Abstract
Purpose :
Refractive errors produce image defocus and may alter quantitative metrics extracted from en face optical coherence tomography angiography (OCTA) images. Up to date it is possible to correct the sphere in commercialized devices using manual or auto-focus, but astigmatism remains uncorrected. This cross-sectional study assessed the qualitative effect of astigmatism on OCTA images and its quantitative analysis.
Methods :
Patients undergoing comprehensive ophthalmic examination, including OCTA, were included. We recruited subjects with a measured astigmatism of 0.5 diopters (D) or more. OCTA scans were acquired with and without astigmatism correction on a single visit using SPECTRALIS HRA-OCT2 (Heidelberg Engineering, Heidelberg, Germany) with follow-up mode activated. Astigmatism correction was applied to either the first or second OCTA image in random order using cylindrical lenses adapted to the OCTA device, with values from -1 to -5. The primary outcome measure was vessel density index (VD) in the nasal macular area of the superficial vascular plexus before and after astigmatism correction. A paired t-test was used for total analysis and a two-way ANOVA test for grouped analysis including astigmatism diopters (0.5-1.25, 1.5-2.25, 2.5-3.25 and >3.5) and axis intervals (with the rule, against the rule or oblique).
Results :
Ninety-one eyes of 91 subjects were included in our cohort, showing a mean astigmatism of 1.25D (95% CI 1.09-1.41) with a mean age of 63.6 years. Mean VD was significantly higher with astigmatic correction (mean 0.308 (95% CI 0.294-0.322) vs. 0.274 (95% CI 0.259-0.288) p<0.0001). Subgroup analysis suggested consistent effects of astigmatism correction depending on the amount of diopters (p=0.047) and regardless of the axis (p=0.119).
Conclusions :
Our data suggest that astigmatic correction increases VD values in OCTA images. Astigmatism correction should therefore be considered in studies that include analyzing quantitative parameters obtained by OCTA. However, further studies will be needed in order to assess the clinical significance of these changes.
This abstract was presented at the 2022 ARVO Annual Meeting, held in Denver, CO, May 1-4, 2022, and virtually.