Abstract
To evaluate the impact of lipid-lowering medications on diabetic retinopathy and diabetic complications requiring intervention in the U.S.Retrospective cohort analysis.Administrative insurance claims drawn from the Truven MarketScan Commercial Claims and Encounters database.Beneficiaries with Type 2 diabetes mellitus (T2DM).Any signs of diabetic retinopathy, as measured by diagnosis codes for non-proliferative diabetic retinopathy (NPDR), proliferative diabetic retinopathy (PDR), diabetic macular edema (DME), and procedure codes for retinopathy treatments (anti-VEGF injections, laser therapy, and vitrectomy).We analyzed a population of 269,782 patients diagnosed with T2DM between 2008 and 2015. 99,233 (37%) of patients were undergoing treatment with lipid-lowering medications. Approximately 6% of patients on lipid-lowering medications had a diagnosis code for NPDR, PDR, or DME, or a procedural code for intravitreal injections, PPV, or laser in their record following diagnosis with diabetes, compared to 6.5% of patients who did not take lipid-lowering medications (p<0.01). In adjusted time-to-event analyses, patients who took lipid-lowering medications prior to diagnosis with T2DM were less likely to progress to any retinopathy diagnosis (HR 0.60, 95% CI 0.55-0.65) and less likely to receive any treatment for retinopathy (HR 0.81, 95% CI 0.78-0.84). These findings were significant at the aggregate level, as well as at the level of individual diagnosis (NPDR HR 0.63, 95% CI 0.57-0.69; PDR HR 0.45, 95% CI 0.37-0.54; DME HR 0.39, 95% CI 0.33-0.45), and at the level of each treatment category (anti-VEGF injection HR 0.81, 95% CI 0.78-0.84; laser HR 0.62, 95% CI 0.47-0.81; vitrectomy HR 0.71, 95% CI 0.59-0.85).We find consistent evidence that patients on lipid-lowering medications are less likely to develop NPDR, PDR, or DME, and modest evidence that these patients are less likely to receive intravitreal injections of anti-VEGF medication, laser treatments, or vitrectomy. Our study validates the findings of studies that have used claims databases in East Asia in relatively homogeneous populations to estimate an association between statin use and retinopathy, replicating them in a U.S. context in a large commercial claims database.
View details for DOI 10.1016/j.ajo.2019.05.029
View details for PubMedID 31194953