Epidemiological studies of the association between tea drinking and primary liver cancer: a meta-analysis.

Many in-vitro and animal studies have provided evidence that tea has many possible anticarcinogenic mechanisms, but epidemiological evidence for the effect of tea consumption on the primary liver cancer risk remains controversial and, to date, there have been no quantitative meta-analyses reported regarding this topic. The aim of this meta-analysis is to evaluate the association between tea consumption and the risk of primary liver cancer from case-control and cohort studies. Epidemiological studies of tea consumption in relation to primary liver cancer were identified by searching MEDLINE, EMBASE, Chinese Bio-medicine Database, and Chinese Wanfang Database, from January 1979 to December 2009. The language of publication was restricted to English and Chinese. Heterogeneity and publication bias were evaluated and the pooled relative risks (RRs) were calculated using a fixed-effect model. A random-effect model was used when statistically significant heterogeneity existed. All data analyses were carried out using R software and the package 'meta'. A total of 13 epidemiological studies consisting of six case-control and seven prospective cohort studies were included. An inverse association with a borderline significance [RR = 0.77; 95% confidence interval (CI) = 0.57-1.03] was found between tea consumption and primary liver cancer. The studies with higher quality showed a similar finding with pooled RR of 0.79 (95% CI = 0.61-1.01). Both men (RR = 0.86; 95% CI = 0.77-0.95) and women (RR = 0.54; 95% CI = 0.37-0.79) showed the preventive effects of tea intake on the development of primary liver cancer. Eight studies conducted on green tea consumption suggested that it was associated with a moderate reduction in risk for primary liver cancer (RR = 0.79; 95% CI = 0.68-0.93). No publication bias was found either from funnel plot visualization or the Egger-weighted regression (P value = 0.34) and the Begg rank correlation (P value = 0.12) methods. These results provide more evidence for a protective effect of tea consumption against the development of primary liver cancer. More well-conducted and large-scale epidemiological studies are needed.