Abstract P195: Assessing the Validity of Predicting Resting Metabolic Rate in Chronic Stroke Patients

Abstract

Chronic stroke patients may have lower resting metabolic rate (RMR) due to disability and ensuing loss of skeletal muscle. Established equations used to estimate RMR based on weight, height, age, or lean body mass in healthy non-stroke individuals may not be accurate for a hemiparetic patient population. The purpose of this study is to determine resting metabolic rate in chronic stroke survivors and compare to RMR calculated with established equations in healthy adults. Adults (n=71; 56 males, 15 females; 40 African American, 27 Caucasian, 4 other / not reported) aged 44-76 years (61 ± 7.5 yrs) who had a history of chronic stroke (> 3 months prior) underwent a 30 minute test after a 12-hour fast to measure RMR by indirect calorimetry, total body DXA scan, and treadmill test (VO2 peak). Estimated RMR was calculated using nine established equations. RMR measured in the total group (1552 ± 319 kcal/day) was significantly different from all nine estimated RMR values (Katch-McArdle 1664 ± 242 kcal/day, P=0.05; Livingston 1671 ± 239 kcal/day, P<0.001; Mifflin 1703 ± 254 kcal/day, P<0.001; Owen 1761 ± 269 kcal/day, P<0.001; Harris Benedict 1782 ± 308 kcal/day, P<0.001; revised Harris-Benedict 1795 ± 306 kcal/day, P<0.001; Cunningham 1818 ± 247 kcal/day, P<0.001; Schofield 2147 ± 301 kcal/day, P<0.001; IMNA 2428 ± 405 kcal/day, P<0.001). Calculated RMR was between 9% and 60% greater than measured RMR, regardless of race. Appendicular lean mass (r=0.65, P<0.001), total lean mass (r=0.65, P<0.001), and VO2 peak (r=0.50, P<0.001) were associated with measured RMR. RMR estimation equations established in healthy adults are not reliable for the chronic stroke population, indicating the need for a more accurate predictive equation to better assist nutritional status in patients with conditions of muscle atrophy.