Premium
Risk of gastrointestinal bleeding and benefit from colorectal cancer reduction from long‐term use of low‐dose aspirin: A retrospective study of 612 509 patients
Author(s) -
Tsoi Kelvin KF,
Chan Felix CH,
Hirai Hoyee W,
Sung Joseph JY
Publication year - 2018
Publication title -
journal of gastroenterology and hepatology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.214
H-Index - 130
eISSN - 1440-1746
pISSN - 0815-9319
DOI - 10.1111/jgh.14261
Subject(s) - medicine , aspirin , hazard ratio , confidence interval , colorectal cancer , gastrointestinal bleeding , incidence (geometry) , population , proportional hazards model , retrospective cohort study , relative risk , cohort study , surgery , gastroenterology , cancer , physics , environmental health , optics
Background and Aim Aspirin, commonly used for prevention of cardiovascular and cerebrovascular diseases, is well known to protect against development of colorectal cancer (CRC) but increases risk of gastrointestinal bleeding (GIB). This cohort study aims to evaluate the benefit of low‐dose aspirin to prevent CRC and its associated risk of GIB. Method A population‐based dataset was used to compare incidence and mortality of CRC and GIB among patients receiving low‐dose aspirin with sex‐matched and age‐matched controls (1:2). A total of 204 170 aspirin users taking aspirin for at least 6 months and 408 339 nonusers were analyzed. Patients' clinical outcomes were documented for up to 14 years or until death. Results A total of 612 509 patients were included; 5118 (2.51%) out of 204 170 aspirin users were diagnosed with CRC; and 2073 (1.02%) died of the malignancy. On the other hand, 13 336 (3.27%) out of 408 339 non‐aspirin users were diagnosed with CRC, and 6953 (1.70%) died. Using the competing risk regression, aspirin usage significantly reduced CRC mortality (subdistribution hazard ratio = 0.59; 95% confidence interval = 0.56 to 0.62). A total of 9483 (4.64%) aspirin users developed GIB, and 820 (0.40%) died, while 11 198 (2.74%) nonusers developed GIB, and 1488 (0.36%) died. Aspirin usage marginally increased risk of bleeding‐related mortality (subdistribution hazard ratio = 1.09; 95% confidence interval = 1.00 to 1.19). Subgroup analyses showed the use of acid‐secreting agents significantly reduced aspirin‐induced mortality. Conclusion The long‐term use of aspirin reduces both incidence and mortality of CRC and at the same time increases incidence and mortality risk of GIB. With combination use of acid‐secreting agents, the bleeding risk can be reduced.