ESCRS - PP01.10 - MORPHOMETRIC CORNEAL REPAIR PATTERNS AFTER DIFFERENT METHODS OF SURFACE PHOTOABLATION FOR MYOPIA: CORRELATION ANALYSIS AND CLUSTERING

MORPHOMETRIC CORNEAL REPAIR PATTERNS AFTER DIFFERENT METHODS OF SURFACE PHOTOABLATION FOR MYOPIA: CORRELATION ANALYSIS AND CLUSTERING

Published 2026 - 30th ESCRS Winter Meeting

Reference: PP01.10 | Type: Presented Poster & Poster | DOI: 10.82333/pmvy-8c29

Authors: Mariia Zhovtoshtan* 1 , Anastasiia Kalinichenko 2

1UAB «Naujas regėjimas»,Vilnius,Lithuania, 2Shupik National Healthcare University of Ukraine,Kyiv,Ukraine

Purpose

To analyze morphometric corneal repair patterns after different methods of surface photoablation for myopia using correlation analysis and clustering.

Setting

Multicenter clinical study (Ukraine)

Methods

A prospective study included 255 patients (510 eyes) who underwent PRK, transepithelial PRK (tPRK), or LASEK. Epithelial healing time was evaluated by slit-lamp biomicroscopy with fluorescein staining, epithelial and stromal thickness (AS-OCT), Corneal Irregularity Measurement (CIM), obtained from a Placido-disk topographer (ZEISS ATLAS with Pathfinder analysis, Carl Zeiss Meditec AG, Jena, Germany), UDVA were assessed along with postoperative complications (delayed epithelialization ≥7 days, epithelial hyperplasia (>20 µm), stromal edema (>20 µm), pathological CIM, subepithelial haze) over 3 months. Pearson coefficient (r) was used to determine correlations between haze and other morphological characteristics (delayed re-epithelialization, epithelial hyperplasia, stromal edema and pathological CIM). K-means clustering (k=2) was performed in Principal Component Analysis (PCA) space to identify reparation patterns.

Results

Stromal edema occurred in 28% after PRK, 19% after LASEK, and 14% after tPRK. Epithelial healing delay ≥7 days was found in 16%, 13%, and 10%, respectively. Pathological CIM was observed in 22%, 20%, and 15%, while haze occurred in 9%, 7%, and 4%. Epithelial hyperplasia occurred in 23% after LASEK, 14% after PRK, and 12% after tPRK. Subepithelial haze was observed in 9% after PRK, 7% after LASEK, and 4% after tPRK. Haze showed strong correlations with stromal edema (r=0.96) and delayed epithelialization (r=0.93) according to correlation analysis. Cluster analysis revealed pathological profiles in 16.5% (PRK), 14.7% (LASEK), and 10.6% (tPRK), identifying subgroups with abnormal reparative patterns.

Conclusion

Transepithelial PRK was associated with the lowest complication rates among surface photoablation methods.
Stromal edema and delayed epithelialization showed the strongest correlation with haze formation. Clustering revealed distinct reparative profiles independent of surgical method. Morphometric assessment may enhance early risk stratification and guide future investigation of novel biological factors in corneal healing.