Almost two thirds of all human corneas processed in Europe for corneal grafting purposes are stored using medium-term organ culture up to 5 weeks, 1 with media containing fetal bovine serum (FBS) as a source of essential growth factors. A storage medium that does not require components of bovine origin would further reduce the risk of transmission of infectious diseases, such as bovine spongiform encephalitis, and would add an additional safety feature to the organ culture method. For this reason, efforts have been made to create a serum-free organ culture medium suitable for keratoplasty. Investigations have been carried out to evaluate the potential usability of Eurosol (Bausch & Lomb Surgical Inc., Irvine, CA), a new tissue culture medium without any components of bovine origin, suitable for corneal storage at 31–37°C for up to 28 days. This morphologic investigation with paired human corneas further evaluates the potential clinical usefulness of this new medium. Scanning electron microscopy was used to describe the ultrastructural appearance of the endothelial cell layer.
MATERIALS AND METHODS
Ten pairs of human corneas were obtained from the Salzburg Eye Bank within 29 hours of death. All corneas were unsuitable for grafting because of potential infections (e.g., hepatitis C) that could not be ruled out with absolute certainty. The mean donor age was 62.8 ± 11 years. The mean time from death to enucleation was 19.3 ± 5.8 hours, and the mean time from death to preparation was 20.3 ± 5.8 hours. The corneoscleral discs (two pairs each) were preserved in organ culture at 31°C for 7, 14, 21, 28, or 35 days. One cornea of each pair was cultivated in conventional medium-term storage medium on minimum essential medium base containing 2% FBS; the second one was stored in Eurosol (Table 1). This medium is formulated to be FBS-free. It contains a high-performance modified Optisol-GS (Bausch & Lomb Surgical Inc.) base medium and proprietary ingredients that simulate FBS, supplemented with additional antioxidants, vitamins, metabolic, nucleotide, and adenosine triphosphate precursors. Because of proprietary interests, the company keeps the exact composition of this medium confidential.
Corneas were stored in glass bottles containing 40 mL of medium. The media were changed weekly. After initial preparation and each time the media were changed, corneas were examined with inverted light microscopy (Olympus IX70, Olympus Optical Co., Tokyo, Japan). Central corneal thickness was measured by means of a micrometer dial attached to the stage of the microscope. With the endothelium in focus, the micrometer dial was set to 0. Thickness was determined by raising the stage until the epithelium was in focus and then recording the micrometer dial reading.
To estimate central endothelial cell density, printouts of a video camera (CCD Camera, Hamamatsu Photonics KK, Hamamatsu City, Japan) attached to the inverted light microscope were used for fixed-frame analysis. Three frames of 0.02 mm2 were randomly selected from the center of each cornea. All cells in the frames were counted and the results were averaged. The cell density per square millimeter for each time point was calculated according to the overall magnification. The system was calibrated by photographs of a micrometer slide. To obtain more information on surface condition and cell shape of endothelial cells at different times, cultivation of two pairs randomly selected was stopped at each of the time points (7, 14, 21, 28, and 35 days of corneal storage). Those corneas were deswollen at 31°C, and scanning electron microscopy of the endothelial surface was performed.
All conventionally stored corneas were deswollen by addition of 5% dextran T500, as commonly practiced in our eye bank. To achieve a system working completely without components of bovine origin, Optisol-GS containing a combination of 2.5% chondroitin sulfate and 1% dextran, was used as deswelling medium for all Eurosol-cultured corneas.
Scanning Electron Microscopy
Corneas were prefixed for 2–3 hours in a mixture of buffered formaldehyde (0.5%, 0.1 M phosphate buffer, pH 7.5, 22°C) and glutardialdehyde (1.5%) followed by a fixation in buffered glutardialdehyde (4%, 0.1 M phosphate buffer, pH 7.5, 2–3 hours). Specimens were rinsed in phosphate buffer (0.13 M), postfixed in buffered osmium tetroxide (1%, 0.13 M phosphate buffer according to Millonig, 20°C, 2 hours), rinsed in phosphate buffer (0.1 M), dehydrated in a graded series of ethanol, and dried to the critical point by carbon dioxide. Dry specimens were mounted with colloid silver (Degussa, Frankfurt am Main, Germany) onto specimen stubs, evaporated with carbon and gold, and examined in a scanning electron microscope (Stereoscan 250, Cambridge, United Kingdom) at an accelerating voltage of 20 kV.
