Non-invasive technique shows promise but needs more study
Oct 3, 2013
by Stephani Sutherland
Electrical stimulation of the motor cortex was established as an effective treatment for pain 20 years ago, but the risks and drawbacks of surgically implanting electrodes in the brain keep many patients from pursuing this invasive treatment (Tsubokawa et al., 1991; Kurata, 1993; Nguyen et al., 2011). What if there was a safer, non-invasive way to deliver analgesic neurostimulation? Transcranial magnetic stimulation (TMS) holds out promise as just such a next-generation pain treatment. In TMS, a magnetic field generated outside the head alters circuit activity inside the brain. TMS was approved in 2008 by the U.S. Food and Drug Administration for treatment of major depression, and researchers are investigating TMS for a number of other neurological conditions, including chronic, intractable pain.
Anne Louise Oaklander, a neurologist and pain researcher at Massachusetts General Hospital, Boston, US, said there is much work to be done, but the potential payoff of TMS for pain makes it well worth pursuing. “It has a huge potential advantage over pain medications,” she said. Drugs move indiscriminately throughout the body, often causing side effects at non-target tissues that limit their use or even prevent them from getting into the clinic. TMS, by contrast, delivers its therapeutic effects directly to the brain, with only minor, local side effects. “That is the trump card of TMS over drugs,” Oaklander said.
Because TMS can modulate brain circuits safely and painlessly, the technique has had tremendous utility for studying pain processing, said Jean-Pascal Lefaucheur, a pioneer in the field of neurostimulation for pain at the Université Paris-Est, Créteil, France. Lefaucheur and his colleagues recently reviewed some of the hundreds of small studies that have examined the effects of TMS on evoked pain in experimental settings (Mylius et al., 2012). (In a separate review, Lefaucheur and colleagues recently discussed the mechanisms of action and the clinical indications of TMS for non-invasive stimulation therapy of pain disorders; see Nizard et al., 2012.)
Clinically, TMS is routinely used as a screening technique to predict a patient’s reaction to cortical stimulation before surgical implantation of electrodes, said Lefaucheur. A person who responds to TMS will almost always benefit from brain stimulation with implanted electrodes, he said. If a patient could receive the same benefit without implantation surgery, all the better.
But few studies have addressed the technique’s clinical efficacy. A 2010 Cochrane Review of non-invasive brain stimulation aimed at chronic pain looked at the accumulated data and concluded that high-frequency TMS showed a small but consistent reduction in patient-reported pain scores compared to sham treatment (O’Connell et al., 2010). Most of the studies’ participants had chronic, intractable neuropathic pain, and TMS produced a small and transient decrease in pain scores by about 15 percent for up to a week. However, inadequate sham controls and blinding may have exaggerated this effect, said lead author Neil O’Connell, Brunel University, Uxbridge, UK.
Although the pooled data involving just 368 subjects in 19 trials did not provide
Neil O’Connell evidence of a clinically meaningful, long-term analgesic effect of TMS, O’Connell said one could yet emerge with further study. It is difficult to synthesize the results of small, heterogeneous studies, he told PRF. While it would be premature to roll out TMS clinically for pain, “I absolutely encourage scientists to study it more,” O’Connell said. “I am not saying it will not be useful [in the clinic], just that substantial uncertainty remains.” An updated review now in preparation will include more studies, but the conclusions for TMS will not change drastically, O’Connell said.
All brain stimulation techniques work by activating neural circuitry. With TMS, a figure eight-shaped plastic paddle containing a coiled wire is placed over the head and briefly charged with high-intensity current, inducing a magnetic field that passes into the brain. Just as a wire placed in a magnetic field will carry current, so do axons. Neurostimulation therapy, Lefaucheur explained, prompts axons to fire action potentials and thereby influences circuits. “It is not [just] a local stimulation. You can stimulate local, short circuits or circuits with distant projections,” he said. Accordingly, TMS of the richly connected cortex can modulate the activity of structures deep within the brain.
Perhaps because its effects go beyond the cortex, TMS seems to affect different aspects of the experience of pain. “Pain is not a monolithic entity,” said David Yeomans, a pain researcher at Stanford University, California, US. Just as pain can arise from a variety of sources, so does it vary in its qualities and characteristics—for example, our discriminative sense feels the bodily sensation of pain, but the emotional aspects of pain cause our suffering.
