Potential of Handheld Optical Coherence Tomography to Determine Cause of Infantile Nystagmus in Children by Using Foveal Morphology

Objective To investigate the feasibility of handheld (HH) ultra-high-resolution spectral-domain optical coherence tomography (SD-OCT) in young children with nystagmus, to determine its sensitivity and specificity in classifying foveal abnormalities, and to investigate its potential to determine the cause of infantile nystagmus with the use of foveal morphology. Design Prospective, case-control study. Participants and Controls A total of 50 patients with nystagmus and 50 healthy control subjects (mean age, 3.2 years; range, 0–8 years). Methods Each patient was scanned using HH SD-OCT (Bioptigen Inc., Research Triangle Park, NC) without sedation, and foveal morphology was classified into 1 of 4 categories: (1) typical foveal hypoplasia (predicting clinical diagnosis of albinism, PAX6 mutations, or isolated foveal hypoplasia); (2) atypical foveal hypoplasia (predicting achromatopsia); (3) other foveal changes (corresponding to retinal dystrophies); and (4) normal fovea (predicting idiopathic or manifest latent nystagmus). An independent interpretation of the HH SD-OCT scans by masked examiners was performed, and the sensitivity and specificity of the predicted diagnosis were calculated. Main Outcome Measures The success rate of image acquisition and sensitivity and specificity of the HH SD-OCT in classifying foveal abnormalities. Results In 94% of examinations, HH SD-OCT was successful. Twenty-three patients had typical foveal hypoplasia (category 1). Of these patients, 21 were diagnosed with albinism and 2 were diagnosed with PAX6 mutations. Five patients were classified as atypical (category 2) and diagnosed with achromatopsia. Six patients had other abnormal foveal morphology (category 3) and were diagnosed with retinal dystrophy. Sixteen patients had normal foveal morphology (category 4). Of these patients, 12 were diagnosed with idiopathic nystagmus and 4 were diagnosed with manifest latent nystagmus. Sensitivities of HH SD-OCT for classifying typical or atypical foveal hypoplasia, other abnormal foveal morphology, and normal morphology were 92.8%, 86.7%, 41.1%, and 88.4%, respectively, with specificities of 91.4%, 94.8%, 97.7% and 95.1%, respectively. Conclusions We demonstrate excellent feasibility of HH SD-OCT in the diagnosis of conditions associated with infantile nystagmus. The HH SD-OCT classification of foveal abnormalities was highly sensitive and specific. This classification was used to determine the underlying cause of infantile nystagmus. Ha