Assessing Retinal Structure in Patients with Parkinson's Disease

The retina is an extension of the central nervous system (CNS), and ocular symptoms can precede manifestations of CNS disorders. Given that several neurodegenerative conditions that affect the brain exhibit ocular symptoms, the retina may be an accessible biomarker to monitor disease progression. Dopamine, the key neurotransmitter related to Parkinson's disease (PD), is contained in amacrine and interplexiform cells, which reside in specific retinal layers. Understanding how loss of dopaminergic cells affects retinal anatomy could be relevant for monitoring disease progression. Here, our objective is to evaluate retinal structure (foveal pit morphology and thickness) in patients with PD.

Thirty-three Caucasian subjects diagnosed with PD and 40 age-matched Caucasian control subjects underwent retinal imaging with spectral-domain optical coherence tomography (SD-OCT). Axial length measurements were used to correct the lateral scale of each macular volume scan. From these corrected volumes, foveal morphology was quantified with previously described algorithms, and Early Treatment Diabetic Retinopathy Study (ETDRS) grids of retinal thickness were generated and incorporated into a logistic regression model to predict PD.

Interocular foveal morphology measurements were highly symmetrical in PD patients and control subjects. There were no significant differences in foveal pit morphology between PD patients and control subjects. Using a model incorporating sex and axial length corrected ETDRS regions, we generated a receiver operating characteristic curve with a C-statistic of 0.80.

Our study, which to our knowledge is the first to properly scale OCT measurements when quantifying retinal thickness, demonstrates that PD patients retain foveal symmetry between eyes. When constructing a model to predict PD, sex, along with the center 1 mm and temporal outer ETDRS regions, were significant predictors of PD. In addition to proper scaling of OCT measures, gender and racial differences in retinal anatomy should be considered in building future predictive PD models when using OCT.