Influenza is associated with an increased odds of Parkinson Disease diagnosis after 10 years

1. Inpatient or outpatient clinic diagnosed influenza infection was associated with an increased odds ratio of diagnosis with Parkinson Disease 10 years later compared to control.

2. The association between influenza and Parkinson Disease diagnosis was stronger >15 years post-infection compared to 10-15 years post-infection.

Evidence Rating Level: 3 (Average)

Study Rundown: The etiology of Parkinson Disease (PD) is poorly understood, with some contributing factors such as genetics and lifestyle/exposures thought to play a role. There is some evidence of an association between infection and PD. This prospective case-control study investigated whether influenza infection was associated with an increased risk of PD at 10-years post-infection. Additionally, this study looked at whether PD risk was associated with other infections. A case group created from new incidental PD diagnosis from the Danish National Patient Registry (DNPR) was compared to a control cohort on their rates of influenza and other infection during the year and during only peak influenza activity. The 10,271 PD patients were comparable to the 51,355 control patients on most diseases (cardiovascular disease, diabetes, Crohn disease, ulcerative colitis, and lung cancer), except the PD patients had a higher rate of COPD or emphysema. Influenza infection was associated with an increased odds of diagnosis with PD, with a stronger association from more remote infections. This association remained in sensitivity analysis, which only considered influenza infection during the peak season. Patients with a non-influenza infection were also mildly associated with PD diagnosis, but unlike influenza, the more recent infections were more strongly associated with PD. One strength of the study is that it analyzed all residents of Denmark through its publicly funded data registries. Additionally, the code-based data registration in these registries is likely to have been consistent for decades, decreasing variability. A limitation of the study is that influenza infections are not laboratory-confirmed in the DNPR, potentially leading to exposure misclassification bias. This was addressed using a sensitivity analysis, which only considered peak flu season influenza infections that are less likely affected by this misclassification. H pylori or hepatitis C had a low incidence and so have an insufficient sample size for powering analysis. When investigating “other infections”, the study was necessarily limited to only severe, medically attended infections; influenza/other infections that do not present to hospital or clinic would be overlooked. For the same reason, this study may not be generalizable to less severe infections. Finally, the important confounder of smoking was not considered in this study as none of the registries recorded smoking history and so proxies like COPD, emphysema, and lung cancer were instead used. Additionally, PD family history could not be removed as a confounder.

Click to read the study in JAMA Neurology

Relevant Reading: A case of probable Parkinson’s disease after SARS-CoV-2 infection

In-Depth [case-control study]: Population-based data from the DNPR was collected on the incidence of new incidental PD diagnosis in adults aged at least 35 years old with no indication the event was drug-induced. A five-fold larger control cohort of age and sex-matched individuals without a diagnosis of PD or secondary parkinsonism was created using the Danish Civil Registration System. Patients with “infection” included any with an inpatient admission or outpatient clinic contact that lead to any diagnosis of infection. A secondary sensitivity analysis only considered influenza infection during peak “flu season”, when there was 3-fold more influenza diagnoses than average (more than 96 per month) by the DNPR. Of the 10,271 eligible PD diagnosis, 38.7% were female, 86.3% were older than 60 years when diagnosed, and the average age was 71.4±10.6 years old. Fewer PD patients compared to the 51,355 controls had COPD or emphysema (6.5% vs 7.7%). Influenza infection >10 years prior to PD diagnosis was associated with increase odds of PD diagnosis (1.73, 95% CI = 1.11-2.71), including in influenza infection that only occurred during peak season (OR = 1.52, 95% CI = 0.80-2.89). The longer between influenza infection and PD diagnosis, the greater the odds of PD diagnosis (10-15 years: OR = 1.33, 95% CI = 0.54-3.27; >15 years: OR = 1.91, 95% CI = 1.14-3.19). 31.4% of the PD patients and 30.1% of the controls had some infection. Having any given infection was mildly associated with a diagnosis of PD (OR = 1.09, 95% CI = 1.04-1.14). Unlike with influenza, more recent infections with pneumonia, gastrointestinal infection, miscellaneous bacterial infections, septicemia, or male genital infections were more associated with PD diagnosis (<5 years: OR = 1.27, 95% CI = 1.18-1.36) compared to more remote infections that were not associated with PD diagnosis (>10 years: OR = 1.04, 95% CI = 0.98-1.10). Urinary tract infections were associated with an increased odds of PD diagnosis more so at >10 years (OR = 1.19, 95% CI = 1.01-1.40).

Image: PD

©2021 2 Minute Medicine, Inc. All rights reserved. No works may be reproduced without expressed written consent from 2 Minute Medicine, Inc. Inquire about licensing here. No article should be construed as medical advice and is not intended as such by the authors or by 2 Minute Medicine, Inc.