Search

Purpose: To report a detailed surgical procedure of tissue engineered endothelial keratoplasty (TEEK) in a rabbit model and its postoperative evaluation. Methods: TEEKs were prepared 7 days before transplantation by seeding human or rabbit corneal endothelial cells on either femtosecond laser‐cut ultrathin human stromal lamellae (fs‐UTSL) or femtosecond laser‐cut human anterior lens capsule (fs‐HALC). Thirty transplantations were performed on aphakic eyes. Recombinant tissue plasminogen activator (rTPA) was used throughout the surgery. The native endothelium was removed by full‐surface scraping and central descemetorhexis. The transplantation was performed as a human Descemet's membrane endothelial keratoplasty. Controls included Descemetorhexis only and transplantation of carrier alone. Postoperative follow‐up was performed by slit lamp and optical coherence tomography, followed by histology. Results: Controls remained oedematous. No fibrin occurred during surgery. All but three TEEKs adhered immediately. One/6 fs‐UTSL and 9/16 fs‐HALC cleared perfectly (p = 0.161). All failures could be explained by at least one of the following causes intraoperative bleeding, vitreous prolapsus, early partial detachment, postoperative irido corneal synechiea/angle closure. Presumed immune rejection was observed in three rabbits only after 4 weeks. Immunostaining with anti‐human CD166 allowed to perfectly differentiate human cells from rabbit cells. In successful TEEK at 3 or 4 weeks, human cells formed a normal endothelium and started migrating outside the carrier. Conclusion: Though the transplantation of a TEEK in rabbits is a complex model with many causes of failure, established procedure including use of rTPA allows reliable preclinical study. In addition, we suggest that fs‐HALC might be a potential carrier for TEEK. [ABSTRACT FROM AUTHOR]

Purpose: To report the long‐term outcomes of large diameter epiretinal lyophilized amniotic membranes (lAMs) in recurrent or persistent macular holes (MHs) with or without rhegmatogenous retinal detachment (RRD), in a prospective interventional case series. Methods: Ten eyes of 10 patients underwent pars plana vitrectomy for MH‐associated RRD (n = 5) or persistent MH without RRD (n = 5), in a university Hospital. A 3 or 4 mm diameter disc of lAM, stained with 0.06% trypan blue, was inserted with a catheter through a sclerotomy and positioned over the MH. Gas or silicone‐oil tamponade was used. At 1 year, the main outcome was anatomic success defined as complete MH closure. Secondary outcomes were best corrected visual acuity (BCVA) recovery, changes in ellipsoid zone (EZ) and external limiting membrane (ELM) defects, complications. Mean follow‐up was 13.8 ± 2.9 months (range, 12–18). Results: Mean baseline data were minimum and maximum diameters, respectively, 945 ± 330 and 1507 ± 717 μm; axial length 26.58 ± 3.38 mm; and number of prior surgeries 1.4 ± 0.96. At 1 year, anatomic success was achieved in eight eyes (80%), and two had reduced diameter of MH. All RRDs were reattached without recurrence. Mean logMAR BCVA improved from 1.92 ± 0.58 to 1.17 ± 0.57 (p < 0.001), with nine eyes (90%) achieving ≥0.3 logMAR improvement. Mean EZ and ELM defects decreased (p = 0.004, p = 0.003, respectively). Postoperative complications were RRD (n = 1) reattached by subsequent surgery, lAM slightly retracted under silicone (n = 1), foveal atrophy after early lAM displacement (n = 1). Conclusion: A 1‐year follow‐up highlighted that epiretinal large discs of blue‐stained lAM can help safely close refractory MHs, and provide satisfactory visual recovery. [ABSTRACT FROM AUTHOR]

Since the 1980s, corneal storage has been carried out by eye banks throughout the world, mainly by two methods: short‐term storage at 4°C (America, India), and medium‐to‐long‐term organoculture at 31–34°C (Europe, Australia). In these two techniques, known as passive, the cornea is simply immersed in a storage medium. However, the control of corneal hydration in vivo depends both on the proper functioning of the corneal endothelial cells (responsible for the imbibition pressure IP) and on the presence of a normal intraocular pressure (IOP), both of which are opposed to the stromal swelling pressure (SP): IP = SP‐IOP. By restoring IOP ex vivo in an active storage machine (ASM) that separates an endothelial chamber from an epithelial chamber, corneal hydration is better controlled, thus preventing the formation of deep endothelial folds responsible for endothelial over mortality. The active control of corneal hydration eliminates the need to add macromolecules to deswell the cornea (continuously for cold storage or during the last 48–72 h for organoculture). The toxicity of this forced deswelling has been demonstrated for organ culture. We performed two successive preclinical studies comparing ASM and organoculture with paired human corneas having more than 2000 cells/mm2 at the time of the first endothelial count. The organoculture medium (CorneaMax, Eurobio, les Ulis, France) was identical in both groups. The ASM was an experimental version from our laboratory, including a PEEK (polyetheretherketone) corneal support, a peristaltic pump, an Edwards pressure sensor, a microsolenoid valve and a microcontroller managing the flow rate (2.6 μL/min) and the IOP (21 mmHg). In the first study, for 52 pairs of corneas stored for 4 weeks, the ASM corneas had, at the end of storage, +23% viable endothelial cells compared to passive organoculture. Furthermore, ASM also significantly increased Na+/K+/ATPase protein expression by 3.8 ± 1.4. In the second study, for 12 pairs of corneas stored for 3 months, the ASM corneas had +53% viable endothelial cells at the end of storage compared to passive organoculture. One third of corneas still had 2000 cells/mm2 or more in ASM versus none in organoculture. The ASM significantly increased Na+/K+/ATPase protein expression by 2.9 ± 0.8. The ASM has 2 transparent windows which allow all quality controls to be performed without deconditioning the cornea, and in particular to perform endothelial counting by specular microscopy on a cornea which remains thin. The ASM thus combines the advantages of both methods of storage: long‐term storage without massive oedema and without toxic dextran and a closed system allowing non‐invasive controls. As with the perfusion machines for vascularized organs, the restoration of near‐physiological conditions improves grafts' quality. We have transferred this innovation to THEA laboratories (Clermont‐Ferrand, world specialist in ophthalmology) which is industrializing the ASM via its spin‐off SINCLER. The industrial version will take the form of a single‐use cassette incorporating a flow/pressure regulator, a media reservoir and a waste bin. The cassette will be plugged into a computer‐controlled multi‐place incubator for IOP and media flow management. The ASM could revolutionize eye banking worldwide. [ABSTRACT FROM AUTHOR]

Aims/Purpose: Viability markers are essential for cellular research and tissue transplantation. However, existing options such as Trypan Blue lack specificity, and some clinically used markers are cytotoxic. Therefore, we synthesized the compound F4267, a non‐fluorescent marker that releases fluorescein upon hydrolysis. The aim of this study was to evaluate the quality of F4267 labelling of endothelial cells and its cytotoxicity in immortalized cells and organ‐cultured endothelium. Methods: We compared the marking capability of F4267 and Calcein‐AM using immortalized cells and endothelium from corneal pairs. The quality of fluorescent labelling was assessed using F4267 at 4 μM and Calcein‐AM at 4 μM. Cytotoxicity of F4267 was evaluated at varying concentrations of 40 and 1000 μM on cells, and at 40 μM on corneal endothelium, with single, repeated, or prolonged exposures. The final density and viability of corneal endothelium were measured using HEC1 staining. Results: F4267 (40 μM) exhibited fluorescent labelling quality equivalent to Calcein‐AM (4 μM), allowing clear differentiation between living, dying, and acellular areas. In HCE‐2 cells, cytotoxicity was reduced by 19% with F4267 (40 μM) compared to Calcein‐AM (4 μM). Prolonged exposure (24 h) to HCEC‐B4G12 cells showed no cytotoxicity with F4267 (40 and 100 μM), but resulted in destruction of the cell layer with Calcein‐AM (4 μM). F4267 (40 μM) also demonstrated significantly lower cytotoxicity, reducing mortality by 10% (single exposure) and 17% (repeated exposure) in organ‐cultured corneas. Conclusions: F4267 shows promise as a marker for assessing viable endothelial cells in organ culture, providing a safe alternative to existing markers. Its release of fluorescein with excellent cell tolerance supports its clinical potential. The integration of viable endothelial cell density assessment would enhance the value of corneal banks. Further validation is required, as well as exploration of broader applications. Reference 1. Pipparelli et al., IOVS 2011. [ABSTRACT FROM AUTHOR]

Aims/Purpose: Retroillumination observation of the anterior segment is a routine examination method as old as the slit lamp, but capturing good quality images is difficult because of patient glare, the presence of corneal and or lens reflections, and limited resolution. We have modified a slit lamp to overcome these limitations and present the first series of images obtained. Methods: On a Topcon SL‐D7 slit lamp, we replaced the light source with a far‐red LED (Thorlabs, M780L3, 200 mW, 800 mA at 780 nm) and added a monochrome camera with sufficient resolution (Baumer, VCXU‐123 M). Reflection suppression was achieved by modifying the optical path (patent pending). In a clinical trial validated by an ethics committee (IDRCB: 2021‐A01496‐35), we acquired images of 5 different diseases after pupillary dilation: cataract, posterior capsule opacification, hereditary stromal dystrophy, Fuchs' corneal endothelial dystrophy (FECD), Bowman's membrane dystrophy. Results: Our innovative prototype device made it possible to obtain images of 3000 × 4096 pixels for 13.3 × 9.7 mm (×40 magnification). In 100% of the cases, one or more clear images on the area of interest were obtained without glare, and without disturbing reflections in 90% of the cases. Some pseudophakic patients showed minimal to moderate light reflection. The images were sufficiently resolved to allow precise analysis of disease areas, such as each guttae of the FECD. Conclusions: Our retroilluminator prototype provides, for the first time to our knowledge, high‐resolution images of the different structures of the anterior segment, easy to acquire and likely to facilitate the diagnosis and follow‐up of patients. Improvements to the prototype are underway to remove some persistent reflections. One of the first applications will be the classification of the severity of endothelial damage in FECD. [ABSTRACT FROM AUTHOR]

