*Article* **The E**ff**ect of Ca**ff**eine on the Risk and Progression of Parkinson's Disease: A Meta-Analysis**

**Chien Tai Hong 1,2, Lung Chan 1,2,\*,**† **and Chyi-Huey Bai 3,4,5,\*,**†


Received: 6 May 2020; Accepted: 22 June 2020; Published: 22 June 2020

**Abstract:** Coffee and caffeine are speculated to be associated with the reduced risk of Parkinson's disease (PD). The present study aimed to investigate the disease-modifying potential of caffeine on PD, either for healthy people or patients, through a meta-analysis. The electronic databases were searched using terms related to PD and coffee and caffeinated food products. Articles were included only upon fulfillment of clear diagnostic criteria for PD and details regarding their caffeine content. Reference lists of relevant articles were reviewed to identify eligible studies not shortlisted using these terms. In total, the present study enrolled 13 studies, nine were categorized into a healthy cohort and the rest into a PD cohort. The individuals in the healthy cohort with regular caffeine consumption had a significantly lower risk of PD during follow-up evaluation (hazard ratio (HR) = 0.797, 95% CI = 0.748–0.849, *p* < 0.001). The outcomes of disease progression in PD cohorts included dyskinesia, motor fluctuation, symptom onset, and levodopa initiation. Individuals consuming caffeine presented a significantly lower rate of PD progression (HR = 0.834, 95% CI = 0.707–0.984, *p* = 0.03). In conclusion, caffeine modified disease risk and progression in PD, among both healthy individuals or those with PD. Potential biological benefits, such as those obtained from adenosine 2A receptor antagonism, may require further investigation for designing new drugs.

**Keywords:** caffeine; Parkinson's disease; risk; progression; meta-analysis

#### **1. Introduction**

Parkinson's disease (PD) is one of the most common neurodegenerative diseases, second only to Alzheimer's disease (AD). Risk factors for PD include genetic mutations, environmental toxins, and lifestyle [1]. An epidemiological study reported some protective factors for PD worldwide, such as female sex, physical activity, and smoking [2]. The consumption of coffee or caffeinated food is associated with the reduction of the risk of PD. Patients with PD are less frequent habitual consumers of caffeinated food [3,4]. The consumption of either tea or coffee exhibited similar effects on the reduction of the risk of PD [5]. In a similar manner, the protective effect of coffee was also noted in dementia and AD [6], whereby caffeine reversed the cognitive impairment and decreased the amyloid burden in transgenic AD mice model [7].

Caffeine is an adenosine A2A receptor antagonist [8]. Different types of adenosine receptors (A1, A2A, A2B, and A3) are widely distributed in the brain. Adenosine A2A receptors are coupled with G-proteins and exclusively expressed in dopaminergic neurons. The activation of adenosine A2A receptors causes an increase in intracellular cAMP levels and the extracellular release of glutamate, resulting in neural excitotoxicity [9]. The neuroprotective effects of caffeine involved the antagonism of the adenosine A2A receptor, down-regulating the down streaming phosphatidylinositol 3-kinase (PI3K)/protein kinase B (AKT) signaling pathway, and avoiding excessive calcium releasing-related neurotoxicity and neuroinflammation [10], which has been experimentally demonstrated in several in vivo models of PD [11–14].

Whether caffeine can reduce the risk and halt the progression of PD remains unclear. In large-scale cohort studies, caffeine consumption was inconsistently associated with a low risk of PD during follow-up [15–18]. However, among patients diagnosed with PD, the administration of caffeine tablets did not modify the disease course [19]. Furthermore, caffeine metabolism varies among patients with PD [20], thus potentially resulting in inconsistent protective effects. This study investigated the association between caffeine and PD progression. Considering disease progression was the primary temporal outcome, only cohort studies rather than case–control studies were included herein, because case–control studies cannot delineate this temporal association.
