AbstractThe relationship between dopaminergic treatment and freezing of gait (FOG) in Parkinson’s disease (PD) is complex: levodopa is the most effective symptomatic treatment for FOG, but long-term pulsatile levodopa treatment has also been linked to an increase in the occurrence of FOG. This concept, however, continues to be debated. Here, we compared the occurrence of FOG between a levodopa-naive PD cohort and a levodopa-treated cohort. Forty-nine treatment-naive patients and 150 levodopa-treated patients were included. The time since first motor symptoms was at least 5 years. Disease severity was assessed using the MDS-UPDRS part III. Occurrence of FOG was assessed subjectively (new freezing-of-gait-questionnaire) and objectively (rapid turns test and Timed Up-and-Go test). The presence of FOG was compared between the levodopa-treated and levodopa-naive groups using a chi-square test of homogeneity. We also performed a binomial Firth logistic regression with disease duration, disease severity, country of inclusion, location of measurement, and executive function as covariates. Subjective FOG was more common in the levodopa-treated cohort (n = 41, 27%) compared to the levodopa-naive cohort (n = 2, 4%, p < 0.001). The association between FOG and levodopa treatment remained after adjustment for covariates (OR = 6.04, 95%Cl [1.60, 33.44], p = 0.006). Objectively verified FOG was more common in the levodopa-treated cohort (n = 21, 14%) compared to the levodopa-naive cohort (n = 1, 2%, p = 0.02). We found an association between long-term pulsatile levodopa treatment and an increased occurrence of FOG. Future studies should further explore the role of nonphysiological stimulation of dopamine receptors in generating FOG, as a basis for possible prevention studies.