Corneal Cross-Linking Induces Increased Corneal Refractive Index, Which Generates Measurement Error of Central Corneal Thickness.

Purpose: This study investigated the impact of a higher-than-expected corneal index of refraction (nc) on measured central corneal thickness (CCT) by Scheimpflug tomography and anterior-segment OCT. Design: Theoretical analysis with prospective cohort study validation. Participants: Twenty-four eyes from 23 participants met the criteria for inclusion in data analysis. Methods: Two models were developed to represent side-view imaging with Scheimpflug geometry (model 1) and interferometric imaging in OCT (model 2). For both models, predicted CCT was calculated for various nc values, and CCT measurement error was quantified. For validation, CCT was measured using Scheimpflug tomography and OCT in prospectively recruited subjects with keratoconus before and after corneal cross-linking (CXL). Central corneal thickness was compared between baseline and follow-up with paired t-test for each device and between devices using Wilcoxon signed rank tests for nonparametric data with significance threshold of P < 0.05. Device-specific equations for CCT and n were solved iteratively at follow-up for those subjects with minimal difference in CCT at baseline to account for measurement error. Main Outcome Measures: Central corneal thickness and nc. Results: In side-view model 1, a negative relationship between predicted CCT and nc was found, as well as a negative association between CCT measurement error and nc. OCT model 2 produced a positive association between predicted CCT and nc and a positive association between CCT measurement error and nc. As validation, CCT was stable (mean ± standard deviation ΔCCT = -0.57 ± 8.56 microns; P = 0.75, paired t-test) between the pre- and post-operative measurements made with OCT. However, CCT measured with Scheimpflug tomography was significantly lower after CXL than before it (-11.0 ± 11.3 microns; P < 0.001). Conclusions: The models indicate that an unknown increase in nc after CXL results in underestimation of CCT from Scheimpflug-based devices and overestimation of CCT from OCT. The larger the difference in measured CCT between devices, the larger the increase in corneal refractive index that has occurred. Clinical validation aligns with both models, indicating the actual CCT is between those reported by each technology. Corneal cross-linking may alter the accuracy of diagnostic devices that rely on nc to determine CCT. Financial Disclosures: Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.