The effect of astigmatism on vessel density measurements derived from optical coherence tomography Angiography (OCTA)

Abstract

Purpose : OCTA density metrics may be useful to monitor diseases that affect the microvasculature of the eye. To better understand how factors unrelated to disease may affect the measurements, we did a study to analyze the effects of astigmatism on OCTA quantitative metrics.

Methods : Normal eyes without astigmatism were imaged without contact lenses, and with contact lenses to induce +0.75 diopters cylinder (DC), +1.75DC, and +2.75DC astigmatism at axis 180 utilizing a 3x3 mm OCTA scan pattern on CIRRUS™ HD-OCT 5000 with AngioPlex^® OCT Angiography (ZEISS, Dublin, CA) images. Vessel density (VD) and perfusion density (PD) were analyzed in four ETDRS (Early Treatment Diabetic Retinopathy Study) quadrants. Univariate linear regressions were performed to correlate quantitative metrics to the level of induced astigmatic cylinder.

Results : 15 eyes (15 patients) were imaged. Linear regression analysis showed that every +1 D increase in induced astigmatic cylinder was associated with a 0.58 mm/mm² decrease in ETDRS inner mean VD (p < 0.001); 0.57 mm/mm² decrease in full mean VD (p < 0.001); 0.60 decrease inner superior VD (p < 0.001); 0.56 mm/mm² decrease in ETDRS inner inferior VD (p = 0.002); 0.66 mm/mm² decrease in ETDRS inner nasal VD (p < 0.001); and a 0.49 mm/mm² decrease in ETDRS inner temporal VD (p = 0.005), a 0.008 decrease in inner mean PD (p = 0.001); 0.008 decrease in full mean PD (p < 0.001); 0.008 decrease in ETDRS inner superior PD (p = 0.001); 0.007 decrease in ETDRS inner inferior PD (p = 0.013); 0.009 decrease in ETDRS inner nasal PD (p = 0.001); and a 0.007 decrease in ETDRS inner temporal PD (p = 0.043). Figure 1 illustrates the density measurements in a single eye, imaged with no correction (upper left) and induced astigmatism (upper right: +0.75D, lower left: +1.75D, lower right: +2.75D).

Conclusions : Increased levels of induced astigmatism correlate with reduced VD and PD with each degree of +1D resulting in approximately a 3% reduction in the measured density. This may be due to the magnification effect of the astigmatism. This should be considered when monitoring OCTA parameters over time in patients whose astigmatic errors may change due to conditions or procedures including keratoconus and other corneal degenerations or refractive or cataract surgeries.

This is a 2020 ARVO Annual Meeting abstract.