Measuring and modeling intraocular light scatter with Shack-Hartmann wavefront sensing and the effects of nuclear cataract on the measurement of wavefront error

Purpose. The purpose of this research is to determine if Shack/Hartmann (S/H) wavefront sensing (SHWS) can be used to objectively quantify ocular forward scatter. Methods. Patient S/H images from an study of nuclear cataract were analyzed to extract scattering data by examining characteristics of the lenslet point spread functions. Physical and computer eye models with simulated cataract were developed to control variables and to test the underlying assumptions for using SHWS to measure aberrations and light scatter from nuclear cataract. Results. (1) For patients with nuclear opalescence (NO) >=2.5, forward scatter metrics in a multiple regression analysis account for 33% of variance in Mesopic Low Contrast acuity. Prediction of visual acuity was improved by employing a multiple regression analysis that included both backscatter and forward scatter metrics (R2 = 51%) for Mesopic High Contrast acuity. (2) The physical and computer models identified areas of instrument noise (e.g., stray light and unwanted reflections) improving the design of a second generation SHWS for measuring both wavefront error and scatter. (3) Exposure time had the most influence on, and pupil size had negligible influence on forward scatter metrics. Scatter metric MAX_SD predicted changes in simulated cataract up to R2 = 92%. There were small but significant differences (alpha = 0.05) between 1.5-pass and 1-pass wavefront measurements inclusive of variable simulated nuclear cataract and exposure; however, these differences were not visually significant. Improvements to the SHWS imaging hardware, software, and test protocol were implemented in a second generation SHWS to be used in a longitudinal cataract study. Conclusions. Forward light scatter in real eyes can be quantified using a SHWS. In the presence of clinically significant nuclear opalescence, forward scatter metrics predicted acuity better than the LOCS III NO backscatter metric. The superiority of forward scatter metrics over back scatter metrics is attributed to the transition to mostly Mie scatter associated with larger particles and increased density. Both physical and computer cataract models can be used to provide insight into what is being captured in a S/H image and to improve instrument design. The ability of forward scatter metrics derived from S/H measurements to predict cataract severity for a longitudinal study is optimized by selecting a patient specific exposure at the initial cataract assessment to maximize the dynamic range of the system. Under various exposure and nuclear cataract conditions, the difference in RMS wavefront error between a 1.5-pass S/H image and a 1-pass S/H image does not reach the threshold of being visually significant for Photopic High Contrast and Mesopic Low Contrast acuity. (Abstract shortened by UMI.)