Purpose: The lens has a monolayer of epithelial cells on its anterior surface that contributes to its homeostasis. A quantification of the viability of these cells is essential to evaluate the safety of new therapies targeting the lens (cataract, presbyopia). The observation of this whole tissue is made difficult by its optical properties (transparency and refraction) and only simplified models have been used until now to measure epithelial viability (primary cell cultures or partial surgical specimens). Aims: To develop a method to quantify the viability of epithelial cells on whole ex vivo lenses. Methods: Ten fresh pairs of 6‐month‐old porcine lens were retrieved from local slaughterhouse. For each pair, one lens was randomly selected to undergo a localized lesion and the other lens remained intact. The lesion was made on the anterior surface with a steel hexagonal rod (to make the mark easily identifiable) previously immersed in liquid nitrogen for one minute. Both lenses were incubated for 1 h at 20°C in a HEC mixture. Images were acquired with a macroscope and analysed with ImageJ. Calcein‐AM and Ethidium images were used to calculate the area covered by living epithelial cells. Hoescht images allowed to count cell nuclei per unit area. Viable epithelial cell density (vECD) was defined as the number of viable cells per unit area. The lenses were immersed (isotonic saline) inside a black container during the measurements to minimize reflections and refractive phenomena or drying of the samples. Results: The vECD median on the ten pairs was 2840 cells/mm2 [10th‐90th percentiles = 2479–3494] for cryo‐applied lenses versus 3364 cells/mm2 [2919–3739] for healthy lenses (p = 0.002). Conclusions: The triple HEC staining adapted to the whole lens provides an unrivalled measure of pan‐epithelial viability without the need for a tedious capsule dissection step that destroys the lens architecture. Other studies will be necessary to determine its ability to discriminate small variations or type of mortality. [ABSTRACT FROM AUTHOR]

Background: In the clinical practice, transparent films are used as sterile interfaces in in vivo dermatologic imaging in order to prevent the transmissions of infections. However, in our experience, the use of a transparent film can alter skin images. Our study aimed to compare the optical quality of a series of different plastic films used as interfaces in order to understand if some might be more suitable for imaging. Materials and methods: We tested the optical properties of 11 different protective transparent films that are marketed in France with a transparency meter and a spectrophotometer. Results: Transmission, minimal diffusion, amount of gray, and contrast were obtained for each transparent film. Transmission ranged from 93.24% to 96.88% (mean 95.36; standard deviation SD 1.02), minimal diffusion from 88.28% to 123.87% (mean 101.04; standard deviation SD 10.02) and contrast from 11.01 to 15.88 (mean 13.93 and SD 1.3). For some films, the transmission was lower at lower wavelengths. Conclusion: All tested films had excellent optical properties. However, some of them had better optical qualities and seemed more suitable for their use in dermatologic imaging. [ABSTRACT FROM AUTHOR]

Abstract: Amoebae pose a potential health risk as pathogenic agents i. e. in the central nervous system or the cornea, early diagnostics are important to implement rapid treatments and reduce impact on human health. Thus, development of selective fluorescent tags for Amoebae is an important issue. Analogues of Miltefosine, a drug effective against various strains of free‐living amoebae, have been synthesized. They are derivatives of the 11‐aminoundecylphosphocholine. Via a peptide bond, a strong emitting BODIPY was embedded to the structure. This compound (1) was used to selectively stain for Acanthamoeba castellanii. In order to test the effect of the insertion of the amide function in the alkylbackbone of this modified Miltefosine, a model molecule (2) was prepared by replacing the dye by the anisidyl moiety. Anti‐amoebic activity was evaluated for both. [ABSTRACT FROM AUTHOR]

Abstract: Purpose: To describe and validate the micro‐instillation of fluorescein on the ocular surface by a disposable calibrated inoculation loop to improve corneal and conjunctival staining quality. Methods: Accuracy and precision of the volume of 0.5% sodium fluorescein collected by a single use 1 μl‐calibrated inoculation loop were measured using a precision balance. Twenty patients (40 eyes) suffering from dry eye syndrome were enrolled in a prospective interventional nonrandomized study. Fluorescein was instilled with the loop, and slit‐lamp images were taken within 30 seconds using cobalt blue light with and without a yellow barrier filter. For comparison, after a washout period, the same images were retaken after instillation of one drop of fluorescein from a single‐dose unit. The main outcome measure was the staining quality assessed by three experts, blind to the instillation method. Patient discomfort (tolerance, by a questionnaire) was also compared. Results: The mean volume collected by the loop was 1.18 ± 0.12 μl, compared with 33.70 ± 6.10 μl using the single‐dose unit. The loop avoided excess dye responsible for unpleasant tearing, masking of lesions and rapid diffusion into the stroma. Micro‐instillation greatly improved image quality without losing information. The yellow filter further improved image contrast. Tolerance was excellent. Conclusion: The 1 μl‐calibrated inoculation loop is a safe, convenient, inexpensive, disposable, sterile, well‐tolerated tool for reproducible micro‐instillation of commercial fluorescein. By greatly improving staining quality, it will help standardize assessment of dry eye severity. [ABSTRACT FROM AUTHOR]

Background Ex vivo confocal microscopy is a recent imaging technique for the perioperative control of skin tumour margins. Up to date, it has been used in the fluorescence mode and with vertical sections of the specimen margins. The aim of this study was to evaluate its use in the reflectance mode and with a horizontal ('en face') scanning of the surgical specimen in a series of basal cell carcinoma of the eyelid. Design Prospective consecutive cohort study was performed at the University Hospital of Saint-Etienne, France. Participants Forty-one patients with 42 basal cell carcinoma of the eyelid participated in this study. Methods Basal cell carcinomas were excised with a 2-mm-wide clinically safe margin. The surgical specimens were analysed under ex vivo confocal microscopy in the reflectance mode and with an en face scanning in order to control at a microscopic level if the margins were free from tumour invasion. Histopathogical examination was later performed in order to compare the results. Main Outcome Measures Sensitivity and specificity of ex vivo confocal microscopy for the presence of tumour-free margins. Results Ex vivo confocal microscopy results were consistent with histopathology in all cases (tumour-free margins in 40 out of 42 samples; sensitivity and specificity of 100%). Conclusions Ex vivo confocal microscopy in the reflectance mode with an 'en face' scanning can control tumour margins of eyelid basal cell carcinomas and optimize their surgical management. This procedure has the advantage on the fluorescent mode of not needing any contrast agent to examine the samples. [ABSTRACT FROM AUTHOR]

Corneal endothelial diseases are common almost everywhere in the world and have 3 main origins: primary diseases, largely dominated by Fuchs' endothelial corneal dystrophy (FECD), iatrogenic diseases which mainly complicate lens surgery, whether cataract or clear lens, but also other surgeries. The frequency of these complications is very low but the number of these surgeries being gigantic (probably around 50 million cases/year/worlwide), the number of post‐surgical bullous keratopathy remains important. Post‐keratoplasty endothelial failure (early or late) is the 3rd cause. All these diseases "consume" 50% of corneal grafts in the world and a large part of the population does not have access to any graft. The number of corneal donors is absolutely insufficient to meet the world demand and corneal bioengineering appears to be a realistic solution to complete the offer. Because of its relative simplicity and the demonstrated effectiveness of endothelial keratoplasty, endothelial bioengineering is logically the first to be developed. The Japanese team of 2 Kyoto universities (Prof. S. Kinoshita, N. Koizumi and N. Okumura) is a pioneer in this field and has demonstrated the effectiveness of injection therapy, which consists of cultivating mature endothelial cells from a small number of corneas from young donors, then injecting them into the anterior chamber of patients. The industrialization of the process is underway in 2 different start‐ups, one in the United States, the other in Japan. In addition to injection therapy, the bioengineering of endothelial grafts (Tissue engineered endothelial keratoplasty, TEEK), which consists of cultivating endothelial cells on a "corneo‐compatible" carrier to exactly reproduce a DMEK or a DSAEK, has passed the preclinical stage. This presentation will detail these different processes, their current limitations and the short, medium and long term perspectives. These Advanced Therapy Medicinal products (ATMP) are clearly revolutionizing corneal transplantation. By offering a new treatment for endothelial diseases, they will make it possible to redirect donor corneas to other corneal diseases that have no therapeutic alternative. Finally, this paper will also discuss the prospects of medical and medico‐surgical treatments for FECD, which are at the frontier of endothelial bioengineering. [ABSTRACT FROM AUTHOR]