We used independent (two-sided) t tests and adjusted the corresponding type I error rates based on the Bonferroni inequality, 3,4 because the observations were dependent over time. Hence, an individual difference was regarded as significant only if the corresponding p value was smaller than 0.05/12 |mS 0.0042. Therefore, an overall type I error rate of 5% could be achieved. 3
One goal of the study was to compare means from both groups and to test their equality. A power analysis based on the independent t test (two-sided) was performed to estimate the minimal differences between both groups, whereby the analysis aimed at a power of 90%. 4
Endothelial Cell Density
There was continuous endothelial cell loss with time in all corneas (Fig. 1). The mean endothelial cell density after 35 days of organ culture was 2,675 cells/mm2 for the Eurosol-cultured corneas and 2,650 cells/mm2 for the corneas cultured in FBS-containing medium. No significant differences in endothelial cell density between serum-free–cultured and FBS medium–cultured corneas could be found at any time point because all p values were larger than 0.042.
Power analysis showed a power of 90% for the assumed differences at the given time points (±276 cells/mm2 on day 7, ±339 cells/mm2 on day 14, ±375 cells/mm2 on day 21, ±483 cells/mm2 on day 28, and ±1,115 cells/mm2 on day 35). For day 21, a common standard deviation of 124.6 was observed in both groups. With the assumption that an actual difference between both group means was ±375, the study would have a power of 90% (regarding the sample size of n = 6 observations in group 1 and n = 6 observations in group 2).
After 28 days of cultivation, a mean cell loss of 6.6% was found in conventional stored corneas and 6.5% in Eurosol stored corneas. After 35 days, the two conventional cultured corneas showed a mean cell loss of 13%; the mean cell loss in Eurosol-cultured corneas was 7.9%.
All corneas showed a swelling to almost twice their original thickness mostly during the first week of cultivation, followed by a continual slight increase in corneal thickness after that time (Fig. 2). The mean corneal thickness for Eurosol-cultured corneas after 35 days was 985 μm. The mean for conventional cultured–medium corneas was 1,000 μm.
No significant difference in central corneal thickness between the two cultivation groups could be found at any time point. Again, the corresponding power analysis showed a power of 90% for the assumed differences at the given time points (±209 μm on day 7, ±144 μm on day 14, ±97 μm on day 21, ±94 μm on day 28, and ±223 μm on day 35). Using Optisol-GS as a deswelling medium was found as effective as using conventional medium and 5% dextran T500.
Scanning Electron Microscopy
Scanning electron microscopy showed an intact endothelial cell layer over the entire posterior corneal surface in each specimen. Just one cornea stored in Eurosol for seven days showed a small area of lysed cells in the periphery of the cornea (Fig. 3A). On the second cornea of this pair, cultured in the conventional Salzburg medium for 7 days, a cluster of rounded dead cells was found in the periphery of the cornea lying over intact endothelium (Fig. 3B). It was assumed that these cells had been expelled from the endothelial monolayer after their death. Moderate folding and wrinkling of the posterior cornea was noted in all samples (Fig. 4). The endothelial monolayer, however, was complete in these areas also.
In comparison with the endothelial cell layer of freshly enucleated eyes of healthy patients with a regular hexagonal cell structure, the rate of pleomorphism and polymegathism increased with time, without a notable difference between the two groups during the first 4 weeks of culture. A central cilium was often detectable on the endothelial cells of Eurosol-cultured (Fig. 5A) and conventional medium—cultured specimens.
