When we last reviewed neurostimulation devices 3 years ago, we concluded that there was some promise—but more sizzle than beef. Now there are more devices and more data. But is there more beef? Maybe.
We generally think of neurostimulation as a new technology, but its history is long. Beginning in the 1870s through the turn of the last century, electrotherapy, adopted enthusiastically in the U.S and Europe after its introduction by German pioneers, became standard treatment for melancholia, neurasthenia, and related conditions. The most commonly used instruments were, in fact, substantially similar to those used today in transcranial direct current stimulation (tDCS). In light of current debates about the extent of the placebo effect in brain devices, it seems instructive that over 130 years ago, internationally recognized experts convened a symposium to debate the following question: “Are the positive results produced by electrotherapy based on suggestion?” Concerns about observer bias, the absence of a clear explanatory model, and the fact that electrical treatments were “repeatedly suspected of attaining results through suggestion only” led to their eventual fall from favor. By the early 1990s, in fact, electrotherapy had essentially disappeared from mainstream practice.
Thus cautioned, lets sally forth into today’s therapeutic “electro-landscape.” We’ll restrict our focus to the treatment of depression, and specifically (where data are available) to treatment-resistant depression (TRD). Up for your consideration are: transcranial magnetic stimulation (TMS); magnetic seizure therapy (MST); vagus nerve stimulation (VNS); transcutaneous vagal nerve stimulation (t-VNS); transcranial direct current stimulation (tDCS); and deep brain stimulation (DBS).
Transcranial Magnetic Stimulation (TMS)
What is it? Transcranial magnetic stimulation (TMS) is a non-invasive procedure that uses magnetic fields to stimulate nerve cells in the brain. The magnetic coils look different depending on the device used. In the Neuronetics device, the patient reclines in what looks like a dentist chair and a coil is lowered onto the scalp near the forehead. In the Brainsway device, the patient sits in a standard chair and the device is more like a helmet that is put over the entire head. Standard treatments occur once a weekday for six weeks.
How available is it for patients? Many psychiatrists are offering TMS therapy to patients; it is available in most areas.
Does it work? In 2008, the FDA approved the rTMS as a treatment for adults with MDD who “have not responded to a single antidepressant medication in the current episode.” As recently as early 2014 (Cusin C, Dougherty D, Biol Mood Anxiety Disord 2012; 2(1): 14), meta-analyses generally emphasized the paucity of well-designed trials and the often less-than-overwhelming results. For instance, the one large NIMH-sponsored, randomized, sham-controlled study “showed a statistically significant difference between the treatments, but overall low rates for both response and remission.” According to authors Cusin and Dougherty: “Based on published data, the role for TMS in the treatment of TRD is still unclear.”
Earlier this year, however, a team at Chapel Hill published the results of a meta-analysis of rTMS in a more narrowly defined population: “patients with major depressive disorder and 2 or more prior antidepressant treatment failures.” (Gaynes B et al, J Clin Psychiatry 2014; 75(5):477-489). They limited their data to “good- or fair-quality studies comparing rTMS with a sham-controlled treatment in TRD patients…published from January 1, 1980, through March 20, 2013.” Their results were distinctly more encouraging: “rTMS was beneficial compared with sham for all outcomes…produced a greater decrease in depressive severity (high strength of evidence), and greater response rates (high strength of evidence)…finally, rTMS was more likely to produce remission (moderate strength of evidence); patients receiving rTMS were more than 5 times as likely to achieve remission as those receiving sham (relative risk = 5.07; 95% CI, 2.50-10.30).
Conclusion: For carefully diagnosed cases of TRD, it appears reasonable to consider rTMS.
Magnetic Seizure Therapy (MST)
What is it? The patient sits (or lies) in a reclining chair. Small magnetic coils are housed in two round pads attached to a pole that comes out of the main body of the machine. The pads are placed on either side of the scalp, just behind the temples. The operator activates the device and electricity is pulsed into the magnetic coils housed inside the pads. It is similar to the TMS pulse, but the pulses are at a higher intensity and frequency so that they produce a seizure. The patient is under general anesthesia for the procedure. This process directly stimulates the portion of the brain that regulates the mood. MST resembles ECT in the number and scheduling of treatments (typically 2-3 per week for 4 to 6 weeks) and the need for anesthesia.
