Epithelial thickness gradients as a diagnostic biomarker for early-stage keratoconus.

PURPOSE: To evaluate peripheral-central epithelial thickness (ET) gradients in eyes with keratoconus and regular astigmatism, and to determine their potential utility as a non-invasive biomarker for disease detection in early-stage keratoconus. METHODS: This retrospective cross-sectional analyzed ET with spectral-domain OCT at the cone apex and at 3 mm in the superior, temporal, nasal, and inferior directions. Epithelial thickness gradients (ΔET) were calculated as the difference between peripheral and central ET. Statistical comparisons were performed using non-parametric tests, with significance set at p < 0.05. RESULTS: A total of 396 eyes (208 keratoconus, 188 regular astigmatism) were analyzed. Central ET was significantly reduced in KC (38.55 ± 5.42 µm) vs controls (48.78 ± 3.94 µm; p < 0.001). While peripheral thickness showed minor differences, the ET gradients (ΔET) were markedly steeper in KC across all quadrants (temporal, superior, nasal, inferior, p < 0.001). Within Stage I KC, eyes with more advanced disease exhibited steeper ΔET values, particularly in the nasal and temporal directions. Mean ΔET, demonstrated significant differences between KC (8.43 ± 4.97 µm) and controls (- 1.55 ± 1.77 µm; p < 0.001), and correlated positively with steep keratometry (r = 0.521, p < 0.001) and negatively with thinnest corneal thickness (TCT) (r = - 0.641, p < 0.001). CONCLUSION: ΔET is markedly steeper in keratoconus than in regular astigmatism-even in Stage I-and correlates with steep keratometry and TCT. With areas under the curves (AUCs) up to 0.976, ΔET provides a rapid, non-invasive epithelial-based biomarker for early detection and staging of keratoconus. Its utility for monitoring progression remains to be established in prospective longitudinal studies.