Adolescents treated with neuroleptics are especially vulnerable to weight gain, with resulting psychological distress and noncompliance (1, 2).
Olanzapine, as well as other atypical antipsychotic medications, is associated with more weight gain than classical neuroleptics. It has been estimated among adults that 10 weeks of olanzapine treatment produces an average weight gain of 4.2 kg (3).
The mechanisms of neuroleptic weight gain are unknown. Weight gain results from an increase in caloric intake or a slowing in body metabolism, with a decrease in resting energy expenditure or a reduction in physical activity energy expenditure. We know of no reported studies on neuroleptic-induced changes in energy balance. In animals, these agents do not necessarily induce weight gain (4).
The aim of this pilot study was to examine caloric intake and energy expenditure as possible contributors to olanzapine-induced weight gain in adolescents with schizophrenia.
Male schizophrenic adolescent inpatients treated with olanzapine (N=10) or haloperidol (N=10) were studied. The diagnosis of schizophrenia was established according to DSM-IV criteria following a structured psychiatric interview, the Schedule for Affective Disorders and Schizophrenia for School-Age Children (5). Consensus among three senior child psychiatrists (D.G., G.Z., A.A.) was required. Patients receiving other medications that affect weight (such as lithium, antidepressants, or valproate) were excluded. The mean age of the olanzapine-treated group was 17.0 years (SD=1.6), the mean duration of time from first hospitalization was 33.6 months (SD=34.7), the mean duration of current hospitalization was 6.6 months (SD=8.1), and the mean accumulated lifetime hospitalization was 16.8 months (SD=16.6). The mean olanzapine dose was 14.0 mg/day (SD=4.1). Before the study, one patient had been drug naive, one had been receiving clomipramine, and eight had been receiving a neuroleptic medication other than olanzapine. The initial mean body weight for the olanzapine-treated patients was 73.8 kg (SD=23.8), and the mean body mass index was 24.5 kg/m2 (SD=5.9). The normal mean body mass index of male 17-year-olds in the United States is 21.2 (SD=4.5) (6); Israeli norms are unavailable. The mean washout period was 17.6 days (SD=27.8).
Ten schizophrenic male patients treated with haloperidol served as a comparison group. These patients did not differ from the olanzapine group in age (mean=17.0 years, SD=1.6), duration of time from first hospitalization (mean=32.3 months, SD=35.9), duration of current hospitalization (mean=5.8 months, SD=7.8), or accumulated lifetime hospitalization (mean=15.8 months SD=19.9). The mean haloperidol dose was 6.5 mg/day (SD=3.4).
The Geha Medical Health Center Review Board approved the study. All patients and their parents signed informed consent forms after the nature of the study had been fully explained to them.
Anthropometric data (height, weight, abdominal circumference, midarm circumference, and skinfold thickness over the triceps) were obtained for both patient groups before and after 4 weeks of treatment.
Caloric intake and energy expenditure were measured in the olanzapine-treated patients only. For caloric intake, a clinical dietician closely monitored food intake for 2 consecutive days. All food products and beverages consumed, as well as snacks, were weighed before and after the meal. The patients were allowed free choice of the type and amount of food. The total daily caloric intake, as well as carbohydrate, fat, and protein contents, were calculated from computerized food tables (Dietary Evaluation for Windows, Jerusalem).
Resting energy expenditure was measured by indirect calorimetry after an overnight fast with a Deltatrac II Metabolic Monitor (Datex-Ohmeda, Helsinki, Finland); accuracy was within 4%.
Daily energy expenditure was simultaneously measured over two consecutive days by means of an accelerometry monitor (Computer Science & Application, Shalimar, Fla.) positioned above the right hip and a heart rate monitor (Polar Electro Oy, Kempele, Finland). The accelerometry and heart rate monitors provided a minute-by-minute recording of movement and heart rate, respectively. Physical activity was defined as light, moderate, vigorous, or very vigorous by accelerometer and heart rate.
Differences between means were analyzed by using paired t tests.
As seen in t1, the body mass index of the patients with schizophrenia increased 5.3% over 4 weeks, which was a significant change.
A nonsignificant increase in body mass index occurred in the haloperidol group from 21.8 kg/m2 (SD=3.0) at baseline to 22.1 kg/m2 (SD=2.8) after 4 weeks (t=–1.7, df=9, p=0.12). Change in body mass index was significantly higher in the olanzapine than the haloperidol group (t=–3.2, df=18, p=0.005).
In the olanzapine-treated group, caloric intake over the 4 weeks significantly increased by 589 kcal/day or 27.7% (t1). There was no change in diet composition from baseline to week 4 in terms of carbohydrates (53.4% [SD=5.9%] and 56.2% [SD=5.4%], respectively), fats (29.9% [SD=8.7%] and 27.0% [SD=5.5%]), or proteins (16.7% [SD=4.3%] and 16.5% [SD=3.1%]). Abdominal circumference increased significantly (t1). No significant change was observed for the other measures of adiposity.
For the patients treated with olanzapine, no significant difference in resting energy expenditure was observed, and the two measurements of physical activity (heart rate and accelerometry) reflected very low levels both before and after treatment, with no changes.
It should be noted that because of the small group sizes, the power to find even large effect sizes (d=0.8) in our study was only 0.62.
Male schizophrenic adolescent inpatients treated with olanzapine gain substantial weight. The weight gain is mediated by an increase in caloric intake and persistent lack of physical activity. Body metabolism remains unchanged.
The mean weight gain in this study is much higher than that reported in adults treated with olanzapine (4).
Our findings support Stanton’s untested hypothesis that antipsychotic-induced weight gain is mediated by an increase in food intake (7).
Experimental data suggest that neuroleptics may stimulate food intake by blocking histaminergic and serotonergic receptors in the lateral hypothalamus. Effects on leptin and insulin have also been postulated (4). We did not find evidence for "carbohydrate craving"(8).
The estimated level of physical activity was low before and after treatment, despite intensive staff efforts to promote activity. Ninety percent of the subjects spent less than 10 continuous minutes in moderate activity during the day, which is much less than that reported for healthy boys (9).
The lack of significant weight gain by the haloperidol comparison group argues against a nonspecific "inpatient effect." Thus, adolescent inpatients receiving typical or no neuroleptics show lower weight gain (1–3, 10).
The limitations of the study are the small patient group size, the open-label nature of the study, and the variable washout period.
Research of the pathophysiology of olanzapine-induced weight gain should focus on the mechanisms of appetite and caloric intake regulation. Dietary programs aimed at reducing food intake and boosting physical activity are essential for all adolescents treated with olanzapine.
Received April 25, 2001; revision received Nov. 20, 2001; accepted Dec. 14, 2001. From the Behavioral Genetic Clinic, the Feinberg Child Study Center, the Nutrition and Dietetics Unit, and the Institute for Endocrinology and Diabetes, Schneider Children’s Medical Center of Israel; the Ribstein Center for Sport Medicine and Research, Wingate Institute for Physical Education and Sports, Netanya, Israel; and the Geha Psychiatric Center, Tel Aviv Medical School, Tel Aviv, Israel. Address reprint requests to Dr. Gothelf, Behavioral Genetic Clinic, Richard E. Feinberg Department of Psychiatry, Schneider Children’s Medical Center of Israel, Kaplan 14 St., P.O. Box 559, Petah Tiqwa 49202 Israel; firstname.lastname@example.org (e-mail). Supported by a feasibility grant from the Israeli Ministry of Health. The authors thank Hilda and Philippe Setton for their support to the behavioral genetics and schizophrenia research program.