Migration of percutaneous endoscopic gastrostomy tube in children.

Abstract

Migration of Percutaneous Endoscopic Gastrostomy Tube in Children To the Editor: We read with interest the article by Kaddu and Tolia (1) that described the migration of a gastrojejunostomy button into the duodenal bulb, leading to repeated vomiting. We report two additional cases of late complications after percutaneous endoscopic gastrostomy (PEG) caused by migration of the internal component of the gastrostomy device. Case 1: A 9-year-old boy with severe neurologic deficiency was brought to the hospital with bilious vomiting of 2 days. Because of severe difficulty swallowing and concomitant failure to thrive (weight, −1.5 SD), a gastrostomy tube (Ansell Medical, Cergy, France) had been inserted percutaneously under endoscopic guidance 6 months earlier, with no complications. Clinical evaluation, laboratory results, and phenytoin levels were normal. Abdominal radiography showed migration of the internal component of the gastrostomy tube through the pylorus, causing intestinal obstruction. The catheter was withdrawn into the stomach, and vomiting stopped immediately. Case 2: The patient was an 11-year-old boy with a history of prenatal asphyxia and severe neurologic deficiency. He had gastroesophageal reflux, severe difficulty swallowing, and failure to thrive (weight, −2 SD). The patient underwent PEG at 10 years of age. Eleven months after the PEG insertion, the patient was admitted to the hospital with severe respiratory distress and vomiting. Chest radiography showed migration of the internal component of the gastrostomy device into the proximal esophagus. The PEG tube was removed endoscopically without difficulty and was replaced with a button-type gastrostomy device (Fig. 1). Because of persistent respiratory distress, the patient underwent computer tomography scan of the chest, which showed a tracheoesophageal fistula caused by the internal component of the tube. The child died of respiratory failure 8 days later, despite surgical closure of the fistula.FIG. 1.: Migration of the gastrostomy device into the proximal esophagus.Percutaneous endoscopic gastrostomy placement is a well-established technique for administering long-term enteral nutrition in pediatric patients and is used with increasing frequency to facilitate feeding, particularly in children with mental retardation. The rate of complication associated with PEG placement ranges from 4% to 26% (2). Various types of complications (early and late, major and minor) have been reported, including wound infections, cutaneous necrosis, gastric metaplasia, granulation tissue, cologastric fistula, occlusion of the tube, intragastrically buried or excluded tube or button, and migration (2,3). The late complications reported in the current report, like the complication that Kaddu and Tolia (1) reported, were caused by migration of the internal component through the pylorus or into the esophagus. In these two cases, the diagnosis should have been made earlier by noting that the length of the external part of the PEG had decreased. Marking the tube, regularly changing and inspecting the external component, and local care and handling by the parents or the medical team should prevent such complications (4). Moreover, children with neurologic impairment require special attention because their symptoms are nonspecific and difficult to identify (lack of verbal expressions of pain, numerous other causes of vomiting). The diagnosis of tube migration should always be suspected in children with PEG in whom nausea, vomiting, or dyspnea develop. Stéphanie Uhlen Karine Mention Laurent Michaud