Prognostic value of energy expenditure and respiratory quotient measuring in patients with liver cirrhosis.

BACKGROUND & AIMS: Resting energy expenditure (REE) and respiratory quotient (RQ) as measured by indirect calorimetry (IC) may correlate with muscle mass and represent prognostic indicators in treating patients with liver cirrhosis. We aimed to assess the correlation of IC-measured REE and RQ with skeletal muscle mass (SM), mortality, and REE values as estimated by Harris-Benedict, European guidelines (EG), and Brazilian guidelines-DITEN (BG) equations in patients with liver cirrhosis.

METHODS

In this prospectively designed study, REE was measured in 126 male patients with liver cirrhosis by IC and predicted by Harris-Benedict, EG (35 kcal/kg current weight), and BG (30 kcal/kg current weight) guidelines. Measurements were obtained at the time of admission to the study. Body composition was determined by whole-body dual-energy X-ray absorptiometry. The association between REE and 3-year survival was investigated.

RESULTS

Cirrhosis etiology was classified as alcohol related (59.0%), viral (20.1%), cryptogenic (11.8%), or other (9.0%). Mean Child-Pugh and MELD indexes were 8.30 ± 2.0 and 14.38 ± 6.12, respectively. RQ showed a moderate correlation with SM (r = 0.64), while IC-measured REE was inversely associated with mortality (multivariate Cox Regression, HR = 0.88, 95% CI: 0.78; 1, p = 0.04). Among the predictive equations for REE, only Harris-Benedict yielded values close to the IC, with a positive Pearson correlation (r = 0.77), excellent accuracy (Cb = 0.98), and positive Lin's concordance correlation (CCC = 0.75). However, a large standard deviation was observed; HB-measured REE did not correlate with mortality.

CONCLUSIONS

RQ and REE, as measured by IC, may be valuable tools for evaluating the severity of cirrhosis, by reflecting SM and predicting mortality, respectively. The predictive equations for REE included in this study cannot replace IC for this purpose. REGISTERED AT: www.clinicalTrials.gov (NCT02421848).