Antifibrotic Effects of Caffeine, Curcumin and Pirfenidone in Primary Human Keratocytes.

We evaluated the small molecules (AFM) caffeine, curcumin and pirfenidone to find non-toxic concentrations reducing the transformation of activated human corneal stromal keratocytes (aCSK) to scar-inducing myofibroblasts (MYO-SF). CSK were isolated from 16 human corneas unsuitable for transplantation and expanded for three passages in control medium (0.5% FBS). Then, aCSK were exposed to concentrations of caffeine of 0−500 µM, curcumin of 0−200 µM, pirfenidone of 0−2.2 nM and the profibrotic cytokine TGF-ß1 (10 ng/mL) for 48 h. Alterations in viability and gene expression were evaluated by cell viability staining (FDA/PI), real-time polymerase chain reaction (RT-PCR) and immunocytochemistry. We found that all AFMs reduced cell counts at high concentrations. The highest concentrations with no toxic effect were 100 µM of caffeine, 20 µM of curcumin and 1.1 nM of pirfenidone. The addition of TGF-ß1 to the control medium effectively transformed aCSK into myofibroblasts (MYO-SF), indicated by a 10-fold increase in α-smooth muscle actin (SMA) expression, a 39% decrease in lumican (LUM) expression and a 98% decrease in ALDH3A1 expression (p < 0.001). The concentrations of 100 µM of caffeine, 20/50 µM of curcumin and 1.1 nM of pirfenidone each significantly reduced SMA expression under TGF-ß1 stimulation (p ≤ 0.024). LUM and ALDH3A1 expression remained low under TGF-ß1 stimulation, independently of AFM supplementation. Immunocytochemistry showed that 100 µM of caffeine, 20 µM of curcumin and 1.1 nM of pirfenidone reduce the conversion rate of aCSK to SMA+ MYO-SF. In conclusion, in aCSK, 100 µM of caffeine, 20 µM of curcumin and 1.1 nM of pirfenidone significantly reduced SMA expression and MYO-SF conversion under TGF-ß1 stimulation, with no influence on cell counts. However, the AFMs were unable to protect aCSK from characteristic marker loss.

Cytoprotective Effects of Human Platelet Lysate during the Xeno-Free Culture of Human Donor Corneas.

We evaluated the suitability of 2% human platelet lysate medium (2%HPL) as a replacement for 2% fetal bovine serum medium (2%FBS) for the xeno-free organ culture of human donor corneas. A total of 32 corneas from 16 human donors were cultured in 2%FBS for 3 days (TP1), then evaluated using phase contrast microscopy (endothelial cell density (ECD) and cell morphology). Following an additional 25-day culture period (TP2) in either 2%FBS or 2%HPL, the pairs were again compared using microscopy; then stroma and Descemet membrane/endothelium (DmE) were processed for next generation sequencing (NGS). At TP2 the ECD was higher in the 2%HPL group (2179 ± 288 cells/mm2) compared to 2%FBS (2113 ± 331 cells/mm2; p = 0.03), and endothelial cell loss was lower (ECL HPL = -0.7% vs. FBS = -3.8%; p = 0.01). There were no significant differences in cell morphology between TP1 and 2, or between 2%HPL and 2%FBS. NGS showed the differential expression of 1644 genes in endothelial cells and 217 genes in stromal cells. It was found that 2%HPL led to the upregulation of cytoprotective, anti-inflammatory and anti-fibrotic genes (HMOX1, SERPINE1, ANGPTL4, LEFTY2, GADD45B, PLIN2, PTX3, GFRA1/2), and the downregulation of pro-inflammatory/apoptotic genes (e.g., CXCL14, SIK1B, PLK5, PPP2R3B, FABP5, MAL, GATA3). 2%HPL is a suitable xeno-free substitution for 2%FBS in human cornea organ culture, inducing less ECL and producing potentially beneficial alterations in gene expression.

Suitability of Different Diagnostic Platforms for Virological Testing of Blood Samples from Cornea Donors.

