This study evaluated the efficacy of nutritional management with and without fluoxetine (FLX) in anorexia nervosa diagnosed according to Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) criteria. Twenty-one patients, with a mean body mass index (BMI) of 15.21+/-2.33 kg/m(2), were treated with nutritional management and FLX at a mean dosage of 30.00+/-9.35 mg (pharmacological group); seventy-four patients, with a mean BMI of 14.24+/-2.16 kg/m(2), were treated only with nutritional management (nutritional group). Clinical evaluation was carried out under single-blind conditions at basal time and after 3, 6, and 12 months by a structured clinical interview, the Eating Disorder Interview based on Longitudinal Interval Follow-Up Evaluation (EDI-LIFE) and using a self-reported questionnaire, the Eating Disorder Inventory (EDI). BMI significantly increased in both the two treatment groups. In addition, the increase shown by the pharmacological group appeared near the beginning of treatment (i.e., at T1) and it was significantly higher than the increase shown by the nutritional group. Physical exercise showed a significant decrease in the pharmacological treatment group. On the other hand, fear of fatness and the scores of the subscales of the EDI significantly decreased in the nutritional treatment group. In terms of weight, the pharmacological group presented the higher amount of therapeutic success.