Proponents claim that, compared with conventional ECT, MST can produce a more precise cortical seizure focus. This would have a few advantages. First, it would eliminate the need for the bite block used in ECT because the masseter (jaw muscle) would not be stimulated. Second, and more importantly, the stimulus would not reach brain structures important for memory, such as the hippocampus. This would lead to less acute post-ictal confusion, faster recovery times, and a lower risk of cognitive and memory impairment, the chief bugaboos associated with ECT.
How available is it for patients? Not FDA approved; it is available to patients with major depression only through a research protocol.
Does it work? Since 2006, a handful of studies have, with a fair degree of consistency, shown MST to be effective for TRD. A non-industry funded review published earlier this year (Cretaz et al, Neural Plast 2015; Epub May 13) describes response rates ranging from 40% to 60% and remission rates ranging from 15% to 30%. The authors were impressed that “most trials were conducted on patients suffering from TRD, who had failed previous therapeutic strategies and therefore had a worse prognosis.”
They also found that MST does indeed cause fewer cognitive side effects than ECT—for example the reorientation time was 2-8 minutes for MST vs. 15-25 minutes for ECT. In addition, MST is much less likely to cause prolonged memory loss. The downside is that MST, while effective, is quite a bit less effective than ECT for TRD. ECT posts remission rates of 50%-70%, at least double the reported remission rates from MST. The authors suggest that MST will gradually become more effective with improvements in lead placement, pulse frequencies, and other parameters.
Conclusion: It’s not clear how MST fits into the range of treatments available. It causes fewer side effects than ECT, but its less effective. And it may not be anymore effective than TMS—which is a much less noxious procedure.
Vagus Nerve Stimulation (VGS)
What is it? Vagus nerve stimulation is a neurosurgical procedure. The patient is put under anesthesia, and the surgeon embeds a silver dollar-shaped device—the stimulator—under the skin of the upper chest, just below the collarbone. A second incision is made on the lower left side of the neck where three small electrodes are wrapped around the vagus nerve. The stimulator in the chest sends electrical impulses to the vagus nerve. The frequency and intensity of impulses can be adjusted non-invasively using a wand that interacts with the device remotely.
How available is it for patients? Though FDA approved, it is not widely available.
Does it work? Based on what many felt was lukewarm evidence (see for example TCPR’s 2006 update), VNS was approved by the FDA in 2005 for treatment-resistant depression. Its invasive nature, potential for messy mechanical problems (for example, any time a battery malfunctions), and side effects (including cough, hoarseness, sore throat, and headache) have resulted in very limited clinical use. One recent review (Cusin C, Dougherty D, Biol Mood Anxiety Disord 2012;2(1):14) was a little more bullish, concluding that while the device is not effective for acute treatment of depression, its effect seems to increase over time. In an open-label extension, patients given VNS were followed and had one-year response rates of up to 34% and remission rates of 15%. But there was no control group for comparison.
Conclusion: VNS is almost certainly ineffective for the first 3 months after implantation, but many become effective over time. We need to see more studies to be sure.
Transcutaneous Vagal Nerve Stimulation (tVNS)
What is it? tVNS is like VNS but without the need for surgery. A branch of the vagus nerve supplies a part of the ear, allowing you to stimulate the nerve with an electrode that is simply placed against the ear. In practice, the whole set looks like wearing headphones and listening to an iPhone. You just charge it up and wear the electrodes for the prescribed amount of time. 15 Minutes once or twice a day for two weeks, and you have a potential treatment for depression.
How available is it for patients? It is approved in some European countries for the treatment of refractory epilepsy and can be prescribed there.
