Although more than four decades have passed since British neurosurgeon McDonald Critchley described severe memory problems in a sample of boxers—presumably the result of repeated head trauma—the relationship between brain injury and Alzheimer's disease (AD) has proven difficult to unravel. For years, there was scant evidence that head injuries endured by nonboxers posed an increased risk for AD; although epidemiologic studies linking the two conditions began appearing in the 1980s, negative studies continue to be published with enough frequency to ensure that the issue remains controversial. And while recent reports have revealed neuropathologic changes in some injured brains that bear a remarkable resemblance to AD, the interpretation and implications of these findings remain uncertain.
Multiple researchers have overwhelming data that boxers have a high rate of dementia. Playing other sports where head injuries are common may also increase the risk of developing dementia, or perhaps cause it to develop at an earlier age. In an article published in NeuroSurgery in 2005, University of North Carolina
Nonetheless, some researchers believe that decoding the AD–brain injury connection will have implications far beyond the possibility of warding off dementia in head trauma patients. "It gives you insight into what may be the base mechanism for Alzheimer's disease," said Gareth W. Roberts, who coauthored several studies on the relationship between the two conditions before becoming Chief Executive Officer of the Cambridge, UK bioinformatics firm Proteom. "And once you have a feeling for what the core pathologic process is, you can ask, 'What kinds of things might stop that?'"
Other investigators envision similarly lofty applications. "I look upon head injury as a paradigm for understanding environmental risk factors for neurodegenerative diseases in general," said John Q. Trojanowski, Professor of Pathology and Laboratory Medicine at the University of Pennsylvania
WHAT HAPPENS AFTER BRAIN INJURY?
Recent studies have provided "very strong evidence that there is a connection between head trauma and at least some of the pathology of Alzheimer's disease," Dr. Trojanowski said. For example, in a report at the recent World Alzheimer Congress 2000, Steven T. DeKosky, and colleagues at the University of Pittsburgh Medical Center reported findings from neocortical samples taken from brain injury patients one and three days after injury. The samples, which were obtained by surgical resection and compared with postmortem samples from neurologically normal controls, revealed increases in amyloid precursor protein (APP), apolipoproteins E and D, and ß-amyloid (Aß). Many of the Aß deposits "had morphologic characteristics of classic amyloid plaques in AD," Dr. DeKosky and colleagues reported.
Several studies have found that Aß deposition occurs in a third of fatal head injury cases, even in children who survived only a few hours. The Aß is generally distributed throughout the brain; its presence does not correlate with cerebral contusions, increased intracranial pressure, or intracranial hematomas. Neurofibrillary tangles may also occur. The nature of pathology depends in part on injury severity—tangles do not seem to occur after mild trauma—but "I don't think there's much insight into how severe the injury has to be" to trigger AD-like pathology, Dr. Trojanowski said.
These changes make sense, Dr. Roberts said, if one accepts the view that Alzheimer's disease is largely an inflammatory process. APP is found in synapses, he noted, and "one of the things we do know happens after brain injury is synaptic remodeling." Moreover, electron microscopy shows that synapses are involved in amyloid plaque formation. This may be a repair process of some sort, he said, but "instead of being shut down when it is appropriate, it just carries on. It becomes a bit like arthritis, where mechanisms that should help you instead become chronically activated and disabling."
It should be noted, however, that the neuropathology of brain injury is by no means a carbon copy of the changes that occur in AD. For example, levels of growth inhibitory factor are increased in reactive astrocytes in experimentally induced brain injury, whereas these levels are reduced throughout the brain in AD. Moreover, the neocortical distribution of neurofibrillary tangles is more superficial in former boxers with dementia pugilistica, or punch drunk syndrome, than in AD patients.
HOW LARGE IS THE RISK?
Despite the array of pathologic evidence linking the disorders, the relationship between AD and head injury remains unsettled from an epidemiologic standpoint. Findings reported last year from the Rotterdam Study, for example, found no increased risk of AD in subjects with a history of head injury. Nonetheless, positive studies outnumber negative ones, and head injury is "becoming more accepted as being associated with the risk of AD," said Brenda L. Plassman, Director of the Program in Epidemiology of Dementia at Duke University Medical Center
In a new report Dr. Plassman and colleagues performed telephone screening of more than 2,000 World War II veterans who had been hospitalized for head injury, pneumonia, or puncture wounds in 1944 or 1945. Subjects who screened positive for possible dementia underwent a three-hour exam that included neuropsychologic testing, neurologic examination, and DNA collection.
