|
|
HK J Paediatr (New Series)
Vol 1. No. 2,
1996
|
|
HK J Paediatr (New Series) 1996;1:194
Proceedings of Clinical Meeting
A Simple Method for Early Detection of Refractive Errors in Infants and Young Children by Photorefraction
CC Lam, CM Tsui, CK Ho, HL Mak CC Lam*, CM Tsui, CK Ho, HL Mak Children Assessment Service*, Department of Health and Ophthalmology, Tuen Mun Hospital
HK J Paediatr (new series) 1996;1:193-206 Annual Scientific Meeting Hong Kong Paediatric Society December 9, 1995 | Amblyopia is one of the commonest causes of visual loss in children, and as this is much more likely to develop in children with refractive errors, their early detection and management is highly desirable. Visual assessment in very young children is however difficult and often postponed, sometimes resulting in irreversible visual impairment by the time of diagnosis. This study examines the application of light refraction principles on a simple instrument for refractive error detection. Eccentric photorefraction involves use of a regular camera with 2 sets of flashes set at its top and side respectively. Two photographs are taken of the child's eyes with each flash set off in turn. Light that passes in and back out through the lens will show a light image whose size is indicative of its refractive power. Prior calibration of different light image sizes versus their known refractive status will enable future table conversion of patient data. 107 children from Tuen Mun Child Assessment Centre were each first assessed with photorefraction and then by an ophthalmologist with full cycloplegic retinoscopy. Cut-off points for positive diagnoses are based on recommendations for young children. 18 children's photographs were technically unsatisfactory for scoring. Analysis of retinoscopy diagnoses within this group showed that there is no significant difference between the proportions of positive and negative results (p = 0.25-0.50), and that exclusion of these cases from the main analysis should not bias results. The remaining 89 photorefraction findings were tested against retinoscopy diagnoses and the agreement in results was statistically significant (p<0.001) yielding a sensitivity of 88% (95% confidence interval 80-96%), specificity of 100% and positive predictive value of 100%. This method is simple, quick, does not require special clinic set-up and could be operated after brief training. Photorefraction may thus be a sensitive, specific and cost effective tool for mass screening for refractive errors in infants and very young children in well baby and general paediatric outpatient clinics.
|
|
|