# Intermediate Term Results of a Novel Minimally Invasive Keratoprosthesis

## TL;DR

A new minimally invasive corneal implant shows better long-term results than existing options in rabbits with severe eye injuries.

## Contribution

A novel minimally invasive keratoprosthesis with a nitinol backplate and DALK implantation technique is introduced.

## Key findings

- mi-KPro maintained clear optics and stable IOP without RPM or neovascularization for up to 12 months.
- PKP grafts failed within 2.2 months with corneal edema and IOP elevation.
- mi-KPro preserved retinal thickness and optic nerve axons compared to PKP and B-KPro.

## Abstract

Keratoprosthesis implantation remains a sight-restoring option in high-risk corneal transplantation where conventional keratoplasty is unlikely to succeed. Existing devices, such as the Boston keratoprosthesis (B-KPro), are very effective but long-term results are limited by glaucoma, infections, retroprosthetic membrane (RPM) formation, and retinal complications. We developed a minimally invasive keratoprosthesis (mi-KPro) that couples an ultrathin nitinol backplate with a poly(methyl methacrylate) optic and is implanted with a modified deep anterior lamellar keratoplasty (DALK) technique.

An experimental study in rabbits.

Thirty New Zealand white rabbits.

Rabbits received severe alkali or acid corneal burns and 1-month later treated with mi-KPro or penetrating keratoplasty (PKP); an additional cohort underwent B-KPro implantation in naïve eyes. Animals were followed for up to 12 months with serial photography, intraocular pressure (IOP) measurements by intracameral manometry and TonoPen, and anterior and posterior segment OCT. Explanted eyes were histologically analyzed.

Endpoints included device anatomic retention, RPM formation, infection, angle status, retinal thickness, and optic nerve axon density on histology.

The modified DALK approach allowed reproducible mi-KPro implantation with the nitinol backplate seated over Descemet membrane, and the optic extending into the anterior chamber through a 2.5-mm trephination. The mi-KPro eyes maintained clear optics without RPM, haze, or neovascularization and showed stable IOP (15.5 ± 2.2 at last follow-up vs. 15.6 ± 1.5 mmHg at baseline; P = 0.138). In contrast, all PKP grafts failed within 2.2 ± 0.7 months with marked corneal edema, neovascularization, and sustained IOP elevation despite timolol. The B-KPro eyes developed early, refractory IOP rise and staphyloma. Tonopen readings strongly correlated with intracameral IOP (r = 0.95, P = 0.001). OCT and histology demonstrated preservation of angle anatomy, retinal thickness, and optic nerve axons in mi-KPro eyes, whereas PKP eyes exhibited significant retinal thinning (P = 0.0014) and axon loss (P = 0.0001).

The mi-KPro combines a minimally invasive surgical approach with a flexible nitinol backplate to achieve long-term retention, stable IOP, and preservation of posterior segment structure in severe chemical injury. These data support mi-KPro as a promising next-generation keratoprosthesis for high-risk corneal blindness.

Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.