Abstract
Purpose::
To compare OCT/SLO cross-sectional tomograms of human retina in vivo with histology to determine the relationship between OCT/SLO images and retinal substructure in normal macular tissue and in tissue overlying choroidal melanoma.
Methods::
OCT/SLO tomograms of approximately 8 micron axial resolution were obtained in vivo from eight patients who subsequently underwent enucleation due to progressive choroidal melanoma. Scans were obtained of both normal macular tissue and of abnormal tissue overlying the tumor. 5 micron resolution histologic sections were obtained in alignment with the positioning of the OCT/SLO scans and the images compared.
Results::
Good correlation was observed between OCT/SLO scans and histology both in normal macular tissue and in tissue overlying choroidal melanoma. In normal macular tissue the main retinal layers could be resolved well with OCT/SLO with the cell body layers (inner and outer nuclear layers, ganglion cell layer) appearing less reflective than the axodendritic layers (inner and outer plexiform layers, nerve fibre layer). The RPE, Bruch’s membrane and inner choroid appeared as a single highly reflective band in the distal retina with the junction between the photoreceptor inner and outer segments appearing as a distinct thinner highly reflective band proximal to this. OCT/SLO tomograms of retinal tissue overlying choroidal melanoma also showed good correlation with histology with many features being accurately visualized including sub-retinal fluid, exudation, intra-retinal oedema, haemorrhage and the disorganization of retinal banding structure.
Conclusions::
OCT/SLO tomograms of the human retina provide valuable information on retinal morphology in vivo. In both normal macular tissue and in abnormal tissue overlying choroidal melanoma, good correlation was observed between OCT/SLO scans and corresponding histologic sections. OCT/SLO is a valuable modality with high clinical relevance and it could have the potential to act as a valid surrogate for retinal histology in vivo.
Keywords: imaging/image analysis: clinical • oncology • pathology: human