1.Effectiveness and safety of epinephrine and norepinephrine for treating non-variceal upper gastrointestinal bleeding: a meta-analysis
Long GAO ; Chen CHANG ; Wenkang FU ; Wenbo MENG ; Wence ZHOU ; Xun LI
Chinese Journal of Digestive Endoscopy 2020;37(5):348-354
Objective:To systematically evaluate the effectiveness and safety of epinephrine injection and norepinephrine spraying for treating non-variceal upper gastrointestinal bleeding.Methods:Databases including the PubMed, Embase, Cochrane Library, CNKI, CBM, VIP and WanFang Data with the retrieval time from inception to December 2018 were searched to collect the related clinical trials. The references of included studies were also retrieved. Studies were screened, data were extracted, and the risk of bias was assessed by 2 reviewers separately. The meta-analysis was conducted by using RevMan 5.3 software. The grading of recommendations, assessment, development and evaluation (GRADE) quality of evidence system was used to assess the results of meta-analysis.Results:A total of 10 studies involving 884 participants were included. The results showed that compared with the norepinephrine spraying, epinephrine injection was superior in the following aspects with significant differences: the effective rate ( RR=1.21, 95% CI: 1.12-1.30, P<0.001); the one-week recurrent bleeding rate ( RR=0.28, 95% CI: 0.17-0.45, P<0.001); the immediate hemostatic rate ( RR=1.38, 95% CI: 1.25-1.52, P<0.001); and the emergent operation rate ( RR=0.35, 95% CI: 0.19-0.63, P<0.001). Conclusion:Epinephrine injection is more effective and safer for treating non-variceal upper gastrointestinal bleeding in comparison with norepinephrine spraying.
2.Front-end access regulation and liability adjustment of infringement in the middle and back-ends of medical artificial intelligence
Chinese Medical Ethics 2025;38(2):150-158
Based on the unique characteristics of medical artificial intelligence (AI) that differ from general medical devices, it was found that the clinical application of medical AI has brought different levels of impacts on the front-end access supervision and regulation, the mid-end informed consent right design, and back-end post-damage liability adjustment. In this regard, it is necessary to appropriately relax the admission review, strengthen the coordination of supervision before and after listing, and simultaneously adjust the comprehensive judgment ideas of “safe and effective” medical AI; clarify the “breadth” of notification and “depth” of disclosure regarding the patients’ informed consent rights; based on current product liability and liability for medical malpractice system, typologically adjust the allocation of liability, while strengthening the supplemental application of medical compulsory liability insurance, so as to enhance the relief of patients’ damage, rationalize risk allocation, and minimize social costs.