Changes in Corneal Power up to 2 Years After Endothelial Keratoplasty: Results From the Randomized Controlled Descemet Endothelial Thickness Comparison Trial.

PURPOSE: To compare changes in corneal power measurements after Descemet membrane endothelial keratoplasty (DMEK) vs ultrathin Descemet stripping automated endothelial keratoplasty (UT-DSAEK). DESIGN: Post hoc subanalysis of the randomized controlled Descemet Endothelial Thickness Comparison Trial. METHODS: A total of 50 eyes (38 patients) with endothelial dysfunction from Fuchs' endothelial dystrophy or pseudophakic bullous keratopathy were randomized to DMEK or UT-DSAEK 1 to 2 days before surgery. Total corneal refractive power (TCRP) and anterior/posterior simulated keratometry were obtained using Scheimpflug imaging preoperatively (baseline) and postoperatively at 3, 6, 12, and 24 months. Spectacle refractions were performed at 6, 12, and 24 months after surgery. SETTING: Hospital centers. RESULTS: The mean hyperopic shift of TCRP from baseline to 12 months was 0.80 ± 1.1 (P = .002) in the DMEK group and 0.69 ± 0.84 (P < .001) in the UT-DSAEK group. Posterior corneal curvature (average K from simulated keratometry) steepened (more negative dioptric power) by 0.42 ± 0.10 (P < .001) in DMEK and 0.54 ± 0.09 (P < .001) in UT-DSAEK. The mean change in TCRP and posterior corneal curvature did not differ between DMEK and UT-DSAEK (TCRP, P = .71; posterior average K from simulated keratometry, P = .36). CONCLUSIONS: Sustained steepening in posterior corneal curvature with loss in total corneal power contributes to hyperopic shifts after endothelial keratoplasty. Changes in corneal measurements do not differ between DMEK and UT-DSAEK. NOTE: Publication of this article is sponsored by the American Ophthalmological Society.