Objective: We aimed to predict fall risk in PD patients using SBR data obtained from PPMI.

Background: While dopamine is currently considered to be the primary neurotransmitter implicated in PD motor symptoms, its role in axial symptoms is unclear. Falls can reflect axial dysfunction such as walking, balance, and gait related difficulties, and are associated with significant morbidity and mortality in PD patients. Accurately predicting fall risk using data obtained only in early stages of motor PD may help clinicians to proactively prevent them.

Method: Dopamine Transporter (DaT)-Single Photon Emission Computed Tomography (SPECT) Striatal Binding Ratio (SBR) and Determination of Falls longitudinal data from 244 PD patients were obtained from the PPMI database. We created random forest models in Python to predict presence of falls, and falls with injury in the past year in PD patients using initial SBR values for the right and left caudate and putamen, and SBR speed of decline, defined as difference between first and second DaTScan divided by months in between.

Results: [figure1]
The model predicted the following outcomes: No Fall, Fall w/o Injury, and Fall w/ Injury. F₁ score for No Fall = 0.604. F­₁ score for Fall w/o Injury = 0.4. F₁ score for Fall w/ Injury = 0.211. Macro-averaged F₁ score = 0.405.

Conclusion: SBR is shown to be a poor predictor of falling in PD patients. While the model was able to accurately identify patients who did not fall with relative success, it correctly classified patients who did fall, and whose falls resulted in injury less than 50% of the time. These results indicate that while SBR is a known predictor of parkinsonian disorders associated to dopaminergic denervation, it has not yet shown significant help to predict certain clinical outcomes in patients, and highlights the need to identify biomarkers to accurately predict important clinical outcomes in patients such as risk of falling.

« Back to 2022 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/dopaminergic-denervation-as-predictors-of-falls-in-pd-patients/