Pump-assisted versus gravity-controlled enteral nutrition in long-term percutaneous endoscopic gastrostomy patients: a prospective controlled trial.

Background: Vomiting, aspiration, flatulence, and diarrhea are well-known negative side effects of enteral nutrition through percutaneous endoscopic gastrostomy (PEG). However, it is not yet clarified if pump-assisted (PA) or gravity-controlled (GC) application is the more comfortable and safe choice for long-term nutrition through PEG.

Methods: This was a prospective, randomized, crossover study. Fifty long-term PEG patients were fed by PA nutrition (G1) and 50 patients were fed by GC nutrition (G2). Six weeks of observation (O1) was followed by a switch of method of nutritional application in both groups and an additional 6 weeks of observation (O2). Daily determination of comfort and safety was done with a standardized questionnaire. Evaluation of blood glucose levels on days 1, 21, and 42 during O1 and O2.

Results: The patients in both groups had the same medical conditions and were of the same age and sex. Far less flatulence (p < .0006) and epigastric fullness (p < .0003) was discovered in G1 during O1. Also, significantly less regurgitation (p < .0002) and vomiting of feeding diet (p < .0001) in G1 versus G2 could be observed. The rate of diarrhea (p < .0003) in G2 was higher than in G1. The daily profile of blood glucose was significantly better (p < .0008) in G1 than in G2. After the nutritional application was changed in O2, the PA group (G2) again showed a significantly better rate of flatulence, epigastric fullness, regurgitation, vomiting, diarrhea, and daily profile of blood glucose. Ninety-six percent of the patients in G2 preferred further nutrition by PA after finishing this study. All patients in G1 continued their accustomed nutrition by PA.

Conclusion: Nutrition through PA showed not only a higher comfort rate but also increased safety, which was expressed through a low rate of regurgitation and vomiting. PA presented better glucose metabolization manifested in improved blood glucose levels. As a result of this prospective study, PA is preferable to GC and preferred by patients with long-term PEG nutrition.