[Transcatheter embolization for endoscopically unmanageable acute nonvariceal upper gastrointestinal hemorrhage]

Evaluate the efficacy of endovascular embolization for patients with endoscopically unmanageable acute nonvariceal upper gastrointestinal hemorrhage as well as the factors that may influence mortality. MATERIALS AND METHODS. Retrospective study over a 4-year period including a historical cohort of 37 consecutive patients (22 men), with a mean age of 69.2 years (range, 22-93 years). In most cases (54%), the hemorrhage stemmed from a gastrointestinal ulcer. Technical, primary clinical, and secondary clinical success rates, as well as complication rates, were calculated. Several clinical and angiographic parameters were compared to the early mortality rate using Kruskal-Wallis or Fisher tests.

Technical, primary clinical, secondary clinical success rates, and complication rates were, respectively, 89.2%, 83.8%, 88.9%, and 10.8%. The early mortality rate was 32.4%. The APACHE II and IGS II scores were strongly correlated with mortality (p=0.001 and p=0.003, respectively).

Endovascular embolization in patients with endoscopically unmanageable acute nonvariceal upper gastrointestinal hemorrhage is effective. However, the mortality rate remains high because of the changes in the clinical condition of these patients.