Thiazolidinediones (TZDs) are widely used as a treatment for diabetes and may also prevent progression to diabetes in those with impaired glucose metabolism. However, adverse event reports from clinical trials of rosiglitazone and pioglitazone, the currently available TZDs, indicate that TZD use increases fracture risk in middle-aged women. Whether these effects are seen in older women and whether TZD effects are modified by estrogen levels is not known. To address these questions, we used data from the Womens Health Initiative clinical trial and observational study. Women who self-reported a diagnosis of diabetes or use of diabetes medication during WHI were included in these analyses. Of 13,932 diabetic women, 1,802 (13%) reported any TZD use (troglitazone, rosiglitazone and pioglitazone combined) during an average of 6.95 years of follow-up. Diabetic women experienced 2,614 self-reported clinical fractures. Hormone therapy (HT) users (N=5,563) included women enrolled in the intervention arm of the HT trial and those reporting oral estrogen use at baseline. The effect of any TZD use before fracture was assessed using Cox proportional hazards models for time to first clinical fracture. Any TZD use was modeled as a time dependent covariate with date of first use determined from self report at clinic visits. Models were adjusted for age, race, education, income, BMI, history of falls, smoking, alcohol use, total calcium intake, total vitamin D intake, trial arm (Calcium/Vitamin D and Diet Modification), use of medications at baseline: bisphosphonates, estrogen, oral steroid, SERM, thiazide diuretic, statin, thyroid, insulin, sulfonylureas, metformin.
In unadjusted and adjusted models, the effect of TZD use on fracture risk varied by HT use (Table). In non-HT users, TZD use was associated with a 42% increased risk of any fracture while in HT users, TZD use was associated with only a 22% increased risk. This apparent protective effect of HT use suggests that higher endogenous estrogen levels might also provide protection against the negative skeletal effects of TZD use, but this remains to be tested.
Our results indicate that, similar to findings in middle-aged women, older women are at increased risk of fracture with TZD use. HT may confer some protection from the negative skeletal effects of TZD use.
http://www.asbmr.org/image.axd?id=989342b1-bf1f-47b6-8bf5-46f9a5cd2c70&t=633747985769570000 - use this link to access accompanying table.