Intersession repeatability of visual function and OCTA metrics in eyes with diabetic macular ischaemia.

Objective: To estimate inter-session coefficient of repeatability (CR) of visual function and OCTA metrics over 3 months in diabetic macular ischaemia (DMI) in stable laser-treated proliferative diabetic retinopathy (PDR) patients.

Methods: This prospective study recruited patients with stable PDR for at least 6 months following pan-retinal photocoagulation with visual acuity of at least 54 ETDRS letters. DMI was confirmed on OCTA as FAZ area of at least 0.5 mm or parafoveal capillary dropout in at least one quadrant if the FAZ area was less than 0.5 mm. Repeatability was assessed at baseline and 3 months by calculating the coefficients of repeatability (CR) and intraclass correlation coefficient (ICC). The Bland Altman (BA) plots including 95% CI bands for the bias (mean difference) and upper and lower limits of agreement (LOA) were used to visualise the agreement of the measurements.

Results: 88 eligible eyes of 63 participants were included in the analysis. The CR for best corrected visual acuity (BCVA) was 10.1 ETDRS letters (95% CI 8.5-11.7 ETDRS letters), and for low luminance visual acuity (LLVA) was 12.4 letters (95% CI 10.4-14.4). The ICC for BCVA and LLVA was 0.81 and 0.75 respectively. For Square Root of Foveal Avascular Zone (SQRT-FAZ) area, CR and ICC was 0.043 mm and 1.00 respectively. The Bland-Altman plots suggest that the level of agreement is wider for vascular density (VD) metrics compared to FAZ parameters.

Conclusion: We recommend using the FAZ metrics for quantitative measurements of DMI outcomes due to the excellent repeatability of manually corrected FAZ area.