Abstract
Purpose :
The population incidence of negative dysphotopsia (ND) varies between 0.20% for severe cases and 25% for mild to moderate ones. While several biometric factors and mechanisms were previously evaluated, the current theoretical models cannot predict the ND occurrence for a given ocular biometry. The present study describes the development of a theoretical method enabling the potential prediction of the ND occurrence.
Methods :
Within a prospective study, post-operative data sets were acquired from two patients with ND complaints and implanted with monofocal IOLs. The measurements included pupil size, anterior chamber depth and axial length (IOLMaster 700), anterior corneal topography (Atlas), IOL tilt and decentration (Purkinje Meter) and refractive errors (Topcon). A Harms tangent screen subjective method was used to localize in the visual field the perceived shadow (HvF). The Liou-Brennan eye model (Liou and Brennan, JOSA A 1997) was customized using the biometrical data to build personalized sequential eye models (Zemax OpticStudio 14). With the same biometry, non-sequential (NSC) eye models were constructed with the IOL’s geometry modelled as a 3D CAD object. To account for the edge glare relative intensity, the edge frosting effect was retro-fitted by a Lambertian scattering surface. Using a polychromatic wide divergence angle light source, irradiance maps were computed using a polar coordinates detector and the outcome was compared with the HvF data.
Results :
The mean IOL tilt was 3.54 ± (-7.63 – 12.38) degrees and the mean decentration was 0.21 ± (0.02 – 0.46) mm. The evaluated patients reported the presence of shadows in the peripheral visual field between approximately 62 and 85 degrees. The average residual refraction difference between the theoretical and the measured data was 0.54D (0.34 – 0.73D) for spherical equivalent (M) and 0.25D (0.06 – 0.43D) for vertical astigmatism (J0). For the evaluated ocular biometries, the irradiance maps revealed a close similarity to the HvF ones in terms of angular localization of the shadows.
Conclusions :
The study proposed an ND evaluation method based on patient specific ocular biometry. For a given IOL geometry and implantation position, these models enable the localization in the visual field and the origin identification of ND.
This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.