AbstractPurpose To evaluate the prevalence of keratoconus (KC) and associated risk factors in a previously undiagnosed population at a Brazilian university hospital, using advanced tomographic and biomechanical tools. Design Prospective, cross-sectional observational study. Methods A total of 521 participants (1,041 eyes) from the staff and students of Gaffrée and Guinle University Hospital, Federal University of the State of Rio de Janeiro (UNIRIO), were examined using corneal tomography (Pentacam AXL®; Oculus Optikgeräte GmbH, Wetzlar, Germany) and biomechanical analysis (Corvis ST®; Oculus Optikgeräte GmbH, Wetzlar, Germany). Exclusion criteria included self-reported keratoconus, prior corneal surgery, known corneal pathology, and suspected contact lens warpage. A structured questionnaire assessing demographics, eye rubbing habits, and family history was administered to all participants before diagnostic examinations. A single experienced examiner performed subjective topometric map evaluations. Objective parameters included the Tomographic and Biomechanical Index version 2 (TBIv2), Corneal Biomechanical Index (CBI), Pentacam Random Forest Index (PRFI), Belin-Ambrósio Deviation Index (BAD-Dv3), and Topographic Keratoconus Classification (TKC). Statistical analyses included chi-square tests, odds ratios (OR), and relative risks (RR). All prevalence rates are reported per eye. Results Subjectively, 36.8% of eyes were classified as normal, 31.4% with regular astigmatism, and 5.8% as ectatic or ectasia-suspect (including 4.2% suspected KC, 1.1% KC, 0.5% PMD). Contact lens warpage (15.1%) and other irregular patterns (10.9%) were also noted. Objectively, 5.86% had altered TKC, 13.16% had pachymetry Conclusions This study revealed a notable prevalence of undiagnosed clinical and subclinical keratoconus, with eye rubbing as a strong behavioral risk factor. Multimodal diagnostics allowed more precise characterization of corneal biomechanical susceptibility. The clinical findings support the importance of early screening and behavioral education.
Barros et al. (Sun,) studied this question.