Poster 212

by in  Poster Session 3

Setting High Referral Thresholds in Non-Cycloplegic Refraction Screening:  Most Missed Children Have Borderline Refractive Risk Factors

Oliver Ehrt, MD, Prof.
Department of Ophthalmology, Ludwig-Maximilians-University
Muenchen, Germany


Introduction: Amblyopia should be detected as early as possible. However, in early live without reliable visual acuity we can only detect amblyogenic risk factors. Over referrals and over prescription of glasses must be avoided by setting high referral thresholds in non-cycloplegic refraction screening. The aim of the study was to evaluate which children will be missed by videorefractometry.

Methods: Five studies with PlusoptiX (A04-A12) from 2005 – 2015 (n=883 children aged 0.5 to 7y) were analyzed. Amblyogenic refractive error was defined by German guideline (hyperopia >3dpt, astigmatism >1dpt, anisometropia >1dpt in cycloplegic retinoscopy). Thresholds for referral were set at >2dpt, >0.75dpt and >1dpt respectively for high specificity (94%).

Results: Prevalence of risk factors were: hyperopia 11%, astigmatism 20%, anisometropia 6%. Overall sensitivity was 80%.
26% of hyperopes were missed on screening. Of those 39% had severe hyperopia > +4dpt.
24% of children with astigmatism were missed. Of those only 5% had severe astigmatism > 2dpt.
5 of 43 children with anisometropia were missed. Of those only two had anisometropia > 2dpt.

Discussion: Although astigmatism was the most frequent amblyopia risk factor in this amblyopia enriched population, only 2 children with severe astigmatism went undetected. As expected in non-cycloplegic screening, severe hyperopia was the most frequent overlooked risk factor (9 children).

Conclusion: Overall, a sensitivity of 80% is acceptable because most missed children will have mild, if any amblyopia and can be treated later when it is picked up with visual acuity testing at a later age.

Print Friendly, PDF & Email

Leave a Comment

Your email address will not be published. Required fields are marked *

Comment *