Comparison of the findings of endothelial specular microscopy before and after corneal cross-linking

Structural Changes in Thin Keratoconic Corneas Following Crosslinking with Hypotonic Riboflavin: Findings on In Vivo Confocal Microscopy

Purpose

To report structural changes observable in in vivo confocal microscopy (IVCM) in keratoconic corneas < 400 μm treated with hypotonic riboflavin and collagen crosslinking (CXL).

Methods

Ten eyes of ten patients with progressive keratoconus and corneal thickness between 350 and 399 μm underwent CXL with hypotonic riboflavin. IVCM was performed preoperatively and at one month, three months, and six months after the procedure.

Results

IVCM analysis one month postoperatively showed complete absence of the subepithelial nerve plexus with gradual regeneration over six months in 8 of the 10 eyes, and poor regeneration in the remaining 2 eyes. The anterior stroma showed extracellular lacunae and hyper-reflective cytoplasm in a honeycomb appearance signifying edema at one month which gradually decreased over six months post CXL. Stromal keratocyte apoptosis was evident in the anterior stroma in all cases and extended to the posterior stroma in four eyes with gradual regeneration evident at three and six months. The specular endothelial count decreased by 8% (P = 0.005) post-CXL, but no corneas developed clinical signs of endothelial trauma.

Conclusion

IVCM analysis of thin corneas after hypotonic CXL showed posterior corneal structural changes. Posterior stromal changes were accompanied by a decrease in the endothelial cell count. This case series was a preliminary feasibility study that might necessitate conducting a well-designed controlled study.

A Review of Corneal Endotheliitis and Endotheliopathy: Differential Diagnosis, Evaluation, and Treatment

The corneal endothelium plays an integral role in regulating corneal hydration and clarity. Endotheliitis, defined as inflammation of the corneal endothelium, may disrupt endothelial function and cause subsequent visual changes. Corneal endotheliitis is characterized by corneal edema, the presence of keratic precipitates, anterior chamber inflammation, and occasionally limbal injection, neovascularization, and co-existing or superimposed uveitis. The disorder is classified into four subgroups: linear, sectoral, disciform, and diffuse. Its etiology is extensive and, although commonly viral, may be medication-related, procedural, fungal, zoological, environmental, or systemic. Not all cases of endothelial dysfunction leading to corneal edema are inflammatory in nature. Therefore, it is imperative that practitioners consider a broad differential for patients presenting with possible endotheliitis, as well as familiarize themselves with appropriate diagnostic and therapeutic modalities.

Quantitative assessment of corneal vibrations during intraocular pressure measurement with the air-puff method in patients with keratoconus

BACKGROUND

One of the current methods for measuring intraocular pressure is the air-puff method. A tonometer which uses this method is the Corvis device. With the ultra-high-speed (UHS) Scheimpflug camera, it is also possible to observe corneal deformation during measurement. The use of modern image analysis and processing methods allows for analysis of higher harmonics of corneal deflection above 100 Hz.

METHOD

493 eyes of healthy subjects and 279 eyes of patients with keratoconus were used in the measurements. For each eye, 140 corneal deformation images were recorded during intraocular pressure measurement. Each image was recorded every 230 µs and had a resolution of 200 × 576 pixels. A new, original algorithm for image analysis and processing has been proposed. It enables to separate the eyeball reaction as well as low-frequency and high-frequency corneal deformations from the eye response to an air puff. Furthermore, a method for classification of healthy subjects and patients with keratoconus based on decision trees has been proposed.

RESULTS

The obtained results confirm the possibility to distinguish between patients with keratoconus and healthy subjects. The features used in this classification are directly related to corneal vibrations. They are only available in the proposed software and provide specificity of 98%, sensitivity-85%, and accuracy-92%. This confirms the usefulness of the proposed method in this type of classification that uses corneal vibrations during intraocular pressure measurement with the Corvis tonometer.

DISCUSSION

With the new proposed algorithm for image analysis and processing allowing for the separation of individual features from a corneal deformation image, it is possible to: automatically measure corneal vibrations in a few characteristic points of the cornea, obtain fully repeatable measurement of vibrations for the same registered sequence of images and measure vibration parameters for large inter-individual variability in patients.