Effects of anti‐thrombotic drugs on all‐cause mortality after upper gastrointestinal bleeding in Japan: A multicenter study with 2,205 cases

OBJECTS Although anti-thrombotic use is recognized as a risk factor for upper gastro-intestinal bleeding (UGIB), there has been no clear evidence that it worsens the outcomes after the bleeding. The aim of this study is to investigate the effects of anti-thrombotic agents on in-hospital mortality following UGIB. METHODS Information on clinical parameters, including usage of anti-thrombotic agents was retrospectively collected from consecutive patients with UGIB at 12 high-volume centers in Japan between 2011 and 2018. The all-cause in-hospital mortality rate was evaluated according to the usage of anti-thrombotic agents. RESULTS Clinical data were collected from 2,205 patients with endoscopically-confirmed UGIB. Six hundred forty-five (29.3%) patients used anti-thrombotic agents. The all-cause in-hospital mortality rate was 5.7% (125 deaths). After excluding 29 cases in which death occurred due to end-stage malignancy, 96 deaths (bleeding related, n=21; non-bleeding-related, n=75) were considered "preventable". Overall, the "preventable" mortality rate in anti-thrombotic users was significantly higher than that in non-users (6.0% vs. 3.7%, P<0.05). However, the "preventable" mortality of anti-thrombotic users showed a marked improvement over time; although the rate in users remained significantly higher than that in non-users until 2015 (7.3% vs. 4.2%, P<0.05), after 2016, the difference was no longer statistically significant (4.8% vs. 3.5%). CONCLUSIONS Although the usage of anti-thrombotic agents worsened the outcomes after UGIB, the situation has recently been improving. We speculate that the recent revision of the Japanese guidelines on the management of anti-thrombotic treatment after UGIB may have partly contributed to improving the survival of users of anti-thrombotic agents.