In 1817, Dr. James Parkinson first described the “shaking palsy” that came to bear his name. Most often thought of as a motor disease, research over the last two centuries has established that Parkinson’s disease in fact follows a long and varied clinical course, whereby sleep disturbance, loss of the sense of smell, depressive mood, and even constipation predate the appearance of the signature slow and quavering movements. But what about cognitive decline? In a large prospective analysis, M. Arfan Ikram and investigators at Erasmus University in Rotterdam, The Netherlands, found that poor cognitive function in otherwise healthy people increases their risk of developing PD years later. People with a combination of mild motor problems and low scores on cognitive tests were most at risk, facing almost three times the odds of developing PD than people with neither. The findings were published September 25 in JAMA Neurology. 

  • In a longitudinal study, poor cognitive function at baseline increased the risk of later PD.
  • Cognitive function appears to form part of prodromal PD.
  • What causes the early cognitive decline is unclear.

The results indicate that cognitive function should be considered part of prodromal PD, wrote Ethan Brown and Caroline Tanner of the University of California, San Francisco, in an accompanying editorial.

Half of people with PD develop dementia, and milder cognitive problems affect others. Typically, cognitive troubles are considered a consequence of late-stage PD, or caused by an unrelated comorbidity such as Alzheimer’s. On the contrary, emerging work suggests that cognition change can occur early as an integral part of the PD process (Ross et al., 2012; Buchman et al., 2016Chahine et al., 2016). 

A recently published case-control study from Ikram and colleagues documented a decline in executive function and processing speed as early as seven years before PD diagnosis (Darweesh et al., 2017). That work used data from the Rotterdam study, a long-running investigation of risk factors for neurological and other diseases in essentially the entire elderly population of one district in that city.

For the present study, first author Sirwan Darweesh drew on the same data, but looked for change over time. Between 2002 and 2008, the investigators measured baseline cognitive function in 7,386 participants who had no signs of parkinsonism or dementia, testing executive function, processing speed, semantic fluency, and memory. They measured manual dexterity by having subjects place small pegs in a pegboard; the task requires fine motor skills and poor performance is a harbinger of PD. They then followed the participants for an average of eight years, and as long as 15 years until they developed dementia, parkinsonism (motor symptoms, but without a diagnosis of PD), or had a diagnosis of PD. Most cases of PD were diagnosed by a geriatrician, neurologist, or research physician, and all were confirmed after expert case review. The average age of the participants was 65; just over half were women. Of the group, 79, or 1.1 percent, developed parkinsonism, and 57 of them were diagnosed with PD. Twenty-four of the 79 were also diagnosed with incident dementia, 10 before and 14 after the onset of motor symptoms, but none showed other symptoms of dementia with Lewy bodies.

Using a global cognition score to combine results from letter-digit substitution, verbal fluency, and delayed word retrieval tests, the investigators found that poorer cognitive function at baseline increased the risk of parkinsonism or PD. For each standard deviation decrease in global cognitive scores compared with the general population, the risk of parkinsonism increased by 80 percent, and PD increased by 50 percent. The correlation held regardless of age or sex, and was even seen for subjects who developed PD without dementia.

Cognitive impairment alone raised the risk of PD, but its effect increased when subtle motor changes were apparent. People with any parkinsonian signs or lower-than-normal manual dexterity had two times the risk of developing PD, about the same magnitude as those with cognitive decline. People with both motor and cognitive issues had more than a threefold increase in risk. The worse the baseline symptoms, the higher their risk.

Could the study subjects have had early PD, which was just not diagnosed at the start? To evaluate this possibility, the investigators looked at only the subjects who had been followed for eight years or more. The association between cognitive dysfunction and PD held. However, the association disappeared when they excluded subjects who had subtle motor signs at the beginning. This raises the possibility that the at-risk subjects in fact already had very early PD and were perhaps progressing slowly, wrote Brown and Tannin.

Nonetheless, it’s clear that in this cohort cognitive decline occurs early, and often in patients destined to develop PD. Altogether, almost half of participants who received a diagnosis of parkinsonism had signs of cognitive dysfunction, or subtle motor features, or both, at baseline. “Cognitive dysfunction is something that physicians need to keep in mind when considering PD, and not limit their view to the motor dysfunction,” said Ikram. Indeed, until recently, prior dementia ruled out a diagnosis of PD (Postuma et al., 2015). The current work supports the idea that changes in cognition are an integral, if variable, part of a disease that is not tightly localized, but affects the whole brain, Ikram told Alzforum.  

In addition to the global cognitive score, the investigators broke down the association by cognitive domain. Consistent with previous work, they noted that deficits involving executive function, processing speed, and semantic fluency increased risk of PD more than memory deficits, though the differences were not statistically significant. This is consistent with results from other population-based studies of PD, and contrasts with what’s seen in Alzheimer’s disease, which mainly affects memory and executive function domains.

The study did not look at cognitive tests that are more specific for PD, such as visual-spatial ability. “We designed our study 20 years back, and our cognitive test battery was more tailored toward AD than PD,” explained Ikram. “If we had administered tests that we now know are more specific for PD, I expect we would have found even stronger associations,” he said.

