Avaliação dos aspectos tomográficos corneanos em pacientes com ceratocone e ceratocone subclínico, comparativamente às córneas normais

Abstract

Introduction: Keratoconus (KC) is the most common corneal ectatic disease, considered progressive, bilateral, and may be asymmetric. It is a non-inflammatory disease characterized by focal thinning of the cornea associated with an increase in its curvature, through a reduction in the biomechanical strength of collagen fibers. Clinically, KC evolves with irregular and progressive loss of visual acuity. (9) Due to advances in diagnostic methods for its detection, it is possible to diagnose changes that are still subclinical, suggesting susceptibility to ectasia. Its treatment is constantly evolving and depends on the stage of the disease. Objective: The present study aims to evaluate, through numerous parameters extracted from Corneal Tomography exams (Pentacam), the previously existing diagnostic indices for detecting subclinical disease cases or often not yet diagnosed. Casuistic and Method: Retrospective evaluation of Pentacam exams of patients previously divided into 3 groups (Normal Corneas in both eyes, corneas with Keratoconus in both eyes - KC and Manifest Keratoconus in one eye and subclinical in the other - VAE VAE-NT). Using corneal tomography parameters, existing diagnostic indices (BAD-D and BESTi) were calculated, and ROC curves were generated to compare their sensitivity and specificity. Results: In the VAE-VAE-NT group, greater sensitivity and specificity were found in the BESTi index (V1 and V2) compared to the BAD-D index, helping to detect unsuccessful disease cases. In BESTi V1 we obtained a sensitivity of 71.6% and a specificity of 85.7% for diagnosing subclinical KC in normal corneas and BESTi V2, we found a sensitivity of 97.9% and a specificity of 98.6%. Conclusions: New studies on patients with subclinical keratoconus are of great value, as KC is known to be a disease with bilateral, but commonly asymmetrical, involvement. It is advisable to diagnose it in early cases, often with unchanged topography since its morbidity rate is high when left untreated. In this way, this study evaluated diagnostic indices based on artificial intelligence; and verified their sensitivity and specificity. Therefore, collaborating to improve the early detection of ectasia or often the detection in its subclinical forms, with an impact on the management of the patient's treatment and often the recovery of their visual quality.