Predictors of Rebleeding after Hemoclip Therapy for Treating High-risk Bleeding Ulcers: Hemoclip Therapy Alone was Comparable to Combination Treatment with Epinephrine Injection

Oh, Hyo Jeong; Kim, Tae Hyeon; Seo, Geom Seog; Choi, Chang Soo; Cho, Eun Young; Kim, Ki Hoon; Seo, Sung O; Kweon, Ji Hye; Ryu, Han Seung; Choi, Suck Chei; Kim, Haak Cheoul; Shin, Sae Ron

Abstract

BACKGROUND/AIMS: Active bleeding and non-bleeding visible vessels in patients with bleeding peptic ulcer are associated with a high risk of rebleeding. The aim of our study was to define the risk factors associated with failure of endoscopic hemostasis and rebleeding in patients with active peptic ulcer bleeding.
METHODS
We retrospectively reviewed 119 patients (90 men and 29 women; mean age, 60.14+/-14.67 years) with active peptic ulcer bleeding (spurting, oozing and/or non-bleeding visible vessel) and who were treated in Wonkwang Medical Center from January 2002 to January 2007. They were classified to endoscopic hemoclipping alone group (n=75) or endoscopic hemoclipping combined with epinephrine injection group (n=44), according to the therapeutic modality.
RESULTS
Initial hemostasis was achieved in the two groups (100%), and permanent hemostasis was achieved 71.4% in all the patients. Operation was done in eight patients (6.7%), and six patients (5%) in the two groups, respectively, died within 1 month after initial hemostasis because of bleeding related complications. Recurrent bleeding, the duration of the hospital stay, blood transfusion requirements, complications and the operation and mortality rates were not statistically different between the hemoclip alone and combination groups. Univariate analysis showed that rebleeding was related to the presence of shock on admission (p=0.01), complication (p=0.00), the pulse rate (<100/min) on admission (p=0.04), single ulcer (p=0.032), the level of hemoglobin (<8 g/dL) (p=0.02) and the volume of transfusion (<3 units) after the procedure (p=0.005) in all the patients. On the multivariate analysis that was adjusted for age and gender, the hemoglobin level (<8 g/dL) (odds ratio = 10.5) was the only significant predictor for early rebleeding.
CONCLUSIONS
This result may suggest that the combination method does not provide a substantial advantage over hemoclipping alone for the hemostatic management of active peptic ulcers bleeding. A low hemoglobin level on admission may be useful to predict rebleeding after initial endoscopic hemostasis in patients with active peptic ulcer. However, this study was designed retrospectively, so the comparison between these two groups should be re-evaluated prospectively in a large, multicenter trial.