What might cause early cognitive issues in PD?  In their editorial, Brown and Tanner opine that the results argue against the idea that the hallmark pathology of PD, aggregated α-synuclein, always spreads from the enteric nervous system up through the vagal nerve to the substantia nigra, and only later enters the neocortex. Instead, they suggest that pathology could originate instead in non-vagal gastrointestinal inputs or the olfactory bulb, affecting the cortex earlier (Braak and Del Tredici, 2017). Another question is whether α-synuclein alone accounts for every PD symptom, or other pathologies come into play, especially in the elderly population studied, Ikram told Alzforum. 

The work emphasizes the need to understand what causes the cognitive symptoms of PD, and another paper out this week offers some ideas. Writing in the September 25 Nature Neuroscience, Luísa Lopes, University of Lisbon, Portugal, and Tiago Outeiro, Institute of Molecular Medicine, Lisbon, report that α-synuclein oligomers interact with the cellular prion protein to modulate the metabotropic glutamine receptor, mGluR5. This extrasynaptic receptor mediates glutamine toxicity in neurons and has been implicated in the toxicity of Aβ oligomers in AD (May 2016 newsSep 2013 news). Blocking mGluR5-mediated signaling or removing prion proteins reversed both synaptic and cognitive problems in synuclein transgenic mice. “This work is exciting because this [pathway] might be one of the molecular substrates for cognitive effects in PD,” Ikram said.

“These new studies suggest cognitive impairment may be a marker of PD risk, but also a symptom than can be targeted therapeutically. These are exciting times for research, but especially for PD patients and their families,” Outeiro wrote in an email to Alzforum.—Pat McCaffrey


  1. PD is no longer viewed simply as a movement disorder. Numerous non-motor features, including hyposmia, REM sleep disturbances, autonomic dysfunction, and impaired cognitive function, are now known to be present, and to often appear sequentially as disease progresses. The precise molecular determinants of disease progression are still unclear, but it is possible that the spreading of α-synuclein pathology from the periphery to the brain, or from the brain to the periphery, may be related to the progressive nature of the disease.

    Diagnosing PD, or other neurodegenerative disorders, is rather challenging, as we still lack established biomarkers that allow us to detect the initial signs of the disease, prior to the overt onset of the typical symptoms. Therefore, studies like this one now published in JAMA are exciting in that they suggest that poor cognitive function might be used as an early indicator of the risk for PD. The study suffers from some limitations, as carefully explained in the accompanying editorial, but still has numerous implications that are worth pursuing in future studies. In particular, it focuses our attention on cognitive function as an important feature in PD. Interestingly, this links nicely with the findings of our study published on the same day (Ferreira et al., 2017), in which we detail the effects of α-synuclein in the hippocampus, via an interaction with the prion protein (PrP). Our study does not solve all questions related to cognitive impairment in PD, but suggests some novel targets that can be explored to treat an important and debilitating feature in PD.

    In conclusion, these new studies suggest cognitive impairment may be an indicator of the risk for PD and also a symptom than can be targeted therapeutically. These are exciting times for research, but especially for PD patients and their families.


    . α-synuclein interacts with PrP(C) to induce cognitive impairment through mGluR5 and NMDAR2B. Nat Neurosci. 2017 Nov;20(11):1569-1579. Epub 2017 Sep 25 PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Shielding Synaptic Glutamate Receptor from Aβ Preserves Memory in Mice
  2. Glutamate Receptor Links Aβ-Prion Complex with Fyn, Synaptic Damage

Paper Citations

  1. . Pre-motor features of Parkinson's disease: the Honolulu-Asia Aging Study experience. Parkinsonism Relat Disord. 2012 Jan;18 Suppl 1:S199-202. PubMed.
  2. . Incident parkinsonism in older adults without Parkinson disease. Neurology. 2016 Sep 6;87(10):1036-44. Epub 2016 Aug 3 PubMed.
  3. . Cognition in individuals at risk for Parkinson's: Parkinson associated risk syndrome (PARS) study findings. Mov Disord. 2016 Jan;31(1):86-94. Epub 2015 Aug 21 PubMed.
  4. . Trajectories of prediagnostic functioning in Parkinson's disease. Brain. 2017 Feb;140(Pt 2):429-441. Epub 2017 Jan 12 PubMed.
  5. . MDS clinical diagnostic criteria for Parkinson's disease. Mov Disord. 2015 Oct;30(12):1591-601. PubMed.
  6. . Neuropathological Staging of Brain Pathology in Sporadic Parkinson's disease: Separating the Wheat from the Chaff. J Parkinsons Dis. 2017;7(s1):S73-S87. PubMed.

Further Reading


  1. . Past, present, and future of Parkinson's disease: A special essay on the 200th Anniversary of the Shaking Palsy. Mov Disord. 2017 Sep;32(9):1264-1310. PubMed.

Primary Papers

  1. . Association Between Poor Cognitive Functioning and Risk of Incident Parkinsonism: The Rotterdam Study. JAMA Neurol. 2017 Sep 25; PubMed.
  2. . Impaired Cognition and the Risk of Parkinson Disease: Trouble in Mind. JAMA Neurol. 2017 Sep 25; PubMed.
  3. . α-synuclein interacts with PrP(C) to induce cognitive impairment through mGluR5 and NMDAR2B. Nat Neurosci. 2017 Nov;20(11):1569-1579. Epub 2017 Sep 25 PubMed.