Purpose: To compare the capacity of anterior segment optical coherence tomography (AS‐OCT) and Scheimpflug camera in detecting subclinical corneal edema in patients with Fuchs endothelial corneal dystrophy (FECD). Methods: 177 eyes of 103 patients with FECD without obvious corneal edema were included and analysed in this single‐center prospective study (NCT03974230). Tomographic analyses were performed successively using Scheimpflug imaging (Pentacam HR, Oculus) and a swept‐source AS‐OCT (Casia SS‐1000 Tomey). All measurements were performed more than 4 hours after awaking to have a stabilized corneal thickness. The resulting images were analysed by two experienced independent observers. AS‐OCT and Scheimpflug images were analysed independently. The main outcome was the agreement between the ability of the two devices to detect subclinical corneal edema, assessed using inter‐device reliability measured by the Cohen kappa statistic. The criteria for subclinical corneal edema were as follows: the presence of irregular isopachs, displacement of the thinnest point of the cornea, and the presence of posterior surface depression. Results: The agreement between the two devices to detect subclinical corneal edema was high. The inter‐device reliability for loss of parallel isopachs, as measured by the Cohen kappa coefficient, was 0.75; for the displacement of the thinnest point of the cornea, it was 0.65; and for the focal posterior corneal surface depression, it was 0.53. The inter‐device reliability for the detection of subclinical corneal edema was 0.70. Subclinical corneal edema was detected in 77% of cases analysed with both devices. Conclusions: Subclinical corneal edema in patients with FECD can be detected by AS‐OCT as well as by Scheimpflug imaging. Both devices can facilitate surgical decision‐making when evaluating the risk of corneal edema after cataract surgery in eyes with FECD. [ABSTRACT FROM AUTHOR]

Purpose: The corneas preserved in bioreactor (BR) had been shown to have not only a better endothelial viability, but also a more differentiated and stratified epithelium than corneas preserved in organoculture. Purpose: By using BR, we would analyse the respective contribution of corneal (C), limbal (L), and conjunctival (Conj) epithelia in corneal epithelial regeneration. Methods: Five pairs of corneas from body donation to Science were used with a death‐to‐collection time < 20 h. A 3‐ to 5‐mm‐wide conjunctival flange was kept intact. Five patterns were set up by fully mechanical removal of 1, 2, or 3 epithelia: C‐L + Conj+, C‐L‐Conj+, C‐L + Conj‐, C + L‐Conj‐, C‐L‐Conj‐ (control) n = 2 for each pattern. The limbus was destroyed by scraping and thermocoagulation. Corneas were then kept in BR (21 mmHg, 2.5 μl/min of Corneamax Eurobio, 31°C) for 3 weeks to allow epithelial regeneration. The epithelium was then analysed using immunofluorescence (IF) on flat mounted cornea by targeting CK12 (corneal epithelium) and CK15 (limbal epithelium). Cell nuclei were counterstained with DAPI. Corneal transparency was quantified using a transparometer. Results: No epithelium was reconstituted in the controls. In the other 4 models including the C‐L‐Conj+ group, the cornea was transparent and covered by a pluristratified corneal epithelium, characterized by CK12 expression. Conclusions: In this BR model, conjunctival epithelial cells allowed the regeneration of a typical corneal epithelium in model C‐L‐Conj+. The corneal epithelium can also migrate to the limbus and conjunctiva. We hypothesize that all 3 ocular surface epithelia (including the conj) contain stem cells or progenitors capable of migrating throughout the cornea and restoring the corneal epithelium independently of each other. The main difference between our ex vivo model and in vivo situation is absence of neovascularization. This suggests that the main cause of limbic insufficiency is due to the loss of the anti‐angiogenic barrier rather than the loss of limbic stem cells. [ABSTRACT FROM AUTHOR]

Purpose: Fuchs corneal endothelial dystrophy (FECD) causes chronic corneal edema and vision loss, and accounts for more than 50% of keratoplasty indications in occidental countries. The Descemet membrane (DM) gradually thickens because of extracellular matrix deposition and round excrescences named guttae accumulate. Chromatic confocal microscopy (CCM) is on optical device that analyses the nanotopography of surfaces. Aim: to study the feasibility of CCM technology to characterize guttae in FECD. Methods: Descemetorhexis of approximately 8 mm in diameter were obtained during endothelial grafts in 11 patients with FECD and control corneas were obtained from 3 donor corneas. The DM were fixed in 0.5% paraformaldehyde immediately after dissection. The CCM analysis was performed at room temperature on surfaces of 400 x 400 μm to characterize the DM with and without guttae. We measured positive and negative heights (peaks and valleys) relative to the mean altitude, as well as their distribution by asymmetry coefficient. Results: CCM resolution allowed to map nanotopography of the DM of FECD patients and controls. The surface of DM control was flat and homogeneous with maximum positive and negative heights of 1.07 ± 0.09 μm and 1.05 ± 0.07 μm, respectively. The height distribution was uniform with an asymmetry coefficient of 0.002 ± 0.23 μm, reflecting a Gaussian distribution (npeaks = nvalleys). The surface of FECD membranes was completely altered with a maximum peak of 6.05 ± 1.49 μm, up to 5.7 times higher than the control. The height distribution was disturbed with asymmetry value of 1.61 ± 0.79 μm (npeaks > nvalleys) and a maximum valley height (3.09 ± 0.79 μm) that was half that of the peak one. Conclusions: CCM is a promising technology to quantitatively characterize the 3D organization of the DM surface. CCM can be used to classify the different types of surface lesions of the FECD corneas and to follow evolution of DM alteration at different stages of the disease. [ABSTRACT FROM AUTHOR]

Purpose: Fuchs Endothelial Corneal Dystrophy (FECD) is characterized by the progressive slow accumulation of extra cellular matrix (ECM) forming Descemetic excrescents named guttae and by an increase in the thickness of the Descemet membrane (DM). Nevertheless, FECD pathophysiology remains only partially understood. Aim: To present the expression of newly identified proteins, using proteomics methods. Methods: DM were obtained from FECD patients undergoing DMEK and from donor corneas (controls) (University Hospital St‐Etienne). Some samples were analysed using MALDI imaging and shotgun proteomics (Kyoto, Doshisha University) to identified proteins selectively under or over expressed in FECD vs controls. Other samples were used for immunostaining (IS) of the top over‐expressed proteins. We used IS on flat mounts of DM in order to study the differences between the center and the periphery and cross sections to study the DM layer affected. Cross sections were performed using a cryostat on DM embedded in a mixture of gelatin, optimal cutting temperature compound and sucrose. Expression of each protein was observed using a fluorescence microscope. At least 3 different samples were used for each protein. Results: MALDI imaging and shotgun analysis identified 11 potential targets overexpressed in FECD, belonging to different families: mainly collagen, proteogylcans, and TGF beta signalling. Using IS, none of the 11 proteins was expressed in healthy samples, except for fibronectin. In FECD samples we identified proteins selectively expressed in different structures of the diseased DM: embryonic layers, Guttae, posterior fibrous layers. Notable differences were also observed between the periphery and the center. Conclusions: These new potential FECD markers located in different ECM abnormal structures suggest a dynamics phenomenon where different proteins are successively expressed and accumulated with time. [ABSTRACT FROM AUTHOR]

Purpose: Management and follow‐up of secondary unlikely location of Dexamethasone Implant: between sulcus 3‐pieces IOL and intact posterior capsule. Methods: A 66‐year‐old man had a traumatism while cutting wood: he consulted 3 days later for severe endophthalmitis. A CT scan excluded any metallic intraocular foreign body (FB). Initial emergency surgery consisted in phacovitrectomy without implantation, bacteriologic and fungal samples, removal of small wood FB, antibiotic & antifungal intravitreal injections. No zonulolysis was present, no capsular tear occurred. Local and systemic treatment allowed good anatomic recovery. At 1.5‐month he developed Irvine‐Gass syndrome (CME). At 2‐months, conditions allowed secondary implantation with a 3‐pieces IOL in the sulcus due to anterior and posterior capsules fusion. As zonule and lens bag were well preserved, we injected intravitreal DI without adverse event (free in the posterior segment). Forty days after DI injection, he consulted in emergency for a hay straw in his left eye. Actually, the DI migrated to the anterior segment, just between sulcus IOL and intact posterior capsule. He had no pain, no redness, normal IOP, 0.2 logMAR far & near vision, and an impression of a paracentral scotoma. He reported repeated heavy loads lifting with head down in his farm, despite postoperative instructions. Medical treatment based on myotics (isotpopilocarpine), continuation of anti‐inflammatory drops, and prophylactic hypotonizing monotherapy drops associated with monthly complete monitoring (slit lamp/OCT/specular microscopy) was decided in first line‐therapy. Results: Spontaneous disintegration of this ectopic DI took 2 months: CME disappeared completely in 1.5 month; IOL remained stable; no IOP elevation or endothelial damage occurred; vision went to 0.0 logMAR and paracentral scotoma disappeared. Conclusions: This ectopic migration leading to DI sandwich enables good healing from a CME. Various options could have been chosen: among YAG capsulotomy, emergency surgery and medical treatment. Here, rigorous monitoring with precise iconographic sequences allows to estimate efficacy, safety and time of resorption in this unlikely location. DI injection should be really discussed for every trauma background. [ABSTRACT FROM AUTHOR]

Purpose: Quality of the endothelial transplant is a critical parameter in the success of Descemet Membrane Endothelial Keratoplasty (DMEK) and in the graft survival. The peeling techniques, preservation methods and operator's skill level need to be experimentally assessed and validated, as they are key elements influencing graft quality. The most reliable method of quality evaluation of the endothelial transplant is the triple staining with Hoechst‐Calcein AM‐Ethidium (HEC) allowing to determine the total number of viable endothelial cells. However, the test HEC has defects: (1) Unspecific fluorescence of the same filter than that Calcein AM prevents an accurate viability analysis; (2) Incompatibility with immunofluorescence (IF) which could provide additional information. The aim of this study is to develop technical tips to overcome these defects. Methods: Two strategies were employed to improve Calcein AM staining: 1. Increase the specific fluorescence intensity by changing the concentration of Calcein AM and diluent; 2. Decrease unspecific fluorescence by adding the fluorescence quencher Trypan Blue (TB). In order to combine IF after the test HEC, an extensive wash in PBS was performed. In total, 19 human corneas had been used. Results: Calcein AM at 4µM diluted in OptiMEM increased fluorescence intensity by 3‐fold (p = 0.0017, n = 5) compared to conventional staining at 2 µM in PBS. Trypan Blue visibly decreased the unspecific fluorescence of Calcein, reduced the inter‐operator (n = 6) variability of cell count by 42% (p = 0.0027, n = 10) and reduce analysis time by 40% (p = 0.0002, n = 10). To perform IF after HEC test, prolonged washing in PBS was an effective method to remove residual Calcein fluorescence and allow imaging using FITC/Alexa 488 filter. Conclusions: This study provides effective technical tips for optimising assessment of the endothelial viability and for being able to perform IF after the test HEC. [ABSTRACT FROM AUTHOR]

