Improved outcomes in patients with non-variceal upper gastrointestinal bleeding while undergoing antithrombotic therapy: A propensity score matching analysis

BACKGROUND

This study sought to examine the effect of antithrombotic use on clinical outcomes in non-variceal upper gastrointestinal bleeding (UGIB).

METHODS

Patients consecutively diagnosed with non-variceal UGIB between February 2019 and September 2020 were divided into two groups based on their antithrombotic use: users and non-users. Using propensity score matching (PSM) and multivariable regression analyses, the impact of antithrombotic use prior to UGIB presentation on clinical outcomes was examined.

RESULTS

In the entire cohort, there were 210 and 260 patients in the antithrombotic user and non-user groups, respectively. Using PSM analysis with seven covariates, two matched groups of 157 patients were created at a 1:1 ratio. In the matched cohort, despite their longer hospital stays and a higher rate of intensive care unit admissions, the patients in the user group had lower 30- and 90-day mortality rates (4.5% vs. 14.0 %; p = 0.003 and 8.9% vs. 18.5 %; p = 0.014, respectively). In the entire cohort, multivariable analyses adjusted for confounding factors revealed that antithrombotic use was associated with lower risks of in-hospital (adjusted OR: 0.437; 95 % CI: 0.191-0.999), 30-day (adjusted OR: 0.261; 95 % CI: 0.099-0.689), and 90-day (adjusted OR: 0.386; 95 % CI: 0.182-0.821) mortality.

CONCLUSION

Antithrombotic use prior to UGIB presentation was found to be an independent protective factor for all-cause mortality.

Three Simple Parameters on Admission to the Emergency Department are Predictors for Endoscopic Intervention in Patients with Suspected Nonvariceal Upper Gastrointestinal Bleeding

BACKGROUND

A considerable number of patients with nonvariceal upper gastrointestinal bleeding (UGIB) need endoscopic intervention.

OBJECTIVE

The aim of this study was to determine factors that predict the need for endoscopic intervention at the time of admission to the emergency department.

METHODS

Consecutive patients with International Classification of Diseases, Tenth Revision diagnosis code K92.2 (gastrointestinal hemorrhage) who underwent upper endoscopy between February 2019 and February 2022, including patients diagnosed with nonvariceal UGIB in the emergency department in the study were reviewed retrospectively. The patients were divided into two groups: those treated endoscopically and those not treated endoscopically. These two groups were compared according to clinical and laboratory findings at admission and independent predictors for endoscopic intervention were determined using multivariate regression analysis.

RESULTS

Although 123 patients (30.3%) were treated endoscopically, endoscopic treatment was not required in 283 (69.7%) patients. Syncope, mean arterial pressure (MAP), and blood urea nitrogen (BUN) at admission were independent predictors for endoscopic intervention in the multivariate analysis, after adjusting for endoscopy time. The area under the curve of the syncope+MAP+BUN combination for endoscopic intervention was 0.648 (95% CI 0.588-0.708). Although the syncope+MAP+BUN combination predicted the need for intervention significantly better than pre-endoscopy Rockall and AIMS65 scores (p = 0.010 and p < 0.001, respectively), there was no significant difference in its comparison with the Glasgow-Blatchford score (p = 0.103).

CONCLUSIONS

Syncope, MAP, and BUN at admission were independent predictors for endoscopic therapy in patients with nonvariceal UGIB. Rather than using complicated scores, it would be more practical and easier to predict the need for endoscopic intervention with these three simple parameters, which are included in the Glasgow-Blatchford score.