To our knowledge, this is the first meta-analysis to evaluate the association between CYP7A1 -204A > C polymorphism and GSD and serum lipid levels. In this study, we collected data from 9 papers to evaluate the association of CYP7A1 gene polymorphisms with GSD and serum lipid levels. The results showed that -204A > C polymorphism of CYP7A1 gene related with difference in serum lipids. However, this polymorphism was not associated with GSD.
The -204A > C of CYP7A1 gene is one of the most frequently studied polymorphisms for the association with GSD. Our previous study showed that A allele of CYP7A1 gene might be considered as risk gene for GSD in Chinese patients . Later on, Juzyszyn et al.  and Strivastava et al. , using larger samples from Polish and Indian, did not confirm such association. The samples sizes in the rest studies were relatively small [12, 23]. Herein, by pooling all the previous studies, we demonstrated a lack of association between this polymorphism with GSD.
An obvious difference of gallstone prevalence between populations is present due to different ethnicities. GSD is highly prevalent in Pima Indians, Hispanic, relatively lower in Asian and the lowest in African . The frequency of A allele of -204A > C polymorphism in gallstone-free subjects is lower in Asian population, 18.09% in Chinese , high up to 60.2% in Indian . In Caucasian population, its frequency is between 52.1  and 72.28% . However, when the population was divided into Asian and Caucasian, we did not find any association of -204A > C polymorphism with GSD existed in either ethnicity.
The second aim of our study is to evaluate the association between -204A > C polymorphism and serum lipids. The -204A > C polymorphism was shown to be associated with plasma LDL-C concentrations [8, 27]. Our previous study also found that individuals with A allele tended to have lower LDL-C concentrations . While in this meta-analysis, we found that the genotype AA had significantly lower levels of LDL-C than genotype CC only in patients, but not in controls. Couture et al.  described that the genotype AC had significantly higher TG levels than the genotype CC in women and the C variant was also associated with an increased TC/HDL-C ratio in men. Hofman et al.  found a significant 34% increase of serum TG levels in genotype AA as compared with genotype CC in a healthy normolipidaemic male population. However, our meta-analysis showed that genotype AC had higher TG levels than genotype AA, allele A carriers in healthy population had higher TC levels than genotype CC, but not in the recessive models in cases, and genotype AA had higher HDL-C levels than genotype CC in controls.
Some limitations of this meta-analysis merit serious consideration. First, only the papers published in English and Chinese were included in our study. Any data reported in other languages could not be included which might bring some bias. Second, no adequate information such as source of the subjects, anti-dyslipidemia drug and etc. could be obtained in this meta-analysis. These factors might bring in several possible sources of heterogeneity. Third, most studies have recruited age > 40 years, for whom environmental factors are likely to contribute more prominently than the genetic component during the development of GSD and dyslipidemia.
In conclusion, this meta-analysis did not found any association between CYP7A1 -204A > C polymorphism and GSD. However, this polymorphism was closely related with serum lipid levels.