Risk factors for upper gastrointestinal bleeding among aspirin users: An old issue with new findings from a population-based cohort study
Pei-Jui Luo,Xi-Hsuan Lin,Chung-Chi Lin,Jiing-Chyuan Luo,Hsiao-Yun Hu,Po-Hsiang Ting,Ming-Chih Hou
DOI: https://doi.org/10.1016/j.jfma.2018.10.007
Abstract:Background/purpose: We aimed to identify the risk factors of first-time occurrence of non-variceal upper gastrointestinal bleeding (UGIB) among aspirin users after adjusting for confounding factors like age, gender, underlying co-morbidities, and medications. Methods: Using the National Health Insurance Research Database of Taiwan and matching age, gender, underlying co-morbidities and enrollment time by propensity score, 11105 aspirin users and 11105 controls were identified for comparison from a cohort dataset of 1,000,000 randomly sampled subjects. Cox proportional hazard regression models were used to identify independent risk factors for first-time occurrence of non-variceal UGIB in the study cohort and in the aspirin users after adjusting for age, gender, underlying co-morbidities, and medications (e.g., non-steroidal anti-inflammatory drugs [NSAIDs], cyclooxygenase-2 [COX-2] inhibitors, steroids, thienopyridines, selective serotonin reuptake inhibitors, warfarin, and dipyridamole). Results: By Cox proportional hazard regression analysis, aspirin use increased the risk of first-time occurrence of UGIB (hazard ratio [HR]: 1.48; 95% confidence interval [CI]: 1.28-1.72). Age, male gender, Helicobacter pylori (H. pylori)infection, diabetes, chronic kidney disease (CKD), cirrhosis, history of uncomplicated peptic ulcer disease (PUD), and use of NSAIDs, COX-2 inhibitors, steroids, and thienopyridines were independent risk factors for UGIB among aspirin users. Conclusion: In addition to age, male gender, H. pylori infection, and concomitant use of NSAIDs, COX-2 inhibitors, steroids, and thienopyridines, underlying co-morbidities including diabetes, CKD, cirrhosis, history of PUD are also important risk factors for first-time occurrence of non-variceal UGIB in aspirin users.