[Skip to Navigation]
December 1997

Enteral Feeding Intolerance: An Indicator of Sepsis-Associated Mortality in Burned Children

Author Affiliations

From the Department of Surgery, University of Texas Medical Branch and the Shriners Burn Institute, Galveston Unit, Galveston.

Arch Surg. 1997;132(12):1310-1314. doi:10.1001/archsurg.1997.01430360056010

Objective:  To determine if enteral feeding intolerance (EFI) is associated with sepsis and increased mortality in children with severe burns.

Design:  A survey.

Setting:  A pediatric burn unit.

Patients:  Ninety-one children surviving longer than 5 days with greater than 80% total body surface area burns.

Interventions:  None.

Main Outcome Measures:  Enteral feeding intolerance indicated by high gastric residuals (>150 mL/h) or uncontrollable diarrhea (>2500 mL/d); thrombocytopenia (platelet count < 100×109/L); hyperglycemia (glucose level >11.1 mmol/L [>200 mg/dL]); sepsis (pathogenic bacteremia or fungemia noted on blood culture results); and mortality.

Results:  Neither EFI nor sepsis developed in 71 patients, EFI alone developed in 2 patients, sepsis alone developed in 5 patients, and EFI and sepsis developed in 13 patients. Enteral feeding intolerance and sepsis were associated by contingency table analysis (P<.001). Mortality was 8% (6 patients) in those with neither EFI nor sepsis, 50% (1 patient) in those with EFI alone, 60% (3 patients) in those with sepsis alone, and 77% (10 patients) in those with EFI-associated sepsis. The 2 latter groups were different from the group with neither EFI nor sepsis (P<.05). Enteral feeding intolerance was identified in 70% of patients before sepsis; thrombocytopenia, 64%; and hyperglycemia, 66%. When compared with thrombocytopenia and hyperthermia, no variables were found to be superior to others for predicting sepsis.

Conclusions:  Enteral feeding intolerance was associated with the development of sepsis and increased mortality in children with greater than 80% total body surface area burns. This sign was identified in 70% of the cases before pathogens were found in the blood; no difference could be shown between the identification of EFI, thrombocytopenia, and hyperglycemia before sepsis. These data indicate that the development of EFI should be used as an indicator of infection and should prompt a search for an inciting focus.Arch Surg. 1997;132:1310-1314