RESUMEN
PURPOSE: To assess the association between clinical and perinatal characteristics and subfoveal choroidal thickness in 9-year-old children. METHODS: The study included data from the population-based Generation R cohort, whose participants underwent cycloplegic refractometry, ocular biometry, height, weight and subfoveal choroidal thickness measurements using a swept-source optical coherence tomography (SS-OCT) instrument. Birth parameters were obtained using medical records. Statistical analyses were performed using multivariate regression models adjusted for age, ethnicity and sex. RESULTS: A total of 1018 children (52.5% girls, 47.5% boys) with a mean age of 9.9 ± 0.3 years and a mean cycloplegic spherical equivalent refraction of 0.80 ± 1.1 D in boys and 0.81 ± 1.4 in girls were eligible for analysis. The subfoveal choroid was 17 µm thicker in girls (298 ± 60.6 µm) than in boys (281 ± 55.0 µm; p < 0.001), a difference of 9.1 µm persisting after adjustment for age, ethnicity and axial length (p = 0.017). Subfoveal choroidal thickness decreased with increasing ocular axial length (-16.2 µm/mm, 95% CI -21.2 to -12.4, p < 0.001) and with increasing myopic refraction (-10.0 µm/D, 95% CI 6.8-13.1; p < 0.001, adjusted for age, ethnicity, axial length and sex) while it increased with increasing body height (1.3 µm/cm, 95% CI 0.8 to 1.9, p < 0.001). Additionally, choroidal thickness increased with increasing birthweight (13.0 µm/kg; 95% CI 0.006-0.020; p < 0.001) and increasing size for gestational age (8.2 µm/kg; 95% CI 4.6-11.8; p < 0.001). Smoking up until the time that pregnancy became known was associated with a thinner choroid (p = 0.016). There was no detectable effect of alcohol consumption. The distributions of axial length, refraction and choroidal thickness were narrower than in older populations. CONCLUSION: The subfoveal choroid was thicker in girls than in boys, and higher body height, higher birthweight and larger size for gestational age were associated with a thicker subfoveal choroid. The implications of these findings for myopia development need further evaluation in longitudinal studies.