according to a report published in Movement Disorders. Associations were found for broad-spectrum antibiotics and those that act against anaerobic bacteria and fungi. The timing of antibiotic exposure also seemed to matter.
In a nationwide case-control study, Finnish researchers compared data on antibiotic use in 13,976 individuals diagnosed with Parkinson’s disease between 1998 and 2014 with antibiotic-use data from 40,697 controls. The strongest connection with Parkinson’s disease risk was found for oral exposure to macrolides and lincosamides (adjusted odds ratio up to 1.416). After correction for multiple comparisons, exposure to antianaerobics and tetracyclines 10-15 years before the index date, and antifungal medications 1-5 years before the index date were positively associated with Parkinson’s disease risk. In post hoc analyses, further positive associations were found for broad-spectrum antibiotics.
Tuomas H. Mertsalmi, MD, from the Helsinki University Hospital and coauthors reported that this was the first study to explore a possible connection between antimicrobial use and Parkinson’s disease.
“In Parkinson’s disease, several studies have described alterations of gut microbiota composition, and changes in fecal microbiota abundance have been found to be associated with gastrointestinal and motor symptoms,” they wrote.
Commenting on the delay between the exposure and diagnosis for the most strongly associated antimicrobials, the authors noted that this 10-15 year lag was comparable with what has been found between the peripheral initiation of Parkinson’s disease and its motor manifestation.
“This would also explain the lack of association between antibiotic exposure 1-5 years before index date – if antibiotic exposure could induce or contribute to the pathogenesis of Parkinson’s disease in the gastrointestinal tract, it would probably take several years before the clinical manifestation of Parkinson’s disease,” they wrote.
With regards to the association seen for sulfonamides and trimethoprim – which was 1-5 years before the index date – they speculated this could reflect treatment for urinary tract infections, which individuals with Parkinson’s disease might be more susceptible to in the prodromal phase of the disease.
The authors noted that infectious disease has also been associated with Parkinson’s disease, and that their analysis did not include information about why the antimicrobial agents were prescribed. However, they pointed out that the associations were only for certain antibiotic classes, which makes it unlikely that the association was related to greater burden of infectious disease among individuals with Parkinson’s disease.
The pattern of associations supports the hypothesis that effects on gut microbiota could link antibiotics to Parkinson’s disease. “The link between antibiotic exposure and Parkinson’s disease fits the current view that in a significant proportion of patients the pathology of Parkinson’s disease may originate in the gut, possibly related to microbial changes, years before the onset of typical Parkinson’s disease motor symptoms such as slowness, muscle stiffness, and shaking of the extremities. It was known that bacterial composition of the intestine in patients with Parkinson’s disease is abnormal, but the cause is unclear. Our results suggest that some commonly used antibiotics, which are known to strongly influence the gut microbiota, could be a predisposing factor,” said lead investigator Filip Scheperjans, MD, PhD, from the department of neurology at Helsinki University Hospital.
The findings may have implications for antibiotic prescribing practices in the future, said Dr. Scheperjans. “In addition to the problem of antibiotic resistance, antimicrobial prescribing should also take into account their potentially long-lasting effects on the gut microbiome and the development of certain diseases.”
The study was funded by the Finnish Parkinson Foundation, the Finnish Medical Foundation, the Maire Taponen Foundation, and the Academy of Finland. One author declared relevant patents and his position as founder and chief executive of a private company. No other conflicts of interest were declared.
SOURCE: Mertsalmi TH et al. Mov Disord. 2019 Nov 18. doi: 10.1002/mds.27924.