Background: To minimize the risk of disease transmission in cornea transplantation, donor screening for blood-derived viral infections is mandatory. Ideally, pre-mortem blood samples are used, but based on availability, cadaveric blood samples of cornea donors may also be used. However, serological and nucleic acid amplification tests (NATs) need to be validated for the use of cadaveric specimens. Methods: Hepatitis B virus (HBV), hepatitis C virus (HCV), human immunodeficiency virus (HIV), human T-lymphotropic virus (HTLV) 1/2, and Treponema pallidum (syphilis)-specific serological and/or NAT assays were validated on different platforms (Abbott Alinity i, Alinity m, Roche Cobas 6800, and Roche Cobas AmpliPrep/Cobas TaqMan (CAP/CTM)) using (un)spiked paired pre- and post-mortem cornea donor blood samples from the same individual (up to 23.83 h after death) of 28 individuals in accordance with the specifications of the German Federal Institute for Vaccines and Biomedicines (Paul-Ehrlich-Institut [PEI]). In addition, routinely HBV-, HCV- and HIV-PCR-negative tested post-mortem blood samples of 24 individuals were used to assess NAT specificity. Results: For the majority of serological parameters on the Abbott Alinity i (HBsAg, anti-HBc, anti-HBs, anti-HCV, anti-HIV, anti-HTLV 1/2, and anti-Treponema pallidum), ratios of generated test results of (un)spiked paired pre- and post-mortem blood samples differed ≤25%, with an agreement of qualitative pre- and post-mortem test results ranging from 91.2 to 100%. For NAT parameters (HBV, HCV, and HIV) on the Cobas 6800, Alinity m, and CAP/CTM, no significant deviation in virus concentrations (factor >5) of spiked pre- and post-mortem blood samples could be observed. Ct-values of corresponding internal controls did also not differ significantly (>1.5 Ct-values). In addition, no false-positive test results were generated when specificity was assessed. Conclusion: Overall, fluctuations of test results for serological and NAT parameters in pre- and post-mortem blood samples examined in this study, were only limited and within the range of what is also observed when routinely testing fresh patient specimens. We conclude that all examined assays are eligible for the screening of blood samples taken up to about 24 h after the occurrence of death.

Human platelet lysate as a replacement for fetal bovine serum in human corneal stromal keratocyte and fibroblast culture.

The isolation and propagation of primary human corneal stromal keratocytes (CSK) are crucial for cellular research and corneal tissue engineering. However, this delicate cell type easily transforms into stromal fibroblasts (SF) and scar inducing myofibroblasts (Myo-SF). Current protocols mainly rely on xenogeneic fetal bovine serum (FBS). Human platelet lysate (hPL) could be a viable, potentially autologous, alternative. We found high cell survival with both supplements in CSK and SF. Cell numbers and Ki67+ ratios increased with higher fractions of hPL and FBS in CSK and SF. We detected a loss in CSK marker expression (Col8A2, ALDH3A1 and LUM) with increasing fractions of FBS and hPL in CSK and SF. The expression of the Myo-SF marker SMA increased with higher amounts of FBS but decreased with incremental hPL substitution in both cell types, implying an antifibrotic effect of hPL. Immunohistochemistry confirmed the RT-PCR findings. bFGF and HGF were only found in hPL and could be responsible for suppressing the Myo-SF conversion. Considering all findings, we propose 0.5% hPL as a suitable substitution in CSK culture, as this xeno-free component efficiently preserved CSK characteristics, with non-inferiority in terms of cell viability, cell number and proliferation in comparison to the established 0.5% FBS protocol.

Eye Banks: Future Perspectives. / Hornhautbanken: Perspektiven für die Zukunft.

Technological progress and societal change are transforming medicine, and cornea banks are no exception. New infectiological factors, statutory requirements, management concepts, globalisation and digitalisation are also influencing how such facilities will operate in the future. The goal of providing high quality material to patients with corneal disease remains unaltered. The present article seeks to shed light on the type of material this will involve and under what circumstances it is to be obtained.

Current Challenges Facing Cornea Banks. / Aktuelle Herausforderungen der Hornhautbanken.

