Supine Positioning for Graft Attachment After Descemet Membrane Endothelial Keratoplasty: A Randomized Controlled Trial.

PURPOSE: The Supine Positioning for Descemet Membrane Endothelial Keratoplasty Attachment (SUPER-DMEK) trial assessed the efficacy of prolonged supine head positioning on graft attachment. DESIGN: Randomized controlled trial. METHODS: Participants with Fuchs' dystrophy were randomized to 5 days of supine head positioning (intervention) or to 1 day (control). Participants, surgeons, and investigators were masked until the day after surgery. Adherence to the allocated intervention was monitored using a head sensor. Main outcome measures were area and volume of graft detachment (coprimary end points) 2 weeks after surgery quantified using a validated neural network for image segmentation on anterior segment optical coherence tomography images, and repeat air injection (rebubbling), subjective visual function, and adverse events (secondary end points). RESULTS: A total of 86 participants received the allocated intervention (35 eyes intervention and 51 eyes control). In the intention-to-treat analysis, the mean area of graft detachment was 28.6% in the intervention arm and 27.5% in the control arm (adjusted between-arm difference, 1.3; 95% CI, -8.7 to 11.4; P = .80). Results for volume of detachment and as-treated analyses based on head position sensor data indicated no potentially clinically relevant effect of prolonged supine positioning on graft attachment. Results were not compatible with a relevant treatment effect on rebubbling or subjective visual function. Adverse events, most commonly back pain, were more common and more severe with the intervention. CONCLUSIONS: In this randomized controlled trial, graft attachment was not improved with prolonged supine head positioning. Prolonged supine positioning frequently caused back pain. Prolonged supine positioning after Descemet membrane endothelial keratoplasty for Fuchs' dystrophy may not be needed in routine practice.

Rotational alignment of corneal endothelial grafts and risk of graft detachment after Descemet membrane endothelial keratoplasty: a double-masked pseudo-randomized study.

PURPOSE: The posterior cornea is rotationally asymmetric, and Descemet membrane endothelial keratoplasty (DMEK) grafts preferentially scroll vertically. This prospective study assessed whether graft attachment after DMEK differed depending on the rotational alignment of the donor graft in the recipient eye. METHODS: Pseudo-randomization and blinding of the graft orientation in the recipient's eye were possible by procedural separation: (1) The eye bank recorded the position of an orientation marker in the donor cornea; (2) the surgeon preparing the DMEK graft recorded an upside-down marker relative to the eye bank marker; and (3) the surgeon assessed the position of the upside-down marker in the recipient after DMEK. Surgeons were masked towards the eye bank marker. Using mixed-effects models, we assessed graft attachment relative to the rotational alignment of the donor graft. RESULTS: Postoperatively, the graft was not fully attached in 59 of 179 eyes (33%). A second air fill (rebubbling) was performed in 11%. The graft axis was in line with the recipient cornea axis in 40%, oblique in 28% and orthogonal in 32%. We did not detect an elevated risk of incomplete attachment (odds ratio [OR], 1.16; 95% CI, 0.61-2.20), risk of rebubbling (OR, 1.25; 95% CI, 0.47-3.31) or larger areas of graft detachment in non-aligned grafts compared to aligned grafts. CONCLUSION: Rotational alignment was not strongly associated with the risk of incomplete graft attachment, although modestly elevated risks cannot be ruled out. Efforts are needed to reduce the need for rebubbling after DMEK and to identify modifiable risk factors for graft detachment.