Descemet Graft Research Articles

Resource maximization during COVID-19 crunch - A novel graft marking technique to use one cornea for two recipients for either Descemet membrane endothelial keratoplasty or deep anterior lamellar keratoplasty.

During pandemic, eye banks worldwide faced drastic reduction of corneal retrieval. Since precut tissues are uncommon in India, the onus is on corneal surgeons to maximize the donor corneal usage. The stumbling block making the stromal part unusable in the graft preparation of the Descemet membrane endothelial Keratoplasty is the stromal window creation to place the orientation mark on the stromal side of the folded graft. To overcome this, we simply marked directly on the folded Descemet graft, after partly stripping it from the donor cornea. Then, we unfolded, punched, and created two grafts after completely stripping it, retaining the Descemet part for the endothelial Keratoplasty and the stromal part for lamellar Keratoplasty. Initially, we used a special F marker, and later we marked an L using just a spatula. This simple technique needing no extra donor tissue or instrumentation maximizes the tissue utility with proper patient selection.

DMEK IN KERATOCONUS WITH ACUTE HYDROPS: A CASE REPORT

We describe the initial results of a Descemet Membrane Endothelial Keratoplasty (DMEK) procedure for acute corneal hydrops in a 45-years female with keratoconus that presented with severe visual loss in the left eye (OS). The patients’ best-corrected visual acuity at presentation in the right eye was 20/80 and hand motion in OS. Slit-lamp examination revealed an extensive Descemet’s membrane tear and stromal corneal edema in OS. Two months after the DMEK procedure, the patient presented with a best corrected visual acuity of 20/200 in the affected eye, corneal edema improvement, and an attached Descemet graft.

Новая стратегия кератопластики: расслоение и разделение роговицы донора.

Purpose: to report the first case of the same day use of a single donor cornea tissue for 5 consecutive transplantations in patients with endothelial dystrophy and keratoconus. Material and methods. The Descemet membrane (DM) and the endothelium were fully detached from the stroma of the donor on the endothelial side, using first a circular and then a direct corneal cutter, forming four partial triangle-shaped Descemet grafts (each constituted a 1/4 Descemet graft). The stroma of the donor's cornea was separated by a circular cutter of a preset diameter. One patient with keratoconus underwent deep anterior lamellar keratoplasty (DALK) using the big bubble technique. Then, each of the four Descemet grafts was sequentially transplanted by the Descemet membrane endothelial keratoplasty (DMEK) method to 4 patients with Fuchs dystrophy and pseudophakia. Results. During and after the surgery, there were no complications requiring additional intervention. Six months after DALK, the best corrected visual acuity (BCVA) was 0.66. Six months after four operations each using ¼ DMEK the average BCVA was 0.95 ± 0.1 (from 0.8 to 1.0) and the endothelial cell density was, respectively, 2839 after DALK and 1680 ± 254 cells/mm2 (from 1492 to 2039 cells/mm2) after ¼ DMEK. Conclusion. One donor cornea can be successfully transplanted to at least 5 patients. Despite the fact that the implementation of DALK and several operations using the ¼ DMEK technique within one day is technically difficult, this approach can contribute to a drastic reduction in the deficit of donor tissue. This approach requires a thorough selection of patients and an extensive surgical experience. For citation: Oganesyan O.G., Makarov P.V., Grdikanyan А.А., Getadaryan V.R. A new strategy of keratoplasty: laminating and splitting the donor cornea. Russian ophthalmological journal. 2018; 11 (3): 11-8 (In Russian). doi: 10.21516/2072-0076-2018-11-3-11-18. The full English version is available online at http://roj.igb.ru/eng/issues

Descemet Membrane Endothelial Transfer: Ultimate Outcome.

To evaluate the clinical outcome of 16 eyes undergoing Descemet membrane endothelial transfer (DMET). In this retrospective cohort study, a consecutive series of 16 eyes from 16 patients was evaluated after subtotal detachment of the Descemet graft after a Descemet membrane endothelial keratoplasty procedure (n = 8) or intended DMET (n = 8) for either Fuchs endothelial dystrophy (n = 10) or bullous keratopathy (BK; n = 6). All 8 Descemet membrane endothelial keratoplasty procedures were complicated by subtotal detachment of the donor graft. The remaining 8 eyes that underwent a DMET procedure were uneventful and no postoperative complications occurred, except 1 eye with BK that experienced a postoperative wound leak. Throughout all postoperative time points, the partially attached status of all Descemet grafts was maintained. Although all eyes operated on for Fuchs endothelial dystrophy showed initial central corneal clearance, no eye operated for BK demonstrated any degree of corneal deturgescence. Ultimately, all 16 corneas decompensated and 15 of the 16 patients elected retransplantation, while 1 patient declined further surgery for health reasons. Retransplantation was performed on average 10.3 (±7.4) months (range, 3-31 mo) postoperatively. Ultimately, regardless of the etiology of endothelial dysfunction, DMET fails to provide satisfactory results in the long term; durable corneal clearance may therefore require the presence of a nearly completely attached Descemet graft.

