Predicting long-term visual outcome in children with birth weight under 1001 g.

PURPOSE: To examine the visual acuity development of extremely low birth weight children and to determine factors that are predictive of long-term outcome.

METHODS

This is a prospective observational longitudinal cohort study of children with birth weight less than 1001 g. One hundred thirty-nine children were recruited. Retinopathy of prematurity (ROP) examinations were graded according to the International Classification for ROP. Grating acuity was assessed monocularly with Teller acuity cards. All children were assessed before 24 months corrected age; 123 of the cohort had a grating acuity assessment at over 3 years. For the children who were capable, an assessment of recognition acuity was measured with the Electronic Visual Acuity system.

RESULTS

Data are presented for the right eye and the ages reported are adjusted for prematurity to allow comparison with normative data. Initial grating acuity was compared with the late grating and recognition acuity, but in both cases analysis showed no statistically significant association. However, the relative risk analysis showed that, if the slope was abnormal, there was a 5.5 times higher risk of abnormal recognition acuity. Eyes with zone 1 disease were associated with a worse visual acuity outcome, but zone 1 disease also occurred more frequently in children with lower birth weight and gestational age.

CONCLUSIONS

Early measurements of visual acuity may be misleading in terms of the visual prognosis. The factor that was most predictive of a poor late visual acuity outcome was the rate of development, as calculated by the slope of the early visual acuity measurements.