Home / Specialties / Nephrology / SGLT-2 Inhibitors and the Risk for Urosepsis

SGLT-2 Inhibitors and the Risk for Urosepsis

Sep 26, 2020
Editor: David L. Joffe, BSPharm, CDE, FACA

Author: Jessica Rogers, PharmD Candidate, University of Colorado Skaggs School of Pharmacy and Pharmaceutical Sciences 

Are SGLT-2 inhibitors responsible for urosepsis or Fourniers gangrene? 

The FDA has issued two warnings regarding SGLT-2 inhibitors, the first in 2015 regarding the increased risk of severe UTI and one released in 2018 regarding a possible association with Fourniers gangrene. However, these warnings were issued due to FAERS reporting, and surrounding evidence remains conflicting. Additionally, there has yet to be specific evidence correlating this class of drugs and these conditions. Several meta-analyses of randomized control trials and observational studies have been completed, yielding conflicting results. More recent meta-analyses have not found an increased risk between SGLT-2 inhibitors and UTI, while older studies did find an increased risk. Several of the meta-analyses that have detected an increased risk have done so when comparing SGLT-2 inhibitor therapy to placebo, but not compared to active comparators. 


Additionally, several observational studies have been completed that have found similar or lower UTI rates compared to DPP-4 inhibitors. There is not as much data regarding urosepsis, with a recent meta-analysis finding no significant increase in risk for SGLT-2s than placebo, and a routine care study that found no increased risk than DPP-4 inhibitors and GLP-1 agonists. Fournier’s gangrene incidence is rare overall, making it difficult for a difference to be detected. A meta-analysis was completed from three RCTs that failed to detect an association, but this could have been due to the small number of events detected. Additionally, three observational studies were completed, which found a similar rate compared to DPP-4 inhibitors. As the use of SGLT-2 inhibitors increases, it is essential to assess both potential safety issues.  

This study aimed to determine whether SGLT-2 inhibitors are associated with an increased urosepsis rate compared with DPP-4 inhibitors among adults with type 2 diabetes, and to describe the risk of Fournier’s gangrene. This was a matched cohort study that used an overall new-user design with time-conditional propensity scoring. Patients were identified from seven Canadian provinces. The matched study cohort consisted of new SGLT-2 inhibitors (including empagliflozin, dapagliflozin, or canagliflozin) and users of DPP-4 inhibitors (alogliptin, linagliptin, saxagliptin, or vildagliptin), both either alone or in combination with other anti-diabetic medications. The SGLT-2 inhibitor group was further split into incident new users: those who did not receive a DPP-4 inhibitor within the past year; and prevalent users: those who did receive treatment with a DPP-4 inhibitor within the previous year. The primary outcome was urosepsis, defined as a hospitalization with a diagnosis code for either acute pyelonephritis, UTI, or acute cystitis, as well as a corresponding code for sepsis. The secondary outcome was Fournier’s gangrene, defined by diagnosis codes. Incidence rates and confidence intervals were estimated using the Poisson distribution. Cox proportional hazards models were used to estimate hazard ratios and confidence intervals for the primary outcome.  

Matched study cohorts consisted of 208,244 patients in each of the two groups. There were 189 urosepsis events among patients taking SGLT-2 inhibitors (incidence rate of 1.00 per 1000 person-years; 95% CI: 0.87-1.16). Among patients taking DPP-4 inhibitors, there were 368 cases of urosepsis (incidence rate of 2.03 per 1000 person-years; 95% CI 0.83-2.24). Looking at the secondary outcome of Fourniers gangrene, the incidence rates were similar between patient groups (number of events: 15 vs. 25; incidence rate of 0.08 vs. 0.14 per 1000 person-years; 95% CI 0.05-1.13 vs. 0.83-2.24). Additionally, there was no evidence of effect modification by age, sex, prior insulin use, or specific SGLT-2 inhibitor molecule.  

In conclusion, there was a lower urosepsis rate and a similar rate of Fourniers gangrene among patients taking SGLT-2 inhibitors than patients taking DPP-4 inhibitors. This study’s results can reassure the risk of both complications in patients taking this class of medication. Several limitations identified include the short duration of follow-up (0.9 years), making it impossible to detect long-term SGLT-2 inhibitor use events. Additionally, there may be confounding by indication if patients at high risk of UTI were preferentially started on a DPP-4 inhibitor. Future studies should include a more extended follow-up period to determine the risk of urosepsis and Fourniers gangrene in patients on long-term therapy.   

Practice Pearls: 

  • Rates of urosepsis were lower among patients with type 2 diabetes treated with SGLT-2 inhibitors than patients taking DPP-4 inhibitors. 
  • The incidence of Fourniers gangrene was similar between patients taking SGLT-2 inhibitors and those taking DPP-4 inhibitors.  
  • The short duration of follow-up limits the application of this study for patients on long-term SGLT-2 inhibitor therapy.  


Fisher, Anat et al. Sodium-glucose Cotransporter 2 Inhibitors and the Risk for Urosepsis – A Multi-site Prevalent New-user Cohort Study. Diabetes, obesity & metabolism 


Jessica Rogers, PharmD candidate, University of Colorado Skaggs School of Pharmacy and Pharmaceutical Sciences