The links between traumatic brain injury and Alzheimer disease have been of great interest for many years. However, the importance of amyloid-β–related neurodegenerative pathophysiologic processes after traumatic brain injury is still unknown. In this review, we present a brief overview of the scientific evidence regarding traumatic brain injury as a contributor to Alzheimer disease and describe recent results showing significant changes in brain extracellular amyloid-β dynamics in patients with severe brain injury. We then discuss the clinical significance of these findings with their implications for translational neurobiology and conclude with further directions for traumatic brain injury and Alzheimer disease research.
Moderate to severe traumatic brain injury (TBI) is a well-documented environmental risk factor for the later development of dementia of the Alzheimer type.1- 3 This increased risk could in principle be owing to 1 or more of the following: (1) impaired cognitive reserve, (2) acceleration of the underlying neurodegenerative processes that normally cause this type of dementia later in life, and (3) as yet undefined factors.
Direct evidence favoring impaired cognitive reserve is difficult to obtain because cognitive reserve is not well defined from a pathophysiologic perspective. However, the findings that cerebral infarction, poor social network support, and low educational attainment also increase the risk of dementia of the Alzheimer type4 provide some support for this concept.
Evidence favoring a TBI-related acceleration of Alzheimer disease (AD)–related pathophysiologic processes comes from human studies and experimental animal data. First, amyloid β (Aβ) plaques and intra-axonal Aβ deposits have been found in approximately one-third of patients with fatal TBI who did not have preexisting AD, Down syndrome, or clinical dementia.5- 13 A biopsy study of TBI survivors requiring decompressive temporal lobectomy has confirmed this finding,14,15 suggesting it is also relevant to patients with nonfatal TBI.
Second, a genetic risk factor for AD, the ε4 allele of the apolipoprotein E gene (ApoE4 ; GenBank AF261279), also increases the risk of adverse clinical outcomes after TBI.16 Although ApoE has many functions, the primary role of ApoE4 in the development of AD may be to promote the development of Aβ plaque pathology.17 In a large autopsy series,18 45% of patients with TBI who had 1 or more ApoE4 allele had Aβ plaque pathology, whereas only 10% of patients with TBI without an ApoE4 allele had such pathology. This finding needs to be replicated, however.
Third, experimental studies in a pig model of traumatic axonal injury reliably reproduced this Aβ plaque pathology, further strengthening the evidence for a causal role of TBI.19,20 Traumatic axonal injury may also play a key role in patients with TBI because colocalization of Aβ with several of the enzymes involved in cleaving the amyloid precursor protein has been detected at sites of axonal injury in human TBI autopsy samples.13
Taken together, these results raise the possibility that TBI may increase the levels of Aβ in the brain, accelerating the Aβ-related pathophysiologic processes believed to be a root cause of dementia of the Alzheimer type. The specific model put forward by Chen et al20 involves coaccumulation of amyloid precursor protein with its proteolytic enzymes at sites of axonal injury, increased intracellular production of Aβ, release of Aβ from injured axons into the extracellular space, and deposition of Aβ into extracellular Aβ plaques. Ultimately, this process could play a role in the observed link between a history of brain trauma and an increased risk of developing AD.
Since 2003, it has been possible to measure Aβ in the cerebral interstitial fluid (ISF) of mice using intracerebral microdialysis.21 The extracellular space is believed to be the crucial site for Aβ aggregation and toxicity.22 These early studies demonstrated the presence of Aβ in the ISF of amyloid precursor protein transgenic mouse brains before the onset of AD-like Aβ plaque pathology. Furthermore, they showed a reduction in Aβ clearance rate after the onset of Aβ deposition, likely reflecting plaque-associated changes in amyloid metabolism. Subsequent experiments indicated that neuronal activity and more specifically synaptically coupled endocytic activity was directly correlated with extracellular Aβ concentrations as measured by microdialysis.23,24 These data were in agreement with previous findings showing that neuronal activity modulates the formation and secretion of Aβ peptides in hippocampal slices overexpressing amyloid precursor protein.25 Because neuronal activity is likely to be reduced in the setting of TBI, these findings raised an intriguing alternative possibility: if the same relationship between extracellular Aβ and synaptically coupled endocytosis exists in humans, TBI could in fact decrease extracellular Aβ owing to reductions in neuronal and synaptic activity.