Cornea transplants are tissue transplants and, as such, must be distinguished from organ transplants (e.g. heart or kidney transplants). However, tissue transplants can only be performed if there are enough donors available to attend to patients in need. Unfortunately, there are too few organ and tissue donors in Germany. All steps involved in processing donor tissues must be performed in accordance with the highest quality standards. All tasks and measures are aimed at improving patient care in the surgical units that are to be supplied. Cornea banks are subject to complex requirements, whose implementation is essential in terms of both infrastructure and personnel. The analysis and identification of essential topics reveal central fields of action that are decisive for implementing the challenges facing cornea banks. Questions of employee qualification, strategic questions due to new transplantation techniques, and changes in the societal perception of organ and tissue donation require the development of strategies that should have a holistic and sustainable effect.

Supplementation of organ culture medium with dextran is not required in pre-stripped human donor tissue for DMEK surgery.

To assess corneal endothelial cell (CEC) quantity and quality in eye-bank prepared lamellar grafts for Descemet Membrane Endothelial Keratoplasty (DMEK) from organ cultured donor corneas that have not undergone de-swelling by media supplementation with dextran. Prior to graft preparation, corneas that had not undergone de-swelling (n = 30) were placed into fresh storage medium without dextran (KMI). Corneas in the control group (n = 30) were placed in dehydration medium containing 5% dextran (KMII). Subtotal stripping of Descemet's membrane (DM) was performed manually. Following graft preparation, 10 corneas of each group were cultured further in their respective medium for 24 h, 72 h, or 120 h, respectively. Before and after DM stripping, as well as at the end of culture, CEC numbers were obtained. At the end of culture, CEC morphology was graded using a scoring system and CEC metabolism was assessed by detection of adenosine triphosphate. At 24 h after DM stripping, mean CEC counts (in cells/mm2) were 2204 in corneas stored in KMII, and 2391 in corneas stored in KMI (p = 0.003). This corresponds to a mean relative CEC loss of 12.4% with dextran versus 9.7% without dextran (p = 0.04). At 72 h, CEC counts were 1946 in KMII, and 2289 in KMI (p = 0.004). This corresponds to CEC loss of 23% with dextran versus 14% without dextran (p = 0.009). At 120 h, CEC counts were 2047 in KMII, and 2230 in KMI (p = 0.14). This corresponds to CEC loss of 22.7% with dextran versus 17.2% without dextran (p = 0.14). Also, at 120 h after DM stripping, 6/10 corneas fell below a threshold of 2000 cells/mm2 if stored in medium containing dextran, versus 1/9 corneas if stored without dextran (p = 0.003). Morphological assessment of CEC quality revealed equal scores for cell polymorphism (median = 1), granulation (median = 0) and segmentation (median = 1) in all groups. Lower ATP/protein ratios were observed in corneas stored in medium without dextran at 24 h (p < 0.001), 72 h (p < 0.001), and 120 h (p = 0.02). Abandoning the use of dextran in corneas destined for DMEK surgery leads to increased CEC counts and thereby serves to reduce tissue loss.

Cryopreservation of amniotic membrane with and without glycerol additive.

PURPOSE: Amniotic membrane (AM) is an essential tool in ocular surface reconstruction. In this study, we analyzed the differential effects of glycerol and straight storage at - 80 °C for up to 6 months on the structural, biological, and mechanical properties of amniotic membrane (AM). METHODS: Human placentae of 11 different subjects were analyzed. AMs were stored at - 80 °C, either with a 1:1 mixture of Dulbecco's modified Eagle medium and glycerol (glycerol) or without any medium or additives (straight). Histological image analysis, tensile strength, cell viability, and basic fibroblast growth factor (bFGF) secretion were evaluated at 0.5, 1, 3, and 6 months. RESULTS: Histologically, neither glycerol nor straight storage significantly altered the epithelial or stromal structure of the AM. However, the cell number of the stroma was significantly reduced during the freezing process, independently of the storage method (p = 0.05-0.001). Tensile strength and Young's modulus were not influenced by the storage method, but longer storage periods significantly increased the tensile strength of the AMs (p = 0.028). Cell viability was higher in glycerol rather than straight AM samples for up to 3 months of storage (p = 0.047-0.03). Secretion of bFGF at 3 months of storage was significantly higher in glycerol versus straight frozen AM samples (p = 0.04). DISCUSSION: Glycerol led to higher cell viability and higher bFGF secretion for up to 3 months of AM storage. However, no significant differences between the two methods were observed at 6 months of storage at - 80 °C.

