Effect of gastric versus post-pyloric feeding on the incidence of pneumonia in critically ill patients: observations from traditional and Bayesian random-effects meta-analysis.

Jing Jiyong,Huang Tiancha,Wang Huiqin,Jin Jingfen

DOI: https://doi.org/10.1016/j.clnu.2012.07.002

IF: 7.643

2013-01-01

Clinical Nutrition

Abstract:Administration of enteral feeding is associated with a higher risk of nosocomial pneumonia. Herein, we systematically review the impact of gastric versus post-pyloric feeding on the incidence of pneumonia.We searched the MEDLINE, EMBASE, Web of Science, and CCTRD (1966 to August 2011) for studies comparing gastric and post-pyloric feeding in critically ill patients. Two reviewers reviewed the quality of the studies and performed data extraction independently. Main outcome measures were the incidence of nosocomial pneumonia, aspiration, and vomiting. The meta-analysis was performed using traditional and Bayesian random-effects model.Our initial searches yielded 563 studies. Of these, we identified 15 randomized clinical trials enrolling 966 participants. Post-pyloric feeding was associated with reduction in pneumonia compared with gastric feeding (relative risk [RR] 0.63, 95% confidence interval [CI] 0.48-0.83, p = 0.001; I² = 0%). The risk of aspiration (RR, 1.11; 95% CI, 0.80-1.53, p = 0.55; I² = 0%) and vomiting (RR, 0.80; 95% CI, 0.38-1.67, p = 0.56; I² = 65.3%) were not significantly different between patients treated with gastric and post-pyloric feeding.Comparing with gastric feeding, post-pyloric route can reduce incidence of pneumonia in critically ill patients.