Purpose: To evaluate the outcomes and graft survival rates after ultrathin (UT) Descemet's stripping automated endothelial keratoplasty (DSAEK) using the microkeratome-assisted double-pass technique.
Design: Prospective, consecutive, interventional case series.
Participants: Patients with endothelial decompensation of various causes (Fuchs endothelial dystrophy, pseudophakic or aphakic bullous keratopathy, failed previous graft, herpetic endotheliitis, or buphthalmus; n = 285 grafts).
Intervention: Donor preparation was performed using the microkeratome-assisted double-pass technique. Stripping of the Descemet's membrane was performed under air and the graft was delivered into the anterior chamber using the pull-through technique through a 3-mm clear-cornea incision using a modified Busin glide.
Main outcome measures: Best spectacle-corrected visual acuity (BSCVA), manifest refraction, endothelial cell density, and graft thickness (GT).
Results: Excluding all eyes with pre-existing ocular comorbidities, mean BSCVA at 3, 6, 12, and 24 months was 0.16, 0.11, 0.08, and 0.04 logarithm of the minimum angle of resolution units, respectively. The percentage of patients achieving BSCVA of 20/20 or better at 3, 6, 12, and 24 months was 12.3%, 26.3%, 39.5%, and 48.8%, respectively. A statistically significant (P < 0.0001) hyperopic shift of 0.78 ± 0.59 diopters (D; range, -0.75 to 1.75 D) was found at 1 year. The endothelial cell loss at 3, 6, 12, and 24 months was 29.8 ± 14.3%, 33 ± 15.5%, 35.6 ± 14.1%, and 36.6 ± 16.0%, respectively. The mean central GT recorded 3 months after surgery was 78.28 ± 28.89 μm. Complications included microkeratome failure to achieve perfect dissection in 21 donor tissues (7.2%), with 6 (2.1%) being discarded; total graft detachment in 11 cases (3.9%); primary failure in 4 cases (1.4%); and secondary failure in 4 additional cases (1.4%). Kaplan-Meier cumulative probability of a rejection episode at 3, 6, 12, and 24 months was 0%, 0.4%, 2.4%, and 3.3%, respectively.
Conclusions: The visual outcomes of UT DSAEK are comparable with those published for Descemet's membrane endothelial keratoplasty and better than those reported after DSAEK in terms of both speed of visual recovery and percentage of patients with 20/20 final visual acuity. However, unlike with Descemet's membrane endothelial keratoplasty, preparation and delivery of donor tissue are neither difficult nor time consuming. Complications of UT DSAEK do not differ substantially from those recorded with standard DSAEK but are much less frequent than those reported after Descemet's membrane endothelial keratoplasty.
Financial disclosure(s): Proprietary or commercial disclosure may be found after the references.
Copyright © 2013 American Academy of Ophthalmology. Published by Elsevier Inc. All rights reserved.