A, The corneoscleral rim was mounted endothelial side up on a custom-made holder, and the DM was then loosened with a hockey stick knife from the scleral spur in a central direction. B, The corneoscleral rim was divided into 2 equal halves. C, The DM was then completely stripped from the posterior stroma to obtain a semicircular sheet of DM. D, The DM rolls with the endothelium on the outside formed spontaneously after immersion in saline. In this case, the 2 Descemet rolls formed with the axis of the roll perpendicular to the straight edge of the semicircular graft.
Corneal thickness (expressed in micrometers according to Pentacam software; Oculus) is indicated according to the color-coded information. The full yellow lines outline the approximate position of the hemi-DMEK graft, and the location of the peripheral graft detachment in case 1 is highlighted by a dotted yellow line. Note that corneal deturgescence starts in the graft’s center and progresses outward from 1 month to 6 months. N indicates nasal; OD, right eye; and T, temporal.
Lam FC, Baydoun L, Dirisamer M, Lie J, Dapena I, Melles GRJ. Hemi–Descemet Membrane Endothelial Keratoplasty TransplantationA Potential Method for Increasing the Pool of Endothelial Graft Tissue. JAMA Ophthalmol. 2014;132(12):1469-1473. doi:10.1001/jamaophthalmol.2014.3328
This study evaluates the technical feasibility and clinical outcomes of a Descemet membrane endothelial keratoplasty (DMEK) technique that could increase the availability of donor tissue for DMEK.
To evaluate the clinical outcome of using a semicircular, large-diameter Descemet membrane graft in DMEK (hemi-DMEK), potentially allowing the use of a single donor cornea for 2 DMEK procedures.
Design, Setting, and Participants
A prospective, interventional case series was conducted at a tertiary referral center. Three eyes of 3 patients with Fuchs endothelial dystrophy received a hemi-DMEK.
Transplantation of a semicircular, large-diameter hemi-DMEK graft.
Main Outcomes and Measures
Best-corrected visual acuity, endothelial cell density, pachymetry, and intraoperative and postoperative complications.
The patients’ best-corrected visual acuity at 6 months was 0.70 (Snellen equivalent, 20/29), 0.50 (20/40 [amblyopic eye]), and 1.20 (20/17). At 1 month, endothelial cell density decreased by 49%, 31%, and 39%, respectively, and endothelial cell migration appeared to continue for up to 6 months. Central corneal thicknesses decreased from 682, 707, and 681 μm before surgery to 523, 534, and 489 μm, respectively, at 6 months. No intraoperative or postoperative complications were seen.
Conclusions and Relevance
Hemi-DMEK (using half-moon–shaped grafts) is technically feasible and may provide visual outcomes similar to those obtained with routine DMEK (full-moon–shaped graft). If so, this technique may have the potential to double the availability of donor endothelial tissue for DMEK surgery.
In the past decade, endothelial keratoplasty (EK) may have become the preferred treatment in corneal endothelial dysfunction, and the latest refinement of the procedure, Descemet membrane EK (DMEK), may provide the best and fastest visual recovery.1- 3 There is, however, a significant shortage of donor corneal tissue in many parts of the world.4,5 This shortage may be lessened by the use of split donor tissue for DMEK and deep anterior lamellar keratoplasty.6- 8 However, the use of split donor tissue does not solve the increasing demand for endothelial grafts.
Currently, only the central portion of the Descemet-endothelial complex (8.5- to 9.5-mm diameter) is harvested; the peripheral rim (ie, approximately half of the graft surface area) is discarded. In contrast to penetrating keratoplasty and Descemet stripping (automated) EK, there is no optical or technical reason to use only the central portion of the donor tissue in DMEK, since a DMEK graft is very thin and uniform in thickness. The aim of the present study was to explore the feasibility and describe the clinical outcomes of harvesting and transplanting a half-moon (semicircular) hemi-DMEK graft.
Dutch Medisch Ethische Toetsingscommissie provided institutional review board approval, and all patients provided written informed consent. The study was conducted according to the Declaration of Helsinki. The participants did not receive financial compensation.
Three pseudophakic eyes of 3 women (aged 66, 72, and 65 years) underwent hemi-DMEK surgery for decompensated Fuchs endothelial dystrophy. One patient was amblyopic in the operated eye.
Corneoscleral buttons were excised from donor globes (donor ages 49, 70, and 67 years) obtained less than 36 hours post mortem. After 1 week in organ culture (CorneaMax, Eurobio) and endothelial cell evaluation, the corneoscleral buttons were mounted endothelial side up on a custom-made holder with a suction cup. Uveal remnants were removed, and the DM was loosened with a hockey stick knife (DORC International) from the scleral spur in a central direction so that the peripheral DM including the trabecular meshwork was detached (Figure 1A).
The corneoscleral buttons were then divided into 2 equal halves with a surgical knife (No. 10 knife; Swann-Morton) (Figure 1B). The DM was then completely stripped from the posterior stroma so that 2 semicircular sheets of DM (2 half-moon–shaped grafts) with an endothelial monolayer were obtained (Figure 1C). In all cases, a DM roll with the endothelium on the outside formed spontaneously after immersion of the DM in saline, with the axis of the roll at any angle to the straight edge of the semicircular graft (Figure 1D).
Immediately after preparation, endothelial cell appearance, density, and viability were evaluated. The hemi-DMEK rolls were then stored in organ culture medium until the time of transplantation.
