Endoscopic YAG-laser treatment in massive upper gastrointestinal bleeding. Report of a controlled randomized study.

In patients with massive upper gastrointestinal bleeding the results of endoscopic control by coagulation with YAG-laser have been compared with the results of an active surgical approach with early operation in a controlled randomized study. The laser coagulation was done at admission in patients in whom an actively bleeding lesion was diagnosed at endoscopy. Of the 66 patients belonging to the laser group, 23 were bleeding at admission and endoscopy, and in 15 of these coagulation was attempted and initial haemostasis was achieved in 14. Seven of the latter rebled and five of them were then operated upon with on postoperative death. Two patients died from bleeding oesophageal varices and hepatic failure. In the patient in whom haemostasis was not achieved, an operation was done for a duodenal ulcer. She died postoperatively. Of the 43 patients belonging to the laser group and not bleeding at endoscopy, nine patients rebled and three of them were operated upon, the with on postoperative death. An additional four patients died in this group, three of bleeding oesophageal varices in combination with hepatic failure and one of bleeding from an unknown source. Sixty-nine patients belonged to the control group and in 19 of these patients bleeding lesions were diagnosed at endoscopy. Five of these were operated upon, with two postoperative deaths. An additional three patients died of bleeding oesophageal varices and hepatic failure. Of the 50 patients belonging to the control group and not bleeding at endoscopy, eight later showed signs of recurrent bleeding. Four of them had oesophageal varices and two died. The other four with recurrent bleeding were operated upon with no postoperative mortality. Thus nine patients (five with oesophageal varices) died in the laser group and seven (five with oesophageal varices) in the control group. There was no statistically significant difference between the two groups regarding mortality, need of blood transfusion or time of treatment in the hospital. However, the material is too small to make any definite conclusions, and further studies are necessary in which laser treatment is compared with an aggressive surgical policy with early operations.