One of the worst symptoms that people with mental illness experience is the auditory hallucinations of schizophrenia. Sadly, drugs often fail to relieve patients of hallucinations, but help may be on the way in the form of a treatment called repetitive transcranial magnetic stimulation (rTMS). It has benefited a handful of patients considerably, Ralph Hoffman, M.D., a psychiatrist with Yale University School of Medicine and his colleagues reported in the March 25 Lancet.

"I think it is a very promising area," Sarah Lisanby, M.D., codirector of the brain-behavior clinic at Columbia University College of Physicians and Surgeons in New York City, told Psychiatric News.

Mark George, M.D., a psychiatrist with the Medical University of South Carolina in Charleston and a major figure in the TMS world, agrees: "It is extremely interesting and fascinating work and with enormous potential in terms of treatment. . . ."

In 1995 PET scans of patients with schizophrenia suggested where auditory hallucinations take place in the brain. They are the left temporoparietal cortex, the thalamus, the hippocampus, and the striatal regions. Such a finding didn’t come as a surprise to investigators in this area. The left temporoparietal cortex is known to be critical to the perception of spoken speech, the thalamus to be involved in auditory processing, and the hippocampus to engage in short-term memory functions necessary for tracking speech over a period of time. Only the striatal regions’ involvement in speech processing was in question.

Then during 1998 and 1999, researchers showed that repetitive transcranial magnetic stimulation (rTMS) could dampen the activity of nerves in a brain area directly stimulated by it.

So Hoffman and his colleagues wondered whether delivering rTMS to the left temporoparietal cortex could subdue the auditory hallucinations that patients with schizophrenia experience. To test that hypothesis, they decided to set up a small pilot study.

Although it would have been nice to see whether rTMS sent to the thalamus, hippocampus, and striatal regions might also muffle auditory hallucinations, Hoffman noted, they could not test this possibility since these areas lie too deep in the brain for rTMS to reach.

They recruited for their study 12 patients with schizophrenia who had been suffering from daily auditory hallucinations for at least six months. Three of the patients, in fact, had been plagued by voices nonstop. Each patient was given four days of active treatment followed by four days of sham (placebo) treatment, or vice versa, with several days separating each protocol. Neither patients nor clinical staff were told which patient was getting which treatment at which time.

Active treatment consisted of 40 minutes of rTMS to the left temporoparietal area over a four-day period—that is, four minutes the first day, eight minutes the second day, 12 minutes the third day, and 16 minutes the fourth day. Sham treatment consisted of stimulation to the same area of the scalp where rTMS had been delivered, but without brain penetration, and like rTMS, it was given for a total of 40 minutes over a four-day period—four minutes the first day, eight minutes the second day, and so forth.

At the start of the rTMS trial, patients were asked to describe their voices—frequency, loudness, content, number of voices, emotional distress, whatever was salient for them—and their description was arbitrarily given a score of 10. Then after each rTMS treatment they were asked to compare their hallucinations with how they had been at baseline. If their hallucinations were more severe, they gave them a number greater than 10, but if their hallucinations were less severe, they gave them a number less than 10. Then at the start of the sham trial the patients were once again asked to assess their hallucinations. Their hallucinations were once again assigned the number 10, and after they got each sham treatment they were asked to rate their hallucinations compared with how they had been at the start of the sham trial.

Eight of the 12 patients reported a reduction in hallucination severity after getting rTMS, but not after getting the sham treatment, the Yale researchers found. "Practically speaking," Hoffman told Psychiatric News, "what it really boiled down to is three out of the 12 went into total remission in terms of their hallucinations. And there were a number of patients whom we would consider had significant improvement, but they did not go into full remission. In four patients the improvements were negligible. . . ."

When asked whether rTMS could be widely used to reduce auditory hallucinations in patients with schizophrenia, Hoffman replied: "That is a big question, but a question that we are taking very seriously." In other words, they are now conducting a more extensive study than the one reported in The Lancet.

Whereas the patients reported in the published study received a total of 40 minutes of rTMS stimulation distributed over four days, in the current study patients are getting 132 minutes of rTMS distributed over nine days. Although the number of patients tested so far is small—six getting rTMS and nine a sham treatment—the data thus far suggest that about 75 percent of the patients getting rTMS are improving, compared with only 30 percent of those getting the sham, and the effects in the former seem to be relatively sustained for a number of months.

"So if that replicates in a large group of patients," Hoffman said, "we think it would be very promising as a potential clinical tool. . . . We are very excited about these data. We are eager to bring more patients into this more extensive protocol."

Although George is also enthusiastic about the research by Hoffman and his team—"It is good work from a good center, with a kind of double-blind design," he said—he cautions that their experimental results need to be replicated at another site before psychiatrists can firmly believe that they are real.

Before rTMS can be endorsed as a treatment for auditory hallucinations, he added, its mechanisms of action in combating such hallucinations need to be better understood as well.

"We know that it is likely depolarizing nerves right underneath the skull when it is being fired," he explained. "But the effects that Hoffman is seeing. . .are long term and tend to build up over several weeks, so that it is obviously some downstream effect of the TMS interacting with whatever is going on to produce hallucinations.

"[So] on the one hand, it is appropriate to be very excited; on the other hand, there is a huge mechanistic gap in between that cries out to be better understood."