Intracerebral microdialysis can also be used in patients monitored in the intensive care unit.26 Clinically approved, commercially available sterile microdialysis catheters can be placed at the same time that another clinically indicated intracranial neurosurgical procedure is being performed with little additional risk to the patient. Among other applications, brain microdialysis has been used clinically to detect early signs of metabolic deterioration, which may provide an early warning of impending secondary insults after acute brain injury.27
We adapted these clinically approved microdialysis methods to allow recovery of Aβ by adding sterile human albumin to the perfusion fluid to block nonspecific binding of Aβ to the catheters and tubing. We then studied 18 patients with severe brain injuries who were affected by TBI or aneurysmal subarachnoid hemorrhage using these microdialysis methods.28 Our principal hypothesis was that there would be an acute increase in extracellular Aβ after TBI in accordance with the model of Chen et al,20 but the results we found were more consistent with an alternative model.
Specifically, there were increasing trends in brain ISF Aβ concentrations during several hours to days in most patients, although the specific pattern of these trends was variable. Interestingly, we found that brain ISF Aβ levels tracked the patients' overall neurological status, as assessed using the Glasgow Coma Scale. In particular, brain ISF Aβ levels increased as patients' neurological status improved, remained stable in clinically stable patients, and appeared to decline when neurological status worsened (Figure 1). We also measured the Aβ1-40 and Aβ1-42 species from pooled samples of a subset of these patients; their concentrations appeared to correlate with those of total Aβ.
Brain interstitial fluid (ISF) amyloid-β (Aβ) concentrations and neurological status. A-D, Examples of the course of changes in brain ISF Aβ concentrations and changes in neurological status, as reflected by the Glasgow Coma Score (GCS). Changes in Aβ appear to track (A and B), and in some cases even precede (C and D), neurological status changes. E, Correlation of change in brain ISF Aβ from baseline with changes in neurological status across 13 patients in which serial GCS measurements could be reliably obtained (n = 173 paired measurements) (Spearman r = 0.50, P < .001 overall; Spearman r = 0.82, P < .001 for change in GCS score ≥2).
Independently, Marklund et al29 measured ISF Aβ1-40 and Aβ1-42 in 8 severely brain-injured patients using a similar microdialysis technique. They reported nonsignificantly higher interstitial Aβ1-42 levels in 3 patients with diffuse axonal injury compared with 5 patients with focal cerebral injuries. Interestingly and in concordance with our findings, a patient with rapid clinical improvement and good recovery had relatively high Aβ levels (case 2 in their study), whereas 1 with persistent coma and poor outcome had undetectable levels of Aβ (case 7 in their study).
All these findings, in conjunction with the experimental microdialysis studies discussed herein, support the alternative hypothesis that extracellular Aβ may be decreased after brain injury owing to or in conjunction with injury-related suppression of neuronal activity. In this light, previous results involving ventricular cerebrospinal fluid (CSF) measurements of Aβ could be reinterpreted; observations of an increase in ventricular CSF Aβ levels over time in TBI and subarachnoid hemorrhage patients could be related to recovery of neurological status rather than secondary injury cascades.30- 32 This may also help explain the findings that ventricular CSF from severely injured aneurysmal subarachnoid hemorrhage patients had lower Aβ levels than ventricular CSF from otherwise healthy, neurologically intact patients with chronic hydrocephalus with suspected shunt dysfunction.30
Suppression of neuronal activity is a likely feature of many types of acute brain injury. This finding may be a consequence of direct disruption of neuronal membranes, energy failure, sodium channel inactivation related to massive depolarization driven by glutamate release,33 or other processes. Interestingly, in a study combining intracerebral electroencephalographic recordings and microdialysis, TBI-induced suppression of neuronal firing was associated with characteristic posttraumatic alterations of microdialysis-measured metabolic markers, such as reduced extracellular glucose levels and increased lactate-pyruvate ratio.34 In our study, we found a strong association between similar metabolic alterations and reduced levels of extracellular Aβ; brain ISF Aβ levels were positively correlated with cerebral glucose levels and negatively correlated with cerebral lactate-pyruvate ratios (Figure 2 in Brody et al28). This finding indirectly suggests that reduced Aβ is associated with suppression of neuronal firing.
In areas of the brain adjacent to sites of macroscopic injury, reduced neuronal activity owing to repeated waves of depression of electrocorticographic activity (cortical spreading depressions) have been documented in a large proportion of patients.35 Electrocorticography was not performed in our study, but it is likely that similar cortical spreading depressions occurred in our patients near sites of macroscopic injury as well. We therefore predicted that patients in whom microdialysis catheters were placed close to areas of focal brain damage (eg, contusions, infarctions apparent on computed tomographic scans) would have lower Aβ levels than patients in whom microdialysis catheters were placed in normal-appearing tissue (based on computed tomographic scans). Reanalysis of our published data with the inclusion of data from several additional patients confirmed this prediction (Figure 2). Taken together, these findings are consistent with the hypothesis that brain extracellular Aβ levels in humans are reduced after acute brain injury as a consequence of reduction of brain neuronal and synaptic activity (Figure 3A).
