Descemet stripping automated endothelial keratoplasty: effect of intraoperative lenticule thickness on visual outcome and endothelial cell density.

Purpose: To determine the correlation between the intraoperative donor lenticule thickness in Descemet stripping automated endothelial keratoplasty (DSAEK) and both the best-corrected visual acuity (BCVA) and endothelial cell density (ECD) at 6 months. To describe relevant surgical considerations with regard to the choice of microkeratome head.

Design: Prospective case series.

Methods: Thirty-five patients (37 eyes) undergoing DSAEK were included in this prospective case series. Intraoperative donor lenticule thickness, assessed by ultrasound pachymetry, was compared with BCVA and ECD 6 months postoperatively.

Results: Mean BCVA logarithm of the minimal angle of resolution (logMAR) was 0.58 preoperatively and 0.29 6 months postoperatively. Three eyes were excluded from BCVA analysis. The average donor lamellae thickness was 175 μm. Mean ECD loss was 39% 6 months postoperatively compared with that preoperatively. There was no correlation between donor lamella pachymetry and BCVA logMAR and ECD at 6 months in this series (P = 0.13 and P = 0.09, respectively). Analogous to our findings, a review of published reports on donor pachymetry in DSAEK did not provide data to suggest a correlation between donor corneal thickness and postoperative BCVA or ECD.

Conclusions: Donor lenticule thickness does not have a direct effect on 6-month postoperative visual acuity or endothelial cells. Efforts should be made to reduce surgical trauma to the lenticule rather than adhering to a strict depth when cutting the graft. We propose using the Moria 350 μm microkeratome head for corneal thickness more than 600 μm and the 300 footplate for corneal thickness less than 600 μm.