High carbohydrate is preferable to high lipid parenteral nutrition in healthy dogs undergoing prolonged sedation.

Parenteral nutrition (PN) is commonly used in intensive care units (ICUs) and is associated with earlier hospital outcome. However, there is scarcity of information about the metabolic effects of PN caloric distribution for dogs. Considering the high tolerance of dogs to lipids and, also, that hospitalized animals usually present insulin resistance, PN formulation with high fat instead high glucose can provide metabolic benefits in this specie. This study evaluated two PN protocols, based on high lipid or high carbohydrate in 12 healthy dogs under sedation/ventilation during 24 h. For baseline data, blood samples were collected 24 h before the study beginning. After fasting, the dogs were anesthetized and put under mechanical ventilation without energy support for 12 h to obtain: daily energy expenditure (DEE), respiratory quotient (RQ), oxygen consumption (VO2), carbon dioxide production (VCO2), lactate, glucose, cholesterol, and triglycerides concentrations. After, the dogs were allocated into two groups: lipid-based energy group (LEG) and carbohydrate-based energy group (CEG). Both groups received the PN infusions at a rate of 3 mL/kg/h for 12 h. Blood tests were performed 12, 24, and 48 h after infusion's completion. VO2 increased after PN in LEG, increasing energy expenditure compared to CEG. RQ remained close to 1 in CEG, indicating carbohydrate preferential consumption. Triglycerides increased in both groups after propofol infusion, remaining higher in LEG until the end of the evaluation. Glycaemia increased in CEG compared to baseline. In conclusion, both PN protocols can be used in healthy animals undergoing prolonged sedation protocols. However, high lipid PN had higher VO2 and DEE, and resulted in higher triglycerides concentrations and lower glycaemia indexes than carbohydrate, making high carbohydrate PN preferable to high lipid PN. Therefore, for use in critically ill patients, the data obtained in this study should be extrapolated, taking into consideration the specificity of each case.