Mostrar el registro completo
NOTA: Los ítems de Digital.CSIC están protegidos por copyright, con todos los derechos reservados, a menos que se indique lo contrario.
Por favor, use este identificador para citar o enlazar a este item:
Compartir / Impacto:
|Visualizar otros formatos: MARC | Dublin Core | RDF | ORE | MODS | METS | DIDL|
The changes of shape of the human cornea with age
|Autor:||Navarro, Rafael; Rozema, J. J.; Tassignon, Marie-José|
|Fecha de publicación:||2012|
|Resumen:||The shape of the aging cornea was studied in a group of 407 nor-mal eyes. The most significant changes found were: a small increase of the curvature, eccentricity and irregularity of the anterior surface with age. For both the anterior and posterior surfaces the misalign-ment with the keratometric axis increased with age. [Introduction]: There are several studies on the mean corneal shape based on fit-ting the surface topography to a given model. The most popular models divide the surface elevation S into a regular basis surface, B (with a straightforward optical interpretation, such as spheres, coni-coids, 3 axes ellipsoids or biconics) plus a residual R = S – B which accounts for local irregularities and departures from the basis sur-face B. It is also common to fit the residual to a Zernike polynomial expansion. In an earlier study the geometry and optical proper-ties of the mean cornea were analyzed using a general 3-axis ellip-soid, so that we could determine the position (x0, y0, z0) and orienta-tion (α, β, γ) of the optical axis in the 3D space. The strength of that B model was patent as it provided significant lower fitting errors than standard (canonical) models. Here we apply a similar approach to a larger set (407 corneas), covering a wide range of ages (4 - 79 years). The measurements were taken with a Scheimpflug system (PentacamTM) which provides topographies of both front and back surfaces. The B model was also improved and generalized to a general biconic defined by 10 parameters: apex radii (RMax, Rmin), conic constants (QMax, Qmin) plus position and orientation in space. The residual R was analyzed by a 8th order Zernike polynomial expansion.|
[Discussion]: The RMS residual (fit error) shows marked differences between the anterior (<RMSant> = 5.7 μm) and posterior (<RMSpos> = 14.6 μm) surfaces (Sant and Spos). This suggests that the biconic is a good model but only for the anterior surface. Interestingly, this residual slightly increases with age for Sant but shows the opposite trend for Spos. Fig. 1 shows the evolution of the horizontal (β) and vertical (α) angles between the biconic and keratometric axes. The linear regression shows a clear trend to increase both angles (misalignment) with age. Even though the predictability is poor due to a high intersubject variability, such trend has a high statistical significance (p-values 0.0236 and 0.0033 for βant, and βpos; and even lower p-values for α). In addition there is a clear misalignment between Sant and Spos of ~2.5º for both angles. Similar plots were obtained for radii, conic constants and apex coordinates for both surfaces. The radii of the anterior surface were found to decrease slightly but significantly (P < 0.01) with age (<RMax>ant = 7.85 - 0.0047·age mm; <Rmin>ant = 7.70 - 0.0040·age), as well as the conic constant along the meridian of maximum curvature for both surfaces (<Qmin>ant = -0.30-0.0025·age; <Qmin>pos = - 0.55 -0.0024·age). Several parameters, on the other hand, remained constant with age, such as the apical radii of the back surface (<RMax>pos = 6.27 ± 0.26 mm; <Rmin>pos = 5.92 ± 0.30 mm); the conic constant along the axis of minimum cur-vature for both anterior and posterior surfaces (<QMax>ant = -0.38 ± 0.13; <QMax>pos = -0.49 ± 0.14 mm) do not change significantly with age; and Euler angle γ, corresponding with the astigmatism axis, which on average is close to vertical, but shows a large inter-subject variability. The statistical analysis presented so far is somewhat preliminary in the sense that it is based on the complete set without removing out-liers. Nevertheless, it is worth noting that this will probably affect the exact values of parameters, but the main trends and conclusions are expected to be basically the same.
[Conclusions]: These results confirm previous findings, such as the increase of the corneal power, or the tendency of the cornea to become more prolate with age. In addition, the biconic model permits to iden-tify that the conic constant has a maximum change along the merid-ian of maximum curvature (no change in the orthogonal one). A new finding (to our knowledge) is that our results suggest a significant progressive tip/tilt of the optical axis of the (best fit) biconic with age, for both anterior and posterior corneal surfaces.
|Descripción:||Póster presentado al 6th EOS Topical Meeting on Visual and Physiological Optics celebrado en Dublin (Irlanda) del 20 al 22 de agosto de 2012.|
|Aparece en las colecciones:||(ICMA) Comunicaciones congresos|
Ficheros en este ítem:
|human cornea with age.pdf||319,9 kB||Adobe PDF|