Review of: Garber AK et al, JAMA Pediatrics 2021;175(1):19–27

The standard of care for inpatients with anorexia nervosa (AN) and dangerously low weight is to start low-calorie refeeding (LCR). LCR starts the refeeding process at a very low calorie count (~1200 kcal/day) and progresses slowly. If malnourished patients move too quickly in this process, they may experience refeeding syndrome—potentially fatal shifts in fluids and electrolytes. The disadvantages of LCR include slow rates of weight gain and lengthy, expensive hospitalizations.

Some retrospective studies support the safety and efficacy of higher-calorie refeeding (HCR), but they have generally not included atypical anorexia nervosa (AAN), a new diagnosis describing patients who demonstrate restrictive behaviors and fear of weight gain but whose weight is in the normal range. Medically unstable patients with AAN comprise nearly one-third of patients in inpatient eating disorder programs.

In this multicenter randomized controlled trial, researchers compared LCR with HCR in medically unstable adolescents and young adults with AN or AAN. Exclusion criteria included BMI less than 60% of median BMI, recent hospitalization, pregnancy, chronic illnesses, suicidality, or psychosis. Participants in the HCR group (n = 60) initially consumed 2000 kcal/day, with a daily increase of 200 kcal, and those in the LCR group (n = 51) initially consumed 1400 kcal/day, with an increase of 200 kcal every other day. Measures of medical stability included heart rate, systolic blood pressure, temperature, and weight. Blood draws occurred daily for the first week and then every other day unless increased frequency was necessary.

The study found that the HCR group achieved medical stability significantly faster than the LCR group (p = 0.01) and had hospital stays that were four days shorter. The HCR group gained an additional 0.8 kg compared to the LCR group. However, the proportion of patients achieving medical stability did not differ between groups. The HCR group’s shorter hospitalizations resulted in savings of over $19,000 in hospital charges per participant.

The study’s findings cannot be generalized to extremely malnourished anorexic patients as they were not included in this study.

CHPR’s TAKE
HCR appears safe and effective both for patients with AN as well as those with AAN. Compared to LCR, HCR stabilizes patients faster, resulting in shorter and less costly ­hospitalizations.