Figure 1. Histopathologic analysis of split (A, B, D, and E) and full-thickness (C and F) donor corneas after 1 (A-C) and 3 (D-F) additional weeks of organ culture showing more epithelial and stromal edema after Descemet stripping with marked loss of keratocytes in the anterior stroma at 3 culture weeks and Descemet membrane with intact and viable endothelium up to 3 culture weeks without significant differences between stripped and nonstripped buttons (hematoxylin-eosin, original magnification ×200).
Figure 2. Transmission electron microscopy of split (A-C and G-I) and full-thickness (D-F and J-L) donor corneas after 1 (A-F) and 3 (G-L) additional weeks of organ culture showing more epithelial (A, D, G, and J) and stromal (B, E, H, and K) edema after Descemet stripping and Descemet membrane with intact and viable endothelium (C, F, I, and L) up to 3 culture weeks without significant differences between stripped and nonstripped buttons (scale bars = 2.5 μm).
Customize your JAMA Network experience by selecting one or more topics from the list below.
Heindl LM, Riss S, Adler W, Steven P, Hos D, Cursiefen C. Corneal Graft Alterations After Descemet Stripping: Implications for Split Cornea Transplantation. JAMA Ophthalmol. 2013;131(5):687–689. doi:10.1001/jamaophthalmol.2013.794
Author Affiliations: Department of Ophthalmology, University Hospital of Cologne, Cologne (Drs Heindl, Steven, Hos, and Cursiefen), and Eye Hospital (Dr Riss) and Departments of Ophthalmology (Dr Riss) and Medical Informatics, Biometry, and Epidemiology (Dr Adler), Friedrich-Alexander University Erlangen-Nürnberg, Erlangen, Germany.
In recent years, there has been tremendous progress in improving lamellar keratoplasty techniques such as deep anterior lamellar keratoplasty1,2 and Descemet membrane endothelial keratoplasty.3,4 Splitting of a single donor cornea into an anterior part (including epithelium, its basement membrane, Bowman layer, and stroma) for use in a deep anterior lamellar keratoplasty procedure in a patient with anterior stromal disease (eg, keratoconus) and into a posterior part (endothelium–Descemet membrane layer) for use in a Descemet membrane endothelial keratoplasty procedure in a patient with endothelial disease (eg, Fuchs endothelial dystrophy) can reduce the need and cost for corneal donor tissue by up to 47%.5-7 Since it is thus far unclear what time limits are acceptable for storing anterior and posterior grafts in split cornea transplantation, we investigated split corneal tissue for temporal morphologic alterations after Descemet stripping.
Eighteen pairs of healthy human donor corneas unsuitable for corneal transplantation were included in this experimental study performed in conformance with the tenets of the Declaration of Helsinki. All corneoscleral buttons were organ cultured in Dulbecco modified Eagle medium containing streptomycin, penicillin (Biochrom), and fetal calf serum (Linaris) at 34°C. The mean (SD) donor age was 60 (24) years, the mean (SD) postmortem time was 10 (3) hours, and the mean (SD) preservation time prior to splitting was 300 (181) hours.
Sixteen corneas were split into anterior and posterior lamellas by Descemet stripping5 and stored for 4 weeks in organ culture to assess the time course of central corneal thickness using slitlamp-adapted optical coherence tomography (Heidelberg Engineering) and endothelial cell count using phase-contrast microscopy (Olympus BX51).
For histologic and ultrastructural analyses, 10 split and 10 full-thickness buttons were compared after 1 and 3 weeks in culture using Mann-Whitney test. P < .05 was considered statistically significant.
For anterior donor lamellas, the mean (SD) central corneal thickness increased nonlinearly from 770 (140) μm prior to splitting to 1057 (128) μm at 1 hour, 1122 (118) μm at 1 day, 1179 (131) μm at 1 week, and 1230 (139) μm at 4 weeks after stripping.
The mean (SD) endothelial cell count of posterior lenticules decreased linearly from 2683 (142) cells/mm2 prior to splitting to 2517 (227) cells/mm2 postoperatively, 2468 (238) cells/mm2 at 1 week, 2336 (393) cells/mm2 at 2 weeks, 2269 (366) cells/mm2 at 3 weeks, and 2006 (293) cells/mm2 at 4 weeks after stripping (ie, mean loss of 128 cells/mm2 per week).
By light microscopy (Figure 1) and electron microscopy (Figure 2), corneal epithelium and stroma revealed significantly more edematous alterations after stripping than full-thickness corneas (P = .02), with a significant increase from 1 to 3 culture weeks (P = .02) and significant anterior keratocyte loss within 3 culture weeks (P = .02). Descemet membranes showed an intact and viable endothelium up to 3 culture weeks in split and nonsplit buttons.
Longer storage of split donor tissue would simplify the logistics of split cornea transplantation.7 However, it was thus far unclear which (potentially irreversible) graft alterations occur after longer storage of the anterior and posterior lamellas.
In this study, anterior grafts revealed chronic edematous changes with anterior keratocyte loss starting after 1 week in culture, and posterior grafts showed a sharp increase of endothelial cell loss between 3 and 4 weeks in culture. These morphologic findings suggest that anterior lenticules can be stored reliably up to 1 week and posterior lenticules—depending on endothelial cell count prior to stripping—up to 3 weeks before grafting. However, no conclusions regarding the reversibility of those alterations can be drawn. Therefore, probably an even longer interval between splitting and grafting may be feasible. Further studies are necessary, especially to define the minimum tolerable donor endothelial cell count for Descemet membrane endothelial keratoplasty and the effect of different preservation media on graft alterations.
Nevertheless, our data support the safety of anterior donor tissue stored in organ culture up to 1 week as well as posterior donor tissue stored in organ culture up to 3 weeks for use in deep anterior lamellar keratoplasty and Descemet membrane endothelial keratoplasty surgery,simplifying the clinical feasibility of split cornea transplantation.
Correspondence: Dr Heindl, Department of Ophthalmology, University Hospital of Cologne, Kerpener Strasse 62, Cologne, 50924 Germany (ludwig.heindl @uk-koeln.de).
Conflict of Interest Disclosures: None reported.
Funding/Support: This work was supported by the German Research Foundation (Priority Research Project SFB 643: B10).
Additional Contributions: Gottfried O. H. Naumann, MD, PhD, Department of Ophthalmology, University of Erlangen, provided expert histopathologic advice; Claudia Köhne, MTA, Institute of Pathology, University of Cologne, and Franziska Bucher, MD, and Anna K. Schumacher, Department of Ophthalmology, University of Cologne, provided excellent technical support; and Kerstin Blüthner, PhD, Department of Ophthalmology, University of Erlangen, and Sigrid Roters, MD, and Hans G. Simons, MTA, Department of Ophthalmology, University of Cologne, provided outstanding eye bank management.