Angiographic intervention in patients with gastrointestinal bleeding

Purpose To assess the clinical effectiveness of angiography and transcatheter intervention in patients with gastrointestinal (GI) bleeding Materials and Methods Clinical records of 76 patients who underwent angiography and embolization for gastrointestinal bleeding were reviewed after obtaining IRB approval. Patient demographics, history, angiographic findings, treatment and outcomes were recorded. The technical success, 24hr and 30day rebleeding and mortality rates were calculated. Multivariate analyses were performed to assess the factors associated with 24hr and 30day rebleeding. Results There were 53 men and 23 women with a mean age of 66 years. 22 patients (29%) underwent endoscopy prior to angiography. All endoscopies were positive: 15/22 (68%) patients with upper GI bleeding and 7/22 (32%) of patients with lower GI bleeding. Tc 99m labeled RBC scan, available in 7 (10%), was positive for bleeding in 6 (85%). Patients received blood transfusions (n=47, 62%) and vasoconstrictors (n=12, 16%) prior to angiography. Positive angiographic findings were seen in 58 (active contrast extravasation in 43 (56.5%), and pseudoaneurysm in 15 (20%). Angiography was negative for GI bleeding in 18. Embolization was performed in 75 (99%). This included empiric embolization in 17. The agents used were coils (n=35, 46%), gel foam (n=12, 16%), microparticles (n=14, 18%), glue (n=1, 1%) or a combination of these (n=13, 17%). Embolization was technically successful in 74 (98.6%). 24 hr and 30 day rebleeding rates were 11 (14.5%) and 14 (18.5%) respectively. Of the 11 who had 24 hr rebleeding, 7 required surgery and two had successful repeat embolization. Of the 14 who had rebleeding at 30 days, 5 required surgery, 5 underwent successful repeat embolization. 24 hr and 30 day mortality rates were 7 (9%) and 7 (9%) respectively. Ischemic complications following embolization were seen in 3 (4%) of which two required surgery. Prior GI bleeding was an independent variable associated with 30 day rebleeding. Conclusion Angiographic intervention has high technical success and acceptable clinical success with minimal complications for patients presenting with GI bleeding. Repeat intervention may increase overall clinical success.