IMPORTANCE: Diabetic nephropathy and diabetic retinopathy share many similarities in pathophysiological processes. Preclinical studies have shown that sodium-glucose cotransporter 2 inhibitors (SGLT2is) have a protective role in the risk of diabetic retinopathy.
OBJECTIVE: To compare the risk of sight-threatening retinopathy associated with SGLT2is and other second-line glucose-lowering medications (including pioglitazone, sulfonylureas, and dipeptidyl peptidase-4 inhibitors [DPP-4is]) in patients with type 2 diabetes (T2D).
DESIGN, SETTING, AND PARTICIPANTS: This cohort study in Taiwan applied a new-user and active-comparator design. Patient demographic and clinical data were obtained from the National Health Insurance Research Database. Adult patients with newly diagnosed T2D from January 1, 2009, to December 31, 2019, were recruited and followed up until December 31, 2020. Propensity score matching was used to identify pairs of patients treated with SGLT2i vs DPP-4i, SGLT2i vs pioglitazone, and SGLT2i vs sulfonylurea from January 1, 2016, to December 31, 2019. Data were analyzed between August 18, 2022, and May 5, 2023.
EXPOSURES: Treatment with SGLT2i, DPP-4i, pioglitazone, and sulfonylureas starting on January 1, 2016.
MAIN OUTCOMES AND MEASURES: The main outcome was sight-threatening retinopathy in participants. Cox proportional hazards regression models were used to assess relative hazards of sight-threatening retinopathy between the matched case and control groups.
RESULTS: A total of 3 544 383 patients with newly diagnosed T2D were identified. After 1:1 propensity score matching, 65 930 pairs of patients treated with SGLT2i vs DPP-4i, 93 760 pairs treated with SGLT2i vs pioglitazone, and 42 121 pairs treated with SGLT2i vs sulfonylurea were identified. These matched patients included 236 574 males (58.6%), with a mean (SD) age of 56.9 (11.8) years. In the matched cohorts, SGLT2i had a significantly lower risk of sight-threatening retinopathy than DPP-4i (adjusted hazard ratio [AHR], 0.57; 95% CI, 0.51-0.63), pioglitazone (AHR, 0.75; 95% CI, 0.69-0.81), and sulfonylureas (AHR, 0.62; 95% CI, 0.53-0.71). The Kaplan-Meier curves showed that SGLT2i was associated with a significantly lower cumulative incidence of sight-threatening retinopathy than DPP-4i (3.52 vs 6.13; P < .001), pioglitazone (4.32 vs 5.76; P < .001), and sulfonylureas (2.94 vs 4.67; P < .001).
CONCLUSIONS AND RELEVANCE: This cohort study found that SGLT2i was associated with a lower risk of sight-threatening retinopathy compared with DPP-4i, pioglitazone, and sulfonylureas. This finding suggests that SGLT2i may play a role not only in reduced risk of diabetic nephropathy but also in the slow progression of diabetic retinopathy in patients with T2D.
