Prospective implementation of algorithmic patient selection for gastrostomy tube placement consultations: a pre- and post-intervention analysis

Background: Studies have shown high but variable mortality following gastrostomy tube (GT) placement. There are no standard practice guidelines for GT placement. Aim: To evaluate if implementation of patient selection and prognosis algorithms for GT insertion has favorable effects on patient outcomes. Methods: This was a pre-, post-cohort analysis in a Veterans Affairs hospital. We implemented a patient selection algorithm aided by the Sheffield Gastrostomy Scoring System (SGSS) in July 2015. We reviewed all referrals to the inpatient service for a GT between July 2014 and June 2016 (pre-, post- implementation), and collected albumin and SGSS at time of consultation, date of GT insertion, and outcomes including vital status and albumin 30 days post-consultation. Patients were followed until 30 October 2016. We compared outcomes pre- and post-implementation. Results: A total of 126 referrals were reviewed, 68 pre- and 58 post-algorithm implementation. Seventy-seven GTs were placed; 58 (75.3%) fulfilled the algorithm-appropriate indications. The mean SGSS was significantly lower among patients in whom GT was placed for algorithm-appropriate indications 2.03 (SD =0.86) vs inappropriate indications (2.59, SD =0.63; p<0.001). Sixty-five (51.6%) patients died by conclusion of study. Thirty day mortality after GT placement was 26.2% (post- (22.4%) vs pre- (29.4%)). Changes in serum albumin at day 30 was non-significant. Conclusion The use of algorithm guidance by the prospective use of the SGSS was associated with a higher likelihood of GT placement both overall and for algorithm-appropriate indications.