AIMS

To investigate the risk of diabetic macular oedema (DMO) associated with the use of sodium-glucose cotransporter-2 (SGLT2) inhibitors in patients with type 2 diabetes mellitus (T2DM).

MATERIALS AND METHODS

We conducted a retrospective cohort study by analysing a large multi-institutional electronic medical records database in Taiwan. We included adult patients with T2DM without DMO newly receiving either SGLT2 inhibitors or glucagon-like peptide-1 receptor agonists (GLP-1RAs) during the period 2016 to 2018. We used propensity scores with inverse probability of treatment weighting to generate comparable groups. The study outcome was incident DMO, determined by clinical diagnosis during outpatient visits or admissions. We followed patients from the index date to either DMO occurrence, last clinical visit, patient death, or December 31, 2020. We performed Cox proportional hazards regression models to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for the risk of DMO.

RESULTS

We included 9986 new users of SGLT2 inhibitors (mean [SD] age 59.6 (12.1) years, median [interquartile range {IQR}] glycated haemoglobin [HbA1c] 70 (61-81)mmol/mol, estimated glomerular filtration rate [eGFR] 89.1 [71.4-108.7] mL/min/1.73 m and urine albumin-creatinine ratio [UACR] 26.1 [9.7-117.6] mg/g) and 1067 new users of GLP-1RAs (mean [SD] age 58.4 (41.5) years, median [IQR] HbA1c 73 [64-84] mmol/mol, eGFR 91.6 [68.6-114.0] mL/min/1.73 m and UACR 37.6 [11.1-153.2] mg/g) with similar baseline characteristics. Lower DMO risks were observed among patients newly receiving SGLT2 inhibitors (7.9/1000 person-years), compared to those receiving GLP-1RAs (10.7/1000 person-years) with an HR of 0.75 (95% CI 0.64-0.88).

CONCLUSIONS

Our findings suggest use of SGLT2 inhibitors was associated with lower risk of DMO in T2DM patients in clinical practice, compared to use of GLP-1RAs. Future studies are necessary to confirm this observation.