This is a further meta-analysis of eleven observational studies to evaluate the effect of coffee on the occurrence and progression of NAFLD. The present meta-analysis suggested that coffee did not significantly decrease the occurrence and progression of NAFLD in the participants being obese or general. In addition, different kinds of coffee failed to show any beneficial efficacy on NAFLD.
The main finding of our meta-analysis, seems to contradict the end points of previous meta-analysis [20], which was performed to appraise the influence of the amount of coffee consumption on NAFLD occurrence. The previous meta-analysis included seven articles, resulting in 4825 cases, and revealed that coffee intake level more than 3 cups was observed lower risk of NAFLD than < 2 cups per day, in other words, the risk of NAFLD was inversely associated with coffee consumption. Our analysis did not contain three studies[23–25], included in the previous meta-analysis, because one of them provided data on chronic liver disease not NAFLD[23], and the data in the study [24] was inconsistent, as well as other one study[25], was just an abstract from a meeting, which did not give sufficient information to calculate. Besides, the present meta-analysis comprised other seven studies [8, 9, 11–14, 17]. In addition, compared with the previous meta-analysis, the diagnosis of NAFLD in the present meta-analysis was made by liver biopsy in four additional articles, which was more exact than ultrasonography. What’s more, more statistical methods were used in the present meta-analysis. The data was analyzed by binary variables and continuous variables.
In an attempt to produce robust results, we formulated rigorous inclusion criteria and included studies that clearly stated the amount of coffee consumption and specifically diagnosed of NAFLD or liver fibrosis in our meta-analysis. Using four articles from the previous meta-analysis as a base, we included seven other studies. According to 7093 cases and 20625 non cases, the present meta-analysis suggested that coffee drinking did not significantly decrease the occurrence and the progression of NAFLD. Binary variables including four articles showed the pooled ORs: 0.84 (95% CI: 0.43–1.62), suggesting not a significant relationship between occurrence of NAFLD and regular coffee drinking. Subgroup analysis of obese patient and general population showed the same result in the binary meta-analysis. It was consistent with the conclusion of Shen et al [19]. Also the continuous variables including seven articles illustrated the pooled Std. Mean difference IV. Random. was − 0.12(95%CI :-0.31-0.07), which showed not statistically significant connection between coffee drinking and the progression of NAFLD. And subgroup analysis of progression of NAFLD, liver fibrosis, revealed the pooled SMD IV. Random. was − 0.11(95%CI :-0.47-0.24), showing the same result. In the continuous meta-analysis, other subgroup analysis of methods of diagnosis of NAFLD, liver biopsy and ultrasonography, types of coffee, regular coffee and espresso coffee, suggested non beneficial efficacy of coffee drinking on NAFLD. That was different from the finding of Shen et al and Chen et al, while supported by Veronese et al, Graeter et al. Moreover, exclusion of any single study and sensitivity analyses based on various exclusion criteria did not materially alter the pooled results, which adds robustness to our main finding.
Although there were many different aspects, they had some similarities. The occurrence of NAFLD was the primary outcome and was consistently reported in many of the included studies in both the previous meta-analysis and our meta-analysis. These two meta-analyses showed not a significant association between the occurrence of NAFLD and coffee drinking within 2 cups per day. This was the same with the outcome form Shen et al[19], demonstrating that total caffeine intake is not associated with the prevalence or hepatic fibrosis of NAFLD. In addition, people in the two analysis did not the same race, from different regions. So the outcome was more creditable for clinic recommendation.
Some epidemiologic studies do not have the same suggestion, and different authors in meta-analysis have inconsistent conclusions, although coffee is regarded as a nonpharmacological tool of primary and secondary prevention of NAFLD [7]. In other words, it is not clear that whether the relevance of coffee consumption and NAFLD is very close.Therefore, we performed the further meta-analysis with more studies and statistical analysis technique to verify the relationship of coffee and NAFLD. And the present meta-analysis suggested that coffee did not significantly decrease the occurrence and progression of NAFLD in the participants being obese or general. In addition, different kinds of coffee failed to show any beneficial efficacy on NAFLD. Therefore, coffee is not recommended for people to decrease the occurrence and slow the progression of NAFLD.
One potential limitation of the present meta-analysis was the various diagnostic criteria of NAFLD used among studies, liver biopsy and ultrasonography. Another limitation was that we were unable to assess the components of coffee on NAFLD, such as caffeine, sugar. Then, further studies should focus on the following points. Firstly, it is necessary to pay more attention on components, timing, and duration of coffee for the occurrence and progression of NAFLD; to date, a great variability exists in the literature. Secondly, early diagnosis is crucial for the the occurrence and progression of NAFLD, so the liver biopsy is essentially used to diagnose NAFLD.
In conclusion, the emerging evidence showed coffee drinking did not significantly decrease the occurrence and progression of NAFLD in the participants. In addition, different kinds of coffee failed to show any beneficial efficacy on NAFLD. Thus, coffee is not recommended for people to decrease the occurrence and slow the progression of NAFLD. Moreover, the results of this meta-analysis should be interpreted with caution because we unable to assess the components of coffee on NAFLD. Further study is needed to assess to the influence of main ingredient of coffee, caffeine, on NAFLD.