Purchase this article with an account.
Rida Shahzad, Hema Radhakrishnan, Sue Shawcross, Waheedh Alshemmri, Victoria Rimmer, Chantal Hillarby; The Use of Corneal Cross-Linking for the Treatment of Bacterial Keratitis. Development of a Human Ex-Vivo Model. Invest. Ophthalmol. Vis. Sci. 2019;60(9):2534. doi: https://doi.org/.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Bacterial keratitis is one of the leading vision-threatening pathologies, being a major cause of corneal ulceration and opacification, and needs vigilant diagnosis and aggressive treatment. This is a laboratory-based study aimed to establish an experimental in vitro model of bacterial keratitis with Staphylococcus epidermidis on donor human corneas, and to study the efficacy of corneal collagen cross-linking (CXL) in eradicating the infection using histological staining techniques.
Two ex vivo corneas were dissected into four quarters each, and de-epithelized. The four quarters were classified into the following study groups: (1) Control (not inoculated, not treated with CXL) (2) CXL (not inoculated, cross-linked only) (3) Infected (inoculated, not treated with CXL) (4) Infected & CXL (inoculated and treated). Staphylococcus epidermidis was used to inoculate corneas in the infected groups. CXL was performed in group 2 without prior inoculation, and in group 4 twenty-four hours after inoculation. The procedure followed the Dresden protocol: corneas were administered with 1% riboflavin for 30 minutes at 5-minute intervals, followed by illumination with ultraviolet-A (UVA) light for 30 minutes with simultaneous instillation of riboflavin every 5 minutes. All corneal tissues were subsequently fixed, frozen, sectioned, and stained with H&E and PAS staining techniques. Images of each group were obtained after examination under a light microscope, and the bacterial cell count was estimated under 20x magnification. Data was analyzed using Two-tailed t-test.
There was a significant reduction in bacterial population in the Infected & CXL group as compared to the Infected and untreated group, observed with both H&E (p=0.005) and PAS (p=0.03) staining. The procedure also caused apoptosis of keratocytes in the anterior section of the stroma, and more extensive keratocyte death was observed in the Infected & CXL group as compared to the CXL-only group.
A model of human bacterial keratitis was successfully developed. The CXL procedure appears to be an effective treatment for bacterial keratitis in terms of eradication of S. epidermidis and may reduce intensity of infection. The use of this model will allow further investigation/development of CXL treatment of microbial keratitis in a controlled environment.
This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.
Bacterial cell count of Infected and Infected+CXL groups
This PDF is available to Subscribers Only