OBJECTIVES: We sought to conduct an assessment of the practice of gastrostomy (G) tube placement across an entire city, which would reflect usual clinical care as compared with referral center practice. METHODS: We reviewed and retrospectively extracted data from patient records for all percutaneous endoscopic G (PEG) and radiological percutaneous G (RPG) tube placements at six Winnipeg hospitals between 1 April 2005 and 31 March 2007. Results: A total of 418 patients had G tubes (376 PEG, 42 RPG) inserted during the study period. The most common indications were cerebrovascular accidents (25%), head and neck cancer (23%), and head trauma (10%). The position of the external bolster was not documented in 38% of patients. The median time to the first complication was 10 days, initiation of feeding was 48 hours, and tube removal was 40 days. Complications developed in 102 (24%) patients. Patients with RPG tubes had more infections and were less likely to receive prophylactic antibiotics (P<0.001). In multivariate analysis, complications were more likely to occur in patients with RPG tubes and after insertions by lowest procedure volume physicians. Overall mortality was 12% within 30 days of G-tube placement. Death of one patient was directly related to peritonitis after G-tube insertion. Conclusions: In usual clinical practice, there is an underuse of prophylactic antibiotics and a delay in the institution of nutritional support after G-tube placement. A small but significant proportion of patients may develop major complications, with associated risk of mortality. The higher complication rate after procedures performed by lowest volume physicians needs further evaluation.
ASJC Scopus subject areas