BMC Ophthalmology
Peripheral pachymetry, eccentricity, and composite tomographic-biomechanical indices for screening forme fruste/suspect keratoconus
By Wei Zhang, Ao Li, Hanyu Jiang and 5 others.
Abstract
Early identification of forme fruste or suspect keratoconus is challenging because curvature-based parameters may remain near normal. This study evaluated Pentacam HR- and Corvis ST-derived parameters across normal, forme fruste/suspect keratoconus, and clinical keratoconus eyes. This single-center retrospective cross-sectional study included eyes examined at Beijing Tongren Hospital from January 2020 to December 2022. Eyes were classified as normal (NOR), forme fruste/suspect keratoconus (DUN), or clinical keratoconus (KCN). Pentacam-derived structural parameters and Corvis ST-derived composite indices were compared among groups. Generalized estimating equations and one-eye-per-patient analyses were used to assess inter-eye correlation. Receiver operating characteristic curves with 95% confidence intervals and DeLong comparisons were generated for NOR-versus-DUN discrimination. The study included 600 eyes: 272 NOR eyes from 136 patients, 51 DUN eyes from 51 patients, and 277 KCN eyes from 168 patients. Kmax was similar between NOR and DUN eyes (45.14 ± 1.64 D vs. 45.24 ± 1.79 D, P = 0.994), but higher in KCN eyes (55.47 ± 9.20 D; both P < 0.001). PachyMin decreased stepwise from NOR to DUN and KCN eyes (533.16 ± 29.42, 506.45 ± 26.49, and 461.88 ± 46.54 μm; all pairwise P < 0.001). For NOR-versus-DUN discrimination, EccInf had an AUC of 0.814 (95% CI, 0.727–0.902), followed by Cor.Vol (0.770; 95% CI, 0.695–0.844). Among composite indices, TBI had the numerically highest AUC (0.759; 95% CI, 0.680–0.839), followed by BAD-D (0.751; 95% CI, 0.677–0.824), CBI (0.668; 95% CI, 0.583–0.753), and SSI (0.666; 95% CI, 0.588–0.744). Sensitivity analyses did not materially change the interpretation. Pentacam HR and Corvis ST parameters showed distinct patterns across the NOR-DUN-KCN spectrum. Peripheral structural parameters and composite indices helped distinguish clinically defined NOR from DUN eyes, supporting layered screening interpretation rather than parameter superiority or longitudinal disease staging.



