Aim: This meta-analysis was conducted to provide more precise evidence for association between primary biliary cirrhosis (PBC) and smoking and some other factors.
Methods: We searched the databases PubMed, EMBASE, Cochrane Library and China National Knowledge Infrastructure up to 31 December 2010. Data were extracted by two persons independently. Homogeneity of effects across studies was assessed using the χ2-test statistic and quantified by I2. Odds ratio (OR) and 95% confidence intervals (CI) were calculated based on fixed- or random-effects models. The publication bias was analyzed by Egger and Begg tests.
Results: A total of five studies were selected according to inclusion criteria. With the fixed-effects model, the pooled OR for PBC and smoking and family history of PBC were 1.67 (95% CI = 1.41–1.92) and 7.56 (95% CI = 1.90–13.22). With the random-effects model, the pooled OR for thyroid disease and urinary tract infection (UTI) were 3.08 (95% CI = 0.84–5.32) and 2.02 (95% CI = 1.40–2.65), respectively. No evidence of publication bias was observed by means of Begg and Egger tests for the factors.
Conclusion: This meta-analysis suggested that smoking, family history of PBC and UTI were strongly associated with PBC in a white population by systematic review of five existing studies, and the association remains to be validated in other populations.