IMAGINE that you're a 55-year-old male with high blood pressure who develops left-sided chest pain radiating into your left arm while running to catch a flight to Myrtle Beach. Airport paramedics rush you to the nearest emergency department, where you quickly undergo a chest x-ray. That's it, just a chest x-ray. Now imagine that the emergency physician looks at your chest x-ray, then looks at you and says, "Well, it could be a heart attack, or it could be nothing. Go home and rest."

Crazy, right? Where's the EKG? Where's the serum troponin? Where's the transthoracic ultrasound, stress echo, and cardiac catheterization? Where's the cardiologist and the 23-hour hospital stay? Who can imagine being a heart attack patient and not undergoing any of these diagnostic aids? Well, this imagined scenario is exactly how modern medicine currently cares for a mild traumatic brain injury (TBI), otherwise known as a concussion. The simplest way to sum up the state of the art of mild TBI care: "Well, it could be a brain injury, or it could be nothing. Go home and rest."

While this approach may have been acceptable in years past, it no longer is. There is growing recognition that mild TBI is a public health problem, a not-so-silent epidemic. Thanks to the surveillance efforts of the Centers of Disease Control and Prevention, we now know that mild TBI affects nearly 1.2 million Americans annually, mostly from motor vehicle collisions and falls.1 This makes mild TBI more common than stroke, multiple sclerosis, and Huntington's disease combined, giving it the dubious distinction as the most frequently occurring brain disorder.2 Mild TBI has also been dubbed the signature injury of the conflicts in Iraq and Afghanistan, potentially affecting 320 000 US soldiers to date.3

But it is not just the sheer magnitude of mild TBI that has led the medical community away from a nihilistic approach to this injury. It was the overwhelming evidence that linked mild TBI to a host of clinically important problems. The November/December 2009 issue of Journal of Head Trauma Rehabilitation (JHTR) chronicled the myriad problems that are directly attributable to TBI, including cognitive difficulties, seizures, postconcussive symptoms, pituitary insufficiency, dementia of the Alzheimer's type, parkinsonism, mood disorders, reduced social functioning, and unemployment.4-7 While the research evidence to date shows that many of these conditions occur with moderate and severe TBI, several complications included mild TBI as well (eg, unprovoked seizures, aggression, depression, completed suicide, dementia of the Alzheimer's type, parkinsonism, ocular and visual motor deterioration).

It is ironic, then, that this quite common neurologic disorder with multiple downstream consequences lacks an objective diagnostic aid. The current standard for diagnosing mild TBI relies on a witnessed or self-reported loss of consciousness for less than 20 minutes, amnesia for less than 24 hours, or any period of confusion or being dazed at the time of the injury.1,8 While efforts to standardize a clinical definition of mild TBI are to be applauded, there are multiple situations where this diagnostic approach is problematic. For example, preverbal children are not able to articulate their experiences around a blow to the head, excluding them from a mild TBI diagnosis. Memory of injury events is confounded by drug and alcohol intoxication as well as underlying dementia. Patients who seize or pass out and then hit their head may have no memory of events due to the underlying medical condition, not brain injury. But beyond these difficult-to-diagnose patients, asking someone with impaired memory to tell you exactly what did-or did not happen-at the time of their injury just seems counterintuitive. The very nature of this injury, thus, prevents an accurate assessment of its diagnosis as currently framed.

To be fair, mild TBI is not entirely without diagnostic aids. Computed tomography (CT) scanning is commonly employed as a quasi-diagnostic tool after a concussion, but it is used to detect hemorrhage, not injury to neurons and glia. Hemorrhage occurs in about 3% to 10% of mild TBI patients (most are contusions), with less than 1% requiring neurosurgery, which can be lifesaving.9 In the remaining 90%, the CT is normal and provides no information on the presence or absence of brain injury. A small percentage of such patients will have postconcussive symptoms lasting more than a year, suggesting the presence of underlying brain injury that is just not visible on CT scan.

"Go home and rest" is the typical advice given to mild TBI patients with normal head CT scans. But this is no longer acceptable. The disability associated with mild TBI has sparked an intense effort to develop treatments for this injury. It has become increasingly clear that lack of an accurate and objective method to diagnosis mild TBI is an obstacle to the development of potential therapies.

In this topical issue of JHTR-Biomarkers of Mild Traumatic Brain Injury-we highlight 6 cutting-edge studies attempting to objectify the diagnosis of brain injury after a concussion. The National Institute of Health defines biomarker as "a characteristic that is objectively measured and evaluated as an indicator of normal biologic processes, pathogenic processes, or pharmacologic responses to a therapeutic intervention."10 In the case of a pathogenic process such as mild TBI, an appropriate biomarker would detect a particular aspect of the process that begins with stretch injury to axons and ends with disrupted brain function. To set the stage for our 6 biomarker studies, it may be helpful to briefly review the nature of brain injury after concussion.

