.
The stimulator consists of a battery-powered pulse generator, wire leads, and one of two types of electrodes. Cylindric electrodes are inserted percutaneously, while flattened, paddle-shaped electrodes require an open surgical insertion. Either the stimulator can be set to deliver a predetermined pattern of stimulation, or the patient may control the amplitude, pulse width, and frequency using a remote-control device. The pulse generator can be implanted in the clavicular, pectoral, abdominal, or buttocks area. Unfortunately, lead migration and breakage have both been common adverse events, noted Dr. Dodick, because of the high degree of movement of the neck.
An alternative is the microstimulator, which combines leads, battery, and programmable microchip into a device smaller than a AAA battery. There is less clinical trial experience with this device, but it has promise, Dr. Dodick said.
Trial Outcomes
A series of small trials, most with fewer than 20 patients, have shown the potential of the treatment. In Dr. Dodick’s trial of 13 patients with chronic migraine, eight had “an excellent outcome,” with at least a 50% improvement in one or more aspects of headache severity, frequency, or disability. Three patients became worse in one or two of these parameters, but none was worse in all three.
When he first analyzed the data from this trial, Dr. Dodick said, “I was a bit disappointed. If you are going to spend $40,000 to $60,000 to go through the surgery, you want to [hit] a homerun with these patients. But it was the patients who taught me that if you suffer through 25 years of incapacitating headaches, in which nothing else works, a stand-up double is really good. ‘You might not be happy,’ they’d say, ‘but we’re really happy.’”
Encouraging results have also been published from small trials in chronic cluster headache. Patients with hemicrania continua appear to respond even more robustly, with more than 90% of patients improving by at least 50%.
Numerous device-related adverse events—lead migration, breakage, and battery failure—have been reported in all trials. Adverse events were “not inconsequential” in Dr. Dodick’s trials, he said, with four patients requiring lead repositioning. In one study, the device in a patient with cluster headache had to be explanted due to persistent dysesthesia when the stimulator was on. Adverse events are reportedly lower with the microstimulator, “since there is nothing to displace,” Dr. Dodick stated, but the number of patients implanted is also smaller.
Results from the first sham-controlled, single-blind study of occipital nerve stimulation in chronic migraine were also presented. In this multicenter trial, 28 patients received stimulator implantation and controlled the stimulator themselves, 16 received implantation but without stimulation, and 17 remained on medication only. All patients had had migraines at least 15 days per month with severity greater than 5 on a 10-point scale, and the headaches were uncontrolled by medication. A positive response to treatment was defined as a 50% or better reduction in frequency or a 3-point reduction in severity. Positive responses were seen in 40% of the real-stimulation group, 6% of the sham-stimulated group, and none of the medication-only group.
Two other controlled trials are ongoing in the United States, with results expected within the year. “Hopefully, by next year, we’ll have a good idea of the potential usefulness of this modality in patients with chronic migraine,” Dr. Dodick said.
Delayed Benefits
An intriguing and important phenomenon characteristic of occipital nerve stimulation is that there is a significant delay between the onset of stimulation and the onset of therapeutic benefit, noted Dr. Dodick. Two months is typical, and up to six months has been observed; in some cases, patients get worse before they get better. “I think that has very important implications,” he indicated. First, it suggests that the therapeutic effect involves some element of neural plasticity, rather than simple blockade, pointing to a central, rather than peripheral, origin for the pain.
“It also tells me that there is no way we can predict who is going to benefit from long-term stimulation based on a short-term trial,” he related. Initially, patients underwent a seven-day stimulation trial to determine whether they were likely to improve with long-term treatment. “We are no longer doing that, since you cannot predict who is going to improve from a trial stimulation period.”
The delay in onset of benefit also makes the predictive value of occipital nerve blockade questionable. “Every one of our patients had had an occipital block,” Dr. Dodick said. “But if a three- to seven-day trial stimulation period isn’t predictive, a short-term block probably won’t be either.” Patients unresponsive to blockade may respond to stimulation, and vice versa. “At the end of the day, the answer is unknown.”
There is also a cumulative reduction in frequency and severity of headache. “The patients get better over time,” commented Dr. Dodick. “I tell patients that what you see at six months is probably what you’re going to get, but you really won’t know how good you will be until the six months have elapsed.”
Occipital nerve stimulation may also be appropriate for anterior pain. He noted that most of the patients receiving treatment had anterior as well as posterior pain, and when treatment relieved posterior pain, the anterior pain improved as well. One patient with hemicrania continua and only periorbital pain improved significantly with occipital nerve stimulation. “Is posterior pain necessary for a response to stimulation? I suggest probably not,” said Dr. Dodick. The ability of occipital nerve stimulation to treat anterior pain again points to a central mechanism, he added.
Numerous practical issues in occipital nerve stimulation therapy remain to be clarified. “Anesthesia is a major issue,” explained Dr. Dodick. “We prefer conscious sedation, because we want the patient to be able to report any occurrence and distribution of paresthesias, to help localize the electrode.” Neurosurgeons and anesthesiologists do not always like this, however, as it complicates their tasks. They prefer full anesthesia, although the best approach remains to be determined.
The best location for placement of the pulse generator is also unknown, as is the best set of stimulation parameters. Lead migration and battery replacement remain unacceptably high. “The technology must improve,” Dr. Dodick said. “Microstimulators are preferred if they are proved to be effective.”
Is bilateral stimulation necessary in patients with only unilateral pain? So far, results suggest not. “In all our patients with unilateral pain, we treat unilaterally, and not one has gone on to develop contralateral pain,” he observed. In such cases, bilateral treatment is “not necessary.” Can unilateral stimulation treat bilateral pain? “If it is acting through a central mechanism, it might be, but we don’t have the answer yet.”
Despite its encouraging successes, occipital nerve stimulation is not treating the cause or causes of headache pain, stated Dr. Dodick. He pointed out that in hemicrania continua, successfully treated patients continue to experience the autonomic effects of the disorder, including rhinorrhea and excess tearing. “Whatever is generating these attacks, we’re not affecting it,” he said.

—Richard Robinson

Occipital Nerve Stimulation for Headache