Association of Common Anti-Inflammatory Medications with Reduced Risk of Vision-Threatening Diabetic Retinopathy in Type 2 Diabetes.

OBJECTIVE: To determine whether adults with type 2 diabetes mellitus (T2DM) using cetirizine, ibuprofen, or prednisone have a reduced incidence of diabetic macular edema (DME) and proliferative diabetic retinopathy (PDR). DESIGN: Retrospective propensity score-matched cohort study. PARTICIPANTS: Adults with T2DM and documented eye-care in the TriNetX US Collaborative Network who had two documented uses of cetirizine, ibuprofen, prednisone, fenofibrate, or gabapentin between January 1, 2005, and March 1, 2025. METHODS: Each drug cohort was 1:1 propensity score-matched on 44 covariates against non-users; fenofibrate served as a positive control, and gabapentin served as both a negative control and an active comparator. Residual confounding was assessed using eight ophthalmic negative-control outcomes (NCOs) and E-values. Nine sensitivity analyses tested comparison to users of gabapentin, alternative outcome windows, a new-user design, glycemic strata, matching for drug indication, and exposure duration. MAIN OUTCOME MEASURES: Incident DME and PDR within 3 years of the index date. RESULTS: Matched-pair counts ranged from 18,762 to 86,846. Cetirizine was associated with reduced DME (hazard ratio [HR], 0.62; 95% confidence interval [CI], 0.51-0.74) and PDR (HR, 0.59; 95% CI, 0.46-0.77); ibuprofen with reduced DME (HR, 0.63; 95% CI, 0.56-0.69) and PDR (HR, 0.48; 95% CI, 0.41-0.56); and prednisone with reduced DME (HR, 0.46; 95% CI, 0.41-0.52) and PDR (HR, 0.36; 95% CI, 0.30-0.43). Fenofibrate showed associations of similar magnitude (DME HR, 0.57; 95% CI, 0.48-0.68; PDR HR, 0.59; 95% CI, 0.45-0.77); gabapentin showed no association (DME HR, 0.95; 95% CI, 0.88-1.02; PDR HR, 1.04; 95% CI, 0.93-1.15). Point E-values for the investigational drugs ranged from 2.57 to 4.97 (lower 95% CI bound, 1.93 to 4.06). The NCO panel was near null for cetirizine and ibuprofen and showed mild protective bias for prednisone and fenofibrate. Findings persisted across the nine sensitivity analyses. CONCLUSIONS: Cetirizine, ibuprofen, and prednisone were associated with reduced incidence of DME and PDR in adults with T2DM, with effect sizes comparable to those of the fenofibrate positive control. These findings support prospective evaluation of anti-inflammatory medications for the prevention of vision-threatening diabetic retinopathy.