Background Levodopa is the mainstay of treatment for Parkinson's disease (PD) but is associated with potentially-disabling drug-induced motor fluctuations and dyskinesias. Previous studies have suggested that both of these complications develop in about 40% patients treated with levodopa for 4–6 years 1. However, most of these studies were not in truly-representative PD cohorts and there is little evidence as to which factors influence the onset of these complications.

Methods We used data from an established, community-based, incident cohort of individuals with Parkinson's disease with ongoing, long-term follow-up. For this analysis we only included patients who have been treated with dopaminergic drugs. We gathered demographic and clinical data at baseline and each year (including asking patients about development of complications, taking detailed drug histories and determining UPDRS motor and complications of treatment scores). From drug histories we calculated the cumulative levodopa-equivalent dose (LED) at 3 years from diagnosis (mg levodopa-equivalent dose×number of days of treatment×10⁻⁵; one unit is equivalent to nearly 100 mg levodopa a day for 3 years). We performed survival analysis with Kaplan-Meier curves and Cox proportional hazards modelling to assess which factors independently influence the development of motor complications.

Results 195 incident PD patients (59.5% male, mean age at diagnosis 72.1 years) were included in the analysis. 135 patients (69%) started treatment with levodopa, 44 (23%) with a dopamine agonist (24 of whom subsequently received levodopa) and 16 (8%) with a monoamine oxidase B inhibitor (9 of whom subsequently received a dopamine agonist and 12 levodopa). Mean daily LED was 234mg (SD 125). 32 patients (16.4%) had developed motor fluctuations and 43 (22.1%) had developed dyskinesias (see Figure 1A and 1B) after a mean follow-up duration from diagnosis of 52 months (and mean treatment duration of 43 months). Only 11 patients (5.6%) developed dyskinesias which were problematic to the extent that treatment changes were warranted. Kaplan-Meier analysis showed that about 30% had motor fluctuations and about 40% had dyskinesias by 5 years of treatment with dopaminergic therapy. Factors independently associated with the development of motor fluctuations were age at diagnosis in years (Hazard ratio [HR] 0.96 [95% confidence interval 0.93–0.99], p=0.007) and cumulative LED (HR 1.53 [1.23–1.91], p<0.001). The development of dyskinesias was independently associated with female sex (HR 2.02 [1.09–3.75], p=0.02), motor UPDRS score at diagnosis (HR 1.03 [1.00–1.06], p=0.04) and cumulative levodopa-equivalent dose (HR 1.37 [1.10–1.07], p=0.004).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Kaplan-Meier estimates of the probability that PD patients on a dopaminergic therapy will be free from motor fluctuations (A) and dyskinesias (B).

Conclusion In a truly-representative PD cohort, after 5 years of treatment, about 30% develop motor fluctuations and about 40% develop dyskinesias, most of which were mild. Younger age and higher exposure to dopaminergic therapy are associated with motor fluctuations and female sex, higher parkinsonian impairment at diagnosis and higher cumulative LED are associated with dyskinesia onset.