January 10, 2008 — Metformin and lifestyle interventions, alone or in combination, were effective for antipsychotic-induced weight gain and abnormalities in insulin sensitivity, in a randomized, placebo-controlled trial of 128 adult patients with schizophrenia at a mental health institute in China.

The study is published in the January 9 issue of the Journal of the American Medical Association.

"The important findings of our study are that it is possible to attenuate antipsychotic-induced weight gain by using lifestyle intervention and metformin," Ren-Rong Wu, MD, of Central South University, Changsha, in Hunan, China, told Medscape Psychiatry. For patients with weight gain, lifestyle intervention plus metformin should be considered first, and if the patients cannot tolerate or they adhere poorly to lifestyle intervention, metformin alone should be considered, he added.

Second-generation atypical antipsychotic medications have been used increasingly to manage patients with a variety of psychotic disorders and severe behavioral disturbances, but there has been growing concern about their potentially serious adverse effects of weight gain, hyperlipidemia, and glucose intolerance, the group writes. A recent study reported that 78.8% of patients receiving atypical antipsychotic medications increased their baseline weight by more than 7%, they note. Clozapine and olanzapine produce the most weight gain, quetiapine and risperidone produce intermediate weight gain, and ziprasidone and aripiprazole produce the least weight gain, they write.

The mechanism underlying the weight gain might be associated with increased appetite or altered insulin sensitivity. Metformin, a drug that inhibits hepatic production of glucose, was found to reduce body weight in patients with type 2 diabetes and in obese individuals without diabetes.

Diabetes Medication and Lifestyle Changes

The researchers conducted a double-blind, placebo-controlled trial to test the efficacy of metformin with or without lifestyle intervention for preventing additional weight gain or causing weight loss in patients who had experienced substantial weight gain during the first year of treatment with atypical antipsychotic agents.

Participants were recruited from the schizophrenia outpatient clinic in a hospital in Changsha, China. The study group comprised 128 adult patients aged 18 to 45 years who had a first episode of schizophrenia, and, during their first year of treatment with an atypical antipsychotic agent — clozapine, olanzapine, risperidone, or sulpiride — had gained more than 10% of their predrug body weight.

Patients continued their atypical antipsychotic medication and were randomized to 12 weeks of 1 of 4 treatments: 750 mg/day of metformin plus lifestyle intervention, 750 mg/day of metformin alone, lifestyle intervention plus placebo, or placebo alone.

The lifestyle intervention consisted of 3 parts: a psychoeducational program, dietary intervention based on the American Heart Association step 2 diet (which allows less than 30% of total energy fraction from fat), and an individualized exercise program (such as walking 30 minutes a day).

The primary outcomes included changes in weight, body mass index (BMI), waist circumference, fasting glucose level, fasting insulin level, and insulin resistance index (IRI). Secondary outcomes included changes in Positive and Negative Symptom Scale scores and adverse effects.

Significant Decrease in Weight

After 12 weeks of treatment, patients in all groups except placebo had statistically significant decreases in mean weight, BMI, waist circumference, insulin levels, and IRI. These measurements continued to increase among patients in the placebo-alone group.

Lifestyle intervention plus metformin had the greatest effect on weight loss, and metformin alone was more effective than lifestyle intervention plus placebo in increasing insulin sensitivity and reversing weight gain in these patients with schizophrenia and significant weight gain from atypical antipsychotic agents.

Table. Change in Outcomes at 12 Weeks vs Baseline*

*BMI indicates body mass index; IRI, insulin resistance index.

All patients maintained relatively stable psychiatric improvement. There were no significant differences in the frequency and types of adverse events among the 4 treatment groups.

It is not known whether the improvement in weight and insulin sensitivity could be sustained beyond 12 weeks, what the effect of a different dose of metformin would be, or whether these results could be generalized to Western populations, the group notes.

"May Be Useful in Specific Patient Group"

Weight gain on atypical antipsychotics is indeed a serious problem, Kenneth E. Towbin, MD, from the Mood and Anxiety Disorders Program at the National Institute of Mental Health, in Bethesda, Maryland, told Medscape Psychiatry. "It's a kind of 2-hit phenomena," he added, where having schizophrenia or bipolar disease itself may lead to weight gain and subsequent problems such as cardiovascular disease, but treatment with atypical antipsychotics increases these problems, and risks accumulate over time.

Clinicians need to think about whether the drug dose is the lowest one needed to control the patient's psychotic symptoms, and lower maintenance doses might be needed, Dr. Towbin said. In addition, since about only half of patients receiving atypical antipsychotics will fall prey to metabolic syndrome, obesity, or type 2 diabetes, it is important to establish baseline levels and closely monitor patients who are treated with these drugs, especially when doses are increased.

