Patients with diabetes mellitus (DM) have increased morbidity and mortality rates, according to the study authors, and some research has suggested they have a greater likelihood of fracture. “While the biological mechanisms underlying the positive association between DM and the risk of fracture remain largely obscure, a higher propensity of falling, impairment of bone quality, and direct or indirect treatment effects of glucose‐lowering medications on bone have been suggested to explain the elevated risk of fracture among DM patients,” they reported in their meta-analysis, published in Obesity Reviews. And while past studies have found a link between thiazolidinedione and fracture, data are limited on the association between thiazolidinedione as well as insulin, sulphonylureas, and metformin and fracture.
The researchers queried PubMed and Web of Science databases for relevant observational studies from inception through February using a combination of these search terms: (diabetes OR insulin OR glargine OR lispro OR biguanidine OR metformin OR thiazolidinediones OR pioglitazone OR rosiglitazone OR sulfonylureas OR sulphonylureas OR acetohexamide OR carbutamide OR chlorpropamide OR glycyclamide OR metahexamide OR tolazamide OR tolbutamide OR glibenclamide OR glibornuride OR gliclazide OR glipizide OR gliquidone OR glisoxepide OR glyclopyramide OR glimepiride) and fracture.
Metformin May Be Safest Bet for Low Fracture Risk
The final analysis included 33 studies, of which 26 were cohort and seven were case-control studies. An estimated 345,133 fractures were reported in about 6,847,158 patients. Studies reported on the use of metformin (n = 12), insulin (n = 23), sulphonylureas (n = 10), and thiazolidinediones (n = 14). Fracture risk was increased among patients using insulin (relative risk [RR] 1.49, 95% confidence interval [CI] 1.29-1.73), sulphonylureas (RR 1.30, 95% CI 1.18-1.43), and thiazolidinediones (RR 1.24, 95% CI 1.13-1.35), while fracture risk was decreased among patients using metformin (RR 0.86, 95% CI 0.75-0.99).
When assessing fracture risk associated with different types of thiazolidinediones, RRs were similar between pioglitazone (1.38, 95% CI 1.23-1.54) and rosiglitazone (1.34, 95% CI 1.14-1.58) use.
“The positive association between insulin or thiazolidinedione use and the risk of fracture was attenuated to non‐significant in the studies that adjusted for certain variables (falls or previous fracture for insulin and osteoporosis/osteoporosis medications, steroid medications, or classic glucose‐lowering medications for thiazolidinediones); this result was possibly driven by the contradictory studies reporting a significant reduction in risk with the use of insulin and thiazolidinediones, as the associations became significant after exclusion of these studies (data not shown),” the researchers further observed. “Furthermore, the association between thiazolidinedione use and the risk of hip fracture became significant (RR 1.21, 95% CI 1.03, 1.43) after exclusion of a contradictory study.”
The study authors concluded that thiazolidinedione use should be discouraged among patients at a high risk of fracture. Insulin and sulphonylurea use may increase fracture risk as well, “which is most likely attributed to an increased risk of hypoglycaemia‐induced falls,” they said. Therefore, when prescribing insulin and sulphonylureas, a patient’s risk of hypoglycemia should be considered. Metformin appeared to have a good safety profile in terms of fracture risk. Still, the researchers called for randomized controlled trials to determine whether certain medications do increase fracture risk or if other confounding factors are at play.