Purchase this article with an account.
Stanley Park, Sarah Hilkert, Julie Lange, Andrew Hendershot, Amit Tandon, Mark Slabaugh, Julie Meier; Optokinetic Reflex: a Novel Screening Tool for Autism Spectrum Disorders. Invest. Ophthalmol. Vis. Sci. 2016;57(12):1548.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Autism spectrum disorders are pediatric developmental disorders that detrimentally affect social interaction and communication. Early diagnosis and subsequent intervention is critical for improved behavioral outcomes. Current diagnostic measures such as M-CHAT are subjective and place responsibility for diagnosis on parents and caregivers. Eye movement abnormalities associated with cerebellar pathology are well established in autistic patients. These include changes in gaze holding, saccades and smooth pursuits. The optokinetic nystagmus (OKN) drum is an easily adapted tool that can examine these particular movements. The purpose of this study is to assess the viability of the OKN reflex as a possible objective screening tool for autism spectrum disorders.
We compared the OKN response of 6 children aged 3 to 5 years with autism spectrum disorders and 17 age-matched controls. Eye movements were analyzed using a video recording of eye movements taken from a smartphone which was simultaneously displaying an OKN stimulus. The video recordings were sent to three masked ophthalmologists who gave a response of either "atypical" or "typical." Findings were recorded as "typical" if the child demonstrated a classic OKN reflex- slow pursuit in direction of stripe movement with a fast reflex saccade. Findings were recorded as "atypical" if the child demonstrated lack of gaze fixation, absence of stereotypic behavior or no response.
OKN response analysis resulted in a sensitivity of 66.7% and specificity of 82.3%. The positive predictive value was 57.14% and the negative predictive value was 87.50%. Interobserver correlation was 0.475 using a mean of paired Cohen’s kappa.
OKN response may be an objective screening tool that can serve as a useful diagnostic aid for autism spectrum disorder. Specificity and negative predictive value were relatively high while interobserver correlation was moderate. A larger sample size and training for screeners may improve the reproducibility of this technique.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.
This PDF is available to Subscribers Only