Early postoperative alterations in infant energy use increase the risk of overfeeding

Energy needs in infants are decreased after surgery because of growth inhibition (resulting from catabolic stress metabolism), decreased insensible losses, and inactivity. Using standardized formulas that account for growth, activity, and insensible losses during this stress period can lead to overfeeding in excess of 200% of the actual measured requirement. Overfeeding during this acute injury period can result in increased CO2 production from lipogenesis. This study determined the effects of a reduced rate of mixed caloric repletion on infant energy use during the early postoperative period.

C-reactive protein (CRP), oxygen consumption (VO2), carbon dioxide production (VCO2), measured energy expenditure (MEE), and total urinary nitrogen (TUN) were measured serially in seven infants (average age, 78 days) during the first 72 hours after abdominal or thoracic surgery. Nonprotein respiratory quotient (RQnp), and values for oxidation of carbohydrate (Ce) and fat (Fe) were calculated. Injury severity was stratified based on serum CRP concentrations of > or = 6.0 mg/dL (high stress) or < 6.0 mg/dL (low stress). Recovery from acute stress was analyzed by comparing studies in which CRP had decreased to < or = 2.0 mg/dL (resolving stress group) with those in which CRP values were greater than 2.0 mg/dL (acute stress group).

Average total caloric intake (64.56 +/- 18.51 kcal/kg/d; approximately 50% of predicted energy requirement) exceeded average MEE (42.90 +/- 9.98 kcal/kg/d) by approximately 50%. Average TUN was 0.18 +/- 0.07 g/kg/d (high stress 0.2 +/- 0.05 versus low stress 0.16 +/- 0.09 g/kg/d). Average RQnp was 1.05 +/- 0.13 and average Ce was 37.28 +/- 16.86 kcal/kg/d. The average calculated Fe was 0.0 +/- 12.27 kcal/kg/d, reflecting approximately equal amounts of fat oxidized compared with fat generated from excess glucose (lipogenesis). When individual studies were analyzed at a CRP cutpoint of 2.0 mg/dL, overfeeding (RQ > 1.0) was significantly less likely in the resolving (2/6 studies, 33.4%) versus acute stress (9/13 studies, 69.2%, Z test P < .001) group. Five of seven (5/7) patients (9/19 individual studies) had negative Fe values (average -9.89 +/- 10.02) reflecting net lipogenesis. The RQnp for these nine studies was 1.14 +/- 0.11 versus 0.97 +/- 0.09 for the remaining 10, and this difference was significant (P < .01). A significant correlation existed between carbohydrate intake and VCO2 (Pearson r = .6951, P < .01). In addition, there was a good correlation between carbohydrate intake and VCO2 (Pearson r = .6591, P < .01). The coefficient of variation for MEE was 8.0% (low stress) versus 30.2% (high stress).

Lipogenesis with increased CO2 production is substantial, even at reduced caloric delivery rates that exceeded MEE by only 50%, during the early postoperative acute metabolic stress period in infants. These data suggest that caloric requirements during stress are likely equal to or only minimally in excess of actual MEE. Intersubject variability, especially in more severely stressed infants, underscores the importance of serial measurements of energy expenditure to enable precise caloric delivery and avoid overfeeding. In the absence of calorimetric measurement, the data suggest that PBMR (predicted basal metabolic rate) should be used to estimate caloric delivery until CRP values are < or = 2.0 mg/dL.