RESUMEN
PURPOSE: To evaluate the clinical outcomes after repeat Descemet membrane endothelial keratoplasty (DMEK) for technical failure (TF) and secondary graft failure (SGF). METHODS: Retrospective analysis of 49 eyes that underwent repeat DMEK either for TF (ie, persistent graft detachment, n = 24) or for SGF (ie, late endothelial graft failure, n = 25). Surgery indications for primary DMEK were Fuchs endothelial corneal dystrophy (FECD, 80%) and bullous keratopathy (BK, 20%). Main outcome measures were best-corrected visual acuity (BCVA), endothelial cell density (ECD), corneal backscattering, pachymetry, and graft survival. Outcomes were compared with an age-matched control group of 49 primary DMEK eyes. RESULTS: Logarithm of the minimum angle of resolution BCVA improved from 0.92 ± 0.6 before to 0.20 ± 0.3 at 1 year after repeat DMEK with better outcomes for eyes with TF than those with SGF (P = 0.046). Donor ECD decreased from 2618 ± 171 cells/mm2 before to 1247 ± 422 cells/mm2 at 1 year postoperatively, with no difference between technical TF and SGF eyes (P > 0.05). One-year BCVA and ECD outcomes were better in the control group than in the repeat DMEK group (P < 0.05). Five-year graft survival probability after repeat DMEK was better for TF than for SGF eyes (100% vs. 75%, P = 0.010) and better for eyes with FECD than BK as primary indication for surgery (92% vs. 65%, P = 0.042). CONCLUSIONS: Repeat DMEK gives acceptable clinical outcomes especially when performed for TF in the early period after primary DMEK. Long-term graft survival probability after repeat DMEK is comparable to primary DMEK for FECD eyes, whereas BK eyes may show an elevated risk to develop graft failure again.
RESUMEN
Keratoprosthesis (KPro) implantation is frequently the only recourse for patients with severe corneal disease. However, problems arise due to inadequate biointegration of the KPro, particularly the PMMA optical cylinder, such as tissue detachment, tissue melting, or eye-threatening infection in the interface. Here, using the AuroKPro as a model prosthesis, a surface functionalization approachâcoating the optical cylinder with nanohydroxyapatite (nHAp)âwas trialed in rabbit eyes with and without a proceeding chemical injury. In chemically injured eyes, which simulated total limbal epithelial stem cell deficiency, clear benefits were conferred by the coating. The total modified Hackett-McDonald score and area of tissue apposition differences 12 weeks after implantation were 5.0 and 22.5%, respectively. Mechanical push-in tests revealed that 31.8% greater work was required to detach the tissues. These differences were less marked in uninjured eyes, which showed total score and tissue apposition differences of 2.5 and 11.5%, respectively, and a work difference of 23.5%. The improved biointegration could be contributed by the attenuated expression of fibronectin (p = 0.036), collagen 3A1 (p = 0.033), and α-smooth muscle actin (p = 0.045)âproteins typically upregulated during nonadherent fibrous capsule envelopment of bioinert materialâadjacent to the optical cylinders. The coating also appeared to induce a less immunogenic milieu in the ocular surface tissue, evidenced by the markedly lower expression of tear proteins associated with immune and stimulus responses. Collectively, the level of these tear proteins in eyes with coated prostheses was 1.1 ± 13.0% of naïve eyes: substantially lower than with noncoated KPros (246.5 ± 79.3% of naïve, p = 0.038). Together, our results indicated that nHAp coating may reduce the risk of prosthesis failure in severely injured eyes, which are representative of the cohort of KPro patients.
