Relevance Of Transparency Recovery And Keratocyte Repopulation Of The Graft After Full-bed Deep Lamellar Keratoplasty Using Cryopreserved Tissue

PURPOSE. We previously developed a stromal hooking and viscoelastic detaching technique to perform deep lamellar keratoplasty, which enables complete removal of recipient stroma in full bed, named as full-bed deep lamellar keratoplasty (FBDLK). In this procedure, cryopreserved corneal grafts (-20℃), an edema and acellular corneal tissue, were used. Favorable prognosis was proved after FBDLK in clinic, whereas the exact procedures of transparency recovery and keratocyte repopulation in the cryopreserved corneal tissue after FBDLK were still unknown. It is meaningful for generalization of this new technique by evaluating transparency recovery and keratocyte repopulation in the cryopreserved corneal grafts after FBDLK.METHODS. There were141eyes of133patients included. Surgery indications included keratoconus (n=56), herpetic simplex corneal scarring (n=41), traumatic corneal scarring (n=25), and corneal dystrophy (n=11). All subjects were scheduled to visit at2days,7days, and0.5,1,3,6,12, and24months postoperatively. Corneal clarity was evaluated with slit-lamp photo at each visit, and ultrasonic corneal pathymetry and in vivo confocal microscopy (IVCM) were performed at0.5,1,3,6,12, and24months postoperatively. Data were collected prospectively and analyzed retrospectively. Main measurements of corneal clarity included central corneal thickness, corneal edema, stromal haze (dust-like or mass-like opacity in the stroma layer with local or diffused distribution, in contrast to edematous opacity), and donor-host interface haze. Corneal stroma was divided into three parts as anterior stroma (1/4depth), mid stroma (1/2depth) and posterior stroma (1/4depth), and stromal haze was evaluated in the three stromal layers. Main measurements of IVCM included morphological changes and quantitive analysis of amorphous acellular reflective material and repopulation of keratocytes. The association of haze and reflective structures in IVCM was also investigated.RESULTS. Reepithelialization of the cryopreserved corneal grafts finished at a mean duration of2.9±1.5days (range,1-7days). Central corneal thickness decreased to normal level up to3months and was relatively steady at12months postoperatively. Edematous opacity relieved quickly up to1month after surgery. Stromal haze appeared as early as1month postoperatively, which peaked at3months and almost resolved at12months postoperatively. Haze grade in the anterior stroma was significantly higher than those in the mid and posterior stroma during3-6months postoperatively. Haze grade was distinctive with individuals. Younger patients were prone to form significant stromal haze in the anterior stroma (β=0.93;95%CI=0.87-0.99. P=0.015). Residual cellular reflective structures could be observed in a few eyes (33.1%) with IVCM. They disappeared within1month, followed with accumulation of amorphous acellular reflective materials and keratocytes. The change trends of amorphous acellular reflective material and stromal haze were similar, and they associated significantly during3-24months postoperatively (r=0.46-0.51, P=0.000-0.001). Significant haze was correspondence to accumulation of amorphous acellular reflective materials in the anterior stroma, and haze relieving was correspondence to reducing of amorphous acellular reflective materials and increasing of normal-appearance keratocytes. Keratocyte density increased gradually and it was significantly higher in the anterior stroma than those in the mid and posterior stroma from3to24months postoperatively. Keratocyte density of the cryopreserved corneal grafts reached to28%,19%. and9%of normal corneal level in the anterior, mid. and posterior stroma. respectively, at2years postoperatively. Patients with younger age and significant anterior stromal haze at3months postoperatively were prone to repopulate more keratocytes at1year. Donor-host interface haze was apparent within1month, and it resolved from3to12months postoperatively. Donor-host interface haze was significantly correlated with amorphous acellular reflective material at0.5-12months posteratively (r=0.23-0.44, P= 0.000-0.02).CONCLUSIONS. The cryopreserved corneal grafts resolved of edematous opacity up to1month and restored normal corneal thickness at3months after FBDLK. Stromal haze may be an important phenomenon of keratocyte repopulation, corresponding to accumulation of amorphous acellular reflective structures, peaked at3months and resolved at1year postoperatively. Repopulation of keratocytes was significant higher in the anterior stroma than those in the mid and posterior stroma. Keratocyte density of the cryopreserved corneal grafts was much lower than normal level at2years after FBDLK.