Preliminary outcome of hemi-Descemet membrane endothelial keratoplasty for Fuchs endothelial dystrophy

Background/aimsTo evaluate the clinical outcome of a full-diameter, untrephined, semicircular Descemet graft in a consecutive series of Descemet membrane endothelial keratoplasty (hemi-DMEK), potentially allowing the harvesting of two grafts from...

Causes that influence the detachment rate after Descemet membrane endothelial keratoplasty.

To investigate Descemet graft (DG) detachment rate after Descemet membrane endothelial keratoplasty (DMEK) in relation to DG position. A total of 175 consecutive pseudophakic eyes that underwent DMEK (175 eyes for Fuchs endothelial dystrophy) from September 2009 through February 2014 at the Tübingen Eye Hospital DG position were studied retrospectively by surgical video at the end of an operation. A group of 45 eyes showed a decentration of the DG with a stromal gap of ≥1.5mm over at least 3clock hours between the descematorhexis edge and the DG. DG detachment was documented at a mean follow-up of 13.9 ± 3.7months after surgery. DG detachment was defined as a detachment of 20% or more of the DG surface area. Various donor characteristics and patient characteristics were analyzed. The best spectacle-corrected visual acuity (BCVA) in the group of eyes with central well-positioned DG differed significantly from those of eyes with decentered DG. The preoperative BCVA in the central well-positioned DG group was 0.63 ± 0.40 logMAR, and in the decentered DG group 0.91 ± 0.51 logMAR (P < 0.001). The postoperative BCVA in the group of eyes with central well-positioned DG was 0.12 ± 0.11 logMAR, and in the group with decentered DG 0.23 ± 0.29 logMAR (P < 0.001). Endothelial cell density and patient characteristics such as age, gender, and intraocular pressure did not differ significantly between the two groups. The group of eyes with central well-positioned DG showed DG detachment in 12%; the group with decentered DG findings had DG detachment in 87% (P < 0.001)at the 12 month follow up. The present findings demonstrate the importance of central well-positioned DG and the relation of disease severity. Central well-positioned DG may reduce the incidence of DG detachment. Overlapping of the donor DG and the host Descemet membrane seems to be responsible for DG detachment. One possible way to enhance graft adhesion could be a larger descematorhexis, which avoids an overlapping. The second possible way could be not waiting too long for surgery to reduce disease severity.

Simple novel surgical device to facilitate preparation of endothelial grafts for DMEK

AbstractPurpose Descemet’s membrane and endothelial keratoplasty (DMEK) is the only technique to achieve a posterior corneal transplant without any interface. It is however difficult to achieve, mainly during the preparation of the graft. We propose a simple novel surgical device to facilitate this preparation.Methods We used 20 corneas for research in this study. The device is in the form of a circular blade with a diameter of 8.5 mm. The blade is broken in two opposite points on 4 mm. The blade is applied to the cornea in order to cut Descemet's membrane and deep stroma. Preparation of the endothelial graft is performed on an artificial anterior chamber using a simple technique with a 27 gauge cannula to detach Descemet’s membrane. The dissection is performed under the microscope by a corneal bank technician in 8 cases, by a resident in 4 cases by a cataract surgeon in 8 cases. The three operators followed a specific course on the same day of dissection but had never tried to dissect a DMEK graft before. Endothelial cell density (ECD) was measured before dissection and 3 days after storage in organ culture. Histological analysis was performed.Results Descemet graft could be easily dissected in 19 cases and resident failed dissection in one case. ECD was 2534±275 cells/mm² before dissection and 2412+/‐360 cells/mm², after 3 days in organ culture (p&lt;0.01). Histological sections confirm that the cleavage was between DM and posterior stroma.Conclusion We report here a reliable and efficient technique for the preparation of pure descemetic grafts. Commercial interest

