Characteristics of Keratoconus and Pellucid Marginal Degeneration in Mean Curvature Maps

Abstract

To characterize keratoconus and pellucid marginal degeneration (PMD) on mean curvature maps. Observational case series. A retrospective analysis of clinical records of 19 keratoconus eyes and eight PMD eyes. Keratometric mean curvature maps were computed from a placido-disk-based corneal topography system. The peak location and amplitude of distortions were quantified by fitting two-dimensional Gaussian functions. Computer simulations of keratoconic corneal topography were used to help interpret the results. Keratoconus and PMD eyes had a characteristic localized increase in convexity (cone) on mean curvature maps. Computer simulations showed that the peak on the mean curvature map accurately represented the peak of the cone-like distortion. However, the peak location and appearance of the same conic distortion on axial and tangential maps were greatly influenced by coexisting astigmatism. The Gaussian function provided an excellent model of keratoconus on mean curvature maps, with a mean cross-correlation of 0.80. According to Gaussian fitting, the vertical cone peak locations were -1.10 +/- 0.43 mm (mean +/- SD) in keratoconus eyes and -1.94 +/- 0.53 mm in PMD eyes (P < .01). The locations of peaks on axial and tangential maps differed considerably from those on mean curvature maps. Advanced keratoconus tended to have higher cones (amplitude 13.31 +/- 6.87 diopters) than early keratoconus (amplitude 10.65 +/- 1.56 diopters). Gaussian fitting of mean curvature maps accurately quantifies the peak location and amplitude of the cone in keratoconus and PMD eyes. This new topographic analysis might be useful in the diagnosis and tracking of corneal ectatic diseases.