%0 Journal Article
%T Overt gastrointestinal bleeding in patients with cancer: Clinical characteristics and outcomes.
%A Kerbage A
%A Hamadeh N
%A Hashash JG
%A Rockey D
%A Barada K
%J Am J Med Sci
%V 0
%N 0
%D 2024 May 31
%M 38825073
%F 3.462
%R 10.1016/j.amjms.2024.05.023
%X <B>BACKGROUND: </B>The aim of this study was to compare the clinical characteristics and outcomes of gastrointestinal bleeding (GIB) between cancer patients (CP) and non-cancer patients (NCP).<BR><B>METHODS: </B>This was a prospective study of patients admitted with overt GIB between 2013 and 2021. GIB etiology, management and outcomes including rebleeding and mortality, were compared between CP and NCP, and among patients with different types of cancer. The associations with categorical variables were assessed with the Chi-square test, and the t-test was used for continuous variables.<BR><B>RESULTS: </B>Of 674 patients admitted for GIB, 144 (21%) had cancer. 121(84%) CP had active disease, 49% had stage 4 cancer, and 78% had solid tumors, of whom 28 (20%) had luminal GI cancers. The most common were colorectal cancer, prostate cancer, and lymphomas. Compared to NCP, CP had higher age-adjusted Charlson Comorbidity Index, and were less likely to undergo endoscopy or endoscopic therapy. Severe GIB was equally prevalent in both groups, but CP had more severe anemia. Peptic ulcer was the most common etiology in both groups. Of 28 luminal cancer patients, 17(59%) bled from their tumors. Nine patients bled from cancer metastasis to the GI lumen. CP had higher in-hospital, one-month, one-year, and end-of-follow-up mortality. Length of hospital stay and re-bleeding rates did not differ between CP and NCP.<BR><B>CONCLUSIONS: </B>CP with GIB are less likely to have diagnostic and therapeutic endoscopy and have higher mortality than NCP. Steps to identify CP at risk for GIB and to improve their outcomes merit further investigation.