Descemet Membrane Endothelial Keratoplasty

To evaluate the learning curve of standardized "no-touch" Descemet membrane endothelial keratoplasty (DMEK) of a single surgeon in a clinic without an in-house eye bank facility. For 25 eyes of 22 patients, Descemet graft preparation and DMEK surgery were performed according to the protocols of the Netherlands Institute for Innovative Ocular Surgery with minor modifications. Best spectacle-corrected visual acuity, subjective refraction, Scheimpflug tomography, and endothelial cell density of the donor tissue were documented before and at 1, 3, and 6 months after the surgery; intraoperative and postoperative complications were recorded. No donor tissue was lost owing to preparation failure of the Descemet graft allowing all surgeries to be completed as planned. At 6 months after the surgery, 83% of eyes reached ≥ 20/40 (≥ 0.5), 48% ≥ 20/28 (≥ 0.8), and 30% (7/23) ≥ 20/20 (≥ 1.0) (n = 23). The mean change in both spherical equivalent and refractive cylinder in the same interval was minimal (0.03 diopter). Mean donor endothelial cell density decreased from 2444 cells per square millimeter (± 198 cells/mm2) before the surgery to 1331 cells per square millimeter (± 491 cells/mm2) at 6 months after the surgery. Partial graft detachment requiring rebubbling occurred in 9 cases (36%). One eye developed primary graft failure. This case series shows that DMEK by a beginning surgeon can achieve good results without preparing the graft in an eye bank. Use of backup tissue can be avoided by strictly adhering to the Netherlands Institute for Innovative Ocular Surgery techniques and by harvesting the graft the day before surgery to avoid last-minute cancellation.

Outcome of phacoemulsification after Descemet membrane endothelial keratoplasty (DMEK)

AbstractPurpose To evaluate the incidence of secondary cataract after Descemet membrane endothelial keratoplasty (DMEK), and the feasibility and outcome of phacoemulsification cataract surgery after DMEK.Methods From a series of 106 consecutive phakic eyes that underwent DMEK for Fuchs endothelial dystrophy or bullous keratopathy, five (4.7%) required cataract surgery 9.2 (± 3.8) months (range 4 to 14 months) after the initial DMEK. Outcomes after phacoemulsification were retrospectively assessed by reviewing the change in visual acuity, refractive error, endothelial cell density, pachymetry and the incidence of complications.Results All phacoemulsification surgeries were uneventful and no dislocations and/or detachments of the Descemet graft were observed. At 6 to 12 months after phacoemulsification, all eyes reached a BCVA of ≥20/30 (0.6) and were within 0.5D of the intraocular lens power calculations. Endothelial cell density decreased from on average 1535 (± 195) cells/mm2 before, to 1158 (± 250) cells/mm2. No significant changes in pachymetry values were observed up to 6 to 12 months postoperatively. All corneas remained clear throughout the study period.Conclusion Phacoemulsification after DMEK can be performed with minimal risk of graft dislocation and may provide good refractive and visual outcomes with an acceptable decrease in endothelial cell density.

Outcomes of phacoemulsification after Descemet membrane endothelial keratoplasty

To evaluate the feasibility and outcomes of phacoemulsification after Descemet membrane endothelial keratoplasty (DMEK) for Fuchs endothelial dystrophy and bullous keratopathy. Tertiary referral center. Comparative case series. The case notes of all phakic DMEK patients who subsequently had cataract surgery were reviewed, and data from a prospectively recorded database were analyzed. This included demographic details, visual acuity, corneal pachymetry, endothelial cell density (ECD), refractive outcomes, and complications. From a series of 106 consecutive phakic DMEK eyes, 5 eyes (4.7%) required phacoemulsification a mean of 9.2 months ± 3.7 (SD) (range 4 to 14 months) after the initial DMEK. All phacoemulsification procedures were uneventful, and no dislocations and/or detachments of the Descemet graft were observed. At 6 to 12 months, all eyes reached a corrected distance visual acuity of 20/30 (0.6) or better and were within ±0.50 diopter of the target refraction. Endothelial cell density decreased from a mean of 1535 ± 195 cells/mm(2) before phacoemulsification to 1158 ± 250 cells/mm(2) 6 to 12 months after phacoemulsification. No significant changes in pachymetry values were observed, and all corneas remained clear throughout the study. Phacoemulsification after DMEK can be performed with minimal risk for graft detachment. The postoperative refractive outcomes were predictable, and visual acuity is likely to improve; there was an acceptable decrease in ECD. Dr. Melles is a consultant to D.O.R.C. International BV/Dutch Ophthalmic USA. No author has a financial or proprietary interest in any material or method mentioned.

