Background: Enteral feeding provides nutrients for patients who require endotracheal tubes and mechanical ventilation. There is a presumed increase in the risk of ventilator-associated pneumonia (VAP) with tube feeding. This has stimulated the development of procedures for duodenal intubation and small intestinal (SI) feeding as primary prophylaxes to prevent VAP.
Objective: To investigate the rate of VAP and adequacy of nutrient delivery with gastric (G) vs. SI feeding.
Design: A prospective, randomized, controlled trial.
Setting: A medical intensive care unit of a county hospital.
Patients: A total of 44 endotracheally intubated, mechanically ventilated patients requiring enteral nutrition.
Intervention: Subjects were randomized to receive enteral nutrition via G or SI feeding. Protocols directed the placement of the feeding tube and the infusion of enteral nutrition and defined the radiographic and clinical criteria for a diagnosis of VAP.
Measurements and Outcomes: The incidence of VAP and the adequacy of nutritional supplementation were prospectively followed. The relative risk of VAP with SI was 1.1 (95% confidence interval 0.96–2.44) compared with G. The SI group received a greater percentage of their caloric requirements (SI 69 ± 7% vs. G 47 ± 7%, mean ± sem, p < .05). Mortality did not differ between G (26 ± 9%) and SI (24 ± 10, p = .86).
Conclusions: There is no clear difference in the incidence of VAP in SI compared with G enteral nutrition. Patients given feeding into the SI do receive higher calorie and protein intakes.
The risk factors associated with nosocomial pneumonia (NP) are organ system failure (1), the method of airway management (2), the patient’s age, exposure to antibiotics, stress ulcer prophylaxis, level of consciousness (3), and the positioning of the patient (4). Epidemiologic studies have failed to demonstrate a correlation between NP and gastric stress prophylaxis or the site of enteral feeding (5). Prod’hom et al. (6), in a randomized controlled trial, demonstrated a significant reduction in NP when sucralfate was used in lieu of H2 blockers or antacids. However, this study, as well as many others, failed to control for the site of enteral feeding (6). Although other studies have shown that the site of enteral feeding is associated with NP (7–14) and maintenance of nutritional status, a causal relationship remains unproven (15, 16). The American Thoracic Society’s consensus statement concludes that distal enteral feeding sites probably decrease the risk of NP (17). Proponents of small intestinal (SI) feeding focus on the pylorus as a physiologic barrier that will reduce reflux of feedings and consequently lower the incidence of NP. Others claim that no proven reduction in the risk of NP accompanies duodenal or jejunal feeding when compared with the gastric (G) route (16). Two randomized controlled trials compared the risk of aspiration in intensive care unit (ICU) patients fed by the SI or G route and found no significant difference in the occurrence of NP (7, 18). One of the groups demonstrated equivalent risk of NP and retrospectively found improved nutrient delivery with SI feeding (7). A subsequent study compared outcomes with G vs. jejunal feeding tubes placed endoscopically (18). These authors prospectively demonstrated increased nutrient intake with jejunal feeding, but there was no significant decrease in NP. This study had a small number of subjects, and in their analysis the authors excluded patients who had failed attempts at feeding tube placement. The power of their study was limited because the risk of NP was low in their subjects. Because of methodologic problems, these studies, although consistent with the analysis by Kollef (5), do not eliminate a possible prophylactic benefit of SI feeding to prevent NP. Because there is evidence that the aspiration of oropharyngeal secretions is the principal source of aspiration leading to pneumonia (19) and that reflux and aspiration occur regardless of the position of the feeding tube, recommendations directing clinicians to use postpyloric enteral feeding sites should be reevaluated.
If the risk of NP with its consequent morbidity theoretically is reduced by selective intubation and feeding into the SI, the use of low-risk, low-cost placement techniques is a reasonable recommendation (17, 20, 21). Improvement in delivery of nutritional goals justifies a similar, but less definitive, recommendation. When the cost and complications from placement of the feeding tubes are significant, as with endoscopic or fluoroscopic intervention, a clear demonstration of clinical benefit, that is, reduced incidence of NP or improved clinical outcomes, should be required (22).
By selecting a population at high risk for pneumonia, we determined that small differences in NP could be detected. To test the hypothesis that distal enteral feeding, beyond the pylorus, provides a primary prophylaxis against NP, we developed a bedside technique for duodenal feeding tube placement in mechanically ventilated patients. This population has a higher incidence of a specific NP, ventilator-associated pneumonia (VAP). This study compared the rates of VAP and the adequacy of calorie and protein delivery with G vs. SI feeding in critically ill patients while mechanically ventilated.