Magnetic resonance imaging of the brains of patients with schizophrenia suggests that while treatment with an antipsychotic drug may alleviate symptoms, it may also contribute to reductions in brain volume.

"It is possible that, although antipsychotics relieve psychosis and its attendant suffering, these drugs may not arrest the pathophysiological processes underlying schizophrenia and may even aggravate progressive brain tissue volume reductions," said Beng-Choon Ho, M.D., an associate professor of psychiatry at the University of Iowa Carver College of Medicine, and colleagues in the February Archives of General Psychiatry.

The reduction in brain volume in schizophrenia occurs not because brain cells die off but rather because dendrites shrink and dendritic spines shrink, causing shrinkage in the synaptic connections in the cortex, explained Jeffrey Lieberman, M.D., another schizophrenia researcher not affiliated with the Iowa study.

"That's why in people with schizophrenia, thinking becomes more stereotyped, routinized, and concrete," said Lieberman, a professor and chair of the Department of Psychiatry at Columbia University College of Physicians and Surgeons and director of the New York State Psychiatric Institute.

"They don't have the elaborate richness of synaptic connections to allow for the cognitive and intellectual processes to occur," he said in an interview with Psychiatric News.

To see what effect antipsychotic drug treatment might have on the brain, the Iowa researchers performed a series of magnetic resonance imaging scans on 211 patients at baseline and an average of three years later. Most patients (n=139) had an additional scan about three years after that, and some had fourth (n=82) and fifth (n=31) scans at similar intervals.

During the course of the study, patients were given "treatment as usual" in the community by their own physicians using typical or atypical antipsychotic drugs or clozapine. Ho and colleagues controlled for dose, illness severity, and substance use.

They also used alternative ways of accounting for any treatment variations.

"Since our findings do not change when we analyze the data using different measures of illness severity, it suggests that our results are fairly robust," Ho told Psychiatric News. "The potential confounding influence from more severely ill patients needing higher doses of antipsychotics becomes less likely."

The longer the patients were followed up, the greater their changes in brain volume. As time passed, they lost total cerebral tissue, gray matter, and subcortical tissue, but saw enlargement of parietal white matter, lateral ventricles, and sulcal cerebrospinal fluid.

Brain volume changed not only over time but also in parallel with antipsychotic medication treatment. Patients who got the highest mean daily doses of antipsychotics had smaller frontal gray matter volumes and larger lateral ventricles.

"Antipsychotic treatment had a significant influence on brain volumes," even after adjusting for illness duration, illness severity, and substance misuse, the researchers noted.

The meaning of reduced brain volumes associated with antipsychotic drug use among patients with schizophrenia is unclear, they pointed out.

Some prior research seems to indicate that changes visible after treatment are similar to those attributable to the effects of the disease. Other studies disagree.

"Medications may reduce the volume in a healthy brain [as seen in primate studies], but disease is also a contributor," said Lieberman. "If you compared treated and untreated patients, the untreated patients might show a greater volume reduction."

Ethical considerations prevented including an untreated control group in the Iowa study.

However, data are available from brain autopsies of untreated patients done in the first half of the 20th century, prior to the introduction of effective antipsychotic medications, said Lieberman. Those studies documented a variety of changes in the brains of people with schizophrenia—aplasia, atrophy, hydrocephalus, and others, he noted. "So it is likely the disease itself that contributes to those neuropathologies," he said.

An editorial by David Lewis, M.D., of the University of Pittsburgh accompanying the report of Ho's study offers another experimental alternative based on the increasing use of antipsychotics for nonschizophrenia diagnoses such as mood disorders. A long-term comparison of brain volume changes in patients with and without exposure to those drugs might reveal whether the illness of schizophrenia or the antipsychotics taken to treat it was responsible for the brain changes Ho and colleagues reported.

"[A] positive finding would suggest a conserved medication effect that is independent of diagnosis or underlying disease," said Lewis. The clinical significance of brain volume changes is also open for debate, he said. The changes may reflect a positive result of therapy and not an adverse effect.

Ho suggested that different pathways initiated by antipsychotic medications lead separately to alleviation of symptoms and brain volume changes.

He also pointed out that the study reports only an association but not a causal relationship between higher antipsychotic dosages and greater brain volume reductions. However, that association, along with animal studies by others, suggests that antipsychotic medication is a mediating factor.

"Our finding does not mean that the disease itself is not a contributing factor to brain anatomical differences in schizophrenia," he said. "There are likely multiple causes for smaller brain volumes in people with schizophrenia—genes, environmental factors, antipsychotics, street drugs, lifestyle, and so on."

Finally, Lieberman raised concerns that some patients or practitioners might draw unwarranted conclusions about the study's findings and decide to stop treatment.

Ho and colleagues also suggested avoiding precipitous changes in the use of antipsychotics to treat schizophrenia.

"We do not advocate that people stop taking their medications, because many patients with schizophrenia benefit from antipsychotic medications and from remaining on treatment," he said. He does, however, recommend using the lowest doses necessary to control symptoms.

Ho and colleagues recommended additional caution when antipsychotics are prescribed for patients diagnosed with disorders other than schizophrenia, such as bipolar or depressive disorders.

The study was supported by NIMH.

An abstract of "Long-Term Antipsychotic Treatment and Brain Volumes" is posted at <http://archpsyc.ama-assn.org/cgi/content/abstract/68/2/128>.