Objective To explore the relationship between periodontitis and postmenopausal osteoporosis.Methods Databases were electronically searched from PubMed (1966 to December, 2010), EMbase (1974 to December, 2010), CBM (1978 to December, 2010), VIP (1989 to December, 2010), CNKI (1979 to December, 2010) and WanFang Data (January, 2007 to December, 2010), and the references listed in all papers were also retrieved. The literature was screened according to the inclusion and exclusion criteria by two reviewers independently; the methodology quality was evaluated after data abstraction; and then the RevMan 5.0 software was used for meta-analyses. Results Four trials were included. Among the total 678 patients involved, 263 were postmenopausal osteoporosis patients, while the other 415 were non-osteoporosis patients. The results of meta-analyses showed that: a) Clinical attachment loss (CAL) of the postmenopausal osteoporosis patients was significantly higher than that of the non-osteoporosis patients (WMD=0.60, 95%CI 0.23 to 0.96); b) The level of gingival recession of the postmenopausal osteoporosis patients was significantly higher than that of the non-osteoporosis patients (WMD=0.78, 95%CI 0.41 to 1.14); c) There were no significant differences in plaque index (PI), gingival index (GI) and periodontal probing depth (PPD) between the two groups (WMD=0.17, 95%CI 0.00 to 0.35; WMD=0.05, 95%CI –0.09 to 0.19; and WMD=–0.08, 95%CI –0.24 to 0.09); d) The results of one study indicated that the rate of periodontitis in the postmenopausal osteoporosis patients was higher than that of the non-osteoporosis patients (OR=2.45, 95%CI 1.38 to 4.34, Plt;0.01); the severe alveolar crest height loss was related to osteoporosis (OR=4.20, 95%CI 1.57 to 11.22, Plt;0.01). Conclusion Postmenopausal osteoporosis patients are more prone to suffer from periodontitis or turn to the worse stage of periodontitis. In consideration of the factors such as small scales and incomplete measure indexes of the included studies, which have influences on the intensity and comprehensiveness of this conclusion, more high-quality studies are required.