Predicting visual outcome after open globe injury using classification and regression tree model: the Moradabad...

Area of Science:

• Ophthalmology
• Trauma Surgery

Background:

• Open globe injuries (OGIs) represent a significant cause of vision loss worldwide.
• Predicting visual outcomes in OGI patients is essential for effective management.

Purpose of the Study:

• To identify factors associated with visual outcomes in patients with open globe injuries.
• To evaluate the predictive capability of the Classification and Regression Tree (CART) model for visual prognosis in OGI.

Main Methods:

• Retrospective case series of 157 patients with OGIs presenting to a tertiary eye care institute.
• Multivariate analysis using binomial logistic regression and CART model for outcome prediction.
• Assessment of visual outcomes, risk factors, and postoperative complications.

Main Results:

• Univariate analysis identified 9 predictors of poor visual outcome.
• Relative afferent pupillary defect (RAPD), poor presenting visual acuity, adnexal injuries, and injury location were significant predictors of vision loss.
• Absence of RAPD correlated with a 79% chance of vision survival; 68% of patients with RAPD and initial visual acuity <6/60 had poor vision.

Conclusions:

• The CART model effectively predicts final visual acuity based on initial prognostic factors.
• Identifying key risk factors like RAPD and presenting visual acuity can guide OGI management and improve visual outcomes.