Abstract
Purpose :
To quantify and examine the distribution of inflammatory dendritic cells (DCs) in large-area mosaic images of the cornea obtained by In Vivo Confocal Microscopy (IVCM).
Methods :
82 subjects were included in the study. Using laser-scanning in vivo confocal microscopy (Hidelberg HRT 3 with Rostck corneal module), the central cornea in both eyes of study subjects (163 eyes) was imaged. Subsequently images from the sub-basal layer were used to assemble the best possible mosaics. Mosaics were analyzed manually by an experienced examiner using image J. Sub-basal inflammatory cells were distinguished, demarcated and grouped into three distinct categories: Type 1) Mature dendritic cells (mDCs); Type 2) Immature dendritic cells (imDCs); Type 3) Globular cells (GCs); reflective, circular-shaped, with no visible dendritic processes and variable size.
Results :
Average size of mosaic images was 6.0 mm2, comprising an average of 37 single confocal fields of view. The process of designation, categorization and counting of inflammatory cells for large-area mosaics took an average time of 19.38 minutes per mosaic image. The DC counting time varied between values as high as 147 minutes for a complex mosaic where corneal sub-basal layer was extensively dominated with variety of inflammatory cells and values as low as 1 minute for simple mosaic images with very few cells in the sub-basal layer. The most prevalent inflammatory cells in this study were type 2 immature dendritic cells (imDCs) and least prevalent cells were type 1 mature dendritic cells (mDCs).
Complete quantitative analysis and geographic distribution patterns are presented.
Conclusions :
Inflammatory cells in the corneal sub-basal layer can be analyzed in vivo in a large field of view of the central cornea. The results provide information regarding the inflammatory status of the cornea in a more complete manner than would be possible from single image analysis, thereby possibly facilitating better diagnosis and management of inflammatory or other pathologic conditions.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.