Intraocular Graft Unfolding Techniques in Descemet Membrane Endothelial Keratoplasty

To define various Descemet graft unfolding techniques in Descemet membrane endothelial keratoplasty. In a retrospective analysis, the surgical videos of 100 consecutive Descemet membrane endothelial keratoplasty cases with at least 6 months of follow-up were evaluated and categorized. The Descemet graft unfolding methods were categorized into 4 basic techniques and 3 auxiliary techniques. All Descemet membrane endothelial keratoplasty surgical procedures could be completed using (a combination of) 4 Descemet graft unfolding techniques: (1) standardized no-touch graft unfolding using a double roll, (2) carpet unrolling while fixating 1 graft edge (Dirisamer technique), (3) small air bubble-assisted unrolling (Dapena maneuver), (4) the single sliding cannula maneuver. Additional maneuvers included turning over the graft when oriented upside down (flushing); manual graft centration with a cannula; and bubble bumping to unfold peripheral inward folds. In 73% of surgical procedures, technique 1 was used, while a combination of techniques was used in 44% and auxiliary techniques in 62%. None of the techniques showed a correlation with the best-corrected visual acuity, endothelial cell density, or postoperative complication rate (P > .10). Descemet membrane endothelial keratoplasty may be further facilitated by using controlled techniques for unfolding the Descemet graft inside the recipient anterior chamber, either as stand-alone techniques or used in various combinations.

Identifying causes for poor visual outcome after DSEK/DSAEK following secondary DMEK in the same eye

To identify causes of reduced visual acuity after Descemet stripping (automated) endothelial keratoplasty (DSEK/DSAEK) and to determine whether such eyes can be successfully 'repaired' with a secondary Descemet membrane endothelial keratoplasty (DMEK). Twelve eyes of 12 patients, who underwent secondary DMEK to manage poor visual outcome after initial DSEK/DSAEK, were evaluated with biomicroscopy, Pentacam imaging, and specular and confocal microscopy, before and at 1, 3 and 6 months after DMEK. Four causes of reduced optical quality of the transplanted host cornea could be identified in DSEK/DSAEK: five eyes (42%) showed large host-Descemet remnants within the visual axis during surgery; six eyes (50%) irregular graft thickness; six eyes subtle 'stromal waves'; and nine eyes (75%) high reflectivity at the donor-to-host interface. After DMEK graft replacement, all corneas cleared and achieved a best corrected visual acuity of ≥20/25 (≥0.8), except for one with a partial Descemet graft detachment. Pachymetry values decreased from 670 (±112) μm before to 517 (±57) μm after secondary DMEK. Higher-order aberrations (Coma and Trefoil) at the posterior surface tended to be lower (p = 0.07) in DMEK grafts than in DSEK/DSAEK grafts. Host-Descemet remnants at the donor-to-host interface, interface reflectivity, graft thickness irregularity and donor stromal contraction may contribute to poor visual outcome after DSEK/DSAEK, without causing permanent host corneal damage, as in most cases, complete visual recovery could be achieved by performing a secondary DMEK.

Recipient Endothelium May Relate to Corneal Clearance in Descemet Membrane Endothelial Transfer

To describe corneal clearance after re-endothelialization of the recipient posterior stroma through Descemet membrane endothelial transfer (DMET) (ie, a "free-floating" donor Descemet graft in the recipient anterior chamber after descemetorhexis), in managing corneal endothelial disorders. Nonrandomized prospective study at a tertiary referral center. Twelve eyes enrolled in our study, 7 suffering from Fuchs endothelial dystrophy and 5 with bullous keratopathy. The clinical outcome was monitored by biomicroscopy, optical coherence tomography, confocal microscopy, endothelial cell density, and pachymetry measurements. All eyes operated on for Fuchs endothelial dystrophy showed corneal clearance, with pachymetry values returning to normal (533 ±47 μm). The denuded recipient stroma re-endothelialized with an average endothelial cell density of 797 (± 743) cells/mm(2) at 6 months after surgery. In contrast, none of the bullous keratopathy eyes showed any improvement throughout the follow-up period. DMET may be effective in the management of Fuchs endothelial dystrophy (primarily a Descemet membrane disorder), but not in bullous keratopathy (primarily an endothelial depletion). Apparently, the indication for surgery (ie, a "dystrophy" vs a "depletion" of recipient endothelial cells) relates to the capacity of the cornea to clear. This suggests that the remaining rim of recipient endothelium (after descemetorhexis) is involved in the re-endothelialization of the recipient posterior stroma after DMET.

