MeasurementsPatients with cirrhosis are frequently malnourished, 1-3 and this may have a detrimental effect on outcome. 4-7 Adequate nutritional support must be provided if the outlook for these patients is to be optimized, [8][9][10][11] but this requires a knowledge of their nutrient requirements. Detailed studies have been undertaken on protein and nitrogen balance in these patients and consistent increases in their requirements clearly demonstrated. 12-15 However, there is no consensus in the data on energy expenditure, and thus energy requirements, in this patient population. [16][17][18][19][20][21][22] Plauth et al. 23 recommended that energy requirements should be determined in patients with cirrhosis by measurement of their resting energy expenditure (REE) using indirect calorimetry. They suggest, but do not advocate, that if the facilities for measurement are not available, then energy expenditure could be estimated by use of the Harris-Benedict prediction formulae. 24 However, it has been shown that these formulae may underestimate mean measured values of REE in patients with cirrhosis by up to 13%, 17,18,[25][26][27][28][29][30][31][32][33] though presently, it is unclear whether this discrepancy is equally evident in all population subgroups.A number of alternative prediction formulae are available, 34-39 but the accuracy of the REE estimates derived from these has not been studied systematically in patients with chronic liver disease. A prediction formula developed specifically for use in this patient population, which incorporated a measure of fat-free mass (FFM) derived using bioelectrical impedance analysis, has been proposed by Mü ller et al. 40 but this has not been validated.The aims of this study were, therefore: 1) to compare values for REE predicted using the Harris-Benedict, 24 Schofield, 36 Mifflin, 39 Cunningham, 35 and Owen 37,38 formulae with measured values in both patients with cirrhosis and a comparable group of healthy volunteers; 2) to compare values for REE predicted using the ''disease-specific'' Mü ller 40 formula with measured values in patients with cirrhosis; 3) to determine if the predictions made in the patients with cirrhosis are adversely affected by malnutrition, fluid retention, the etiology of the liver injury, or the degree of hepatic decompensation; and 4) to derive an equation for predicting REE in this patient population and to assess its validity.
PATIENTS AND METHODSThe patient population comprised 56 men and 44 women, with a mean (range) age of 48.5 (26-71) years, with cirrhosis of varying etiology and severity. The etiology of the liver injury was determined using historical, clinical, laboratory, radiological, and histological variables, while the functional severity of the liver disease was assessed using Pugh' s modification of Child' s grading system 41 (Table 1). Fifty-nine patients were assessed as in-patients, while the remaining 41 were assessed as out-patients. All patients with alcohol-related cirrhosis had been abstinent from alcohol for periods ra...