Abstract
To investigate how sodium-glucose co-transporter 2 inhibitors (SGLT2is) add-on therapy for metformin affects diabetic retinopathy (DR) progression in patients with type 2 diabetes mellitus (T2DM). This nationwide population-based study conducted from January 1, 2016, to December 31, 2018 involved 3,432,911 adults with T2DM in Taiwan. To adjust for potential confounders, data on sex, age, income, comorbidities, diabetes complication severity index score, staging of kidney disease, anti-diabetic medications, and index year were included. The outcome was DR progression, determined by procedure codes or the addition of ICD-9-CM or ICD-10-CM codes to the medical records of the patients during the study. Sensitivity analyses were performed to validate the findings. The adjusted hazard ratio (aHR) of DR progression was 0.89 for the SGLT2is add-on group, relative to the control group [95% confidence interval (CI) 0.81-0.99, P = 0.026]. The Kaplan-Meier curve of the cumulative incidence rate showed that the cumulative incidence of DR progression was considerably decreased in the SGLT2is cohort (log-rank P = 0.0261). The use of SGLT2is for less than 1 year and 1-2 years were associated with a significant increase in the risk of DR progression (aHR 1.56 and 1.88, respectively); however, the risk markedly reduced if the SGLT2is regimen was used for more than 2 years (aHR 0.41, 95% Cl 0.35-0.48; P < 0.001). The serial sensitivity analysis showed consistent findings. The aHR of DR progression was 0.82 for the SGLT2is cohort relative to the non-SGLT2is cohort based on the fundoscopy or indirect ophthalmoscopy findings within 1 year before the outcome date (95% Cl 0.71-0.95; P = 0.009). Co-administration of metformin and SGLT2is may reduce the risk of DR progression. Short-term use of SGLT2is may markedly increase the risk of DR, whereas prolonged use SGLT2is may significantly decrease it.