The Use of Higher Dose Steroids Increases the Risk of Rebleeding After Endoscopic Hemostasis for Peptic Ulcer Bleeding
- 147 Downloads
Previous studies have shown that several factors such as hemodynamic instability at admission are risk factors for rebleeding of peptic ulcer bleeding. However, whether steroid use increases the risk of rebleeding remains elusive.
This study aimed to clarify the risk factors for rebleeding after endoscopic hemostasis for peptic ulcer bleeding.
A total of 185 patients who underwent endoscopic hemostasis for peptic ulcer bleeding at our institution between 2005 and 2017 were retrospectively analyzed. We evaluated factors, including comorbid conditions, in-hospital onset, and steroid use, associated with rebleeding by logistic regression analysis. In addition, we investigated the association between the dose of steroids and rebleeding.
The rebleeding rate after endoscopic hemostasis for peptic ulcer bleeding was 14.6%. In the multivariate analysis, the independent risk factors for rebleeding were steroid use (odds ratio 4.56, p = 0.015), multiple ulcers (4.43, p = 0.005), number of comorbidities ≥ 3 3.18, p = 0.026), hemodynamic instability (3.06, p = 0.039), and number of comorbidities ≥ 3 (2.93, p = 0.047). Furthermore, the use of higher dose steroids (≥ 20 mg per day in prednisolone; 10.55, p = 0.002), but not lower dose (< 20 mg per day in prednisolone), was an independent risk factor for rebleeding in the multivariate analysis. The relationship between steroid use and rebleeding was observed in a dose-dependent manner (p for trend = 0.002).
This study first revealed that using higher dose steroids was an independent risk factor for rebleeding after endoscopic hemostasis for peptic ulcer bleeding, with a dose-response relation.
KeywordsSteroid Rebleeding Peptic ulcer bleeding Dose-dependent relation
Peptic ulcer bleeding
Nonsteroid antiinflammatory drugs
Proton pump inhibitor
American Society of Anesthesiologists
Percutaneous endoscopic gastrostomy
Endoscopic naso-biliary drainage
Compliance with ethical standards
Conflict of interest
All authors have no conflicts to declare.