Ex vivo magnetic resonance imaging of crystalline lens dimensions in chicken

PURPOSE A reduction in the power of the crystalline lens during childhood is thought to be important in the emmetropization of the maturing eye. However, in humans and model organisms, little is known about the factors that determine the dimensions of the crystalline lens and in particular whether these different parameters (axial thickness, surface curvatures, equatorial diameter, and volume) are under a common source of control or regulated independently of other aspects of eye size and shape. METHODS Using chickens from a broiler-layer experimental cross as a model system, three-dimensional magnetic resonance imaging (MRI) scans were obtained at 115-microm isotropic resolution for one eye of 501 individuals aged 3-weeks old. After fixation with paraformaldehyde, the excised eyes were scanned overnight (16 h) in groups of 16 arranged in a 2x2x4 array. Lens dimensions were calculated from each image by fitting a three-dimensional mesh model to the lens, using the semi-automated analysis program mri3dX. The lens dimensions were compared to measures of eye and body size obtained in vivo using techniques that included keratometry and A-scan ultrasonography. RESULTS A striking finding was that axial lens thickness measured using ex vivo MRI was only weakly correlated with lens thickness measured in vivo by ultrasonography (r=0.19, p<0.001). In addition, the MRI lens thickness estimates had a lower mean value and much higher variance. Indeed, about one-third of crystalline lenses showed a kidney-shaped appearance instead of the typical biconvex shape. Since repeat MRI scans of the same eye showed a high degree of reproducibility for the scanning and mri3dX analysis steps (the correlation in repeat lens thickness measurements was r=0.95, p<0.001) and a recent report has shown that paraformaldehyde fixation induces a loss of water from the human crystalline lens, it is likely that the tissue fixation step caused a variable degree of shrinkage and a change in shape to the lenses examined here. Despite this serious source of imprecision, we found significant correlations between lens volume and eye/body size (p<0.001) and between lens equatorial diameter and eye/body size (p<0.001) in these chickens. CONCLUSIONS Our results suggest that certain aspects of lens size (specifically, lens volume and equatorial diameter) are controlled by factors that also regulate the size of the eye and body (presumably, predominantly genetic factors). However, since it has been shown previously that axial lens thickness is regulated almost independently of eye and body size, these results suggest that different systems might operate to control lens volume/diameter and lens thickness in normal chickens.