Is retinal optical coherence tomography an imaging biomarker for Parkinson disease?

Abstract

Purpose: To assess the peripapillary retinal nerve fiber layer (RNFL) thickness in patients with Parkinson disease (PD) by optical coherence tomography (OCT), determining whether there is any correlation among these measurements and the duration and severity of the disease in order to obtain an imaging biomarker for PD.

Methods: Peripapillary RNFL thickness was measured by OCT in 102 eyes from 52 patients diagnosed with PD (mean age: 65.98 ± 8.7 years; range: 45 - 84). These data were compared with those from 97 eyes from 50 age-matched healthy controls. Independent two-samples Student's t-test for parametric numbers was used for determining whether the values of RNFL thickness differ between PD eyes and controls. The Pearson correlation coefficient was used to assess whether there is any correlation between the decreased peripapillary RNFL thickness and both disease duration and PD severity. We also conducted a multiple linear regression model to identify the most influential variables in the UPDRS score and the values of the differential intercept coefficients to develop a formula that will allow us to make predictions.

Results: PD patients showed a statistically significant decrease in average peripapillary RNFL thickness (93±9 µm, range: 74 to 117) compared with control eyes (101±7 µm, range: 81-117) (p<0.001; Student's t test). Besides, a statistically significant reduction of peripapillary RNFL thickness was observed in every quadrant: inferior (p<0.001), superior (p<0.001), nasal (p<0.001) and temporal quadrant (p=0.017) in PD patients. Furthermore, we found a significant inverse correlation between disease duration and average peripapillary RNFL thickness (r= -0.303; p=0.002). Likewise, there were differences among average peripapillary RNFL thickness and disease severity measured by UPDRS (r= -0.615; p<0.001). Thus, we developed a regression model in order to define an equation that predicts the UPDRS score from the average RNFL thickness.

Conclusions: PD patients showed a decreased peripapillary RNFL thickness evaluated by OCT. We also noted that the further evolution and severity of PD, the lower RNFL thickness. These results suggest that axonal degeneration would be present in the retina of PD patients and our regression equation to predict UPDRS score from peripapillary RNFL thickness measured by OCT might be used as an imaging biomarker in PD detection or disease progression.

Keywords: 550 imaging/image analysis: clinical • 613 neuro-ophthalmology: optic nerve • 612 neuro-ophthalmology: diagnosis