Introduction Patients with cirrhosis are frequently malnourished and this has a detrimental effect on outcome. Accurate assessment of resting energy expenditure (REE) would facilitate management by providing an objective measure on which to base individualised recommendations for nutritional support. REE can be accurately measured using indirect calorimetry but this technique cannot be used easily in the clinical setting. A number of portable bedside techniques for estimating REE have now been developed although they have not, to date, been validated in this patient population. The aim of the present study was to validate the use of alternative methods for assessing REE against the ‘gold standard’ of indirect calorimetry.

Methods The study population comprised 19 patients (12 men: seven women; mean [range] age, 60.3 [31–83] yr) with cirrhosis; 11 (57.9%) were adequately nourished, three (15.8%) moderately malnourished and five (26.3%) severely malnourished. REE was measured, under standardised conditions, using a precision, indirect calorimeter (Cortex Metamax 3 B). Additional REE measurements were undertaken, at the same sitting, using: (i) the Medgem Handheld indirect calorimeter; (ii) the Sensewear Pro 3 armband direct calorimeter; and, (iii) the Bodystat Quadscan 4000 bioelectric impedance analyzer. REE was also predicted using the generic Harris-Benedict, Schofield, Mifflin, and Cunningham equations and the disease-specific Müller and Morgan & Madden equations.¹

Results The mean (±1 SD) REE measured using the Metamax indirect calorimeter was 1368±456.4 Kcal/24 hr. REE measurments provided by the alternative methods ranged from 1188.1 kcal/24 hr less to 722.6 Kcal/24 hr more than the Metamax values. The Medgem provided the ‘least inaccurate’ REE measurements but values still differed by approximately ±700 Kcal/24 hr. The differences in measurement values between the portable devices and the ‘gold standard’ were inconsistent and bidirectional. Likewise, REE values obtained using the prediction equations ranged from 1018.6 kcal/24 hr less to 1468.3 Kcal/24vhr more than the Metamax values. The generic Mifflin equation provided the ‘least inaccurate’ REE estimates but values still ranged from 887.7 Kcal less to 558.4 Kcal/24 hr more than the Metamax values.

Conclusion Measurements of REE provided using these bedside techniques and estimated using the predication equations were not sufficiently accurate to be of value, in the clinical setting, either for the assessment of nutritional requirements nor for monitoring over time to assess disease progression or responses to nutritional therapy.

Reference 1 Madden AM, Morgan MY. Resting energy expenditure should be measured in patients with cirrhosis, not predicted. Hepatology 1999;30:655–664.

Disclosure of Interest None Declared