Purpose: The corneal bioreactor (BR) patented by our laboratory allows to restore a physiological epithelium of porcine or human corneas. The measurement of the transepithelial electrical resistance (TEER), correlated with the presence of tight junctions, is a robust parameter for assessing the integrity of the corneal epithelium. Aim: To integrate the measurement of TEER with BR and to evaluate its relevance on porcine corneas. Methods: Two Ag/AgCl electrodes (WPI, Saratosa, FL) were placed, one in an endothelial chamber filled with Corneamax storage medium (Eurobio, Les Ulis, France), the other in an epithelial chamber and under 3 conditions: immersed in the conservation medium; immersed in the conservation medium while isolating the cornea from the sclera thanks to an applicator cone usually used for iontophoresis (Opia Technologies, Paris, France); in direct contact with the epithelium with interposition of Gel‐larmes (Théa, Clermont‐Ferrand, France). The electric current frequencies tested ranged from 50 Hz to 1 MHz. Three fresh pig corneas (with and without epithelium) were placed in BR and the TEER measured in the 3 conditions, with and without epithelium. Controls: Fresh pig sclera and Corneamax alone in BR. Results: Results are expressed as median. TEER of controls: 302 Ω (sclera) and 266 Ω (Corneamax alone) TEER of corneas without epithelium: 321 Ω, in the 3 experimental conditions, thus close to the control. TEER of corneas with epithelium: 397 Ω with electrode immersed in the medium, thus close to the sclera (397 Ω). In the other 2 conditions, the TEER was higher, with a greater variability, depending on the current frequency: from 1456 Ω to 143 Ω with the applicator cone and from 2335 Ω to 242 Ω with the Gel‐Larmes. Conclusions: The sclera placed in BR is a conductive tissue. It is possible to determine the TEER of a cornea ex vivo in the BR, if the cornea is isolated from the sclera and the epithelium present on the cornea is healthy. [ABSTRACT FROM AUTHOR]

Purpose: The corneal bioreactor (BR) in our laboratory, with an intraocular pressure (IOP) of 21 mmHg and the renewal of the culture medium, allows to maintain the homeostasis of the 3 layers of the cornea. We showed that an IOP reduced to 10 mmHg allowed to simulate a corneal edema (760 ± 57 µm) with a healthy epithelium. Aim: To compare 2 hyperosmolar eye drops in the evolution of corneal deturgescence and epithelial ulcer healing: ODM5 composed mainly of NaCl 5% and hyaluronic acid (Horus Pharma, Saint‐Laurent‐du‐Var, France) and Muro 128 composed mainly of NaCl 5% and various excipients (Bausch+Lomb, Bridgewater, NJ) Methods: Beforehand, 3 pairs of human corneas were preserved for 2 weeks in the BR filled with CorneaMax (Eurobio, France), with a pressure maintained at 10 mmHg. A central epithelial ulcer of 6.64 ± 0.04 mm in diameter was made (D0). Each cornea received a drop of Muro 128 or ODM5 four times a day. Corneal central volume measurement (Casia 1, Tomey) was performed daily at the beginning and end of the day. Ulcer healing was monitored daily by fluorescein labelling and quantified by image analysis (Fiji) taken with a macroscope. At D14, the corneas were removed and fixed for immunostaining of E‐Cad (cell architecture), Zo‐1 (tight junctions), MUC16 (mucins), K3K12 (cell maturity) and for histology (Saffron Eosin Hemathein). Results: Mean daily volume loss was comparable: 2.43 ± 1.55 mm3 with ODM5 vs. 2.18 ± 1.71 mm3 with Muro 128 (p = 0.698 3). Corneal ulcers in all 3 pairs healed at D4, D5, D5 with ODM5 vs. D5, D13, and 1 cornea never healed with Muro 128. In IHC and histology, the epithelium was of better quality with ODM5. Conclusions: The efficacy of the 2 eye drops is comparable. Epithelial preservation is partly better with ODM5, possibly related to the presence of hyaluronic acid. [ABSTRACT FROM AUTHOR]

Purpose: The aim of this study is to report the use of Vicryl 10‐0 (polyglactin 910) resorbable monofilament in corneal trauma during COVID‐19 pandemic: structural and functional results, and follow‐up adaptation during health crisis. Methods: During the COVID‐19 pandemic, nine patients were selected from the overall population of open or closed globe surgical emergencies (n = 37) in a french University Hospital. Selected patients were eligible for corneal sutures with Vicryl 10‐0, thus replacing the traditional 10‐0 nonabsorbable monofilament nylon suture. Postoperative treatment combined anti‐inflammatory drugs, topical antibiotics and lubricating drops. Outpatient visits were performed at D10, M2, M6 then every six month. Interim visits were performed if needed between D10 and M2 by teleconsultation. Patients received complete examination including different imaging. The main outcome was best corrected visual acuity. Secondary outcomes included average corneal astigmatism and the occurrence of complications. Results: Mean distance and near acuity improved from 0.711 ± 0.805 & 0.556 ± 0.305 logMAR preoperatively to 0.056 ± 0.073 & 0.333 ± 0.100 logMAR postoperatively (p = 0.032 and p = 0.043 respectively). Mean astigmatism decreased from 4.91 ± 4.21 diopters preoperatively to 0.99 ± 0.57 after suture resorption (p = 0.020). There were no serious adverse events or repeated surgery. Conclusions: Polyglactin 910 10‐0 could provide a maximum infectious risk reduction; a shorter topical treatment and antibiotic prophylaxis. The absorbable nature of the sutures simplifies logistics by reducing the number of early outpatient visits before M2. It avoids suture removal, especially for the pediatric population. Interim teleconsultation(s) on demand are interesting to evaluate the need for an additional face‐to‐face consultation. Vicryl 10‐0 sutures are ideally suited to the COVID‐19 pandemic logistical constraints, as a safe and effective alternative to nylon 10‐0 for the management of eligible corneal wounds. [ABSTRACT FROM AUTHOR]

. Purpose: Study the clinical and microbiological characteristics and the prognostic factors of post-traumatic endophthalmitis. Methods: Seventeen eyes were included between 2004 and 2010, with clinical and microbiological data collected prospectively. Conventional cultures and panbacterial PCR were performed on aqueous and vitreous samples. Results: Clinical signs of endophthalmitis were observed soon after trauma (1.5 ± 2.5 days). Laceration with an intraocular foreign body (IOFB) was noted in 53% of the patients. At admission, all patients had aqueous humour (71%) and/or vitreous (53%) samples. Fifteen patients (88%) underwent a pars plana vitrectomy. Bacteria were identified in 77% of the cases: Staphylococcus epidermidis ( n = 5), Streptococcus ( n = 4), Bacillus ( n = 2), Pseudomonas stuzeri ( n = 1), and Streptococcus salivarius and Gemella haemolysans (multibacterial infection, n = 1). Progression toward phthisis was observed in 35% of the cases; 41% of the patients recuperated visual acuity (VA) ≥20/40. A good final visual prognosis (≥20/40) was significantly associated with initial VA better than light perception (0% versus 70%, p = 0.01) and absence of pupillary fibrin membrane (80% versus 20%, p = 0.05). There was no correlation between visual prognosis and age, the type of laceration (corneal or scleral) or presence of an IOFB. We found a statistical trend toward an association between bacterial virulence and poor final VA. Conclusion: This series showed that better final VA outcomes were associated with initial VA better than light perception, S. epidermidis or culture-negative cases and absence of retinal detachment during the clinical course. [ABSTRACT FROM AUTHOR]

The article presents a study that determines the characteristics of yellow globules in balloon cell naevus. It cites the case of an eight-year-old boy presented with yellowish-brown congenital papule on her right cheek and his histological examination revealed a dermal melanocytic naevus with confluent nests and sheets of large clear cells. It notes the result which shows that yellow or white globules associated with a brown pigmentation is an indication of balloon cell naevus.

