SLET is a relatively novel treatment for LSCD, which has long-term efficacy with a success rate of 76%, and 67% of successful cases attained 20/60 or a better vision (12). In two other large-sized studies, the success rate of SLET was 70% and 83.8%, and 70% and 64.7% of patients gained two or more lines of improvement of VA (16, 17). The success rate of allo-SLET is lower than auto-SLET, because it cannot avoid lysis of the implant due to immune rejection (18). In the present study, primary successful outcomes were observed in 10 (76.92%) eyes. The success rates of allo-SLET and auto-SLET were 80% and 66.67%, respectively. Besides, 60% of successful cases gained a significant improvement in VA. Compared to previous studies, the surgical success rate of the modified SLET was similar, whereas the rate of improvement in VA was not remarkable. This is related to the fact that patients included in this study had severe corneal opacity, and SLET could restore structure and function of the corneal epithelium, while it could not treat the residual corneal stromal opacity.
In traditional SLET, hAM supports the graft regeneration and acts as a scaffold for regenerating epithelium emerging out of the limbal tissue explants. In the current study, the limbus tissue directly contacted the corneal stromal, which indicated that the corneal stromal and hAM wrapped the LSCs and provided an in vivo medium for it. In addition, hAM has the effect of promoting tissue epithelialization, inhibiting formation of new blood vessels and scars, and anti-inflammatory, which has been considered as one of the important biomaterials to promote corneal regeneration and to heal the wound (19, 20). According to a previous study, from day 3 after amniotic membrane transplantation (AMT), hAM epithelium showed progressive signs of degradation, becoming undetectable at day 15, and hAM tissues were no longer detectable after 8 weeks (21). In a case report presented by Chaudhuri et al., anterior segment optical coherence tomography (AS-OCT) was administered at the first 2 weeks for a patient who underwent SLET, and showed that hAM settled down and adhered to the cornea on day 10 after SLET, while there was a complete epithelialization on the limbal side on day 14 (22). We found that the time of hAM dissolution coincides with the time of corneal epithelialization. Therefore, we speculate that when the hAM is placed under the limbal tissue, the application and dissolution of hAM may affect the regeneration rate of the limbal tissue. The use of hAM as a soft contact lens can take the advantage of its biological features, while avoiding the influence of its dissolution on epithelial regeneration.
To our knowledge, the corneal stroma has a nutritional effect on LSCs, thus, the hAM is not used as a supporting substance. A variety of cellular products in the stroma, including growth factors/cytokines, extracellular matrix (ECM) components, and kinases, can support normal corneal development and homeostasis (23). After wound healing, corneal epithelial cells release interleukin-1α (IL-1α) and interleukin-1β (IL-1β), accelerating epithelial coverage of the wound (24). Lumican in EMC maintains corneal transparency and promotes corneal epithelial wound healing by modulating the synthesis of collagen fibrils (25). IKKβ (inhibitor of NF-κB [nuclear factor κB] kinase β) in keratocytes is required for corneal epithelial wound healing via repression of oxidative stress and attenuating hepatic fibrogenesis. Besides, corneal stromal stem cells have a therapeutic effect and can restore the ECM organization and corneal transparency in vivo (26). Therefore, corneal stroma can serve as a potential nutritional source for the proliferation, motility, and differentiation of LSCs and cornea.
The renewal of the corneal epithelium and the differentiation of LSCs can be recognized by IVCM and IC based on the cell morphology and limbal tissue modifications. In agreement with previous researches, multi-layered corneal epithelia without conjunctival epithelia intruded in the central cornea were mainly detected in succeeded SLET (27). A number of scholars suggested that during corneal healing, cells first repopulate the limbus, and then, heal centripetally (28). However, in the current research, limbal crypts or palisade-like structures were not detectable in our patients. We found that after SLET, immature corneal epithelial cells and epithelial transition zone appeared around the limbus tissue and the area of mature epithelium gradually increased over time. This finding is consistent with Mittal et al.’s result (29). We can therefore speculate that, each tissue may form an island of epithelial cells. In the early postoperative period, the islands are surrounded by immature epithelial cells, which can be produced by LSCs, and then, the cells gradually proliferate and differentiate into mature epithelium, and the area of limbal tissues gradually decreases.
The expansion of the epithelial cell islands makes epithelialization of the cornea, thus, we cut the transplants into more small pieces to obtain more cell islands. However, we found that different tissues grew at different rates, probably due to the different sizes of limbal explants. Kethiri et al. demonstrated that a minimal amount of 0.3 mm2 live tissue or ≥ 0.5 mm2 cadaver tissue would be sufficient for the expansion of LSCs in vitro (30). Hence, surgery should make a balanced relationship between the number and the size of corneal tissue fragments, in order to ensure optimal tissue area, while increasing the number of epithelial cell islands. In contrast to other reports, to obtain more limbal tissues with a sufficient size, we expanded the sampling area of limbus in the donor eyes as in CLAu, while no complications appeared in the donor eyes.
Our team first reported the application of SLET in China, and achieved promising clinical results. In the present study, conjunctival cyst was first reported in the recipient eye, which is a rare postoperative complication. It may be due to implantation of conjunctival epithelial cells into the subconjunctiva during surgery, followed by proliferation of epithelial cells and degeneration of the central part, or neoformative corneal epithelial cells may grow inward and extend into the subconjunctiva and degenerate in the central part, thereby forming a cyst. However, it could be removed by a simple resection with good outcome.
The limitations of this study included small sample size, short follow-up period in some patients, and only 2 patients analyzed the epithelial morphology. Accordingly, extension of sample size, a longer observation period, and performing IVCM and IC in pre- and post-operation are required to confirm our findings.
In conclusion, modified SLET seems to be a safe and effective technique for treating LSCD. The corneal stroma and hAM wrapped around the graft can provide protection and nutrition to the LSCs without negatively influencing clinical outcomes. IVCM and IC are highly significant to diagnose LSCD and characterize the healing process on the ocular surface after SLET.