Morphological Differentiation of Corneal Inflammatory Cells

Area of Science:

• Ophthalmology
• Cell Biology
• Medical Imaging

Background:

• Corneal confocal microscopy (CCM) analyzes corneal nerve fibers and corneal inflammatory cells (CICs).
• CICs are potential biomarkers for disease activity in chronic autoinflammatory diseases.
• Current lack of standardized criteria hinders reliable CIC morphological characterization.

Purpose of the Study:

• To establish a standardized protocol for morphological classification of CICs.
• To assess the protocol's applicability and interrater reliability.

Main Methods:

• Systematic literature review to identify CIC definitions.
• Development and iterative refinement of a structured classification algorithm.
• Application of the optimized protocol by multiple raters and calculation of interrater reliability using Fleiss kappa.

Main Results:

• Literature search revealed no uniform criteria for differentiating CIC types.
• Initial protocol showed low interrater agreement (K = 0.09).
• Revised protocol achieved significantly higher interrater reliability (K = 0.64 with 3 raters, K = 0.61 with 4 raters).

Conclusions:

• Existing literature-based criteria for CICs lead to high error rates.
• A standardized protocol with clearly defined cell types yields reproducible results.
• The developed protocol is suitable for future evaluation of CICs in corneal confocal microscopy.