BACKGROUND: We report the largest European series of patients in whom both ventriculoperitoneal shunts (VPS) and percutaneous endoscopic gastrostomies (PEG) have been inserted with the aim of determining if this combination is safe or if there is an increased risk of VPS infection. PATIENTS AND METHODS: The paper and electronic records of 302 patients who had a ventriculoperitoneal (VP) shunt inserted in the regional Neurosciences unit at Salford Royal NHS Foundation Trust between 2002 and 2007 were reviewed. RESULTS: A total of 24 patients with VP shunts had 26 PEG inserted. Thirteen PEG were inserted in 11 patients with a pre-existing VP shunt. The median age was 58 years (21-77 Yrs) with seven male and 17 female patients. In total, five patients developed a shunt infection (20.8%) compared to the overall rate of VP shunt infection for Salford Royal NHS Foundation Trust of 7% (P = 0.017). The increase in number of VP shunt infections when the procedures were done more than 10 days apart (2/14) was not significant (P = 0.25). CONCLUSION: In patients who need long-term enteral feeding following a VP shunt insertion it may be prudent to delay insertion of a PEG for at least 10 days to reduce VPS infection. In stable patients who have had a VP shunt inserted on previous hospital admissions PEG insertion need not be avoided because of concern regarding cerebrospinal fluid or shunt infection.
BACKGROUND: We report the largest European series of patients in whom both ventriculoperitoneal shunts (VPS) and percutaneous endoscopic gastrostomies (PEG) have been inserted with the aim of determining if this combination is safe or if there is an increased risk of VPS infection. PATIENTS AND METHODS: The paper and electronic records of 302 patients who had a ventriculoperitoneal (VP) shunt inserted in the regional Neurosciences unit at Salford Royal NHS Foundation Trust between 2002 and 2007 were reviewed. RESULTS: A total of 24 patients with VP shunts had 26 PEG inserted. Thirteen PEG were inserted in 11 patients with a pre-existing VP shunt. The median age was 58 years (21-77 Yrs) with seven male and 17 female patients. In total, five patients developed a shunt infection (20.8%) compared to the overall rate of VP shunt infection for Salford Royal NHS Foundation Trust of 7% (P = 0.017). The increase in number of VP shunt infections when the procedures were done more than 10 days apart (2/14) was not significant (P = 0.25). CONCLUSION: In patients who need long-term enteral feeding following a VP shunt insertion it may be prudent to delay insertion of a PEG for at least 10 days to reduce VPS infection. In stable patients who have had a VP shunt inserted on previous hospital admissions PEG insertion need not be avoided because of concern regarding cerebrospinal fluid or shunt infection.