This study represents a European prospective clinical multicenter trial and was undertaken to evaluate the applicability of the biofragmentable anastomosis ring (BAR) as a routine anastomotic tool in teaching hospitals. The trial results analyzed consisted of 1666 BAR anastomoses performed in 1360 patients from March 1989 to May 1996 in the upper (1042 anastomoses) and lower (624 anastomoses) gastrointestinal (GI) tract. Only patients selected for elective procedures and having previously undergone orthograde bowel cleansing were entered into the trial. In the upper GI tract six anastomoses (0.58%) developed clinically relevant and radiologically detectable leaks with indications for reoperation. In the lower GI tract 42 (6.73%) anastomoses showed a radiologically detectable leak with clinical manifestations in 28 cases (4.48%). Reoperation was performed in 18 cases (2.80%). The overall leakage rate with clinical relevance was 2.04%. Three gastrojejunostomy episodes of bleeding were observed (0.18%) at the BAR anastomotic site. During the early postoperative course there was no ileus due to obstruction of a BAR anastomosis. Reintroduction of diet after the operation was not delayed. In two centers a follow-up evaluation reported no BAR-related late anastomotic stenoses. There were no intraoperative deaths, but 54 patients died postoperatively. Peritonitis following anastomotic leakage was responsible for postoperative deaths in four cases; three of them were related to BAR anastomoses. In conclusion, the BAR anastomotic procedure is an established, rapid, simple to learn, highly standardized, safe technique with the advantage of no persistent foreign material in the anastomotic region and therefore no induction of stenosis. At present, the application of anastomoses in various segments of the GI tract, from the stomach to the middle third of the rectum, can be recommended.