Cortical stimulation seems to affect both these components, said Lefaucheur. Improvement of the sensory discrimination of pain might arise from modulation of descending inhibition circuits, which pass through the thalamus en route to brainstem structures and the spinal cord. In contrast, changes in the emotional aspects of pain likely arise from effects on the brain’s limbic circuitry, he explained.
Like any treatment, people do not respond uniformly to TMS; patients may see improvement in one aspect of the pain experience but not the other.
Further complicating the clinical understanding of TMS are the endless variations in its delivery. In clinical studies, the motor cortex has emerged as the clear winner in terms of where to target TMS for pain; the few studies that were aimed at dorsolateral prefrontal cortex—the bullseye for treating depression—were ineffective for pain (O’Connell et al., 2010). But the clarity ends there when it comes to the details of effective stimulation. Where, for example, within the motor cortex should one stimulate? One might aim intuitively at the cortical real estate representative of the painful area, but experts agree that the analgesic effects do not correspond to the somatotopic map.
Repetitive TMS (rTMS), in which a series of pulses is delivered in rapid succession within a single session, has emerged as the preferred method, but the protocol can be highly variable in details such as pulse frequency, stimulation intensity, and timing and duration of the treatments—for how long and how often should rTMS be delivered? The exact best technique is likely to vary for different people with different types of pain. Most clinical studies have used a pulse frequency somewhere between 1 and 10 hertz, with high-frequency stimulation consistently producing better effects than low-frequency (O’Connell et al., 2010).
It now seems clear that any long-lasting effects on pain will require multiple sessions of rTMS. In addition to the clinical evidence for the treatment’s transient nature, one current theory of how rTMS works also fits with the need for ongoing sessions. Multiple sessions of rTMS might affect the brain’s connectivity much like learning does. When you learn to play the piano or speak a new language, Oaklander said, “you create new synapses and you lose others; you change your brain.” If TMS engages the same types of synaptic plasticity, many treatments might be required to reshape signaling in circuits molded by chronic pain. As for how many, Oaklander said, “Nobody knows how often it needs to be repeated.” Treatments might be required daily at first and then as often as weekly for months or even years.
The updated Cochrane Review will include several studies of multiple sessions, O’Connell said, which were absent from the 2010 review. In one recent small study, a course of 14 rTMS sessions over 21 weeks for fibromyalgia resulted in long-term improvements in pain and quality-of-life scores (Mhalla et al., 2011). But even repeated treatments might not change the brain enough to provide lasting relief for some. The first randomized, multicenter, sham-controlled trial of repeated rTMS for neuropathic pain showed that 10 daily 5 Hz sessions provided only short-lived benefits with no cumulative effects (Hosomi et al., 2013).
Youichi Saitoh, a neurosurgeon at Osaka University in Japan and lead author of that study, believes that relief from neuropathic pain will require indefinite treatment with rTMS. Even in patients with implanted electrodes who have used neurostimulation for 10 years, analgesia does not last beyond about a day, Saitoh told PRF in an email. That leads him to think that rTMS does not lead to permanent neuroplastic changes in the pain processing system, he said.
Some researchers are investigating whether multiple sessions of rTMS cause structural remodeling in the brain. Neuropathic pain leads to well-documented structural changes in the brain, for example in the cingulate cortex (May, 2008). Yeomans and his colleagues plan to use brain imaging to investigate whether those changes might be reversed following a course of treatment with rTMS.
In practice, the delivery of TMS over multiple sessions remains a challenge because of the need to reproducibly target a specific brain area from outside of the head. Newly developed magnetic resonance imaging (MRI)-guided neuro-navigation can help the technician to hit the intended mark, but it remains to be seen whether the exceedingly expensive technique will be worth the price tag.
