Morbidity in neurologically impaired children after percutaneous endoscopic versus Stamm gastrostomy

Neurologically impaired children frequently require a feeding gastrostomy. Few reports are available comparing the incidence of postoperative complications and symptomatic gastroesophageal reflux after endoscopic versus operative Stamm gastrostomy in this group of children. We undertook a retrospective study of 63 consecutive neurologically impaired children requiring a feeding gastrostomy, with an average of 23 months of follow-up. No child had symptomatic gastroesophageal reflux. Thirty children had a percutaneous endoscopic gastrostomy and 33 had a Stamm gastrostomy, depending on the preference of the surgeon. The two groups were comparable in age range, cause of neurologic impairment, and indication for gastrostomy. Minor complications occurred in 30%. All three major complications occurred after Stamm gastrostomy, including two postoperative deaths. Symptomatic gastroesophageal reflux developed in 60%. The incidence of fundoplication after gastrostomy was 10% in the percutaneous endoscopic gastrostomy group and 39% after Stamm gastrostomy (p < .025). Morbidity was lower after percutaneous endoscopic gastrostomy than after Stamm gastrostomy in this group of neurologically impaired children. Fundoplication for symptomatic gastroesophageal reflux was infrequent after percutaneous endoscopic gastrostomy and significantly more common after Stamm gastrostomy. Percutaneous endoscopic gastrostomy is recommended as the initial procedure in neurologically impaired children without symptomatic gastroesophageal reflux who require a feeding gastrostomy.