Introduction Percutaneous endoscopic gastrostomy (PEG) placement has been a well recognised method for delivery of enteral nutrition for many years. A more recent and less invasive alternative has been radiologically inserted gastrostomy (RIG). Although both are accepted methods for enteral feeding, there is a lack of direct comparative data of the two techniques. Current practice is case dependent with lack of evidence based guidelines regarding appropriate choice of procedure.

Methods A 1-year retrospective analysis of all PEG and RIG procedures undertaken within the University Hospitals Leicester NHS Trust between September 2007 and 2008. Patients were identified using endoscopy records; radiology records; and Nutrition Nurse files. Notes from all those identified were retrieved and reviewed. A patient satisfaction questionnaire was posted to all surviving patients to evaluate the patient experience.

Results 129 PEGs and 49 RIGs were identified. Notes were reviewed in 75% PEGs (n=97), and 67% RIGs (n=33). Patient demographics were similar in both groups. Main indication for PEG was CVA or neurodegenerative disorders (46% all PEG procedures). In contrast, the most common indication for RIG placement was oesophageal carcinoma or ENT pathology (42% all RIG placement). 18% RIGs were inserted due to displacement of existing tube or blockage. Complications were more common in the RIG placement patients (48% vs 8% for PEG). Main complications in RIGs vs PEGs included displacement (12% vs 1%), unsuccessful placement (8% vs 2%), blocked tube (6% vs 0%), symptomatic hypotension (6% vs 0%), and abdominal pain/vomiting (4% vs 2%). 30-Day Mortality was 0.8% in the PEG cohort, and 2% in the RIG cohort. Patient Experience: Only 31% PEG (36/116) and 13% RIG (5/39) returned a completed questionnaire. Patients reported greater discomfort in the PEG group, with 17% indicating moderately severe/severe pain during the procedure. No responders in the RIG cohort indicated this level of discomfort but there was no indication of type or quantity of analgesia used.

Conclusion There appears to be a greater risk of complication following RIG placement compared to PEG placement, with higher overall 30-day mortality. It is acknowledged that these groups are not directly comparable, and the higher mortality rate in the RIG group may reflect the underlying disease process. The high rates of displacement or blockage of RIG are also of concern, and further research should evaluate most effective type of RIG for placement. There is a need for a robust prospective randomised controlled trial evaluating PEG placement vs RIG placement for provision of nutritional support. This will be vital for subsequent generation of evidence based guidelines and enable patients to be fully informed regarding the options for feeding.