Diabetes drug linked to fractures

Two frequently prescribed diabetes drugs - pioglitazone and rosiglitazone - increase the risk of fractures. Patients who use these drugs for a year or longer are more than twice as likely as nonusers to sustain a fracture. Pioglitazone and rosiglitazone belong to a drug class called thiazolidinediones. The use of these anti-diabetes agents is most strongly linked to fractures of the hip and wrist, and the elevated risk is seen in both men and women, independent of age. Along with an increased fracture risk, the use of these thiazolidinediones has also been linked to reduced bone formation and accelerated bone loss. Swiss researchers studied 1,020 patients who experienced a low-trauma fracture and 3,728 matched controls and associates and sought to verify and expand on the association between thiazolidinedione use and fractures. Given that prior research had linked rosiglitazone use with fractures of the upper and lower extremities, the researchers wanted to determine if thiazolidinediones also increased the risk of fractures at other sites. In addition, they wanted to know if these risks were related to age or sex. After adjusting the data for the effects of age, body mass index, use of other diabetes agents, and other possible confounders, the use of rosiglitazone or pioglitazone for about 12 to 18 months increased the odds of fracture by 2.5 and 3 fold, respectively. Particularly high risks were noted for fractures of the hip and thigh bone, and the wrist and forearm, with increased odds of 4.5 and 3 fold, respectively. There was also evidence of a dose effect, meaning the risk increased as the dosage increased.The associations were largely unaffected by patient age or gender. No significant fracture risk was associated with other classes of oral diabetes agents. The findings, although consistent with recently reported data from a randomised trial, are based on relatively few thiazolidinediones-exposed patients and need to be confirmed by additional observational studies and by controlled clinical trials.