O1046: Evolving Surgical Decision Making in Gastrostomy Tube Concurrent Fundoplication: A National Retrospective Cohort Study

Background: Studies have demonstrated marked practice variability in performing an antireflux procedure, or concurrent fundoplication, at the time of gastrostomy tube (GT) placement. This variability has been attributed to evolving evidence on outcomes after fundoplication as well as non-clinical factors such as surgeon preference, institutional practice, and influence from referring physicians or families. This study examines trends in the sources of variation regarding the decision to perform a concurrent fundoplication versus GT alone.

Methods: This was a retrospective cohort study of children (0-17 years) who underwent GT placement with or without fundoplication from January 2018 – December 2023 in the National Surgical Quality Improvement Program-Pediatric database. We employed variance component analysis, with patients clustered within surgeons and surgeons clustered within hospitals, to quantify how much of the variability in treatment assignment (i.e. the decision to perform concurrent fundoplication) is attributable to different levels - patient, surgeon, or hospital. Then, the years 2018 and 2023 were separately analyzed via multilevel logistic regression with clustering for hospital to determine the patient clinical characteristics associated with concurrent fundoplication. These two analyses included the same set of clinical characteristic predictor variables.

Results: Over the study period, 23851 patients underwent GT placement sampled by NSQIP-Ped, with 5% (1187/23851) of patients receiving a concurrent fundoplication. Patient-level contribution to variance in the decision to perform fundoplication rose from 14% in 2018 to 51% in 2023, while surgeon- and hospital-level contribution fell from 86% in 2018 to 49% in 2023 (Figure 1). Of patient clinical characteristics, in 2018, patients had a higher odds of undergoing concurrent fundoplication if they had gastrointestinal disease (odds ratio [OR]: 7.26, confidence interval [CI]: 4.98–10.57), structural pulmonary abnormalities (OR: 2.21, CI: 1.52–3.20), or had received a blood transfusion within 48 hours (OR: 3.36, CI: 1.30–8.66) (Table 1). In 2023, structural pulmonary disease and recent transfusion no longer showed statistically significant associations, while gastrointestinal disease remained associated with fundoplication (Table 1). In addition, patients with impaired cognitive status had lower odds of undergoing concurrent fundoplication (OR: 0.54, CI: 0.34–0.86).

Conclusion: Though non-clinical factors previously dominated the decision of whether to perform a concurrent fundoplication with GT, patient clinical characteristics have recently played an increasingly prominent role. Additionally, the specific patient factors associated with fundoplication have shifted over time, potentially reflecting evolving clinical judgment. In particular, perhaps as a result of more contemporary literature, children with neurological impairment are now at lower odds of receiving concurrent fundoplication. Further studies are needed to compare outcomes of fundoplication versus gastrojejunostomy tube in children with significant reflux precluding gastric feedings.