Medical Complications in Atypical Anorexia Nervosa

Reviewed by Anne Marie O'Melia, MS, MD, FAAP

Whitelaw, M., Lee, K. J., Gilbertson, H., & Sawyer, S. M. (2018). Predictors of complications in anorexia nervosa and atypical anorexia nervosa: degree of underweight or extent and recency of weight loss? Journal of Adolescent Health, 63(6), 717-723.

https://doi.org/10.1016/j.jadohealth.2018.08.019

Admission guidelines for medical treatment of restricting type eating disorders often list low body weight as a reason for inpatient stabilization and careful refeeding. Diagnostic criteria for Anorexia Nervosa (AN) in the Diagnostic and Statistical Manual of Mental Disorder, 5th edition includes “significantly low weight” as a core feature of the illness. Patients with Atypical AN (AAN) meet all of the diagnostic criteria for AN, except that they are within or above the normal weight range despite having lost a significant amount of weight before presentation.   A recent study published in the Journal of Adolescent Health shows that patients with AAN carry the same medical risks as patients with AN.

This study highlights the need to carefully screen for eating disorders in all adolescents, including understanding any changes in their relationship with food, and to monitor all adolescent weight loss, including understanding methods of weight loss.

The Melbourne based eating disorders research team show that total weight loss and recent weight loss were stronger predictors than admission weight for several important clinical complications in teenagers with restrictive eating disorders. The study compared total weight loss and recent weight loss with admission weight as predictors of physical and psychological complications in teenagers aged 12 to 19 years with AN or AAN using data from retrospective (2005 to 2010) and prospective (2011 to 2013) studies. They collected data for each participant via chart review of medical records from day 1 up to 28 days following admission or until discharge.

Predictors of complications included total weight loss (from lifetime maximum), recent weight loss (from past 3 months), and admission weight. The investigators measured electrolyte balance, clinical, anthropometric, and psychometric markers during admission. Participants’ weights were taken twice weekly during hospitalization. Indicators measured during the clinical course were biochemistry measures, clinical observations, nutrition prescription, anthropometry, psychotropic use, psychometric measures and length of stay.

In total, 118 participants with AN and 53 with AAN were included in the study. They were found to have similar risks of electrolyte derangement, vital sign instability and dangerously low heart rate. Greater total weight loss and recent weight loss were stronger predictors than admission weight for the incidence and severity of bradycardia (low heart rate). Total weight loss was the strongest predictor for hypophosphatemia, developed by 41% of participants during treatment. Importantly, no complication was independently associated with simply admitting at a low weight, the hallmark of anorexia.

Why is this important?

This study illustrates that most adolescents who express an interest in losing weight, and all adolescents with recent, rapid weight loss need to be evaluated for the onset of an eating disorder. They may need specialized medical treatment to avoid life-threatening complications of refeeding syndrome. A patient with AAN will have a body weight that could be classified as being in a “normal” or “healthy” weight range but they are actually just as sick as someone with typical anorexia.  This study helps us understand the need to advocate for patients with AAN and to educate providers, families, teachers and coaches about the need to screen for and treat eating disorders in patients that may not initially look to be at high risk.