Comparison of the efficacy and safety of super-selective and selective transcatheter arterial embolization in non-variceal gastrointestinal bleeding

Jiawei ZhongKang LiChen TaiyuFoqiang LiaoXu Shu^*

• The First Affiliated Hospital of Nanchang University, Nanchang, China

Background: Vascular interventional therapy (TAE) is an effective treatment for most abdominal organ hemorrhages, especially for non-variceal gastrointestinal bleeding (NVGIB) unresponsive to endoscopy. While selective and super selective embolization are two key interventional hemostasis methods, research on their application in total gastrointestinal bleeding is limited. This study compares these two techniques to assess their safety in non-variceal gastrointestinal bleeding hemostasis. Methods: We conducted a retrospective study on patients with NVGIB who received vascular interventional therapy from August 2014 to October 2024, comparing the clinical characteristics and outcomes of patients with selective or super-selective embolization. The primary outcome was clinical success, and secondary outcomes included technical success, rebleeding (overall and within 3 days), transfusion requirements, need for additional therapies, complications, and mortality. Results A total of 116 patients with NVGIB who received vascular interventional therapy were included. Among the 88 patients with non-variceal upper gastrointestinal bleeding (NVUGIB), 48 and 40 were treated with super-selective and selective embolization, respectively. Gastroduodenal artery was the most common embolized vessel in both groups. All cases achieved technical success. 85.42% of the super-selective embolization group achieved clinical success, and 70.00% of the selective embolization group achieved clinical success (P=0.080). The rebleeding within 3 days rate of the super-selective embolization group was significantly lower than that in the selective embolization group (8.33% vs 27.50%, P=0.017). The bleeding related mortality was 6.25% in the super-selective embolization group and 7.50% in the selective embolization group. In the subgroup of 28 patients with lower gastrointestinal bleeding (LGIB), no significant differences in clinical outcomes were observed between the two embolization approaches. However, it is noteworthy that all 5 cases of post-procedural intestinal ischemia occurred in this LGIB subgroup. The embolic material used had a significant impact on early rebleeding and the subsequent need for additional therapy in LGIB(P<0.05), but not in UGIB. Conclusion: For refractory NVUGIB, super-selective TAE compared to selective TAE reduces early rebleeding. Decision-making should prioritize patient transfusion needs, which was the sole independent predictor of rebleeding. The embolization strategy for LGIB should carefully consider the choice of embolic material and the inherent risk of intestinal ischemia.