Retinal vascular permeability as a preclinical biomarker for diabetic retinopathy

Abstract

Purpose : Early diabetic retinopathy (DR) lacks sensitive and reliable biomarkers that can detect subtle vascular changes before clinical signs manifest. The purpose of this study was to investigate if retinal vascular permeability can be determined as a preclinical biomarker for diabetic patients with no diabetic retinopathy (DMnoDR) via use of the dynamic tracer kinetic model (DTKM) derived extraction fraction based on conventional fluorescein video-angiogram (FVA). Extraction fraction (i.e., the fraction of imaging agent that leaks in a single pass of blood through the tissue) has a direct relation to retinal vascular permeability.

Methods : Six healthy control subjects and six DMnoDR subjects underwent a routine FVA using a Heidelberg SPECTRALIS® (55°, two-60 sec movie). Optimized dose of sodium fluorescein dye (0.5 ml of 10%) was used to obtain the FVAs. Each movie frame underwent preprocessing of removing motion and blinking artifacts and oversaturated pixels. The arterial input concentration and tissue concentration were selected by arterial and tissue signals averaged over manually selected regions of interest. DTKM analysis were performed to obtain pixel-by-pixel mapping of extraction fraction.

Results : Control subjects had no ocular history. The average age and body mass index (BMI) were 28 ± 7 y and 25 ± 5 kg/m², respectively. The average age and BMI for DMnoDR patients were 50 ± 10 y and 35 ± 3 kg/m², respectively. The average duration of diabetes was 10 ± 9 years. The average hemoglobin A1c (HbA1c) was 6.7 ± 0.9 %. The extraction fraction of control subjects was 0.05 ± 0.03 and for DMnoDR patients was 0.13 ± 0.09. Extraction fraction was found to be significantly greater in DMnoDR subjects than in healthy control subjects (p<0.05) .

Conclusions : The current results showed an increase in the DTKM derived extraction fraction in diabetic patients with no retinopathy. The retinal vascular permeability may be a good preclinical biomarker for early stages of diabetic retinopathy.

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