Abstract
Purpose:
To correlate visual acuity (VA) and light increment sensitivity (LIS) with birth weight (BW) and gestational age (GA) in school aged former premature children, and to study the prevalence of strabismus, amblyopia and refractive errors in the same patient cohort.
Methods:
One hundred fifty five children (125 former prematures originally participating in a prospective postnatal telemedical study with digital wide-field retinal imaging, and 30 age-matched term born children) were enrolled in the GRANIT study (Giessen RetinAl maturatioN study in former premature InfanTs) at 6-12 years. The premature participants were categorized by the results of the ROP-screening: prematures with laser-treated ROP (t-ROP; n=10), with spontaneously resolved ROP (sr-ROP; n=40), without ROP (no-ROP; n=75). In addition, they were categorized according to birth weight (BW) into <1000g, 1000-1500g, >1500g, and GA into ≤28 Weeks (W), >28 - ≤32 W, and >32 W.<br /> All participants underwent a comprehensive examination protocol, including orthoptic examination for strabismus, amblyopia, and refraction. VA was assessed with Landolt C and ETDRS letters, and LIS measured at 0°, 2,8° and 8° in the visual field using the MP1 (Nidek, Italy). Ophthalmological and neonatal intensive care parameters from the initial study were included in the analysis.
Results:
Prevalence of strabismus and amblyopia was highest in the t-ROP group (50%, 20%) and lowest in the no-ROP group (17%, 9%). Participants in the t-ROP group were high myopic (mean refraction -4.4Dpt), while participants in no-ROP (mean refraction 0.62Dpt) and sr-ROP (mean refraction 0.75Dpt) were not. Central LIS (0°) was significantly lower in all groups except for BW >1500g (ANOVA, p<0.05). No significant differences were found for LIS at 2.8° and 8° in all groups. All children born ≤28W GA had significantly reduced VA. Multiple comparison test (Holm-Sidak or Dunn’s method) among the groups BW < 1000g or GA ≤28 W did not reveal significant treatment-related differences.
Conclusions:
Prevalence of strabismus, amblyopia and refractive errors increase with severity of ROP, while low BW and GA were associated with significantly decreased central visual function, regardless of the severity of retinal changes. Central LIS and VA data correlate and increase the robustness of data sets describing foveal function.