HCV RNA Testing of Plasma Samples from Cornea Donors: Suitability of Plasma Samples Stored at 4 °C for up to 8 Days.

BACKGROUND: The HCV RNA testing of potential cornea donors frequently relies on blood samples stored pre mortem. The recommended storage time of maximum 72 h frequently excludes a significant fraction of donors. METHODS: The influence of storage time of EDTA plasma samples at 4 °C on the viral load measured with the Roche HCV Quantitative Test vs. 2.0 was evaluated for 43 samples from HCV-positive individuals. RESULTS: The mean reduction of the viral load after 4 °C storage for 6-8 days was 0.46 log10 IU/ml (range +0.17 to -1.66 log10 IU/ml). After 1-3 days a mean loss of 0.19 log10 IU/ml (range +0.30 to -1.41 log10 IU/ml) and after 3-5 days of 0.32 log10 IU/ml (range +0.36 to -1.81 log10 IU/ml) was observed. In 23.3% of samples, a viral load reduction ≥ 1 log10 IU/ml (1.0-1.81 log10 IU/ml) was found after prolonged storage (5-8 days). In none of the samples did the HCV load fall below the detection limit. CONCLUSION: Plasma storage for up to 8 days can quantitatively reduce the HCV RNA load, yet has no influence on the reliability of a qualitative HCV RNA detection by this ultrasensitive test to determine the HCV status of serologically negative cornea donors.

P34-A132†Air-lift cultivation of human corneal donor tissues.

PURPOSE: Corneal donor tissue can be used for a number of different reconstructive surgical operations involving the rehabilitation of injured or degraded anterior, posterior and intermediate corneal lamellas.Potential corneal donor tissues undergo a rigorous screening process including medical evaluation of the endothelial and stromal layers. Depending on this assessment, the tissue´s scope of use is often narrowed down to few types of emergency procedures mainly due to an insufficient number of viable endothelial cells or divergent cell morphology.In addition to all these limitations, one must not ignore the sometimes critical post-preparation degeneration caused by the submerged culturing process itself, leading to epithelial debridement and stromal edema. All these factors reduce the already short supply of donor corneas. In this study, we aim to optimize this culturing process to avoid tissue degradation. METHODS: We used an organ culturing system based on our long established Ex Vivo Eye Irritation Test System (EVEIT) (Spöler et al., 2015). This bioreactor has been modified in size and shape to accommodate human-sized corneal explants. The established mechanisms for supplying the cornea with nutrients and physiologically relevant pressure conditions were adapted to support sterility. Human donor corneas that failed the initial quality protocol and which are released for research were obtained from our cornea bank and inserted in the EVEIT culture system. Corneal integrity was observed during the cultivation period of 19 days. RESULTS: The human cornea observed maintained transparency in contrast to what generally can be observed in the established European culturing system with submersion of the cornea. Final endothelial layer examinations confirmed the presence of viable endothelial cells, as documented during initial corneal bank quality control. CONCLUSION: With this proof of principle, we confirmed that we can maintain the integrity of the human donor cornea in our modified EVEIT organ culture system. Further investigation, optimization and confirmation will be pursued to meet medical regulations.

P05-A152†Human platelet lysate for cornea organ culture.

