Purpose: To compare the choroidal thickness (CT) of diabetic eyes (different stages of disease) with controls, using swept-source optical coherence tomography. Methods: A multicenter, prospective, cross-sectional study of diabetic and nondiabetic subjects using swept-source optical coherence tomography imaging. Choroidal thickness maps, according to the nine Early Treatment Diabetic Retinopathy Study (ETDRS) subfields, were obtained using automated software. Mean CT was calculated as the mean value within the ETDRS grid, and central CT as the mean in the central 1 mm. Diabetic eyes were divided into four groups: no diabetic retinopathy (No DR), nonproliferative DR (NPDR), NPDR with diabetic macular edema (NPDR + DME), and proliferative DR (PDR). Multilevel mixed linear models were performed for analyses. Results: The authors included 50 control and 160 diabetic eyes (n = 27 No DR, n = 51 NPDR, n = 61 NPDR + DME, and n = 21 PDR). Mean CT (ß = −42.9, P = 0.022) and central CT (ß = −50.2, P = 0.013) were statistically significantly thinner in PDR eyes compared with controls, even after adjusting for confounding factors. Controlling for age, DR eyes presented a significantly decreased central CT than diabetic eyes without retinopathy (β = −36.2, P = 0.009). Conclusion: Swept-source optical coherence tomography demonstrates a significant reduction of CT in PDR compared with controls. In the foveal region, the choroid appears to be thinner in DR eyes than in diabetic eyes without retinopathy.

To analyze the effect of anti-vascular endothelial growth factor agents (anti-VEGF) in submacular choroidal thickness (CT) of diabetic retinopathy (DR) patients. Cross-sectional study, which included 25 DR patients (50 eyes) divided in 2 groups, according to DR stage and previous treatments: nonproliferative DR and diffuse diabetic macular edema in both eyes, submitted to macular laser in both eyes and anti-VEGF injection only in 1 eye (nonproliferative diabetic retinopathy + diabetic macular edema group, n = 11); and proliferative DR in both eyes, treated with panretinal photocoagulation in both eyes and anti-VEGF injection only in 1 eye (proliferative diabetic retinopathy group, n = 14). In the study visit, all patients underwent optical coherence tomography with enhanced depth imaging protocol. Choroidal segmentation was performed manually. The medium CT in central macular area (CCT) and the CT in centrofoveal B-scan were obtained automatically. The 25 eyes treated with anti-VEGF showed a reduction on CCT (P = 0.002) and subfoveal CT (P = 0.004), compared with the fellow eyes treated with laser only. Independent evaluation of PDR group revealed similar results (CCT, P = 0.02; subfoveal CT, P = 0.03). In nonproliferative diabetic retinopathy + diabetic macular edema group, CCT was also significantly thinner in eyes treated with anti-VEGF (P = 0.04). A correlation between the number of injections and a thinner CT was found in this group (P = 0.03) and in the evaluation of all eyes together (P = 0.03). Diabetic eyes treated with anti-VEGF agents have reduced CT.