Abstract
Purpose :
To measure refractions and eye lengths in the periphery of eyes with a range of refractive errors and to devise numerical models of these eyes consistent with both sets of measurements.
Methods :
Central and peripheral cycloplegic refractions were obtained (Grand Seiko Open-field Autorefractor) in 22 healthy subjects (12–39 years old) at eccentricities from 40° nasal (N) to 40° temporal (T), by directing their gaze using a fixation target at 1 m. Then, eye lengths and corneal curvatures were measured (LenStar Optical Biometer) at N/T horizontal eccentricities 0°, 10°, 20°, and 30°; optical path lengths of the cornea, anterior chamber, lens, vitreous, and retina were exported and transformed into geometrical lengths using custom software. A complete optical model was developed (Zemax optical design code) that incorporated all of these parameters.
Results :
Relative to central refractive errors, eccentric peripheral refractive errors were hyperopic in myopes (0.49±1.10 D at 30° T), but myopic in emmetropes (-0.93±0.47 D at 30° T) and hyperopes (-1.96±1.62 D at 30° T). These refractions corresponded to central and peripheral axial length measures (myopes at 0°: 25.24±1.73 mm, 30° T: 24.65±1.68 mm; emmetropes 0°: 23.46±1.06 mm, 30° T: 23.02±1.14mm; hyperopes 0°: 23.81±0.23 mm, 30° T: 23.10±0.01 mm). For one hyperope, one emmetrope, and one myope, optimized Zemax eye models, supplied with the measured corneal curvatures, path lengths, and individual refractions, captured most of the optical behavior of the eye with only (1) a small tilt (~2°) on the lens and (2) a fitted posterior lens curvature. With a finite diameter (2 mm) refractor beam, coma was the dominant remaining aberration. In the examined cases, significant differences were found in the "shape-factor" (ratio of anterior to posterior curvature at given optical power) of the lens.
Conclusions :
In the modeling, a single fitted parameter, representing the posterior curvature of the lens, produced substantial agreement among off-axis refractive profiles and eye lengths. Thus, the resultant, self-consistent model eyes can derive previously unmeasurable lenticular parameters that may be helpful in clinical studies. The shape factor for the lens appears to depend upon the refractive state of the eye which, in turn, may be implicated in refractive development; longitudinal study of refractive development will be needed to determine the predictive validity of such an indicator.
This is a 2021 ARVO Annual Meeting abstract.