Outcomes of radiologically inserted gastrostomy versus percutaneous endoscopic gastrostomy

Gastrostomy insertion either via radiological (radiologically inserted gastrostomy, RIG) or endoscopic (percutaneous endoscopic gastrostomy, PEG) approaches are widely practiced throughout Australia. The purpose of this study was to compare outcomes of inserted tubes and cost evaluation by both methods.

A retrospective cohort study conducted on all-cause gastrostomy insertions at a quaternary Australian Hospital, Royal Brisbane and Womens' Hospital (RBWH) between January 2012 and August 2015. Current referral pattern is first-line gastrostomy and second-line radiological insertion.

A total of 402 gastrostomy tubes were inserted with a total of 307 PEG tubes and 95 RIG tubes, with follow-up to one calendar year. Mean patient age was 61 years ± 14.2 years with 76% male patients. A total of 84% of patients were head and neck cancer patients; major indications for insertion include prophylaxis (58%), dysphagia (32%) and NBM (2.5%). Patient groups were heterogeneous with varied indications for insertion including prophylaxis, dysphagia, decompression, NBM and treatment side effects. Outcomes measured included the following: complications, premature tube failure prior to expected removal and, overall tube outcome/ disposition. A lower incidence of minor complications was observed with the RIG group than the PEG group without differences in major complications over time. Tube failure due to either blockage or dislodgement was assessed. Multivariate analysis of all-cause dislodgement found 'method of insertion' a predictor of dislodgement with RIG 5.4(OR) times more likely to be dislodged than PEG. Competing risk analysis demonstrates equipment as a significant cause of dislodgement occurring more commonly with RIG than PEG tubes. Tubes were removed more often in the PEG group because a large volume were prophylactic. Tubes were replaced more often in the RIG group, with tube blockage and equipment as causes for tube replacement in this group. Replacements occur either in suite or bedside. Costing data were limited with only 94 patients' costing data qualifying for a limited unit cost evaluation, with radiologically inserted tubes marginally more expensive than tubes inserted endoscopically.

Both are safe procedures, with improved techniques; radiologically inserted gastrostomies have an improved profile with respect to dislodgement rates than previously reported in the literature. Radiological tubes remain limited by equipment factors with balloon failure an ongoing issue. Cost analysis was hindered by poor documentation; however, the opportunity cost remains an important advantage of radiological insertion at peripheral sites, increasingly relevant for health service delivery in our geographically vast state.