We performed this research to determine the prevalence of low bone mineral density in patients given levothyroxine and to investigate the effects of replacement and suppressive doses of levothyroxine, age, body mass index, and TSH level on bone mineral density. One hundred and ten Asian women taking levothyroxine for at least six months were grouped into levothyroxine replacement group (levothyroxine dose that will maintain TSH level of 0.5-5.0 ulU / mL) and TSH-suppressive group (dose that will maintain TSH level of less than 0.5 uIU/mL). Bone mineral density (BMD) was measured by dual-energy x-ray absorptiometry at the lumbar spine, femoral neck and total hip. The prevalence of low bone mineral density on levothyroxine replacement and TSH-suppressive groups were 58 % and 45.8%, respectively. In the levothyroxine replacement group significant associations between levothyroxine dose, as well as body mass index, and low BMD in both spine and femoral neck were detected. An inverse relationship between duration of therapy and low bone density was also observed. Even normal TSH level in this group correlated with low bone mass in the total hip. Age and menopausal status were also significantly associated with low bone density. In the group of patients taking TSH-suppressive doses of levothyroxine, only age showed an inverse correlation with BMD. This may be secondary to the small population size generated for this group.