Incidence of infectious endophthalmitis after keratoplasty surgery: an updated systematic review and meta-analysis

Introduction Infectious endophthalmitis is a rare but vision-threatening complication following keratoplasty. With the evolution of corneal transplantation techniques toward endothelial keratoplasty (EK) and anterior lamellar keratoplasty (ALK), contemporary pooled incidence estimates stratified by surgical technique and geographic region are required to inform perioperative prevention strategies and postoperative surveillance. This study aimed to provide an updated systematic review and single-arm meta-analysis evaluating the incidence and risk factors of infectious endophthalmitis after keratoplasty. Methods This systematic review and meta-analysis was conducted according to PRISMA guidelines. Five databases (PubMed, Scopus, Web of Science, ScienceDirect, and Cochrane Library) were searched from inception to December 31, 2025. Observational studies reporting incidence of infectious endophthalmitis after keratoplasty were included. A random-effects single-arm meta-analysis of proportions with logit transformation was performed. Subgroup analyses were conducted by keratoplasty type, geographic region, study period, and follow-up duration. Heterogeneity was assessed using the I 2 statistic and τ 2 , and prediction intervals (PIs) were calculated when appropriate. Results Twenty-one studies comprising multiple keratoplasty procedures were included. The pooled incidence of infectious endophthalmitis after keratoplasty was 0.40% (95% CI 0.30–0.50), with a prediction interval of 0.08–1.95%, indicating substantial between-study variability. Significant heterogeneity was observed (I 2 = 95.75%, τ 2 = 0.59, p < 0.001). Stratified analyses showed higher incidence after penetrating keratoplasty (0.50%; 95% CI 0.30–0.60) compared with ALK (0.20%; 95% CI 0.01–0.30) and EK (0.20%; 95% CI 0.20–0.30) (p < 0.0001). By region, incidence was 0.40% in North America, 0.30% in Asia, and 0.70% in Europe (p = 0.007). Incidence varied by follow-up duration: 0.20% (≤ 1 month), 0.60% (1–12 months), and 0.30% (> 12 months) (p = 0.007). Frequently reported risk factors included penetrating keratoplasty, combined intraocular procedures, anterior vitrectomy, donor-related contamination, delayed suture removal, and higher systemic comorbidity burden. Visual outcomes were generally poor, with frequent graft failure and limited visual recovery. Conclusion Post-keratoplasty infectious endophthalmitis remains uncommon but clinically consequential. Penetrating keratoplasty showed a higher pooled incidence than lamellar techniques in this synthesis, and meaningful geographic variation persists. These findings indicate lower pooled incidence estimates with lamellar techniques in this synthesis and highlight the importance of rigorous perioperative infection-prevention strategies.