Full-Field Optical Coherence Tomography of Human Donor and Pathological Corneas

Abstract

Purpose: To evaluate the performance of a full-field optical coherence tomography (FF-OCT) system in the study of human donor and pathological corneas and assess its suitability for use in eye banks.

Methods: Our study was carried out using an FF-OCT system developed for non-invasive imaging of tissue structures in depth with ultrahigh resolution (1 µm in all directions). Images were acquired from eight stored human donor corneas (either edematous or after deswelling) and five surgical specimens of corneas with various diseases (bullous keratopathy, lattice corneal dystrophy, stromal scar after keratitis, keratoconus and Fuchs dystrophy). They were compared with standard histology and pre-operative spectral domain OCT.

Results: The FF-OCT device enabled a precise visualization of the cells and the different structures (epithelium, basement membrane, Bowman’s layer, stroma, Descemet’s membrane and endothelium) in normal corneas. Specific lesions in various corneal diseases could also be easily identified, such as corneal edema, epithelium and Bowman’s layer irregularities, breaks, or scars (keratoconus), stromal opacities, deposits, fibrosis (stromal corneal scar, bullous keratopathy, lattice corneal dystrophy) and Descemet’s membrane thickening and guttae (Fuchs dystrophy). FF-OCT image features were comparable to the details provided by conventional histology. Higher resolution could be demonstrated with FF-OCT when compared with spectral domain OCT.

Conclusion: FF-OCT is a powerful non-invasive imaging tool that allows detailed study of corneal structures. Images correlate well with conventional histology. Further studies should evaluate the benefit of this technique as a complement to current assessment methods of human donor corneas.

• Corneal imaging
• corneal storage
• corneal transplantation
• optical coherence tomography

Declaration of interest

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this article.

Fabrice Harms is affiliated with LLTech SAS. Use of the LightCT Scanner was provided free of charge for the purpose of this study.