149 - The Effect of Pre-Admission Energy Balance on Short-Term Medical Outcomes: Findings from the Study of Refeeding to Optimize iNpatient Gains (StRONG)

Poster Number: 149
Publication Number: 149.580

Background: Negative energy balance contributes to bradycardia and other vital sign instabilities warranting hospitalization in adolescents and young adults (AYA) with anorexia nervosa (AN) or atypical anorexia nervosa (AAN). The Study of Refeeding to Optimize Inpatient Gains (StRONG) trial is a multicenter randomized controlled trial comparing higher-calorie refeeding (HCR), starting at 2000 kcal/d and advancing by 200 kcal/d, versus lower-calorie refeeding (LCR), starting with 1400 kcal/d and advancing by 200 kcal every other day.

Objective: The purpose of this secondary data analysis from the StRONG trial is to examine the effect of pre-admission, self-reported exercise and dietary restriction on short-term refeeding outcomes.

Design/Methods: Secondary analysis of data from the StRONG trial included 120 AYA hospitalized with medical instability secondary to AN or AAN. Admission assessments: Average Metabolic Equivalents (METs, kcal/kg/d) from self-reported physical activity for one week prior; dietary intake from 24-hr recall of one day prior (kcal/d); total energy expenditure (kcal/d) from Estimated Energy Requirements (< or =18 yrs) or Mifflen St. Jeor equations (>18 yrs) using assessed METs. Energy balance = dietary intake – total energy expenditure. Dependent variable: time to restore medical stability (heart rate (HR) ≥ 45 bpm; systolic blood pressure (SBP) ≥ 90 mmHg, temperature ≥ 35.6° C, orthostatic increase in HR ≤ 35 bpm and decrease SBP ≤ 20 mmHg, and ≥ 75% median BMI (%mBMI)). Unpaired t-tests compared groups; adjusted linear regressions examined independent variables (energy balance, METs) on hospital outcomes.

Results: Among 84 of 111 trial participants with METs data, 92% were female, mean (SD) age was 16.4 (2.5) years, 56% were diagnosed with AN, %median BMI was 84.7 (11.9), median energy balance was -838 (777) kcal. Greater negative energy balance, but not METs alone, predicted more days to restore medical stability [ = -.15 (CI -.2 , 0), p=.04], even on HCR.

Conclusion(s): AYA with AN and AAN who had greater energy imbalance due to combined higher levels of exercise and dietary restriction required longer medical hospitalizations, even when treated with HCR. Findings support the assessment of both physical activity and nutrition histories on admission to the medical hospital to inform individualized treatment protocols.