Predicting the onset of freezing of gait in Parkinson's disease.

BACKGROUND

Freezing of gait is a debilitating symptom of Parkinson's disease associated with high risks of falls and poor quality of life. While productive therapy for FoG is still underway, early prediction of FoG could help high-risk PD patients to take preventive measures. In this study, we predicted the onset of FoG in de novo PD patients using a battery of risk factors from patients enrolled in PPMI cohort.

METHODS

Baseline characteristics were compared between subjects who developed FoG (68 patients, 37.2%, pre-FoG group) during the five-year follow up and subjects who did not (115 patients, 62.8%, non-FoG group). A multivariate logistic regression model was built based on backward stepwise selection of factors that were associated with FoG onset in the univariate analysis. ROC curves were used to assess sensitivity and specificity of the predictive model.

RESULTS

At baseline, age, PIGD score, cognitive functions, autonomic functions, sleep behavior, fatigue and striatal DAT uptake were significantly different in the pre-FoG group relative to the non-FoG group. However, there was no difference in genetic characteristics between the two patient sets. Univariate analysis showed several motor and non-motor factors that correlated with FoG, including PIGD score, MDS-UPDRS part II score, SDMT score, HVLT Immediate/Total Recall, MOCA, Epworth Sleepiness Scale, fatigue, SCOPA-AUT gastrointestinal score, SCOPA-AUT urinary score and CSF biomarker Abeta. Multivariate logistic analysis stressed that high PIGD score, fatigue, worse SDMT performance and low levels of Abeta were independent risk factors for FoG onset in PD patients.

CONCLUSIONS

Combining motor and non-motor features including PIGD score, poor cognitive functions and CSF Abeta can identify PD patients with high risk of FoG onset.