Share this article on:


Laino, Charlene


Honolulu, HI — Patients who suffer a head injury with loss of consciousness are at increased risk of developing Parkinson disease years later, according to a study of twins presented at the AAN Annual Meeting here in April.



“While it's well established that post-traumatic parkinsonism may occur following cumulative head trauma, this study suggests that short, brief periods of unconsciousness – minutes to hours – also place one at risk,” said chief investigator Samuel M. Goldman, MD, a clinical research scientist at the Parkinson's Institute in Sunnyvale, CA.

The good news, he said, is that the study revealed a 30- to 40-year window between head injury and the development of Parkinson disease, “so there is a huge opportunity to intervene.”

This is not the first study to suggest a link between mild head injury and Parkinson disease, Dr. Goldman noted. Several case-control studies found an increased frequency of prior head injury in Parkinson patients, and more epidemiological studies are showing a significant positive association between the two. But, he added, other trials have found no significant link.

“What's novel about our research is that we studied discordant Parkinson disease twins, about half of whom were identical,” Dr. Goldman said. “Since twins share many behavioral traits and environmental experiences as well as similar or identical genotypes, they offer a powerful design that controls for all sorts of potential confounders, such as genetics and socio-economic status.”

Back to Top | Article Outline


What is notable about the research, according to other experts at the meeting, is that an association between head injury and Parkinson disease makes sense from a pathophysiological point of view.

“This is a plausible hypothesis,” said Andrew Grandinetti, PhD, an epidemiologist at the Pacific Biomedical Research Center at the University of Hawaii at Manoa, who has done extensive work with Parkinson patients in the Honolulu-Asia Aging Study. “The biological changes that occur in head trauma are similar to those we see in Parkinson disease.”

Why does head injury make a patient susceptible to developing the neurodegenerative disease marked by tremors, rigidity, bradykinesia, and loss of postural reflexes?

According to Dr. Goldman, the same characteristics of the inflammatory cascade that is initiated by mild to moderate head injury can be seen in the Parkinson brain at autopsy, with impairment of the blood-brain barrier, edema, leukocyte infiltration, microglial activation, excitotoxicity, free radical production, lipid peroxidation, and finally, upregulation and release of inflammatory cytokines.

“White blood cells are recruited from the periphery and became activated in the brain,” Dr. Goldman said.“The brain starts secreting cytokines and you also see impaired mitochondrial function.”

In Parkinson disease brains at autopsy, “you see increased levels of the same cytokines, impaired motor function, active inflammation in the substantia nigra, and mitochondrial dysfunction,” he said.

Back to Top | Article Outline


The MPTP model offers further proof that a single remote toxic insult can result in active ongoing nigral inflammation and Parkinson disease year later, said Dr. Goldman, referring to the dozens of young people who exhibited symptoms of an advanced form of the progressive disease after illegally injecting an impure form of synthetic meperidine (Demerol) in the early 1980s. MPTP, or 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine, is a neurotoxin that induces a parkinsonian-type syndrome in animals.

“Twenty years later, we can see inflammation in their brains,” Dr. Goldman said. “It shows that it is not unreasonable to think that a process that happened years ago is contributing to parkinsonism now.”

Back to Top | Article Outline


For the study, Dr. Goldman's team identified twin pairs in which at least one twin had Parkinson disease from the National Academy of Sciences/National Research Council World War II Veterans Twin Cohort, a database of 31,848 Caucasian male veterans born between 1917 and 1927.

“We started screening for Parkinson disease by phone in the early 1990s,” he said. “Then we went out and directly examined anyone we thought might have the disease based on the telephone interview.”

Once cases were identified, the men and their twins were asked to participate in another telephone interview in which they were asked if they had ever sustained a head injury that had resulted in a loss of consciousness or amnesia, the length of unconsciousness, whether they were hospitalized, and the date of the injury.

They identified 142 twin pairs discordant for Parkinson disease, including 60 homozygous pairs. About three-fourths, or 186, of the twins completed the interview.



After excluding head injuries occurring after diagnosis, 29 men (15.6 percent) reported at least one prior head injury, for which 17 (7.8 percent) were hospitalized. The mean age at first head injury was 26.8 years, and the duration between head injury and Parkinson disease symptoms ranged from two to 70 years, with a mean of 38 years. The mean age of Parkinson disease diagnosis was 64.2 years.

Back to Top | Article Outline


In the 26 pairs in which the researchers were able to obtain complete histories on both twins, the twin with Parkinson disease was three times more likely to have sustained a prior head injury, Dr. Goldman said. “This was statistically significant.”

The magnitude of increased risk was similar for identical and fraternal twins, but these values did not reach statistical significance, perhaps because of the smaller number of cases, he said.

The researchers then repeated the analysis, truncating observations in both twins 10 years prior to when the affected twin developed Parkinson disease to ensure that the injury was not due to a Parkinson-related fall.

Again, the twin who had sustained a head injury was three times more likely to have developed Parkinson disease, Dr. Goldman reported.

The study also showed that having two head injuries was worse than one, with twins who had two head injures having a 4.7-fold increased risk of Parkinson disease.

When the researchers divided the injured twins into four categories based on the length of unconsciousness – fewer than 5 minutes, 5 to 29 minutes, 30 to 59 minutes and an hour or longer – no significant association was found between duration of unconsciousness and Parkinson disease. Similarly, hospitalization was not significantly associated with Parkinson disease.

Back to Top | Article Outline


Dr. Goldman acknowledged that, like all retrospective studies, this trial was susceptible to recall bias. “It's possible that cases had selective ‘enhancement’ of head injury recall,” he said. “But given that we asked if they were unconscious, you'd think it's harder to be wrong about that.”

Also, the persistence of the association – when the analysis was limited to head injuries that occurred at least 10 years prior to diagnosis – reduces the likelihood that early“pre-clinical” Parkinson disease might have caused the head injury, he said.

Dr. Grandinetti agreed that steps were taken to reduce recall bias. But, he added, it would be useful to do a follow-up study looking at hospital records to confirm that the patients really lost consciousness. “A review of medical records would add to the study,” he said.

Cynthia Comella, MD, Associate Professor of Neurology at Rush-Presbyterian-St. Luke's Medical Center in Chicago, IL, also called for follow-up studies. “This is an intriguing observation that now needs to be looked at statistically.”

©2003 American Academy of Neurology