Midterm Results of Descemet Membrane Endothelial Keratoplasty: 4 to 7 Years Clinical Outcome.

PURPOSE: To evaluate the midterm outcomes of Descemet membrane endothelial keratoplasty (DMEK) up to 4-7 years postoperatively. DESIGN: Retrospective, consecutive interventional case series. METHODS: A total of 250 consecutive eyes of 209 patients who underwent DMEK at our institute and had potentially at least 4 years of follow-up. Main outcome measures were best-corrected visual acuity (BCVA), endothelial cell density (ECD), central corneal thickness (CCT), graft survival rate, and postoperative complications. RESULTS: At 6 months postoperatively, 73% of the eyes reached a BCVA of ≥20/25 (0.8) and 44% ≥20/20 (1.0), and remained stable up to 7 years (P = .7114). ECD dropped by 33.9% in the first 6 months, and then declined by a yearly rate of 9.0%. CCT did not show a significant change after 6 months up to 7 years (P = .8447). The cumulative graft survival rate at 4 and 7 years was 0.96 (95% confidence interval [0.93, 0.99]). Rebubbling procedures were performed in 4.4% of eyes, all within the first 6 postoperative months. Repeat transplantations were performed in 15.2% of cases to manage greater than one third graft detachment (11.6%), primary graft failure (1.6%), or secondary graft failure (2.0%), with the majority (79%) of repeat transplantations performed within the first year. After 6 months, the main complications were allograft rejection (2.4%) and secondary graft failure (2.0%). CONCLUSIONS: The visual acuity level achieved at 6 months after DMEK may remain stable up to at least 4-7 years; ECD shows a constant, slow decrease; and complications after the first 6 months occur in <5% of cases.

Evaluation of Early Graft Detachment After Descemet Membrane Endothelial Keratoplasty Using New Swept-Source Optical Coherence Tomography.

PURPOSE: To evaluate the ability of Swept-source optical coherence tomography (SS-OCT) in detecting early endothelial graft detachment after Descemet membrane endothelial keratoplasty (DMEK) and to determine the possible causes of graft detachment using 2 different protocols for anterior segment (AS) imaging. METHODS: Eight eyes from 8 consecutive DMEK surgeries were evaluated. High-resolution images were captured with SS-OCT using 2 different radial protocols for AS imaging (16- and 6-mm radial protocols). Central and peripheral corneal changes were evaluated. RESULTS: All eyes developed early graft detachment detectable with the SS-OCT. The 16-mm radial protocol scanned the cornea detecting early graft detachment in 6 out of 8 cases. The 6-mm radial protocol provided detailed images of the cornea, detecting early detachment in all cases. Stromal irregularities and small remnants only detectable using SS-OCT were the possible cause of detachment in 3 cases. One of the 8 eyes had air reinjection. The detachment in that eye was apparent by slit-lamp examination. The detachments spontaneously resolved in the other 7 eyes. AS SS-OCT evaluation was found to have added diagnostic value to biomicroscopy in 5 of these eyes (64.5%). CONCLUSIONS: AS SS-OCT is sensitive in detecting early graft detachment after DMEK. The 16-mm radial protocol of Triton SS-OCT is superior regarding quick scan of the entire cornea providing a general visualization of the DMEK graft. The 6-mm radial protocol provides detailed corneal imaging of a selected area, which is useful for evaluating structural alterations that are not detectable with larger scans and may affect the adherence of the endothelial graft.

"Salt and Pepper Endothelium" Recurring After Descemet Membrane Endothelial Keratoplasty.

PURPOSE: To describe the presence of "salt and pepper endothelium", that is, typical cellular inclusion bodies in a patient with Fuchs endothelial corneal dystrophy (FECD), that recurred in the donor corneal endothelial cells after Descemet membrane endothelial keratoplasty (DMEK). METHODS: A 76-year-old man underwent DMEK for FECD in his left eye. Routine specular microscopy imaging, best-corrected visual acuity measurements, and pachymetry measurements were performed before and after surgery. RESULTS: Besides large guttae indicating FECD, preoperative specular microscopy images showed variable-sized dark cellular inclusion bodies in the endothelial cells. One month after DMEK, donor endothelial cells appeared normal; however, at 3 months, the typical inclusion bodies reappeared and progressed slowly within a 4-year follow-up period. Both best-corrected visual acuity and pachymetry were stable throughout the study period. CONCLUSIONS: "Salt and pepper endothelium" recurred after the host tissue was exchanged by donor Descemet membrane, that is, a DMEK graft. These changes may indicate that either the donor corneal endothelial cell morphology is modified by adjacent tissue structures or that it is completely replaced by recipient endothelium within the first months after surgery.

Repeat Descemet membrane endothelial keratoplasty after complicated primary Descemet membrane endothelial keratoplasty.

