Purpose: To test the diagnostic ability of spectral domain optical coherence tomography for the detection of Parkinson disease using retinal nerve fiber layer and retinal thickness parameters. Retinal pigment epithelium produces levodopa.
Methods: Patients with Parkinson disease (n = 111) and healthy subjects (n = 200) were enrolled. The Spectralis optical coherence tomography was used to obtain retinal nerve fiber layer thickness and retinal measurements. Two linear discriminant functions (LDFs) were developed, one using retinal nerve fiber layer parameters and another using retinal thickness. A validating set was used to test the performance of both LDFs. Receiver operating characteristic curves were plotted and compared with the standard parameters provided by optical coherence tomography for both LDFs. Sensitivity and specificity were used to evaluate diagnostic performance.
Results: The Retinal LDF combines only retinal thickness parameters and provided the best performance: 31.173 + 0.026 × temporal outer - 0.267 × superior outer + 0.159 × nasal outer - 0.197 × inferior outer - 0.060 × superior inner + 0.049 × foveal thickness. The largest areas under the receiver operating characteristic curve were 0.902 for Retinal LDF. The Retinal LDF yielded the highest sensitivity values.
Conclusion: Measurements of retinal thickness differentiate between subjects who are healthy and those with advanced Parkinson disease.