Long-term outcome and efficiency of symptom-selective approach to assess gastroesophageal reflux prior to gastrostomy in neurologically impaired children.

ABSTRACT

PURPOSE: To evaluate the necessity of preoperative screening for gastroesophageal reflux (GER) prior to gastrostomy in neurologically impaired children. METHODS: Medical records of neurologically impaired children, who have undergone laparoscopic gastrostomy between January, 2004 and June, 2018, were retrospectively reviewed. Before the year of 2014, all patients who required gastrostomy had been routinely screened for GER pre-operatively, but after the year of 2014, only the ones with GER-related symptoms were tested. The characteristics and outcomes of Routine Screening (RS) and Selective Screening (SS) periods were compared. RESULTS: There were 55 and 54 patients in the RS and SS periods, respectively. Demographics, primary pathologies, and mean follow-up durations (> 2 years) were similar. The rate of GER screening was significantly lower in the SS period (29.6% vs. 63.6%). The rate of Laparoscopic Nissen Fundoplication (LNF) combined with gastrostomy was significantly lower in the SS period (14.8% vs. 38.2%). During follow-up, the rates of new-onset GER symptoms (13% vs. 11.7%) and LNF requirement later on (6.5% vs. 8.8%) were statistically similar between the two periods. CONCLUSION: Routine screening for GER is not necessary prior to gastrostomy in neurologically impaired children. Symptom-selective screening algorithm is safe and efficient in the long term.