Although the maximal storage time in the normal routine of the Salzburg cornea bank is 28 days, the observation period of two pairs of corneas was expanded to 35 days. The scanning electron microscopy pictures of the Eurosol-cultured corneas after 35 days of storage showed a regular cell structure with a low degree of polymegathism, a cell diameter of approximately 20 μm, and a smooth surface (Fig. 5B), whereas the cornea stored in conventional medium showed a cell diameter of 12–22 μm and a higher degree of irregularity (Fig. 5C).
A number of reports have shown evidence that iatrogenic transmission was responsible for several cases of Creutzfeldt-Jakob disease (CJD) caused by injection of pituitary growth hormone 5,6 or by direct implantation of dura mater grafts. 7 Furthermore, three probable cases of transmission of sporadic CJD by corneal transplantation 8–12 have been published. Recently, the epidemic of bovine spongiform encephalopathy, occurring mainly in the United Kingdom, and the possibility of transmission to humans by consumption of infected meat causing a new form of CJD, designated variant CJD, 13,14 has resulted in great public concern.
Because FBS is an essential factor in media for medium-term corneal storage and because no reliable laboratory testing for CJD or BSE is available at the moment, 10 a potential risk of transferring prion disease by corneal tissue using FBS cannot be ruled out. Because only serum is processed that originates from animals raised in countries where bovine spongiform encephalopathy has not yet occurred, this way of transmission can be considered to be rather unlikely.
Conversely, theoretically, if prion-infected FBS accidentally had been distributed and used for corneal culture, a huge number of recipients would be grafted with potentially contaminated corneas. Therefore, having a serum-free corneal storage medium available would confer an additional safety aspect to the organ culture method.
Because the quality of the donor endothelium is an important factor in the success of a penetrating keratoplasty, several criteria have been defined and widely accepted as indicators of the structural and functional integrity of the corneal endothelium. 15 Corneas that shall be used for transplantation should not have an endothelial cell count of less than 2,000 cells per square millimeter because of the regularly extensive cell loss after corneal transplantation. All corneas of this series had more than 2,500 cells per square millimeter.
Neither the Eurosol-stored nor the conventionally stored corneas lost more than 16% of their endothelial cells during their observation period. This is less than the 20% cell loss that has generally been adopted, during an observation period of 30 days, as the limit for acceptance for a cornea intended for human grafting. 15 The mean endothelial cell loss between the beginning and end of the preservation period is similar to that of other studies. 16,17 Endothelial cell death is, to a certain degree, an unavoidable finding even under the best conditions in organ culture corneal storage.
Correlation between endothelial cell loss and increasing preservation time was also found by other authors. 15,18 Because endothelial cells have only a limited capacity to divide, they must migrate and enlarge to cover the bare areas of Descemet's membrane that result from cell death. 19,20 Because of this repair mechanism, the endothelial monolayer remains intact during organ culture, 15,18,21,22 but pleomorphism and polymegathism, to a certain extent, are the result. Therefore, cell death in a population of hexagonal cells was found to result in pleomorphism by reformation of the remaining cells. 23 Vital staining of endothelial cells, as routinely used to assess endothelial viability throughout the storage period, was not performed in this study because of potential interference with scanning electron microscopy fixation.
A central cilium is normally found with scanning electron microscopy on endothelial cells of normal healthy corneas. 24 Therefore, we defined the presence of a central cilium on cultured corneas as a sign of vitality and conservation of cell structure. Swelling of the cornea induces Descemet's folds (Fig. 4), but the endothelial cell layer should remain intact in these regions. This was the case in Eurosol-cultured corneas and in corneas cultured in the conventional medium.
According to these criteria, all corneas processed in this study would have been considered transplantable. Because of the relatively small number of corneas in this series and because clinical data are not yet available, a final evaluation of this new medium is not yet confidently possible. Nevertheless, this pilot study with paired human corneas indicates a potential clinical applicability of the tested serum-free storage medium, offering an alternative to conventional FBS-containing media.
1. European eye bank activities in the last five years (1994–1998). In:European Eye Bank Association (EEBA) Directory
, 8th ed. Meeting of the European Eye Bank Association, Arhus, Denmark, 2000:12–21.
