To analyze differences in the stromal bed according to the method used to obtain the disc for endothelial transplant with 1) an automated microkeratome for Descemet stripping automated endothelial keratoplasty (DSAEK), 2) homogeneous dissection of Descemet membrane (Descemet membrane endothelial keratoplasty [DMEK]), or 3) a manual microkeratome used for DSAEK.Methods:
We analyzed the stromal bed of 25 consecutive corneas used for endothelial transplantation, divided into 3 groups: 1) 11 samples cut with an automated microkeratome, 2) 5 samples used for DMEK, 3) 9 samples cut with a manual microkeratome. A scanning electron microscope was used to obtain an image of the center of the stromal bed. The irregularity index of the surface complementary to the stromal surface of the disc obtained for grafting was calculated with the Canny algorithm (0: completely smooth; 1: completely irregular).Results:
At all thresholds studied (20, 30, 50, and 70), the mean irregularity index for group C (35.2, 24.4, 13.7, and 8.8, respectively) was higher than that of group A (26.2, 14.8, 6.7, and 4.0, respectively), which in turn was higher than that of group B (7.0, 4.2, 2.4, and 1.8, respectively). Differences were statistically significant among all groups and for all thresholds.Conclusions:
Irregularity of the stromal bed after any dissection can be quantified using the Canny method. The use of an automated microkeratome for DSAEK reduces irregularity and helps obtain a surface more similar to that obtained with DMEK than with a manual microkeratome.