Causes of Glaucoma After Descemet Membrane Endothelial Keratoplasty

To describe the incidence and causes of glaucoma after Descemet membrane endothelial keratoplasty (DMEK). Nonrandomized prospective cohort study at a tertiary referral center. The incidence of glaucoma was evaluated in the first 275 consecutive eyes that underwent DMEK for Fuchs endothelial dystrophy (260 eyes) or bullous keratopathy (15 eyes). Glaucoma was defined as a postoperative intraocular pressure (IOP) elevation of ≥24 mm Hg, or ≥10 mm Hg from the preoperative baseline. If possible, the cause of glaucoma was identified, and best-corrected visual acuity (BCVA), endothelial cell density (ECD), and postoperative course were documented, with a mean follow-up of 22 (± 13) months. Overall, 18 eyes (6.5%) showed postoperative glaucoma after DMEK. Seven eyes (2.5%) had an exacerbation of a pre-existing glaucoma. Eleven eyes (4%) presented with a de novo IOP elevation, associated with air bubble-induced mechanical angle closure (2%), steroid response (0.7%), or peripheral anterior synechiae (0.4%), or without detectable cause (0.7%). Two eyes (0.7%) required glaucoma surgery after DMEK. At 6 months, all eyes had a BCVA of ≥20/40 (≥0.5), and 81% reached ≥20/25 (≥0.8) (n = 16); mean ECD was 1660 (± 554) cells/mm(2) (n = 15) (P > .1). Glaucoma after DMEK may be a relatively frequent complication that could be avoided by reducing the residual postoperative air bubble to 30% in phakic eyes, applying a population-specific steroid regime, and avoiding decentration of the Descemet graft. Eyes with a history of glaucoma may need close IOP monitoring in the first postoperative months, especially in eyes with an angle-supported phakic intraocular lens.

Incidence of Early Allograft Rejection After Descemet Membrane Endothelial Keratoplasty

To report the incidence of early allograft rejection after Descemet membrane endothelial keratoplasty (DMEK), that is, transplantation of isolated Descemet membrane with its endothelium. The first series of 120 eyes of 105 patients operated on for Fuchs endothelial dystrophy or pseudophakic bullous keratopathy, with an average 2 years of follow-up after 9.0- to 10.0-mm-diameter DMEK, enrolled in our study. During the entire study period, only 1 of the eyes showed any signs of a cellular immune response to the Descemet graft. A 76-year-old patient presented with discomfort, reduced visual acuity to counting fingers, corneal decompensation, and a Khodadoust line in the central cornea 4 months after (decentered) DMEK. Intensified topical corticoid therapy resulted in a complete visual recovery to 20/25 (0.8) within weeks. A "classic" allograft rejection (with an appearance similar to that after penetrating keratoplasty) can occur after DMEK. However, compared with the earlier (endothelial) keratoplasty procedures, DMEK may be associated with a lower rejection rate of ≤ 1%, despite transplant diameters of ± 9.5 mm. The apparent immune tolerance in DMEK may result from either less "upregulation" or more "downregulation" of the immune system.

Optical Coherence Tomography, Scheimpflug Imaging, and Slit-Lamp Biomicroscopy in the Early Detection of Graft Detachment After Descemet Membrane Endothelial Keratoplasty

To evaluate the efficacy of anterior segment optical coherence tomography (OCT), Scheimpflug imaging, and slit-lamp biomicroscopy in the early detection of a (partial) graft detachment after Descemet membrane endothelial keratoplasty (DMEK). Anterior segment OCT, Scheimpflug imaging, and slit-lamp biomicroscopy were performed in 120 eyes of 110 patients after DMEK. Seventy-eight eyes showed a normal corneal clearance, and the attached Descemet grafts could not be identified with any of the imaging techniques. Forty-two eyes showed persistent stromal edema in the first postoperative month. In transplanted corneas that (partially) did not clear in the early postoperative period, OCT had an added diagnostic value in 36% of cases (15 of 42 eyes) in visualizing whether the graft was detached and, in particular, to discriminate between a "flat" graft detachment and delayed corneal clearance. In contrast, in the presence of corneal edema, Scheimpflug imaging did not provide more information than slit-lamp biomicroscopy in the detection of a graft detachment. Anterior segment OCT may be an effective tool in the detection of an early graft detachment after DMEK, to determine if secondary surgical intervention is indicated or is to be avoided.