PURPOSE: Comprehensible concerns have been raised regarding the safety of FBS-based culture media. In this talk we discuss the benefits of using human platelet lysate (HPL) for the xeno-free culture of human donor corneas, isolated corneal stromal keratocytes (CSK) and stromal fibroblasts (SF). METHODS: 32 human corneas unsuitable for transplantation from 16 human donors were cultured for 25-days in either 2%FBS or 2%HPL medium and compared by phase contrast microscopy (ECD and morphology), and next generation sequencing (NGS). Effects of 0.5%FBS, 5%FBS, 0.5%HPL, 2%HPL and 10%hPL on cultured human CSK and SF were evaluated. RESULTS: Differential cornea culture showed lower endothelial cell loss in the 2%HPL vs. 2%FBS group (ECL hPL=-0.7% vs. FBS=-3.8%; p=0.01). 2%HPL led to the upregulation of cytoprotective, anti-inflammatory and anti-fibrotic genes (e.g. HMOX1, SERPINE1, ANGPTL4, LEFTY2) and the downregulation of pro-inflammatory/apoptotic genes (e.g. CXCL14, SIK1B, PLK5, PPP2R3B). CSK/SF cell viability remained high in all groups (98-100%). Cell numbers and proliferation rates increased (p=0.024-0.001), CSK marker expression decreased with higher fractions of HPL and FBS (p<0.001). SMA1 increased with higher amounts of FBS (p=0.003) but decreased with incremental HPL substitution in both cell types (p=0.014). HPL contained more TGF-ß1 (100%hPL 1.861±0.231ng/ml vs. 100%FBS 0.015±0.010ng/ml, p<0.001). bFGF and HGF were only detectable in 100% hPL (bFGF 0.067±0.017ng/ml, HGF 1.074±0.050ng/ml). CONCLUSION: 2%HPL is a suitable xeno-free substitution for 2%FBS in human cornea organ culture, inducing less ECL and potentially beneficial alterations in gene expression. CSK and SF can be cultured with xeno-free hPL. To maintain CSK characteristics substitution must remain minimal (0.5% hPL/FBS). hPL contains the antifibrotic HGF und bFGF, suppressing myofibroblast conversion.

P32-A134†Cytoprotective effects of human platelet lysate during the xeno-free culture of human donor corneas.

PURPOSE: We evaluated the suitability of 2% human platelet lysate (2%HPL) to replace 2% fetal bovine serum containing medium (2%FBS) for the xeno-free organ culture of human donor corneas. METHODS: 32 human corneas unsuitable for transplantation from 16 human donors (age 69.3±15.7years) were collected 38.5±17.1 hours after death. They were first cultured in 2%FBS containing medium for 3 days (time point TP1), then evaluated by phase contrast microscopy (endothelial cell density (ECD) and cell morphology. Following an additional 25-days culture period (time point TP2) in either 2%FBS or 2%HPL medium the pairs were again compared by phase contrast microscopy (ECD and morphology), stroma and Descemet membrane/endothelium (DmE) were processed for next generation sequencing (NGS). RESULTS: ECD did not differ between the 2%HPL and 2%FBS group at TP1 (p=0.87). At TP2 the ECD was higher in the 2%HPL group (2179±288cells/mm2) compared to 2%FBS (2113±331cells/mm2; p=0.03), and endothelial cell loss was lower (ECL hPL=-0.7% vs. FBS=-3.8%; p=0.01). There were no significant differences in cell morphology, neither between TP1 and 2 nor between 2%HPL and 2%FBS. NGS showed the differential expression of 1644 genes in endothelial and 217 genes in stromal cells. 2%HPL led to the upregulation of cytoprotective, anti-inflammatory and anti-fibrotic genes (e.g. HMOX1, SERPINE1, ANGPTL4, LEFTY2, GADD45B, PLIN2, PTX3, GFRA1/2) and the downregulation of pro-inflammatory/apoptotic genes (e.g. CXCL14, SIK1B, PLK5, PPP2R3B, SLURP1, FABP5, MAL, GATA3). CONCLUSION: 2%HPL is a suitable xeno-free substitution for 2%FBS in human cornea organ culture, inducing less ECL and potentially beneficial alterations in gene expression.

Cryopreservation in a Standard Freezer: -28 °C as Alternative Storage Temperature for Amniotic Membrane Transplantation.

