Longitudinal Detection of Radiation-Induced Peripapillary and Macular Retinal Capillary Ischemia Using OCT Angiography

Abstract

Purpose: To study longitudinal changes in retinal capillary circulation in eyes treated with iodine 125 (I¹²⁵) plaque brachytherapy for uveal melanoma using OCT angiography (OCTA).

Design: Longitudinal prospective study of 21 patients undergoing treatment for uveal melanoma with I¹²⁵ plaque brachytherapy. Eyes with melanoma were imaged with OCTA before treatment and at 12-month intervals until 2 years after brachytherapy.

Participants: After institutional review board approval, participants were enrolled prospectively from an academic ocular oncology clinic.

Methods: Peripapillary (4.5 × 4.5-mm) and macular (3 × 3-mm) OCTA scans were acquired with AngioVue (Optovue, Inc, Fremont, CA).

Main outcome measures: The peripapillary nerve fiber layer plexus capillary density (NFLP_CD), macular superficial vascular complex vessel density (mSVC_VD), and foveal avascular zone (FAZ) area were calculated.

Results: Before treatment, no significant difference was found in the NFLP_CD, mSVC_VD, or FAZ area between eyes with melanoma and normal fellow eyes. By 24 months, 11 eyes had developed clinical signs of radiation retinopathy, radiation optic neuropathy, or both. In treated eyes, the NFLP_CD (48.4±4.1%) was reduced at 12 months (46.7±5.0%; P = 0.04, Wilcoxon signed-rank test) and 24 months (44.5±6.1%; P < 0.001). Similarly, the mSVC_VD (48.4 2±3.6%) was reduced in treated eyes at 12 months (43.5±5.9%; P = 0.01) and 24 months (37.4±9.1%; P < 0.001). The FAZ area (0.26±0.11 mm²) increased in treated eyes at 12 months (0.35±0.22 mm²; P = 0.009) and 24 months (0.81±1.03 mm²; P = 0.001). When only eyes with clinically evident radiation changes were evaluated, the changes in NFLP_CD, mSVC_VD, and FAZ area were more pronounced. OCT angiography measurements correlated with both radiation dose and visual acuity. The mSVC_VD measured at 12 months was found to predict the development of clinically apparent radiation retinopathy within 1 year.

Conclusions: OCT angiography demonstrated early emergence of peripapillary and macular capillary vasculature changes after I¹²⁵ plaque brachytherapy. OCT angiography provided a quantitative measurement of retinal capillary changes associated with ischemia that correlated with visual acuity and radiation dose and may predict future development of radiation-induced retinal toxicity.

Figure 1.

Graphs showing the (A) macular superficial vascular complex vessel density (mSVC_VD), (B) foveal avascular zone (FAZ) area, and (C) peripapillary nerve fiber layer capillary density (NFLP_CD) for all patients at baseline, 12 months, and 24 months. Each line represents a single eye treated with plaque brachytherapy. Eyes with radiation retinopathy (RR) or papillopathy at 24 months are shown as solid red lines. Eyes without radiation retinopathy or papillopathy at 24 months are shown as dashed blue lines.

Figure 2.

Representative example of macular superficial vascular complex and foveal avascular zone (represented by yellow line) at baseline, 12 months, and 24 months for a patient with no radiation retinopathy at 24 months (top row) and a patient who demonstrated radiation retinopathy by 24 months (bottom row).

Figure 3.

Representative example of peripapillary nerve fiber layer capillary density at baseline, 12 months, and 24 months for a patient with no radiation retinopathy at 24 months (top row) and a patient who demonstrated radiation retinopathy by 24 months (bottom row).

Figure 4.

A, Scatterplot showing that peripapillary nerve fiber layer capillary density (NFLP_CD) at 24 months correlates with the radiation dose (Gy) to the optic nerve as measured by radiation dose to 50% of the optic disc (D50 disc). The relationship between NFLP_CD and D50 disc was examined using Pearson’s correlation coefficient. Each eye is indicated with a point on the plot. A best-fit line is shown. B, Scatterplot showing a similar trend between the radiation dose to 50% of the fovea and the macular superficial vascular complex vessel density (mSVC_VD), although this was not statistically significant. P< 0.05 was used to define significance.

Figure 5.

Correlation between the logarithm of the minimum angle of resolution (logMAR) visual acuity (VA) at 24 months and OCT angiography measurements. Each eye is indicated with a point on the plot. A, The relationship between macular superficial vascular complex vessel density (mSVC_VD) and logMAR VA was examined at 24 months using Pearson’s correlation coefficient. B, The relationship between foveal avascular zone (FAZ) area and logMAR VA was examined at 24 months using Pearson’s correlation coefficient. C, The relationship between peripapillary nerve fiber layer capillary density (NFLP_CD) and logMAR VA was examined at 24 months using Pearson’s correlation coefficient. The asterisks indicate an outlier with severe macular ischemia. This eye was not included in the Pearson’s correlation coefficient presented. P< 0.05 was used to define significance.