Analysing the ocular biometry of new-born infants

To model and analyse the ocular biometry of new-born infants.

This work is based on previously published data of a cohort of 66 new-born infants aged 0-3 days. After exclusion of seven myopic subjects, the available retinoscopy, keratometry and ultrasound biometry data were analysed, along with calculated parameters such as lens power and whole eye power.

Male infants have significantly flatter corneas that female infants (Mann-Whitney U test, p < 0.001), which was associated with a difference in gestational age between genders (multiple linear regression; p = 0.043). No other gender-based differences were seen. Corneal curvature (Pearson, r = 0.575; p < 0.001), lens power (r = -0.681; p < 0.001), and anterior chamber depth (r = 0.654; p < 0.001) were all correlated to axial length, but not refraction (r = -0.114; p = 0.42). Most ocular parameters were associated with gestational age (linear regression analysis; p < 0.05), rather than birth length, birth weight, fertilization method or parental myopia (all p > 0.05), suggesting scaled eye growth during the last weeks before birth. Multivariate Gaussian analysis demonstrated that a statistical eye model can be defined that generates synthetic data that is significantly equal to the original data (non-parametric Mann-Whitney test for equality; all p < 0.05), with similar variability (non-parametric Levene test; all p > 0.05).

The eye undergoes a scaled growth until birth, at which time male and female infants have similar values. The models presented may serve as an early biometry reference.