Abstract
An increasing number of observational studies have described an association between aspirin use and a reduced risk of incident hepatocellular carcinoma. We performed this meta-analysis to provide a comprehensive and updated aggregate assessment of the effect of aspirin on HCC incidence. Two independent authors performed a systematic search of the literature, utilising the Medline, Embase, Scopus, and PubMed databases. A total of 16 studies (12 cohort studies, and 4 case-control studies) were selected for inclusion, with a large number of studies excluded, due to an overlapping study population. The pooled analysis of cohort studies involving a total population of approximately 2.5 million subjects, 822,680 aspirin users, and 20,626 HCC cases demonstrated a 30% reduced risk of HCC associated with aspirin use (adjusted HR 0.70, 95%CI 0.60-0.81). There was a similar but non-significant association observed across the case-control studies (adjusted OR 0.60, 95%CI 0.32-1.15, p = 0.13), which involved a total of 1961 HCC cases. In a subgroup meta-analysis of patients with cirrhosis, the relationship between aspirin use and incident HCC diminished to non-significance (adjusted HR 0.96, 95%CI 0.84-1.09). Aspirin use was associated with a statistically significant increase in bleeding events when all relevant studies were pooled together (adjusted HR 1.11, 95%CI 1.02-1.22). Prospectively collected data should be sought, to define the optimal patient group in which aspirin is safe and effective for the chemoprophylaxis of HCC.