The use of sodium-glucose co-transporter-2 inhibitors or glucagon-like peptide-1 receptor agonists versus sulfonylureas and the risk of lower limb amputations: a nation-wide cohort study

Background: Numerous studies have investigated the potential association of sodium-glucose co-transporter-2 inhibitors (SGLT2-Is) with an increased risk of lower limb amputations (LLAs), but have produced conflicting results. Particularly studies comparing SGLT2-Is to glucagon-like peptide-1 receptor agonists (GLP1-RAs) seem to find a higher LLA risk with SGLT2-I use. This raises the question whether the results are driven by a protective GLP1-RA-effect rather than a harmful SGLT2-I-effect. GLP1-RAs could promote wound healing and therefore reduce the risk of LLAs, but the associations between both drug classes and LLA remain uncertain. Therefore, the aim of the current study was to investigate the risk of LLA and diabetic foot ulcer (DFU) with SGLT2-I use and GLP1-RA use versus sulfonylurea use. Methods: A retrospective population-based cohort study was conducted using data from the Danish National Health Service (2013–2018). The study population (N = 74,475) consisted of type 2 diabetes patients aged 18 + who received a first ever prescription of an SGLT2-I, GLP1-RA or sulfonylurea. The date of the first prescription defined the start of follow-up. Time-varying Cox proportional hazards models estimated the hazard ratios (HRs) of LLA and DFU with current SGLT2-I use and GLP1-RA use versus current SU use. The models were adjusted for age, sex, socio-economic variables, comorbidities and concomitant drug use. Results: Current SGLT2-I use was not associated with a higher risk of LLA versus sulfonylureas {adjusted HR 1.10 [95% confidence interval (CI) 0.71–1.70]}. Current GLP1-RA use, on the other hand, was associated with a lower risk of LLA [adjusted HR 0.57 (95%CI 0.39–0.84)] compared to sulfonylureas. The risk of DFU was similar to that with sulfonylureas with both exposures of interest. Conclusion: SGLT2-I use was not associated with a higher risk of LLA, but GLP1-RAs with a lower risk of LLA. Previous studies reporting a higher risk of LLA with SGLT2-I use compared to GLP1-RA use might have been looking at a protective GLP1-RA effect, rather than a harmful SGLT2-I effect.