Amyloid-β (Aβ) levels adjacent to sites of macroscopic injury vs in normal-appearing tissue after traumatic brain injury. Initial interstitial fluid (ISF) Aβ levels were significantly lower when catheters were placed adjacent to sites of macroscopic injury (n = 4) compared with when catheters were placed in normal-appearing tissue (n = 8), as assessed by computed tomographic scans. White arrows indicate microdialysis catheter tips. P = .02, Mann-Whitney test.
Schematic view of brain interstitial fluid (ISF) amyloid-β (Aβ) dynamics in the setting of acute brain injury. A, Observed changes in ISF Aβ in patients with acute severe brain injury. Preinjury Aβ levels were unknown in patients, but after injury Aβ levels tracked the patients' global neurological status, as assessed using the Glasgow Coma Scale. In particular, brain ISF Aβ levels increased as patients' neurological status improved, remained stable in clinically stable patients, and appeared to decline when neurological status worsened. B, Hypothesized model of ISF Aβ dynamics after acute brain injury. Soluble Aβ levels are likely reduced after injury owing to reduction of synaptic activity. However, such reduced levels of soluble extracellular Aβ could also reflect insoluble Aβ aggregation in the extracellular space and/or intracellular Aβ accumulation.
However, we cannot completely rule out the possibility that the reduced levels of ISF Aβ observed in the patients with TBI were a reflection of extracellular Aβ deposition into insoluble aggregates or intracellular Aβ retention (Figure 3B). Similar reductions in brain extracellular soluble Aβ owing to Aβ being retained in insoluble forms have been hypothesized to explain the reduced levels of Aβ1-42 seen in the lumbar CSF of patients with AD.36 Clearly, the relationship between intracellular and extracellular Aβ is complex. For example, Billings et al37 found that water maze training (which may increase synaptic activity) was associated with increased total soluble Aβ but decreased insoluble Aβ and oligomeric Aβ in transgenic mice with Aβ plaque pathology.
Furthermore, it is formally possible that Aβ levels are initially higher than normal after TBI and then increase further in concert with clinical recovery. For logistical reasons, most microdialysis catheters were placed 12 to 24 hours after injury, leaving the possibility that an early “spike” in extracellular Aβ levels could have occurred within the first few hours of injury. Obviously, preinjury Aβ levels cannot be assessed directly in humans.
We therefore are in the process of going back to mouse models to directly address these mechanistic questions about the handling of Aβ after TBI. Combined experimental TBI, Aβ microdialysis, and electrophysiologic studies in transgenic mice are feasible. To date, these have revealed that extracellular Aβ levels and electroencephalographic activity are immediately reduced in the hippocampus after controlled cortical-impact TBI.38 These experimental results are concordant with the alternative hypothesis discussed herein.
Fundamentally, there are many still-unresolved questions regarding the interaction between TBI and neurodegenerative processes related to Alzheimer disease. First, the aggregation state of the soluble extracellular Aβ was not investigated in any of these experimental or clinical microdialysis studies. It is possible that minor but potentially important toxic Aβ subspecies, such as oligomers and protofibrils, could be elevated after TBI, even if total ISF Aβ levels are reduced. Such toxic subspecies could represent a pathophysiologic link between TBI and dementia. Microdialysis-based methods for assessing the aggregation state of Aβ in the extracellular space of the human brain will be of great interest.
Second, the relationship between traumatic axonal injury and extracellular brain Aβ dynamics requires clarification. Advanced magnetic resonance imaging methods, such as diffusion tensor imaging, can provide much more detailed information regarding axonal integrity after TBI than standard methods, such as computed tomography or conventional magnetic resonance imaging. Combined microdialysis and diffusion tensor imaging studies could be used to address this issue.
Third, the dynamics of tau-related pathophysiologic processes after TBI have not been thoroughly assessed. Tau pathology in the form of neurofibrillary tangles is another hallmark of AD and has been described in a subset of TBI patients.12,15
Finally, the relationship between genetic factors and Aβ handling after TBI should be readdressed using microdialysis-based approaches. As described herein, ApoE genotype may have a substantial effect on Aβ deposition, but ApoE genotype effects on soluble extracellular Aβ dynamics have not been determined. Likewise, polymorphisms in the promoter region of neprilysin, 1 of the key Aβ-degrading enzymes, appear to affect Aβ deposition after TBI39 but, again, their role in extracellular Aβ has not been addressed, to our knowledge.
