Refractive astigmatism and the toricity of ocular components in human infants.

PURPOSE: Many studies have characterized astigmatism in infancy, but few have been longitudinal or contained ocular component data. This study characterized the frequency, orientation, and longitudinal change with age of infant astigmatism. Additional factors investigated were the influence of early astigmatism on emmetropization and its relation to corneal and lenticular toricity.
METHODS: Three hundred two infants were enrolled in the study. Of these, 298 provided data for at least one visit at 3 +/- 1 months, 9 +/- 1 months, 18 +/- 2 months, and 36 +/- 3 months. Testing included cycloplegic retinoscopy (cyclopentolate 1%), video-based keratophakometry, and ultrasonography over the closed eyelid.
RESULTS: Astigmatism > or =1.00 DC was common at 3 months of age (41.6%) but decreased in prevalence to 4.1% by 36 months (p < 0.0001). The most common orientation was with-the-rule at 3 months (37.0% compared with 2.7% for against-the-rule) but against-the-rule at 36 months (3.2% compared with 0.9% for with-the-rule). Most of the change in the average value of the horizontal/vertical component of astigmatism (J0) occurred between 3 and 9 months (-0.26 +/- 0.36 D; p < 0.0001) with no significant change between 9 and 36 months (-0.05 +/- 0.36 D; p=0.09). Spherical equivalent refractive error was not correlated with J0 at 3 and 9 months (R=0.002, p=0.48 and R=0.001, p=0.56, respectively). The two were only weakly correlated at 18 and 36 months (R=0.06 for each age, p <0.0001, p=0.0002, respectively). Changes in spherical equivalent between 3 and 9 months were unrelated to either the initial value of J0 (partial R for J0=0.0001; p=0.85) or the change in J0 (partial R for change in J0=0.0031; p=0.31). Across all the ages, corneal toricity was with-the-rule, and lenticular toricity was against-the-rule (produced by the toricity of the posterior lens surface). The cornea and anterior lens surface became more spherical with age, contributing to the shift away from with-the-rule refractive astigmatism. Toricity of all the refractive surfaces became less variable with age.
CONCLUSIONS: Consistent with many reports, astigmatism was common in early infancy but decreased in prevalence with age, particularly when with-the-rule in orientation. The reduction in percentage of infants with astigmatism appeared to be caused by decreases in the toricity of the cornea and the anterior lens combined with decreases in the variability of corneal and lenticular surfaces. Astigmatism in infancy appeared to be unrelated to emmetropization of spherical equivalent refractive error.