Peering inside depressed people's heads, California scientists got a new glimpse of the power of a pill.
Actually, three pills: two types of antidepressant, plus a "dummy" pill, or placebo. The scientists wanted to know - without having to wait weeks for patients to report that they're feeling better - whether the effects of antidepressants could be detected quickly in the brain's electrical activity.
Yes, they could, the study found. Yet effects were also seen among patients who improved on the placebo, the scientists report this month in the American Journal of Psychiatry.
"We were just looking at the placebo group as a control group," says Dr. Andrew Leuchter, a psychiatry professor at the University of California, Los Angeles, who led the study. "It was really quite a surprise to us when we could see that they had significant changes in brain functions."
In addition, those changes were very different from the changes seen in people who responded to medication. Yet the two groups had similar symptom reduction over time.
In both cases, the difference was most significant in the brain's prefrontal region, mainly in the right hemisphere. People responding to the antidepressants had a decrease in activity there; people responding to the placebo had an increase.
The findings highlight the importance of comparing the patient and doctors' analyses of symptoms with biological brain data, says Diego Pizzagalli, a psychology and neuroscience research fellow at the University of Wisconsin-Madison. Although both are important, the two types of information are dramatically distinct in this study, he notes. "I think it would be a mistake to just focus on one of the two."
Scientists so far have had a hard time gauging the effects of placebo on brain activity, says Dr. Wayne Drevets, of the National Institute of Mental Health. But in the UCLA study, the prefrontal difference "was a pretty impressive result," he says. "Statistically, it looks kind of compelling."
Much of the function of the prefrontal cortex is to try to inhibit expression or emotional thought, says Drevets, who specializes in brain imaging and depression.
The UCLA researchers compared a group given the antidepressant Effexor with a group given a placebo. They also compared a group given Prozac with another placebo group. None of the 51 patients or the scientists knew who was taking what.
Over eight weeks, the team measured electrical activity in the subjects' brains five times, and monitored whether depression symptoms were improving. In the end, 52 percent of the people who had taken antidepressants felt better, along with 38 percent on the placebo.
Then the scientists noted the brain activity trends among the different groups.
"Dogma has been that placebo is no treatment," says Leuchter. "Placebo is not no treatment."
The new research tries to address an important question, says Dr. Eric Nestler, chairman of psychiatry at the University of Texas Southwestern Medical Center at Dallas. "You take two depressed patients and give them medication: One gets better just by placebo response, the other really needs the active medication to get better. The question is why," he says. "The related question is are they getting better in the same way?"
Yet the caveats to such research are huge - for one thing, "we don't even know if the people have the same disease," Nestler notes.
"Out of 100 depressed patients, it's quite likely that there are 10 different diseases contained in what we call depression," he says. "We're not at the point yet that we can differentiate depression" the same way doctors can tell whether a cough is caused by pneumonia or a cold.
Pizzagalli says the new study may help address this very question. "The point is that possibly neuroscience will be able one day to identify subtypes of depression with very specific functional impairment in the brain," he says.
The UCLA researchers also found that among those whose symptoms improved, people didn't report feeling much better till about two weeks after starting on the pills. Yet changes in brain activity registered in just a few days in the medication responders, Leuchter says - vs. one to two weeks in placebo responders.
His study showed that the placebo effect "seems to be an effect that grows over time."
But factors other than an inert pill may play a role, Leuchter says. A patient who has been depressed for months, then resolves to seek help, may be primed for treatment. Patients in a study also receive personal attention that could improve their mood - for the duration of the study, anyway. And they have the study's schedule and other expectations to meet.
"Whatever process is going on may be additive over time, and that may be why these processes have a certain latency that you don't see with medication," Leuchter says.
Still, he says, the study strongly suggests that among depression patients, there are different brain pathways to improvement. He wonders whether the pathways would be similar in people given psychotherapy vs. a placebo form of therapy - or whether results from the psychotherapy group would look more like the group who responded to placebo in the current study.
Leuchter also notes that patients with many other illnesses respond to placebo. "If someone gets better on a placebo for their hypertension or asthma or chronic pain, would we see the same changes in the prefrontal regions of the brain?" he asks.
In the current study, the effects were not lasting among those who responded to the placebo, Leuchter says. "At the end of eight weeks, we told people what they were getting, and within a month of the end of the study, most of the placebo responders required medication. I don't want people to think we don't need medication anymore."
Indeed, that would be the wrong message, says UT Southwestern's Nestler. "People who are seriously ill with depression really need treatment."
The UCLA study involved patients who met the diagnostic guidelines for major depression but whose degree of illness was not severe, Nestler points out. "So the placebo responses they're seeing are high, but they're placebo responses for people with mild depression.
"The kind of person who can't get out of bed in the morning, is despondent, can't work, can't do anything - that person's not going to show a placebo response."
Placebo response is "a tougher issue to tease out in depression," agrees Drevets of the NIMH. "If you look at a depressed patient today and again in three or four weeks depressive symptoms fluctuate. They're intermittent."
The fluctuations can be dramatic, he says. "It's not like cancer where the tumor is persistent over time."
Because depression has varied forms, Drevets takes extra steps in his brain imaging studies to find subjects whose disease is more likely to be biologically based. For instance, he considers whether depression or manic depression runs in subjects' families, and whether bouts of depression have recurred.
Other studies often don't take such steps, "so they get a much more heterogeneous group of people. And this just increases the placebo effect by washing out the true differences," Drevets says. "I get much more dramatic responses to medicines. My bet is because I'm getting such a robust medication effect that they would also not have so much of a placebo effect."
Still, the new study is novel because it examined the brain's reaction to treatment over time, says Pizzagalli. The repeated glimpses of brain activity were possible because the research used a technique called quantitative electroencephalography, or QEEG - a relatively inexpensive, noninvasive method that Pizzagalli expects will continue to yield important data. QEEG takes the electrical brain data from a standard EEG and digitizes it, Leuchter says; from this, researchers can better quantify the intensity of the electrical signal.
With a technique such as PET (positron emission tomography) scanning, Pizzagalli says, "it's not always possible to make assessments (over time), for health reasons, because you use a radioactive tracer."
In addition, at a time when some studies are questioning whether the placebo effect even exists, the new research shows that placebo has a "very important and remarkable affect" on the brain, Pizzagalli says. "It shows the power, so to speak, of the mind for affecting biology."
Such research may someday lead to an important diagnostic tool, Nestler says.
"If there's a high school student, for example, who's having a tough time with a girlfriend, do they have bad depression, or is it one of those lifetime things that takes time and experience to get used to?" he says. "We have no way to distinguish that right now."
---
(c) 2002, The Dallas Morning News.