The answers to these questions will help address whether TBI in fact accelerates the neurodegenerative processes underlying AD. If so, interventions designed to block or reverse these processes could be of great benefit. For example, if toxic Aβ subspecies are produced after TBI, microdialysis-based pharmacodynamic and pharmacokinetic studies could help assess candidate therapeutics targeting the productions or effects of these subspecies. A recent study40 describing beneficial effects of inhibiting the secretase enzymes required for Aβ production in experimental TBI underscores the potential for the development of such therapeutics. However, the preclinical findings of Loane et al40 need to be interpreted cautiously because the targeted secretase enzymes have other roles and the effects of their inhibition may not be necessarily or exclusively related to Aβ.
In the broader context, such pharmacodynamic and pharmacokinetic studies in patients with TBI could help drive forward the development of therapeutics for Alzheimer disease. Although our study and that of Marklund et al29 demonstrate that it is possible to perform Aβ microdialysis measurements in patients with brain injuries, it has not yet been ethically or logistically possible to directly assess Aβ dynamics in the brain of patients with AD. Intracranial procedures are commonly indicated in patients with TBI but not commonly performed in those with AD.
On the other hand, if TBI does not accelerate specific AD-related pathophysiologic processes, it may be more fruitful to focus on enhancing cognitive function and cognitive reserve using general restorative and rehabilitative approaches. Furthermore, there may be other, non–Aβ-related secondary injury cascades that should be targeted to improve long-term cognitive outcomes after TBI. Microdialysis-based assessments in patients with TBI could similarly play a key role in therapeutic development.
As a final note, if further verification of the link between synaptic activity and extracellular Aβ in the human brain can be obtained, this link could potentially be useful in improving the clinical monitoring of patients with severe brain injuries. Specifically, Aβ could serve as an independent and objective real-time measure of neuronal and synaptic activity in the local region around the catheter. Questions about the prognostic value of Aβ-based microdialysis measurements and assay standardization issues clearly must be addressed before any such clinical monitoring could be routinely useful.
Correspondence: David L. Brody, MD, PhD, 660 S Euclid Ave, Campus Box 8111, Washington University in St Louis, St Louis, MO 63110 (firstname.lastname@example.org).
Accepted for Publication: November 19, 2009.
Author Contributions:Study concept and design: Magnoni and Brody. Acquisition of data: Magnoni and Brody. Analysis and interpretation of data: Magnoni and Brody. Drafting of the manuscript: Magnoni and Brody. Critical revision of the manuscript for important intellectual content: Brody. Statistical analysis: Brody. Obtained funding: Brody. Administrative, technical, and material support: Brody.
Financial Disclosure: None reported.
Funding/Support: This work was supported by the National Institutes of Health and the Burroughs Wellcome Fund.
Thank you for submitting a comment on this article. It will be reviewed by JAMA Neurology editors. You will be notified when your comment has been published. Comments should not exceed 500 words of text and 10 references.
Do not submit personal medical questions or information that could identify a specific patient, questions about a particular case, or general inquiries to an author. Only content that has not been published, posted, or submitted elsewhere should be submitted. By submitting this Comment, you and any coauthors transfer copyright to the journal if your Comment is posted.
* = Required Field
Disclosure of Any Conflicts of Interest*
Indicate all relevant conflicts of interest of each author below, including all relevant financial interests, activities, and relationships within the past 3 years including, but not limited to, employment, affiliation, grants or funding, consultancies, honoraria or payment, speakers’ bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued. If all authors have none, check "No potential conflicts or relevant financial interests" in the box below. Please also indicate any funding received in support of this work. The information will be posted with your response.
Some tools below are only available to our subscribers or users with an online account.
Download citation file:
Web of Science® Times Cited: 26
Customize your page view by dragging & repositioning the boxes below.
Enter your username and email address. We'll send you a link to reset your password.
Enter your username and email address. We'll send instructions on how to reset your password to the email address we have on record.
Athens and Shibboleth are access management services that provide single sign-on to protected resources. They replace the multiple user names and passwords necessary to access subscription-based content with a single user name and password that can be entered once per session. It operates independently of a user's location or IP address. If your institution uses Athens or Shibboleth authentication, please contact your site administrator to receive your user name and password.