Comparison of the femtosecond laser (IntraLase) versus manual microkeratome (Moria ALTK) in dissection of the donor in endothelial keratoplasty - Initial study in eye bank eyes

Purpose: To determine the safety and efficacy of a femtosecond laser (IntraLase) and manual microkeratome (Moria ALTK) in creating precut endothelial keratoplasty donor tissue. Methods: Sixteen corneoscleral buttons from 8 donors were evaluated within 2 days of the death of the donor. The mean donor age was 72 years, and mean death-to-preservation time was 11 hours. Eight eyes underwent deep lamellar keratectomy by using the femtosecond laser (IntraLase: firing rate, 30 kHz; lamellar cut energy, 7.4 mu J; side cut energy, 5.5 mu J; spot size, 10 mu m; diameter, 9.0 mm; depth, 400 mu m; spiral pattern), whereas the other 8 eyes were cut by using the Moria ALTK microkeratome (350-mu m head). Ultrasonic pachymetry and endothelial cell density (ECD) were performed before and after keratectomy. The residual stromal bed was examined with electron microscopy to determine the smoothness of the surface. Cell viability was assessed by using a transferase dUTP nick end labeling (TUNEL) assay. Results: The mean preoperative pachymetry was similar in the microkeratome group and femtosecond laser group (P = 0.239). The microkeratome group obtained a consistently deeper keratectomy of 446 +/- 25 versus 400 +/- 41 mu m in the laser group (P = 0.023). Similarly, the residual stromal bed was thinner in the microkeratome group (115 +/- 28.5 vs. 177 +/- 42 mu m; P = 0.005). There was no statistically significant difference in the ECD between the 2 groups preoperatively or at 48 hours after keratectomy. Compared with the preoperative state, there was a 1% and 4% reduction of ECD in the microkeratome and femtosecond laser groups, respectively. Scanning electron microscopy of the stromal surface consistently showed a smoother contour in the manual microkeratome group. TUNEL assays indicate no significant endothelial cell loss in either the microkeratome group or the femtosecond laser group. Conclusions: The femtosecond laser (30 kHz) and the manual microkeratome are equally effective in creating precut endothelial keratoplasty donor tissue, with no detrimental effect on endothelial cell density. The microkeratome creates a smoother stromal surface and thinner endothelial discs. The femtosecond laser lamellar dissection depth is less deep, and the stromal surface is less smooth. This particular feature of femtosecond laser keratectomy may improve disc adherence, which continues to be a problem in endothelial keratoplasty. A prospective, randomized study is needed to evaluate postoperative vision and disc adherence by using both technologies in endothelial keratoplasty.