Abstract
Background/Aims
Acute peptic ulcer bleeding is the most common cause of non-variceal upper gastrointestinal bleeding (NVUGIB). Endoscopic hemostasis is the standard treatment. However, various conditions complicate endoscopic hemostasis. Transarterial visceral embolization (TAE) may be helpful as a rescue therapy. This study aimed to investigate the factors associated with rebleeding after TAE.
Methods
We retrospectively investigated the records of 156 patients treated with TAE between January 2007 and December 2021. Rebleeding was defined as the presence of melena, hematemesis, or hematochezia, with a fall (>2.0 g/dl) in hemoglobin level or shock after TAE. The primary outcomes were rebleeding rate and 30-day mortality.
Results
Seventy patients with peptic ulcer bleeding were selected, and rebleeding within a month after TAE occurred in 15 patients (21.4%). Among the patients included in rebleeding group, significant increases were observed in the prevalence of thrombocytopenia (73.3% vs. 16.4%, p<.001) and ulcers >1 cm (93.3% vs 54.5%, p = .014). The mean AIMS65 (albumin, international normalized ratio, mental status, systolic blood pressure, age >65 years) score (2.3 vs 1.4, p = .009) was significantly higher in the rebleeding group. Multivariate logistic analysis revealed that thrombocytopenia (odds ratio 31.92, 95% confidence interval 6.24–270.6, p<.001) and larger ulcer size (odds ratio 27.19, 95% confidence interval 3.27–677.7, p=.010) significantly increased the risk of rebleeding after TAE.
Conclusion
TAE was effective in the treatment of patients with high-risk peptic ulcer bleeding. AIMS65 score was a significant predictor of rebleeding after TAE, and thrombocytopenia and larger ulcer size increased the risk of rebleeding after TAE.
Acknowledgements
We thank Ji Young Choi for her help with drawing figures.
Authors’ contributions
Conception and design: Jun-young Seo
Analysis and interpretation of the data: Ji Hoon Yu, Jun-young Seo
Drafting of the article: Jeong Woo Lee, Jun-young Seo
Critical revision of the article for intellectual content: Jun-young Seo, Ju Sang Park, Sang Jong Park, Sang-Jung Kim, Eun Jeong Jang, Sang Woon Park and Jae Woo Yeon
Disclosure statement
No potential conflict of interest was reported by the author(s).