Abstract
BACKGROUND: Although the current guidelines recommend endoscopic combination therapy, endoscopic epinephrine injection (EI) monotherapy is still a simple, common and effective modality for treating peptic ulcer bleeding (PUB). However, the rebleeding risk after EI monotherapy is still high, and identifying rebleeding patients after EI monotherapy is unclear, which is highly important in clinical practice. This study aimed to identify risk factors and constructed a predictive nomogram related to rebleeding after EI monotherapy. METHODS: We consecutively and retrospectively analyzed 360 PUB patients who underwent EI monotherapy between March 2014 and July 2021 in our center. Then we identified independent risk factors associated with rebleeding after initial endoscopic EI monotherapy by multivariate logistic regression. A predictive nomogram was developed and validated based on the above predictors. RESULTS: Among all PUB patients enrolled, 51 (14.2%) had recurrent hemorrhage within 30 days after endoscopic EI monotherapy. After multivariate logistic regression, shock [odds ratio (OR) = 12.691, 95% confidence interval (CI) 5.129-31.399, p < 0.001], Rockall score (OR = 1.877, 95% CI 1.250-2.820, p = 0.002), tachycardia (heart rate > 100 beats/min) (OR = 2.610, 95% CI 1.098-6.203, p = 0.030), prolonged prothrombin time (PT > 13 s) (OR = 2.387, 95% CI 1.019-5.588, p = 0.045) and gastric ulcer (OR = 2.258, 95% CI 1.003-5.084, p = 0.049) were associated with an increased risk of rebleeding after an initial EI monotherapy treatment. A nomogram incorporating these independent high-risk factors showed good discrimination, with an area under the receiver operating characteristic curve (AUROC) of 0.876 (95% CI 0.817-0.934) (p < 0.001). CONCLUSIONS: We developed a predictive nomogram of rebleeding after EI monotherapy, which had excellent prediction accuracy. This predictive nomogram can be conveniently used to identify low-risk rebleeding patients after EI monotherapy, allowing for decision-making in a clinical setting.