Long-term Outcomes in Fellow Eyes Comparing DSAEK and DMEK for Treatment of Fuchs Corneal Dystrophy

The progress and future of corneal endothelial transplantation

Endothelial transplantation has recently been accepted worldwide, in the long history of corneal transplantation. The introduction of endothelial keratoplasty (Descemet stripping automated endothelial keratoplasty and Descemet membrane endothelial keratoplasty) has enabled us to expand the surgical indications owing to the low incidence of rejection and quick recovery of visual function. New technologies have been developed to ensure stable postoperative outcomes with a shorter learning curve, such as transplantation using cultured human endothelial cells and induced pluripotent stem cells (iPS) or new devices such as artificial endothelium. This review discusses the history and characteristics of corneal transplantation alongside new treatment options that may offer hope for patients with endothelial disease in the future.

Simple Modification of a DSAEK Carrier Enables Graft Insertion by Injection Without Compromising Endothelial Cell Viability

PURPOSE

We evaluate the quality and feasibility of preloading Descemet stripping automated endothelial keratoplasty (DSAEK) grafts into a modified EndoGlide Ultrathin system for graft injection.

METHODS

DSAEK grafts were prepared by experienced processing technicians at 2 separate locations, loaded into a modified EndoGlide Ultrathin, and placed in storage media. Grafts processed at one location were shipped cross-country overnight to the other location and were examined on arrival for positioning within the modified EndoGlide Ultrathin. All grafts were ejected and analyzed for endothelial cell loss (ECL) with calcein acetoxymethyl staining and FIJI segmentation. A subset of grafts was measured by optical coherence tomography for graft thickness 1 hour after cut, 1 hour after loading, and 1 day after loading.

RESULTS

No grafts were displaced from the modified carrier over 3 shipping events (n = 9), and all grafts (n = 18) were successfully ejected. Grafts loaded into the modified carrier and ejected exhibited no more cell loss than grafts loaded into the standard carrier and removed by pull-through (14.0% ± 2.8% vs. 12.2% ± 3.4%, respectively, P = 0.24). Carrier modification skills can be successfully transferred as grafts loaded by a processing technician new to carrier modification were within the acceptable limit of 25% ECL for transplant DSAEK grafts. Graft thickness increased significantly (P < 0.05) between the postcut and 1-hour postload measurement and the postcut and 24-hour postload measurement.

CONCLUSIONS

The EndoGlide Ultrathin can be modified to enable its use for graft injection while not compromising the ability to use the pull-through method for graft delivery. Preloaded DSAEK grafts swell significantly during the 24-hour storage period, and patterns of ECL may be linked to swelling.

Clinical outcome analysis of two approaches to trypan blue dyeing for DMEK

To evaluate the clinical implications of the different trypan blue dyeing techniques used during liquid bubble (LBT) and manual peel (MPT) DMEK lenticule preparation techniques. This study retrospectively compared the degree to which endothelial cells are preserved using selective Descemet Membrane (DM) staining (LBT) versus bath-staining (MPT) when performed by a single surgeon, sourced from a single eye bank. Endothelial cell density measured after the 3-month follow-up was 1805 and 1916 cells/mm2 respectively, differing significantly (p = 0.012). A double-scroll graft formation was found and maintained until implantation in 94% of preparations with bath staining and 50% of preparations using selective DM staining. Preoperative visual acuity was comparable between preparation techniques at 0.4 logMAR as well as postoperatively, at an average of 0.1 logMAR. Reducing chemical stress on the endothelium by avoiding any contact with trypan blue allows for a significantly higher degree of cell preservation. However, achieving the often-desired double-scroll graft formation was possible less frequently. It remains unclear which factors define the differences graft scrolling behavior observed between LBT and MPT.

Outcomes of Donor Corneas of Different Thicknesses From Endothelial Keratoplasty in Deep Anterior Lamellar Keratoplasty for Keratoconus

PURPOSE

To evaluate the outcomes of divided residual donor corneas obtained from endothelial keratoplasty in keratoconus with deep anterior lamellar keratoplasty (DALK).

METHODS

In this retrospective, comparative, clinical study, 103 keratoconic eyes that underwent DALK were enrolled; 67 eyes received thin grafts from Descemet stripping automated endothelial keratoplasty, and 36 received thick grafts from Descemet membrane endothelial keratoplasty. Baseline and postoperative central corneal thickness (CCT), inferior corneal thickness, uncorrected distance visual acuity, corrected distance visual acuity, corneal astigmatism, mean keratometry, biomechanical properties, and complication rates were measured.