Does it work? A single pilot study randomly assigned 37 depressed patients to tVNS (worn for 15-30 minutes a day for two weeks) or to sham treatment. Active treatment beat sham on the Beck Depression Inventory but not the HAM-D scale. There were no side effects. (Hein E et al, J Neural Transm 2013:120(5):821-827).
Conclusion: tVNS has promise, but its much too soon to tell.
Transcranial Direct Current Stimulation (tDCS)
What is it? The device involves placing two electrodes—one at each temple—onto the patient’s head with an elastic band and flipping a switch. A relatively weak, direct (as opposed to alternating) electrical current is delivered for 20-30 minutes, 5 days a week, for 2-3 weeks.
How available is it for patients? No FDA approvals. Do it yourself kits are available to anyone. Other, more legitimate devices are available by prescription.
Does it work? At the time of our last review, the available evidence was discouraging: two blinded, sham-controlled studies showed modest symptom improvement, but not significant difference between treatment and control groups (for information, see NIH pages http://1.usa.gov/1fhBHK9 and also http://1.usa.gov/1gjILX0). What’s happened since then?
As mentioned in this issue’s accompanying piece by Drs. Sahlem and Borckhardt (p. 8), a double-blind 2013 study randomized 120 subjects with MDD to active tDCS, sham tDCS, sertraline, or placebo. Active tDCS was better than sham, and the combination of sertraline and electricity was more effective than either treatment alone. Subsequent meta-analyses, however, have been mixed (of three, one concluded that tDCS “has promise”; one found “a medium, significant effect size in outcomes”; and one euphemistically described the clinical utility of tDCS as “unclear when clinically relevant outcomes such as response and remission rates are considered”).
Conclusion: Larger studies still needed!
Deep Brain Stimulation (DBS)
What is it? In this invasive form of treatment, electrodes are surgically implanted in deep brain structures, including the subgenual cingulate cortex (SCC; also known as the Brodmann area 25. Ref: Drevets et al, CNS Spectr 2008;3(8):663-681) and the ventral striatum/ventral cortex, that have been implicated in depression.
How available is it for patients? It is available with proper consultation and assessment by doctors. This may require consults with multiple specialists, including psychiatrists, psychologists, neurologists, and neurosurgeons, to determine if the device is appropriate.
Does it work? After a series of very impressive open-label trials showing robust responses and a few side effects, several large, multi-center randomized, sham-controlled studies were launched, accompanied by fanfare and great expectations. Disappointment has followed: At least two of these, one sponsored by Medtronic (http://bit.ly/1Hqsk1G) and the other by St. Jude Medical (http://www.sjm.com/broaden), have reportedly been shut down due to lack of measurable benefit in the treatment arms. According to one review author, “researchers involved remain hopeful that modifications of inclusion criteria and technique might ultimately result in a demonstrable clinical benefit for some subset of patients…” (http://bit.ly/1fCL7jE).
Reflections on the Past and Future of TMS
Mark George, MD
Distinguished University Professor, Layton McCurdy Endowed Chair; Director, Brain Stimulation Laboratory
Dr. George discloses that he is an unpaid consultant to Brainsonix, Brainsway, Cerval/Neostim, Mecta, Neuronetics, NeoSync, Nervive, and Puretech Ventures. Dr. Carlat has reviewed this interview and found no evidence of bias in this educational activity.
TCPR: Dr. George, I guess if there is a “father of TMS” you would be one to potentially qualify for that role. When did you get involved in this research?
Dr. George: There are several modern parents, with Tony Barker inventing the modern machine in 1985, and John Rothwell and Mark Hallett developing the neurophysiological and neurological aspects. In the area of psychiatry, back in 1993 I had this then quite heretical idea that you could noninvasively, nonconvulsively stimulate the prefrontal cortex and ease people’s depression. It was heretical because people thought the only way you could stimulate the brain to treat depression was ECT, and that required a seizure. But nevertheless, we did the early trials, and now, over 20 years later, there are three large randomized controlled trials with Class I evidence showing that TMS is effective for depression. The first was funded by Neuronetics, which got FDA approval around 2008. Then we did a trial without industry funding, the NIH OPT-TMS (optimization of TMS) trial which also found positive results compared to sham (Borckhardt J, Brain Stimul 2013;6(6):925-928). And then the most recent large trial was published a couple of months ago and funded by Brainsway (George M et al Brain Stimul 2014;7(3):421-431), and their multisite international trial was positive as well and they got FDA approval about a year ago.