Human amniotic membrane (hAM) is usually stored at -80 °C. However, in many regions, cryopreservation at -80 °C is not feasible, making hAM unavailable. Therefore, the possibility of cryopreservation at -28 °C (household freezer) was investigated. hAMs (n = 8) were stored at -80 °C or -28 °C for a mean time of 8.2 months. hAM thickness, epithelial integrity and basement membrane were assessed histologically. The collagen content, concentration of hepatocyte growth factor (HGF) and basic fibroblast growth factor (bFGF) were determined. Elastic modulus and tensile strength were measured. The mean thickness of hAM stored at -28 °C was 33.1 ± 21.6 µm (range 9.7-74.9); thickness at -80 °C was 30.8 ± 14.7 µm (range 13.1-50.7; p = 0.72). Mean collagen content, epithelial cell number and integrity score showed no significant difference between samples stored at -28 °C or -80 °C. Basement membrane proteins were well preserved in both groups. Mean tensile strength and elastic modulus were not significantly different. Concentration of bFGF at -28 °C was 1063.2 ± 680.3 pg/g (range 369.2-2534.2), and 1312.1 ± 778.2 pg/g (range 496.2-2442.7) at -80 °C (p = 0.11). HGF was 5322.0 ± 2729.3 pg/g (range 603.3-9149.8) at -28 °C, and 11338.5 ± 6121.8 pg/g (range 4143.5 to 19806.7) at -80 °C (p = 0.02). No microbiological contamination was detected in any sample. The cryopreservation of hAM at -28 °C has no overt disadvantages compared to -80 °C; the essential characteristics of hAM are preserved. This temperature could be used in an alternative storage method whenever storage at -80 °C is unavailable.

RT-PCR Testing of Organ Culture Medium for Corneal Storage Fails to Detect SARS-CoV-2 Infection Due to Lack of Viral Replication.

Concerns of possible transmission of SARS-CoV-2 from donors to patients by corneal transplantation have caused a decline in corneal transplantations. Graft culture media are routinely tested for infectious risks, but it is unclear whether this constitutes a viable means to avoid transmitting SARS-CoV-2 via keratoplasty. We found that SARS-CoV-2 RNA was not present in the medium after seven days of organ culture of corneas from donors (n = 4), who were SARS-CoV-2-positive upon tissue procurement. These medium samples showed no presence of viral RNA. To pursue this question under controlled conditions and further exclude the possibility of productive infection in corneal grafts, we inoculated corneoscleral discs from healthy donors (n = 8) with infectious SARS-CoV-2 and performed PCR testing of the culture medium at various time points. After seven days of culture, we also tested for SARS-CoV-2 RNA within the inoculated corneal tissue. The medium from tissue samples inoculated with SARS-CoV-2 showed no increase in viral RNA, which may indicate lack of viral replication in these corneal grafts. SARS-CoV-2-RNA was, however, detected on or in corneal tissue seven days after inoculation. Our data suggest that corneal grafts may not be permissive for replication of SARS-CoV-2 and demonstrates that PCR testing of culture media cannot safely exclude that tissue has been exposed to SARS-CoV-2. It also demonstrates the difficulty to differentiate between virus adherence and virus replication by PCR testing in SARS-CoV-2 exposed tissue.

SARS-CoV-2: Impact on, Risk Assessment and Countermeasures in German Eye Banks.

INTRODUCTION: Since the beginning of the COVID-19 pandemic there has been some debate regarding the risk of transmission through tissue transplantation and tissue banking processes. AIM OF THE STUDY: To analyze the changes that SARS-CoV-2 has caused regarding the harvesting of corneal donor tissue and eye bank activities in Germany. METHODS: A questionnaire was provided to 26 eye banks in Germany, consisting of questions about adaptations made in the screening of potential donors and the harvesting of corneal tissue following the pandemic spread of SARS-CoV-2. RESULTS: Eighteen eye banks actively reduced recruitment of donors and two banks ceased all activity. Additional diagnostic screening was performed in eight banks, using conjunctival swabs and/or nasopharyngeal swabs. In six eye banks, additional protective measures, such as FFP2 masks and/or facial shields, were implemented. Overall, a mean reduction in the number of obtained donor tissues of 17% was observed. DISCUSSION: Conjunctival and/or nasopharyngeal swabs of donors have been implemented by a minority. Reasons for not performing additional tests may be moderate sensitivity and lack of validation for postmortem use of RT-PCR testing. Also, the hazard of SARS-CoV-2 entering the corneal donor pool with subsequent transmission might be perceived as theoretical. Face shields provide a sufficient barrier against splash and splatter contamination but may be insufficient against aerosols. Additional face masks would provide support against aerosols, but it remains debatable if corneal harvesting can be considered an aerosol-producing procedure. In the future we expect to see changes in current guidelines because of a surge in scientific activities to improve our understanding of the risks involved with cornea donation in the COVID-19 pandemic, and because current practice may reduce the availability of donor corneas due to new exclusion criteria while the demand remains unchanged.

