Microkeratomes yield smoother stromal bed than femtosecond laser in DSAEK prep

Three cutting systems offered variable stromal bed smoothness but preserved the corneal endothelium, according to a study.

Two mechanical microkeratomes created a smoother stromal bed than a femtosecond laser in the preparation of donor tissue for Descemet’s stripping automated endothelial keratoplasty, a study found.

Investigators compared stromal bed smoothness and endothelial cell density after preparation of donor tissue using the Moria 300-µm microkeratome, Horizon 300-µm disposable microkeratome (Refractive Technologies) and IntraLase femtosecond laser (Abbott Medical Optics).

“Both microkeratomes that were tested, the Moria as well as the Horizon, performed quite well in terms of creating a smooth lamellar bed,” V. Vinod Mootha, MD, the corresponding author, told Ocular Surgery News. “That was demonstrated by our study using subjective as well as objective criteria looking at the electron microscopic images of the surfaces that were created.”

However, both microkeratomes and the laser yielded similar central endothelial cell density and viability, the study authors said.

“All the systems showed that there was good preservation of the endothelium, which is also an important finding because we want to preserve the endothelium at each step of the procedure, starting from donor preparation,” Dr. Mootha said.

Ongoing research aims to ensure that current methods of donor tissue creation meet strict criteria for lamellar bed smoothness and endothelial viability, Dr. Mootha said.

“We’re trying everything possible to ensure that we have high-quality donor tissue to give our patients the best possible outcomes given our current understanding and technology,” he said.

Study results were published in Cornea.

Cutting, testing, scoring

The masked study included 15 donor corneas. One cornea from each of five donor pairs was mounted on a Moria ALTK artificial chamber and cut with the Moria microkeratome. The other cornea was mounted on a plastic artificial chamber and cut with the Horizon microkeratome.

The five remaining corneas were mounted on the Moria artificial chamber and cut with the IntraLase FS60 femtosecond laser.

The donor corneas were bisected, with half used for live-dead assay to assess central endothelial viability. The other half underwent scanning electron microscopy (SEM) of the stromal bed at magnifications of 14, 40 and 200.

Surface roughness of the stroma was graded based on SEM images. A qualitative roughness scale ranged from 1 (smoothest) to 5 (roughest). The qualitative scores were based on examinations of SEM images by two masked observers. The roughness evaluation module included in Scanning Probe Image Processor software (SPIP v.3.3.5, Image Metrology) was used to assess roughness of ×200 SEM images.

Specular microscopy was used to measure central endothelial cell density before and after tissue dissection.

Stromal and endothelial outcomes

Study results showed that both mechanical microkeratomes produced smooth stromal beds. Based on masked observer assessments, the Horizon microkeratome produced a smoother stromal bed than did the Moria microkeratome or IntraLase. Mean qualitative surface roughness scores were 1.8 for the Horizon, 3.7 for the Moria and 4.1 for the IntraLase. The difference was statistically significant (P < .001).

However, the Moria microkeratome scored best when ×200 SEM images of the lamellar beds were analyzed using imaging software designed to assess smoothness of nano-engineered surfaces. Mean quantitative surface roughness scores were 1,626 for the Moria, 2,183.2 for the Horizon and 2,882.2 for the IntraLase, with lower scores indicating smoother surfaces. The difference was statistically significant (P = .003).

Central endothelial cell density decreased 2.93% in the IntraLase group, 2.1% in the Moria group and 0.32% in the Horizon group; the difference was statistically insignificant.

All three systems appeared to yield satisfactory endothelial cell viability, the authors said.

Respective pros and cons

The Horizon microkeratome is disposable and easy to use but requires the use of a specially calibrated pachymeter. However, air is used to pressurize the donor cornea on the plastic artificial anterior chamber. Standard ultrasound pachymeters are calibrated to measure corneal thickness with fluid behind the endothelium.

“Since the interface is from the endothelium to air, there’s a much greater change in the index of refraction,” Dr. Mootha said. “You can’t really accurately measure the residual stroma that’s been left after the cut has been made.”

The femtosecond laser has shown efficacy in refractive surgery, but more development is needed to improve its utility in cutting donor tissue, Dr. Mootha said.

“The femtosecond, in my opinion, is the instrument of choice for creating a LASIK flap, but in terms of making deeper cuts in the cornea, I think more research needs to be done to consistently create a smooth deep lamellar surface,” he said. “Further research is required to refine its use, but it certainly shows great potential given its programmability and reproducibility.”

In addition, existing femtosecond lasers are cost-prohibitive for most eye banks and surgeons, Dr. Mootha said. – by Matt Hasson

Reference:

• Mootha VV, Heck E, Verity SM, et al. Comparative study of Descemet stripping automated endothelial keratoplasty donor preparation by Moria CBm microkeratome, Horizon microkeratome and IntraLase FS60. Cornea. 2011;30(3):320-324.

• V. Vinod Mootha, MD, can be reached at Department of Ophthalmology, 5323 Harry Hines Boulevard, University of Texas Southwestern Medical Center, Dallas, TX 75390-9057; email: vinod.mootha@utsouthwestern.edu.
• Disclosure: Dr. Mootha has no relevant financial disclosures.

This study shows that while two different microkeratomes readily achieved relatively smooth deep stromal dissections, further optimization is required to produce an equally smooth deep dissection with a femtosecond laser. The biggest problem with femtosecond laser deep stromal dissection is the occurrence of gross surface irregularities, which may not be readily apparent unless the entire dissection plane is examined at low magnification with side lighting. A contributing factor is that compression folds develop in the posterior stroma during applanation, and these seem to occur even with curved applanating devices. Also, the posterior stroma seems to separate less smoothly with a femtosecond laser than a blade. The gross abnormalities have led to poorer visual results with femtosecond laser DSAEK. Further work with femtosecond lasers is warranted because the laser can produce a more precise depth of cut than a microkeratome. On the other hand, the laser is a much more expensive approach, so better patient outcomes must be demonstrated to justify the extra cost.

– Marianne O. Price, PhD
Executive Director, Cornea Research Foundation of America, Indianapolis
Disclosure: Dr. Price receives travel grants from Moria and consulting fees from AMO.