Comparison of estimated energy requirements using predictive equations with total energy expenditure measured by the doubly labelled water method in acute spinal cord injury.

Abstract

Prospective, observational OBJECTIVES: To evaluate agreement between a reference method (doubly labelled water, DLW) of total energy expenditure (TEE) and published equations for estimating energy requirements in acute spinal cord injury (SCI). Victoria, Australia METHODS: Twenty participants (18 male) within 8 weeks of traumatic SCI completed DLW, anthropometric and dietary intake assessments. Energy requirements were predicted using Harris-Benedict, Schofield, Henry, Nelson, Buchholz and Chun equations, multiplied by a combined activity and stress factor of 1.3, and the ratio method (kJ/kg body weight). Fat-free mass (FFM) and fat mass (FM) were calculated from TBW-derived DLW and from bioelectrical impedance spectroscopy (BIS). Median time since injury was 41 days. Median TEE was 9.1 MJ. Fair agreement was found between TEE and predicted energy requirements for the Chun (rc = 0.39), the Harris-Benedict equation (rc = 0.30), the ratio method (rc = 0.23) and the Buchholz (rc = 0.31) and Nelson equations (rc = 0.35), which incorporate measures of FFM and/or FM. Other equations showed weak concordance with DLW. When two hypermetabolic patients were removed, agreement between TEE and predicted energy requirements using the Buchholz equation increased to substantial (rc = 0.72) and using the Nelson (rc = 0.53) and Chun equations (rc = 0.53) increased to moderate. The Buchholz equation had the smallest limits of agreement (-2.4-2.3 MJ/d). The population-specific Buchholz equation that incorporates FFM, predicted from either BIS or DLW, demonstrated the best agreement in patients with acute SCI. The study was funded by grants from the Institute for Safety, Compensation and Recovery Research (ISCRR Project # NGE-E-13-078) and Austin Medical Research Foundation. M Panisset was supported by an Australian Postgraduate Award.