Detection of contamination during organ culture of the human cornea.

PURPOSE: Corneas harvested post-mortem are at risk of contamination, therefore antibiotic additives are used in cold storage and organ culture systems. In the latter, sterility testing of the medium is part of the standard protocol. Intuitively, testing after longer organ culture periods should be more likely to detect contaminations than early testing, but may delay allocation. This study evaluates whether an optimal time for detection of donor cornea contamination can be identified. METHODS: The study complies with the Declaration of Helsinki. All procedures were supervised using a certified quality management system (ISO 9001:2000). Donor corneas harvested by enucleation or corneoscleral excision over 5 consecutive years were processed according to German and EC laws and guidelines. The corneas were stored in a closed organ culture system in 100 ml MEM containing penicillin, streptomycin and amphotericin B at 31 degrees C for up to 28 days without media exchange. In 762 corneas, 10 ml samples of medium, obtained between days 3 and 8 of culture, were tested for sterility in an automated detection system (BacT/ALERT, bioMérieux). In 424 corneas, a second sterility test was performed from the same medium before release. Contamination detection probabilities were related to the culture duration before the primary test (Cochran-Armitage). RESULTS: Overall, 19 contaminations were found. Contaminations were bilateral in four donors. One contamination was apparent by macroscopic inspection of the medium prior to the primary sterility test; 12 were detected upon primary sterility testing. Furthermore, six primarily undetected contaminations were observed: five in the secondary sterility test and one suspected microscopically. In most cases, contamination could also be seen by medium turbidity and acidification, but in three cases macroscopic medium changes were significantly delayed or absent. No trends were found between the times of primary sterility sampling and both positive and false negative test outcome probabilities. CONCLUSION: Detection probability of contaminations in organ culture media does not increase between days 3 and 8; therefore, sterility can be tested on day 3. With the recommended follow-up after sterility testing being 7 days, microbiologic release can take place after 10 days of culture. Nevertheless, the testing is not failsafe, and should always be combined with macroscopic inspection of the media.

Assessment of Corneal Endothelium during Continued Organ Culture of Pre-Stripped Human Donor Tissue for DMEK Surgery.

PURPOSE: To measure corneal endothelial cell (EC) quality and quantity following Descemet membrane (DM) stripping of human donor corneas and continued storage in organ culture medium containing dextran. METHODS: DM stripping was performed in 30 organ cultured, corneoscleral discs. Corneas were divided into 3 groups of 10 corneas each. Baseline mean EC density (cells/mm2) was 2,372 (SD ± 259) in group 1, 2,540 (SD ± 266) in group 2, and 2,665 (SD ± 263) in group 3. Following subtotal DM stripping, culture was continued at 31°C for 24 hours (group 1), 72 hours (group 2), and 120 hours (group 3), respectively. EC density was measured before stripping and at the end of culture. At the end of culture, corneal EC morphology was graded using a scoring system and EC viability was measured by detection of adenosine triphosphate. RESULTS: At the end of culture, mean EC density was 2,159 (SD ± 293) in group 1, 1,946 (SD ± 182) in group 2, and 2,047 (SD ± 225) in group 3. This constitutes an EC loss of 9,1% (SD ± 5,3%) in group 1, 23,0 % (SD ± 6,5%) in group 2, and 22,7% (SD ± 9,1%) in group 3 (p < 0.001). After completion of follow-up, all groups contained corneas with EC counts < 2,000 cells/mm2. Cell morphology scores did not differ between the three experimental groups. EC viability measurements showed a tendency toward lower readings with extended length of culture. CONCLUSIONS: Corneal EC loss does occur following DM stripping and continued organ culture. EC loss increases with storage past 24 hours, but donor corneas may fall below 2,000 cells/mm2 independently of storage duration. The use of eye bank prepared donor lamellae for Descemet Membrane Endothelial Keratoplasty (DMEK) may increase patient safety by offering standardized quality control before tissue release.