Patterns of Corneal Endothelialization and Corneal Clearance After Descemet Membrane Endothelial Keratoplasty for Fuchs Endothelial Dystrophy

To describe various endothelial migration healing patterns after Descemet membrane endothelial keratoplasty (DMEK), and to determine the contribution of the donor and host endothelium in the clearance of a transplanted cornea. Nonrandomized, prospective clinical study. In a total of 150 consecutive eyes that underwent DMEK (ie, transplantation of an isolated Descemet graft, for Fuchs endothelial dystrophy), re-endothelialization patterns were studied. Of these eyes, 36 showed a "stromal gap" between the "descemetorhexis edge" and the graft, or (partial) graft detachment. Endothelialization patterns of the host posterior stroma were documented at 1, 3, 6, 9, 12, and 24 months after surgery with Pentacam imaging, specular microscopy, optical coherence tomography, confocal microscopy, and slit-lamp biomicroscopy. Complete corneal clearance was seen in 28 of 36 eyes (78%) with a stromal gap, or (partial) detachment, progressing from the periphery toward the center; and 27 of 34 eyes (79%) with normal visual potential reached a visual acuity of ≥20/40 (≥0.5) or better. In 3 eyes that had the Descemet graft implanted upside-down, a "reversed corneal clearance pattern" was observed (ie, persistent edema where the graft was attached), while the area overlying the detachment cleared. One case that had a "descemetorhexis" performed without endothelial graft implantation showed persistent stromal edema. The presence of donor endothelium in the recipient anterior chamber may be required for endothelial migration and/or recovery of corneal clarity. Re-endothelialization may be associated with massive endothelial migration and some form of cell signaling to draw donor endothelial cells toward the recipient posterior stroma ("homing").

Endothelial keratoplasty: DSEK/DSAEK or DMEK - the thinner the better?

Endothelial keratoplasty has been adopted worldwide as an alternative to penetrating keratoplasty in the treatment of corneal endothelial disorders. Descemet stripping (automated) endothelial keratoplasty (DSEK/DSAEK) may be the current standard, whereas Descemet membrane endothelial keratoplasty (DMEK), that is, isolated transplantation of Descemet membrane, may allow further improvement of clinical outcome. DSEK/DSAEK may still have three major challenges: suboptimal visual acuity and relatively slow visual rehabilitation, limited accessibility due to required investments in equipment or the purchase of predissected tissue, and a drop in donor endothelial cell density in the early postoperative phase. Although DMEK may allow much quicker and (near) complete visual rehabilitation as well as easier logistics in donor preparation, the surgical technique may initially require more training to obtain consistent outcomes. Compared with DSEK/DSAEK, DMEK may have higher clinical potential with 75% of cases reaching 20/25 or better (> or =0.8) within 1-3 months. Furthermore, preparation of isolated Descemet grafts does not require large investments and may increase overall donor tissue availability. Hence, corneal surgeons may consider 'to make the switch' from DSEK/DSAEK to DMEK.

Causes of Primary Donor Failure in Descemet Membrane Endothelial Keratoplasty

To determine the causes of primary failure of donor Descemet membrane transplants in Descemet membrane endothelial keratoplasty (DMEK). Laboratory and clinical study. From a larger series of eyes that underwent DMEK for Fuchs endothelial dystrophy or bullous keratopathy, 11 transplanted corneas did not clear within the first week after surgery. During a secondary 'Descemet stripping endothelial keratoplasty' procedure, the first graft was carefully removed and analyzed with light microscopy. For each patient, the surgical video of the initial DMEK procedure was analyzed to determine the cause of transplant failure. With light microscopy, all but two explanted transplants had a endothelial cell density of 1700 cells/mm2 or higher. Three Descemet grafts may have been positioned upside down and three may have failed due to manipulation. One transplant could not be properly attached onto the recipient posterior stroma because of insufficient air-support in a post-vitrectomy eye. For four eyes, no obvious cause of failure could be determined. Unexpectedly, graft failure after DMEK may not primarily result from tissue damage during implantation and associated endothelial cell loss. Detached grafts may therefore be managed by early reposition. When the graft is attached after surgery but fails to clear, secondary surgical intervention may be postponed, because these corneas may clear spontaneously after several weeks.