Researchers studying changes in levels of stress and appetite hormones in women recovering from the eating disorder anorexia nervosa (AN) report that it may be possible to predict certain serious post-treatment risks that can jeopardize recovery.  

Their discovery involves an observed relationship between levels of the stress hormone cortisol and the risk that a recovering AN patient will suffer from binge eating symptoms in the months following successful completion of a weight-restoration program.

The research, published in the Journal of Personalized Medicine, was conducted by a team led by Jessica H. Baker, Ph.D., of the University of North Carolina (UNC) at Chapel Hill, and included Cynthia M. Bulik, Ph.D., a 2017 BBRF Distinguished Investigator at UNC and the Karolinska Institute in Sweden. Ya-Ke Wu, Ph.D., RN, was first author of the study.

Anorexia nervosa (AN) is an eating disorder defined by an individual's restriction of food intake to the point that it causes a significant and dangerous decrease in body weight. Affecting mostly (but not exclusively) women, AN typically involves anxiety toward food, fear of weight gain, and a failure to recognize the seriousness of sharp reductions in body weight. Untreated, the disorder can lead to starvation and loss of life.

People with AN fall into two subtypes. One, called "restrictive," involves a consistent avoidance of food. In the other subtype, called "binge-eating/purging," the affected individual ingests food, in some cases in large quantities, but then purges the food before it can be digested. In bulimia nervosa (BN), as well as in binge-eating disorder (BED) and the binge eating/purging subtype of AN, the individual experiences a loss of control while eating (i.e., a feeling that one cannot stop/control how much one is eating), regardless of the amount of food consumed.  

The loss of control over eating is thought by researchers to be a more important predictor of poor outcomes in BN and binge-eating /purging AN patients than the quantity of food ingested. In recent years, this has led to a distinction between "objective" and "subjective" types of binge eating. In objective binge eating, the quantity of food ingested is indeed abnormally large, given context and body type; in subjective binge eating, the individual has the perception of eating a vast quantity of food but actually is not doing so. Both forms of binge eating involve the experience or perception of the affected person of losing control over their food intake, which can have serious psychological and biological consequences.

The UNC research team notes that even though the portions of food consumed by AN patients during renourishment treatments (usually conducted over a period of weeks on an inpatient basis, in a medical facility) is not objectively large, they often contain energy-dense foods. Thus, the researchers say, "patients with AN may be at risk of developing binge eating when consuming a high caloric diet daily in order to achieve target weight goals."

The team recruited 75 women, aged 15 or older, receiving renourishment treatment in an academic medical center, of whom 73 were included in the final analysis. Among other things, the women contributed blood samples at three time points:  admission, discharge, and 3 months after discharge.  Levels of key hormones were measured each time, including levels of the "appetite hormone," ghrelin, the "satiety hormone," leptin, and various hormones involved in the body's stress response, including cortisol.

Three months post-treatment, leptin and ghrelin levels observed during renourishment were noted to have changed over time, but in neither case were these levels found to be associated with self-reported binge eating in the 73 women in the prior 4 weeks.

In contrast, cortisol levels were positively associated with binge eating risk in the same interval.  Specifically, higher cortisol levels at the beginning of and after the treatment period ended tended to predict that a participant would develop subjective binge eating behavior—the self-reported perception of some of the women that they had lost control of their eating and were eating too much, even though in objective terms they may not have been.

"Our findings suggest that cortisol might serve as an early marker of the feeling of losing control over eating," the team said. While the relation was seen only in subjective binge eating, they said that the small size of their sample, and hence, the limited statistical power of the study, may have prevented them from seeing a similar relation between cortisol levels and risk of objective binge eating in women following weight restoration treatment.

In future studies, the team aims to uncover biological mechanisms underlying loss of control over eating, or its perception. The broader aim is to identify and validate biomarkers of binge-eating behaviors, which in turn might reveal that within the various subtypes of AN, personalized treatments based on the biomarkers might be possible, potentially improving outcomes.