Clinical ocular prediction model of postoperative ametropic amblyopia in patients with congenital ectopia lentis.

INTRODUCTION

Despite prompt and appropriate surgical management, a considerable proportion of patients with congenital ectopia lentis (CEL) suffer from postoperative ametropic amblyopia. To predict and identify at-risk patients early, and ensure timely amblyopia treatment, we conducted a thorough investigation into the onset and progression patterns of postoperative amblyopia in patients with CEL. Moreover, an ocular prediction model was constructed for amblyopia.

METHODS

In this prospective cohort study, amblyopia analysis was conducted to reveal the prevalence of postoperative amblyopia at different time points of follow-up. Comparative analysis and logistic regression analysis were performed for the development of an amblyopia prediction model. Receiver Operating Characteristic (ROC) analysis, calibration plots, and decision curve analysis (DCA) were used to evaluate the performance of the model. A nomogram was created to determine the probability of postoperative amblyopia. Amblyopia was diagnosed according to the most recent edition of the Amblyopia Preferred Practice Pattern.

RESULTS

A total of 889 eyes from 677 patients operated for CEL were enrolled in this study. In the pediatric cohort, the prevalence of amblyopia showed a decreasing trend with follow-up time from 1 month to 3.5 years. A prediction model based on preoperative best-corrected visual acuity (BCVA) and cardiac phenotype was established to predict postoperative amblyopia. For effective individual prediction, a nomogram was created. With great calibration, discrimination, and clinical usefulness, the prediction model demonstrated good performance.

CONCLUSION

The findings underscore that the prevalence of ametropic amblyopia in pediatric CEL patients who underwent lens surgery exhibited a marked decline over time. The prediction model established with preoperative BCVA and cardiac phenotype can provide accurate and individualized predictions of postoperative amblyopia, and it has the potential to assist ophthalmologists in rapidly identifying high-risk patients.