RESULTS

Six months after transplantation, the group receiving thin grafts had a CCT of only 455.1 ± 43.0 μm, whereas that of the group receiving thick grafts was 546.7 ± 44.2 μm. Both CCT and inferior corneal thickness in the thin group were significantly lower than those in the thick group (measured with Pentacam at 36 months, P < 0.001) and remained throughout the 5-year follow-up period. Both procedures had comparable postoperative logarithm of the minimum angle of resolution UDVAs, logarithm of the minimum angle of resolution corrected distance visual acuity, astigmatism, and mean keratometry values (36 months; P = 0.335, 0.286, 0.680, and 0.365, respectively). Corneal biomechanical analysis revealed that the thin group had a significantly higher stiffness parameter at the first applanation than the thick group at the 2-year follow-up (P = 0.036) while other parameters were equivalent.

CONCLUSIONS

The outcomes of keratoplasty with donor tissue are comparable regardless of the thickness of the graft, which suggests that transplantation with either type of the split corneal procedure for DALK in patients with keratoconus is feasible.

Mediators of Visual Acuity in Descemet Membrane Endothelial Keratoplasty and Ultrathin Descemet Stripping Automated Endothelial Keratoplasty

PURPOSE

The aim of this study was to investigate mediators of visual acuity in ultrathin Descemet stripping automated endothelial keratoplasty (UT-DSAEK) compared with Descemet membrane endothelial keratoplasty (DMEK).

METHODS

This was a prespecified secondary analysis of the Descemet Endothelial Thickness Comparison Trial, a prospective, randomized controlled trial comparing UT-DSAEK with DMEK. Subjects with Fuchs endothelial dystrophy or pseudophakic bullous keratopathy presenting to 2 academic centers were enrolled and randomized to either UT-DSAEK (n = 25 eyes) or DMEK (n = 25 eyes). Higher order aberrations (HOAs) and corneal densitometry were measured with Pentacam Scheimpflug imaging at 3, 6, 12, and 24 months.

RESULTS

The posterior corneal surface at the 6.0-mm optical zone had significantly less total HOAs ( P <0.001) in the DMEK group compared with UT-DSAEK at 24 months. Anterior and posterior corneal densitometry improved from baseline to 24 months for both UT-DSAEK and DMEK, but there was no significant difference between the 2 groups. Corneal densitometry and posterior HOAs were both associated with best-corrected visual acuity ( P <0.05). DMEK had 1.3 logarithm of the minimum angle of resolution better visual acuity compared with UT-DSAEK at 24 months. Approximately 64% of this effect was mediated through posterior HOAs, whereas none was mediated through anterior HOAs or densitometry.

CONCLUSIONS

Decreased posterior HOAs mediate better visual acuity and account for improved vision after DMEK compared with UT-DSAEK. Corneal light scatter as measured by densitometry is similar between UT-DSAEK and DMEK, indicating that the increased thickness and stromal-stromal interface in UT-DSAEK do not significantly affect visual acuity.

Comparison of Long-Term Outcomes of DSEK and DMEK in Fuchs Endothelial Dystrophy

PURPOSE

This study aimed to compare the long-term endothelial cell loss, graft survival, and clinical outcomes in patients with Fuchs endothelial dystrophy (FED) after Descemet stripping endothelial keratoplasty (DSEK) and Descemet membrane endothelial keratoplasty (DMEK) using a standardized surgical protocol.

METHODS

Three hundred and six consecutive DSEK and DMEK grafts of 223 patients with FED performed by 8 surgeons between January 2006 and August 2022 were analyzed. The primary outcome measures were graft survival, endothelial cell loss, and best spectacle-corrected visual acuity.

RESULTS

At 5 years, graft survival was 96% for both DSEK and DMEK eyes. The mean percentage of endothelial cell loss was 57.7 ± 17.1 in DSEK and 56.8 ± 15.2 in DMEK eyes ( P = 0.430). The mean best spectacle-corrected visual acuity was 0.13 ± 0.14 logMAR in DSEK and 0.01 ± 0.18 logMAR in DMEK grafts ( P <0.00001) at 5 years postoperatively. Rebubbling was performed in 7.8% DSEK and 2.1% DMEK grafts ( P = 0.441). Cox regression identified rejection episodes (HR 6.5; 95% CI: 1.70-24.8; P = 0.0062) as a significant contributing factor for graft failure.

