Прогнозирование клинико-морфофункциональных результатов кросслинкинга роговичного коллагена у детей

Abstract

Purpose. To develop methods of predicting corrected distance visual acuity (CDVA) and minimum corneal thickness (СT) 12 months after crosslinking in children using preoperative data. Material and methods. This study was conducted on 30 eyes in 30 patients (21 (68%) boys, 9 (32%) girls) with a median age of 16 [15; 17] years (from 12 to 17 years) who underwent the procedure of accelerated «epi-off» crosslinking. All patients underwent standard and special examination, including high-tech methods (Sheimpflug camera). Multiple regression was used to develop models for predicting CDVA and the minimum thickness of the cornea. The follow-up period was 12 months. Results. A model was developed for predicting CDVA 12 months after crosslinking, depending on the CDVA data and keratometry K2 before surgery, described by the formula: CDVA12 months after = –0,833 + 0,907 × CDVA before + 0,020 ×K2 (Adj R² =0.739, p<0.000). A model has been developed for predicting the minimum СT 12 months after crosslinking in children, depending on the data of the minimum СT and CDVA before surgery, described by the formula СT12 months after = 49,868 + 0,866 ×TRbefore + 14.551 × CDVAbefore (Adj R² =0,950 (p<0,001). Conclusion. The developed methods for assessing the clinical and morphofunctional results of corneal collagen crosslinking in children can significantly help in choosing the optimal tactics after the treatment of keratoconus in children using crosslinking. Key words: keratoconus, keratoconus in children, corneal collagen crosslinking, artificial intelligence, multiple regression