Hospital resource utilization associated with pediatric gastrostomy devices

Objective: Pediatric gastrostomy devices (GD) are associated with frequent planned and unplanned ongoing care. The objective was to examine the association between planned or unplanned repeat intervention and 1) the type of tube at initial placement and 2) the method of initial device placement: operative, endoscopic or image-guided. Methods: We performed a retrospective review from January 1, 2010 – December 31, 2011 of consecutive patients that received a GD or gastrojejunostomy device (GJ) at a freestanding pediatric hospital. We estimated the risk of hospital utilization in terms of repeat interventions or device-related emergency department visits for initial GJ versus GD placement and based on method of initial device placement. Results: Over the two year period, 275 (92.0%) GDs and 24 (8.0%) GJs were placed. One hundred forty-five (48.5%) were placed surgically, 113 (37.8%) endoscopically, and 41 (13.7%) with an image-guided technique. Repeat utilization occurred in 60.9% of patients. Multivariate Poisson regression estimated that GJs compared to GDs were 2.05 (95% CI: 1.01 - 4.16) times as likely to have repeat hospital utilization. Multivariate Cox regression estimated that endoscopic and image-guided placements were associated with higher rates of hospital utilization than surgical placement (HR 2.56; 95% CI: 1.73 - 3.79 and 2.07; 95% CI: 1.20 - 3.58, respectively). Conclusions: Resource utilization after GD/GJ placement is not equal relative to the method of initial placement. Process standardization requires further investigation to reduce this health care burden.