TCPR: How effective is TMS?
Dr. George: We’ve found that across multiple studies the chance of remission in treatment-resistant patients is about 30%, and the chance of response is about 50%-60%.
TCPR: And in terms of length of the effect, how long-lasting do we think TMS is?
Dr. George: We don’t have a lot of good data on durability of TMS. We know that ECT has excellent acute efficacy rates, 60%-70%, but the effect is not very durable (George M et al, Arch Gen Psychiatry 2010;67(5):507-516). You have over a 50% chance of relapse at least six months despite aggressive medication use. So far in the TMS literature, the effects appear to be more durable than ECT and maybe more durable than medications. Neuronetics was required by the FDA to do some long-temr studies as a condition of approval. They found that if you are a remitter after the acute course of TMS, a year later you have about a 90% chance of still being a remitter—with the caveat that you have been on medications and sometimes required repeat tune-up sessions of TMS. That study has not yet been published, but the findings have been published in abstract form and they are encouraging. (Links to Neuronetics ongoing research can be found on their site here: http://bit.ly/1TOZXTv.)
TCPR: How often to patients get TMS tune-ups?
Dr. George: That seems to be highly variable. Our practice here in Charleston is to give people an initial course of TMS, and if they remit we do not automatically do maintenance TMS, but we maintain them on medication. If they do start to relapse we will go back in and do some sessions for a week or two, which often will bring them back to remission. We then talk to them about maintenance TMS. And there are no good rules yet or good studies about how to do maintenance. People are generally tapered to a treatment every week and then every two weeks attempting to try to reduce treatment to once a month. Many patients don’t need repeat treatments, but others seem to, and unfortunately we don’t have any way right now to predict who is going to need that level of TMS.
TCPR: When I’ve read the TMS studies, I’ve always been concerned about whether the treatment is truly double-blinded, because the active treatment causes a clicking and scalp sensation. The concern is that patients and treaters might guess which is the active treatment arm, leading to a larger placebo effect in that group.
Dr. George: Yes, this was a real problem with the early studies of TMS. The treater is standing right beside the person every treatment session, which might be every weekday for an hour for four weeks, so that is a lot of exposure to a treater. To prevent positive treatment expectations, you have to prevent anybody who comes into contact with the patient from knowing which arm he or she is assigned to. The first Neuronetics trial did not rise to his level of methodology. They tried to keep the coil operators uninformed, but in an informal poll of everybody who was involved in that trial, all treaters figured out which patients were getting real TMS almost immediately. So that was not truly a double-blind trial.
TCPR: And could the patients guess which treatment they were getting as well?
Dr. George: They might have been able to, but in that trial, which was industry sponsored, they did not ask that question. Don’t ask, don’t tell. With the NIH-funded trial, we worked really hard and we came up with an active sham condition. We put a small ECT pad that was connected to a greatly reduced ECT machine and later a small TENS unit (transcutaneous electrical nerve stimulation) underneath the TMS coil. The patients in the sham condition get a small electrical discharge in those pads underneath the coil precisely at the time of the supposed TMS. So they get the exact same pain sensation as real TMS. In addition, the real versus the fake coils have a different tone to them when they discharge, and so we used noise-dampening ear- plugs in the patients and the treaters. And with all of that work we were able to come up with an active sham condition that was truly a sham and was truly blind. And we did ask patients and treaters and raters across the trial what they thought they were getting, and they were not able to guess better than chance. So the first Neuronetics trial wasn’t truly double blind by the most rigorous definition; the OPT-TMS was truly double-blind, and the Brainsway group built on the technology that we developed in the OPT-TMS, and theirs was also truly double-blind.
