Backgrounds/aims: We performed a meta-analysis of observational epidemiological studies to examine the association between oral contraceptives (OC) and hepatocellular carcinoma (HCC).
Methods: Two independent researchers conducted PubMed searches followed by systematic abstraction of studies that compared OC use between patients with HCC and a group of controls. Pooling of ORs was conducted using a random effects model. Heterogeneity and publication bias among studies were examined.
Results: Twelve case-control studies that included 739 cases and 5223 controls met the inclusion and exclusion criteria. The pooled estimate of ORs (age- and sex-matched only) from all 12 studies was 1.57 (95% CI=0.96-2.54, p=0.07) with a heterogeneity of I(2)=39.9. Exclusion of one large multi-national European study decreased the heterogeneity to I(2)=16.9 and increased the pooled OR to 1.70 (95% CI=1.12-2.59, p=0.01). Eight studies reported adjusted ORs (in addition to age and sex); the pooled estimate was 1.45 (95% CI=0.93-2.27, p=0.11) with a heterogeneity of I(2)=20.4. Only few studies identified or adjusted for other HCC risk factors. Six studies showed a significant 2- to 20-fold increase in HCC risk with longer durations of OC use; however, the reporting was too inconsistent to allow meta-analysis.
Conclusions: The evidence is inconclusive to establish a relation between oral contraceptives and the risk of hepatocellular carcinoma. Future studies should focus on the duration, intermittency, and recency of OC use.