Surgical procedures were performed using previously described techniques with a few adjustments.2,9 After insertion into the anterior chamber, the hemi-DMEK graft was oriented endothelial side down by careful, indirect manipulation with air and fluid. While the anterior chamber was maintained with fluid and air, the graft was gently spread over the iris and oriented with the widest graft diameter across the longest horizontal meridian. An air bubble was then injected underneath the donor DM to position the tissue onto the recipient’s posterior stroma.2 The anterior chamber was completely filled with air for 60 to 90 minutes, followed by partial air-fluid exchange to leave the eye pressurized with a minimum of 50% air fill. Postoperative management was identical to that used after standard DMEK surgery.1,2
The surgery was uneventful in all cases. The hemi-DMEK graft formed a single roll either with the long axis in line with the straight edge of the semicircular graft or with the long axis at an oblique angle to the straight edge, so that an asymmetric double roll was obtained.2 However, standard graft unfolding techniques10 proved to be effective in unfolding the hemi-DMEK roll. Centering the hemi-DMEK graft required some additional maneuvers and adjustments. The graft was positioned with its longer dimension running along the longer horizontal meridian of the eye. In addition, the hemi-DMEK roll was positioned eccentrically before unfolding to ensure that the semicircular hemi-DMEK graft was well centered after unfolding.
After surgery, the hemi-DMEK grafts were fully attached in all cases except for a small, persistent peripheral detachment in case 1 that developed by week 1 (Table). No rebubbling procedures were required, and the postoperative course for all 3 cases was uneventful.
All eyes demonstrated improved corneal clarity and an improvement in best-corrected visual acuity at 6 months (Table), and no visual symptoms or disturbances were reported. Endothelial cell density (ECD) decrease at 6 months was 57%, 38%, and 63% compared with the preoperative eye-bank values (Table).
Owing to the difference in shape between the semicircular DMEK graft and the circular descemetorhexis, areas of bare corneal stroma were present in all cases. These initially edematous denuded areas showed deturgescence starting from the area adjacent to the graft (Figure 2) to clear by 6 months. In case 3, the previously bare cornea stromal area had a mean (SD) ECD of 770 (10) cells/mm2 and the area of attached hemi-DMEK had a mean ECD of 1073 (10) cells/mm2 at 6 months.
From a technical point of view, hemi-DMEK grafts could be produced using only a slight modification of our standardized harvesting technique.11,12 Tissue stripping was no more complex, and no tissue loss was encountered due to preparation failure. During surgery, all hemi-DMEK grafts could be unrolled and positioned using our standardized unfolding techniques.10 The only modification required was to position the hemi-DMEK roll so that the long, straight edge of the unfolded graft would run along the longest horizontal meridian of the eye and the hemi-DMEK roll was slightly eccentric before unfolding. This was done to ensure that the unfolded hemi-DMEK graft would be centered.
Our series may have been too small to evaluate the potential effects of the hemi-DMEK graft on visual outcome, since one of the 3 eyes was amblyopic. However, another eye, which had normal visual potential, reached a best-corrected visual acuity of 1.20 (Snellen equivalent, 20/17) at 6 months, which may indicate that visual recovery with hemi-DMEK (half-moon–shaped graft) may be similar to that with standard DMEK (full-moon–shaped graft). A return to normal pachymetry values across the cornea in all cases may be indicative of the eye being able to reach its full visual potential after hemi-DMEK.
Endothelial cell density decreased in the first month after hemi-DMEK by 31% to 49%, followed by a decrease between the second and sixth months of 7% to 24%. The mean ECD decrease following standard DMEK is approximately 30% to 35% at 6 months postoperatively.1,3 Most of this decrease has been shown to have occurred by 1 month; further decreases were relatively small, at approximately 7% per year.13,14 We postulate that these differences may be explained by variations in cell migration between hemi-DMEK and standard DMEK. A different migratory pattern may be expected in hemi-DMEK because a semicircular transplant was positioned onto a circular denuded stroma bed (after a circular descemetorhexis), leaving a larger gap of bare corneal stroma that has to be repopulated with donor or host endothelial cells.
We have demonstrated that hemi-DMEK is technically feasible and can be performed with possible good clinical outcomes. However, the limited number of cases in the present report does not permit one to determine with confidence the precise frequency of good outcomes.
Submitted for Publication: April 2, 2014; final revision received June 26, 2014; accepted July 4, 2014.
Corresponding Author: Gerrit R. J. Melles, MD, PhD, Netherlands Institute for Innovative Ocular Surgery, Laan op Zuid 88, Rotterdam, Zuid-Holland 3071 AA, the Netherlands (email@example.com).
Published Online: September 11, 2014. doi:10.1001/jamaophthalmol.2014.3328.
Author Contributions: Dr Melles had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Dirisamer, Dapena, Melles.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Lam, Lie, Melles.
Critical revision of the manuscript for important intellectual content: Lam, Baydoun, Dirisamer, Dapena, Melles.
Statistical analysis: Lam.
Administrative, technical, or material support: Lam, Baydoun, Dirisamer, Lie.
Study supervision: Dirisamer, Dapena, Melles.
Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Lam received the Pfizer Ophthalmic Fellowship through the Royal College of Ophthalmologists in London to support him in his corneal fellowship at the Netherlands Institute for Innovative Ocular Surgery. This fellowship grant is unrelated to this study. Drs Baydoun and Dapena received a World Ophthalmology Congress 2014 travel grant unrelated to this study. Dr Melles is a consultant for DORC International/Dutch Ophthalmic USA. No other disclosures are reported.