CI: Initiation of Enteral Nutrition (2012)

Study Design:
- Click here for explanation of classification scheme.
Quality Rating:
Research Purpose:
To determine if the outcomes of early administration of enteral feeding to mechanically ventilated patients (pt) is beneficial or detrimental when compared to delayed enteral feeding. To study in incidence of infection among patients receiving enteral nutrition.
Inclusion Criteria:
Pt > 18 yrs, admitted to the Intensive Care Unit (ICU), expected to require mechanical ventilation >24 hrs.
Exclusion Criteria:
Pt transferred to ICU temporarily because (b/c) of a lack of available beds in another hospital, had received prior mechanical ventilation during the same hospital stay, contraindication to enteral (EN) feeding (pancreatitis, short gut syndrome, or malabsorption syndrome), classified as being malnourished by CU nutrition support team, received EN or parenteral (PN) nutrition before admission to ICU, if pt’s physician prescribed different nutrition support strategy, pt or family refused to give informed consent for study participation, pt could not tolerate placement of orogastric or nasogastric tube.
Description of Study Protocol:
Study assigned mechanically ventilated patients in the ICU of Barnes-Jewish Hospital in St. Louis, MO, to the early-feeding group or the late feeding group. Patients in the early-feeding (EF) group were scheduled to receive their total EN needs on the first day of ventilation. Late group (LF) patients received 20% of estimated EN daily total requirements for first 4 days of ventilation and 100% beginning on day 5.
Nutritional requirements were estimated by the nutrition support team w/ a goal of 25 kcal/kg of ideal body weight (IBW) and 1.0-1.3 g protein/kg IBW. EN was given via bolus feeds run by gravity into the stomach. Formula used was Osmolite HN providing 1.06 kal/ml, w/ 16.7% protein, 29.0% fat, and 54.3% carbohydrate. Medications and free water were also administered through the orogastric tube. EN feedings began at 60 ml every 4 hours (hrs) and were increased by 60 ml every 3 feeds until goal was met. If residuals exceeded 150 ml, next feeding was held. After 3 feedings were held, small bowel tube was placed and feedings administered by drip. Patients admitted on odd numbered days were assigned EF and those admitted on even days were assigned LF.
General pt characteristics were recorded upon study admission including diagnosis, severity of illness based on Acute Physiology and Chronic Health Evaluation (APACHE II) and  predicted mortality based on APACHE II. Mechanically ventilated patients had orogastric tube placed on day 1 of ventilation, bed elevated to >30 degrees (expect during procedures or patients w/ hypotension). Cultures for vancomycin-resistant enterococci and C difficile were taken at ICU admission, if diarrhea present, and at discharge or death.
Data Collection Summary:
Primary outcome measures included occurrence of ventilator-associated pneumonia. Secondary outcome measures included duration of mechanical ventilation, length of stay (LOS) in both ICU and hospital, hospital mortality, diarrhea (from Clostridium difficile infection), need for gastrostomy tube, and total ICU days on antibiotics.
Description of Actual Data Sample:
189 patients were evaluated for the study. 27 excluded b/c they received EN or PN earlier in their hospital stay, 12 b/c of extubation w/in 12 hrs of admission to ICU. 150 patients were enrolled in the study (75 in the EF group and 75 in the LF group).
Summary of Results:
  • During first 5 days EF received 27.9% of caloric needs and 26.9% of protein needs compared to LF group that received 7.0% of caloric needs and 7.7% of protein needs.
  • 44 (58.7%) of EF pts and 19 (25.3%) of LF pts had at least one feeding w/held b/c of gastric residual volumes (p < 0.001).
  • 25.3% of EF pts developed ventilator-associated pneumonia during the first 96 hrs vs 14.7% of LF pts (p=0.102).
  • 24.0% of EF pts developed pneumonia after 96 hrs vs 16.0% of LF pts (p=0.221)
  • Pts developing pneumonia had higher Clinical Pulmonary Infection Score (CPIS).
  • Duration of ventilation was 6.1 ± 6.2 days for EF and 5.7 ± 8.2 days for LF (p = 0.847).
  • Total antibiotic days were 12.4 ± 9.9 for EF vs 16.7 ± 12.5 for LF (p < 0.001).
  • ICU LOS for EF was 13.6 ± 14.2 days vs 9.8 ± 7.4 days for LF (p = 0.043)
  • Hospital LOS was 22.9 ± 19.7 days for EF vs 16.7±  12.5 days for LF (p = 0.023)
  • No statistical difference observed in number of mortalities per group.
Author Conclusion:
More incidence of ventilator-associated pneumonia and diarrhea and longer LOS were observed in pts in the EF group. The study suggests that suboptimal infusion does more harm than almost no infusion.
Funding Source:
Reviewer Comments:
  • Unblinded study
  • Delivery of nutrition was very low in both groups.
  • Author notes limitations of generalizability of the study and unusal method of delivery of nutrition to ICU patients (Bolus)