Green Tea Consumption and Liver Disease

Xi Jin; Ruo-heng Zheng; You-ming Li


Liver International. 2008;28(7):990-996. 

In This Article


After a thorough literature search of the Western and Chinese literature, a total of 472 articles were identified. However, most of them were experiments on rats, out of which, 442 were excluded by title screening. Abstracts of the remaining 30 articles were reviewed and 17 articles were retrieved in full text, where review articles, studies without a control group, wrong study design and other inappropriate studies were excluded after sufficient discussion between two independent investigators (Figure 1).

Figure 1.

Summary of article selection process.

Generally, 10 enrolled studies[19,20,21,22,23,24,25,26,27,28] included four randomized-controlled clinical trials (RCTs), two cohort, one case-control and three cross-sectional studies that provided data on the association between green tea consumption and liver diseases ( Table 1 ). Among them, eight studies were conducted in China, one in Japan and the other in the USA. In terms of study population, eligible subjects consisted of men and women, except for four studies that recruited only men.[19,20,22,25] Besides, most of the studies used population-based controls and regular green tea consumption was recorded. Nevertheless, one study[21] used subjects from a population who were Hepatic B virus surface (Hbs) antigen (Ag) carriers while another study[20] recruited subjects who were Hbs-Ag and α-fetoprotein (AFP) positive. Furthermore, most of the studies used adjustments such as age, sex, smoking, drinking to control potential confounders and the study periods varied from <6 months to more than 6 years. Finally, there were only four studies reporting the response rate[19,22,23,24] of subjects, which varied from 85 to 98%.

For stringency of systematic review, here we presented the overall evaluation of methodological quality using criteria previously set by Downs and Black.[17] As can be seen from Figure 2, the score for each eligible study ranged from 13 to 25 and, intriguingly, the highest one is not the only double-blinded RCT[20] but the cohort study conducted by Ruhl and Everhart,[26] which, to some extent, reflected the relatively low quality of Chinese studies in general. Nevertheless, the rest of the studies were scored from high to low mainly following the type of study as an RCT, cohort, case-control and cross-sectional study.

Figure 2.

Scores for the assessment of methodological quality of enrolled studies.

It is worth briefly introducing the study conducted by Ruhl and Everhart.[26] In this cohort study, the authors gathered data from participants in the first National Health and Nutrition Examination Survey (1971-1975) and found the protective role of increased coffee and tea intake against chronic liver disease as described in Table 1 and Table 2 . To overcome the limitation of insufficient information, they conducted another detailed survey in 1982-1984 and found significant protective effects of coffee but not tea, although the latter also showed a protective tendency. Also, the effect of protection was limited to persons at a higher risk for liver diseases including heavier alcohol intake, overweight, diabetes or high iron saturation. Despite scoring the highest among the selected articles, there are also several limitations of this paper that should be acknowledged. The first and most important one is the mixed evaluation of coffee and tea and obviously the former was much more emphasized, which underweighed this paper in terms of the current theme. Besides, the aetiology of the chronic liver disease was not given and relying on hospitalization and death certificate diagnoses may underestimate the incidence of chronic liver disease.

As demonstrated in Table 2 , the studies included here were published between the years 1995 and 2005, with numbers of subjects ranging from 52 to 29 090 [this figure was extracted from the cross-sectional studies conducted by Shen et al.[22]] and a combined total of 38 990. The effect of green tea consumption on liver disease was presented by RR or OR whereas the P-value was displayed if the former markers were unavailable ( Table 2 ). Among 10 studies, eight yielded statistically significant results showing a protective role of green tea against liver disease but two studies[20,24] only showed a partial tendency. Also, four studies[22,23,24,25,26] showed a positive association between green tea intake and attenuation of liver disease. Moreover, the dosage and duration of tea consumption also differed in different trials and three studies were sponsored by local pharmaceutical companies.[20,27,28] However, further meta-analysis assessing the summary effects of green tea was precluded because of study heterogeneity caused by the existence of varied disease outcomes, green tea dosage, study design and so on.

Because we investigated the association between green tea consumption and liver diseases, the outcomes varied as incidence of primary liver cancer in two studies,[20,24] mortality of primary liver cancer in two studies,[19,22] incidence of liver disorders[25] and cirrhosis[21] in two studies, incidence of chronic liver disease in one study[26] and incidence[23] plus attenuation[27,28] of fatty liver disease in three studies. When considering the protective effects of green tea against subgroups of liver diseases, it seems that they are more effective in fatty liver disease and liver disorders than in liver cancer as two studies involving liver cancer[20,24] did not show statistical protective effect. In general, the common limitation of all studies is the lack of concealment, even in the only double-blinded RCT. Other methodological weaknesses included mixed evaluation of tea and coffee effects on liver disease,[26] no comparison of subjects' characteristics between sample and control groups,[28] no randomization,[21,22,23,24,25,26] selection bias[22,24] and so on.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.