Calculation of crystalline lens power using corneal topography and whole-eye biometry with extended-depth Optical Coherence Tomography

Abstract

Purpose: To compare crystalline lens power calculated from anterior and posterior corneal topography and whole-eye biometry with crystalline lens power predicted with the method of Bennett.

Methods: Measurements were performed on both eyes of each of 10 consenting subjects with spherical equivalent refractive error ranging from -5.4D to 4.0D (mean value +/- SD = -2.1 +/- 2.7D). Anterior and posterior mean corneal radius over the central 3mm zone was measured with the Pentacam (Oculus, Arlington, WA). Corneal thickness, anterior chamber depth, lens thickness and vitreous depth were measured using a custom extended-depth Optical Coherence Tomography (OCT) system with optical switch (Ruggeri et al, Biomedical Optics Express, 2012). The effective power of the crystalline lens was calculated from the biometric data assuming a fixed value of anterior to posterior radius of curvature ratio (1.8) and an equivalent refractive index of 1.43 based on the measurements of Dubbelman et al (Vis Res, 2001). An error analysis shows that the uncertainty in the lens power predicted with this method is +/-0.6D. The lens power was compared with the lens power predicted using the Bennett method (Bennett, Ophth Phys Opt, 1988; Rozema et al, Invest Ophth Vis Sci 2011). The Bennett method calculates the corneal power from the anterior radius of curvature only and assumes a lens equivalent index of 1.416 and a ratio of anterior to posterior lens radius of curvature (1.66) corresponding to the relaxed lens of the Gullstrand eye model.

Results: Predicted lens power ranged from 21.4D to 27.8D, with a mean value (+/-SD) of 24.2+/-1.8D. The mean difference (+/-2SD) between the predicted lens power and the power obtained using the Bennett method was 3.2D+/-0.5D. When the Bennett method was modified to take into account the posterior corneal surface in the calculation of corneal power, the mean difference between methods decreased to 0.3D+/-0.1D.

Conclusions: Whole-eye biometry combined with corneal topography allows prediction of the crystalline lens power with an uncertainty of +/-0.6D, without requiring knowledge of the lens curvatures. The method of Bennett was found to significantly underestimate lens power compared to the presented method possibly due to estimation errors in the corneal power.

Keywords: 630 optical properties • 677 refractive error development • 653 presbyopia