Special Section Series Sponsored By Moria

Ultra-thin Grafts

The next advance in corneal lamellar surgery

Francis W. Price, JR., MD and Massimo Busin, MD

The Rationale for Ultra-Thin Endothelial Grafts

With the ongoing evolution in endothelial keratoplasty techniques, thinner grafts make sense

Francis W. Price, Jr., MD

Endothelial keratoplasty (EK) has rapidly become the standard of care for Fuch's dystrophy and other forms of endothelial dysfunction. Since the procedure was first introduced just 7 years ago, adoption rates have grown steadily, with EK now accounting for 45% of all U.S. corneal transplant cases (Fig. 1).

As surgeons have moved away from full-thickness penetrating keratoplasty (PK), EK techniques have become increasingly precise. We have seen a transition from deep lamellar endothelial keratoplasty (DLEK) to Descemet-stripping endothelial keratoplasty (DSEK or DSAEK), and now to exploration of Descemet membrane endothelial keratoplasty (DMEK or DMAEK) — all in the attempt to target a replacement of the diseased tissue, leaving the healthy portions of the cornea intact.

Over time, our goals have changed, as well. In the beginning, my primary motivation for seeking out alternatives to PK was safety. The large, open-sky PK incisions left the cornea weakened and the patient vulnerable to losing an otherwise successful graft due to a fall or inadvertent blow to the eye.

Endothelial keratoplasty has eliminated those structural deficiencies in the cornea, making surgery much safer. In addition, the transition to EK has been nothing short of life-changing for our patients. Visual recovery occurs within a few weeks, rather than a year or more, and visual outcomes are much better and more predictable. With smaller, more stable incisions, we don't induce severe irregular astigmatism and there is no need to restrict patient activities after the initial 2-month postoperative period.

Fig. 1: Rate of PK vs. EK in the U.S.

DSAEK and beyond

The most common EK procedure today is DSAEK. Although we don't have enough experience yet to say with authority how its long-term (10+ years) safety compares to that of PK, we do know that DSAEK is essentially a refractive-neutral procedure, with all the advantages of visual recovery and safety described above, and that it can safely be performed under topical anesthesia.¹ So why the continued proliferation of keratoplasty acronyms and techniques?

The answer is that the desire for better visual results—once just a bonus of safer surgery—is now driving the evolution of EK.

While post-DSAEK vision is much better than anything we have seen in the past, most patients still don't achieve 20/20 vision. There are likely a number of factors contributing to the reduction in best-corrected visual acuity (BCVA). For example, a 5.0-mm DSAEK incision, while much smaller than a PK incision, is still large enough to induce astigmatism or other aberrations. Irregularities in the donor may increase the vision, or the hyperopic shift caused by transplanted stroma may play a role.

I believe that a mismatch in the corneal curvature between the donor and host tissue is likely the most significant factor limiting the visual outcome. The variance in normal Ks, which can range from the mid 30s to the low 50s, is certainly a confounding variable in our analysis of postoperative results. When a very steep donor cornea is paired with a very flat recipient, for example, we are likely to get folds or wrinkles as the donor tissue adheres to the mismatched host cornea, degrading the optical image. Thicker grafts magnify any differences in curvature.

Descemet's membrane endothelial keratoplasty (DMEK) is one possible solution. Because one is transplanting the bare minimum layer of tissue, the graft is extremely thin. And it seems to be producing far more results in the 20/15 to 20/25 range than DSAEK.² In our center, we are achieving about a 2-line improvement in visual acuity with DMEK compared to DSAEK. However, the very thin tissue is difficult to manipulate in some eyes and challenging to orient and unfold correctly.

DMAEK is a hybrid of DSAEK and DMEK that involves a central DMEK with a thicker outer ring. The visual results are similar to DMEK (73% are 20/25 or better at 3 months³) with better ease of handling. Donor preparation is still quite challenging, although some eye banks are beginning to offer pre-cut tissue for DMAEK.

But the biggest problem with both DMEK and DMAEK is the postoperative management. Our rate of re-injecting air is dramatically higher for DMEK and DMAEK, with 40% and 20%, respectively, requiring a re-bubbling procedure, compared to less than 4% of DSAEK cases. We've also seen higher rates of tissue loss with the newer procedures, at least during the learning curve. With DMEK, I initially reported a tissue loss rate of 7%, although that is now down to 0% with more experience. And with DMAEK, my tissue loss rate was 13%. I am now using DMAEK grafts prepared by my local eye bank, but I know the eye bank may still lose more tissue preparing these grafts than they do in preparing DSAEK grafts.

Is thinner the answer?

Given the challenges of DMEK/DMAEK, some have wondered whether simply cutting thinner DSAEK grafts might be the answer.

In a recently published paper, Neff, Holland and colleagues reported that grafts ≤131 μm, as measured by postoperative AS-OCT, provided a statistically significant improvement in BSCVA compared to thicker grafts.⁴

If thinner DSAEK grafts could be cut intentionally, there is good reason to believe they might provide the same visual benefits of DMEK/DMAEK but with greater ease of preparation and handling and with a lower rate of dislocation.