TCPR: So what do you see happening with the future of TMS?
Dr. George: I’m a bit of a dreamer. When we started this, we chose some of these ways that we are doing it for no particularly good reason. My boss Bob Post and I had to make a lot of educated guesses. The idea of once a day with weekends off—I just modeled that on ECT. Could we make it more efficient? So, for example, a lot of people can’t travel to a doctor’s office every day; they live far away from a TMS device. So can you bake the cake faster? Can you get people undepressed faster? Or can you do all the treatments in a day? Different studies are now experimenting with that. I just came from a meeting and learned about a study in Korea in which they gave an entire course of TMS—30 treatments—in three days. They randomized people to intensive TMS versus the standard once a day weekdays for six weeks, and found the results were equivalent after 6 weeks.
TCPR: Is accelerated TMS off-label?
Dr. George: It has not been formally studied, but yes, we have recently treated a cohort of new onset nontreatment-resistant depression and we had a 100% remission rate. There was one person in that group who had been wrongly diagnosed as having Alzheimer’s disease, and when I took a history I found a pattern of chronic recurrent untreated mood disorder. When we treated him, his dementia turned out to be pseudodementia and it went away. So people are using it in depression in the setting of cognitive decline and with good results. Unlike ECT, it doesn’t have any potentially negative cognitive side effects.
TCPR: I want to touch on an ethical issue. Psychiatrists who lease these expensive machines have a strong financial incentive to refer patients to the treatment. Other specialists have long wrestled with those kinds of issues, but not psychiatrists. Do you have any thoughts about that?
Dr. George: That’s an important issue, and the level of conflict of interest depends to some extent on your treatment model. In one model treaters view themselves as referral centers, and most of the patients that they treat are not their patients. And in that model I think there is less of a conflict of interest, because the incentive is for the psychiatrist to do their best job with TMS so they will get future referrals. That is the way ECT has commonly occurred; there are usually one or two ECT providers in a town, and then different psychiatrists will refer to that ECT practice and so there is less self-referral. Now, if you have a small practice and you own or lease the device and you self-refer your own patients, that could be more problematic. At the minimum, you would certainly need to inform your patients about the potential economic incentives so that they are aware of it. You might also consider a policy in which someone else who is not economically linked to the decision should chime in to say yes or no.
TCPR: Seeing TMS being adopted so widely must be pretty satisfying to you after all the fundamental work you did on it.
Dr. George: In psychiatry we don’t have many good examples of rags to riches, taking a far out idea into a clinically usable technique. TMS for depression is thus unique. So I have been very satisfied with the way it has gone and I think the future is promising. But what I say to everybody is that I would love in 20 years to come back and find that no one is using TMS and that we have figured out something that is even better. And that TMS as a technology was just a bridge to even more effective ways of modifying the brain therapeutically. So I am not necessarily wedded to TMS at all; it is one technology and it may become a bridge to even more effective ways of modifying neural circuits and helping people who are suffering.
TCPR: Thank you Dr. George.
Fisher Wallace and Alpha-Stim for Depression? Claims vs. Evidence
Gregory L. Sahlem, MD, clinical instructor of Psychiatry and Behavioral Sciences at the Medical University of South Carolina
Jefferey J. Borkhardt, PhD, associate professor and director of the Biobehavioral Medicine Division and Brain Stimulation Laboratory at the Medical University of South Carolina.
Both Drs. Sahlem and Borckhardt have disclosed that they have no relevant relationships or financial interests in any commercial company pertaining to this educational activity.
Carly Simon swears by it. The daytime show “The Doctors” gave it a glowing review. Ads for it seem to be invading psychiatrists’ Google search results. We’re talking, of course, about the Fisher Wallace Stimulator, touted by the manufacturer as being an effective treatment for depression, anxiety, insomnia, and pain. (http://www.fisherwallace.com/). The Alpha-Stim device makes similar claims (http://www.alpha-stim.com/). How do these devices work? Are they actually effective? And where do they fit into the rapidly expanding array of neuromodulators?
