EndoArt in Corneal Endothelial Disease: Systematic Review and Meta-Analysis of Clinical Outcomes and Complications

The purpose of this study was to systematically evaluate the safety, clinical outcomes, and complications of EndoArt (EyeYon Medical, Nes Ziona, Israel) as an alternative to donor endothelial keratoplasty for chronic corneal oedema. MEDLINE/EMBASE, PubMed, Cochrane Library, and Google Scholar were searched for studies published between 1st Jan 2020 and 17th May 2025 reporting clinical outcomes following EndoArt implantation. A total of 12 studies were included. Eligible studies included case reports, case series, retrospective observational studies, and single-arm interventional studies. Overlapping cohorts were resolved by retaining the most complete datasets. Risk of bias was assessed by the Joanna Briggs Institute critical appraisal tools and the Methodological Index for Non-randomised Studies. Quantitative synthesis was pre-specified for outcomes with ≥2 studies; meta-analysis used standardised mean differences or pooled event rates with heterogeneity. Four studies (56 eyes) were included for quantitative analysis. Pooled visual acuity improved (SMD -0.728; 95% CI -1.277 to -0.180; p=0.009); mean visual acuity changed from 1.561 to 1.212 logMAR with low heterogeneity (I² = 20.428%, p = 0.285). Central corneal thickness decreased (SMD -1.420; 95% CI -1.867 to -0.974; p<0.001); mean central corneal thickness fell from 755.46 to 554.07 µm with low heterogeneity (I² = 0%). The pooled re-bubbling rate was 60.2% (95% CI 38.7 to 78.4%; I² = 52.88%). The most frequently reported adverse events included partial implant detachment and raised intraocular pressure. EndoArt achieves meaningful deturgescence and visual acuity gains, but at the cost of high detachment and rebubbling rates. Given current evidence, EndoArt is likely best suited as salvage therapy, offering independence from donor tissue. No immune response was reported. Larger cohort comparative studies are required to refine indications and technique.