Background: Diabetes is associated with an increased risk of nephrolithiasis. Mechanistic studies suggested that sodium–glucose cotransporter 2 inhibitors (SGLT2i) might be associated with a lower risk of nephrolithiasis, despite limited evidence from clinical studies.
Objectives: To further examine the risk of nephrolithiasis in SGLT2i users compared with dipeptidyl peptidase-4 inhibitors (DPP4i) users using a real-world population of type 2 diabetes (T2D) patients.
Methods: Using claims data from 15% random samples of national Medicare beneficiaries, we identified individuals with type 2 diabetes who initiated SGLT2i or DPP4i between 1/1/2017-12/31/2018. We defined the index date as the day of the first filled prescription of any SGLT2i or DPP4i, with no use of either in the prior year. We followed the cohort until the first medical encounter with diagnosis of nephrolithiasis, death, or 12/31/2018. Inverse probability treatment weighting (IPTW) was used to balance the baseline covariates, including sociodemographic, prior comorbidities and medications.
Results: Of the 116,506 included individuals (age 72±10 years, 53% women), 0.96% developed nephrolithiasis over a median follow-up of 360 days. After IPTW, the SGLT2i and DPP4i groups were well balanced in the baseline characteristics. The incidence rate of nephrolithiasis was 9.89 (95% confidence interval [CI] 8.49-11.52) and 11.02 (95% CI 10.34-11.73) event per 1000 person-years in the SGLT2i group and DPP4i group, respectively. SGLT2i use was associated with a significantly lower risk of nephrolithiasis compared with DPP4i use (hazard ratio [HR] 0.81; 95% CI 0.66-0.99). Subgroup analysis demonstrated that the use of SGLT2i was significantly associated with a lower risk of nephrolithiasis compared with the use of DPP4i in males (HR 0.75; 95% CI 0.58-0.98), but not in females (HR 0.75; 95% CI 0.65-1.26). SGLT2i, compared with DPP4i, was associated with a significantly lower risk of neprholithiasis among non-Hispanic Black (NHB) individuals (HR 0.22; 95% CI 0.07-0.64), but not in the other racial/ethnic groups (HR [95% CI] for non-Hispanic White: 0.91 [0.71-1.12]; HR [95% CI] for Hispanic: 1.09 [0.50-2.37]). This effect was also evident in patients without baseline nephrolithiasis (HR 0.68; 95% CI 0.53-0.88).
Conclusions: SGLT2i was associated with a lower risk of nephrolithiasis, and the effect was more pronounced in men, NHB, and those without a history of nephrolithiasis. Our findings reflect real-world evidence on the association between SGLT2i use and the risk of nephrolithiasis and help guide the selection of glucose-lowering drugs in patients with T2D.
