RESEARCH AND PROFESSIONAL BRES main meals or between-meal snacks or in the percentage of energy intake from car- bohydrates, fats and proteins. Restrained Eating Score An increase in the restrained eating score was apparent only in the placebo-treated group: the percentage with very high scores shifted from only 25% at the onset of the studyto 89% after 6 months (P<.01) andto 67% after 12 months (not significant). DISCUSSION According to Wurtman and Wurtman (12), intake at main meals does not explain the obese status of carbohydrate-craving sub- jects. In contrast, carbohydrate-rich snacks that subjects tend to eat in the late after- noon and evening to ameliorate feelings of depression and stress are responsible for 25% of the total daily energy intake and for 33% of the total carbohydrate intake. Wurtman et al (9) found that dexfenflur- amine invariably decreased the intake of carbohydrate-rich snacks in these subjects who had free choice of foods (9). In our study, snacks also accounted for 28% of the total energy intake and ac- counted for 33% of the total carbohydrate intake. However, the snack times reported by Wurtman et al (9,12) do not conform to Dutch habits. We found a significant (P<.001) reduc- tion in the energy, fat, and protein content in the main dishes of both of our study groups. Main dishes and between-meal snacks were lower in energy and between- meal snacks were lower in fat and protein in the dexfenfluramine-treated group. Probably, compliance with this restricted diet, supplemented by adherence to the general rule of high-fiber and low-fat in- take, overruled the effects of dexfenflur- amine on food selection. Although not reflected in the weight losses, the dexfenfluramine-treated pa- tients complied better with the suggested diet. The similarity in weight losses expe- rienced by the two study groups may be explained by small numbers and by the strict medical and dietary guidance, which motivated every patient and so enhanced the placebo response. Apparently, for a similar weight benefit, dexfenfluramine- treated patients experienced less every- day preoccupation and stress while diet- ing--afinding that agrees with similar find- ings in rats (6,13,14). APPLICATIONS Our study failed to demonstrate that dexfenfluramine had an effect on diet com- position of patients who did not crave carbohydrates, were under strict dietary guidance, and did not have free food choices. However, their enhanced dietary compliance and reduced preoccupation and stress while dieting should challenge dietitians to define groups of patients with eating behavior characteristics and food preferences who may benefit from pro- longed dexfenfluramine therapy. References 1. Garrow JS. Treatment of obesity. Lancet. 1992; 340:409-413. 2. Guy-Grand B, Crepaldi G, Lefebre P, Turner P, Apfelbaum M, Gries A. International trial of long-term dexfenfluramine in obesity. Lancet. 1989; 2:1142-1144. 3. Orthen-Gambill N, Karanek RB. Differential effects of amphetamine and fenfluramine on dietary self-selection in rats. Pharmacol BiochemBehav. 1982; 16:303-309. 4. Fantino M, Faion F, Rolland Y. Effect of dexfenfluramine onbodyweight set-point; study in the rat with hoarding behaviour. Appetite. 1987; 7(suppl):115-126. 5. Stunkard AJ. Anorectic agents lower a body weight set-point. Life Sci. 1982; 30:2043-2055. 6. Lacour F, Ravel D, Kerdelhu B, Espinal J, Duhault J. Stress-related overeating; d- fenfluramine and neuroendocrine responses in the aging rat. In: Ailhaud G, Guy-Grand B, Lafontan M, Ricquier D, eds. Obesity in Eu- rope 91. London, England: John Libbey; 1991:55-56. 7. NathanC. Dexfenfluramineandbodyweight in overweight patients. In: Vague J, Bj6rntorp P, Guy-Grand B, Rebuff6-Scrive M, Vague P, eds. Metabolic Complications of Human Obesities. Amsterdam, The Netherlands: Excerpta Medica; 1985:229-234. 8. Wurtman RJ, Wurtman JJ. Carbohydrate craving, obesity and brain serotonin. Appetite. 1986; 7(suppl):99-103. 9. Wurtman J, Wurtman R, Reynolds S, Tsay R, Chew B. Fenfluramine suppresses snack intake among carbohydrate cravers but not among noncarbohydrate cravers. Int JEating Disord. 1987; 6:687-699. 10. Strien van T, Frijters JER, Bergers GPA, Defares PB. The Dutch Eating Behaviour Ques- tionnaire (DEBQ) for assessment of restrained, emotional and external eating behaviour. Int J Eating Disord. 1986; 5:295-315. 11. Strien van T. Eating Behaviour, Person- ality Traits and Body Mass. Lisse, The Neth- erlands: Swets & Zeitlinger; 1986. 12. WurtmanJJ, Wurtman RJ. d-Fenfluramine selectively decreases carbohydrate but not pro- tein intake in obese subjects. IntMonogr Obes. 1983; 1:79-84. 13. Brindley DN, Saxi in J, Shadidullah H, ArmstrongM, Margiapane EH. Dexfenfluramine: relationship between decrease of the body weight setpoint and metabolic effects. In: Vague J, Bj6rntorp P, Guy-Grand B, Rebuffb-Scrive M, Vague P, eds. Metabolic Complications of Human Obesities. Amsterdam, The Nether- lands: Excerpta Medica; 1985:207-217. 14. Garattini S, Samanin R. d-Fenfluramine and salbutamol: two drugs causing anorexia through different neurochemical mechanisms. Int Monogr Obes. 1983; 1:151-157. Longitudinal changes in zinc status in untrained men: Effects of two different 12-week exercise training programs and zinc supplementation MELINDA M. MANORE, PhD, RD; JULIE M. HELLEKSEN, MS, RD; JEAN MERKEL, MS, RD; JAMES S. SKINNER, PhD inc is essential to metabolic pathways used during physical activity (1). Ath- letes reportedly have altered zinc me- tabolism (2-5) and may be at risk for poor zinc status. This hypothesis is based on research showing that dietary zinc intakes may be low in athletes who emphasize a high-carbohydrate diet and restrict intake of flesh food (6,7) and that exercise can increase zinc losses through sweat (8) and urine (3,9). Researchers also report lower plasma and serum zinc levels in endurance athletes vs sedentary control subjects (2- 4,10) and in individuals after acute exer- cise (9,11). However, studies on the effect of exercise training or strenuous daily ex- At the time of the study, all of the authors were located at Arizona State Univer- sity, Tempe, AZ 85287-2502. M. M. Manore (corresponding author) was an associate professor of nutrition and J M. Helleksen and J Merkel were gradu- ate students in the Food and Nutrition Laboratory, Department of Family Re- sources and Human Development, and J. S. Skinner was aprofessorofexercise science in the Department of Exercise and Physical Education. Currently, J M. Helleksen is a clinical dietitian at Humana Hospital, Brandon, Fla; J. Merkel is a diabetes educator with CIGNA Health Plan of Arizona, Phoenix, Ariz. JOURNAL OF THE AMERICAN DIETETIC ASSOCIATION / 1165