Femtosecond lasers have been proposed for this use. However, current femtosecond laser technology cannot provide consistently high quality, thin graft tissue. The cuts are not as smooth as with a microkeratome, perhaps due to some compression or irregularity of posterior stroma.

A double-pass ultra-thin DSAEK technique that can be performed with Moria's new microkeratome system has been introduced by Prof. Busin of Italy and is demonstrating good early results.⁵

As endothelial keratoplasty continues to evolve, I expect we will see an ongoing emphasis on techniques that improve surgical efficiency and provide better visual outcomes for our patients with endothelial dysfunction.

Moria's New System for Creating Ultra-Thin Corneal Lamellar Grafts

Worldwide, most endothelial keratoplasty (EK) grafts are created with a Moria microkeratome and artificial chamber. The new Moria system for creating ultra-thin grafts consists of seven single-use microkeratome heads ranging from 50 to 300 microns, a single-use artificial chamber, special inflation tubing and syringe sets, and the turbine-driven Moria CB Microkeratome with control unit. The heads can be used in various combinations in a two-pass technique to create ultra-thin grafts or individually to create conventional grafts. The artificial chamber with rotating guide ring allows initiation of cuts in opposite quadrants of the cornea, to produce large diameter grafts that are much more planar than conventional grafts. The inflation tubing and infusion syringe sets help to consistently maintain pressure in the system. The new Mini Busin Spatula and Forceps facilitate handling and insertion of the ultra-thin grafts.

References

1. Price MO, Price FW Jr. Endothelial keratoplasty a review. Clin Experiment Ophthalmol 2010;38(2):128-40.
2. Price MO, Giebel AW, Fairchild KM, Price FW Jr. Descemet's membrane endothelial keratoplasty: prospective multicenter study of visual and refractive outcomes and endothelial survival. Ophthalmology 2009;116(12):2361-8.
3. McCauley MB, Price MO, Fairchild KM, et al. Prospective study of visual outcomes and endothelial survival with Descemet membrane automated endothelial keratoplasty. Cornea 2011;30(3):315-9.
4. Neff KD, Biber JM, Holland EJ. Comparison of central corneal graft thickness to visual acuity outcomes in endothelial keratoplasty. Cornea 2011;30(4):388-91.
5. Busin M, Beltz J, Patel A, Scorcia V. Ultrathin DSAEK: Future of endothelial keratoplasty? American Society of Cataract and Refractive Surgery 2011 Annual Symposium, Course 28-306, March 28, 2011, San Diego, CA.

Dr. Price is in private practice at Price Vision Group in Indianapolis, IN, and is the founder and board president of the Corneal Research Foundation of America. He performs more than 400 corneal transplantation procedures annually, is a well-known instructor of EK techniques and has authored 37 peer-reviewed papers on EK. Contact him at (317) 814-2823 or fprice@pricevisiongroup.net.

Ultra-Thin DSAEK: Personal Experience

A new technique produces excellent visual results with the ease of handling and preparation of conventional DSAEK

Massimo Busin, MD

The ideal keratoplasty procedure is one that takes place in a closed system and offers rapid visual rehabilitation and improved visual outcomes with a low risk of complications. Descemet-stripping automated endothelial keratoplasty (DSAEK) fulfills these requirements and has therefore become the gold standard for surgical treatment of endothelial decompensation.

DSAEK works well regardless of phakic status, IOP, or factors that pose limitations for other techniques. In my hands, 73% of DSAEK patients without co-morbidities can see 20/40 or better at one month after surgery and 81% at three months. Four-fifths of patients achieve 20/40 BSCVA – and they get to that point in a matter of weeks – so we have certainly come a long way from our post-PK results.^1,2

However, only one-third of my DSAEK patients achieve 20/20 or better acuity, and others have reported even lower rates. It is very frustrating that sometimes patients with clear grafts and the potential for 20/20 vision cannot achieve that potential.

I began performing Descemet membrane endothelial keratoplasty (DMEK) in 2006 and was initially enthusiastic about its potential to improve visual acuity. There were reports that 20% to 45% of eyes achieved BSCVA of 20/20 or better with DMEK. However, there are a number of disadvantages to DMEK: It requires considerable surgical skill and prolonged surgical time. Unlike DSAEK, it is not suitable for all eyes. For example, DMEK is not a good choice in cases with concomitant ocular pathology or posterior stromal opacification; a shallow or poorly visualized anterior chamber; or when there is communication between the anterior chamber and vitreous cavity, through which the donor endothelium may easily dislocate posteriorly.

Finally, we have seen a high rate of tissue loss (up to 16% in some reports) with DMEK, an unacceptably high detachment rate of up to 63%, and a failure rate of up to 8%, which is higher than that for other procedures. I think all these factors will limit the widespread use of DMEK in clinical practice.

We know that some eyes can achieve 20/20 vision with DSAEK, particularly when the graft is thinner than average.³ I wanted to find a way to make those results more reproducible in an easy-to-perform and widely applicable procedure, so I devised a new approach, “ultra-thin” (UT) DSAEK. In this procedure, the surgical technique is similar to that of DSAEK, except that the graft is created with two microkeratome passes. Together, the two passes serve to debulk and refine the donor tissue. The microkeratome is essential. At the moment, there is no doubt that the quality of microkeratome EK grafts is far superior to that of femtosecond laser EK grafts.