CONCLUSIONS

DMEK had superior visual acuity outcomes compared with DSEK in these patients up to 5 years after surgery. At 5 years, there was no significant difference in graft survival or endothelial cell loss between DSEK and DMEK eyes with FED. We propose that our standardized technique reduces the need for rebubbling.

Size and Shape Matter: Cell Viability of Preloaded Descemet Stripping Automated Endothelial Keratoplasty Grafts in Three Different Carriers

PURPOSE

The aim of this study was to examine endothelial cell loss (ECL) associated with preloaded Descemet stripping automated endothelial keratoplasty (DSAEK) grafts loaded into 3 carriers of different size and shape.

METHODS

Thirty-six donor corneas were prepared for DSAEK and loaded into an EndoGlide Ultrathin (control) (2.0 mm × 3.5 mm lumen, 4.5 mm/4.9 mm incision for scleral tunnel/clear corneal insertion), Descemet membrane endothelial keratoplasty EndoGlide (experimental) (1.1 mm × 1.7 mm lumen, 2.65 mm incision), or round glass Jones tube (experimental) (1.8 mm lumen, 3.0 mm incision). Control grafts were stored for 6 days in Optisol-GS and experimental grafts stored for 24 hours in Life4C before analysis using Calcein AM staining. Grafts were imaged and ECL was analyzed by FIJI segmentation. The statistical significance of ECL was determined using 1-way ANOVA and Tukey post hoc analysis.

RESULTS

There were no significant differences in donor characteristics for grafts in each cohort. ECL for grafts loaded into the EndoGlide Ultrathin was 10.3% ± 2.3% (graft thickness: 60-189 μm, n = 9). ECL for grafts loaded into the Descemet membrane endothelial keratoplasty EndoGlide was 22.2% ± 7.1% (graft thickness: 38-63 μm, n = 9). ECL for thin grafts (34-60 μm, n = 9) loaded into the Jones tube was 24.0% ± 5.0%. ECL for thick grafts (92-119 μm, n = 9) loaded into the Jones tube was 34.2% ± 6.1% ECL ( P < 0.001). Combined regression analysis revealed that graft thickness is directly correlated to ECL ( P < 0.01).

CONCLUSIONS

The size and shape of the carrier can influence the cell viability of preloaded DSAEK grafts. Compared with a larger carrier, smaller lumen carriers are associated with greater ECL. In smaller lumen carriers, ECL increases with increasing DSAEK graft thickness.

Long-Term Outcomes of Descemet Membrane Endothelial Keratoplasty: Effect of Surgical Indication and Disease Severity

PURPOSE

The aim of this study was to evaluate clinical outcomes and graft survival in a large patient cohort up to 10 years after Descemet membrane endothelial keratoplasty (DMEK) based on surgical indication and Fuchs endothelial corneal dystrophy (FECD) severity.

METHODS

The cohort in this retrospective study included 750 eyes that underwent DMEK for FECD (86%), bullous keratopathy (BK, 9%), and other indications (5%). Based on the modified Krachmer grading, 186 eyes (29%) had moderate FECD (Krachmer grade 3-4) and 440 eyes (68%) had advanced FECD (Krachmer grade 5-6). Main outcome measures were best-corrected visual acuity (BCVA), central corneal thickness, endothelial cell density (ECD), endothelial cell loss (ECL), postoperative complication rate, and graft survival.

RESULTS

The mean 10-year BCVA was 0.08 ± 0.28 logMAR (n = 96), and the BCVA remained stable between 5 and 10 years postoperatively (all P > 0.05). The mean ECL at 5 and 10 years was 56% ± 17% (n = 460) and 66% ± 14% (n = 96), respectively ( P < 0.05). The 10-year graft survival rate for the total cohort was 0.85 (95% confidence interval, 0.82-0.89). BCVA and ECD differed at 1 to 5 years between eyes with different surgical indications (all P < 0.05), and FECD eyes had higher graft survival compared with BK eyes (0.90 vs. 0.60, P = 0.001). Moderate FECD eyes had better BCVA, ECD, and graft survival outcomes than advanced FECD eyes (all P < 0.05). Minor graft detachment ( P = 0.04) and lower donor ECD ( P = 0.01) were related to lower 10-year ECD.

CONCLUSIONS

Long-term outcomes and graft survival after DMEK were better in FECD eyes than BK eyes and were also related to FECD disease severity. Eyes operated for moderate FECD showed the highest graft survival probability and excellent long-term outcomes.

