Race/ethnicity and socioeconomic-based analysis of vascular alterations on OCTA in patients with diabetes

Abstract

Purpose : Racial and ethnic minorities have a higher prevalence of diabetic retinopathy (DR) and present at
advanced stages.^1–3 Two small studies have measured DR in racial/ethnic groups using optical coherence tomography angiography (OCTA). However, neither included adjustments for socioeconomic (SES) factors.^4,5 SES factors are important as they are not equally distributed between racial groups. We performed a retrospective case-control study using OCTA to investigate microvascular differences between racial/ethnic groups. Area deprivation index (ADI) was used to approximate SES.

Methods : We obtained 3x3mm macular OCTA scans from diabetic patients without other ocular comorbidities or prior DR treatments. One eye per patient was included in the analysis. Parameters analyzed included foveal avascular zone (FAZ), superficial capillary plexus (SCP) and deep capillary plexus (DCP) for parafoveal vessel density (VD), vessel length density (VLD), and flow index (FI). SES was measured using patients’ addresses and ADI. Multivariable linear regression was used to adjust estimates for potential confounders of age, SS, hypertension, gender, duration of disease, and HbgA1C. P value of <0.001 was considered significant.

Results : 227 controls and 1993 diabetics were included in the study. Racial breakdown was 43% Hispanic, 24.4% NH Asian, 6.8% NH Black, 9.6% NH Caucasians, and 16.3% Other. All comparisons used Hispanic group the referent. Among controls, Caucasian patients had smaller FAZ area (ß = -0.089, P < 0.001) and FAZ perimeter (ß = -0.325, P < 0.001). Among all diabetics, Caucasians had smaller FAZ area (ß = -0.100, P < 0.001) and FAZ perimeter (ß = -0.406, P < 0.001), but higher deep VD (ß = 1.227, P < 0.001) and deep VLD (ß = 0.362, P < 0.001). Asian participants also had higher deep VD (ß = 0.566, P < 0.001). Lastly, the addition of SES scores, both at a national and state level, did not modify any of the racial regression models significantly.

Conclusions : Our findings from a large county population suggest that at baseline, there are differences in the microvascular alterations between racial/ethnic groups that cannot be explained by their SES. Albeit statistically significant, further investigation is needed to determine whether these differences are meaningful enough to change clinical DR practices.

This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.

 

Baseline characteristics