This appears to be the first study to do a head-to-head comparison of lifestyle changes vs metformin, said Dr. Towbin. Other study strengths include the fact that it also looked at individuals who had both interventions as well as individuals who had neither one. The study was powered with sufficient numbers, the lifestyle interventions were clearly described, and the metformin doses were "reasonable," he added.

Follow-up studies are needed to see what happens over the long-term, since metformin poses certain risks such as problems of B₁₂ and folate deficiency. It would also be interesting to see if lifestyle changes can be maintained over a longer period, he noted.

"I think the clinical implications ... are that it is yet one more study that shows that metformin may be useful in that specific group of patients that shows very significant weight gain, or metabolic syndrome, or the beginning of type 2 diabetes while on neuroleptics," Dr. Towbin said. It does not imply that all patients should be started on metformin at the same time as they are started on atypical antipsychotics, he noted.

The research was supported by National Key Technologies Research and Development Program from the Ministry of Science and Technology of the People's Republic of China. One of the study authors has obtained funding. The remaining study authors have disclosed no relevant financial relationships.

JAMA. 2008;299:185-193.

Clinical Context

Study Highlights

• Included were outpatient adults aged 18 to 45 years from 1 mental health institute with a first psychotic episode of schizophrenia diagnosed with the Structured Clinical Interview of the Diagnostic and Statistical Manual of Mential Disorders, Fourth Edition, Axis I Disorders, Clinical Version, who had gained at least 10% of their body weight since before the use of atypical antipsychotic drugs.
• Patients had used only 1 of 4 atypical antipsychotic agents (clozapine, olanzapine, risperidone, or sulpiride), had shown stable improvement in 1 year with dose change of not more than 25% within 3 months, and had a caregiver to ensure adherence to protocol.
• Excluded were those with liver or renal dysfunction, cardiovascular disease, diabetes, substance abuse, other psychiatric diagnoses, or those who were not able to perform lifestyle modification.
• The atypical antipsychotic medications remained at a fixed dose during the study, and adherence to protocol was defined as taking 80% or more of assigned medication.
• The lifestyle intervention included psychoeducation and dietary and exercise programs.
• Psychoeducation was administered at baseline, and at weeks 4, 8, and 12; the dietary intervention used the American Heart Association step 2 diet; and the exercise program consisted of treadmill sessions 7 times a week for 30 minutes per session to attain 70% of heart rate reserve.
• Primary outcomes were change in body weight, BMI, waist circumference, fasting glucose levels, insulin levels, and IRI calculated by the formula of homeostasis assessment for insulin resistance model.
• Psychiatric symptoms and adverse effects were monitored.
• Follow-up visits occurred at weeks 4, 8, and 12.
• Of 128 patients, 32 each were assigned to 4 groups: lifestyle intervention with or without metformin 750 mg daily or identical-appearing placebo.
• Study completion was more than 90% in each group, and adherence was similar.
• Mean age of the patients was 26 years, 50% were men, 8% were current smokers, mean duration of schizophrenia was 9 months, 7% had a family history of diabetes, and a similar number were taking each of the 4 atypical antipsychotic medications.
• Mean baseline BMI was 24.5 kg/m², weight was 64 kg, and fasting glucose level was 95 mg/dL.
• The lifestyle-plus-metformin group had the greatest decrease in weight (7.3% or 4.7 kg), BMI (mean, 1.8 kg/m²), and waist circumference at each follow-up; the lifestyle-plus-placebo group had decreases in weight (mean, 1.4 kg or 2.2%) and BMI (mean, 0.5 kg/m²) at each follow-up, but waist circumference decreased only at the first visit and not at weeks 8 and 12.
• The metformin-alone group had a mean BMI reduction of 1.2 kg/m².
• The placebo-alone group gained a mean of 3.1 kg vs baseline (4.8%), and BMI increased by 1.2 kg/m².
• Waist circumference decreased by 2.0 cm in the combined and 1.3 cm in the metformin-alone groups, and it increased by 2.2 cm in the placebo-alone group.
• IRI decreased by 3.6 in the combined, 1.0 in the lifestyle-plus-placebo, and 3.5 in the metformin-alone groups, and it increased by 0.4 in the placebo-alone group.
• Fasting glucose levels decreased in all intervention groups but not in the placebo-alone group.
• Among the 4 groups, the combined group had superior outcomes for all measures vs the other 3 groups.
• There were no significant correlations between initial weight and weight changes at 12 weeks for any of the groups.
• There were no significant differences in adverse effects among the 4 groups, and no hypoglycemia was reported.

Pearls for Practice

• The greatest benefit against the adverse metabolic effects of atypical antipsychotic use in patients with schizophrenia is seen with a combination of lifestyle intervention and metformin.
• The use of lifestyle intervention with and without metformin for 12 weeks is associated with improvements in metabolic indices of weight, BMI, fasting glucose levels, and IRI among patients with schizophrenia who are receiving atypical antipsychotic drugs.

CME Test