To identify demographic and clinical factors associated with freezing of gait (FoG) in a multi-centric Mexican Parkinson’s disease (PD) cohort.

FoG is common in PD with a prevalence of 54.8%. FoG represents a disabling phenomenon with a decay in patient’s quality of life. It also represents a risk factor for falls, fractures, and loss of independence. Identifying factors associated with FoG aids development of early intervention strategies.

An epidemiological, observational, cross-sectional, comparative, and analytical study was designed using the multi-centric PD cohort of the PSG-Mexico 2017-2018. A total of 306 patients were included. A univariate analysis was used to find associations between clinical and demographic variables and the presence of FoG (MDS-UPDRS items 2.13 and 3.11). Significantly associated variables were used to construct a multivariate regression model to determine predictive factors.

28.1% (86/306) reported symptoms of FoG. Motor PD subtypes: PIGD-PD (OR=12.31, p=.001) and TD (OR=0.06, p=.001), use of levodopa (OR=2.74, p=.015), use of dopamine agonists (OR=3.46, p=.06), use of pramipexole or rotigotine (OR=12.41, p=.001), use of COMT inhibitors (OR=5.425, p=.009), presence of hallucinations (OR=1.94, p=.03), apathy (OR=2.55, p=.004), sleeping problems (OR=1.74, p=.03), drooling (OR=4.36, p=<.001), and “off” painful dystonia (OR=7.21, p=.01) were significantly associated with FoG.

Multivariate analysis proved that PIGD (OR=20.1, p=<.0001), TD (OR=4.8, p=.02), drooling (OR=3.34, p=.004), disease duration (OR=1.11, p=.002), and severity of symptoms (OR=1.055, p=<.0001), were significant predictors for the presence of symptoms of FoG. Absence of symptoms of cognitive impairment showed to be a protective factor (OR= -1.054, p=.004). The model showed an accuracy 80.7%, sensitivity 71.4%, specificity 83.1%, PPV 52.3%, and NPV 91.8%.  

Prevalence and associated factors with FoG are similar to previously reported data. Our results suggest that motor subtype, disease duration, severity of disease predict patients at risk of FoG. A previously non-reported factor appeared to be “drooling”.