P514 - HOW CAN BODY COMPOSITION IMPROVE ENERGY EXPENDITURE ESTIMATES?

P514

HOW CAN BODY COMPOSITION IMPROVE ENERGY EXPENDITURE ESTIMATES?

A. Zabalegui^1,*, M. F. Mucarzel¹, R. Cartiel¹, F. Palmas¹, V. Rodriguez¹, R. Burgos¹

¹Vall d'hebron hospital, Barcelona, Spain

Rationale: Tailoring nutritional support to energy requirements (ER) is key in hospitalised patients receiving parenteral nutrition(PN). Indirect calorimetry (IC) is the gold standard for ER measurement; however, limited access often leads to the use of predictive equations (PE). Bioelectrical impedance (BIA) estimates ER through body composition-based formulas. The aim of our study was to evaluate the applicability of predictive equations and BIA in estimating ER in patients on PN support.

Methods: single-centre, cross-sectional study was conducted in hospitalised patients requiring PN. Both IC and BIA were performed. ER was measured using IC (Cosmed Q-NRG+) and estimated using PE: Harris-Benedict (HB), European Society for Clinical Nutrition and Metabolism (ESPEN) recommendations, and BIA-based basal metabolic rate estimates (Bodystat Quadscan 4000 and Akern Nutrilab). Correlation, strength of association, and agreement with IC were assessed using Bland–Altman analysis.

Results: A total of 179 patients were included (64.2% male; mean age 60.4 years). Table 1 shows the mean energy expenditure (EE) measured by IC and estimated using PE, along with correlation, strength of association, and agreement data.

Table 1.

All predictive equations showed a statistically significant correlation with IC, although predictive accuracy varied. BIA was the only method that demonstrated acceptable agreement with measured EE.

Conclusion: Predictive equations do not accurately reflect the actual ER of hospitalised patients on PN. BIA appears to be the most accurate method, likely due to the inclusion of body composition parameters in its estimation formula.

Disclosure of Interest: None declared