Purpose: Penetrating keratoplasty (PK) remains the first technic of corneal transplantation worldwide and rejection a common cause of graft failure. Its prevention has been based for decades on repeated instillation of steroid eye drops (dexamethasone, prednisolone, rimexolone). Nevertheless, the overall survival of corneal transplants, all indications combined, is lower than that of some organ transplants. Both compliance and the discontinuous nature of the eye drops may be involved. The improvement of immunosuppression could avoid numerous graft failures. In a rabbit model of PK, we previously demonstrated that a subconjunctival dexamethasone implant (Ozurdex) was well tolerated and prevented rejection as efficiently as steroid eye drops trice daily during the first 5‐6 weeks. We then designed a clinical trial to first investigate the tolerance and safety of subconjunctival Ozurdex after PK. Methods: Single centre, phase II non‐randomized tolerance and safety pilot study on 14 patients (NCT02834260). Designed to analyze the risk of increased intraocular pressure (IOP, primary endpoint at month 1), possible discomfort, resorption time, and collect initial efficacy data. The implant was injected at the 12 o'clock position, 5 mm from the limbus, at the end of PK, in order to deliver dexamethasone immediately. A steroid eye drop relay was planned at the time of complete resorption of the implant. Patients were monitored at regular intervals during 12 months: IOP, discomfort and redness scores, implant status, rejection episode, endothelial cell density, and central corneal thickness by OCT. An independent adverse event committee gave its opinion after the first patient, then at regular intervals. Results: No increase in IOP, no adverse event related to the implant was observed. The average duration of resorption was 6 weeks. The therapeutic relay with dexamethasone eye drops was uneventful. Conclusions: His is the first proof of concept that dropless immunosuppression is possible after penetrating keratoplasty. Ozurdex, chosen here because it was the only ophthalmological steroid implant marketed in France, obviously resorbs too quickly, but this work paves the way for the use of other implants that release a steroid or another immunosuppressive agent in the very long term. [ABSTRACT FROM AUTHOR]

Purpose: In France, the adoption by an eyebank of a new technique for graft preparation and/or a new CE‐marked storage medium requires the submission of a "process" file to the competent health authority. This file must include pre‐clinical data on corneas stored in this medium or prepared with the new technique. Aim: To analyze the endothelial alterations induced by the pre‐cutting with the microkeratome of corneas stored in Tissue‐C organoculture medium and deswelled in Carry‐C medium. Methods: Ten pairs of corneas (donation for scientific purposes) were retrieved by in situ corneo‐scleral cutting and immersed in 100mL of Tissue‐C (Alchemia, Italy) and stored at 31°C in a dry incubator for 30 days (D). At the beginning (D2 to D5) and end (D30) of storage, the usual quality control was performed: central endothelial cell density (ECD) after incubation with 0.9% NaCl, transparency, thickness (OCT) and microbiological control. At D30, the corneas were transferred to a deswelling Dextran‐containing medium (Carry‐C, Alchemia) for 24 H. After randomization, one cornea of each pair was pre‐cut with the MORIA linear microkeratome (DSAEK, single pass, objective #150 µm) and put back in Carry‐C for 72H. The endothelial viability (viable ECD) of the all corneas was then measured by triple Hoechst‐Ethidium‐Calcein‐AM staining as we previously described (IOVS2011, Cornea2014). Results: The initial ECD was 2401 ± 815 for the DSAEK group and 2298 ± 805 cells/mm2 for the control group (p = 0.386) and at D30, respectively, from 1945 ± 587 versus 1868 ± 577 cells/mm2 (p = 0.421). One cornea was perforated. Graft thickness was 187 ± 112 um. For the other 9 pairs, the viable ECD (by definition always lower than bank ECD) was 1061 ± 448 for the DSAEKs and 1149 ± 409 cells/mm2 for the controls (p = 0.066). The pre‐cut grafts had fewer endothelial folds. Conclusions: Pre‐cutting by bank of corneas stored in Tissue‐C/Carry‐C does not induce significant additional cell loss compared to control. [ABSTRACT FROM AUTHOR]

Purpose: The permanent regeneration of the human corneal epithelium involves the proliferation of basal cells of corneal epithelium for compensating for daily desquamation (Majo et al. Nature 2008) and of limbal stem cells (LSCs, Cotsarelis, Cell 1989) located in limbal crypts that can be stimulated in case of important epithelial lesion. Many markers of LSCs had been described in the literature but some are controversial, particularly for their specificity. Aim: To evaluate the specificity of the main LSC markers proposed in the literature and map their expression on the surface of the cornea. Methods: Referenced or potential markers of LCSs were tested on cross‐sections of fresh human corneas (body donation to Sciences, time between death‐retrieval‐fixation <21 hr) using a standard technique of immunohistochemistry: ABCB5, ΔNp63, P63, ABCG2, Pax6, CK15, CK17, N‐cadherin, Connexin 43, EGFR, vimentin, Sox2, OCT 4. Secondly, antibodies that showed positivity on cross sections were used for immunolabelling flat‐mounted whole cornea to map their expression. Results: None of 13 markers was exclusively located at the basal layer of the limbal epithelium. Nevertheless, CK15 appears to be the most promising marker because of its very strong staining in the limbal basal cell layer. However, its staining was also found, less intense, in basal layer of conjunctival epithelium and medium layers of limbal epithelium. On flat‐mounted corneas, CK15 was expressed in a circular ring corresponding to the anatomical limbus, but CK15+ cells were not restricted within the Vogt palisades (or limbal crypts on cross sections). Conclusions: A marker of cells located exclusively in limbal crypts or Vogt palisades and presumed to be the true LSCs is still missing. CK15 seems the most specific markers for limbal basal cells but not restricted to specific niches. [ABSTRACT FROM AUTHOR]

Purpose: Compliance issues are often reported in ophthalmology for glaucoma which is one of the leading causes of blindness and visual disability and for dry eye syndrome (DES) requiring multiple frequent instillations. The smart electronic devices have emerged as one of the most reliable means of improving adherence to treatment, allowing objective quantification of the taken treatments. Aim: to report the pilot use in DES of two new electronic devices designed for the monitoring of adherence to eye drop medications. Methods: The KaliDROP for eye drop bottles, and the KaliJAR for single‐dose‐units (SDU; Kali‐CARE, Santa Clara, CA, USA) were 3G wireless devices that recorded and transmitted in real respectively the number of instillations (when the bottle in pressed and reversed) and the number of SDU discarded after use. Both were tested by one of the author suffering from DES. The Kali Drop was tested during 28 consecutive days with an ABAK bottle of Thealose (Thea laboratories). The KaliJAR was tested during 47 consecutive days with SDU of Thealoz Duo Gel (Thea laboratories). In parallel, he noted on paper the date and time of each eye drop instillation. The agreement with the data recorded in the software was analyzed. Results: On the 354 instillations noted on paper, 352 (99.4%) were recorded by the KaliDROP device. Two instillations were not recorded, and one instillation was wrongly detected. On the 2200 SDUs introduced in the KaliJAR, 2136 (97.1%) were recorded. 64 SDUs (2.9%) were not recorded but none was wrongly detected. This percentage of false negative could be explained by the fact that 30–50 SDUs were inserted immediately one after the other. The number of undetected SDU would be virtually null if each SDU was discarded separately as normally recommended. Conclusions: These two innovative simple‐to‐use devices could find strong applications in therapeutic education and clinical trials. [ABSTRACT FROM AUTHOR]

Purpose: We have recently developed and characterized a unique model of herpetic keratitis reproducing dendritic or geographic ulceration on human corneas stored in our corneal Bioreactor (BR). The BR is a multi‐application experimentation platform developed to keep the cornea alive for prolonged periods of time in a physiological environment of intraocular pressure and environmental renewal. Aim: To evaluate whether this model can be used to test the efficacy of antiviral eye drops. Methods: HSV‐1 keratitis was obtained as previously described: paired corneas discarded from our eye bank were stored in the BR for 14 days (D) at 21 mmHg and 2.6 µl/min of medium renewal (CorneaMax, Eurobio, France) to allow epithelial regeneration. A superficial linear epithelial lesion was then made in the centre, followed incubation of the epithelial surface for 1 hr in a solution of 106 Plaque‐Forming Unit/mL HSV‐1. Three days post‐infection, 0.15% ganciclovir ophthalmic gel used with active molecule at 3.75 µg/ml was instilled on the cornea inside the BR, 3 times/day during 14D and 2 times/day during 5D. The gel was rinsed after 10 min with BSS. The paired cornea was treated similarly with the vehicle (carbomer 974P). Changes in epithelial ulceration size were monitored daily by non‐invasive imaging through the transparent BR windows. Histology and immunolabeling (HSV‐1 gpB, epithelial markers) were done at D40. Results: For the control corneas, the entire epithelium desquamated at D10. At D40 all keratocytes expressed gpB and endothelial cells partly disappeared. For ganciclovir‐treated corneas, only half of the epithelium desquamated at D14, then it began to regenerate during the third week to reach 1 to 3 epithelial layers at D40. Epithelial and endothelial cells were not infected (gpB‐). Some keratocytes were infected only in the periphery. Conclusions: Model of herpetic keratitis in the BR has a high potential to evaluate antiviral eye drops efficacy. [ABSTRACT FROM AUTHOR]

Purpose: The detection and selection of corneas operated on for refractive surgery (LASIK or photorefractive keratectomy (PKR) by excimer) will become a major concern for eye banks in the coming years because this surgery is often forgotten during the interview with the deceased's relatives. Passive storage in organoculture (OC) prevents the detection of the traces of this surgery because of stromal edema making the cornea less transparent and wrinkled. On the other hand, the active storage machine (ASM) restores intraocular pressure, restores the cornea to its original shape, respects transparency and incorporates non‐invasive controls. It is, therefore, likely to facilitate the detection of refractive surgeries. We present here the case of 2 corneas operated on with PKR and stored successively in OC and ASM. Methods: The 2 corneas of a 49‐year‐old donor operated 20 years earlier by PKR for ‐2 and ‐3 diopters myopia were stored in OC for 14 days and then placed in ASM for 48 hr. Transparency (transparometer, BiiGC), thickness map and OCT topography (Casia 1, Tomey) were performed under the 2 storage conditions. At the end of the observation, histology and electron microscopy were performed. Results: Traces of PKR remained unnoticed in OC while they were evident in ASM: central epithelial anomaly, central thinning and flattening of central keratometry were found respectively in transparometry, pachymetry map and OCT topography. Histology and ultrastructure confirmed the absence of Bowman's membrane, pathognomonic of PRK. Conclusions: By placing the cornea under physiological conditions, and in particular by triggering its deswelling thanks to the intraocular pressure and by restoring its natural curvature, the ASM should allow effective detection of refractive surgery traces. [ABSTRACT FROM AUTHOR]

Purpose: The HEC (Hoechst‐Ethidium‐Calcein) labeling, which we described in 2011, is a simple destructive test unanimously adopted to determine the pan‐endothelial viable endothelial cell density (vECD) using image analysis (e.g. using CorneaJ, ImageJ dedicated plugin). Sometimes, the endothelial surface folded by stromal edema remains difficult to analyze. We present here the HECplus technique, which, with an optimization of image acquisition and analysis, further increases the accuracy of vECD determination, particularly in endothelial folds. Methods: 10 pairs of corneas stored in organoculture at 31°C were analyzed. The endothelial side was incubated with 4 µM Calcein‐AM, 1 µM Ethidium and 11 µM Hoescht 33342 for 45 min at RT. Fluorescence was acquired in 3 channels first with a macroscope, in 3D with an image stacking of the entire graft then with a microscope after flat mounting. A 2 mm plastic pellet emitting constant fluorescence at 510 nm was imaged at the same time to normalize the images. Only one sharp image was reconstructed in ImageJ from the stacks of images. The final standardized images were then analyzed by CorneaJ (HEC group) and then manually by delimiting all viable surfaces on a graphic tablet (HECplus group) Results: The vECD was significantly higher by 8.2 ± 1.9% in the HECplus group versus HEC (2059 ± 127 versus 1989 ± 156 cells/mm2). Endothelial cells died preferentially in the top of the folds and were better preserved in the bottom of the folds. While HEC wrongly considered some folds as dead, HECplus allowed the error to be corrected. The folding pattern was similar between the 2 corneas of the same pair but differed from one donor to another. The normalization of images for calcein showed overall differences in the level of fluorescence between corneas, suggesting differences in metabolic activity. Conclusions: We recommend applying the HECplus protocol when a very high degree of accuracy is required, for example when comparing storage media. [ABSTRACT FROM AUTHOR]

Purpose: The preclinical testing of the efficacy and safety of hyperosmotic anti‐edematous eye drops is complicated by the lack of a simple in vivo and ex vivo animal model. Our patented corneal bioreactor (BR) allows prolonged preservation of the human cornea by restoring physiological intraocular pressure and renewing the culture medium. Under certain experimental conditions, the corneas preserved in BR are nevertheless thicker than normal (684 ± 52 um, A J Transplantation 2019). Aim: To use this characteristic of corneas stored in the BR to test the efficacy of new hyperosmotic eye drops prototypes. Methods: Human corneas (n = 4) discarded by our eye banks were used whatever their endothelial status. They were first stored for 2 weeks into the BR filled with CorneaMax (Eurobio, France), with 21 mmHg in the endothelial chamber to allow regeneration of a normal epithelium as previously shown. Using a specific BR lid allowing sterile instillation in front of corneal center, each cornea was treated topically with 3 drops, every 2 hr, repeated 3 times. The epithelial chamber was emptied just before instillation and filled again with CorneaMax 2 min after instillation. Three corneas were treated with a commercial eyedrop containing 5% NaCl and 0.15% sodium hyaluronate and one cornea with Balanced Salt Solution (BSS) as a control. Central corneal volume (on an 8mm disc, CCV8 in mm3) was measured with OCT (Casia, Tomey) immediately before and after each instillation. The decrease in CCV was calculated for each instillation and before the first and the third instillation. Results: For corneas treated with NACl 5%, CCV decreased by (mean ± SD) 1.0 ± 0.4, 0.7 ± 0.1 and 0.6 ± 0.2 mm3, respectively, for the first, the second and the third instillation, and globally by 2.4 ± 1.1 mm3. The cornea treated with BSS remained unchanged. Conclusions: The BR is an efficient tool to quantify the efficacy of hyperosmotic eye drops designed to deswell human cornea. [ABSTRACT FROM AUTHOR]

Purpose: Corneal endothelial cells (ECs) are continuously exposed to two physical stimuli: intraocular pressure (IOP) on their apical surface and contact with the Descemet membrane on their basal surface. We suppose that IOP could act on mechanoreceptors which could induce mechanotransduction in ECs: YAP (Yes Associated Protein) a transcriptional coactivator and Zo‐1 included in apical tight‐junctions. In numerous cell types, mechanical stimuli induce YAP nuclear translocation, which regulates genes expression involved in cell survival. Aim: To study the localization of YAP in human ECs and its link with Zo‐1 homeostasis, using our corneal bioreactor (BR) capable of restoring a transcorneal pressure gradient equivalent to the IOP. Methods: Human corneas were placed in the BR, during 14 days at 21 mmHg with circulation of a storage medium (CorneaMax, Eurobio, France). For control, paired corneas were stored in a flask without IOP in the same medium. Immunolabelling for YAP and Zo‐1 was done on paraformaldehyde of methanol fixed whole corneas by incubating primary and secondary antibodies on the endothelial side only. After flat mount, the entire endothelium was observed by epifluorescence microscope (IX81, Olympus, Japan). Results: Several endothelial areas of corneas stored in the BR had regular hexagonal EC, well‐organized tight‐junctions, and a homogeneous nuclear localization of YAP. Corneas stored without IOP presented some areas without EC and other areas covered by EC with irregular shape, disorganized tight‐junctions and an heterogeneous localization of YAP, with some cells having focal expression in their nucleus and others cytoplasmic expression. Conclusions: The bioreactor is a promising tool to study the role of mechanotransduction in human EC survival and function. IOP seems able alter expression of YAP and of Zo‐1. IOP‐induced nuclear translocation of YAP in various corneal areas seems linked with integrity of tight‐junctions. [ABSTRACT FROM AUTHOR]

Purpose: Staphylococcus aureus (SA) is the second most frequent pathogen of bacterial keratitis. Fibronectin‐binding proteins (FnBPs) which are staphylococcal adhesins are found in 95% of clinical isolates and are shown to be important adhesins or invasion factors in ulcerative keratitis. Aim: to study the role of the fibronectin (Fn) in SA adherence on corneal epithelium and its role in the initiation of the keratitis. Methods: We used 4 strains of SA: 8325‐4 and DU5883 (8325‐4 isogenic strain lacking in fnbA/B), and SA113 and SA113 srtA‐ (Isogenic strain lacking in Sortase A). Levels of internalisation were measured by lysostaphin protection assay on human corneal epithelial HCE‐2 cells. SA adherence level was calculated by determining the ratio between the SA area and area of corneal epithelial ulceration blocked by Fn antibodies or not. The surface of epithelial ulceration was defined with Fn immunostaining. Results: On cells, intracellular loads of 8325‐4 and DU5883 strains were, respectively, 4.2 and 2.6 log CFU/106 cells (p < 0.0001) and intracellular loads of SA113 and SA113 srtA‐ strains were, respectively, 4.3 and 3.9 log CFU/106 cells (p < 0.0001). Onto intact and injured corneal epithelium, SA (8325‐4) adherence level was, respectively, 0.04 and 4.36 MFI (Mean Fluorescence Intensity) (p = 0.0003). On injured corneal epithelium, adherence level of 8325‐4 and DU5883 was significantly reduced from 3.7% to 0.5%, respectively (p < 0.001), whereas adherence level of SA113 Wt and srtA‐ was not significantly reduced from 1.5% to 0.7%, respectively (p = 0.1873). Blockage of Fn showed a significant decrease of 87%, 85% and 86% of the 8325‐4, SA13 WT and SA113 SrtA‐ adhesion level, respectively, but a not significant decrease for DU5883 (43%). Conclusions: Capacity of SA to bind the epithelium of intact cornea was weak, whereas corneal epithelium ulceration significantly increased its adherence. Adherence is mediated by the presence of Fn inside the epithelium, FnBPs and adhesine proteins in SA. [ABSTRACT FROM AUTHOR]

Purpose: TGF‐βs trigger ENdothelial‐Mesenchymal Transition (EnMT) in corneal endothelial cells (ECs) in animal models of endothelial wound healing. All the 3 isoforms of TGF‐β were detected in human aqueous humor and their receptors were detected in ECs. More and more evidence indicated that central and peripheral ECs have a different status. Aim: To compare the effects of the 3 isoforms of TGF‐β1, 2 and 3 on ex vivo corneal endothelium and to compare their effects on central and peripheral endothelium. Methods: 10 pairs of fresh human corneas (body donation to Sciences) were incubated in a serum free medium for 2 weeks at 31°C. One cornea of each pair was incubated with 10 ng/mL of TGF β1 or 2 or 3, the paired cornea without TGF‐β or with another isoform of TGF‐β (n = 2 for each comparison). Immunolabelling using markers of ECs (ZO‐1, CD166, COX IV and N‐cadherin) and of mesenchymal cells (MCs) (CD44, Collagen 1 and fibronectin) was performed of flat‐mounted corneas. Staining was observed with confocal microscopy to map the regional expression of each marker. Results: In the controls, MC markers were expressed only by ECs located at the extreme periphery and by trabecular cells. The 3 isoforms of TGF‐β induced the loss of EC markers and the increase of MC markers in central and peripheral endothelium. The effects of TGF‐β1 and 2 were higher than that of TGF‐β3. TGF‐β‐induced alterations were more pronounced in central than in peripheral endothelium. Conclusions: Normal ECs at the extreme periphery and trabecular cells share characteristics of MCs. These ECs could then derive from trabecular cells. TGF‐β (10 ng/ml) induces EnMT in the center of the endothelium of normal corneas. Physiological concentration of TGF‐β might contribute to maintain peripheral ECs in an undifferentiated state able to slowly proliferate and contribute to endothelial homeostasis. [ABSTRACT FROM AUTHOR]

Purpose: Our patented corneal bioreactor (BR) improves the survival of human corneas ex vivo in the long term by restoring intraocular pressure and renewal of storage medium in a sterile environment. Aim: To use the BR to test the efficacy of eye drops containing devitalized ground amniotic membrane of umbilical cord extract (AMUCE) to treat epithelial ulcerations. AMUCE contains high level of proteoglycans, collagen and physiological growth factors. Methods: Only pairs of human corneas were used to obtain best possible controls. They were either corneas discarded by our eye bank (cell density between 800 and 2000 cells/mm2) or specific donation for research (normal corneas). They were first stored for 2 weeks into the BR filled with CorneaMax (Eurobio, France), with 21 mmHg in the endothelial chamber to allow regeneration of a normal epithelium as previously shown. Then, a 6.0 mm diameter calibrated epithelial ulcer was made. Using a specific BR lid allowing sterile handling, each cornea was treated topically 3 times daily either by the AMUCE eye drops (TBF, France) or by the vehicle (0.9% NaCl). Closure of the ulceration was monitored daily using fluorescein staining quantified by image analysis (FIJI) of macro‐zoom microscopy. After wound healing corneas were fixed for histology (Hematein Eosin Saffron) and immunolabeling of epithelial markers. Results: Three pairs were used. Wound healing was obtained at D3, D7, D5 with the AMUCE and D3, D6, D6 with the vehicle (no significant difference). Nevertheless, drug‐treated corneas had a thicker epithelium than with vehicle only. Histology and immunolabeling for E‐Cadherin and cytokeratin K3/K12 confirmed that epithelium of AMUCE‐treated corneas was more differentiated than with vehicle only. Conclusions: The BR allows controlled studies of epithelial wound healing on human corneas with repeated instillations. By improving epithelial maturation, AMUCE eye drop is an interesting candidate for epithelium regeneration. [ABSTRACT FROM AUTHOR]

Purpose: We present the development and characterization of an electromechanical device that quantifies the elasticity of the crystalline lens ex vivo by measuring its reaction force when it is compressed by a rod moved step by step. Our specifications required a robust system, fast to allow serial analysis, capable of measurements on an immersed lens, and highly resolved to detect small changes in elasticity. Methods: A first simple manual device was constructed to dimension the measuring system. The lenses (human and porcine) immersed in a buffer solution were placed on a precision weighing balance and compressed from top to bottom by a rod mounted on a micrometer screw. This system made it possible to define, among other things, the resolution of the balance (1 mg, 0.01 milli‐Newton). In a second device, the rod system was motorized with a resolution of 0.625 µm pitch in order to automate the measurement, and the balance data were integrated in real time into software executable on a desktop computer. A control camera (30 µm/pixel) was added to facilitate the approach of the rod in contact with the lens and to document the deformation experienced. Results: The built‐in calibration of the balance allows direct measurement in Newton. The device allowed determining the very homogeneous behavior of the lens (elastic reaction versus displacement) as well as the breaking point of the capsule which is much higher than the elastic deformation (at about 100 times the weight of the lens). Automation also allows the measurement of lens relaxation times during pulsed stimulation, for example by compression/decompression of 1 mm on a porcine lens. Conclusions: With a measured effective fidelity of 5 µm and 0.01 mN, this direct compression system allows the mechanical elastrometry of the lens. The measurement of a large series of human and porcine lenses of different ages is in progress to determine its accuracy and repeatability. It will be useful for any research on lens biomechanics. [ABSTRACT FROM AUTHOR]

Purpose: Most recent studies about FECD highlighted its complex and heterogeneous genetic background. Yet, the sequence of cellular mechanisms leading to the formation of Descemet excrescents (Guttae), endothelial cell (EC) death and loss of pumping functions remains poorly understood. Aim: To present new phenotypic characteristics of ECs observed in late onset FCED, opening new perspectives in understanding its pathogenesis. Methods: Thirty corneal central buttons were collected from FECD patients during graft. They were compared with fresh normal corneas (body donation to Sciences), and with 10 whole corneas with early stage of FECD (discarded from our eye bank). The morphology of ECs and of guttae, the expression of EC markers, of stem cell markers, of proliferation marker and of extracellular matrix was studied using the immunostaining of flat mount of corneas. Results: Central ECs of FECD specimens and extreme peripheral ECs of fresh corneas shared common features: weaker expression of the usual CE markers (Na/K ATPase, ZO‐1, COXIV, CD166, N‐cadherin), expression of mesenchymal stem cells markers (nestin, c‐Myc, CD44), presence of proliferating cells (Ki67), presence of numerous dead cells (Ethidium), numerous cells containing pigments, isolated and confluent guttae, thicker Descemet membrane (DM), and localization of collagen I, IV, V and VIII. In corneas with early stages of FECD, the peripheral radial rows of EC and Descemet striae normally observed in healthy corneas were here significantly enlarged. Conclusions: These new features of ECs and DM suggest that central ECs of FECD acquire or retain characteristics of peripheral ECs of healthy corneas, corresponding to less differentiated cells. We, therefore, suggest a new physiopathology mechanism involving the abnormal dedifferentiation of ECs in the centre of the cornea, resulting in the slow formation of the plaque‐like endothelial structure, typical of FCED. [ABSTRACT FROM AUTHOR]

Purpose: The deswelling of a corneal graft at the end of storage in the eye bank is essential before perforating or deep anterior lamellar keratoplasty to allow good donor/recipient congruence and to accelerate visual recovery. In France, CorneaJet (Eurobio, France) containing 5% Dextran T500 is the most commonly used medium, but its deswelling kinetics is not referenced and the duration of 48 or 72 hr is chosen purely by habit while the increase in endothelial toxicity over time is documented. Aim: To determine the corneal deswelling kinetics in CorneaJet. Methods: 12 corneas discarded by our eye bank were used. They had an endothelial cell density at reception (day 2 to day 4) of 1713 ± 389 cell/mm2 (mean ± SD). All had been stored in organoculture in CorneaMax medium at 31°C for at least 1 week. The corneas were immersed in 50 ml vials of CorneaJet (same batch) and were stored at 31°C. Central corneal thickness (CCT in µm) and central corneal volume (on an 8mm disc, CCV8 in mm3) were measured by optical coherence tomography (OCT, Casia I, Tomey, Japan). Transparency was evaluated on a prototype measuring contrast transmission (relative scale in %). Measurements were performed before immersion in CorneaJet and repeated at hour (H)3, H6, H9, H12, H18, H24, H30, H36 and H48. Results: All data were expressed as mean ± SD respectively at the different times. CCT: 1284 ± 249; 1031 ± 233; 897 ± 22; 773 ± 160; 741 ± 123; 673 ± 75; 626 ± 27; 638 ± 41; 645 ± 49; 652 ± 55. CCV8: 73 ± 9; 61 ± 9; 53 ± 9; 47 ± 8; 44 ± 6; 40 ± 4; 38 ± 3; 37 ± 3; 37 ± 3 and 37 ± 3. Transparency: 13 ± 4; 19 ± 5; 24 ± 6; 29 ± 6; 33 ± 7; 40 ± 7 41 ± 7; 44 ± 9; 45 ± 8 and 44 ± 11. Conclusions: Corneal deswelling in CorneaJet is rapid during the first 12 hr and reaches a plateau at 20 hr. After 20 hr, transparency and thickness are ideal for surgery. [ABSTRACT FROM AUTHOR]

Purpose: Lens elasticity can be calculated from the deformation observed when it is submitted to different rotational speed in a specific device called lens spinner. Up to now, the accuracy of this technique is limited by dynamic blur due to small movements of the lens during rotation and requires expensive high‐speed cameras and intense light sources that can be harmful to the lens. Aim: to assess whether a stroboscopic system could overcome this limit. Methods: The critical parameter was the synchronization of the light source with the position of the lens on the rotative plate. We thus designed a digital control system combined with a stepper motor that allowed reaching more than 2000 rpm while maintaining an angular accuracy of 0.2 degrees with respect to position of the illumination source. This allowed obtaining sharp 8‐megapixel images with a standard commercially available camera. We analyzed fresh lenses from 6‐month‐ and 3‐year‐old pigs and lenses from human donors (specific retrieval authorization for scientific purpose during corneal donation, from the Agence de la Biomédecine). Results: Lens imaging was obtained by overlaying images of several short flashes: images referenced by their angular position allowed a 360‐degree reconstruction of the geometry of the lens (one image every 45 degrees). The light flux remained moderate with 10 flashes of 100 microseconds with a 20W white LED source. The maximal rotational speed for imaging was 2000 rpm allowing a wide range of investigation as the limit of the resistance of the porcine lens capsule was observed at much higher speeds. Conclusions: Compared to pre‐existing lens spinner systems designed to essentially maintain a stable rotational speed, the fine angular position control combined with the stroboscopic lighting lens spinner strongly improves the reading of the lens' geometry by an order of magnitude of 10 and will be a very useful device for research in lens biomechanics. [ABSTRACT FROM AUTHOR]

Purpose: Our patented corneal bioreactor (BR) improves the long‐term survival of human corneas ex vivo by restoring intraocular pressure and renewal of storage medium. Aim: To use the BR to test the efficacy of eye drops containing devitalized ground amniotic membrane of umbilical cord extract (AMUCE) to treat deep stromal ulcerations. AMUCE contains high level of proteoglycans, collagen and physiological growth factors. Methods: Only pairs of human corneas (with ECD > 800 cells/mm2) were used to obtain best possible controls. They were first stored for 2 weeks into the BR filled with CorneaMax (Eurobio, France), with 21 mmHg in the endothelial chamber to allow regeneration of a normal epithelium. Then, an 8.0 mm diameter calibrated epithelial ulcer was made with a spatula and a 6 mm diameter × 100 µm in depth was made with Excimer laser in the stroma. Using a specific BR lid allowing sterile handling, each cornea was treated topically 4 times daily either by the AMUCE eye drops (TBF, France) or by the vehicle (0.9% NaCl). Closure of the ulceration was monitored daily using fluorescein staining quantified by image analysis (FIJI) of macro‐zoom microscopy. After wound healing corneas were fixed for histology (HES) and immunolabeling of epithelial markers. Results: Three pairs were used. Epithelium wound healing was obtained at D3, D9, D10 with AMUCE and D3, D9, D19 with the vehicle. Histology performed in ulcerations centre showed 3‐6 epithelial layers with AMUCE and 1‐4 with the vehicle. At D21, significant epithelial detachment was observed only for 2 corneas treated with the vehicle. For one pair, immunolabelling for E‐cadherin and Cytokeratin K3/K12 showed better epithelial differentiation with AMUCE than with vehicle. No evidence of stromal regeneration was observed in the 6 corneas. Conclusions: AMUCE has potential for a faster and lasting epithelium regeneration (stratification and maturation). Nevertheless, it has no impact on stromal regeneration in this model. [ABSTRACT FROM AUTHOR]

Purpose: We recently showed that corneas stored in the corneal bioreactor (BR) that restores IOP and storage medium renewal had increased endothelial viability and enzymatic pump expression, less stromal edema and posterior folds and more stratified and differentiated epithelium than paired corneas stored passively in the same medium. Aim: To use this model to analyze the respective contribution of corneal(C), limbal(L) and conjunctival(Conj) epithelium in regeneration of corneal epithelium. Methods: Five pairs of fresh human corneas (body donation to Sciences, death to retrieval time <20 hr) were used. During retrieval, a rim of conjunctiva was left intact. Five models were then built by totally mechanically removing 1, 2 or 3 epithelial population(s): C‐L+Conj+, C‐L‐Conj+, C‐L+Conj‐, C+L‐Conj‐, C‐L‐Conj‐ (control). Corneas were then stored for 1 month in the BR (21 mmHg, 2.5 µl/min of CorneaMax, Eurobio) to allow epithelial regrowth. The epithelia were then examined by immunostaining of flat mounted whole cornea using cytokeratine CK12 (corneal epithelium), CK15 (limbal epithelium) and Hoechst (all cells nuclei). Transparency was assessed by a contrast transmission analyzer. Results: No epithelium regenerated in the controls. In the 4 other models, the cornea was covered by a multilayered corneal epithelium, even in the C‐L‐Conj+, characterized by expression of K3/K12 and a transparent tissue. Conclusions: The 3 ocular surface epithelia (comprising the conjunctival one) contain stem‐cells or progenitors able to migrate onto the whole cornea and regenerate corneal epithelium independently of each other. In this ex vivo model of corneas deprived of blood supply and innervation, the neovascularization cannot develop. This suggests that the corneal conjunctivalization characteristic of limbal stem cell deficiency is not due to the absence of a progenitor capable of transdifferentiating into corneal epithelium but mainly to the absence of an anti‐angiogenic barrier. [ABSTRACT FROM AUTHOR]

Purpose: With very photophobic patients, the advantages of red or near infrared (NIR) light to develop new ophthalmology imaging devices seem obvious: no or little glare, possibility of long signal integration, no phototoxicity, and lesser autofluorescence of ocular tissues. Nevertheless, in this range, the shortest possible wavelength facilitates signal detection. Aim: To determine the maximal irradiance tolerated with 6 wavelengths: 2 red, 2 far red and 1 NIR lights in order to determine the shortest wavelength well tolerated by patients, in comparison with the standard cobalt blue light of ophthalmology slit lamp. Methods: Study design: interventional, monocentric, single group assignment study on 30 eyes of 30 patients with infectious keratitis after informed consent. Thanks to a customized machine from our laboratory, the photophobic eye was exposed to the 6 lights with increasing intensity. The patients switched off the light when the discomfort was too elevated. The maximal cumulative irradiance possible at 482, 650, 675, 700, 750 and 800 nm were respectively 171, 689, 759, 862, 920 and 889 mW/cm2. Results: The maximal cumulative irradiance tolerated increased significantly with wavelength (p < 0.001). However, the difference was not significant between each increment. Thus, red at 675 nm gave a significantly higher cumulative irradiance than blue at 482 nm. Red at 700 nm did not provide significant gain; however, far red at 750 nm still provided additional gain, but not at 800 nm. The shortest wavelengths were stopped more quickly and more than 50% of patients reached the maximum irradiance delivered by the source at 750 and 800 nm. Conclusions: We demonstrate that a light source at 750 and 800 nm can be used for ophthalmic imaging with good tolerance in the photophobic patient. [ABSTRACT FROM AUTHOR]

Purpose: Treatment of complex macular holes (MH) is challenging. The use of amniotic membrane (AM) has been described but is not yet codified, and we need to collect more cases. Aim: To describe tips and tricks for the use of a large patch of lyophilized AM (lAM) in this indication. Methods: Consecutive 10 cases series: two patients with severe MH‐induced rhegmatogenous retinal detachment (RRD), three with recurring RRD + persistant MH, five with persistant MH after inner limitant flap. Our IRB approved the study and all patients signed an informed consent. lAM (Visio Amtrix, TBF) was chosen because of its availability, easy handling, thinness and transparency. A 3‐ to 4‐mm diameter disc was trephined with a disposable punch, stained with 0.06% Trypan Blue (after a first failure due to poor visualization of an unstained lAM) and rinsed with BSS. Stained lAM were inserted through a 23G sclerotomy using a standard perfusion catheter and unfolded on top of the MH using a forceps and a backflush needle. Tamponade was done with gas (17% C2F6, Arceole; n = 6) or heavy silicone (Densiron 68, Fluoron; n = 4) depending of the case. Routine follow‐up was done for 12 months, including SD‐OCT (Spectralis, Heidelberg) and color fundus photography (CR2‐AF, Canon). Results: Ten eyes of 10 patients of 62 ± 9 years with preoperative mean BCVA 1/80 (1.9 logMAR) and a mean minimal MH diameter of 942 ± 333 μm underwent surgery. The blue staining disappeared within 15 days. Anatomic success occurred for all patients: six/10 were closed, four/10 had reduced diameter. BCVA improved in all cases. The five RRD were reattached without recurrence. Apart one failure (lAM left unfolded) with migration at day 1 and MH closed with central scar, no secondary displacement of the patch occurred. Conclusions: A large patch of lAM seems a new method to cure complex MH in different situations. Trypan blue staining is, in our hand, absolutely necessary to allow correct positioning. Complete unfolding seems also required to allow good adherence over the MH. [ABSTRACT FROM AUTHOR]

Purpose: The standard version of our corneal active storage machine (ASM) maintains corneas viability much better than standard organ culture over 1 and 3 months (Am J Transplant 2019, Transplantation 2020). While it aims at recreating a physiologic environment (IOP and medium renewal), corneas were nevertheless immersed from both sides, probably contributing to slightly higher thickness than in vivo. Aim: To assess whether epithelial alternate exposure to air and liquid (AirLifting, AL) could further improve quality of corneas stored into the ASM. Methods: Twenty paired human corneas (initial ECD > 2000 cells/mm2 and <10% difference between right and left) were randomized to be stored for 1 month in the ASM either fully immersed or with AL. All other parameters were similar: IOP (21 mmHg), endothelial medium flow (2.6 μL/min) and nature (CMax, Eurobio). AL was done with CMax (6 cycles air‐liquid/min). Specular microscopy ECD, OCT thickness, transparency were monitored at Day (D)2, D26, D28. Pancorneal viable ECD (Hoechst/Ethidium/Calcein‐AM), histology, immunostaining, western blot for epithelial cells and endothelial cells (ECs) structure/functions proteins were done at D28. Results: At D2, ECD was 2412 ± 222 versus 2385 ± 251 cells/mm2 (p = 0.349) respectively with and without AL and 2109 ± 165 versus 2119 ± 199 cells/mm2 (p = 0.621) at D26. The viable ECD at D28 was 2170 ± 321 versus 2185 ± 275cells/mm2 (p = 0.794). All corneas remained thin transparent during storage and at D28 only, corneas were thinner with AL: 667 ± 60μm versus 695 ± 55μm without (p = 0.048). Epithelial quality was similar, but the epithelium was thicker with AL (49 ± 20) than without (23 ± 7μm) (p = 0.001) and had more cell layers. Endothelial cells structure/functions were similar. Conclusions: By mimicking blinking, AL enhances epithelium structure during storage into the ASM. It may also slightly help controlling corneal thickness. ASM with AL may be useful for penetrating and anterior keratoplasty as well as for research works. [ABSTRACT FROM AUTHOR]

ABSTRACT. Purpose: The onset of acute retinal necrosis secondary to herpes simplex encephalitis is exceptional. We report such an association in an immunocompetent patient, in whom the genome of herpes simplex virus type 1 (HSV-1) was identified successively at both sites of infection. Methods: Polymerase chain reaction (PCR) assay of HSV-1 in cerebrospinal fluid (CSF) and aqueous humour. Results: An immunocompetent patient aged 40 years presented with HSV-1 encephalitis, which was confirmed by imaging, viral serology and identification of the HSV-1 genome in the CSF. The subject's immunological profile was normal. The patient was treated with foscavir. Six weeks after clinical recovery and negative PCR, the patient presented with a unilateral acute retinal necrosis syndrome. Polymerase chain reaction of the aqueous humour was positive, while serology and PCR of the CSF remained negative. Conclusion: Identification of the HSV-1 genome at the two successive sites of infection stresses the possibility of brain-to-eye transmission of HSV-1. [ABSTRACT FROM AUTHOR]