Background/Purpose: Risk of osteoporotic fracture in type II Diabetes Mellitus is conflicting and possibly related to the anti-diabetic medication used. There is evidence to suggest protective effect of some medications (e.g. metformin and sulfonylureas) while others have been associated with increased risk for fractures (e.g. thiazolidinedione and DPP4-inhibitors).  Thus, we examined the association of oral anti-diabetic medications with hip fractures risk in community-dwelling older adults with type 2 Diabetes Mellitus.

Methods: We conducted a nested case-control study in men and women ≥ 40 years old with incident (new-onset) type II Diabetes Mellitus, from The Health Improvement Network (THIN), an electronic medical records database from the United Kingdom. For each incident (new-onset) hip fracture case, we selected up to 4 controls matched by age and sex. We categorized the anti-diabetic medications into the following categories: 1) Current use (last prescription within 1 year) of Metformin alone; 2) Current use of Sulfonylureas alone; 3) Current use of Thiazolidinediones or DPP4 inhibitors alone; 4) Current use of combinations or others; 5) Current use of any drug with recent switch in category; 6) Remote use of any medications (last prescription >1 year ago) as the reference category.  In a multivariable conditional logistic model, we adjusted for alcoholism/heavy drinking, BMI, fall/high risk of falls, smoking and medications (ace inhibitors, anti-osteoporosis medications, anti-seizure drugs, beta-blockers, diuretics, estrogen, glucocorticoids).

Results: Among 2673 subjects (mean age 72y; 70% women; 447 incident hip fracture cases and 2226 controls), the hip fractures cases had lower BMI and were more likely to have alcoholism and falls. As shown in Table, compared to remote use of any oral anti-diabetic medications, the odds for hip fracture was reduced by 36% with current metformin use (OR 0.64, 95% CI 0.49- 0.84), no different with current use of sulfonylureas (OR 1.02, 95%CI 0.76-1.38) and slightly increased, although not statistically significant, with current Thiazolidinedione or DPP-4 inhibitors (OR 1.29, 95%CI 0.69-2.43) use.

Conclusion:   We found protective effect of metformin use, no effect of sulfonylureas and possibly adverse effect of Thiazolidinedione and DPP-4 inhibitors on hip fracture risk in this large cohort of older diabetic patients. Clinicians caring for older diabetic patients need to consider fracture risk while selecting anti-diabetic medications.