Anterior Segment Oct Detachment Metrics And Their Role In Predicting Rebubbling After Dmek Surgery
Published 2026 - 44th Congress of the ESCRS
Reference: POCOR260 | Type: Free paper | DOI: 10.82333/mv6f-r366
Authors: Dr Seyfi Aydin* 1 , Dr Yusuf Berk Akbas 1 , Dr Fahri Onur Aydin 1 , Dr Ali Ceylan 2 , Dr Ebrar Kumantas Aydin 1 , Dr Idil Celen Arabaci 1 , Dr Seda Duran Guler 1 , A/Prof. Burcin Kepez Yildiz 1
1Basaksehir Cam and Sakura City Hospital,Istanbul,Türkiye, 2Basaksehir Cam and Sakura City Hospital,Istanbul,Türkiye;Akdeniz University Faculty of Medicine,Antalya,Türkiye
Purpose
To determine whether quantitative anterior segment optical coherence tomography (AS-OCT) graft detachment metrics can predict the need for postoperative rebubbling after Descemet membrane endothelial keratoplasty (DMEK).
Setting
Single tertiary referral eye centre (Başakşehir Çam and Sakura City Hospital, Istanbul, Turkey).
Methods
Prospective single-centre study of 50 eyes undergoing DMEK for endothelial failure at our clinic between July 2024 and July 2025. BCVA (logMAR), IOP were recorded preoperatively and at months 1, 3 and 6 postoperatively. AS-OCT was performed postoperatively and graft detachment was quantified using ImageJ (NIH): maximum detachment height and detachment area were measured, and normalised ratios (detachment area/cornea area and detachment area/graft area) were calculated. Eyes were stratified by rebubbling requirement. Rebubbling was performed when clinically indicated; ROC analysis determined cut-offs predicting rebubbling.
Results
BCVA improved from 1.3±0.6 logMAR to 0.4±0.4 at postoperative 6 months (p<0.001). Graft detachment occurred in 32/50 eyes (64%); 19/50 (38%) required rebubbling. Detachment height and area were higher in rebubbled eyes(n:19) than in non-rebubbled eyes(n:13). ROC cut-offs; detachment height 167 um (AUC 0.923), detachment area 108000 um2 (AUC 0.976), area/cornea 6.50% (AUC 0.927), area/graft 20.42% (AUC 0.915).
Conclusions
Early postoperative AS-OCT-derived detachment metrics accurately predict rebubbling requirement after DMEK. ROC-derived cut-offs may help standardise follow-up and guide timely intervention.