[Optical coherence tomography--high-resolution tissue imaging, but not histology!]

Abstract

Optical coherence tomography (OCT) is a non-invasive and fast method for retinal in vivo imaging. It was introduced in 1991 and became commercially available in 1996. With higher resolution spectral domain OCT (SD-OCT), all histological retinal layers can be identified with an axial resolution of 1-5 µm today. Numerous papers have correlated OCT and histological images of physiological retina from different species. Reliable attribution of the reflective bands in OCT images to histological layers has been facilitated even for those layers that are displayed differently in OCT and histology, particularly in the outer retina. In contrast, the correlation of OCT and histology of pathologically altered human retina has only rarely been described due to very limited histological availability. The studies that presented such a correlation on animal material were able to identify some artifacts of histological processing. They also showed that OCT reflectivity changes do not necessarily correlate with cell density in the retina, such that a retinal scar, which is rich in fibres, may appear isoreflective to surrounding healthy retina. This issue, however, has mostly been neglected. Morphometric analyses revealed significantly larger diameters in OCT images compared to histology of photocoagulation lesions. Obviously, OCT reflectivity changes indicate a different tissue condition than altered HE-staining properties in histology. While coarse retinal structure alterations may be located and followed very well with OCT, it is almost impossible to derive from an OCT image the histological change that causes the alteration on a cellular level. This leads to frequent misinterpretations of OCT images. This article aims to pinpoint the pitfalls of OCT interpretation with examples from the current literature, and it encourages the reader to be attentive and critical when interpreting OCT images.