Diagnosis and percutaneous treatment of gastrointestinal hemorrhage. Long-term experience.

Objective: to report our experience in the diagnosis and treatment of gastrointestinal hemorrhage.
Method: from April 1987 to April 1997, 196 patients with gastrointestinal hemorrhage (134 men and 62 women) were studied. 165 (84%) were diagnosed as presenting upper gastrointestinal hemorrhage, and 31 (16%) presented lower gastrointestinal hemorrhage. The patients were studied with endoscopy and arteriography, and embolization was prescribed in 131 (67%). Patients with bleeding from esophageal varices were excluded from this study.
Results: a bleeding point was identified angiographically in 33% (n = 65) patients. 131 (67%) patients were treated with therapeutic embolization, which was successful in 89% (n = 116) patients. The bleeding was resolved in 80% (n = 93) of the patients. Complications included arterial spasm (n = 12), pain (n = 24), coil migration (n = 8), allergic reaction (n = 2) and celiac trunk dissection (n = 2). During follow-up 16 patients presented rebleeding that stopped after reembolization in 9 cases, whereas in 7 cases surgery was needed.
Conclusions: in our experience, diagnostic angiography and percutaneous therapeutic embolization are effective, less aggressive methods that lead to few complications. Both methods have become indispensable tools in managing patients with gastrointestinal hemorrhage that does not respond to conservative therapy. Even in patients with no evidence of angiographic bleeding, embolization in selected patients is successful.