Gender effect on time to levodopa-induced dyskinesias.

Levodopa-induced dyskinesias (LID) are commonly observed during long-term treatment of patients with Parkinson's disease (PD). The impact of non-pharmacological factors on the latency to LID appearance is not known. The aim of the paper was to identify factors associated with time to appearance of LID. Consecutive PD patients treated with levodopa (n = 155) were included in this historical prospective analysis and LID and non-LID groups were compared. The relationship between possible risk factors and the time of LID onset was explored using the Kaplan-Meier method and the Cox multivariate regression model, controlling for the confounding effects of gender, age of disease onset, time to initiation of levodopa treatment, and history of smoking. Patients with LID (57.4%) were significantly younger at disease onset and had a slightly longer latency from diagnosis to levodopa treatment than those without; disease duration and age had no effect on LID appearance. Female gender was associated with a shorter time to LID and the median time to LID was 6 years for males and 4 years for females (p = 0.004). In the multivariate survival analysis a younger age of onset of PD and a longer time from diagnosis to levodopa treatment initiation were also associated with a shorter time to LID appearance (p = 0.030 and 0.036, respectively). Female gender is associated with a significantly shorter latency to LID appearance. Younger age at PD diagnosis and a longer time until starting levodopa are associated with both higher likelihood to develop LID, and a shorter latency until LID were observed.