Descemet Membrane Endothelial Keratoplasty (DMEK) Reduces the Corneal Epithelial Thickness in Fuchs' Patients

Fuchs' endothelial corneal dystrophy (FECD) is the occurrence of corneal edema due to endothelial cell dystrophy. Descemet membrane endothelial keratoplasty (DMEK) is considered to be the gold standard of treatment. The aim of this study was to investigate the changes in the corneal epithelial thickness of FECD patients before and after DMEK and to compare these results with a healthy control cohort. In this retrospective analysis, 38 eyes of patients with FECD that were treated with DMEK and 35 healthy control eyes received anterior segment optical coherence tomography (OCT; Optovue, XR-Avanti, Fremont, CA, USA). The corneal epithelial thicknesses in different locations were analyzed and compared between the preoperative, postoperative, and control cohorts. The median follow-up time was 9 months. There was a significant degression of the mean epithelial thickness after DMEK in the central, paracentral, and mid-peripheral zones (p < 0.01) of the cornea. The total corneal thickness and stromal thickness decreased significantly as well. No significant differences were observed between the postoperative and control cohorts. In conclusion, the FECD patients had an increased epithelial thickness compared to the healthy controls, which decreased significantly after DMEK and reached thickness levels comparable to those of healthy control eyes. This study emphasized the importance of distinguishing between the corneal layers in anterior segment pathologies and surgical procedures. Moreover, it accentuated the fact that the structural alterations in FECD extend beyond the corneal stroma.

Intraocular Lens Calcification After Pseudophakic Endothelial Keratoplasty

PURPOSE

To determine the incidence and to document risk factors for intraocular lens (IOL) calcification after pseudophakic endothelial keratoplasty.

METHODS

We retrospectively studied 2700 consecutive penetrating, anterior lamellar and endothelial keratoplasties carried out between December 1992 and June 2022 at the National Eye Hospital, Paris, France.

DESIGN

Retrospective cohort study.

RESULTS

All IOL calcification cases were associated with endothelial keratoplasty. Out of 588 endothelial keratoplasty procedures, 576 eyes were pseudophakic at the end of surgery. Fourteen cases of IOL calcification were observed during follow-up. The cumulative incidence of IOL calcification after endothelial keratoplasty was 4.5%±1.3% at 60 months. Hydrophilic acrylic IOL material (P < .001) and use of SF₆ for anterior chamber tamponade (P = .001) were significantly and independently associated with the cumulative incidence of IOL calcification.

CONCLUSION

The incidence of IOL calcifications seems to be around 5%. Ophthalmologists should avoid hydrophilic acrylic IOLs in patients with endothelial disorders. When the patient already has a hydrophilic IOL, SF₆ should be avoided. The only effective treatment is IOL exchange.

Glasses-Assisted 3D Display System-Guided Descemet Membrane Endothelial Keratoplasty Tissue Preparation

PURPOSE

The aim of this study was to evaluate the feasibility of Descemet membrane endothelial keratoplasty (DMEK) tissue preparation using a glasses-assisted 3-dimensional (3D) display system and to compare it with a conventional surgical microscope.

METHODS

Healthy pairs of human corneas suitable for penetrating keratoplasty surgery were selected for this study. The tissues were randomly divided into 2 groups. Each pair of corneas had 1 cornea (group 1) prepared with NGENUITY (Alcon) with a 5-second staining time with vision blue, and the fellow cornea (group 2) was prepared using a OPMI Lumera 700 surgical microscope (Carl Zeiss Meditec, Jena, Germany) with a 30-second staining time. DMEK graft preparation time, speed of stripping, graft width, and endothelial cell loss were evaluated.

RESULTS

Twenty-eight pairs of corneas were included in this study. The graft preparation time was significantly higher in the 3D group than in the conventional group (498 ± 147 vs. 418 ± 85 seconds, P value = 0.031). The mean speed of stripping was 0.59 ± 0.081 mm/s in group 1 and 0.089 ± 0.005 mm/s in group 2 ( P value = 0.024). The mean endothelial cell density in group 1 and group 2 before tissue preparation was 2162 ± 115.21 and 2153 ± 122.45, respectively ( P value > 0.1). After tissue preparation, the endothelial cell density reduced to 1911 ± 150.72 in group 1 and 1998 ± 90.72 in group 2 ( P value = P value > 0.05). The graft width was 5.05 ± 0.71 mm in group 1 and 4.92 ± 0.23 mm in group 2 ( P value > 0.05).

CONCLUSIONS

DMEK tissue preparation with 3D display system NGENUITY is feasible with a slightly increased preparation time. The improved visualization allows a reduced staining time that could be beneficial for eye banks because it may reduce the toxic effect of staining colorants.