Comparison of the standard predictive equation for calculation of resting energy expenditure with indirect calorimetry in hospitalized and healthy dogs

OBJECTIVE: To determine the level of clinical agreement between 2 methods for the measurement of resting energy expenditure (REE). DESIGN: Prospective case series. ANIMALS: 77 dogs. PROCEDURE: Oxygen consumption (Vo2) and CO2 production (Vco2) were measured with an open-flow indirect calorimeter in healthy (n = 10) and ill (67) dogs. Measurements were collected at 3 time periods on 2 days. The Vo2 and the Vco2 measurements were then used to calculate the REE values. RESULTS: Mean values of measured (MREE) and predicted (PREE) REEs in healthy dogs and a dog with medical illnesses or trauma were not significantly different. There was a significant difference on day 2 between the MREE and PREE in the group of dogs recovering from major surgery. More importantly, there was significant variation between the PREE and MREE on an individual-dog basis. The PREE only agreed to within +/- 20% of the MREE in 51% to 57% of the dogs. CONCLUSIONS AND CLINICAL RELEVANCE: The level of agreement between these two methods for determining the 24-hour REE was poor in individual dogs. The level of disagreement between the 2 methods indicates that these methods may not be used interchangeably in a clinical setting. Measurement of REE by use of indirect calorimetry may be the only reliable method of determining REE in an individual ill or healthy dog.