Long-Term Outcomes of EK for Failed Grafts

A retrospective study investigating the long-term outcomes of endothelial keratoplasty (EK) in eyes with a history of failed corneal transplantation confirmed graft survival time decreases after every subsequent graft and showed that eyes having failed multiple grafts retained good visual potential.

“We wanted to better understand long-term outcomes of EK in eyes with prior failed corneal transplants to help guide clinical decision making for their further management,” said Erika Ellis MD, PhD. “Our findings showing they retained good visual potential despite experiencing decreased graft survival time support efforts to continue pursuing alternative treatments that can provide sustained functional improvement for these patients. Perhaps emerging keratoprostheses and cell-based treatment options may improve outcomes in patients with multiple graft failures.”

Data for the study were extracted from records of patients who had undergone Descemet stripping endothelial keratoplasty (DSEK) or Descemet membrane endothelial keratoplasty (DMEK) after prior keratoplasty. The EK procedures were performed over a 10-year period (2014–2024) by 2 UCLA Jules Stein Eye Institute surgeons. The analysis included 295 EK procedures (60% DSEK, 40% DMEK) performed for prior graft failure in 221 eyes of 208 patients. Mean follow-up was about 33 months.

The team explored the effect of the number of prior grafts on graft survival by comparing results for 252 eyes with 1 to 2 prior grafts and 43 eyes with 3 or more prior grafts.

“When we looked at survival for all our grafts after prior graft failure, we found the 50% survival rate was approximately 3 years, and the 5-year survival rate was about 30%,” said Dr Ellis, explaining the rationale for this analysis. “However, within the subgroup of eyes that had 3 or more prior grafts, the 50% survival rate was just about 2 years, and the 5-year survival rate was less than 5%.”

The results for the eyes included in the retrospective chart review showed that while the graft failure rate at 1 year was higher among eyes with 3 or more prior failed grafts than among those with 1 to 2 (12% versus 8%), there was no difference between the 2 groups in the rate of repeat EK within the first year.

“This is probably because we know if the current graft fails in less than 1 year, the subsequent graft is likely to fail in even less time. Therefore, these patients are put in a difficult situation, and many choose to defer another surgery for some time,” Dr Ellis said.

Data on change in vision after repeat graft showed that 88.4% of eyes with 3 or more prior grafts and 84.7% of eyes with 1 to 2 prior grafts achieved a significant improvement in corrected distance visual acuity (CDVA) after undergoing EK (defined as an improvement from preoperative VA of 20/200 or worse to better than 20/200 or an improvement of at least 2 lines for patients with preoperative VA of 20/200 or better).

The study also analysed complication rates after EK as a secondary outcome and found that eyes with 3 or more prior grafts had a higher rate of graft rejection, although the difference compared to the group with fewer failed grafts did not achieve statistical significance.

Dr Ellis spoke at the 2025 ASCRS annual meeting in Los Angeles.

Erika M Ellis MD, PhD is an ophthalmology resident at Jules Stein Eye Institute, University of California, Los Angeles, US. emellis@mednet.ucla.edu