The role of corneal endothelial morphology in graft assessment and prediction of endothelial cell loss during organ culture of human donor corneas.

PURPOSE: Endothelial assessment is crucial in the release of corneas for grafting. We retrospectively analysed the role of endothelial morphology parameters in predicting endothelial cell loss during organ culture. METHODS: Human donor corneas were cultured in minimal essential medium with 2% fetal calf serum and antibiotics. Initial endothelial morphology was assessed microscopically using score parameters polymegethism (POL), pleomorphism (PLE), granulation (GRA), vacuolization (VAC), segmentation of cell membranes (SEG), Descemet's folds (DF), trypan blue-positive cells (TBPC) and endothelial cell-free areas (ECFA). Some corneas were primarily rejected based on endothelial assessment. Endothelial cell density (ECD) was assessed at the beginning (I-ECD) and end of culture. Corneas were then placed in dehydration medium (as above + 5% dextran 500). In a subgroup, ECD was reassessed after dehydration. Endothelial cell loss during culture (ECL@Culture) and culture+dehydration (ECL-Culture&Dehydration) were calculated. Data were given as mean ± SD and analysed using multiple linear and logistic regression. Odds ratios (OR) and 95% confidence intervals (CI) were calculated. RESULT: I-ECD was 2812 ± 360/mm2 (n = 2356). The decision to reject a cornea due to endothelial assessment was associated negatively with I-ECD (OR = 0.77/100 cells, CI 0.7-0.82) and positively with ECFA (OR = 2.7, CI 1.69-4.35), SEG (OR =1.3, CI 1.01-1.68) and donor age (OR = 1.26/decade, CI 1.33-1.41). ECL@Culture was 153 ± 201/mm2 (n = 1277), ECL@Culture&Dehydration was 169 ± 183/mm2 (n = 918). ECL@Culture was associated positively with donor age, I-ECD, GRA and TBPC, and negatively with PLE, and DF. ECL@Culture&Dehydration was associated positively with age, sex, initial ECD, POL, PLE, VAC and TBPC. CONCLUSION: Morphological parameters displayed associations with the exclusion of corneas from culture and with endothelial cell loss. Appropriate parameter selection for screening purposes may help improve graft quality.

Conjunctival and intraocular swabs for the microbiological assessment of donor corneas.

PURPOSE: In this study, we investigated the associations between conjunctival (co) and intraocular (io) swabs and their implications for the contamination rates of organ-cultured corneas. METHODS: A total of 4177 swabs from 1054 corneas of 527 donors were acquired from the conjunctiva, after disinfection with 5% polyvinylpyrrolidone-iodine solution, and also from the anterior chamber after corneoscleral trepanation (io). Samples were incubated at 22.5 ± 2.5°C and 32.5 ± 2.5°C in thioglycollate broth for 14 days. Donor corneas were cultured in a closed system at 31°C. Microbial differentiation was performed for positive cultures. RESULTS: A higher temperature (32.5°C) and the intraocular swab retrieving localization led to significantly higher swab positive rates (32.5°C versus 22.5°C, odds 1.65, p < 0.0001; io versus co, odds 1,53, p < 0.0001). Death-to-collection time and laterality (left or right eye) had no significant influence on swab positivity. The cause of death significantly influenced the positive rates (p < 0.0001). Detection at 32.5°C occurred significantly earlier than at 22.5°C (p < 0.0001). The overall comparison of detected species showed no significant differences in the variety between intraocular and conjunctival swabs. During the study period, six contaminations of organ-cultured corneas occurred: four times Pseudomonas aeruginosa and once each Candida albicans and Staphylococcus hominis were found. Swap results and cornea contaminations were not significantly correlated. CONCLUSIONS: Co and io swabs show high microbial colonization rates, even after standard disinfection. Io swabs generally reproduce the co microbial range, most likely due to a mobilization and diversion of microorganisms during the trepanation procedure. Swab results do not yield a valuable tool to predict contaminations of organ-cultured corneas.