Impact of Coffee on Liver Diseases

A Systematic Review

Sammy Saab; Divya Mallam; Gerald A. Cox II; Myron J. Tong


Liver International. 2014;34(4):495-504. 

In This Article


Coffee is one of the most commonly consumed beverages in the world.[2] There is increasing evidence that daily consumption of 2–3 cups of coffee has significant health benefits. Not only has coffee been associated with a decrease in a number of liver diseases, but its consumption may also decrease mortality.[9] Thus, coffee appears to have 'hepatoprotective' health benefits.[68] Coffee is composed of over one hundred compounds, any of which could be responsible for its beneficial effects.[50] It is possible not one compound in particular, but the synergistic effect of multiple compounds, which provides the health benefits described.

Not all types of coffee may be beneficial in liver disease. Numerous studies have shown a hepatoprotective role for filtered coffee, and a potentially deleterious effect for unfiltered coffee.[26,28] It was postulated that this difference is due to the presence of kahweol and cafestol, which are caffeine diterpenes that are released from ground coffee beans but removed by paper filters.[28,69] Moreover, another study found that espresso coffee had no beneficial effect on liver disease, particularly in NAFLD.[50] In the US, filtered coffee is one of the main types of coffee consumed, whereas in Europe, espresso coffee is more commonly consumed.[50] Anty et al. postulated that perhaps espresso coffee was not found to be beneficial in NAFLD because of the sucrose added to the coffee.[50] Sucrose is composed of glucose and fructose, and fructose has been associated with increased severity of hepatic fibrosis in NASH.[79]

There are a number of proposed mechanisms for the hepatoprotective effects of caffeine ( Table 6 ). In rat studies, methylxanthine caffeine has been implicated in the hepatic fibrinogenesis pathway by (i) downregulating transforming growth factor beta-1 (TGFB-1)-induced connective tissue growth factor (CTGF) production in hepatocytes via promotion of breakdown of SMAD2 (a TGF-B effector protein), (ii) inhibition of SMAD3 phosphorylation, and (iii), by upregulation of the PPAR-gamma receptor.[70] The antioxidant hepatoprotective effects of coffee may also be induced by UDP glucoronosyltransferases (UGT1A).[71] Caffeine has also been implicated to have antifibrotic effects via its influence on hepatic stellate cells (HSC) through inhibition of focal adhesion kinase (FAK) and actin synthesis, stimulation of HSC apoptosis, induction of intracellular F-actin and cAMP expression, and via inhibition of procollagen type 1C and alpha-smooth muscle actin expression.[72]

Caffeine as well as cafestol and kahweol may have anticarcinogenic effects by upregulation of antioxidant-responsive element (ARE)-regulating signaling ( Table 6 ).[73,74] The ARE sequence plays a key role in carcinogenesis as it has been found on the promoter of numerous genes involved in detoxification processes. Furthermore, animal models and in-vitro studies indicate that kahweol and cafestol may deregulate enzymes involved in detoxification of carcinogens.[75,76] These compounds also induce glutathione-S-transferase and gamma-glutamylcysteine synthetase (GCS), leading to protection against mutagenesis, and inhibit N-acetyltransferase.[75,78]

Although caffeine is a major component of coffee, studies evaluating non-coffee caffeinated sources have revealed inconsistent evidence of hepatoprotective effects.[30,33] With regards to tea consumption, studies have found no statistically significant association between tea intake and risk of cirrhosis,[33,34] death from cirrhosis,[36] chronic liver disease,[32] HCC,[58] or death due to HCC.[56,64,80] Most studies did not specify which type of tea participants consumed. However, Inoue et al. studied green tea and Kurozawa et al. studied green, black, and oolong tea.[60,64]

Coffee preparation methods include filtered, unfiltered, and espresso, and can also vary in its roast profile (medium vs. dark). Differences in preparation method (filtered, unfiltered, espresso) as well as type of roast play a role in the composition of coffee. Filtered coffee does not contain cafestol and kahweol; however, filtration of coffee better preserves chlorogenic acids than the barista method of espresso preparation.[50] The various degrees of roast refer to the internal bean temperatures found during roasting. Darker roasts have had higher roasting temperatures. Caffeine content also varies between types of coffee [generic brewed coffee (95–200 mg per 8 oz), espresso (40–75 mg per 1 oz), generic instant coffee (27–173 per 8 oz)].

There are numerous limitations when interpreting the studies regarding the health benefits of coffee. Many of the larger studies, including those by Freedman et al., Modi et al., Hu et al., and Molloy et al., did not necessarily account for differences in socioeconomic status or other dietary factors.[31,42,49,61] Although one would argue that perhaps patients who had greater coffee intake were likely healthier, Freedman et al. found that coffee drinkers tended to have poorer overall health (P = 0.29) and vitality scores (P = 0.018) compared to non-coffee drinkers. In addition, participants who drank coffee may have had higher cigarette use and alcohol consumption.[42] Also, many studies collected data on coffee intake at only one time point, thus, the coffee intake noted may not have accurately reflected participants' intake over time.[30,34,36,38] If it is assumed that caffeine is indeed responsible for the hepatoprotective effects of coffee, then another potential limitation is the variation of caffeine content of coffee within and among coffee shops.[80] Furthermore, many studies failed to define coffee cup size.[24,30,33,36] Although it is clear that coffee intake has hepatoprotective effects, the lack of standardization of coffee cup size amongst various studies leads to ambiguity regarding how much coffee intake is necessary for these effects.

Our study is limited in that it is based mostly upon observational studies with inherent biases, including recall bias in retrospective studies, as well as selection bias and unmeasurable confounding factors amongst all non-randomized controlled studies.[81] Cross-sectional studies, such as NHANES III, are limited in that they cannot establish a temporal association between coffee intake and study findings.[30,81]

Numerous epidemiological studies suggest that consumption of approximately 3 or more cups of coffee daily will reduce the risk for and severity of hepatotoxicity due to a variety of underlying pathologic processes. While the aforementioned studies provide compelling evidence to suggest that coffee is useful as an alternative medicine in the treatment of the most common types of liver disease, blinded randomized controlled trials must be performed to provide evidence for causation, and to eliminate confounding variables and various types of bias inherent in cross-sectional, cohort, and case-control studies. Additional animal and cell culture studies are also warranted to further elucidate the biochemical basis for the potential beneficial effects of coffee in liver disease patients.