OBJECTIVE: Lower extremity complications significantly impact morbidity and healthcare costs among people with type 2 diabetes (T2D). Evidence regarding the impact of different glucose-lowering medications on these outcomes remains inconclusive.
METHODS: We emulated a target trial using two linked national claims databases (OptumLabs Data Warehouse, Medicare fee-for-service). We included adults with T2D at moderate cardiovascular risk who initiated GLP-1RA, SGLT2i, DPP-4i, or sulfonylurea between 2014-2021, and used propensity score inverse probability of treatment weighted Cox proportional hazards models to compare the incidence rates of the primary composite outcome of incident foot ulcer/abscess, osteomyelitis, Charcot arthropathy, or amputation across the four medication classes under the intention-to-treat framework.
RESULTS: The weighted study cohort included 81,998 DPP4i-initiators, 43,734 GLP-1RA-initiators, 57,399 SGLT2i-initiators, and 206,374 sulfonylurea-initiators; they were well balanced on all examined baseline characteristics. Sulfonylurea use was associated with a higher risk of the composite lower extremity complications outcome compared to DPP-4i (HR 1.15; 95%CI 1.11-1.19), GLP-1RA (HR 1.20; 95%CI 1.13-1.28), and SGLT2i (HR 1.08; 95%CI 1.02-1.14). SGLT2i use was also associated with a higher risk compared to GLP-1RA (HR 1.11; 95%CI 1.03-1.21). Amputation events were rare in all treatment groups.
CONCLUSION: We observed greater relative risk of lower extremity complications with sulfonylurea use compared to DPP4i, GLP-1RA, and SGLT2i use, and with SGLT2i use compared to GLP-1RA use. Reassuringly, the absolute differences between the medication classes were <1%. Diabetes management teams may consider these medication-associated risks when selecting glucose-lowering therapies for individuals without high cardiovascular risk, especially those predisposed to lower extremity morbidity.
