James Recht, MD
Clinical Instructor in Psychiatry, Harvard Medical School
Dr. Recht has disclosed that he has no relevant financial or other interests in any commercial companies pertaining to this educational activity.
If your practice is anything like mine, then up to half of your patients with depression are at best partially responsive to the treatment you provide. You may have nodded sympathetically as you read the results of the 2007 STAR*D trials, showing that after two complete trials of standard pharmacotherapy, only half the study participants were truly symptom free. Assuming that we are not making systematic (or lazy) mistakes in our work, it’s only natural that we would look to the evolving field of brain stimulation in hopes of finding more effective treatment options.
In this article, the focus will be restricted to the following non-invasive technologies: transcranial magnetic stimulation, or TMS; magnetic seizure therapy (MST); transcutaneous vagal nerve stimulation (t-VNS); and transcranial direct current stimulation (tDCS). For clarity’s sake, we will limit this review to the treatment of refractory depression. (Other conditions undergoing study include OCD, panic and other anxiety disorders, and even some forms of psychosis.)
Transcranial Magnetic Stimulation (TMS) Since early in this century, TMS has been the object of the most eager (and wishful) anticipation. It remains the single most studied non-invasive therapy, and the only one to have received even limited FDA approval for treatment of depression.
By way of brief review, TMS works by sending pulses of magnetic fields through a metal coil that is placed close to the skull. In most studies, treatment involves using high frequency (10 per second or 10 Hz) pulses to stimulate neurons in and around the left dorsolateral prefrontal cortex. The treatment is time consuming: each session lasts from 30 to 45 minutes; most protocols involve at least four to five sessions each week for six weeks or more.
When TCPR first looked at TMS in late 2005, it appeared to show promise, particularly as a treatment for refractory depression. When we returned to the subject in 2008, however, things were not looking so good—in fact, as a result of disappointing results published in a large and (until then) eagerly anticipated 2007 study, it appeared to us that it was highly unlikely the technology would receive FDA approval. Indeed, the agency’s 2007 review panel consensus was that its “efficacy was not established.”
After that, TMS’s regulatory road remained a bumpy one. In 2008, the FDA issued a qualified approval of the device for “depression that has not responded to one prior course of antidepressant therapy.” The agency’s decision appeared to have been based largely on a post hoc re-analysis of previously published data, and was met with skepticism (see, for example, TCPR, January 2009). The widely-respected journal Neuropsychopharmacology published a highly critical letter from physicians at the consumer watchdog group Public Citizen, expressing their “concern...that FDA has cleared this device, particularly if patients are diverted from effective therapies such as antidepressant medications.” A particularly damning appraisal of a device intended to treat antidepressant-resistant disorders!
The news since 2010 has not been altogether bad for TMS. Late 2010 saw the publication of data from a large prospective study intended to answer critics’ concerns. Crucially, the study included patients with depression defined as “moderately resistant.” For inclusion, subjects were required to have had “insufficient clinical benefit from one to four adequate medication trials or intolerance to three trials.” The results? Just more than 14% of TMS patients achieved full remission, compared with about 5% of sham treatment recipients. Though perhaps unlikely to convince skeptics, this carefully designed study did show a significant (p<.02) difference between TMS and sham treatment (George MS et al, Arch Gen Psychiatry 2010;67(5):507–516).
Researchers have developed at least two specific refinements to the TMS technique: in the duration and frequency of the magnetic pulses, and in the design of the magnetic coil. One example of the former, known as Theta-Burst Stimulation (TBS), uses bursts of high frequency stimulation (5 Hz, the frequency designated “theta” in EEG nomenclature). TBS has been used in several small studies; in one case series involving seven patients, after three weeks of treatment, three patients remitted completely, and another two showed reductions in Hamilton Depression Rating Scale scores of more than 50%. The treatment was described as well tolerated (Holzer M and Padberg F, Brain Stimul 2010;3(3):181–183). Other researchers have claimed significant improvements in efficacy using a more powerful “H coil” that can stimulate deeper subcortical brain structures. Here too, results from a number of small, open trials and case series have been published, but to date no randomized, double-blind studies are available.
Magnetic Seizure Therapy (MST) This technology can be thought of as a sort of hybrid of TMS and ECT. MST uses magnetic coils identical to the ones used in TMS, but at sufficiently higher frequency and intensity to purposely induce one seizure per treatment session. In most other respects, including the number and scheduling of treatments (typically two to three per week for four to six weeks) and the need for anesthesia and muscle relaxation, MST very closely resembles ECT.
So why use MST? Using electrical impulses to cause seizures, as in ECT, is not ideal. Bone and spinal fluid have impedance—that is, they resist the current delivered by ECT, making it quite difficult to deliver ECT in a consistent way. MST uses magnetic impulses, and since our bodies are essentially “transparent” to this form of energy, it may allow more consistent and precisely directed impulses than are possible with ECT. As a result, MST might result in less acute postictal confusion, faster recovery times, and a lower risk of cognitive and memory impairment—the chief bugaboos associated with ECT.
How MST compares with ECT clinically is not yet clear. A handful of case reports have shown good clinical responses, and two small (a total of 30 patients) partially randomized studies that compared MST with ECT seem to support the claims of better tolerability and faster recovery times. Reportedly, a multi-center, randomized, controlled study is currently ongoing. So stay tuned.
Transcutaneous Vagal Nerve Stimulation (t-VNS) Despite what many felt was a lukewarm evidence base (see for example, TCPR, January 2006), VNS was approved by the FDA in 2005 for treatment-resistant depression. Its invasive nature (it requires surgical implantation of a device about the size of an Oreo cookie); the potential for messy technical problems (for example, battery malfunctions); and side effects (including cough, hoarseness, sore throat and headache) have resulted in very limited clinical use.
Transcutaneous VNS may overcome these limitations. It uses a pocket-sized, battery-operated “stimulation unit” attached to a wired electrode that your patient wears like an earphone. Impulses are transferred via the ear electrode through the skin to the auricular branch of the vagus nerve. It looks a lot like an old-fashioned hearing aid. And it requires three to four one-hour “stimulation sessions” each day. OK, but does it work? Answer: only the most preliminary studies have been done so far, with even fewer patients than in the case of MST described previously.
Transcranial Direct Current Stimulation (tDCS) In tDCS, the clinician straps two electrodes—one at each temple—onto the patient’s head with an elastic band, and then delivers a relatively weak, direct (as opposed to alternating) electrical current for 20 to 30 minutes. In the small number of controlled trials that have been published, treatment is delivered daily, for at least two to three weeks. Side effects appear to be limited to itchy scalp and minor skin irritation. However, following some promising results from two smaller studies in 2006 and 2008 and at least two blinded, sham-controlled studies (totaling just over 100 patients) have shown no significant differences in clinical response (Fregni F et al, J Bipolar Disorders 2006;8(2):203–204; Paulo S et al, Int J Neuropsychopharm 2008;11:249–254).
TCPR Verdict: The verdict on brain devices for treatment of depression? Still lots of sizzle, but we could use more beef.