Comparison of equations of resting and total energy expenditure in peritoneal dialysis patients using body composition measurements determined by multi-frequency bioimpedance

Abstract

Waste products of metabolism accumulate in patients with kidney failure and it has been proposed that the amount of dialysis treatment patients require be adjusted for energy expenditure. This requires validation of methods to estimate energy expenditure in dialysis patients. We compared values of resting energy expenditure (REE) estimated in peritoneal dialysis (PD) patients using a selection of available equations with estimates derived using a novel equation recently validated in chronic kidney disease patients (CKD equation). We also determined the relationship of these estimates of REE and of total energy expenditure (TEE - which is REE plus physical activity associated energy expenditure (PAEE) estimated using the Recent Physical Activity Questionnaire) - to bioimpedance-derived parameters of body composition. We studied 118 adult PD patients; 75 male (63.6%), 33 diabetic (28.5%), Caucasoid (42.4%), mean age 59.3±18.2 years and weight 73.1±16.6kg. REE with the CKD equation was 1532±237kcal/day, which was more than that for Mifflin-St. Joer 1425±254, Harris-Benedict 1489±267, Katch-McArdle 1492±243, but less than Cunningham 1648±248kcal/day. Bland Altman mean bias ranged from-107 to 111kcal/day. TEE was 1924 (1700-2262) kcal/day, and on multi-variate analysis was associated with appendicular muscle mass and nitrogen appearance rate (β 34.3, p<0.001 and β 5.6, p=0.002, respectively). With reference to the CKD equation, the majority of standard equations underestimate REE in PD patients. Whereas the Cunningham equation overestimates REE. TEE was associated with appendicular muscle mass and estimated dietary protein intake.