Endoscopic actionable findings in acute coronary syndrome with gastrointestinal bleeding: overt versus occult presentations
N Mahachai, P Pengchata, T GeeratragoolAbstract
Background
Acute coronary syndrome (ACS) requires prompt antithrombotic therapy, which increases the risk of gastrointestinal bleeding (GIB). When GIB occurs during ACS hospitalisation, clinicians face a therapeutic dilemma between bleeding control and continuation of cardioprotective treatment. Although endoscopy is commonly performed in this setting, the clinical yield and actionable findings may differ between overt and occult bleeding presentations. Data comparing these phenotypes in patients with ACS remain limited.
Purpose
To compare endoscopically actionable findings between ACS patients presenting with overt versus occult GIB.
Methods
We conducted a single-centre retrospective cohort study at Siriraj Hospital between 2011 and 2024. The study included adults admitted with ACS who were diagnosed with gastrointestinal bleeding and underwent endoscopy within 30 days. Patients were classified as overt or occult GIB. The primary endpoint was a pragmatic composite of endoscopically actionable outcomes at index endoscopy: endoscopic haemostasis, high-risk bleeding lesions, or a management-change lesions (defined as gastrointestinal cancer/tumour, inflammatory bowel disease, or colitis). Multivariable logistic regression was performed to identify predictors of the composite and management-change endpoints.
Results
Among 119 patients (overt n=84; occult n=35), the composite actionable endpoint occurred in 34.5% overall and was more frequent in overt versus occult GIB (40.5% vs 20.0%; OR 2.72, p=0.032). The difference was driven mainly by management-change lesions (19.0% vs 2.9%; OR 8.00, p=0.021), whereas high-risk bleeding lesions were similar between groups (16.7% vs 11.4%; p=0.467). In adjusted analyses, overt presentation remained independently associated with management-change lesions (Firth aOR 6.31, p=0.030) and showed a borderline association with the composite endpoint (Firth aOR 2.59, p=0.053).
Conclusion
In ACS patients undergoing endoscopy within 30 days of GIB diagnosis, actionable endoscopic findings were common in both bleeding phenotypes. Overt GIB was associated with a substantially higher likelihood of management-changing pathology, supporting timely endoscopic evaluation when clinically indicated.revisedSlide2