The factors that influence the risk of accidents during the replacement of gastrostomy catheters remain unknown and therefore have not yet been thoroughly investigated. We conducted a nationwide questionnaire survey of 415 rehabilitation-training facilities for the replacement of gastrostomy catheters. We received 221 valid responses. Among the catheter replacement methods submitted, the bumper button replacement method was the most widely chosen, comprising 40% of the valid responses. The measures used to prevent accidental erroneous catheterization included examination of the stomach contents, endoscopic examination of the stomach, and the detection of insufflation sounds, although these measures varied widely among the facilities. Fifty-one out of the 221 facilities that responded to the survey experienced various mishaps, of which 20 were due to erroneous catheterization. In approximately 40% of the facilities, there was no operative manual for the replacement nor was the patient's consent taken before performing the procedure. This investigation elucidates the risks involved in the replacement of gastrostomy catheters. This survey also suggests that the methods used for catheter replacement should be re-examined to prevent accidents during the replacement.

We experienced 4 cases of graft-enteric fistulas in which 2 cases were died due to intestinal hemorrhage and the other 2 cases are survived by the surgical treatment of infected graft excision with extra-anatomical bypass or anatomical bypass grafting with omental wrap. The most important thing is to prevent this late complications at the first operation by wrapping of graft with the surrounding tissues or omentum.

Objective：To investigate the characteristics of studies registered in the field of rehabilitation medicine.Methods：The university hospital medical information network clinical trials registry database was searched for domestic clinical trials associated with rehabilitation medicine that were registered after June 2005. We extracted information about studies and analyzed their registration trends and overall characteristics.Results：Among the 21,410 registered trials, we found 529 trials associated with rehabilitation. The purpose of this study was to investigate efficacy in 65％ of the studies. Among these studies, 54％ were parallel-group comparison studies, 50％ were registered retrospectively, and 85％ did not publish any results. In comparison studies, 86％ were randomized controlled studies, and 47％ were open-label trials.Conclusion：An increasing trend of registration was observed. However, we found several problems in registration. Prospective registration is important to decrease publication and outcome reporting biases. Education for the relevant study protocol and registration might improve the quality of clinical study in domestic rehabilitation medicine.