Reply to Gyökeres et al.

 

 Buy Article Permissions and Reprints

The useful comments and suggestions by Gyökeres et al. regarding the optimal treatment for colocutaneous fistula after percutaneous endoscopic gastrostomy (PEG) are impressive and well-taken points. We agree with Dr. Gyökeres that simple conservative procedures such as cutting of the outer PEG tube, pushing the inner part of the tube, and waiting for uneventful spontaneous closure of fistula would also be a good and classic solution for closure of colocutaneous fistula in most cases. With regard to the optimal treatment for gastrocolocutaneous fistula after PEG, however, we would like to emphasize two points. Firstly, how can one ensure safe natural drainage of the retained part of the tube without endoscopic retrieval? In practice, tube migration may accelerate mucosal erosion, and rare but life-threatening complications associated with retained PEG tubes have been reported - bowel perforation and obstruction [1] [2] [3]. Although these complications develop mainly in the small bowel, a retained tube can potentially cause similar complications in the colon, especially in patients in poor general condition or with comorbid diseases. Simple skin-level division of the PEG without endoscopic retrieval may therefore be associated with rare but significant potential risks. Our second point of concern is the optimal timing and consequent cost-effectiveness of treatment. Conservative care does not ensure the complete healing of the fistula in all patients, and it has the disadvantage of the uncertainty of the treatment outcome. In contrast, endoscopic intervention using metallic clips has the advantage of definite and earlier blockage of the fistulous tract, as well as being a safe method of removing the tube without causing more mucosal damage [4]. In our case, after confirmation of the fistula by immediate fistulography using Gastrografin (Schering, Berlin, Germany), gastroscopy revealed a yellowish discharge from the PEG site. We recognized that it was more important to close the gastrocolic fistula rather than the colocutaneous fistula, as the gastrocolic fistula might provide a persistent source of clinical symptoms and delayed wound healing. In addition, we assumed that bulky and hasty bowel preparation for colonoscopic retrieval of the retained tube might accelerate dehiscence of the fistulous tract and frank peritonitis. To achieve preferential closure of the gastrocolic fistula, we allowed 10 days of conservative care. As Gyökeres et al. point out, however, endoscopic clipping of the gastrocolic fistula before conservative care may shorten the hospital stay. Although regrettably we did not do this at the time, we believe that this clipping method would also provide earlier complete healing of the gastrocolic fistula and make rapid colonoscopic retrieval of the retained tube possible. Particularly in view of the fact that patients who have undergone a PEG procedure are generally in poor health and malnourished, and sometimes have long-lasting fistulas, early closure of the fistulous tract using endoscopic techniques is a precondition for treatment. In conclusion, we agree with some of the points made by Gyökeres et al. - that classic conservative treatment instead of endoscopic intervention may also be a good solution for the treatment of fistulas after PEG, and that endoscopic closure with a metallic clip may also be suitable for gastrocolic fistulas. The conservative and endoscopic therapeutic strategies need to be compared with regard to their appropriateness and cost-benefit considerations, and the findings need to be confirmed in larger prospective case studies; at present, it may be concluded that all the possible individual approaches are still open.

References

• 1 Coventry B J, Karatassas A, Gower L, Wilson P. Intestinal passage of the PEG end-piece: is it safe.  J Gastroenterol Hepatol. 1994;  9 311-313
  PubMedGoogle Scholar
• 2 Lambertz M M, Earnshaw P M, Short J, Cumming J G. Small bowel obstruction caused by a retained percutaneous endoscopic gastrostomy gastric flange.  Br J Surg. 1995;  82 951
  PubMedGoogle Scholar
• 3 Steinberg R M, Madhala O, Freud E. et al . Skin level division of percutaneous endoscopic gastrostomy without endoscopy retrieval: a hazardous procedure.  Eur J Pediatr Surg. 2002;  12 127-128
  Article in Thieme ConnectPubMedGoogle Scholar
• 4 Marshall J, Bodnarchuk G, Barthel J. Early accidental dislodgment of PEG tubes.  J Clin Gastroenterol. 1994;  18 210-212
  PubMedGoogle Scholar

D. K. Lee, M.D. 

Dept. of Internal Medicine · Wonju Christian Hospital · Yonsei University Wonju College of Medicine

162, Ilsan-Dong · Wonju, 220-701 · South Korea

Fax: + 82-33-745-6782

Email: gidept@wonju.yonsei.ac.kr