Therapeutic Interventions for Vascular Parkinsonism: A Systematic Review and Meta-analysis.

Background: Vascular parkinsonism (VP) is defined as the presence of parkinsonian syndrome, evidence of cerebrovascular disease, and an established relationship between the two disorders. However, the diagnosis of VP is problematic, particularly for the clinician confronted with moving from diagnosis to treatment. Given the different criteria used in the diagnosis of VP, the effectiveness of available therapeutic interventions for this disease are currently unknown.

Methods: To assess the clinical response of all published therapeutic interventions for VP that have been reported in the literature, we conducted a systematic review looking for VP subjects treated with any therapeutic intervention. To clarify the prevalence of responsiveness to levodopa among VP subjects, we conducted a meta-analysis of 17 observational studies retrieved with the search criteria of our review. Also, four studies were included in a second analysis to explore if nigrostriatal lesion affected the prevalence of levodopa response in VP subjects. Relevant articles were identified from MEDLINE, Scopus, and Web of Science published until June 2017.

Results: 436 non-duplicate citations were identified for screening, 107 articles were assessed for eligibility, and only 23 observational studies were included in this review. No randomized clinical trials were found. Four different therapies were found in the literature; among them, levodopa was the only one repetitively reported. The calculated event rate of levodopa response in VP subjects was of 0.304 [95% confidence interval (CI) of 0.230-0.388]. The overall odds ratio for good response to levodopa in VP with lesion in the nigrostriatal pathway vs. no lesion in the nigrostriatal pathway was 15.15 (95% CI: 5.2-44.17).

Conclusion: Despite the lack of randomized controlled trials, results of this systematic review and meta-analysis show that VP subjects, as operationally defined here, have a low response rate to levodopa; nigrostriatal lesion could be used as a proxy predictor of levodopa response in VP subjects. Other therapies seem to be co-adjuvant. Randomized controlled trials with a clear definition of VP are necessary to be able to assign positive or negative predictive values to available treatments and to recommend any of the therapeutic interventions for these subjects.