Air pollution and Parkinson's disease: A systematic review and meta-analysis up to 2018

Background: Recent epidemiological findings investigate effects of exposure to air pollution on neurodegenerative disease. We performed a systematic review and meta-analysis to investigate the association between air pollution exposure and Parkinson's disease (PD). Methods: We performed an extensive literature search in PubMed and Google Scholar databases and further searched for unpublished results in conference abstracts until November 2018. We identified 102 unique studies referring to air pollution and PD, from which 15 were included in the meta-analyses. We applied random-effects models to combine risk estimates and investigated between studies heterogeneity. We assessed publication bias through plots and the Eggers test in cases of sufficient number of studies. We assessed associations accounting for multi-pollutant exposures and effect modification patterns by sex and smoking habits. Results: We identified 13 reports investigating associations of PD with long-term exposure to regulated air pollutants whilst two reported associations for short-term exposure to PM2.5. The pooled relative risk (RR) for incidence of PD following an increase in long-term exposure for 10 mu g/m(3) in PM2.5 was 1.06 (95% Confidence Interval (CI): 0.99, 1.14) and in NO2 1.01 (95%CI: 0.98, 1.03), while for 5 ppb increase in O-3 1.01 (95% CI: 1.00, 1.02) and for 1 mg/m(3) in CO 1.34 (95% CI: 0.85, 2.10); the pooled RR for a hospital admission due to PD after a 10 mu g/m(3) increase in PM2.5 short-term exposure was 1.03 (95% CI: 1.01, 1.05). There was high heterogeneity between study-specific results for most of the analyses, attributed to different populations under study. Effects were robust to multi-pollutant adjustment while there were indications of higher particles' effects among non smokers. Conclusions: We found weak evidence for an association between air pollution, mostly originating from traffic, and PD. Although meta-analysis increases power to detect small associations in rare outcomes, further research is needed to elaborate our suggestive associations. Such results are of public health significance since population aging in developed countries is expected to increase incidence of PD.