Outcomes of a Standardized, High-Caloric, Inpatient Re-Alimentation Treatment Protocol in 120 Severely Malnourished Adolescents with Anorexia Nervosa

Abstract

Evidence accumulates that, with close medical monitoring and phosphate supplementation, higher-caloric re-alimentation protocols beginning at 2000 kcal/day (HCR) are not associated with an increased incidence of electrolyte abnormalities in patients with anorexia nervosa (AN) but rather result in faster weight gain. These studies are still scant and have largely been performed in adults or moderately malnourished adolescents. Methods: A retrospective chart review of patients with AN aged 12−20 years and with a body mass index (BMI) < 15 kg/m2 alimented according to a standardized treatment protocol in a German clinic specialized in AN was conducted. All patients received 2000 kcal/day from day one. The effect of HCR was examined with respect to laboratory changes and weight development over 4 weeks. Results: In 120 youth (119 (99.2%) females and 1 (0.8%) male, the mean BMI was 13.1 ± 1.1 (range = 10.2−15.0), %mBMI was 62.1 ± 6.0% and weight gain was 0.76 ± 0.22 kg per week, with the highest rate of weight gain during week 1 (1.25 ± 1.28 kg/week). Over 4 weeks, the total weight gain was 3.00 ± 1.92 kg. Nine patients (7.5%) developed mild hypophosphatemia, and none developed refeeding syndrome. Conclusions: Starting re-alimentation with 2000 kcal/d under close medical surveillance, severely malnourished youth with AN met the recommended weight gain targets between 0.5 and 1 kg/week according to current treatment guidelines, without anyone developing refeeding syndrome.

Keywords: eating disorders; energy intake; nutrition; refeeding syndrome.

Figure 1

(a) Development of body weight over 28 days. (b) Development of BMI over 28 days. (c) Development of serum phosphate over 28 days.

Figure 1

(a) Development of body weight over 28 days. (b) Development of BMI over 28 days. (c) Development of serum phosphate over 28 days.

Figure 2

Relationship between serum phosphate and BMI on admission. While the correlation was significant, R² was very low (r² = 0.07), indicating only a weak relationship between the two factors. Patients with a lower BMI on admission tended to have lower serum phosphate values. However, the patient with the lowest BMI of 10.2 had normal serum phosphate values of 1.32 mmol/L, whereas a patient with a BMI of 13.7 presented with a phosphate value of 0.9 mmol/L. Moreover, the BMI of patients with hypophosphatemia in this study ranged from 11.1 to 13.7 kg/m², indicating that low phosphate values are not necessarily exhibited only at lower BMI values.