Purchase this article with an account.
S. Ti, E.M. Espana, M. Grueterich, A. Touhami, S.C. Tseng; Phenotypic Outcome Following Transplantation of Exvivo Expanded Limbal Epithelium on Amniotic Membrane for Total Limbal Deficiency in Rabbits . Invest. Ophthalmol. Vis. Sci. 2003;44(13):3152.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Purpose: To determine the resultant epithelial phenotype following transplantation of exvivo expanded limbal epithelium on amniotic membrane(AM) for total limbal stem cell deficiency(LSCD)in rabbits. Methods: Left eyes of 52 rabbits were rendered total LSCD by n-heptanol debridement of the entire corneal epithelium followed by 360° limbal rim removal. Following cytological verification of LSCD, superficial keratectomy was performed. Group I (n=10) received rabbit AM transplantation, whilst Groups II-IV (n=42) underwent transplantation of autologous LSC cultured on rabbit AM (LSC-AM graft). Clinical outcome was graded as "success" with restoration of smooth avascular cornea, "partial success" if two or more quadrants of corneal surface were smooth, or "failure" if the cornea was revascularized and irregular. Epithelial phenotype was graded "corneal" if expression of K3 keratin was positive and Muc5AC negative; "conjunctival" if K3 keratin was negative but Muc5AC positive, and "mixed" if both K3 keratin and Muc5AC were positive. Results: A long-term follow-up of more than one year was achieved. The failure rate was 100% in Group 1 and this correlated with an epithelial phenotype of "conjunctival" in 9 and "mixed" in 1. Inclusion of expanded LSC improved cumulative survival rates to 18-74% (Groups II-IV). Different suturing techniques, subconjunctival steroid injection, and tarsorrhaphy employed in Groups II (n=17) and III (n=13) did not significantly alter outcome. Group II phenotype was "corneal" in 2, "mixed" in 7, and "conjunctival" in 6 ; group III phenotype was "corneal" in 3, "mixed" in 3 and "conjunctival" in 9.Outcome was best in Group IV(n=12), which received a larger graft and second AM as a patch. This correlated with "corneal" phenotype in 9, "mixed" in 2, and "conjunctival" in 1. Successful corneas showed total dissolution of fibronectin-rich AM stromal matrix, while partial success and failure corneas showed retained fragments of such matrix, surrounded by inflammatory cell infiltration and positive expression of a-smooth muscle actin by fibroblasts. Conclusions: A strong correlation exists between clinical success and resultant corneal epithelial phenotype (Spearman Rank r = 0.798, P<0.0001), and remodeling of matrix without inflammation. Ex vivo expanded LSC can successfully reconstruct corneal surfaces with unilateral total LSCD.
This PDF is available to Subscribers Only