Therapy vs. Medication for Depression, what do we learn from brain imaging?

Richard Friedman MD reviewed in the New York Times an article by Helen Mayberg, MD from JAMA Psychiatry . Dr. Mayberg started with the known premise that some patients do better with medication for depression, others do better with therapy (cognitive-behavioral therapy-- CBT). The patient got 12 weeks of escitalopram (lexapro) or 12 weeks of CBT. Only about 40% responded to either.

But there was a brain difference in the patients who responded to CBT vs. those for whom medication worked. As Dr. Friedman summarizes, "Patients who had low activity in a brain region called the anterior insula (PET scans) measured before treatment responded quite well to C.B.T. but poorly to Lexapro; conversely, those with high activity in this region had an excellent response to Lexapro, but did poorly with C.B.T."

Why might this be so? Here we move into the range of speculation, or educated guess. However, Friedman writes, "the insula is centrally involved in the capacity for emotional self-awareness, cognitive control and decision making, all of which are impaired by depression. Perhaps cognitive behavior therapy has a more powerful effect than an antidepressant in patients with an underactive insula because it teaches patients to control their emotionally disturbing thoughts in a way that an antidepressant cannot."

I might go on to suggest that overactivity in the insula responds more quickly to a chemical treatment, but underactivity requires stimulation such as thought and CBT to improve.

How does this relate to chronic pain and TMS? The insula and related structures appear to be involved in attention processes and cognitive control. The TMS mind-body healing model involves study, education, journaling, psychology and in so doing, may help strengthen the insula. Perhaps this explains, indirectly, the effectiveness of the treatment model.

More studies would be necessary and specifically, it would be fascinating to find 38 patients (65 were initially in the protocol) or even 20 and do a randomized trial of TMS treatment vs. a "placebo" (e.g. reading self-help books on stress, taking an active or inactive medication, pain-killers, etc.) and see what that study would show. Again, the challenge here is that patients must be willing to be randomized and therefore postpone TMS treatment while the protocol was proceeding... this could be 12 weeks as in this study or even 6. But, in my experience, TMS/chronic pain patients are not real willing to wait.

Anyway, this study presents more food for thought on the subject of individualized treatment for depression, based upon initial PET scan or f-MRI im

aging It also presents both the promises and challenges of doing TMS research with randomization and deferral of treatment for 1/2 the subjects. Finally it shows the important effects of both medication and CBT (therapy) on brain regions and clinical improvement.