Abstract

Section 1

Section 2

Section 3

Section 4

Section 5

Discussion
Board

The adverse effects of overeating and a sedentary life style are well documented (1). In rats, the single most effective way to prolong life is by significant and chronic under nutrition (without malnutrition)(2). Voluntary wheel-running, on the other hand, has a much more modest, but nonetheless, beneficial effect on longevity (3). Not surprisingly, there is an important public health campaign to promote weight loss and "get physical" (4). However, preoccupations with weight loss and excessive exercise are risk factors for anorexia nervosa (5) and the female athlete triad.

The female athlete triad is associated with disordered eating, amenorrhea, and osteoporosis (6). The American College of Sports Medicine emphasizes the need to recognize the triad in active females. It also emphasizes that when one component is present the physician should look for the other components (6). Various authors suggest that the initiating factor for the triad is disordered eating (7). An important type of eating disorder is anorexia nervosa, which has one of the highest mortality rates of any psychiatric disorder (8). Competitive female athletes are at increased risk for anorexia nervosa (e.g., 6,9) and numerous authors argue that excess physical activity is a risk factor for anorexia nervosa (5,9,10,11). This a complex disorder with multiple predisposing, precipitating and perpetuating factors (12). Animal modeling is one way to focus on specific aspects of the problem.

The most popular animal model of anorexia nervosa is the weight-loss syndrome produced by freely induced wheel-running in rats maintained on a moderate food deprivation schedule (see 13,14,15). Compared to animals that are exposed to either variable alone, the simultaneous imposition of both variables causes profound weight loss. As body weight progressively declines, running progressively increases. We argued the syndrome models two features of anorexia nervosa: hyperactivity and freely induced weight loss (16). Others argued that is also associated with an "activity anorexia"(11). However, the reduction in food intake is at least partly experimenter induced and animals exposed to the syndrome eat the same amount as animals comparably weight-reduced by simple food restriction (16). On the basis of these considerations, we argued that the term "anorexia" should not be applied and proposed that the syndrome be called an exercise-stress syndrome (16).

Yet other data we collected demonstrated that social isolation from weaning until adulthood increases vulnerability to the syndrome (17); that there are changes in insulin-like growth factor II gene expression (18); that chronic paraventricular hypothalamic alpha-2 adrenergic stimulation increases vulnerability to the syndrome (19); that water deprivation combined with free access to running wheels can produce a similar hyperactivity and weight-loss phenomenon (20); and that so-called glucoprivic stimulation paradoxically reduces appetite in animals subjected to the syndrome as it does in anorexics (16). Other data we collected suggest that, while beta-endorphin and dynorphin changes occur (21), they are not critically involved in the syndrome (see 13). Finally, we generated data several years ago implicating a deficiency in serotonin in the syndrome (22,23,24). Other investigators have made similar conclusions (25,26). The implication that a deficiency in brain serotonin is causally related to anorexia nervosa is paradoxical if the central substrate for the disorder is viewed at the level of the hypothalamus, where the acute administration of serotonin drugs suppresses appetite. However, there is evidence for serotonin abnormalities above the level of the hypothalamus in anorexia nervosa (27). Hence, the eating disorder is itself more than an eating disorder.

From this perspective, the central substrate for anorexia nervosa may importantly relate to the substrate for obsessive-compulsive disorder (OCD) (28). Anorexia nervosa has a variety of similarities with OCD and some authors describe it as an OCD spectrum disorder (29). Others argue that physical activity, starvation, and obsessive compulsiveness are intimately related (9). This argument has clear implications not only for anorexia nervosa but also for the female athlete triad. Serotonin reuptake inhibitors (SRIs) benefit OCD but not other antidepressant drugs (30). This raises the possibility that anorexia nervosa may also be successfully treated with SRIs. Beneficial effects have been reported in some (28) but not all (31) studies. More research is clearly needed on the interrelationship between OCD and anorexia nervosa.

One model well suited for this research is the exercise-stress syndrome. Not only is it a popular model of anorexia nervosa, it is also a model of OCD. This is because the hyperactivity can be attenuated by the selective SRI, fluoxetine, but not by a non-selective tricyclic antidepressant (26). Compulsive exercise is related to both anorexia nervosa (32) and the female athlete triad. Does the exercise-stress syndrome model other components of the triad?

It is well established that the exercise-stress syndrome models two components of the triad: disordered energy balance (33) and anestrous (34). However, to our knowledge, there are no reports demonstrating that the syndrome changes bone homeostasis. The reason for this, however, is that this possibility has not been heretofore considered. There is, however, an excellent reason to consider this proposal. It is well established that the syndrome activates the hypothalamo-pituitary-adrenal axis (HPA) (35). Importantly, the glucocorticoids are the most common cause of osteoporosis that is not menopause or age related (36).

A variety of animal models for anorexia nervosa have been proposed. These include transgenic mice over expressing corticotropin-releasing factor (37); the thin-sow syndrome in pigs (38); the anorexia produced by low-protein diets (39); strain differences in the ingestive effects of morphine in mice (40); restraint-stress anorexia (41); the spontaneous anorexias associated with hibernation (see 14); neurochemical depletion of the noradrenergic innervation of the paraventricular hypothalamic nucleus (see 42); and various contingent-exercised or forced-exercised models that lack either face or construct validity (see 14). However, there are relatively few model systems as pertinent to the female athlete triad as the exercise-stress syndrome. It has face, construct and criterion validity as an animal model of anorexia nervosa (14). This presentation concludes that it also a valid animal model of the female athlete triad. Consequently, the model is positioned to promote research on the psychoneuroendocrinology and psychoneuro-osteology of both anorexia nervosa and the female athlete triad.

Back to the top.

Back to the top.

Back to the top.

Back to the top.

Back to the top.