PURPOSE: To describe the clinical outcome and complications of repeat Descemet membrane endothelial keratoplasty (re-DMEK). DESIGN: Retrospective case series study at a tertiary referral center. PARTICIPANTS: From a series of 550 consecutive DMEK surgeries with ≥ 6 months follow-up, 17 eyes underwent re-DMEK for graft detachment after initial DMEK (n = 14) and/or endothelial graft failure (n = 3). The outcomes were compared with an age-matched control group of uncomplicated primary DMEK surgeries. METHODS: The re-DMEK eyes were evaluated for best-corrected visual acuity (BCVA), densitometry, endothelial cell density (ECD), pachymetry, and intraoperative and postoperative complications. MAIN OUTCOME MEASURES: Feasibility and clinical outcome of re-DMEK. RESULTS: In all eyes, re-DMEK was uneventful. At 12 months, 12 of 14 eyes (86%) achieved a BCVA of ≥ 20/40 (≥ 0.5); 8 of 14 eyes (57%) achieved ≥ 20/25 (≥ 0.8), 3 of 14 eyes (21%) achieved ≥ 20/20 (≥ 1.0), and 1 eye (7%) achieved 20/17 (1.2); 5 eyes were fitted with a contact lens. Average donor ECD decreased from 2580 ± 173 cells/mm(2) before to 1390 ± 466 cells/mm(2) at 6 months after surgery, and pachymetry from 703 ± 126 µm to 515 ± 39 µm, respectively. No difference in densitometry could be detected between re-DMEK and control eyes (P = 0.99). Complications after re-DMEK included primary graft failure (n = 1), secondary graft failure (n = 2), graft detachment requiring rebubbling (n = 1), secondary glaucoma (n = 2), cataract (n = 1), and corneal ulcer (n = 1). One eye received tertiary DMEK. CONCLUSIONS: In the management of persistent graft detachment and graft failure after primary DMEK, re-DMEK proved a feasible procedure. Acceptable BCVA may be achieved, albeit lower than after DMEK in virgin eyes, and some cases may benefit from contact lens fitting. Complications after re-DMEK may be better anticipated than after primary DMEK because graft detachment and graft failure tended to recur, suggesting that intrinsic properties of the host eye play a role in graft adherence and graft failure.

Are polymegethism, pleomorphism, and "poor swelling" valid discard parameters in immediate postmortem evaluation of human donor corneal endothelium?

PURPOSE: To study the validity of endothelial polymegethism, pleomorphism, and "poor swelling" as tissue discard parameters in the immediate postmortem evaluation of human donor corneal endothelium. METHODS: We retrospectively evaluated the quality of the endothelium at first and second evaluations for all processed corneas exhibiting moderate polymegethism, pleomorphism, or "poor swelling" in our eye bank over a 5-year period. RESULTS: Out of 2008 eyes qualifying for our study, 422 corneas (21%) showed polymegethism, pleomorphism, or poor swelling at the first tissue evaluation immediately after excision of the corneoscleral button. In 363 (86%) of these corneas, a normal endothelial mosaic was observed at the second tissue evaluation after 7 to 21 days of organ culture, whereas only 59 (14%) still showed persistent polymegethism, pleomorphism, or "poor swelling" at that time point. CONCLUSIONS: A recovery of normal endothelial cell mosaic and "normal swelling" at the second evaluation suggests that cellular contour parameters do not relate to tissue viability, but rather to a cellular stress reaction. If so, the validity of endothelial cellular contour morphology as an early parameter in assessing the suitability of a donor cornea for transplantation may be reconsidered.

Restoration of the photoreceptor layer and improvement of visual acuity in successfully treated optic disc pit maculopathy: a long follow-up study by optical coherence tomography.

PURPOSE: To investigate by optical coherence tomography (OCT) the evolution of the photoreceptor layer and its association with best-corrected visual acuity (BCVA) in optic disc pit (ODP) maculopathy after successful surgical treatment. METHODS: Fourteen eyes of 14 patients were included in this study, and followed up from 36 to 95 months (mean 57.36 ± 18.32 months). The follow-up period started at the time of complete subretinal fluid absorption. Examination was performed by time-domain OCT before and after treatment. Spectral-domain OCT was used after treatment. Parameters assessed were type of elevation, central foveal thickness, time elapsed from onset to treatment, type of treatment, BCVA, and inner segment outer segment (IS/OS) junction line. The IS/OS junction was characterized after treatment as intact, interrupted, or absent (not distinguishable). RESULTS: Significant restoration of the IS/OS junction line was first noticed between 6 and 12 months after fluid absorption (p = 0.02; Wilcoxon signed rank test). Restoration was continuous up to the 24th month of postoperative examination after fluid absorption (p = 0.14; Wilcoxon signed rank test). BCVA was 0.99 ± 0.38 logMar before treatment, 0.81 ± 0.26 logMar (p = 0.011; paired t-test) immediately after fluid absorption and 0.61 ± 0.33 logMar (p = 0.026; one-way ANOVA) 24 months after fluid resolution. BCVA was significantly positively correlated with the integrity of the IS/OS junction line during follow-up (Pearson r = 0.775; p < 0.001). CONCLUSIONS: The IS/OS junction restoration cannot be detected immediately after fluid resolution in the majority of cases. It became evident 6-12 months later and was completed 24 months after fluid absorption. Improvement in BCVA was noticed only during the first 2 years of follow-up. No significant changes were noticed in BCVA or the IS/OS line after 2 years. Among the studied variables, the final photoreceptor layer condition and BCVA immediately after fluid absorption are the main factors predicting final BCVA after successful surgical treatment of ODP maculopathy.