Endoscopic injection of adrenaline for severe peptic ulcer haemorrhage in high surgical risk patients.

Endoscopic adrenaline-hypertonic injection was attempted in 40 patients admitted for oesophagogastroduodenal ulcer haemorrhage unresponsive to conventional medical treatment and presenting with severe underlying disease or advanced age (less than 80 years). The results were compared with our own historical controls (43 patients) treated by conventional therapy, meeting the same inclusion criteria. Permanent haemostasis was achieved in 32 patients in the injection group and 30 in the control group (NS) but emergency surgery was less frequent in the injection group (2 vs 25, p less than 0.001). Blood transfusion requirements were less in the injection group (8.5 +/- 6.2 vs 10.2 +/- 5.4, p less than 0.05) but length of hospital stay was not really different (15.7 days +/- 9.3 vs 20.9 +/- 14.4). Unfortunately, mortality was not reduced in the injection group (14/40 vs 17/43). Two lethal complications attributable to injection treatment occurred. This treatment could represent an alternative to conventional haemostatic treatment in high surgical risk patients with severe clinical bleeding, avoiding emergency surgery. In spite of the fact that we selected high-risk patients, endoscopic treatment was not able to lower the mortality (about 37%). Due to severe unpredictable side effects and potential risks of long-term massive rebleeding, this treatment should be performed electively in patients with severe clinical bleeding, as first line treatment when surgical risk factors exist or immediately before surgery in low risk patients.