2. Skelnik DL, Lindstrom RL. A new serumtree 31°C corneal preservation medium [ARVO Abstract]. Invest Ophthalmol Vis Sci 1999; 40:S635.
3. Miller J. Simultaneous Statistical Inference
, 2nd ed. New York: Springer Verlag, 1981.
4. Neter J, Wasserman W, Kutner M. Applied Linear Statistical Models: Regression, Analysis of Variance and Experimental Design
, 3rd ed. New York: McGraw Hill College, 1990.
5. Bilette de Villemeur T, Deslys JP, Pradel A, et al. Creutzfeldt-Jakob disease from contaminated growth hormone extracts in France. Neurology 1996; 47:690–5.
6. Markus HS, Duchen LW, Parkin EM, et al. Creutzfeldt-Jakob disease in recipients of human growth hormone in the United Kingdom: a clinical and radiographic study. Q J Med 1992; 82:43–51.
7. Lang CJ, Heckmann JG, Neundoerfer B. Creutzfeld-Jakob disease via dural and corneal transplants. J Neurol Sci 1998; 160:128–39.
8. Duffy P, Wolf J, Collins G, et al. Possible person-to-person transmission of Creutzfeld-Jakob disease. N Engl J Med 1974; 290:692–3.
9. Heckmann JG, Lang CJG, Petruch F. Transmission of Creutzfeldt-Jakob disease via a corneal transplant. J Neurol Neurosurg Psychiatry 1997; 63:388–90.
10. Hogan RN, Brown P, Heck E, et al. Risk of prion disease transmission from ocular donor tissue transplantation. Cornea 1999; 18:2–11.
11. Hogan RN, Cavanagh HD. Transplantation of corneal tissue from donors with diseases of the central nervous system. Cornea 1995; 14:547–53.
12. Uchiyama K, Ishida C, Yago S, et al. An autopsy case of Creutzfeldt-Jakob disease associated with corneal transplantation
. Dementia 1994; 8:466–73.
13. Bruce ME, Will RG, Ironside JW. Transmission to mice indicate that `new variant' CJD is caused by the BSE agent. Nature 1997; 389:498–501.
14. Hill AF, Desbruslais M, Joiner S. The same prion strain causes vCJD and BSE. Nature 1997; 389:448–50.
15. Pels E, Schuchard Y. Organ culture preservation of human corneas. Doc Ophthalmol 1983; 56:147–53.
16. Frueh BE, Böhnke M. Corneal grafting of donor tissue preserved longer than 4 weeks in organ culture. Cornea 1995; 14:463–6.
17. Redbrake C, Salla S, Frantz A. Changes in human donor corneas preserved for longer than 4 weeks. Cornea 1998; 17:62–5.
18. Borderie VM, Kantelip BM, Delbosc BY, et al. Morphology, histology, and ultrastructure of human C31 organ-cultured corneas. Cornea 1995; 14:300–10.
19. Doughman DJ, Van Horn D, Harris JE, et al. The ultrastructure of human organ-cultured cornea. I. Endothelium. Arch Ophthalmol 1974; 92:516–23.
20. Steinhorst U, Böhnke M, Singh G, et al. Elektronenmikroskopische Befunde zur endothelialen Wundheilung in vitro. Fortschr Ophthalmol 1984; 81:288–90.
21. Doughman DJ, Van Horn D, Rodman WP, et al. Human corneal endothelial layer repair during organ culture. Arch Ophthalmol 1976; 94:1791–6.
22. Lindstrom RL, Doughman DJ. A metabolic and electron microscopic study of human organ-cultured cornea. Am J Ophthalmol 1976; 82:72–82.
23. Sperling S. Early morphologic changes in organ cultured human corneal endothelium. Acta Ophthalmol 1978; 56:785–91.
24. Versura P, Bonvicini F, Caramazza R, et al. Scanning electron microscopy
study of human cornea and conjunctiva in normal and various pathologic conditions. Scanning Electron Microscopy