Purpose: To test the reliability and accuracy of the 2Win Refractometer Corneal Reflexes App (CR App) in detecting manifest strabismus in pediatric patients.
Methods: Prospective study involving 167 children with suspected strabismus (mean age 7.6 years; SD = 3.0, range 2-14 years) undergoing the CR App ocular alignment assessment (Pediatric Ophthalmologist) versus the alternate cover test with prism ocular alignment assessment for distance (Orthoptist) as the gold standard. The AAPOS 2013 guidelines for the detection of manifest strabismus in primary position (> 8 PD) were used.
Results: Total sensitivity, specificity, PPV, and NPV for the CR App were 79.2%, 86.2%, 86.4%, and 78.9%, respectively. The overall inconclusive rate was 17.9%, but was 36.3% in children younger than 5. Sensitivity and PPV for vertical deviations were poor (33.3% and 12.5%, respectively). The accuracy of the CR App regarding the degree (in prism diopters) of manifest deviations was tested with the Wilcoxon signed rank sum test: correlation with the gold standard was good for esodeviations (p value = 0.765, not statistically significant) and poorer for exodeviations (p value = 0.056, still not statistically significant), whereas a significant difference (p value = 0.0001) was observed for vertical deviations.
Conclusion: The CR App showed good sensitivity, specificity, PPV, and NPV for manifest strabismus > 8PD in accordance with the AAPOS 2013 guidelines; sensitivity and PPV were poor for vertical deviations. The accuracy of the CR App was good for horizontal deviations, but poor for vertical deviations. The inconclusive result rate was high in younger children.
Keywords: 2Win refractometer (Adaptica); CR App (Corneal Reflexes App); Manifest strabismus; Nonrefractive amblyopia risk factors.