The Safety of Percutaneous Endoscopic Gastrostomy Tube Placement in Patients With Existing Ventriculoperitoneal Shunts

Abstract

Background: Because the insertion of percutaneous endoscopic gastrostomy tubes (PEG) involves disruption of the gastrointestinal tract with potential peritoneal contamination, patients with indwelling ventriculoperitoneal (VP) shunts could be at increased risk of meningitis, a potentially devastating infection. The safety of PEG placement in the presence of a VP shunt is unclear. Methods: A retrospective chart review was performed that included all adult patients with existing VP shunts requiring PEG placement at a single university medical center over an approximate 9-year period from July 1995 to March 2004. Results: Thirty-nine patients who underwent PEG placement 2–564 days after shunt placement were identified. Two patients (5%) subsequently developed meningitis. Cerebrospinal fluid cultures demonstrated Staphylococcus aureus and Enterococcus faecalis. These infections occurred 2 and 15 months after PEG placement, respectively. At the time of PEG placement, 17 patients (44%) were receiving antibiotics for reasons other than operative prophylaxis, and 11 patients (28%) received prophylactic antibiotics. Both infected patients had received antibiotics at the time of PEG placement. Conclusions: These data constitute the largest series of patients with existing VP shunts undergoing PEG placement reported to date. When compared with the published 2%–5% infection rate for patients with VP shunts alone, our data do not suggest an increased risk of infection for patients after PEG placement. As the total number of adult patients requiring a PEG after VP shunt placement is low, multicenter studies should be carried out to better stratify this risk.

Due to potential contamination with enteric contents, patients are theoretically at increased risk for meningitis. In this series of 39 patients, we did not find an increased rate of infection for shunted patients after percutaneous endoscopic gastrostomy placement compared to previously published shunt infection rates.