From 1940s armed forces hospital records, the researchers were able to estimate the severity of each subject's head injury, based on the occurrence of amnesia or skull fracture and the duration of unconsciousness. The findings revealed that "the more severe the injury, the greater the risk of AD and dementia"; the relative risks (compared with controls) ranged from about 2 for moderate head injury to 4 for severe injury. The findings are consistent with those from most other positive epidemiologic studies, Dr. Plassman said. "It's rather striking that all of these studies used different samples, methods, and criteria for head injury, yet all have odds ratios that are pretty close." The relative risk of AD after head injury is roughly similar to that reported for subjects heterozygous for the APOE*E4 allele, she added.
THE GENETIC CONNECTION
The role of genetic vulnerability is suggested by the fact that only a subset of brain injury patients develops amyloid pathology. Many investigators believe APOEgenotype is the key culprit. "There is a clear relationship between having an APOE*E4 allele and your likelihood of developing plaques after a head injury," Dr. Roberts said. "So in a sense the APOE–head injury story gives you the first genetic–environmental interaction in a neurologic disease." This relationship is consistent with APOE's proposed role in the maintenance and repair of neuronal membranes, synaptogenesis, and other processes. Indeed, researchers at the University of Glasgow reported earlier this year that the densities of the ß-amyloid peptides Aß-42 and Aß-40 in head injury patients were related in a dose-dependent manner to APOE*E4 endowment.
However, it is possible that in many cases head injury doesn't induce AD-like pathology so much as accelerate its arrival. A 1989 retrospective study found that a history of head injury was associated with earlier onset of AD. And a report from the Mayo Clinic suggested that AD rates were not elevated among subjects with a history of head trauma, but that the injury hastened the time to AD onset by about eight years.
FUTURE INTERVENTIONS
Can prompt, appropriate treatment after brain injury reduce or prevent the development of AD-like pathology? While the mouse model for AD that Dr. Trojanowski and others have been using has yielded interesting findings, the rodents' lack of tau pathology and "quirky" behavior did not allow for ideal testing of therapeutic interventions.
Until now, researchers have lacked a good animal model for studying the development of Alzheimer's disease. The transgenic mice used in the CNDR contain the human gene that produces the Ab protein. With the aid of techniques developed at the Penn Head Injury Center
Even without head trauma, these mice would eventually develop Ab plaques later in life. With the trauma, they produce symptoms of Alzheimer's disease at a remarkably increased rate.
"Here, we can clearly see a direct cause and effect relationship between repetitive concussions and Alzheimer's," said John Q. Trojanowski. "Using the head trauma model in these mice represents a step forward in our ability to understand the basic molecular mechanisms behind Alzheimer's disease. More importantly, we believe this model system can be used to screen for new medications in the search for a cure."
At present, "aside from telling football players and soccer players to either not play or to wear a helmet, there's not much in the way of interventions," Dr. Trojanowski noted. However, as researchers gain a better understanding of the relationship between trauma, risk factors, and genetic vulnerability, medical advice could theoretically be targeted to a patient's profile: "If you're APOE*E4 homozygous, you should really think twice about playing football. If you're heterozygous for APOE*E4, you'd better wear a helmet and take vitamin E and aspirin for the rest of your life."
Moreover, the inflammatory model of AD pathogenesis offers obvious potential for intervention. Epidemiologic studies have found a reduced rate of AD among people who regularly used nonsteroidal anti-inflammatory drugs. And a study in the Journal of Neurosciencefound that ibuprofen reduced not only inflammation, but Aß plaque burden in a transgenic mouse model for AD.
An interesting recent 2008 year study at Washington University in St. Louis and the University of Milan
"We were trying to understand why traumatic brain injury increases the risk of Alzheimer's disease," said Dr. David Brody of Washington University
The classic theory is that such injuries increase the amount of amyloid beta, which may accelerate the development of sticky clumps of amyloid plaque that is a hallmark of Alzheimer's disease.
To study the classic theory related to the increasing amount of amyloid beta, accelerating the development of sticky clumps of amyloid plaque, the teams placed a small catheter into the brains of the patients to sample fluid in the spaces between cells, where amyloid beta protein accumulates. Then, they took hourly measurements of the fluid to check levels of the protein.
"What we were expecting was that amyloid beta levels would be high immediately after the injury and fall over time," Brody said in a telephone interview. What they saw instead was a gradual increase in levels of this protein as patients recovered brain function. The better the patients got, the higher their amyloid beta levels rose. And in patients whose neurological function worsened, amyloid beta fell.
Brody said the finding suggests that amyloid beta in the human brain may be an indicator of how well brain cells are communicating, something studies in mice have suggested.
And while it also proves it is possible to directly measure amyloid beta in humans, "it raises a lot more questions than it answers," Brody said.
Sources and Additional Information:
