Ophthalmic grafts: evaluation of the storage media for cornea and sclera - PDF Document

Presentation Transcript

  1. Ophthalmic grafts: evaluation of the storage media for cornea and sclera Word count: 20942 Fien Hanssens Student number: 01400412 Supervisor(s): Prof. Dr. Hilde Beele, Dr. Dimitri Roels A dissertation submitted to Ghent University in partial fulfilment of the requirements for the degree of Master of Science in the Biomedical Sciences Academic year: 2018 – 2019

  2. Preface After 5 years Biomedical Sciences, the end of my studies is near. Therefore, I would like to take the opportunity to thank some people who helped me a lot during this period. First of all, I would like to thank my promoter Prof. Dr. Hilde Beele who gave me the chance to work on a very interesting topic. She guided me in all the aspects of this master thesis and was always prepared to give me feedback. My heartfelt thanks to do a lot (!!) of effort for the approval of my ethical committee. My co-promoter Dr. Dimitri Roels and supervisor Dr. ir. Sarah Glorieux provided me very useful information. Dr. Roels was always ready to give me an explanation about the corneal transplantations and helped evaluating the sclerae that are used in this master thesis. Dr. ir. Glorieux always thought along with me during this 2-year period. I could enter her office with every question and she also gave me different insights and a lot of valuable ideas during my thesis period. My special thanks to Prof. Dr. Ilse Claerhout and her staff, who welcomed me. She explained me everything about the differences between the warm and cold storage method and helped me establish my retrospective study. Also Matthias Claeys was always prepared to answer all my questions about the retrospective analysis. Even skyping on a Friday night was not too much for him, thanks a lot Matthias! Leen Pieters from the Morphology Lab deserves a special thanks because she taught me to make histological slides of the sclerae. Hereby, I would also like to thank the complete staff of the Tissue Bank. From day one, you welcomed us with open arms and all of our questions could be asked! You did everything to make sure that my cold storage medium arrived on time, but unfortunately I did not need it. Without Joke, Marieke and Sarah I could never have done the complete procedure of cornea and sclera processing by myself. I cannot thank you enough for all the patience you had with me during these 2 years! On top of that, Joke also made time to score my histological slides and gave me useful remarks that helped me during my analysis. Lore Vandevelde, my fellow student, made sure that this period was also full of fun. She always made time for all my questions (especially about Excel) and was always ready to help me with anything. Because of her, I will never forget my 23rd birthday! She was, and will remain, my greatest support during these years. I am sure that I would never have gotten this far without her! Finally, I would also like to thank my parents, mom & John and dad & Nancy, for their financial and general support during these 5 years. Without them I could never have started this study. They always have my back and would do anything for me. Also my boyfriend Arthur deserves a special thank you. He had to listen to all my frustrations during stressful periods but he always managed to encourage me and to cheer me up.

  3. Table of contents Summary .............................................................................................................................. 1 1.Introduction ................................................................................................................... 2 1.1 Anatomy of the eye .................................................................................................... 2 1.2 The cornea ................................................................................................................. 2 1.3 Corneal transplantation or keratoplasty ...................................................................... 4 1.3.1 Indications ........................................................................................................... 4 1.3.2 Techniques .......................................................................................................... 5 1.3.3 The procedure ..................................................................................................... 7 1.3.3.1 Procurement ................................................................................................. 7 1.3.3.2 Preparation ................................................................................................... 8 1.3.3.3 Evaluation ..................................................................................................... 8 1.3.3.4 Storage ......................................................................................................... 9 i) Cryopreservation .......................................................................................... 9 ii) Cold or hypothermic storage ........................................................................10 iii) Warm storage or organ culture ....................................................................12 iv) Alternative ...................................................................................................12 1.3.3.5 Implantation and follow-up ...........................................................................13 1.4 The sclera .................................................................................................................13 1.5 Scleral transplantation ...............................................................................................14 1.5.1 Indications ..........................................................................................................14 1.5.2 Procedure ...........................................................................................................14 1.5.2.1 Procurement ................................................................................................14 1.5.2.2 Preparation ..................................................................................................14 1.5.2.3 Evaluation ....................................................................................................15 1.5.2.4 Storage ........................................................................................................15 1.6 Research questions ..................................................................................................16 2.Materials and methods ................................................................................................17 2.1 Cornea ......................................................................................................................17 2.1.1 Disinfection of the bulbus oculi (according to SOP_Pp_C_003_06) ....................17 2.1.2 Dissection of the cornea (according to SOP_Pp_C_010_05) ..............................17 2.1.3 Digital cell count endothelial cells (according to SOP_Pp_C_012_02) ................18 2.1.4 Sampling of the storage medium for microbiological control (according to SOP_Pp_C_011_03) .....................................................................................................18 2.1.5 Vital staining with trypan blue and transfer to transport medium (according to SOP_Pp_C_006_08) .....................................................................................................18 2.1.6 Sampling of the transport medium for microbiological control (according to SOP_Pp_C_014) ...........................................................................................................19 2.2 Corneal transplantations: retrospective analysis .......................................................19

  4. 2.2.1 Data collection ....................................................................................................19 2.2.2 Study design .......................................................................................................20 2.2.3 Data analysis ......................................................................................................20 2.3 Sclera .......................................................................................................................21 2.3.1 Preparation scleral grafts (according to SOP_Pp_S_002_07) .............................21 2.3.2 Hematoxylin-eosin staining scleral grafts ............................................................22 3.Results ..........................................................................................................................24 3.1 Corneal transplantation: retrospective analysis .........................................................24 3.1.1 Mean values database ........................................................................................24 3.1.2 Endothelial cell densities (ECDs) ........................................................................25 3.1.2.1 PKP .............................................................................................................25 3.1.2.2 DSAEK ........................................................................................................27 3.1.2.3 Paired-Samples T Test: warm storage .........................................................27 3.1.2.4 Paired-Samples T Test: cold storage ...........................................................28 3.1.3 Breakdown by donor age ....................................................................................29 3.1.3.1 PKP .............................................................................................................29 3.1.3.2 DSAEK ........................................................................................................30 3.1.4 Breakdown by acceptor age................................................................................31 3.1.4.1 PKP .............................................................................................................31 3.1.4.2 DSAEK ........................................................................................................33 3.1.5 Rejection.............................................................................................................34 3.2 Sclera .......................................................................................................................34 3.2.1 Macroscopic evaluation ......................................................................................34 3.2.2 Histological evaluation ........................................................................................35 4.Discussion ....................................................................................................................37 4.1 Corneal transplantations: retrospective analysis .......................................................37 4.1.1 Endothelial cell densities (ECDs) ........................................................................37 4.1.2 Breakdown by donor age ....................................................................................38 4.1.3 Breakdown by acceptor age................................................................................39 4.1.4 Rejection.............................................................................................................39 4.2 Sclera .......................................................................................................................40 5.Conclusion ...................................................................................................................41 6.Reference list ...............................................................................................................42 7.Addendum ...................................................................................................................... i 7.1 Approval ethical committee .......................................................................................... i 7.2 Traceability of tissues ................................................................................................. iii 7.2.1 Empty form .......................................................................................................... iii 7.2.2 Empty register ..................................................................................................... iv

  5. 7.3 Statistical tests (SPSS) ............................................................................................... v 7.3.1 Normal distribution endothelial cell densities (ECDs) ............................................ v 7.3.1.1 PKP ............................................................................................................... v 7.3.1.2 DSAEK ....................................................................................................... viii 7.3.2 Independent Samples T-Test ECDs .................................................................... xii 7.3.2.1 PKP ............................................................................................................. xii 7.3.2.2 DSAEK ........................................................................................................ xii 7.3.3 Paired Samples T-Test ...................................................................................... xiii 7.3.3.1 Warm storage ............................................................................................. xiii 7.3.3.2 Cold storage ............................................................................................... xiii 7.3.4 Independent Samples T-Test donor age ............................................................ xiv 7.3.4.1 PKP ............................................................................................................ xiv 7.3.4.2 DSAEK ........................................................................................................ xv 7.3.5 Independent Samples T-Test acceptor age ....................................................... xvi 7.3.5.1 PKP ............................................................................................................ xvi 7.3.5.2 DSAEK ...................................................................................................... xviii 7.3.6 Chi-square test rejection ..................................................................................... xx 7.3.6.1 PKP ............................................................................................................. xx 7.3.6.2 DSAEK ....................................................................................................... xxi 7.4 Scoring histological coupes ...................................................................................... xxi 7.4.1 Investigator 1 ..................................................................................................... xxi 7.4.2 Investigator 2 ................................................................................................... xxvi

  6. Summary The cornea is the front transparent part of the eye and is composed of different layers. The endothelium plays the most important role in cornea transplantations since endothelial cells cannot multiply. At the Bank for ophthalmic tissues of UZ Ghent, corneas are stored according to the warm method (28°C-37°C) as it enables a long storage time (5 weeks). The cold storage method (2°C-8°C) can, with suitable storage medium, ensure a storage time of 3 weeks and is easier to perform. Both storage methods are compared using a retrospective analysis in order to find out whether there is a significant (p < 0,05) difference in endothelial cell densities (ECDs) in two separate groups of operation techniques. In both groups, the cold storage is preferred above the warm storage. Also significant (p < 0,05) differences between both storage methods were investigated in groups classified according to the donor age, acceptor age and rejection rates. In general, a significant (p < 0,05) result in favor of the cold storage method was found for all research questions. The sclera is an ophthalmic tissue that can be transplanted and is stored in ethanol at 5°C. The 3 major layers are episclera, stroma and lamina fusca. Recently, the manufacturer of ethanol changed so the old and new ethanol are compared in an in vitro analysis. Histological sections are stained with hematoxylin-eosin and each layer is scored between – and +++ based on different parameters. This analysis showed no differences between both ethanol solutions. 1

  7. 1. Introduction 1.1 Anatomy of the eye The eye or the bulbus oculi is divided into three primary layers, with each of these layers being subdivided [1]. The first layer is the outermost supporting layer, consisting of the cornea, the sclera and the limbus in between. The second layer is the middle uveal layer, which consists of the iris, the ciliary body and the choroid. The last layer is the interior layer, which is known as the retina. The eye is also compartmentalized into three chambers: the anterior chamber, the posterior chamber and the vitreous cavity, which can be seen on Figure 1 [1]. In this master thesis, we will focus on the outer layer and more in particular on the cornea and the sclera Figure 1: Cross-section of the human eye [1]. 1.2 The cornea is the front transparent part of the eye (Figure 1, Figure 2) [1, 2]. The thickness and diameter of the cornea are respectively approximately 0.5 mm and 11-12 mm [3]. Since the eye functions as a sense organ to capture specific light stimuli, the cornea must be transparent and able to project those light rays on the retina, which then forms the final image. At the same time, the cornea is responsible for light refraction before the light rays reach the retina [4]. In addition, the cornea must be strong as it must withstand the pressure of the eye and it has to provide protection to the underlying structures. Furthermore, the cornea is avascular and it serves as a barrier that prevents possible pathogenic micro-organisms [4]. The cornea is composed of five different layers, three of them are cellular (epithelium, stroma and endothelium) and two are interface (Bowman’s membrane and Descemet’s membrane) [4]. From outside to inside the following layers can be distinguished: the epithelium, Bowman’s membrane, the stroma, Descemet’s membrane and the endothelium (Figure 2) [2, 5]. The cornea 2

  8. Figure 2: (a): The position of the cornea in the eye, (b): The different layers of the cornea, (c): Microscopic view of the endothelial cells [2] The epithelium of the cornea is made up of a uniform non-keratinized stratified squamous epithelium and functions as a barrier to chemicals, microbes and water [4, 6]. The refractive power of the eye is, amongst other things, due to the interface between the tear film and the cornea. The epithelium can replace itself every 7 – 10 days using involution and apoptosis. Epithelial stem cells are located at the limbus (the transition between cornea and sclera) and migrate to the central part of the cornea in order to provide new corneal epithelial cells. During migration to the central part, the stem cells differentiate into transient amplifying cells and basal cells. The most superficial layer of the epithelium is composed of 2-3 layers of flat polygonal cells (Figure 3) [4]. Those cells have apical microvilli and microplicae that are covered with a glycocalyceal layer. This layer increases the surface of contact and adherence of the tear film. The suprabasal or wing cells lay under the superficial cells and are less flat. Both the superficial cells and the wing cells have tight junctional complexes in order to inhibit tears entering intercellular spaces. The deepest layer consists of basal cells that are attached to the basement membrane in order to ensure adhesion of the epithelium to the underlying corneal layers. The basal cells are capable of mitosis and divide into wing and superficial cells. Figure 3: Cross-section of the corneal epithelial layer [4]. 3

  9. Bowman’s layer ensures that the shape of the cornea is maintained but has no regenerative capability, which means that injury results in a scar. The stroma, the most abundant layer of the cornea, provides mechanical strength and transparency of the cornea. This layer is also the most refracting part of the cornea. The stroma is transparent, which is the result of the precise organization of stromal fibers and extracellular matrix (ECM), from which the stroma is made of. The collagen fibers are arranged in a parallel way and those bundles are called fibrils. Multiple fibrils together are packed in lamellae. Keratocytes are the major cell type of the stroma, also known as specialized fibroblasts, and are involved in maintaining the ECM environment. Descemet’s membrane is an elastic basement membrane that separates the endothelium from the stroma. If Descemet’s membrane is damaged, the membrane curls and Descemet folds can be observed with a slit lamp. The endothelium is a single layer that ensures corneal clarity by removing water from the stroma using endothelial pumps [7]. It has a typical hexagonal structure and is metabolically active (Figure 2). The endothelial cell density (ECD), expressed in number of cells/mm2, changes during life. In normal healthy corneas, the ECD decreases approximately 0.6% per year but there are basically enough spare endothelial cells to maintain the transparency of the cornea during its lifetime. Because endothelial cells cannot regenerate, dead endothelial cells after illness or injury cannot be replaced [8]. In case of disease or injury, the decrease in ECD increases, which leads to an edema in the stroma and loss of transparency. This can only be cured with a corneal transplantation [3]. Although corneal transplantations are not always due to endothelial defects (as seen in 1.3.1 Indications), corneal transplantations require in almost all cases a healthy endothelium [3, 9]. 1.3 Corneal transplantation or keratoplasty 1.3.1 Indications There are multiple indications for corneal transplantations. The most important ones, besides injury, are ulcers, keratoconus and corneal dystrophies [5]. Ulcers can be traumatic, bacterial, fungal, parasitic or viral. However, only fungal and viral ulcers can be treated with a corneal transplantation, in particular with a penetrating keratoplasty (PKP) (see 4.2.2 Techniques) (Figure 4) [10]. An example of a viral ulcer is an ulceration due to herpes simplex virus, which is a recurrent infection that can cause scarring and inflammation of the cornea. Bacterial ulcers can be caused by wearing contact lenses. If the contact lens damages the corneal epithelium, bacteria can enter the eye. In most cases, the treatment for bacterial ulcers is application of topical antibiotics. Conical deformation of the cornea is called keratoconus (Figure 4) [10, 11]. It is a genetic disease that is mostly bilaterally asymmetrical and has typically a reduced pachymetry (corneal thickness) in comparison to the normal thickness of 540-550 µm. There is a risk of acute hydrops, which means that the cornea can tear, and corneal transplantation can be used as therapy. Corneal dystrophies are innate, bilateral and dominantly inherited [12]. They are classified according to their place of appearance, hence according to the corneal layer where the dystrophy is located. Corneal dystrophies can be seen as white spots at the specific damaged corneal layer. Fuchs endothelial dystrophy (FED), which belongs to the corneal dystrophies, will often lead to an edema (pachymetry > 600 µm). Guttae, holes in the endothelium, can be seen, which indicates a decline in endothelial cell density (ECD) (Figure 4) [10]. The genetics of FED remain unclear. Pseudophakic bullous keratopathy (PBK) occurs more common in patients suffering from cataract due to trauma caused by cataract surgery. This happens when the internal lens is emulsified and aspirated from the eye and replaced by an intraocular lens (IOL) [13]. 4

  10. This surgery can lead to bullae (blisters) in the epithelium and stromal edema, resulting in a higher chance of endothelial damage. Therefore, the lens of the donor cornea is checked during corneal procurement in the clean room. If the donor has an intraocular lens, it is possible that the endothelium is damaged and that the cornea is not suitable for transplantation. Figure 4: Indications for corneal transplantation. (A): Bacterial ulcer, (B): Fungal ulcer, (C): Endothelial guttata seen in Fuchs endothelial dystrophy (FED), (D): Keratoconus, [7]. 1.3.2 Techniques There are various techniques to perform a corneal transplantation. The two main groups are perforating keratoplasty (PKP), where the entire cornea of the patient is removed and replaced by a donor cornea, on the one hand, and lamellar keratoplasty (LKP), where only the affected part of the cornea is replaced (Figure 5), on the other hand [5]. The type of corneal transplantation used, depends on the pathology. Penetrating keratoplasty (PKP) is used in case of trauma, ulcers, keratoconus or FED. In those cases, the donor cornea is sutured in the patient’s eye [5]. The sutures are removed after approximately one year. The endothelial cell density is highly important and must be more than 2200 cells/mm2. However, if there is an urgency, also corneas with an ECD of less than 2200 cells/mm² are used because at that moment losing sight is a major problem and the ECD is then of secondary importance. Epithelial defects, Descemet folds and stromal scars have an influence on the postoperative recovery and vision, so those elements have to be mentioned before transplantation. Some of the complications of PKP are unpredictable astigmatism, slow rehabilitation and the prolonged use of topical steroids [14]. Astigmatism is caused by deformation of the cornea. Instead of a round corneal curvature, the shape of the cornea is more like a rugby ball. Then, light rays do not focus on the correct place of the retina, which leads to blurry vision. The chronic use of steroids can also result in side effects. Immune rejection may be derived from the endothelial cells of the donor and can lead to graft failure. Rejection is significantly higher in PKP compared to LKP. A possible reason is the presence of donor endothelial cells. However, most LKP techniques also transplant the endothelial layer. Another possible reason is that more tissue (all five corneal layers) is transplanted with PKP. Both reasons can lead to an increased immune reaction. There is also more trauma to the eye with PKP, which can lead to a higher chance of inflammation [14]. Lamellar keratoplasty (LKP) has some advantages compared to PKP, for example faster recovery time, reduced astigmatism and sometimes less endothelial cell loss, but this depends on the type of lamellar graft that is needed [14]. Some techniques have to manipulate the very thin endothelial layer, which can result in more endothelial cell loss. Thus, LKP has also some disadvantages. 5

  11. This technique is technically demanding because a microkeratome is needed to cut the cornea. Consequently, it results in a prolonged surgical time [15]. The cutting process can also be performed in the eye bank if the adequate equipment is available. The vision results of LKP are not superior to PKP [14]. However, the follow-up time of the patient’s vision was significantly different between PKP and LKP in the study of Akanda et al., so this may be a contributor to the results. Lamellar keratoplasty (LKP) has evolved into different methods where Descemet stripping automated endothelial keratoplasty (DSAEK) is most often used (Figure 5E) [5, 15]. This method is used in endothelial failure: the endothelium and Descemet’s membrane of the patient are removed and replaced by the endothelium, Descemet’s membrane and a thin layer of posterior stroma of the donor. It is obvious that the ECD has to be greater than 2200 cells/mm2 with this method since almost only the endothelium is transplanted. DSAEK is a possible treatment for FED, PBK and other dystrophies. There are no sutures required to keep the graft in place after DSAEK. However, the patient has to lie in a flat position because the pressure on the graft ensures that the graft remains at the correct position. Descemet’s membrane endothelial keratoplasty (DMEK) is a technique where only the Descemet’s membrane and the endothelium are transplanted without the stroma-stroma interface as with DSAEK (Figure 5F) [5, 15]. It could be that this technique provides better and faster vision restoration than with DSAEK due to the normal Descemet’s membrane-stroma interface. However, this technique is very complex to perform because manipulation of those thin layers is very difficult and the endothelial layer can be easily damaged, which limits the use of this technique. (Deep) Anterior lamellar keratoplasty ((D)ALK) is used to replace the epithelium, Bowman’s layer and the stroma of the patient (Figure 5D) [15]. Because the endothelium is not transplanted with this method, the ECD of the donor cornea is not that important. Therefore, the inclusion criteria for donor corneas would be broader and there is no risk of endothelial allograft rejection, which is the most important advantage. ALK is not commonly performed due to the need for special instrumentation and the increased surgical time, but it is a possible treatment for superficial corneal scars, dystrophies and thinning of the cornea. Both perforating and lamellar keratoplasty have advantages and disadvantages. PKP is used when the whole cornea is affected and is easy to implement but there is a higher chance of complications compared with the lamellar technique, for example graft failure. LKP is used when only a part of the cornea is affected. This technique is more difficult but there is a reduced risk of complications. An alternative for corneal transplantation is the use of an artificial cornea, also known as a keratoprosthesis [5]. The most popular one is the Boston keratoprosthesis. This can be used in case of multiple failed corneal transplants or ocular surface diseases for which corneal transplants are likely to fail. In some cases, an artificial cornea is successful but many complications can occur. Glaucoma, for example, is a frequent complication where the optic nerve is damaged, often combined with increased pressure in the eye, resulting in vision loss. In general, the Boston keratoprosthesis is made of a front and back plastic plate with in between the donor corneal tissue, as seen in Figure 5C [16]. 6

  12. Figure 5: Different techniques for corneal transplantation. (A): The 5 different layers of cornea, (B): Penetrating keratoplasty, (C): Boston keratoprosthesis, (D): Deep anterior lamellar keratoplasty (DALK), (E): Descemet stripping automated endothelial keratoplasty (DSAEK), (F): Descemet membrane endothelial keratoplasty (DMEK). The purple color shows the layers that are transplanted with the different techniques [1]. 1.3.3 The procedure The cornea banking and transplantation workflow can be divided in a number of major processes: we start with the procurement, the preparation and the evaluation in the clean room. The storage of the cornea is elaborated in a separate section because different storage methods are described in detail. Finally, the implantation of the corneal graft and the follow-up are described. 1.3.3.1 Procurement After careful selection of the deceased donor, the bulbi oculi are preferably procured within 12 hours. After the initial procurement, the bulbi oculi are preserved in a moist chamber (4°C) for maximum 48 hours until preparation in the clean room [7, 17]. Donor selection is needed because the cornea should not transmit diseases. The procurement of the bulbi oculi should be performed as soon as possible in order to reduce the microbiological contamination on the ocular surface. The contamination occurs after death because of the lack of tears and blinking. The cornea is avascular, which means that the exclusion criteria are less severe than for perfused tissues. For example, patients with solid malignancies can still be a cornea donor [18]. Patients with retinoblastoma, hematological malignancies and malignant tumors or swelling of the anterior eye segment are excluded. In Europe, there is mostly no maximum donor age since the endothelium, which is one of the most important parameters for the evaluation of the corneal quality, is always thoroughly studied during the process in the clean room [3, 9]. There are more changes in the cornea with increasing donor age, such as a decline in the ECD, which can lead to a poor outcome for the acceptor [7]. However, after an evaluation with corneas from donors older than or younger than 66 years, corneas from younger donors showed less post-operative endothelial cell loss but there were no differences detected in the final outcome after 5 years [9]. 7

  13. Also due to the aging population, there is evidence enough that a maximum donor age would only limit the already limited number of donor corneas [15]. In general, human leukocyte antigen (HLA) typing and ABO blood group matching is not necessary for corneal transplants. But the ophthalmologist will ask for a cornea with a known HLA type if there is an increased risk of rejection [15, 18]. Blood (serum or plasma) from the donor is tested for human immunodeficiency virus (HIV)-1, 2 (anti-HIV-1, 2 and HIV1 NAT testing), hepatitis B virus (HBV) (HBsAg, anti-HBc and HBV NAT testing), hepatitis C virus (HCV) (anti-HCV and HCV NAT testing) and syphilis in order to avoid transmission of infections [19]. Nucleic acid testing (NAT) reduces the window period, which means that the identification of infections in blood is faster with NAT than with antibodies or RNA from infections, also known as polymerase chain reaction (PCR) testing. Syphilis is a treatable disease but it is more common in people who also have HIV and HCV, so this is the reason why syphilis is also tested. 1.3.3.2 Preparation The corneas are prepared in a laminar flow cabinet in the clean room of the Tissue Bank. A clean room is a controlled environment with a controlled level of contamination [20]. The level of contamination is defined by the number of particles, with a specific particle size, per cubic meter. The air delivered to the clean room always passes through HEPA (high efficiency particulate air) filters. These filters can capture particles that are 0.3 µm and larger. Upon entering the clean room, there is a changing room and from this room, the class D room can be accessed in order to perform the preparing work. In the class D room, there must be less than 1000 particles of 0.3 µm and less than 200 colony forming units (CFU) per m³ air sample. Through a connection shaft, the class C room can be accessed from the class D room. In the class C room, there must be less than 100 particles of 0.3 µm and less than 100 CFU per m³ air sample. This means that the air in class C is more clean than in class D. The eyes are marked to distinguish the superior and inferior part of the eye or cornea. This mark is also used to locate aberrations and to communicate about it with the surgeons. The preparation of the corneas is done eye by eye. Thus, the right eye is first procured (cornea and sclera) and thereafter, the left eye is handled. After removal of the residual conjunctiva, the pachymetry and the endothelial cell density (ECD) are measured with the specular microscope and the morphology of the cornea is investigated with the slit lamp. Descemet folds, epithelial defects and stromal scars can be detected with the slit lamp. Then, the eye is disinfected with Isobetadine to avoid the chance of microbial contamination [3]. After neutralizing the Isobetadine with Na-thiosulphate and transferring the eye in physiological serum, a microbial control of the limbus is taken. The corneoscleral button is removed from the eye and sucrose is applied on the endothelial side. The endothelium is visualized under the light microscope in order to obtain the endothelial cell density afterwards. Sucrose causes a shrinkage of the endothelial cells, caused by the osmotic effect, resulting in increased intercellular spaces and more visible individual endothelial cells [21]. The sucrose is then rinsed off and the corneoscleral button is stored in the appropriate storage medium. The endothelium is always thoroughly studied because a healthy endothelium is required for transplantation [3, 8]. The corneal grafts are appropriately stored in the clean room until release. 1.3.3.3 Evaluation Immediately after preparation, a first microbial test is performed by taking a swab of the limbus. The swab is flushed in a thioglycolate tube and spread out on a blood plate, a chocolate-agar plate and a Sabouraud dextrose agar. Blood and chocolate-agar plates allow the culture of both rapidly growing and fastidious aerobe and anaerobe micro-organisms [22]. 8

  14. Thioglycolate is suitable for the growth of anaerobes. Sabouraud dextrose agars have a high sugar concentration and a low pH. This makes them selective for filamentous fungi and yeasts. The thioglycolate tube is stored for at least 7 days at 34°C. The chocolate-agar plate and the blood plate are stored for at least 3 days at 34°C. The Sabouraud dextrose agar is stored for at least 7 days at 31°C. After at least 48 hours in the storage medium, a sample of the medium is taken for microbial control. A re-evaluation of the cornea, at least 48 hours before transplantation, occurs to guarantee the quality of the cornea. In this step, a vital staining with trypan blue is done and the cornea is also transferred in transport medium [7]. Transport medium has exactly the same composition as the warm storage medium but it is supplemented with dextran. During warm storage, the cornea swells and is not ready for clinical use. Dextran ensures reducing of the corneal swelling. Therefore, the cornea has to be transferred in transport medium. Since the cold storage medium is very similar to the transport medium, the cornea does not swell during cold storage. Consequently, no transfer in transport medium is needed with the cold storage method. After applying trypan blue, the endothelium is visualized under the light microscope in order to detect dead endothelial cells and to count the ECD subsequently. Normally, endothelial cells are impermeable for trypan blue. However, when the cell membrane is damaged, trypan blue can enter the cytoplasm and stain the dead endothelial cells blue [7]. The staining with trypan blue is an inexpensive and easy method to evaluate cell viability. Afterwards, the cornea is transferred in the transport medium and 10 ml storage medium is centrifuged. The pellet is resuspended in thioglycolate and each time one droplet is applied to a Sabouraud dextrose agar and a blood plate. Endothelial examination is performed by light microscopy, which examines the morphology of the endothelium, and by a determination of the ECD using the specular microscope and/or light microscope [3, 23]. Usually, a minimum of 2200 cells/mm2 is required for a corneal transplantation as this is a crucial factor for a good outcome [21]. After transplantation, small endothelial cell loss can occur due to, for example, surgical trauma, rejection of endothelial cells or other complications [8]. This is the reason why a large number of endothelial cells are initially required. The endothelial cell density and vision is also measured 6 months and 1 year postoperatively at the occasion of a follow-up visit. In this way, the progression of the ECD and vision can be measured over time. At least 24 hours after the cornea was re-evaluated and placed in transport medium, sampling of the transport medium for microbiological control is done. The sample is centrifuged and the pellet is resuspended in thioglycolate. A drop of the resuspended pellet is applied on a blood plate, chocolate-agar plate and Sabouraud dextrose agar. In this way, it is possible to read the microbiological controls at the moment of distribution, right before implantation. 1.3.3.4 Storage There are three methods that can ensure the storage of the cornea. To date, there is no method that assures perfect preservation of the cornea. Each storage method has its advantages and disadvantages [7, 24]. i) Cryopreservation The first method is cryopreservation, the cornea may be stored therein presumably unlimited [3, 7]. Studies are still going on to evaluate this storage time. It must be said that this method is rather exceptional and therefore, not as elaborated as the other storage methods. A first rapprochement was the study of Tripathi et al. In this study, corneas were successfully stored in glycerol for 3 months [25]. However, this is insufficient evidence to be able to speak of an unlimited storage time. Cryopreservation is almost never used because there are a number of disadvantages associated with it [3]. In cryopreservation, there is usually endothelial damage, whereas integrity of the endothelium is actually essential [3]. 9

  15. Therefore, in the article of Tripathi et al., cryopreservation was used for DALK (deep anterior lamellar keratoplasty), in which the donor endothelium does not play a role [25]. When the cornea was stored in glycerol at -80°C, this preservation method did have some advantages, such as maintaining transparency and corneal thickness. However, anhydrous glycerol was used in this study, so preservation of the corneal thickness is rather due to osmolarity effects than to the temperature alone. Since the donor endothelium did not play a role in this study, these results cannot be used for corneal preservation. A healthy endothelium is required for PKP and most LKP techniques. ii) Cold or hypothermic storage The second method is the cold storage (2°C-8°C). This method is very simple and effective but microbiological tests are more difficult to perform [26]. Vials allow inspection of the endothelium by specular microscopy and trypan blue, a vital staining, is used to assess endothelial damage (Figure 6) [15, 23]. Normally, bacteria can grow due to the warm temperature in the warm storage method. However, in the cold storage method, bacteria do not get this chance so they are less likely to be noticed. The solution for this is to take a culture and incubate it at the appropriate warm temperature so that the microbiological control is still accurate [27]. The storage medium has to be warmed preoperative to room temperature in order to enhance the decontamination [23]. Antibiotics accumulate in the cornea during storage and become active in the eye after transplantation as the temperature rises [23]. The thickness of the cornea remains approximately the same, which is a major advantage of the cold storage [23]. A storage medium for cold storage is, in general, supplemented with antibiotics, dextran, chondroitin sulphate and additives to prevent corneal swelling and to improve the storage capacity [23]. A B C Figure 6: Storage media for the warm and cold storage method1. (A): Cold storage medium, (B): Warm storage medium, (C): Warm transport medium [28]. The time between the death of the donor and the storage of the cornea is a crucial factor but differs between eye banks [29]. Usually, the maximum time between death and cold storage is 12 - 24 hours [23]. If the time interval between death and preservation is up to 5 hours and 45 minutes, the corneal quality is better than after a longer time interval [9]. However, these results were obtained by donors up to 42 years, which may influence the results. Worldwide, the cold storage method is used a lot but the warm storage method (28°C-37°C) is still more used in Europe [3]. The reason behind this is that the cornea can be stored longer than in the cold storage method. The time between the death of the donor and the warm storage of the cornea is 24 - 48 hours, which is longer than in the cold storage method [23]. During this period, wound healing can occur. This happens by endothelial cells migrating from the limbus to the center of the cornea. However, the main difference between the cold and warm storage method is the difference in total storage time (Figure 7) [23]. 1 http://www.eurobio-cornea.com/en/cornea-range-xsl-352.html 10

  16. In the early years of the cold storage method, storage media like McCarey-Kaufman (M-K) medium were used [7]. With this cheap medium, a storage for 4 - 10 days was assured [23, 30]. Later, a more expensive medium, Optisol-GS, was introduced as storage medium with a storage period for maximum 14 days (Figure 7) [3, 23]. The main purpose of a storage medium is to maintain endothelial cell viability from the moment the donor cornea is collected until the cornea is transplanted [26]. Optisol-GS contains mainly dextran and chondroitin sulfate to maintain hydration of the stroma [3]. Dextran can be seen as an osmotic agent while chondroitin sulfate acts as a membrane stabilizer [15]. Gentamicin and streptomycin are added as antimicrobial agents in order to be efficacious against several bacteria. The cornea can be stored in Optisol-GS for about seven days to maintain optimal quality. Optisol-GS was the most commonly used storage medium in the United States until 2013 [15]. However, with this storage method, a limited endothelium loss can occur during storage compared to the warm storage method [23]. Figure 7: Endothelial cell damage and cell loss in different storage media after staining with trypan blue. The regression formulas for the MK medium, Optisol-GS and the organ culture are respectively: y = 11.8x + 11.6, y = 0.19x + 4.2 and y = 0.11x -1.8 [17]. In developing countries, the M-K medium was initially used because patients were available for surgery within 24 - 48 hours [30]. Due to an exponential increase in tissue collection, longer storage with Optisol-GS was needed. Since Optisol-GS costs ten times more than the M-K medium, they first used M-K medium. If longer storage was required, it could be an option to transfer the cornea from the M-K medium to Optisol-GS. However, this transfer led to a decreased endothelial cell density and the morphology of the endothelial cells changed. This can be due to mechanical cell loss during the transfer. In the end, all tissues retained a final cell density higher than 2200 cells/mm², which means that the tissues could be used for transplantation. Because the morphology of the endothelial cells changed, a re-evaluation before transplantation is needed. Thus, if this method is used with the necessary precautions, it can be a valuable option to store corneas in developing countries. Since the storage time with Optisol-GS is not very long, there is a need for another storage medium with a longer storage time. Cornea Cold can offer a solution here. It ensures an extended storage time (21 days) and the endothelium has a better chance of survival because the loss of endothelial cells is smaller in Cornea Cold than in Optisol-GS [31]. When using the storage medium Cornea Cold, there is also a reduced swelling of the cornea, the morphology is better preserved and loss of transparency is smaller compared to Optisol-GS [31]. 11

  17. The results of this study indicate that Cornea Cold provides a better quality than Optisol-GS. However, there is a conflict of interest for the manufacturer of Cornea Cold, so the results have to be nuanced. Another type of storage medium, Eusol-C, has a higher endothelial cell loss compared to Optisol-GS [32].. Therefore, this storage medium is not suitable for long-term storage. iii) Warm storage or organ culture Organ culture requires more processes in the clean room and an incubator at 30 - 37°C is needed [23]. The tissue culture medium is usually supplemented with fetal or newborn calf serum (FCS), antibiotics and antimycotics [7, 23]. The fetal or newborn calf serum must come from a bovine spongiform encephalopathy (BSE)-free country in order to prevent transmission of BSE [18]. Dehydrating molecules are excluded from the storage medium because they are taken up and accumulated by corneal cells at warm temperatures (30 – 37°C). Due to this exclusion, the cornea swells during warm storage. Before implantation, the swelling has to be reversed by placing the cornea in a transport medium which is supplemented with dextran (Figure 6). The endothelium cannot be visualized with a specular microscope, as with the cold storage medium, so a light microscope is used. Temporary swelling of the intercellular spaces is also required to visualize the endothelial cells. This is mostly done with sucrose and disappears after a couple of minutes. In the warm storage method, Eagle’s minimum essential medium (MEM) is usually used as storage medium [3]. Fetal bovine serum (FBS), penicillin, streptomycin and amphotericin B are added to this medium. The storage time in this medium is about 4 weeks. The storage period of MEM is considerably longer compared to Optisol-GS, which has a storage time of 7 days (Figure 7). However compared to Cornea Cold, which has a storage time of 21 days, this difference is less clear. Researchers have been looking for a different kind of storage medium that does not contain serum [31]. This storage medium, called SFM (serum-free medium), provides a better survival of both the corneal endothelium and epithelium. There is an increased chance of detecting bacteria and fungi during warm storage. In addition, antibiotics in the storage medium are also more effective [3]. After 48 hours, a sample of the storage medium is taken for microbiological screening [23]. Shortly before implantation, the cornea is evaluated with trypan blue as described in 1.3.3.3 Evaluation. Finally, the cornea is transferred in the transport medium and released for transplantation if no contamination or poor quality is observed. iv) Alternative Another possibility to optimize the storage of corneas is to combine both the cold and the warm storage method [33, 34]. Since endothelial loss can occur in the cold storage method, there was searched for a solution to solve this problem [23, 33]. First the cold storage method, with Optisol-GS as used storage medium, is used for the maximum permitted time (7-10 days) and then further storage takes place according to the warm storage method with as storage medium MEM [33]. If endothelial cell loss occurred after cold storage, the quality of the endothelium was “improved” during warm storage. This improvement needs to be examined critically because endothelial cells have no regeneration potential. This means that the alleged endothelial recovery could be due to elimination of affected endothelial cells and elongation of residual, healthy endothelial cells across the affected region. In that case, the number of endothelial cells remains the same because only their size has increased. Therefore, the quality of the endothelium after cold storage is the same as after warm storage since the quality of the endothelium depends on the number of endothelial cells. 12

  18. 1.3.3.5 The implantation of the donor cornea with PKP and some LKP techniques is secured with sutures. The sutures connect the donor cornea with the limbus of the acceptor, where epithelial stem cells reside. These sutures do not have to be removed immediately and can be retained for a while. If the sutures are removed, they are usually removed after 1 to 2 years. With the LKP techniques DSAEK and DMEK, no sutures are needed to keep the corneal graft in place. Within this technique, an air bubble presses against the corneal graft so the patient must stay at a flat position after implantation. Immediately after implantation eye drops with topical corticoids are prescribed for approximately 6 months. Patients get a schedule, that has to be meticulously followed, for the use of the eye drops. The eye drops are needed to avoid rejection and infection. Immediately after the implantation, the vision is not yet restored. This healing process can last up to 1 year in order to achieve the best vision results. The patient normally has a follow-up visit 1 week postoperatively, every month in the first 6 months postoperatively and 1 year postoperatively. The endothelial cell density is counted 6 months and 1 year postoperatively. 1.4 The sclera The sclera, also known as the white of the eye, is the outer layer of the bulbus oculi (figure 1) [35]. It is an opaque and fibrous structure. The sclera touches both the cornea (anterior) and the dura mater (posterior). The transition between the cornea and the stroma is called the limbus (figure 8) [35]. The functions of the sclera consist of protecting the intraocular structures of the eye, maintaining the shape of the eye, resisting internal and external forces and providing attachment for eye muscles. In contrast to the cornea, the sclera is vascularized which means that there are more contra-indications. For example, donors with malignancies cannot donate their sclerae. The sclerae are generally only prepared in general tissue donors, not in donors who are only eligible for corneal donation. However, the availability of scleral tissues is rather high so donor shortage is not a big issue. The sclera is a ‘popular’ tissue for transplantation due to its strength, flexibility, ease of storage and little or no inflammatory reaction [36]. The composition of the sclera itself is not complex. It is mainly composed of collagen and elastic fibers, forming a dense connective tissue [37]. From outside to inside, the sclera consists of the capsule of Tenon (the fascia bulbi), which is mostly not included in the anatomy of the scleral tissue, the episclera, the scleral stroma and the lamina fusca (Figure 8) [37, 38]. Tenon’s capsule is a thin membrane that envelops the complete eye. The episclera consists of loose, fibrous and elastic tissue and is highly vascularized. The scleral stroma is avascular and contains irregular collagen type I fibers, which is unlike the aligned collagen fibers in the corneal stroma (Figure 8). This layer is the most abundant (approximately 95%) in the sclera. Fibrocyte nuclei are found between the bundles of collagen [39]. The lamina fusca is an avascular, thin brown pigmented layer, due to melanocytes, and consists of collagen and elastic fibers at the inner surface of the sclera. This layer is difficult to distinguish because it is extremely thin. Implantation and follow-up 13

  19. C Figure 8: (Panel A and B): A histological microscopic section of a human sclera stained with hematoxylin-eosin. Empty lacunae between fibers are artefacts due to tissue preparation. Panel A: 10x magnification, panel B: 4x magnification [39]. Panel C: The limbus as transition from cornea to sclera (between the solid lines). C: conjunctiva, TC: Tenon's capsule, CM: ciliary muscle [35]. 1.5 Scleral transplantation 1.5.1 Indications The sclera, obtained from deceased donors, can be used in its entirety or it can also be cut in smaller parts, depending on the indication. The primary indication for the use of scleral grafts are iatrogenic surgical complications [37]. Scleral grafts can be used for orbit reconstruction after enucleation [40]. For this procedure, it is not needed to divide the sclera in different parts because the complete sclera is mostly used. A prosthetic, artificial eye can be wrapped in donor sclera and the eye muscles of the patient can be sutured on the scleral graft in order to mimic the normal movement of the eye. Also eye lid reconstruction can be done with scleral grafts. Therefore, the sclera has mostly to be divided in different parts. 1.5.2 Procedure 1.5.2.1 Procurement This procedure is similar to the procedure of corneal transplantation because the bulbi oculi are procured in its totality in the operating theater. The bulbi oculi are preserved in a moist chamber (4°C) until preparation in the clean room. Since the sclerae have a limited blood supply, selection of the donor is critical in order to avoid transmission of diseases [18]. For example, donors with malignancies are excluded. If the donor has no known diseases, the sclerae are further prepared in the clean room. Just as with the cornea, blood of scleral donors is also tested for HIV, HBV, HCV and syphilis [19]. 1.5.2.2 Preparation In the clean room, the eyes are disinfected to avoid the chance of microbial contamination. After cornea preparation, the sclerae are prepared. Everything that remains in the sclera (the vitreous humor, the lens, the iris, the choroidal and retinal tissue) is removed. Special attention is given to the lens. 14

  20. If the lens has been replaced by an intraocular lens implant (IOL), mostly in older people, it must be noted in case of corneal transplantation because the IOL could damage the corneal endothelium. Afterwards, the sclerae are rinsed in physiological serum and finally stored in a 70% ethanol solution. 1.5.2.3 Evaluation Immediately after preparation, microbiological control is executed on the rinsing liquid and on the storage medium. Both rinsing liquid and storage medium are tested by thioglycolate, a blood plate, a chocolate-agar plate and a Sabouraud dextrose agar. Rinsing liquid (10 ml) and storage medium (5 ml) are centrifuged and the pellet is resuspended with 2 ml thioglycolate. Most of the resuspension (1.5 ml) is transferred back into the thioglycolate tube and the remaining liquid is divided over a blood plate, chocolate-agar plate and Sabouraud dextrose agar. 1.5.2.4 Storage At the Bank for ophthalmic tissues of UZ Ghent, the sclerae are stored in a 70% ethanol solution at 4°C for maximally 5 years. This technique is inexpensive, widely used and is the best regarding bacterial, viral and fungal contaminations [37, 41]. Ethanol promotes coagulation of proteins, which explains why fibrocyte nuclei can be seen after storage in ethanol and not after storage in another medium [40]. Ethanol was considered potential toxic for surrounding tissues but this issue was solved by immersing the sclerae before implantation in balanced salt solution (BSS) for 20 minutes. The organization of the collagen fibrils remained the same as before. Other options for storage of sclerae exist [36, 37, 40]. The first option is storage in glycerin at room temperature for up to 3 months. In here, the tissue presented a more regular pattern than in other media (Figure 9) [36, 37]. This suggests that the sclerae stored in glycerin may have an increased integrity and tensile strength, which is an advantage since sclerae are usually used for restoration of the structural integrity [40]. However, resistant microorganisms are still preserved, which is a problem. Glycerin could be an ideal storage medium if the bactericidal and antiviral properties were increased. There must be kept in mind that glycerin is also used for the storage of viruses [41]. Figure 9: Photomicrographs of the sclera with collagen bundles in a regular pattern (40x magnification). (A): Unpreserved sclera (Picrosirius red), (B): Glycerin-preserved sclera (Masson trichrome) [40]. A second option is gamma radiation, at which the risk of disease transmission is decreased due to the sterilization process [37, 41]. This method is also not perfect because the gamma radiation has additional issues. For example, the heat and pressure are not optimal for the sclera. The third option is freezing [36]. The sclera is then frozen at -20°C in an antibiotic solution for maximally 3 months. Once the graft is thawed, it has to be used within 24 hours and cannot be refrozen. So, a continuous low temperature is needed for storage and transportation. This low temperature can cause ice formation, which can result in aberrations in the scleral structure. 15

  21. A last option is freeze-drying, where the sclera is rapidly frozen followed by dehydration of the sclera under high pressure [36, 37]. This results in a biologically stable material that can be stored and transported for longer periods at room temperature compared to other methods. The advantage of this method is the precise assessment of graft size and shape before surgery. However, the additional equipment needed for this storage method can be seen as a disadvantage. If freeze-drying was compared with storage in a 95% ethanol solution at room temperature, the freeze-drying technique showed better results. 1.6 Research questions This master dissertation consists of two separate parts, each with a number of research questions: 1. Retrospective study of the preservation techniques of the cornea The commonly used warm storage method of the cornea is a labor-intensive process compared to the cold storage method. In addition, it seems that the outcomes after cornea transplantation are often better with cold-stored corneas than with warm-stored corneas. We had the intention to perform a prospective analysis in order to compare the cold storage method with the warm method. But due to the new GDPR rules, it was not feasible to perform this prospective study. Therefor the study is limited to a retrospective analysis. Two techniques, PKP and DSAEK, and 2 preservation methods, cold and warm storage, will be compared using a number of parameters, such as donor age, which is up till now not a parameter to determine donor approval, acceptor age, rejection, endothelial cell density (ECD) preoperatively, 6 months postoperatively and 1 year postoperatively. We isolated the following research questions:  Is there a significant (p < 0,05) difference between the warm and cold storage method in PKP and DSAEK based on the endothelial cell densities (ECDs) and the progressions?  Is there a significant (p < 0,05) difference between the means of the ECDs, measured on 3 time points (preoperatively, 6 months and 1 year postoperatively), in the warm and cold storage method?  Is there a significant (p < 0,05) difference between the warm and cold storage method in PKP and DSAEK in 2 groups of donors (< 66 years and ≥ 66 years) based on the ECDs and progressions?  Is there a significant (p < 0,05) difference between the warm and cold storage method in PKP and DSAEK in 3 groups of acceptors, classified according to their age at transplantation, based on the ECDs and progressions?  Is there a significant (p < 0,05) difference between the warm and cold storage method in PKP and DSAEK based on rejection rates? 2. Histological evaluation of sclerae, preserved in a new storage medium Recently, the manufacturer of the storage medium (ethanol 70%) of the sclerae has changed. Ethanol manufactured at UZ Ghent pharmacy has been replaced by ethanol from the manufacturer Fagron. As the use of a new storage medium might influence the sclera, a validation was carried out using macroscopic evaluation, rigidity testing and evaluation of paraffin coupes with hematoxylin-eosin (HE) staining. 16

  22. 2. Materials and methods Since it was important to understand the complete process of corneal transplantation, I was trained to execute the preparation of the cornea and sclera in the clean room. I had to assist the complete process of cornea and sclera procurement in the clean room 3 times and I had to perform 3 times the procedure under supervision, before a declaration of competence was granted. This means that I am allowed to execute the complete procedure by myself. The process exists of different standard operating procedures (SOP’s) as described in the sections 2.1 Cornea and 2.2 Sclera. A register with data of donors and acceptors is completed by staff members of the Tissue Bank, as legally required. This register contains the following information about the donor: donor code, donor age, blood group, place and date of procurement, date and hour of reception, the receiving person and acceptance or exclusion. The following information about the grafts and acceptors is filled in: graft code, ‘cornea’ or ‘sclera’, microbial results, acceptance or rejection of the graft + date, date and hour of release + name of courier, adrema number ( a unique code for each patient), receiving hospital, date of reception and name and birth date of acceptor. 2.1 Cornea 2.1.1 Disinfection of the bulbus oculi (according to SOP_Pp_C_003_06) After procurement of the bulbus oculi, the superior part is marked by the prelevating person in the operating theater or mortuary and transferred into a moist chamber. First, the right bulbus oculi is disinfected and the right cornea is dissected, followed by the left bulbus oculi. The bulbus oculi is transferred from the moist chamber in a petri plate and the residual conjunctiva and mersuture thread is removed from the bulbus oculi. Then, some physiological serum is applied to the bulbus oculi in order to facilitate the visualization with the specular microscope and slit lamp. The specular microscope visualizes the endothelial cell density (ECD) and the pachymetry of the cornea. Detection of Descemet folds, epithelial defects and stromal scars is possible with the slit lamp. Thereafter, the bulbus oculi is placed in Isobetadine for 2 minutes, followed by Na-thiosulphate for 1 minute in order to neutralize the Isobetadine. Finally, the bulbus oculi is transferred into physiological serum for minimum 1 minute and maximum 5 minutes. 2.1.2 Dissection of the cornea (according to SOP_Pp_C_010_05) After disinfection, the bulbus oculi is placed in a petri plate and a microbiological control of the limbus is taken with a marked cotton swab. The corneoscleral button is excised from the bulbus oculi and placed in a sterile plate. Sucrose is applied to the endothelial side. Simultaneously, the sclera is transferred to physiological serum and is prepared afterwards. The endothelium of the cornea is visualized under the light microscope and at least 5 pictures are taken. A digital cell count is then performed, according to SOP_Pp_C_012_02. The sucrose is rinsed off and the corneoscleral button is attached to a rubber cap using a needle and a thread. The rubber cap is then put on the correct storage medium, here CorneaMax for the warm storage and CorneaCold for the cold storage. The cornea is placed in an incubator at 31°C for the warm storage method and in a refrigerator at 4°C for the cold storage method. 17

  23. The cotton swab for the microbiological control of the limbus is flushed in thioglycolate and spread over a blood plate, chocolate-agar plate and Sabouraud dextrose agar. The thioglycolate tube is stored at 34°C for at least 7 days. The blood plate and chocolate-agar plate are stored at 34°C for at least 3 days. The Sabouraud dextrose agar is stored at 31°C for at least 7 days. 2.1.3 Digital cell count endothelial cells (according to SOP_Pp_C_012_02) The digital cell count is performed on the pictures of the endothelium that were made during the dissection of the cornea. The image processing program ImageJ is opened and the correct picture is selected. A grid is placed on the picture and with the cell counter, the number of endothelial cells can be obtained. In one square, all cells within the lines and the cells on or tangent to 2 sides of the square are counted. The cell number is multiplied with 25 to calculate the total cell number. This is done for each picture and the lowest and highest number of endothelial cells of the 5 pictures are not taken into account. Then the average of the remaining numbers is taken to obtain the endothelial cell density (ECD), which is a parameter for corneal quality. 2.1.4 Sampling of the storage medium for microbiological control (according to SOP_Pp_C_011_03) After at least 48 hours in the warm storage medium, 10 ml of storage medium is aspirated with a needle through the rubber cap. The syringe is detached from the needle and the storage medium is transferred to a centrifuge tube. This tube is then sent to the Laboratory of Clinical Biology for the detection of aerobes, anaerobes, yeast and fungi. After at least 7 days, the result is known. Since the cold storage medium has a smaller volume, only 3 ml is aspirated from the storage medium and sent to the Laboratory of Clinical Biology. 2.1.5 Vital staining with trypan blue and transfer to transport medium (according to SOP_Pp_C_006_08) A re-evaluation of the cornea takes place at least 48 hours before transplantation. The cornea is transferred from the storage medium in a petri plate and trypan blue is applied to the endothelial side for 30 seconds. The vital staining is then rinsed off with physiological serum and physiological serum is also applied to the endothelial side. The endothelium is visualized under the light microscope and at least 5 pictures are taken. During warm storage, the cornea is then transferred in transport medium, here CorneaJet, which is not needed during cold storage. Respectively 10 ml or 3 ml of the used warm or cold storage medium is transferred in a centrifuge tube and centrifuged for 8 minutes at 3500 rpm. The supernatant is poured off and the pellet is resuspended in 2 ml thioglycolate. From this solution, 1 ml is transferred back into the original thioglycolate tube. Each time 1 droplet of the resuspended pellet is applied to the Sabouraud dextrose agar and the blood plate. The thioglycolate tube and the blood plate are stored at 34°C for at least 7 and 3 days respectively. The Sabouraud dextrose agar is stored at 31°C for at least 7 days. Afterwards, a digital cell count is executed on the obtained pictures of the endothelium, as described in 2.1.3. If the endothelial cell density (ECD) is less than 1500 cells/mm², the cornea is rejected. If the ECD is between 1500 and 2200 cells/mm², the cornea is rejected if there are more than 2 corneas in stock. If there are less than 2 corneas in stock, the cornea is released for urgency. If the ECD is greater than 2200 cells/mm², the cornea is released. 18

  24. 2.1.6 Sampling of the transport medium for microbiological control (according to SOP_Pp_C_014) Sampling of the transport medium occurs 24 hours after the cornea is placed in the transport medium. This step only occurs during the warm storage method. The rubber cap is pierced with a needle and 10 ml transport medium is centrifuged for 8 minutes at 3500 rpm. The supernatant is poured off and the pellet is resuspended in 2 ml thioglycolate. From this suspension, 1.5 ml is brought back into the thioglycolate tube and the remaining 0.5 ml is divided by applying each time one droplet on a blood plate, chocolate-agar plate and Sabouraud dextrose agar. The thioglycolate tube is stored at 34°C for at least 7 days. The blood plate and chocolate-agar plate are stored at the same temperature as the thioglycolate tube for at least 3 days. The Sabouraud dextrose agar is stored at 31°C for at least 7 days. 2.2 Corneal transplantations: retrospective analysis 2.2.1 Data collection After receiving approval of the ethical committee (Appendix 7.1), data for the retrospective study was added to data that had already been obtained for the master dissertation of Matthias Claeys in 2016-2017 [42]. The data deals with corneal transplantations in both UZ Ghent and AZ Maria Middelares. In UZ Ghent, warm-stored corneas from the Tissue Bank are used. In AZ Maria Middelares, both warm-stored and cold-stored corneas are used from respectively the Tissue Bank of UZ Ghent and the Cornea Bank of CHU Liege. Corneas obtained from other Eye Banks (Amsterdam, Antwerp and Italy) are not included in this study. Data from corneal transplantations in UZ Ghent is collected from August 2006 until September 2012. Data collection of corneal transplantations in AZ Maria Middelares is performed from December 2012 until August 2018. In order to fulfill the GDPR rules, no names of patients and donors are known by the student. Data from corneal grafts is obtained from paper documents for grafts of CHU Liege and also from a database for grafts of UZ Ghent. The following parameters of corneal grafts are obtained: place of procurement (Ghent/Liege), graft number, date of storage, storage time, storage method (cold/warm), donor age and endothelial cell density after storage (ECD after storage). The following data from patients undergoing a corneal transplantation is collected: date of birth, age at transplantation, gender (m/f), left or right eye, date of operation, indication, type of operation (DSAEK/PKP), rejection (yes/no), in case of rejection: new graft number and date of new graft, endothelial cell density after 6 months (ECD 6m) and endothelial cell density after 1 year (ECD 1y). After 1 year, the patient is referred back to the treating ophthalmologist so the follow-up ends in the hospital. Implementation of all data resulted in data of 703 cornea transplantations. Only the operation techniques DSAEK and PKP are retained. Other techniques are excluded in this study. In total, 361 cornea transplantations are executed with DSAEK and 342 with the PKP technique. If we divide the 703 cornea transplantations in 2 groups according to the storage method, then 404 corneas are stored according to the warm method and 299 corneas according to the cold method. Endothelial cell density (ECD) is measured after storage (so before implantation, ECD after storage), 6 months after operation (ECD 6m) and 1 year after operation (ECD 1y). Both absolute and relative progressions can be calculated from these measurements. The following measurements are calculated: progression in initial endothelial cell density after storage and 6 months postoperatively (ECD after storage - 6m), progression in initial endothelial cell density after storage and 1 year postoperatively (ECD after storage - 1y) and progression in endothelial cell density 6 months postoperatively and 1 year postoperatively (ECD 6m - 1y). Attention has to be drawn to the fact that used corneas always have an ECD of minimum 2200 cells/mm². This is because a minimum of 2200 cells/mm² is imposed to use corneas for corneal transplantation. In some cases, corneas with an ECD of less than 2200 cells/mm² is released in case of urgency. This is because, at that specific moment, the ECD is less important compared to losing sight. 19

  25. 2.2.2 Study design A retrospective, observational study is used for the analysis of the database because the samples and the parameters for the study are set after implementation of the data in the database and no attempts are made to affect the outcome. A group of individuals is mainly similar but differ by a certain characteristic, here the storage method (warm or cold) or the operation technique (DSAEK or PKP). 2.2.3 Data analysis Analysis of all data is executed with the online tool SPPS 25.0 Statistics. The selected parameters from the complete Excel datasheet that are used for data-analysis are the following: acceptor age at transplantation, type operation (DSAEK/PKP), rejection (yes/no), donor age, storage method, ECD after storage (ECD after storage), ECD 6 months postoperatively (ECD 6m), ECD 1 year postoperatively (ECD 1y), absolute and relative progression of ECD after storage and 6 months postoperatively, after storage and 1 year postoperatively and 6 months and 1 year postoperatively. A frequency table is made of parameters that can summarize the dataset in Excel. First the dataset is split, based upon operation technique (PKP/DSAEK) in order to compare the warm and cold storage method afterwards. The following parameters are analyzed: number of corneas, donor age, acceptor age, ECD after storage, ECD 6 months postoperatively and ECD 1 year postoperatively. For each parameter, the mean and standard deviation are calculated. In the variable ‘ECD after storage’, 4 corneas are found with an initial ECD of 0 cells/mm². Since this initial density is not possible for a corneal transplantation, these 4 results were excluded and changed into system missing. This can be done in SPSS via transform, recode into same variables. The first step in the statistical analysis is to evaluate if the variables concerning the endothelial cell densities are normally distributed. This can be done in SPSS via analyze, descriptive statistics and explore. The normality is checked by histograms, boxplots and quantile-quantile (Q-Q) plots because visual interpretation is preferred above the Shapiro-Wilk and Kolmogorov- Smirnov test. Testing the normal distribution is necessary to know which tests must be performed afterwards. The first test performed is to see if there is a significant difference (p-value < 0.05) between the warm and cold storage method in PKP and DSAEK, based on the endothelial cell densities (ECD after storage, 6m and 1y) and the calculated progressions between the endothelial cell densities. The cell densities and its progressions are continuous variables and the storage method is a categorical variable (warm/cold). Because the data are also normally distributed, an Independent-Samples T Test is executed. This test can be found in SPSS via analyze, compare means and Independent-Samples T Test and has to be executed two times: once for the PKP group and once for the DSAEK group. Because the endothelial cell densities are normally distributed and measured on 3 different time points (after storage, 6 months and 1 year postoperatively), a Paired-Samples T Test can be executed. With this test, measurements on 2 time points can be compared: ECD after storage – ECD 6m, ECD after storage – ECD 1y and ECD 6m – ECD 1y. A significant difference (p-value < 0.05) between the mean of the endothelial cell densities is sought. This test is done in 2 groups: warm and cold storage. The Paired-Samples T Test can be found via analyze, compare means and Paired-Samples T Test. After the Paired-Samples T Test, a mean profile plot can be composed because the data are longitudinal. This means that the same variables have repeated observations. This can be done in SPSS via graphs, chart builder and then the option ‘line’. 20

  26. The third test is to see if there is a significant difference (p-value < 0,05) between the warm and cold storage method in PKP and DSAEK in 2 groups of donors (< 66 years and ≥ 66 years), based on the ECDs and progressions. To date, there is no maximum donor age in Europe but some countries take 66 years as limit donor age. Also in literature, the limit of 66 year is often used. This is the reason why the donor age is divided in 2 categorical groups: < 66 years and ≥ 66 years. The cell densities and progressions are continuous variables and the storage method is a categorical variable (warm/cold). Given that the endothelial cell densities and progressions are normally distributed, an Independent-Samples T Test is executed. This test has to be done for the 2 donor groups (< 66 years and ≥ 66 years) in both the PKP and DSAEK group. The acceptors in both the PKP and DSAEK group are divided in three groups, based on their age at transplantation. In each group, approximately the same number of acceptors is included. In the PKP group, the first group is 16 years old until 49 years old (n = 113), the second group is 50 years old until 68 years old (n = 116) and the last group is 68 years old until 94 years old (n = 113). In the DSAEK group, the first group of acceptors is 30 years old until 67 years old (n = 116), the second group is 68 years old until 74 years old (n = 113) and the last group is 75 years old until 96 years old (n = 132). A significant (p < 0,05) difference between the warm and cold storage method in PKP and DSAEK is sought in the three groups of acceptors, based on the ECDs and progressions. The ECDs and progressions are continuous variables and normally distributed. The storage method is a categorical variable, which means that an Independent-Samples T Test is executed in SPSS. Finally, a significant difference (p-value < 0.,05) between cold and warm storage in the PKP and DSAEK group is sought, based on rejection (yes/no). Because the variables are categorical and each variable has two categorical, independent groups (cold/warm and yes/no), a Chi-square test is executed. This test can be found in SPSS via analyze, descriptive statistics and crosstabs. 2.3 Sclera Three sclerae from 3 different donors are used for research: 18S036, 18S038L and 18S057. The two ethanol solutions, both used for scleral storage, have a different composition. The ethanol from UZ Ghent pharmacy is composed of ethanol (70%, CAS 64-17-5), purified water, isopropylalcohol (2%, CAS 67-63-0) and methyletylcetone (2%, CAS 78-93-3). The ethanol from Fagron has the following composition: ethanol (70%, CAS 64-17-5), propan-2-ol (0.7%, CAS 67-63-0), butan-2-on (0.7%, CAS 78-93-3), denatonium benzoate (0.0007%, CAS 3734- 33-6) and purified water. Both ethanol solutions will be compared, based on a hematoxylin- eosin staining of the 3 scleral grafts. First, the grafts have to be prepared in the clean room and thereafter, sections are made for histological stains 2.3.1 Preparation scleral grafts (according to SOP_Pp_S_002_07) The preparation of scleral grafts is associated with the preparation of corneal grafts (according to SOP_Pp_C_003_06 and SOP_Pp_C_010_05). The bulbi oculi are obtained from the operating theater in a moist chamber (4°C) and transferred to the clean room. The bulbi oculi are stripped from resting conjunctiva or sutures. Thereafter, the bulbi oculi are disinfected for 2 minutes in Isobetadine. In order to neutralize the Isobetadine, the bulbi oculi are placed for 1 minute in Na-thiosulphate. Finally, the bulbi oculi are transferred in physiological serum. The sclera is prepared after excision and preparation of the cornea. During corneal preparation, the sclera is preserved in its entirety in physiological serum. 21

  27. The sclera is removed from the physiological serum and emptied by removing the vitreous humor, the lens and the iris. The sclera is two times separately rinsed in 20 ml physiological serum. The second rinsing liquid is kept for microbiological control. The last step is to transfer the sclera in 40 ml 70% ethanol, also known as the storage medium. For this in vitro analysis, both ethanol from UZ Ghent pharmacy and ethanol from Fagron are used. Each sclera is divided into 6 pieces. Three pieces are stored in ethanol from UZ Ghent pharmacy and the other 3 pieces are stored in ethanol from Fagron. Both storage media are preserved at 4°C in the refrigerator. Microbiological control is performed on both the second rinsing liquid and on the storage medium. Both rinsing liquid (10 ml physiological serum) and storage medium (5 ml 70% ethanol) are centrifuged for 8 minutes at 3500 rpm. The supernatant is poured off and the pellet is resuspended in 2 ml thioglycolate. From this suspension, 1.5 ml is transferred back in the original thioglycolate tube and each time 1 drop of the resuspended pellet is applied on a blood plate, chocolate-agar plate and Sabouraud dextrose agar. This procedure has to be done for the rinsing liquid and the storage medium. The thioglycolate tubes are stored at 34°C for minimal 7 days, the blood plate and chocolate-agar plate for minimal 3 days. The Sabouraud dextrose agar is incubated at 31°C for minimal 7 days. 2.3.2 Hematoxylin-eosin staining scleral grafts Before the hematoxylin-eosin staining process is fulfilled, the sclerae are first macroscopically investigated by Dr. Roels after storage in the two different storage media (UZ Ghent pharmacy and Fagron). In addition, the rigidity and flexibility are tested by gently pulling and moving the sclerae with forceps. The histological evaluation with the hematoxylin-eosin staining is performed at the Morphology Lab in UZ Ghent. Therefore, the sclerae have to be transferred from the Tissue Bank to the Morphology Lab and then back to the Tissue bank. Since traceability of tissues always have to be fulfilled, there has to be a form and a register where each transfer is signed with date, hour and other important data in order to know at every moment where the tissue is kept. During this master thesis such a form and register, that can be used at different departments, have been created (Appendix 7.2). In the form the following information have to be filled in: type and serial number of the graft, date and hour of departure on the Tissue Bank + signature, which editing is done on which department, date and hour of entry in and exit out of the specific department + signature, date and hour of return on the Tissue Bank + signature. The register is divided in the following sections: type material/graft, donor number, serial number graft, place of prelevation, date + hour reception, person receiving, date + hour storage, start device storage, date + hour delivery, person delivery and destination. Because the communication is in Dutch, the form and register are also in Dutch. The register and the form are translated in English for this master thesis. Every time the tissues were transferred between the two departments, this form and register were completed. After storage at 4°C in the 2 different storage media (Fagron and UZ Ghent pharmacy ethanol), each part of the sclera is put into formalin for fixation and stored in the refrigerator until further processing. The fixation is needed to stop decay processes of the material. The structure itself is preserved but is somewhat hardened. However, cutting the sclera is still not possible so paraffin has to be used to make the cutting process easier. The disadvantage of paraffin is its impossibility to solve in water. To overcome this problem, the water is expelled by alcohol, also known as dehydration. The sclerae are, in this order, put in different alcohol solutions (50%, 70%, 85%, 96% ethanol and 100% isopropanol). The sclerae are then imbedded in toluene, which is soluble in paraffin, to expel the alcohol. 22

  28. The next step is to embed and orientate the tissues in warmed paraffin. Then the tissues are placed, embedded in paraffin, in a refrigerator. Afterwards, the tissues are placed on a small wooden block and put back in the refrigerator for at least 1 hour. Different slices of 5 µm from each part of the sclerae are cut with a microtome and placed on a slide. During this transfer, the slices can deform or tear, causing artefacts in the histological slide. The slides are put in an oven overnight and the next step is to eliminate the paraffin from the slides because most stains are aqueous. This step is called hydration, the opposite of dehydration at the beginning of this process. This also means that the steps are executed in reversed direction. In this process, the hematoxylin-eosin staining is also executed and the final step is to place a thin glass over the slide with mounting medium. Finally, the tissue preparations can be visualized with a Nikon Eclipse 90i light microscope with different magnifications and low-resolution images of the coupes can be transferred on a computer. The names and descriptions of the slides are hidden to have a double-blind study. The laboratory technician assigned a number to each slide and made a list where each number was followed by the name and the description of the original slide. This list was only viewed at the end for the evaluation of the slides. In order to evaluate the histological sections of the scleral grafts, a number of parameters have been identified that are present in the sclera (Table 1). The parameters are: elastic fibers and blood vessels for the episclera, collagen and fibrocyte nuclei for the stroma and the presence of a thin, brown layer for the lamina fusca. Each parameter is scored by +++, ++, +, +/- or - and stands respectively for excellent presence, very good visible, good/visible presence of the parameter, somewhat present but not clearly distinguishable and not visible/present. Table 1: Parameters to score the histological slides of the sclera Scleral layers +++ ++ + +/- - Elastic fibers Blood vessels Collagen density Fibrocyte nuclei Episclera Stroma Lamina fusca Thin, brown layer Two independent researchers examined the slides, characterized by numbers and letters, under a light microscope with magnification 10x and gave scores based on the table of parameters (Table 1). Afterwards, the numbers and letters were decoded in order to compare the UZ Ghent pharmacy ethanol and the Fagron ethanol solution with each other. Per sclera, the UZ Ghent pharmacy ethanol and the Fagron ethanol are compared with different slices of the tissue. 23

  29. 3. Results 3.1 Corneal transplantation: retrospective analysis 3.1.1 Mean values database Data of the warm and cold storage method in both operation techniques are given in the 2 tables below (Table 2 and 3). These tables give a brief summary of some parameters in the Excel datasheet in order to have a general overview of the corneal transplantations included in the analysis. In the PKP group of the database, more corneas are stored according to the warm storage method (258) than to the cold storage method (84) (Table 2). The mean donor age in the cold storage group is 57,20 years compared to the warm storage group where the donor age is 53,09 years. The mean acceptor age in the warm storage group is 58,20 years and 57,93 years in the cold storage group. The mean ECD after warm storage is 2523,68 cells/mm² compared to the mean ECD after cold storage with 2600,78 cells/mm². After 6 months, the cold storage method has a mean of 2077,43 cells/mm², while the warm storage method has a mean of 1759,27 cells/mm² 6 months postoperatively. After 1 year, the difference is not disappeared. The cold storage method has a mean of 1988,91 cells/mm² compared to 1589,19 cells/mm² 1 year postoperatively after warm storage. Table 2: Frequencies of some parameters in the PKP group for both warm and cold storage groups. PKP Warm storage Cold storage Number of corneas 258 84 Donor age in years (mean ± SD) 53,09 ± 16,68 57,20 ± 14,95 Acceptor age in years (mean ± SD) 58,20 ± 17,02 57,93 ± 15,93 ECD after storage in cells/mm² (mean ± SD) 2523,68 ± 290,13 2600,78 ± 299,37 ECD 6 months in cells/mm² (mean ± SD) 1759,27 ± 568,63 2077,43 ± 558,52 ECD 1 year in cells/mm² (mean ± SD) 1589,19 ± 544,81 1988,91 ± 544,13 In the DSAEK group of the database, more corneas are stored with the cold method (215) compared with the warm method (146) (Table 3). The mean donor age differs 2,25 years between the warm and cold storage method, respectively 59,48 and 61,73 years. The mean acceptor age in the warm storage group is 70,40 years and 70,85 years in the cold storage group. The mean ECD after warm storage is 2551,23 cells/mm² compared to 2598,07 cells/mm² after cold storage. The difference between the mean ECD after 6 months is more visible. The mean ECD 6 months postoperatively after warm storage is 1127,81 cells/mm² compared to 1652,11 cells/mm² after cold storage. After 1 year, the cold storage method has still a higher mean ECD (1578,72 cells/mm²) than the warm storage method (1127,33 cells/mm²). 24

  30. Table 3: Frequencies of some parameters in the DSAEK group for both warm and cold storage groups. DSAEK Warm storage Cold storage Number of corneas 146 215 Donor age in years (mean ± SD) 59,48 ± 15,79 61,73 ± 14,83 Acceptor age in years (mean ± SD) 70,40 ± 16,44 70,85 ± 15,79 ECD after storage in cells/mm² (mean ± SD) 2551,23 ± 281,26 2598, 07 ± 262,78 ECD 6 months in cells/mm² (mean ± SD) 1127,81 ± 563,79 1652,11 ± 553,86 ECD 1 year in cells/mm² (mean ± SD) 1127,33 ± 539,37 1578,72 ± 541,74 3.1.2 Endothelial cell densities (ECDs) 3.1.2.1 PKP The first step in the analysis is checking if the variables are normally distributed. The histogram, boxplot and Q-Q plot of each variable is visualized (Appendix 7.3.1.1). Below the graphs are given for one variable (ECD after storage) in the PKP group (Figure 10 and 11). The histograms show a symmetric pattern, also known as Bell-shaped, which is characteristic for a normal distribution. Outliers are detected because some corneas are released for urgency and then the ECD may be lower. In the Q-Q plot, the data has to be distributed along the diagonal line in order to be considered as normally distributed, which is here the case. The outliers are seen as individual points and not laying on the diagonal. The boxplot, in which the data is divided in quartiles, is also symmetric, which means that the data is considered to be normally distributed. Outliers are also shown as individual points. Some outliers can be detected, especially in the ECD after storage and in the absolute and relative progression of the ECD 6m – 1y. However, it can be argued that the mean of the variables is each time coming from a normal distribution. This is based on the central limit theorem (CLT), which states that if the dataset is sufficiently large, the mean is from a normal distribution. For this complete statistical analysis, the central limit theorem is an important factor that can be extended over the entire dataset. Figure 10: A histogram to check the normal distribution of the variable ‘ECD after storage’. Corneas (336) in this variable have a mean of 2533,43 cells/mm² and a standard deviation of 289,85. The data is normally distributed if the histogram is Bell-shaped. Some outliers are found on the left of the distribution and one on the right, this can be explained by the fact that corneas at urgency can have less than 2200 cells/mm². 25

  31. Figure 11: On the left, a Q-Q plot to check the normal distribution of the variable ‘ECD after storage’. The points have to be distributed along the diagonal line in order to have a normal distribution. Outliers can be seen as individual points. On the right, a boxplot representing the data in quartiles. These quartiles have to be symmetrically divided in order to have a normal distribution. Outliers are detected as individual points. Since the data are normally distributed, an Independent-Samples T Test is executed in order to see if there is a significant difference (p < 0.05) between the warm and cold storage method in PKP, based on the endothelial cell densities (ECDs) and progressions (Table 4). The performance of this test can be found in the Appendix 7.3.2.1. The Levene’s Test has to be executed first in order to know if the variances are equal or not. This can be done by checking the significance level of the Levene’s test. If the significance level is below 0,05; equal variances are not assumed, equal variances are assumed if the significance level is above 0,05. Dependent from the Levene’s Test, the associated row of the variances has to be followed in order to find the significance level of the Independent-Samples T Test. In the ECD after storage, outliers can be detected because in case of urgency, also corneas with less than 2200 cells/mm² are released. The progression 6m – 1y contains also outliers. The last 6 months, corneas can both lose or increase endothelial cells, as seen on the graph. All ECDs and progressions have significant (p < 0,05) results, except for the progression 6 months to 1 year (t(190,68) = -2,48; t(232) = -4,18; t(193) = -4,52; t(228) = -3,07; t(228) = -3,39; t(191) = - 3,62; t(191) = -3,90). From these results, it can be seen that corneas in the PKP group stored according to the cold method have higher ECDs than those stored according to the warm method. Table 4: Summary of the Independent-Samples T Test that compares the warm and cold storage in the PKP group, based on the ECDs and progressions. Significant (p < 0,05) results, indicated in red, are in favor of the cold storage method. PKP Cold storage Mean ± SD (N) 2600,78 ± 213,92 (79) ECD Warm storage Mean ± SD (N) 2523,68 ± 314,83 (257) p-value (cells/mm²) After storage 0,014 After 6 months 1759,27 ± 510,64 (176) 2077,43 ± 476,29 (58) < 0,001 After 1 year 1589,19 ± 522,24 (150) Absolute (cells/mm²) -781,86 ± 467,36 (176) 1988,91 ± 516,33 (45) Absolute (cells/mm²) -561,30 ± 442,72 (54) < 0,001 Relative (%) Relative (%) Progressions Absolute Relative After storage - 6 months -30,87 ± 18,16 (176) -21,42 ± 17,09 (54) 0,002 0,001 After storage - 1 year -953,68 ± 485,54 (150) -37,64 ± 19,21 (150) -648,21 ± 497,00 (43) -24,72 ± 18,95 (43) < 0,001 < 0,001 6 months - 1 year -195,52 ± 326,62 (132) -10,65 ± 21,73 (132) -170,59 ± 325,32 (37) -8,40 ± 16,30 (37) 0,682 0,559 26

  32. 3.1.2.2 Corneal transplantations operated via the DSAEK technique have endothelial cell densities and progressions with a normal distribution. This is established by the central limit theorem, histograms, boxplots and Q-Q plots (Appendix 7.3.1.2). In the variable ECD after storage, outliers can be detected with a very high ECD. The absolute progression 6m – 1y showed outliers whereby the ECD decreased a lot the last 6 months. The relative progression 6m - 1y had outliers on both sides. So the last 6 months, the loss or gain in cells can be large. Within the DSAEK group, an Independent-Samples T Test can be conducted to search for a significant difference (p < 0.05) between the cold and warm storage method, based on the endothelial cell densities and the progressions (Table 5 + Appendix 7.3.2.2). After executing the Levene’s test, the significance level of the Independent-Samples T Test could be obtained (Table 5). All the ECDs and progressions, except the progression 6 months to 1 year postoperatively, have significance levels under 0,05 (t(277,52) = -4,00; t(275) = -8,98; t(229) = -7,16; t(262) = -7,08; t(262) = -8,08; t(220) = -5,59; t(220) = -6,43). In the DSAEK group, cold storage results in higher ECD levels compared to warm storage. Table 5: Summary of the Independent-Samples T Test in the DSAEK group to compare warm and cold storage based on ECDs. Significant (p < 0,05) results, indicated in red, are in favor of the cold storage method. DSAEK DSAEK ECD Warm storage Mean ± SD (N) Cold storage Mean ± SD (N) p-value (cells/mm²) After storage 2551,23 ± 238,11 (142) 2650,56 ± 207,53 (198) < 0,001 After 6 months 1127,81 ± 423,76 (100) 1652,11 ± 489,65 (177) < 0,001 After 1 year 1127,33 ± 452,88 (86) 1578,72 ± 469,55 (145) < 0,001 Absolute (cells/mm²) -1426,09 ± 448,42 (98) -1432,05 ± 474,80 (85) -114,93 ± 195,73 (71) Absolute (cells/mm²) -996,40 ± 492,69 (166) -1062,20 ± 482,01 (137) -86,06 ± 291,62 (133) Relative (%) Relative (%) Progressions Absolute Relative After storage - 6 months After storage - 1 year 6 months - 1 year -55,71 ± 16,39 (98) -55,85 ± 17,84 (85) -7,58 ± 18,09 (71) -37,55 ± 18,36 (166) -40,03 ± 17,79 (137) -3,23 ± 16,28 (133) < 0,001 < 0,001 < 0,001 < 0,001 0,455 0,082 3.1.2.3 With a Paired-Samples T Test, a significant difference (p < 0.05) between the means of the ECDs in the warm storage group can be examined at 3 different time points (after storage, 6 months and 1 year postoperatively) (Table 6 + Appendix 7.3.3.1). As seen in Table 6, the means are significantly different from each other (t(273) = 30,23; t(234) = 32,41; t(202) = 8,23; p < 0,0001). In the first and second pair, the mean of the ECD after storage is significantly higher than the ECD after 6 months and 1 year. If the ECD after 6 months and 1 year is compared, then the ECD after 6 months has significantly more cells than after 1 year. On average, the ECD after storage has 1012 cells/mm² more than the ECD 6 months postoperatively and 1127 cells/mm² more than the ECD 1 year postoperatively. The mean ECD after 6 months has 167 cells/mm² more than after 1 year. After one year, the ECD decreased 44%. This can also be clearly visualized in the mean profile plot, in which at each time point the mean ECD with the error bars is composed (Figure 13). Paired-Samples T Test: warm storage 27

  33. Table 6: Summary of the Paired-Samples T Test in the warm storage group where the ECD at 3 time points is measured and compared, based on the means of the ECDs. Significant (p < 0,05) values, indicated in red, are in favor of the earliest measurements. Warm storage Mean ± SD (N) 2546,96 ± 268,01 (274) - 1534,68 ± 568,28 (274) 2549,35 ± 268,66 (235) - 1422,64 ± 545,22 (235) 1576,98 ± 547,70 (203) - 1409,65 ± 542,21 (203) Pairs of ECDs (cells/mm²) After storage - 6 months After storage - 1 year 6 months - 1 year p-value < 0,001 < 0,001 < 0,001 Figure 12: The mean profile plot of the means of the ECDs after warm storage. The ECDs are measured on 3 time points: after storage, 6 months postoperatively and 1 year postoperatively, as seen on the x- axis. From these measurements, the mean is taken. The mean ECD in cells/mm² can be found on the y-axis. The differences between the time points are significant (p < 0,001), which means that a significant (p < 0,05) decrease in mean ECD occurs over time. 3.1.2.4 Paired-Samples T Test: cold storage The Paired-Samples T Test showed significant results (p < 0,001) in the cold storage group between the means of the ECDs at the 3 time points (after storage, 6 months and 1 year postoperatively) (t(219) = 25,61; t(179) = 25,07; t(169) = 4,54) (Table 7 + Appendix 7.3.3.2). The first 2 pairs of ECDs show significantly higher means of ECDs after storage compared to the means of the ECDs after 6 months and 1 year. When comparing the means of the ECDs after 6 months and 1 year, the mean of the ECD after 6 months has a significantly higher ECD than the mean of the ECD after 1 year. The mean of the ECD after storage has on average 890 cells/mm² and 963 cells/mm² more than respectively the mean of the ECD after 6 months and 1 year. The mean of the ECD after 6 months has on average approximately 104 cells/mm² more than the mean of the ECD after 1 year. After 1 year, the ECD decreased 36%. The mean profile plot shows the mean of the ECDs at the 3 time points, each time with the associated error bars (Figure 14). 28

  34. Table 7: Summary of the Paired-Samples T Test in the cold storage group. The mean ECD is compared at 3 time points (after storage, 6 months and 1 year postoperatively). Significant (p < 0,05) results, indicated in red, are found and are in favor of the earliest measurements. Cold storage Mean ± SD (N) 2646,75 ± 213,09 (220) - 1757,14 ± 526,12 (220) 2640,28 ± 213,46 (180) - 1676,97 ± 512,78 (180) 1752,92 ± 528,21 (170) - 1648,46 ± 516,41 (170) Pairs of ECDs (cells/mm²) After storage - 6 months After storage - 1 year 6 months - 1 year p-value < 0,001 < 0,001 < 0,001 Figure 14: Mean profile plot of the mean ECDs after cold storage. The ECDs are measured on 3 different time points, as seen on the x-axis. From these measurements, the mean is taken and compared to each other. On the y-axis, the mean ECD in cells/mm² can be found. The differences between the time points are significant (p < 0,05), which means that there is significantly decrease in ECD over time. 3.1.3 Breakdown by donor age 3.1.3.1 PKP With the Independent-Samples T Test, a significant (p < 0,05) difference between the warm and cold storage method is sought in donors younger than 66 years, on the one hand, and in donors who are 66 years old and older, on the other hand (Table 8 + Appendix 7.3.4.1). First, the Levene’s test is executed. If the significance level is above 0.05, equal variances are assumed and the associated row is taken into consideration. From this row, the final significance level can be obtained. In the PKP group with donors younger than 66 years, the ECDs after 6 months and 1 year and the associated progressions show significant (p < 0,05) results (t(165) = -4,03; t(138) = -4,02; t(165) = -3,36; t(165) = -3,62; t(138) = -3,57; t(138) = - 3,70). The significance level of the ECD after storage is 0,05; which means that this ECD is practically significant. All the significant results are in favor of the cold storage method. The last 6 months of follow-up, so the progression 6m – 1y, is not significant (p = 0,808; p = 0,847). This means that the last 6 months of follow-up no difference is detected between the warm and cold storage. 29

  35. In donors ≥ 66 years old, no significant results are found in all ECDs and progressions. This means that in older donors approximately the same postoperative outcomes are obtained, independent of the storage method. Thus, only younger donors (< 66 years) show significant (p < 0,05) differences in favor of the cold storage. Table 8: Summary of the Independent-Samples T Test in the PKP group in order to compare the warm and cold storage method in donors < 66 years old and ≥ 66 years old, based on the ECDs and progressions. Significant (p < 0,05) results, indicated in red, are in favor of the cold storage method. PKP Donor age Warm storage Mean ECD ± SD (N) Cold storage Mean ECD ± SD (N) ECD p-value After storage 2569,14 ± 305,81 (193) 2642,81 ± 220,12 (54) 0,050 After 6 months 1792,03 ± 530,85 (130) 2178,70 ± 452,71 (37) < 0,001 After 1 year 1634,42 ± 545,15 (110) 2081,57 ± 523,23 (30) < 0,001 < 66 years Absolute (cells/mm²) -787,71± 486,47 (130) -957,38 ± 506,09 (110) -191,24 ± 318,94 (98) Warm storage Mean ECD ± SD (N) Relative (%) -0,31 ± 0,19 (130) -0,37 ± 0,20 (110) -0,11 ± 0,20 (98) Absolute (cells/mm²) -492,19 ± 417,48 (37) -588,07 ± 489,08 (30) -209,38 ± 356,29 (24) Relative (%) Progressions Absolute Relative After storage - 6 months After storage - 1 year 6 months - 1 year -0,18 ± 0,16 (37) -0,22 ± 0,19 (30) -0,10 ± 0,17 (24) 0,001 < 0,001 < 0,001 < 0,001 0,808 0,847 Cold storage Mean ECD ± SD (N) ECD p-value After storage 2388,78 ± 305,75 (63) 2510,00 ± 170,78 (25) 0,065 After 6 months 1666,70 ± 440,82 (46) 1829,47 ± 469,20 (17) 0,206 After 1 year 1464,80 ± 435,60 (40) 1713,00 ± 428,26 (13) 0,079 Absolute (cells/mm²) -765,33 ± 413,11 (46) -943,51 ± 429,74 (40) -207,82 ± 352,55 (34) Relative (%) -0,32 ± 0,17 (46) -0,39 ± 0,17 (40) -0,10 ± 0,26 (34) Absolute (cells/mm²) -711,71 ± 471,21 (17) -787,00 ± 506,50 (13) -120,18 ± 268,90 (11) Relative (%) Progressions Absolute Relative ≥ 66 years After storage - 6 months After storage - 1 year 6 months - 1 year 3.1.3.2 In the DSAEK group, significant (p < 0,05) results can be found in both donor groups (< 66 and ≥ 66 years old) for all the ECDs and progressions, except for the absolute and relative progression of the ECD the last 6 months of follow-up (< 66 years: t(207) = -3,10; t(158,55) = -5,89; t(135) = -4,62; t(158) = -4,13); t(158) = -4,87; t(132) = -3,53; t(132) = -4,15 and ≥ 66 years: t(127) = -3,32; t(105) = -7,10; t(85) = -5,68; t(102) = -6,09; t(102) = -6,80; t(85) = -4,54; t(85) = -5,13) (Table 9 + Appendix 7.3.4.2). In this progression no difference between the warm and cold storage method is detected. The significant results show all higher ECDs and less endothelial cell loss in favor of the cold storage method. Both donor groups in the DSAEK group are more similar and give a better outcome with the cold storage method. -0,28 ± 0,19 (17) -0,31 ± 0,19 (13) -0,07 ± 0,16 (11) 0,661 0,461 0,280 0,143 0,455 0,670 DSAEK 30

  36. Table 9: Summary of the Independent-Samples T Test in the DSAEK group, in which the warm and cold storage is compared in younger (< 66 years) and older (≥ 66 years) donors, based on the ECDs and progressions. Significant (p < 0,05) results, indicated in red, are in favor of the cold storage method. DSAEK Donor age Warm storage Mean ECD ± SD (N) Cold storage Mean ECD ± SD (N) ECD p-value After storage 2579,89 ± 241,98 (98) 2677,39 ± 212,88 (111) 0,002 After 6 months 1153,88 ± 435,09 (69) 1597,43 ± 524,53 (94) < 0,001 After 1 year 1184,84 ± 430,59 (57) 1557,98 ± 489,23 (80) < 0,001 < 66 years Absolute (cells/mm²) -1414,92 ± 470,19 (68) -1401,40 ± 438,52 (56) -113,41 ± 197,68 (49) Absolute (cells/mm²) -1088,40 ± 512,69 (92) -1108,29 ± 498,52 (78) -66,83 ± 284,41 (71) Relative (%) Relative (%) Progressions Absolute Relative After storage - 6 months After storage - 1 year 6 months - 1 year -0,55 ± 0,17 (68) -0,54 ± 0,17 (56) -0,06 ± 0,18 (49) -0,41 ± 0,19 (92) -0,41 ± 0,18 (78) -0,02 ± 0,17 (71) < 0,001 < 0,001 0,001 < 0,001 0,323 0,149 Warm storage Mean ECD ± SD (N) Cold storage Mean ECD ± SD (N) ECD p-value After storage 2491,56 ± 219,42 (43) 2618,27 ± 196,73 (86) 0,001 After 6 months 1069,77 ± 398,09 (31) 1730,68 ± 451,37 (76) < 0,001 After 1 year 1005,50 ± 487,98 (28) 1615,19 ± 458,07 (59) < 0,001 Absolute (cells/mm²) -1451,41 ± 401,07 (30) -1507,08 ± 544,18 (28) -118,32 ± 195,87 (22) Absolute (cells/mm²) -882,15 ± 443,97 (74) -1001,27 ± 456,29 (59) -109,44 ± 307,85 (57) ≥ 66 years Relative (%) Relative (%) Progressions Absolute Relative After storage - 6 months After storage - 1 year 6 months - 1 year 3.1.4 Breakdown by acceptor age 3.1.4.1 PKP After performing the Independent-Samples T Test in the PKP group for 3 groups of acceptors, some statistical significant (p < 0,05) differences can be found (Table 10 + Appendix 7.3.5.1). In the two first groups of acceptors (16 – 49 years and 50 – 68 years), the ECDs 6 months and 1 year postoperatively and the progressions after storage – 6 months and after storage – 1 year show significant (p < 0,05) results in favor of the cold storage method (group 1: t(86) = - 3,27; t(75) = -3,85; t(85) = -2,76; t(85) = -3,05; t(74) = -3,43; t(32,96) = -3,93 and group 2: t(75) = -2,79; t(62) = -2,60; t(72) = -2,36; t(72) = -2,49; t(61) = -2,49; t(61) = -2,51). The ECD after storage and the progression the last 6 months of follow-up are not significant (p = 0,366; p = 0,175; p = 0,144; p = 0,932; p = 0,775; p = 0,985), which means that no statistical significant difference between the warm and cold storage can be found. The oldest group of acceptors (69 – 94 years) show completely different results compared to the other two groups of acceptors. Only the ECD after storage is here significant in favor of the cold storage method (p = 0,003; t(110) = -3,01). All other ECDs and progressions are not significant in the oldest group, which means that the storage method has no influence on the oldest acceptors. -0,58 ± 0,15 (30) -0,60 ± 0,20 (28) -0,10 ± 0,19 (22) -0,34 ± 0,17 (74) -0,38 ± 0,17 (59) -0,05 ± 0,16 (57) < 0,001 < 0,001 < 0,001 < 0,001 0,900 0,209 31

  37. Table 10: Summary of the Independent-Samples T Test in the PKP group in order to compare the warm and cold storage method in 3 groups of acceptors, based on the ECDs and progressions. Significant (p < 0,05) results, indicated in red, are in favor of the cold storage method. PKP Acceptor age Warm storage Mean ECD ± SD (N) Cold storage Mean ECD ± SD (N) ECD p-value After storage 2628,31 ± 380,58 (86) 2701,19 ± 271,69 (26) 0,366 After 6 months 1899,39 ± 465,58 (66) 2261,73 ± 397,44 (22) 0,002 After 1 year 1693,22 ± 488,13 (58) 2174,84 ± 420,81 (19) < 0,001 16 - 49 years Absolute (cells/mm²) -763,84 ± 436,44 (66) -979,42 ± 489,00 (58) -211,87 ± 303,20 (53) Absolute (cells/mm²) -465,43 ± 413,51 (21) -541,78 ± 412,74 (18) -98,87 ± 180,51 (15) Relative (%) Relative (%) Progressions Absolute Relative After storage - 6 months After storage - 1 year 6 months - 1 year -0,29 ± 0,16 (66) -0,36 ± 0,18 (58) -0,11 ± 0,16 (53) -0,17 ± 0,14 (21) -0,20 ± 0,15 (18) -0,05 ± 0,09 (15) 0,007 0,003 0,001 < 0,001 0,175 0,114 Warm storage Mean ECD ± SD (N) Cold storage Mean ECD ± SD (N) ECD p-value After storage 2551,53 ± 254,01 (85) 2554,85 ± 142,82 (27) 0,932 After 6 months 1710,21 ± 552,40 (56) 2093,38 ± 489,59 (21) 0,007 After 1 year 1609,98 ± 550,53 (47) 2018,59 ± 565,24 (17) 0,012 Absolute (cells/mm²) Absolute (cells/mm²) Relative (%) Relative (%) Progressions Absolute Relative 50 - 68 years After storage - 6 months -831,12 ± 521,30 (56) -0,33 ± 0,20 (56) -503,22 ± 487,62 (18) -0,19 ± 0,19 (18) 0,021 0,015 After storage - 1 year -931,66 ± 500,08 (47) -0,37 ± 0,21 (47) -558,38 ± 573,06 (16) -0,22 ± 0,22 (16) 0,016 0,015 6 months - 1 year -165,46 ± 327,36 (39) -0,09 ± 0,23 (39) -199,46 ± 477,84 (13) -0,09 ± 0,23 (13) 0,775 0,985 Warm storage Mean ECD ± SD (N) Cold storage Mean ECD ± SD (N) ECD p-value After storage 2391,51 ± 245,73 (86) 2548,08 ± 179,15 (26) 0,003 After 6 months 1638,89 ± 486,52 (54) 1784,80 ± 445,79 (15) 0,300 After 1 year 1433,38 ± 508,44 (45) 1540,33 ± 344,28 (9) 0,550 69 - 94 years Absolute (cells/mm²) -752,80 ± 449,20 (54) Absolute (cells/mm²) -765,20 ± 382,57 (15) Progressions Relative (%) Relative (%) Absolute Relative After storage - 6 months -0,32 ± 0,19 (54) -0,30 ± 0,16 (15) 0,923 0,805 After storage - 1 year -943,51 ± 474,99 (45) -0,40 ± 0,20 (45) -1020,78 ± 348,19 (9) -0,40 ± 0,13 (9) 0,646 0,988 6 months - 1 year -203,15 ± 360,59 (40) -0,11 ± 0,27 (40) -248,44 ± 242,74 (9) -0,14 ± 0,14 (9) 0,722 0,803 32

  38. 3.1.4.2 In the youngest (30 – 67 years) group of DSAEK acceptors, all ECDs and progressions are found to be significantly (p < 0,05) different in favor of the cold storage method (t(81) = -3,24; t(81) = -3,55; t(68) = -2,53; t(68) = -3,04; t(85) = -4,23; t(71) = -3,70) (Table 11 + Appendix 7.3.5.2). Except the ECD after storage and the progression the last 6 months of follow-up, which are non-significant, show no differences between both storage methods. Acceptors who are 68 – 74 years old show almost the same results. The only differences are the significant (p < 0,05) result of the ECD after storage in favor of the cold storage method and the not significant result of the absolute progression after storage – 1 year (t(105) = -2,03; t(87) = - 4,02; t(74) = -2,59; t(83) = -3,23; t(83) = -3,66; t(72) = -2,30). However, with a p-value of 0,053; this result is practically significant in favor of the cold storage method. The oldest group of acceptors (75 – 96 years) shows the same results as the youngest acceptor group, except for the ECD after storage which is also significant (p < 0,001) in favor of the cold storage method (t(121) = -3,64; t(79,64) = -7,87; t(80) = -6,66; t(94) = -5,67; t(94) = -6,79; t(76) = -5,33; t(76) = -6,17). Non-significant results show no significant differences between the warm and cold storage method. Table 11: Summary of the Independent-Samples T Test in order to find a significant (p < 0,05) difference between the warm and cold storage method in 3 different groups of acceptor ages, based on the ECDs and progressions. The significant (p < 0,05) results, indicated in red, are all in favor of the cold storage. DSAEK DSAEK Acceptor age Warm storage Mean ECD ± SD (N) Cold storage Mean ECD ± SD (N) ECD p-value After storage 2610,34 ± 238,52 (48) 2674,11 ± 215,16 (62) 0,144 After 6 months 1214,20 ± 451,00 (35) 1659,31 ± 500,68 (52) < 0,001 After 1 year 1224,32 ± 397,70 (31) 1616,90 ± 481,56 (42) < 0,001 30 - 67 years Absolute (cells/mm²) -1395,33 ± 488,12 (34) -1354,68 ± 405,82 (30) -122,44 ± 193,20 (25) Absolute (cells/mm²) -1047,33 ± 475,95 (49) -1066,57 ± 515,87 (40) -68,95 ± 321,72 (37) Relative (%) Relative (%) Progressions Absolute Relative After storage - 6 months After storage - 1 year 6 months - 1 year -0,53 ± 0,18 (34) -0,52 ± 0,16 (30) -0,07 ± 0,15 (25) -0,39 ± 0,18 (49) -0,40 ± 0,19 (40) -0,01 ± 0,19 (37) 0,002 0,001 0,014 0,003 0,459 0,194 Warm storage Mean ECD ± SD (N) Cold storage Mean ECD ± SD (N) ECD p-value After storage 2577,85 ± 203,79 (42) 2662,85 ± 216,45 (65) 0,045 After 6 months 1161,37 ± 413,48 (30) 1598,95 ± 517,95 (59) < 0,001 After 1 year 1178,03 ± 464,39 (29) 1484,09 ± 522,39 (47) 0,012 Absolute (cells/mm²) -1402,57 ± 418,24 (30) -1400,22 ± 497,74 (29) -106,28 ± 180,82 (25) Absolute (cells/mm²) -1055,75 ± 501,05 (55) -1167,42 ± 495,17 (45) -121,83 ± 345,70 (42) 68 - 74 years Relative (%) Relative (%) Progressions Absolute Relative After storage - 6 months After storage - 1 year 6 months - 1 year -0,55 ± 0,16 (30) -0,54 ± 0,19 (29) -0,07 ± 0,15 (25) -0,40 ± 0,19 (55) -0,44 ± 0,19 (45) -0,05 ± 0,19 (42) 0,002 < 0,001 0,053 0,024 0,836 0,693 33

  39. Warm storage Mean ECD ± SD (N) Cold storage Mean ECD ± SD (N) ECD p-value After storage 2475,17 ± 247,32 (52) 2618,75 ± 190,72 (71) < 0,001 After 6 months 1012,66 ± 389,01 (35) 1693,97 ± 456,86 (66) < 0,001 After 1 year 955,12 ± 469,59 (26) 1629,52 ± 405,89 (56) < 0,001 75 – 96 years Absolute (cells/mm²) -1477,61 ± 441,43 (34) -1556,85 ± 514,21 (26) -116,29 ± 223,37 (21) Absolute (cells/mm²) -903,52 ± 492,00 (62) -967,79 ± 430,85 (52) -69,96 ± 218,10 (54) Progressions Relative (%) Relative (%) Absolute Relative After storage - 6 months After storage - 1 year 6 months - 1 year -0,59 ± 0,15 (34) -0,62 ± 0,18 (26) -0,09 ± 0,25 (21) -0,34 ± 0,18 (62) -0,37 ± 0,16 (52) -0,03 ± 0,12 (54) < 0,001 < 0,001 < 0,001 < 0,001 0,415 0,318 3.1.5 Rejection A Chi-square test is performed to compare the number of rejections between the cold and warm storage method in PKP and DSAEK (Table 12 + Appendix 7.3.6). Corneas operated with the PKP technique have 17 rejections of the total 258 corneal transplantations after warm storage, which means 6,59 %. After cold storage, the percentage of rejection is 7,14 % (6 rejections on 84 corneal transplantations). There is no significant result between the cold and warm storage in the PKP group (p = 0,860). In the DSAEK group, 12 rejections have occurred on 146 corneal transplantations (8,22 %) after warm storage and 8 rejections on 215 corneal transplantations (3,72 %) after cold storage. No significant result is obtained by comparing the warm and cold storage method (p = 0,067). However the value 0,067 is close to 0,05 and can be seen as almost significant. Table 12: A summary of the Chi-square test in order to find a significant (p < 0,05) difference between the warm and cold storage method in the PKP and DSAEK group, based on the rejection rates. No significant results are detected. Warm storage Rejection No 241 134 Cold storage Rejection No 78 207 Yes 17 12 Yes 6 8 p-value 0,860 0,067 PKP DSAEK 3.2 3.2.1 Macroscopic evaluation Dr. Roels detected no macroscopic differences between the different sclerae after storage in ethanol from UZ Ghent pharmacy or from Fagron. Also the rigidity and flexibility were largely the same. Dr. Roels could apply the same force to all the scleral parts in the two different ethanol solutions. He had no remarks on the scleral structure itself and gave permission for further investigation on the sclerae. Sclera 34

  40. 3.2.2 Histological evaluation The complete datasheet with results of the histological evaluation by two investigators can be found in the Appendix 7.4. Now, the three sclerae in the UZ Ghent pharmacy ethanol and in the Fagron ethanol are separately discussed, based on the scores that the investigators gave. Some histological slides are not investigated due to some problems that occurred during the process of making the histological sections. For example, if the slice of the tissue is not nicely spread over the slide, folds are formed and those folds are visualized under the microscope while the normal structure of the sclera is lost. Another problem is that if the mounting step is not performed properly, the tissue slice cannot be perfectly visualized and it is difficult to see the different layers of the sclera. However, most of the slides could be examined by the two investigators. By first looking at the elastic fibers of the episclera of the first sclera (18S038L), almost no differences are found between the UZ Ghent pharmacy ethanol and the Fagron ethanol (Figure 15). The elastic fibers in both ethanol solutions get in most cases a score of ++, which means that the elastic fibers are remarkably present. The presence of blood vessels in the episclera is somewhat different in the different ethanol solutions. The parts of the sclera stored in UZ Ghent pharmacy ethanol showed slightly a better preservation of the blood vessels compared to the parts stored in Fagron ethanol. Both storage media had scores between – and + but in the UZ Ghent pharmacy ethanol, blood vessels are more scored with + and blood vessels in Fagron ethanol are more scored with -. The collagen density is in both solutions very good, recognized by the score ++. Fibrocyte nuclei are found in all scleral parts and in both ethanol solutions. They are mostly scored with +. The lamina fusca is seen in most parts of the sclera, even though it is an extremely thin layer. However, in some tissue parts, the lamina fusca is not detected. Figure 15: Sclera (18S038L) stored in Fagron ethanol (left) and UZ Ghent pharmacy ethanol (right) with a scale bar of 100 µm. 35

  41. The second sclera (18S036) is investigated in the same way as the first sclera (Figure 16). The elastic fibers of the episclera are in all tissue parts good visible, except for 1 tissue slide of the second investigator. Blood vessels of the episclera are in most cases not present in both the ethanol solutions. However, the scores vary between ++ and - so the scores are not uniform. The collagen of the stroma is in both the UZ Ghent pharmacy ethanol and the Fagron ethanol dense, which means that the scores are between +/- and ++. Fibrocyte nuclei in the stroma get a score of + in almost all tissue parts, so the nuclei are good visible. The presence of the lamina fusca is very diverse with scores ranging from + to -. In most tissue slices the lamina fusca is not present or visible, characterized by the score -. However, in some tissue slides, the lamina fusca was yet visible. So it is difficult to see differences between the two ethanol solutions in this sclera. Figure 16: Sclera (18S036) stored in Fagron ethanol (left) and UZ Ghent pharmacy ethanol (right) with a scale bar of 100 µm. The third sclera (18S057) that is investigated, showed excellent to very good presence of the elastic fibers of the episclera in the UZ Ghent pharmacy ethanol and in the Fagron ethanol (Figure 17). Blood vessels are present in all the tissue slices but more blood vessels are seen in the UZ Ghent pharmacy ethanol. The collagen density is the same in both ethanol solutions with scores ranging from ++ to +/-. Fibrocyte nuclei are present in all scleral parts, scored by a +. In this sclera the lamina fusca is not present in most tissue parts, characterized by the score -. Figure 17: Sclera (18S057) stored in Fagron ethanol (left) and UZ Ghent pharmacy ethanol (right) with a scale bar of 100 µm. 36

  42. 4. Discussion Normally, a prospective study would have been executed to compare the warm and cold storage method of corneal grafts. However, after many applications and emails to the ethical committee and due to the new GDPR rules, the subject of this master thesis changed into a retrospective analysis. Therefore, a database with encoded data about corneal transplantations was investigated. The corneas in this database are either stored according to the warm method or according to the cold method. Also two operation techniques, PKP or DSAEK, are included in the database. It is pity that the warm and cold storage method could not be prospectively investigated after putting a lot of effort in all the applications. 4.1 Corneal transplantations: retrospective analysis By analyzing the frequency table, it can be seen that the mean donor and acceptor age in all groups is almost similar. However, there is a lot of variance between the different groups if we look at the number of included corneas. The PKP group has more warm-stored corneas and the DSAEK group has more cold-stored corneas. This means that some results can be influenced by the number of corneas included in this analysis. In the future, it would be useful to have the same number of corneas in each group to obtain results that come from similar groups. This is not done in this study because the sample size would decrease immediately and it was aimed to investigate a database with sufficient sample size in order to draw appropriate conclusions. It can be noticed that there is a difference between the ECDs of the warm and cold storage method. In general, the cold storage method has more endothelial cells than the warm storage method. These differences will be further investigated and explained in the following sections. 4.1.1 Endothelial cell densities (ECDs) The warm and cold storage methods are compared in the PKP and DSAEK group. All results are significant, except the absolute and relative progression 6m - 1y. The cold storage gives higher ECDs and lower endothelial cell losses compared to the warm storage method. For patients it might be better to have a corneal transplantation with a cold-stored cornea because the most important parameter, the ECD, shows better outcomes. However, the difference between the amount of cells the last 6 months of follow-up is not significant. This means that during the last 6 months no significant differences between the 2 storage methods are found and no big differences occur in the number of endothelial cells. The amount of cells is already largely determined at 6 months. Thus, the ECD is largely the same after 6 months and after 1 year in the test population. In the test population of the PKP group more corneas are used that are stored according to the warm method. In the test population of the DSAEK group, this is the exact opposite. More cold-stored corneas are used for this operation technique compared to warm-stored corneas. It is not clear if this is only an observation in this dataset or a general fact. It is possible that a surgeon prefers, for example, cold-stored corneas for DSAEK. Perhaps, the numbers of corneas in both the storage methods can influence the obtained results. The means of the ECDs at 3 time points (after storage, 6 months and 1 year postoperatively) are investigated in both the warm and cold storage group. Each pair showed significant results so the 3 time points are significantly different from each other, with each time the earliest mean ECD measurement having the most endothelial cells. This means that the mean ECD significantly decreases after 6 months and 1 year in both the warm and cold storage group. 37

  43. This decrease is also visible with the mean profile plot. On average, the cold storage method has a higher mean ECD at each time point than the warm storage method. The decrease of the mean ECD between the 3 time points is in the warm storage method also greater than in the cold storage method. The decrease between the mean of the ECD after storage and the mean of the ECD after 1 year is the greatest, followed by the decrease between the mean of the ECD after storage and the mean of the ECD 6 months postoperatively. The decrease in mean ECD the last 6 months of follow-up is the smallest as seen on the graph. The cold storage method is associated with a better outcome for patients according to the ECDs, even when the warm storage has more corneas included in this study. A remark for the investigations of the ECDs and the progressions is that other complications than rejection are not taken into account. Although rejection is the principal complication other complications after corneal transplantation are possible confounders of the tests. For a following study, it would be useful to include other complications in order to see which storage method is associated with a particular complication. Moreover, many loss of follow-ups, known as drop-outs, occur in all groups after 6 months and 1 year. From this observation, the following questions can arise: are the patients who do come for a follow-up visit the ones with problems? Perhaps patients without problems after corneal transplantation attach no importance to a follow-up visit? These hypotheses could influence the ECD results in this study. For example, patients with problems could represent a lower ECD and then it would seem that the outcomes are not that promising after corneal transplantation. Furthermore, in some cases the ECD is not determined at the occasion of a follow-up visit. It would be better to further standardize the already standardized procedure in order to keep as much patients as possible in the database. Thus, also the follow-up can be a factor that influences the results of this study. Sometimes, the ECD is also measured after a time span of 2 years. For future research, it could also be interesting to check these ECDs in order to see a long-term evolution of the ECDs in patients [43]. However, in most patients no major differences between the ECD after 1 year and the ECD after 2 years are found. Also the drop-out rate is already high after 6 months and 1 year, which would probably still increase after 2 years. 4.1.2 Breakdown by donor age There is no maximum donor age in almost all European countries. However, in literature and in some countries the maximum donor age is 66 years [15, 43]. Therefore, all donors in both the PKP and DSAEK group are divided in 2 groups according to their donor age (< 66 years or ≥ 66 years). The ECDs and progressions of younger donors (< 66 years) in the PKP and DSAEK group show all better outcomes after cold storage. Younger donors are associated with a higher ECD anyway but it is even ameliorated if the corneas are stored according to the cold method. The ECD is already roughly determined 6 months postoperatively because the ECD after 1 year does not differ that much from the ECD after 6 months. This can also be confirmed by looking at the progression between 6 months and 1 year postoperatively. More interesting is the comparison between the warm and cold storage method in the group of donors ≥ 66 years. In the PKP group, there is no difference between the warm and cold storage method in the older donor group. Thus if it is already certain that the operation technique PKP will be used, it does not matter which storage method is used because both storage methods give approximately the same results regarding the ECDs. This is completely different if the DSAEK group is taken into consideration. Older donors (≥ 66 years) show better outcomes after cold storage than after warm storage. If it is sure that a corneal transplantation will be done according to the DSAEK technique, cold-stored corneas would be preferred. Since only older donors in the PKP group show no significant results, the cold storage method would be the best used method in both operation techniques and in both donor groups regarding the ECDs. It must be noted that there are far more warm-stored corneas in the PKP group and more cold-stored corneas in the DSAEK group, but less distinct than in the PKP group. These observations could possibly influence the results. A solution for this could be to compose subgroups with the same number of corneas. 38

  44. 4.1.3 Breakdown by acceptor age To date, there is not much known or published about acceptor ages in corneal transplantation. Therefore, the acceptor ages in the PKP and DSAEK group are split in three different groups and a significant difference is sought between the warm and cold storage method, based on the ECDs and progressions. In the PKP group, the two youngest acceptor groups (16 – 49 years and 50 – 68 years) show the best results after cold storage. The ECD is already roughly fixed 6 months postoperatively. Due to the non-significant result of the progression 6 months – 1 year, it can be seen that there is not that much cell loss the last 6 months in both the warm and cold storage method. The ECD after storage is not significant but this makes sense because the ECD after storage is preoperatively determined in the clean room and the acceptor age is not known yet. The oldest group of acceptors (69 – 94 years) in the PKP group show no differences between the warm and cold storage method. Only the ECD after storage, which is preoperatively determined, is significant in favor of the cold storage method. This significant difference disappeared completely after 6 months and 1 year. So in the oldest donor group, it does not matter if the corneas are warm-stored or cold-stored. Since the other acceptor groups do show a better outcome after cold storage, the cold storage method is preferred in the PKP group. It must be kept in mind that there are much less cold-stored corneas in the PKP group. This can also be the reason why the oldest acceptor group does not show significant results. In the DSAEK group, all groups of acceptors show preference for the cold storage method. The ECD after 1 year shows no increasing cell loss compared to the ECD after 6 months. For both PKP and DSAEK, the cold storage is the preferred storage method regarding the ECDs and progressions of the acceptors. Also here, no equally divided groups of warm-stored and cold-stored corneas in PKP and DSAEK are composed, which can contribute to the results. It must also be kept in mind that younger acceptors have possibly other indications for corneal transplantation, with a possible better outcome, than older acceptors [44]. 4.1.4 Rejection A significant difference between the warm and cold storage in PKP and DSAEK is searched, based on rejection rates. In many studies this comparison is often forgotten by only looking at the operation techniques and not at the storage methods [14]. A strength of this study is that both storage methods are included with large numbers of patients to examine rejection rates. In both PKP and DSAEK no significant differences between the warm and cold storage are found. So, it does not matter which storage method is used to reduce the rejection rate maximally. However, in the DSAEK group an almost significant result (p = 0,067) is found in favor of the cold storage. This is an important finding in the comparison of the warm and cold storage method. The cold storage method in the DSAEK group tends to be the better storage method, resulting in less rejections based on rejection rates. A possible reason why there are no significant results is perhaps due to the higher amount of corneas stored according to the warm method than according to the cold method. Rejection is a severe complication after corneal transplantation and has to be avoided in all cases. The strength of this retrospective analysis of corneal transplantations is that the number of corneas included in this study is high. This means that there is a sufficient sample size to support the proposed research questions. Also the division of both storage methods and operation techniques ensures that the study is specific and takes into account as many factors as possible. Of course, repeating this analysis with more data would be useful in order to find more clear differences between the warm and cold storage method. For example, a significant difference could perhaps be found, based on rejection rates. 39

  45. The retrospective, observational study itself is seen as the principal limitation because no causal reasoning can be drawn and only associations can be found. In future research, it would be preferred to perform a randomized control trial (RCT). By using this study it would still be possible to draw causal reasons. Another limitation is that vision outcomes are not investigated in this study. It would be useful to include these data in addition in the future because these parameters are also important for corneal acceptors. 4.2 Sclera In the first sclera (18S038L), the elastic fibers and stroma show no differences between the UZ Ghent ethanol and the Fagron ethanol. The presence of blood vessels, which are part of the episclera, is somewhat different in both ethanol solutions. The UZ Ghent pharmacy ethanol has higher scores than the Fagron ethanol, so blood vessels could be better preserved in UZ Ghent pharmacy ethanol but there is a lot of variation in the different tissue slices. The lamina fusca is in most of the tissues present but due to its thinness, it is difficult to make conclusions about the sclerae based on this layer. The second sclera (18S036) has some diversity in its results because it was not cut so well. The episclera and lamina fusca are present in most of the tissues, although there are tissue slices where the episclera and lamina fusca are almost not visible. The stroma is similar in the UZ Ghent pharmacy ethanol and the Fagron ethanol. The last sclera (18S057) showed excellent presence of the elastic fibers and the stroma in both UZ Ghent pharmacy ethanol and Fagron ethanol. Like the first sclera, the blood vessels are better preserved in UZ Ghent pharmacy ethanol than in Fagron ethanol but the differences are not really clear. The lamina fusca is not visible in most of the tissues but this can be due to the cutting process, in which a layer could be removed by cutting the scleral tissue skew or by using a blunt knife. In general no noticeable differences are detected between all the different slides, with sometimes an exception for the blood vessels. If there are differences, it is difficult to know if it is due to the ethanol solution or due to the preparation process of the scleral coupes. Because the differences are not remarkable, it is most likely not due to the ethanol storage solution. This in vitro study encountered some limitations. Due to the cutting process with the microkeratome, the tissue can deform or tear. This is the major cause of artefacts that are often seen in the histological slides. The number of sclerae investigated is here limited to three. In the future, more sclerae could be investigated in order to have a larger group and to make stronger conclusions. However, it seems very doubtful after investigating many histological slides of the 3 different sclerae that significant differences would be overlooked. Only a hematoxylin-eosin staining is executed. By using other specific stains, a more detailed insight could be presented. This is not done because the stains that could be used for the scleral grafts were not suitable for paraffin coupes, which were used in this study, or it was not yet confirmed by other investigators that a specific stain could be used for paraffin coupes. Two of the 3 sclerae that are used, were rejected for release. This can of course influence the results of the histological evaluation. Due to the use of rejected sclerae, not a perfect reflection of healthy corneas can be given but it must be kept in mind that human body material is not abound. It is thus more difficult to perform studies on healthy human body material than on rejected human body material. 40

  46. 5. Conclusion The retrospective analysis of the databank with corneal transplantations was a good alternative for the prospective analysis that could not be executed. In this way, the warm and cold storage method could still be compared. The endothelial cell densities (ECDs) and progressions in both PKP and DSAEK were in favor of the cold storage method. The ECDs decreased in both storage methods, so after 6 months and 1 year the ECD decreases anyway. However, this decline is stronger in the warm storage group. Also if the dataset was split, based on donor and acceptor age, the cold storage method showed generally the best outcomes regarding the ECDs and progressions. The comparison between the warm and cold storage method based on rejection rates showed that the cold storage method reduces the number of rejections after DSAEK. This study gives additional evidence for future prospective research concerning both storage methods in the Bank for ophthalmic tissues of UZ Ghent. In general, it can be said that the cold storage method gives superior results compared to the warm storage method regarding the ECDs and progressions. However, it must be kept in mind that the storage time is shorter during cold storage. In my opinion, it depends on the supply of corneas which storage method is best. On the one hand, if there is a shortage of corneas, which results in a quick release of them, it would be better to use the cold storage method due to the successful results. On the other hand, if there are many corneas and they remain a while in the clean room before release, the warm storage would be preferred because it guarantees a longer storage time. Since promising outcomes in patients can be found after cold storage, the ‘new’ storage method could be implemented at the Bank for ophthalmic tissues of UZ Ghent after a validation period in the clean room. The histological sections of the scleral grafts were scored with different parameters in order to compare two ethanol solutions. No remarkable differences were detected by the 2 individual investigators. These results make it possible to use the new ethanol solution of Fagron instead of the UZ Ghent pharmacy ethanol. 41

  47. 6. Reference list 1. Kels BD, Grzybowski A, Grant-Kels JM (2015) Human ocular anatomy. Clin Dermatol33: 140- 6 2. Navaratnam J, Utheim TP, Rajasekhar VK, Shahdadfar A (2015) Substrates for Expansion of Corneal Endothelial Cells towards Bioengineering of Human Corneal Endothelium. Journal of functional biomaterials6: 917-45 3. Armitage WJ (2011) Preservation of Human Cornea. Transfus Med Hemother38: 143-147 4. DelMonte DW, Kim T (2011) Anatomy and physiology of the cornea. J Cataract Refract Surg 37: 588-98 5. Tan DT, Dart JK, Holland EJ, Kinoshita S (2012) Corneal transplantation. Lancet379: 1749-61 6. Sridhar MS (2018) Anatomy of cornea and ocular surface. Indian journal of ophthalmology66: 190-194 7. Wilson SE, Bourne WM (1989) Corneal preservation. Surv Ophthalmol33: 237-59 8. Armitage WJ, Dick AD, Bourne WM (2003) Predicting endothelial cell loss and long-term corneal graft survival. Investigative ophthalmology & visual science44: 3326-3331 9. Cruz GKP, Ferreira Junior MA, Balbino da Silva S, de Azevedo IC, Santos VEP, Ivo ML (2018) Chronological Factors of the Process of Donation and Classification of Corneal Tissue. Transplantation proceedings50: 827-830 10. Lamm V, Hara H, Mammen A, Dhaliwal D, Cooper DK (2014) Corneal blindness and xenotransplantation. Xenotransplantation21: 99-114 11. Fournie P, Touboul D, Arne JL, Colin J, Malecaze F (2013) [Keratoconus]. Journal francais d'ophtalmologie36: 618-26 12. Eghrari AO, Riazuddin SA, Gottsch JD (2015) Fuchs Corneal Dystrophy. Progress in molecular biology and translational science134: 79-97 13. Pricopie S, Istrate S, Voinea L, Leasu C, Paun V, Radu C (2017) Pseudophakic bullous keratopathy. Romanian journal of ophthalmology61: 90-94 14. Akanda ZZ, Naeem A, Russell E, Belrose J, Si FF, Hodge WG (2015) Graft rejection rate and graft failure rate of penetrating keratoplasty (PKP) vs lamellar procedures: a systematic review. PloS one10: e0119934 15. Lee WB, Meinecke E, Varnum B (2013) The evolution of eye banking and corneal transplantation: a symbiotic relationship. Int Ophthalmol Clin53: 115-29 16. Rock T, Landenberger J, Bramkamp M, Bartz-Schmidt KU, Rock D (2017) The Evolution of Corneal Transplantation. Ann Transplant22: 749-754 17. Gaum L, Reynolds I, Jones MN, Clarkson AJ, Gillan HL, Kaye SB (2012) Tissue and corneal donation and transplantation in the UK. Br J Anaesth108 Suppl 1: i43-7 18. Gezondheidsraad H (2013) Kwaliteitsnormen voor verschillende types van menselijk lichaamsmateriaal die voor toepassing op de mens bestemd zijn: weefsels. In p 57. Brussel: HGR 19. Gezondheidsraad H (2016) Rapportering en interpretatie van biologische testen op stalen afkomstig van donoren van menselijk lichaamsmateriaal. In p 31. Brussel: HGR 20. Technology CA In 21. Thuret G, Manissolle C, Herrag S, Deb N, Campos-Guyotat L, Gain P, Acquart S (2004) Controlled study of the influence of storage medium type on endothelial assessment during corneal organ culture. Br J Ophthalmol88: 579-81 22. Gezondheidsraad H (2014) Microbiologische controle van menselijk lichaamsmateriaal voor toepassing op de mens met maximale vrijwaring van microbiologische veiligheid: Praktische aanbevelingen. In p 43. Brussel: HGR 23. Pels E, Beele H, Claerhout I (2008) Eye bank issues: II. Preservation techniques: warm versus cold storage. Int Ophthalmol28: 155-63 24. Jeng BH (2006) Preserving the cornea: corneal storage media. Curr Opin Ophthalmol17: 332- 7 25. Tripathi H, Mehdi MU, Gupta D, Sen S, Kashyap S, Nag TC, Purwar M, Jassal M, Agrawal AK, Mohanty S, et al. (2016) Long-term preservation of donor corneas in glycerol for keratoplasty: exploring new protocols. Br J Ophthalmol100: 284-90 26. Parekh M, Salvalaio G, Ferrari S, Amoureux MC, Albrecht C, Fortier D, Ponzin D (2014) A quantitative method to evaluate the donor corneal tissue quality used in a comparative study between two hypothermic preservation media. Cell Tissue Bank15: 543-54 27. EEBA (2017) Technical Guidelines for Ocular Tissue (TGOT). In EEBA (ed) p 10. EEBA 28. Eurobio Cornea Range. In Les Ulis: Eurobio 42

  48. 29. and cold storage on metabolic profile of human cornea. Acta Ophthalmol91: 83-7 30. Kapur N, Gunda S, Dixit S, Chauhan L, Acharya M, Mathur U (2018) Effect of transfer of donor corneal tissue from McCarey-Kaufmann medium to Optisol-GS on corneal endothelium. Indian journal of ophthalmology66: 219-222 31. Valtink M, Donath P, Engelmann K, Knels L (2016) Effect of different culture media and deswelling agents on survival of human corneal endothelial and epithelial cells in vitro. Graefes Arch Clin Exp Ophthalmol254: 285-295 32. Yuksel B, Uzunel UD, Kusbeci T (2016) Endothelial Cell Viability of Donor Corneas Preserved in Eusol-C Corneal Storage Medium. Exp Clin Transplant14: 441-444 33. Camposampiero D, Tiso R, Zanetti E, Ruzza A, Bruni A, Ponzin D (2003) Improvement of human corneal endothelium in culture after prolonged hypothermic storage. European Journal of Ophthalmology13: 745-751 34. Haug K, Azqueta A, Johnsen-Soriano S, Shahdadfar A, Drolsum LK, Moe MC, Roger MT, Romero FJ, Collins AR, Nicolaissen B (2013) Donor cornea transfer from Optisol GS to organ culture storage: a two-step procedure to increase donor tissue lifespan. Acta Ophthalmol91: 219-25 35. Van Buskirk EM (1989) The anatomy of the limbus. Eye (Lond)3 ( Pt 2): 101-8 36. Frota AC, Lima Filho AA, Dias AB, Lourenco AC, Antecka E, Burnier MN, Jr. (2008) Freeze- drying as an alternative method of human sclera preservation. Arq Bras Oftalmol71: 137-41 37. Hodge C, Sutton G, Devasahayam R, Georges P, Treloggen J, Cooper S, Petsoglou C (2017) The use of donor scleral patch in ophthalmic surgery. Cell and tissue banking18: 119-128 38. Watson PG, Young RD (2004) Scleral structure, organisation and disease. A review. Exp Eye Res78: 609-23 39. Nicoli S, Ferrari G, Quarta M, Macaluso C, Govoni P, Dallatana D, Santi P (2009) Porcine sclera as a model of human sclera for in vitro transport experiments: histology, SEM, and comparative permeability. Mol Vis15: 259-66 40. Bernardini Roriz MC, Pereira de Avila M, Barbosa De Sousa L (2008) Optical microscopy study of human sclera stored in different media. Am J Ophthalmol146: 458-461 41. Romanchuk KG, Nair P, Grahn B (2003) How long can donor sclera be safely stored? Cornea 22: 569-72 42. Matthias C (2016-2017) Wat is de invloed van de bewaarmethode op de celdensiteit en klinische outcome na lamellaire en penetrerende corneatransplantatie? In Faculty of medicine and health sciences, p 66. UGent 43. Lass JH, Benetz BA, Gal RL, Kollman C, Raghinaru D, Dontchev M, Mannis MJ, Holland EJ, Chow C, McCoy K, et al. (2013) Donor age and factors related to endothelial cell loss 10 years after penetrating keratoplasty: Specular Microscopy Ancillary Study. Ophthalmology120: 2428-2435 44. Chung SH, Kim HK, Kim MS (2010) Corneal endothelial cell loss after penetrating keratoplasty in relation to preoperative recipient endothelial cell density. Ophthalmologica224: 194-8 Kryczka T, Szaflik JP, Szaflik J, Midelfart A (2013) Influence of donor age, post-mortem time 43

  49. 7. Addendum 7.1 Approval ethical committee i

  50. ii