# Impact of Microkeratome Dissection Parameters on Textural Interface Opacities in DSAEK Grafts

## TL;DR

This study shows that using a larger microkeratome head during DSAEK graft preparation can reduce the risk of textural interface opacities.

## Contribution

The study introduces the 'M-TIO' grading scale and identifies specific microkeratome parameters linked to TIOs in DSAEK grafts.

## Key findings

- A larger microkeratome head reduces the risk of TIOs compared to smaller heads.
- The difference between pre-cut and post-cut graft thickness is significantly associated with TIOs.
- Using a slow single-pass technique with a smaller head increases TIO risk.

## Abstract

Background: Textural interface opacities (TIOs) following Descemet’s stripping automated endothelial keratoplasty (DSAEK) have become a significant postoperative concern. Studies have explored possible links such as stromal irregularities and viscoelastic usage, but the exact cause of TIOs remains unclear. PURPOSE: To evaluate the relationship between microkeratome dissection parameters and the development of textural interface opacities in DSAEK grafts utilizing the “M-TIO” grading scale for standardized assessment. Methods: Optical coherence tomography (OCT) images of DSAEK-processed corneal grafts, prepared with the same microkeratome and technique for transplantation at Bascom Palmer Eye Institute, underwent blinded analysis using a newly developed grading scale termed “M-TIO”. This analysis aimed to evaluate and categorize the occurrence of TIO, explore its potential correlation with graft characteristics prior to DSAEK preparation, and assess specific stages of the DSAEK dissection process. Data collected included the size of the microkeratome head used, the difference between the head and the actual stromal cut, and the difference between the pre-cut graft thickness and post-cut DSAEK lenticule thickness. Results: The study retrospectively included 422 donor corneas transplanted from 2019 to 2023. Variables associated with TIO in the final multivariable ordinal logistic model included the difference between the pre-cut graft thickness and the post-cut DSAEK lenticule thickness (OR: 1.57 [99% CI: 1.22 to 2.06] per 50 µm) and microkeratome head (OR: 6.95 [99% CI: 1.04 to 36.60] 300 µm, OR: 4.39 [99% CI: 0.76 to 19.00] 350 µm, and OR: 18.86 [99% CI: 2.35 to 175.91] 400 µm vs 450 or 500 µm, respectively). Conclusions: This study identified a statistically significant association between TIOs and the microkeratome DSAEK preparation, proposing several factors that could help prevent its occurrence. Specifically, creating an ultra-thin DSAEK lenticule from an initially thick graft using a smaller microkeratome head with the slow single-pass technique may increase the risk of TIOs. In contrast, utilizing a larger microkeratome head can improve stromal thickness consistency, reduce technical challenges during graft preparation, and lower the risk of TIOs.