“Just Eat Some Food”

Anorexia nervosa (AN) is a life-threatening eating disorder characterized by deliberate restriction of calories to the point of starvation (less than 85 percent of the individual’s normal healthy weight). AN usually begins at the onset of puberty and is more common among women, affecting 1 percent to 2 percent of the female population and 0.1 percent to 0.2 percent of males. AN is one of the important causes of premature death in women.

Shame, stigma, denial and distorted body image conspire so individuals with AN do not normally seek professional help. Rather, they are usually hospitalized in the midst of a medical crisis. Weight gain is the initial goal of treatment for malnourished and emaciated patients. But refeeding is medically and psychologically complicated, and not without risks. In an excellent review of current research and meta-analysis published in the journal Eating Disorders, Garber, et al, 2016, provide much-needed analysis and recommendations regarding the challenges of refeeding AN patients.

The methodology in this analysis included retrospective chart reviews, examining short-term, meal-based refeeding in malnourished adolescents (ages 14–16) diagnosed with eating disorder and hospitalized for medical stabilization. A unique feature of this review was the sub-analysis of severely malnourished patients (mBMI < 75%) who comprised nearly 30% of the study population. To date, only two prior studies of High Caloric Refeeding (HCR) of severely underweight AN subjects have had adequately powered sample sizes for valid research. In both instances, higher rates of hypophosphatemia during refeeding were observed.

Challenges of Stabilization and Refeeding

Weight gain, early on, is crucial for recovery. Among hospitalized patients, hastened weight gain plus a higher weight at discharge were predictive of improved weight recovery at one year. In outpatient treatment, early and significant weight gain (at least 0.43–0.86 kg/week during the first four weeks) was predictive of improved physical health (weight gain) and psychological recovery at 12 months.

The rate and goals of weight gain among AN patients have been a source of controversy among eating disorder professionals, due primarily to the risks of hypophosphatemia and refeeding syndrome. Accordingly, the authors assert that the best available evidence suggests:

  • Among mildly and moderately malnourished patients, LCR is too conservative and associated with poorer outcomes.
  • Medically managed HCR, with electrolyte correction, is not associated with increased risk for refeeding syndrome.
  • For severely underweight/malnourished hospitalized patients, there is insufficient evidence to change the current standard of care.
  • Parenteral nutrition is not recommended.
  • Nutrient compositions within current recommended ranges are appropriate.
  • More research is needed regarding treatment efficacy and risks associated with treatment in non-hospital settings.
  • The long-term impact of different refeeding approaches is not known.

The excellent analysis of the most current research and subsequent recommendations by these experts identify priorities for future studies on refeeding.

  1. The first is of course patient safety. Half of the participants in both the Maginot et al. and Smith et al. studies developed hypophosphatemia. Accordingly, HCR administered by ED professionals under close medical monitoring and electrolyte replacement in a hospital setting is feasible.
  2. A second priority is to describe feeding protocols in greater detail.
  3. Third, future refeeding studies should go beyond a four-week follow-up period (Smith et al. and Peebles et al) to examine longer-term treatment outcomes.

Why physicians prescribe Low Caloric Refeeding (LCR) to their more malnourished patients is most likely due to the documented risk of refeeding syndrome. Without better data, the decision to exercise caution with more malnourished patients is understandable. Study designs with randomized subjects and better controls are indeed difficult but needed to address these important questions.

Why Does This Matter?

Anorexia Nervosa is a disease with considerable mortality and morbidity. A review of nearly 50 years of research confirms that AN has the highest mortality rate of any psychiatric disorder. It is a major cause of death, either directly from starvation or indirectly due to suicide. Early in my career, I worked to identify changes in pituitary –gonadal axis which occur with starvation and result in loss of menstrual period and fertility. Rehydration, a key to initial treatment, has been shrouded in myth and has not been consistent from eating disorder program or one specialist to another.

Refeeding protocols and the risks of refeeding are critical issues in the initial stabilization of severely malnourished AN patients. Refeeding Syndrome is a life-threatening syndrome among severely underweight and malnourished individuals regardless of the cause.

This important finding was a result our government’s desire to learn how to treat starvation among those who were interned in concentration camps during World War II. The design involved recruiting active-duty serviceman to lose a predetermined percent of their body weight.

The main and unpredicted finding was unfortunate for the participants. Specifically, severe, rapid weight loss resulted in very low blood pressure and reduction in the size and strength of the heart. As a result, cardiovascular collapse can occur during refeeding because a smaller, weaker heart muscle can fail to function at its previous capacity when blood volume increases rapidly, as during HCR. Moreover, high carbohydrate intake, which one would intuitively assume would be the most logical and expedient way to deliver calories to a starving person, can cause increased electrolyte imbalances, which are life threatening and can result in heart failure. Consequently, the first few days and weeks of refeeding AN patients demands close monitoring of body mass and the cardiovascular system. AN is an important disease, which like major depression is very difficult to model in animals. Starvation, on the other hand, is all too common a model and occurs today all around the world. Refeeding and rehydration should be well studied and well understood as an evidence-based, lifesaving initial intervention.