CSF zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA 3-Endorphin and Dynorphin in Bulimia Nervosa Timothy D. Brewerton, M.D., R. Bruce Lydiard, Ph.D., M.D., Michele T . Laraia, R.N., M.S.N., Jennifer E. Shook, Ph.D., and James C. Ballenger, M.D. zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLK Objective: Preclinical and clinical evidence suggests that central opioid dysfunction may play a role in the pathophysiology of the eating disorders. In particular, endogenous opioids are known to regulate feeding behavior, mood, perception, and neuroendocrine function, all of which are disturbed in patients with eating disorders. Although low concentrations of CSF ^-endorphin have been reported in low-weight patients with anorexia nervosa, central opioid activity in normal-weight patients with bulimia nervosa has not been reported. The authors therefore measured CSF concentrations of ^-endorphin and dynorphin in drug-free female patients with DSM-III-R-Je/j'neii bulimia nervosa and normal comparison subjects. Method: After 4 days of a low monoamine diet and overnight bed rest, CSF was obtained (12-26 cc) from 11 women with bulimia and 17 normal comparison subjects (eight women and nine men). Results: The women with bulimia had significantly lower CSF concentrations of ^-en- dorphin than did the female comparison subjects. However, CSF concentrations of dynorphin were not significantly different in patients and female or male comparison subjects. ^-Endor- phin concentrations were inversely correlated with Beck Depression Inventory scores and the depression subscale of the Eating Disorders Inventory. Conclusions: These data support a role for central opiates in the mediation of the pathophysiology of the signs and symptoms of bulimia nervosa, although it is impossible to rule out the effects of depression on the results. (Am J Psychiatry 1992; 149:1086-1090) P rechnical and chnical evidence suggests that altera- tions in endogenous opioid function are associated with alterations in feeding (1-8), mood (8, 9), percep- tion (8, 10), and neuroendocrine function (11), all of which are characteristic of patients with eating disor- ders. In animals, opioid agonists increase and opioid antagonists decrease food intake. These effects appear to be mediated by several receptor types, including K,zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA \i, 5, and O receptors (12, 13). A disturbance of opioid function could also contribute to the hypercortisolism observed in anorexia nervosa (14, 15). Kaye and asso- ciates found that underweight patients with anorexia nervosa had significantly lower CSFzyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA P-endorphin con- centrations than healthy volunteers (16). These low CSF [3-endorphin concentrations normalized after weight restoration. Lower concentrations of plasma (3- endorphin have been reported in normal-weight pa- Received Jan. 30, 1991; revisions received Oct. 8 and Dec. 18,1991; accepted Jan. 10, 1992. From the Department of Psychiatry and Be- havioral Sciences, Medical University of South Carolina. Address re- print requests to Dr. Brewerton, Department of Psychiatry, Medical University of South Carolina, 171 Ashley Ave., Charleston, SC 29425. Copyright © 1992 American Psychiatric Association. tients with bulimia nervosa (17, 18). However, it is highly questionable whether plasma P-endorphin re- flects central P-endorphin function. CSF levels of P-en- dorphin, as well as of dynorphin, have not been re- ported in normal-weight patients with bulimia nervosa. We therefore studied CSF levels of p-endorphin and dynorphin in a group of normal-weight patients with bulimia nervosa and healthy comparison subjects. zyxwvutsrqponml METHOD Patients and comparison subjects were recruited through local newspaper advertisements and then inter- viewed with the Structured Clinical Interview for DSM- III (19), which was modified for DSM-III-R diagnosis. Patients with bulimia nervosa were included if they had experienced binge eating and purging at least three or more times per week for at least 6 months before the study. The mean weekly binge frequency was 8.2 times (SD=5.2) (range=3-19), and the mean weekly vomiting frequency was 4.7 times (SD=6.4) (range=0-19). Five patients had concurrent DSM-///-i?-defined major de- pression, and six did not. Comparison subjects were ex- 1086 Am ] Psychiatry 149:8, August 1992