The contribution of genetic polymorphisms to bone mineral density (BMD) and fracture risk in women is a controversial topic. We evaluated the effect of the XbaI and PvuII polymorphisms of the estrogen receptor a to BMD and fracture risk in a meta-analysis, including published data and additional information from investigators. Five thousand eight hundred thirty-four women from 30 study groups were analyzed with fixed and random effects models. The PvuII polymorphism was not associated with BMD at any skeletal site examined and 95% CIs exclude effects over 0.015 g/cm2 for both the femoral neck and the lumbar spine. Conversely, XX homozygotes (women carrying two copies of the gene variant without an XbaI restriction site) consistently had higher BMD than other subjects. The magnitude of the effect was similar in the femoral neck and lumbar spine (0.014 g/cm2 [95% CI, 0.003-0.025] and 0.015 g/cm2 [95% CI, 0.000-0.030], respectively; no between-study heterogeneity for either). Total body BMD was also significantly higher in XX homozygotes (by 0.039 g/cm2 and 0.029 g/cm2 compared with Xx and xx, respectively). Available data on fractures suggested a protective effect for XX (odds ratio [OR], 0.66 [95% CI, 0.47-0.93] among 1591 women), but not PP (OR, 0.93 [95% CI, 0.72-1.18] among 2,229 women). In summary, we have found that XX homozygotes may have higher BMD and also a decreased risk of fractures when compared with carriers of the x allele, whereas the PvuII polymorphism is not associated with either BMD or fracture risk.