Amblyopia: Detection and treatment in a sample of New Zealand children

Abstract

Purpose: To investigate the efficacy of vision screening in the detection of amblyopia and its risk factors and the effectiveness of amblyopia treatment in anisometropic children aged 4-7 years, focussing on binocular outcomes and cortical processing. Methods: To establish the efficacy of preschool vision screening programmes in New Zealand, a retrospective records review was conducted of children referred to the Ophthalmology Department, South Auckland, who were screened between 1 March 2010 and 28 Feb 2011. Information collected included age at time of screening and first specialist assessment (FSA), ethinicity, social deprivation classification, unaided vision (UVA) at screening, UVA at FSA, discharge rate without treatment required, ophthalmic diagnosis, and improvement in vision. To assess the efficacy of conventional amblyopia therapy, measurements of best corrected visual acuity (BCVA), stereopsis, contrast sensitivity (CS) and global motion (GM) perception using random-dot kinematograms were conducted on a subset of 22 children with anisomtropic amblyopia before and after treatment. These results were compared to a control cohort of 28 children. Results: The vision screening programme had a high false positive referral rate (38.5%) and a low positive predictive value (41.1%). The most common reason for failing vision screening was uncorrected refractive error (N=184), however 0.7% were found to be amblyopic. Odds ratios found that the Keeler logMAR crowded test (odds ratio 1.27) was better at detecting visual defects compared with Kay picture test (odds ratio 0.35) in a hospital setting. In children with anisometropic amblyopia conventional thereapies (glasses, occlusion or penalisation) succesfully improved BCVA by more than nine letters (p<0.001) and also significantly improved contrast sensitivity in the amblyopic eye (p<0.0001). Monocular treatments improved stereopsis when measured with both Frisby (p<0.0001) and StereoFly (p<0.0001) stereotests but not with the TNO stereotest (p=0.1). Motion coherence thresholds improved in the amblyopic eye with conventional therapy (p = 0.0004). Conclusion: Pre-school vision screening is effective in the detection refractive errors and amblyopia which is treatable with conventional amblyopia therapy.