Axonal injury (AI) is currently thought to be the structural substrate behind most post-mild TBI neurologic dysfunction. Studies in animals and humans have revealed AI after all severities of TBI, including mild TBI.11 During AI, stretch-induced calcium influx leads to axonal microtubule and neurofilament destruction and disruption of axonal transport, leading to axonal swelling, disconnection, loss of synaptic connections, and possibly cell death12,13 Biomarkers of AI currently attempt to detect 3 broad aspects of axonal injury: structural changes to cellular elements, leakage of brain-related proteins into the peripheral circulation, and subtle changes in neurologic function. This topical issue of JHTR provides excellent examples of cutting-edge research from all 3 areas.

STRUCTURAL CHANGES

AI after mild TBI involves structural and physiologic changes to cellular and subcelluar elements that are visible in pathologic examination of brain upon autopsy.14 These changes are potentially detectable in vivo with newer forms of neuroimaging. Kou et al review the unique sensitivity of 3 forms of advanced magnetic resonance imaging (MRI)-susceptibility-weighted imaging, diffusion tensor imaging (DTI), and MR spectroscopic imaging-for the detection of axonally injured white matter, shear injury to small blood vessels (microhemorrhages), and abnormal metabolites resulting from mitochondrial failure. Although DTI and MR spectroscopic imaging are still experimental, susceptibility-weighted imaging is currently being performed in many centers and can reveal white matter lesions where CT and conventional T1/T2 MRI are normal (Kou et al, this issue).

By measuring the magnitude and straightness (isotropy) of water movement in and around neurons, DTI may be uniquely suited to detect AI. After mild TBI, axons swell, reducing water movement between axons and making it move straighter. Niogi and Mukherjee review the ever-growing list of studies revealing changes in white matter water diffusion after mild TBI that correlate with abnormal cognitive outcome. DTI is a quantitative, not qualitative, imaging modality. Thousands of water movement measurements must be analyzed and interpreted. Niogi and Mukherjee correctly point out that DTI analysis is still a work in progress, and it is likely that the ideal method has not yet been invented (Niogi and Mukherjee, this issue).

Kelly shows us that functional MRI can reveal abnormal patterns of cortical activation after mild TBI, even when cognitive function is back to normal. A downstream consequence of AI is mitochondrial failure and a change in the ability of the neuron to extract and use oxygen. Functional MRI measures and spatially maps onto the cortex the ratio of oxygenated to deoxygenated hemoglobin. Mild TBI subjects asked to perform a simple memory task are able to do these tasks as well as controls but activate more scattered and laterally located cortical areas. This suggests that the duration of physiological recovery after mild TBI may extend longer than observed clinical recovery (Kelly, this issue).

CENTRAL NERVOUS SYSTEM PROTEIN LEAKAGE INTO SERUM

AI results in the dissolution of the axonal cytoarchitecture, mitochondrial failure, and apoptotic cell death. These processes release a variety of proteins into the interstitial fluid that may be detectable in serum if the blood-brain barrier is damaged and thus open.15 S100B, the astrocyte-related protein, is perhaps the best-studied brain marker to date. It is present in the serum within 30 minutes of mild TBI but is renally cleared by 6 hours. Unden and Romner review the strong evidence that S100 is a sensitive predictor of hemorrhage on head CT after mild TBI. A normal level measured within 3 hours of injury has a negative predictive rate of 99%. Use of this test could reduce unnecessary CT scanning for mild TBI by 30%. Although this test is in clinical use in several European counties, it is not yet approved by the US Food and Drug Administration in the United States (Unden and Romner, this issue).

NEUROLOGIC DYSFUNCTION

Subtle but distinct changes in brain function that directly result from a loss of synaptic connections may now be detectable thanks to recent advances in biomedical engineering. The ability to track a visual object through space is a measure of attentiveness, which is disrupted after mild TBI. Maruta et al report on the use of an eyeglass-type device that measures visual tracking after mild TBI. Gaze error variability was found to be markedly increased compared with that in controls and correlated with working memory and changes in water diffusion on DTI (Maruta et al, this issue). Using a stripped down version of electroencephalogram involving only 3 leads, McCrea et al found significant abnormalities in electrical brain activity after mild TBI that correlated with postconcussive symptoms. While the symptoms resolved by day 4, EEG abnormalities were still present at day 8 postinjury, but not on day 45. These results echo those found by Kelly in that physiological recovery after mild TBI appeared to extend beyond the point of apparent clinical improvement (McCrea et al, this issue).

As these studies so elegantly point out, a new era in TBI care is upon us. The development of an objective biomarker of mild TBI is a necessary first step toward a better understanding of the natural history and pathophysiology of this injury. It is hoped that the continued refinement of biomarkers such as the ones presented here will open the door to randomized controlled trials of treatments designed to interrupt the process of axonal injury or enhance the process of repair and recovery. The overall goal is to reduce the substantial disability from this injury and to improve the lives of the millions of people affected.

-Jeffrey J. Bazarian, MD, MPH, Issue Editor

Department of Emergency Medicine, University of Rochester Medical Center, Rochester, New York

REFERENCES