Multimodal imaging of experimental choroidal neovascularization

Area of Science:

• Ophthalmology
• Medical Imaging
• Animal Models

Background:

• Choroidal neovascularization (CNV) is a key cause of vision loss.
• Accurate in vivo measurement of CNV is crucial for research and treatment evaluation.
• Current imaging modalities require validation against histological standards.

Purpose of the Study:

• To compare in vivo imaging techniques for measuring laser-induced CNV in mice.
• To validate fluorescein angiography (FA), multicolor (MC) imaging, and optical coherence tomography angiography (OCTA) against ex vivo histology.
• To assess interobserver reliability for each imaging modality.

Main Methods:

• A murine laser-induced CNV model was used.
• CNV lesions were imaged using FA, MC, and OCTA at 14 days post-induction.
• Ex vivo isolectin GS-IB4 stained choroidal-retinal pigment epithelium whole mounts served as the gold standard.
• Measurements were compared using Spearman's rank correlation and interobserver reliability was assessed using intraclass correlation coefficients (ICC).

Main Results:

• FA (early and late) and MC imaging showed good correlation with ex vivo histology (Spearman's r = 0.76 and 0.74, respectively).
• Interobserver reliability was high for FA (ICC = 0.97 and 0.96) and good for MC (ICC = 0.85).
• OCTA demonstrated poor correlation (Spearman's r = 0.06) and low interobserver reliability (ICC = 0.60) with histological measurements.

Conclusions:

• FA and MC imaging are reliable tools for in vivo CNV evaluation in rodent models.
• Caution is advised when using OCTA for CNV monitoring in small rodents due to measurement variability.
• Further comparative studies are recommended for OCTA in CNV assessment.