Gastrocutaneous fistula in children after removal of gastrostomy tube: Incidence and predictive factors

Background/Purpose: Removal of a gastrostomy tube may result in a persisten t gastrocutaneous fistula. The authors reviewed their experience to determi ne the incidence of this event in children, and the factors that contribute to its occurrence. Methods: Over a 3-year period, 32 children had a gastrostomy tube placed th at was subsequently removed. Of these, 18 sites closed spontaneously and 14 had a persistent gastrocutaneous fistula (44%). The groups were similar wi th respect to age, sex, and diagnosis. Mean time from insertion to removal of the tube was 6.5 +/- 2 months in the group with spontaneous closure, and 29.8 +/- 8 months in the gastrocutaneous fistula group (P <.05). Results: One of 17 (6%) children who had tubes removed within 8 months of i nsertion had a fistula, compared with 13 of 15 (87%) of those who had one r emoved after 9 or more months (P<.001). Patients with tubes inserted percut aneously (PEG) were less likely to have a fistula than those using the Stam m technique (28% v 55%, not significant), although PEGs also tended to be i n place for a shorter time before removal (12.6 v 21.1 months). Steroids di d not increase the likelihood of a fistula (50% v42%, not significant). Conclusions: The incidence of persistent gastrocutaneous fistula is high af ter gastrostomy tube removal in children. The most important factor predisp osing to this problem appears to be the length of time the tu be is in plac e before its removal. Almost all tubes removed within 8 months will close s pontaneously. J Pediatr Surg 34:1345-1346. Copyright (C) 1999 by W.B. Saund ers Company.