Predictors for outcomes after super-selective mesenteric embolization for lower gastrointestinal tract bleeding.

BACKGROUND: The optimal management of bleeding from the lower gastrointestinal tract (LGIB) remains controversial. We aim to evaluate the efficacy of mesenteric embolization for LGIB and to identify predictors for re-bleeding after the procedure. METHODS: We conducted a retrospective review of all patients who underwent mesenteric embolizations for LGIB in our institution over a 6-year period (from August 2007 to August 2013). Technical success was defined as the absence of contrast extravasation on post-embolization angiogram. Clinical success was defined as the absence of overt LGIB (clinical bleeding with >1 g/dL decrease in haemoglobin) within 30 days post-embolization. RESULTS: Mesenteric embolization was performed in 26 patients with LGIB. Technical success rate was 100%, with no occurrence of post-embolization ischaemia. Clinical success rate was 65.4%, with nine patients re-bleeding within 30 days post-embolization. Three underwent surgery, one underwent re-embolization and five were treated conservatively. Mortality rate was 19.3% (five patients), with two bleeding-associated mortalities. Site and aetiology of LGIB, platelet count and coagulation status prior to embolization, number of packed red blood cells and fresh frozen plasma transfusion were found to be predictors of clinical failure. After Bonferroni's correction (P < 0.005), platelet count of ≤140 × 10(9) /L prior to embolization was the only statistically significant factor associated with re-bleeding (odds ratio = 17.5, 95% confidence interval: 2.364-129.57; P = 0.004). CONCLUSION: Mesenteric embolization was found to be safe and effective in treating LGIB (100% technical success, no post-embolization ischaemia), with 65.4% of cases not requiring further intervention. Low platelet count prior to embolization appears to be associated with clinical failure.