If you want to understand these devices, it’s best to start by learning about their simpler cousin process, transcranial direct current stimulation (tDCS). Picture a battery with a wire attached to the negative end (the anode) going into a light bulb and returning to the battery’s positive end (the cathode). Electrons flow through the wire, heating the lightbulb’s filament to create light. This one-way electrical current is called direct current (DC).
Now, imagine that the lightblub is replaced with your skull, and the wires from the battery go to conductive pads that are kept on either temple by a headband. You are not picturing a simple tDCS device. A small charge will flow to your temple and will stimulate the part of your cortex under the electrode. If we’re talking about, say, the left dorsolateral prefrontal cortex, which is thought to be hypoactive in depression, this stimulation will theoretically increase electrical activity there and ease your depression.
It makes sense, and it may actually work—according to a recently published randomized sham-controlled trial. In the study, 120 mildly treatment-resistant patients with MDD were randomized to tDCS alone, tDCS combined with sertraline, sertraline alone, or placebo, and followed for 6 weeks. Both tDCS and combined tDCS and sertraline worked better than placebo alone or sertraline alone, and the combined treatment posted the best outcomes of all (Brunoni AR et al, JAMA Psychiatry 2013;70(4):383-391). Its not a huge study, and we’d like more data, but we can at least say that this technology looks promising—and the side effects are almost non-existent.
How are the Fisher-Wallace Stimulator and the Alpha-Stim devices relaced to tDCS? They’re very similar except the current going to the skull is alternating current (AC) instead of direct current (DC), and for that reason they are categorized as tACS (transcranial alternating current stimulators.) In DC, the flow of electrons is constant and in one direction, but in AC, the flow changes direction frequently. AC is how electricity is transported to households, and in that context it has the advantage of being more efficient and cheaper to deliver.
Why would AC have any advantage over DC for brain stimulation? It’s not clear. The theoretical advantage of AC is that the brain has its own natural oscillations, and with alternating current you can either try to match the brain’s ongoing frequency by stimulating at that frequency, or disrupt it with alternate frequencies. Both the FWS and Alpha-Stim have patented frequencies (sometimes called “waveforms”), which the manufacturers say are the keys to how their devices modulate neuronal activity.
Regardless of the theoretical mechanism, the key issue is whether the contraptions actually help your patients. The company websites make prominent mention of the fact that their devices have been “cleared” by the FDA for depression and other conditions. But for those not steeped in FDA regulatory policy, this is a potentially misleading statement. The FDA “clears” a device, it does not mean that it has been “approved,” nor does it mean that the agency has reviewed any efficacy data. Instead, it means that the FDA has determined that the device is similar to other devices previously approved, often for indications completely different from the one being marketed. Nonetheless, the company websites states that the devices are proven to be effective for depression.
To dig deeper into these claims, we reviewed both the company-cited publications (http://bit.ly/1BhuAKg) and any other data we could find via standard databases such as PubMed. We were not able to find any published large-scale (meaning 200-400 subjects) randomized, controlled trials for the treatment of MDD. The data that both companies cite as demonstrating efficacy are derived from small studies (none have enrolled more than 70 patients, and most enrolled about 20) focusing on other conditions (such as substance use and anxiety disorders) or on mixed pathologies. Some of these studies included a depression measure, but none specifically targeted MDD. The company websites also cite anecdotal evidence from clinicians. However, none of these sources of evidence, either alone or in aggregate, rise to the level that we usually require when deciding on treatments for our patients. At best, they suggest possible efficacy.
What about tDCS? The evidence for tDCS in depression is stronger (we referenced an impressive result above), although smaller studies have been contradictory (http://bit.ly/1IfSqZk). No tDCS device has yet been cleared for any psychiatric indication, though patients can buy such devices very cheaply from the internet.
James Recht, MD is a psychiatrist based in Cambridge, MA.
Dr. Recht disclosed that he had no relevant relationships or financial interests in any commercial company pertaining to this educational activity.