Development of a prediction formula of Parkinson disease severity by optical coherence tomography.

The aims of this study were to assess the peripapillary retinal nerve fiber layer (RNFL) thickness in patients with Parkinson's disease (PD), to determine its correlation with disease severity, and to define a simple biomarker for predicting clinical severity. One hundred two eyes from 52 patients affected by PD were compared with 97 eyes from 50 age-comparable controls. In all patients, peripapillary RNFL thickness was measured by optical coherence tomography (OCT). We used the Unified Parkinson's Disease Rating Scale (UPDRS) total score and measured responses in the on medication state. Eyes from patients with PD had a statistically significant decrease in average peripapillary RNFL thickness compared with control eyes (P < 0.001). This reduction was observed in every quadrant (inferior, superior, nasal [P < 0.001], and temporal [P = 0.017]) in patients with PD. Furthermore, a strong inverse correlation was found between the PD severity measured according to the UPDRS score and the average peripapillary RNFL thickness (r = -0.615; P < 0.001) and PD duration (r = -0.303; P = 0.002). From these results, we defined a regression equation that predicts the UPDRS score from the above-mentioned variables: UPDRS = 81.6 + 29.6 * log PD duration (years) - 0.6 * RFNL thickness (μm). We observed that, as the evolution and severity of PD progress, the peripapillary RNFL layer thickness, as evaluated by OCT, gradually diminishes. These results suggest that the average peripapillary RNFL thickness measured by OCT might be useful as a biomarker to detect the early onset and progression of PD.