UT-DSAEK graft preparation and transplant

I utilize the Moria artificial chamber. Although my initial experience was with a metal, reusable chamber, Moria now has a disposable version that offers important advantages for the double-pass ultra-thin technique (Fig. 1).

Fig. 1: Moria Single-Use Artificial Chamber

Pachymetry should be performed before and after the first de-bulking step to determine the actual tissue thickness and choose the appropriate microkeratome head(s). The first cut is performed with a 300-μm head, just as one would normally create a conventional DSAEK graft. Of all the factors that can affect the graft depth, the most important is pressure in the system. I recommend keeping the pressure in the artificial chamber constant at 80-90 mm Hg. To maintain this ideal level, the height of the infusion bottle is kept at 120 cm and the tubing is clamped at 50 cm from the entrance into the artificial anterior chamber.

The second or refinement cut is made with a 50- to 200-μm microkeratome head (Fig. 2), depending on the tissue thickness, with the goal of ultimately creating a graft that is approximately 100 μm or less. Each microkeratome pass is performed very slowly and steadily.

Fig. 2: Moria Single-Use CB Heads

For the second cut, the dovetail is moved 180° away from the initial position in order to perform it from a direction opposite to the one used for the de-bulking cut. This is done because the initial part of the cut is usually deeper than the final part. By starting the two cuts from opposite directions, the second cuts deeper exactly where the first cut was shallower. As a result, the risk of perforation is minimized and the final shape of the graft is planar.

In my initial cases, I had to dismantle the lid from the artificial anterior chamber, possibly causing excessive graft manipulation. However, the new Moria disposable chamber has a rotating guide ring that makes it easier to move the location for initiating the second pass. In addition, the new chamber creates a larger-diameter graft. Typically, the diameter of the graft is slightly reduced with each successive pass, but the disposable chamber enables the surgeon performing the 2-pass technique to create a graft that is 1-2 mm larger than one made with the reusable chamber.

The resulting tissue is very thin but stays flat, so it behaves like a DSAEK graft and is much easier to manipulate than a DMEK graft.

For delivery of the graft through a 3.0-mm clear corneal incision, I use a mini-glide (the Busin Mini Spatula) and 23-gauge forceps, both available from Moria. Although others have proposed more complex inserters, I find that especially thin tissue tends to curl and will not lie flat with inserters, so I prefer to use a simple mini-glide to pull the tissue through the incision.

Results

The average thickness of grafts prepared at the Venice Eye Bank for a comparative experimental trial was 73 ± 14 μm in the UT group and 175 ± 38 μm in the DSAEK group. There was no significant difference in endothelial cell loss between corneas undergoing one pass (conventional grafts) or two passes (UT grafts). Ease of handling, although not quite as easy as DSAEK, was far superior to DMEK.

More recently, at the last ESCRS meeting in Paris I showed that in 24 of 26 ultra-thin DSAEK cases, I achieved central graft thicknesses of 100 μm or less (Fig. 3). My personal results with UT-DSAEK are included in Table 1. Fifty percent of the UT cases achieved 20/20 or better BSCVA, which is remarkable.

Fig. 3: At 3 months, 24 of 26 eyes had central graft thickness of <100 μm as measured by OCT.

At one year, the endothelial cell (EC) loss was about the same as we have seen with DSAEK, despite the fact that I transplanted the UT grafts through a much smaller (3.0-mm) incision. In the past, we have seen a higher rate of EC loss with smaller incisions because there is typically more extensive manipulation of the tissue.

When I compare UT-DSAEK to DSAEK and DMEK, my conclusion is that all three offer an acceptable rate of cell loss. With the ultra-thin procedure, the speed of visual recovery is faster than conventional DSAEK and equivalent to DMEK — and the proportiion that achieve final acuity of 20/20 is higher than conventional DSAEK and perhaps also DMEK. In short, this procedure offers us the potential to achieve the visual results of DMMEK with the ease of handling and tissue preparation of DSAEK.

Fig. 4: Ultra-thin DSAEK grafts compared to DMEK grafts

References

1. Price FW Jr, Whitson WE, Marks RG. Progression of visual acuity after penetrating keratoplasty. Ophthalmol 1991;98(8):1177-85.
2. Zaidman GW, Goldman S. A prospective study on the implantation of anterior chamber intraocular lenses during keratoplasty for pseudophakic and aphakic bullous keratopathy. Ophthalmol 1990;97(6):757-62.
3. Neff KD, Biber JM, Holland EJ. Comparison of central corneal graft thickness to visual acuity outcomes in endothelial keratoplasty. Cornea 2011;30(4):388-91.

Prof. Busin is head of the department of ophthalmology at Villa Igea Hospital in Forli, Italy. He is a renowned lecturer and instructor in endothelial keratoplasty techniques. Contact him at +039-0543-419503 or mbusin@yahoo.com.