The Cutting Edge
One property—some would say limitation—of TMS is that it can only reach regions that lie within centimeters of the brain’s surface. Could pain pathways be better
targeted at deeper structures? Several groups are asking that question. Brainsway, a company in Israel, has developed an “H-coil” to deliver what they call deep TMS. In a recent study, researchers used the H-coil to target the area of the motor cortex representing the leg, de
ep in the central sulcus, in subjects with diabetic neuropathy and saw pain relief that lasted up to three weeks (Onesti et al., 2013). In other work, Yeomans and his collaborators used four coils and what he described as “high-level math” to model how the combined coils might “shape” magnetic fields to direct currents deep into the brain (Tzabazis et al., 2013). They aimed at the dorsal anterior cingulate cortex (dACC), an area Yeomans says is activated by any experience of pain, according to neuroimaging studies. In their study, both acute pain in healthy subjects and chronic pain in subjects with fibromyalgia were attenuated by rTMS.
In MRI-guided TMS, the operator holds the figure-eight coil to the patient’s scalp while monitoring the brain stimulation site on the three-dimensional MRI. Stereotactic spheres mounted on the coil identify its position relative to the spheres on the goggles that localize the patient’s head. Credit: Roi Treister, Massachusetts General Hospital, Boston, US
Although leading researchers in the field were supportive of these exploratory forays into the deeper reaches of TMS, they overwhelmingly agreed that what is most needed is a better fundamental understanding of the technique. Who might benefit most from TMS—people with neuropathic pain or other forms of pain? What is the optimal stimulation site, at what device settings, when and for how long? And what will benefits look like? How might TMS treatment interact with analgesic drugs? All of these basic questions remain unanswered. “It is a very complicated problem,” said Lefaucheur. “We need a large series of patients to clearly determine the correlations” among all these factors, he said.
O’Connell agreed and in a 2011 editorial (O’Connell and Wand, 2011) argued that rather than develop new coil configurations, researchers should prioritize robust but basic studies of TMS. “Large, well-controlled studies are needed to test whether that early promise is real,” he said. Oaklander and her team recently wrote a review of TMS (Treister et al., 2013, in press in Rambam Maimonides Medical Journal) and have applied for funding for the planning stage of a large-scale clinical trial of TMS for neuropathic pain. This October, the Radcliffe Institute of Harvard University will host a gathering of scientists, clinicians, and regulatory agents to set out a path to designing such a trial.
Other non-invasive stimulation techniques are being investigated, too, including transcranial direct current stimulation (tDCS) and cranial electrotherapy stimulation (CES), in which electrodes applied to the scalp deliver low-intensity current directly. Much fewer data are available for these modalities compared to TMS: The Cochrane review of six small studies suggested tDCS might provide a slight benefit for pain, but the updated review of data suggests no significant effect over sham stimulation, O’Connell said. CES appeared to provide no benefit in four studies reviewed.
The Bottom Line
Even if the promise of TMS for pain holds up, the technique surely will not be a panacea. “The main limitation for TMS is the short duration of its [analgesic] effect,” said Lefaucheur. “It may not be feasible for some refractory, chronic neuropathic pain,” he said, considering the ongoing need for treatment that is costly, time-consuming, and requires technical expertise for delivery. Lefaucheur has had several patients with neuropathic pain who initially shunned electrode implantation for rTMS, but after a year of the toil of monthly treatments, they eventually opted for surgery.
In Japan, Saitoh and his colleagues are working to make daily rTMS treatment more accessible by developing a cheap, user-friendly rTMS machine for everyday home use. He hopes such a device will eliminate the need for electrode implantation.
Overall, TMS may be more useful to treat acute forms of pain and pain that is not so refractory to other treatments, Lefaucheur concluded. In treating depression, TMS is often used for a few days during an acute depressive episode, and “there can be a synergy between stimulation and [antidepressant] drugs,” Lefaucheur said. But with refractory, chronic pain, “the pain is constant, and the drugs do not work well.” Such conditions may require ongoing treatment in repeated sessions.
Despite the caveats and the need for more study, researchers were hopeful that rTMS could turn out to be useful in the pain clinic. David Brock, Medical Director at Neuronetics, Malvern, Pennsylvania, US, a company that makes the TMS machines widely used in clinics for depression, likens TMS to a Swiss Army knife. “We have figured out how to use one blade—for depression—but it has so many other potential tools. We just have to figure out how to use them,” he said. If TMS can change the disease state of chronic pain without drugs, as it has for depression, Brock said, “that would be a real boon to patients and society.”
Stephani Sutherland, PhD, is a neuroscientist, yogi, and freelance writer in Southern California.
Image credit: Eric Wassermann, via Wikimedia Commons