Reclassifying Traumatic Brain Injury—What Are the Next Steps?

SEATTLE—A new, more reliable, efficient, and valid classification system for traumatic brain injury (TBI) is under development, with the goal of driving more focused research and individualized therapy for patients with head injury, reported Geoffrey Manley, MD, PhD, at the 61st Annual Meeting of the American Academy of Neurology.

Problems With the Current Classification System
TBI classification is “still a symptoms-based classification system,” said Dr. Manley, Professor of Neurosurgery at the University of California, San Francisco, and Chief of Neurosurgery at San Francisco General Hospital.

In fact, Dr. Manley argued, “we probably had a better classification system for head injury 200 years ago than we have today.” In the beginning of the 18th century, autopsies became more routine and many disease conditions, including head injuries, started to be classified based on pathoanatomic features. At that time, TBI was classified as follows: commotio (ie, concussion), contusio (ie, bruising of the brain, contusions), and compresio (ie, compressive injuries such as subdural and epidural hematomas). Today, “we classify TBI as mild, moderate, and severe, and that’s based upon the Glasgow Coma Score (GCS),” said Dr. Manley. “I think that, pathoanatomically, we’re looking at a much more heterogeneous disease process than just mild, moderate, and severe.”

As an example of the insufficiency of classifying TBI based solely on GCS score, Dr. Manley described the case of a 56-year-old executive. The patient fell and was found to have a GCS of 15. “The residents thought he was doing fine, because the day after the injury he’s showing two fingers, he seems to be conversing just fine,” stated Dr. Manley. However, the patient had an orbitofrontal lesion, and at 12-month follow-up he was found to be impulsive, unemployed, and divorced. “So I don’t think this was a mild head injury for this guy,” said Dr. Manley. “We don’t consider this lesion mild just because this patient had a GCS of 14 or 15. We look at this kind of injury, and we say, ‘This is someone who’s going to have social dysregulation in the future.’”

Dr. Manley suggested that because the GCS does not provide information about the pathophysiologic mechanisms behind neurologic deficits, a more multidimensional classification system is needed to develop more targeted treatment and improve outcomes. Cancer, for example, is classified using a multidimensional system. “It’s really the pinnacle of disease classification. It’s a mixture of anatomy, physiology, metabolomics, immunology, and genetically defined diseases,” said Dr. Manley. “We didn’t get as far as we did with cancer by saying ‘You’ve got mild cancer,’ ‘You’ve got moderate cancer,’ and ‘You’re on death’s door.’”

Workshop Recommendations
In October 2007, the NINDS, with support from the Brain Injury Association of America, the Defense and Veterans Brain Injury Center, and the National Institute of Disability and Rehabilitation Research, convened a workshop to discuss the limitations of using the GCS for classification of TBI and the need to develop a more pathoanatomic-based classification system. The organizing committee was cochaired by Dr. Manley, and their recommendations were published in the July 2008 Journal of Neurotrauma.

In order to begin to reclassify TBI, “we really have to restructure the field from the ground up,” said Dr. Manley. “We need to define a common set of demographic data elements so that we’re all collecting the same kind of information across studies, whether they’re funded by the NIH, whether they’re funded by the Department of Defense, whether this is just a study that you’re going to get funded out of your individual funds at your own department. We should all be collecting the same demographic data.”

Dr. Manley emphasized the need for agreement on outcome measures. “We have to do more in terms of getting better outcome measures that define the disabilities that these patients have.” He also suggested that imaging requirements should be better defined. When doing CT and MRI scans, he said, the same kind of sequences should be used so that films can be compared from site to site and across different studies.

Lastly, Dr. Manley recommended defining requirements for genomics and proteomics. Setting standards for collecting, processing, and sorting is essential for investigating biomarkers, he said.

New Horizons in TBI Research
Dr. Manley noted that in the upcoming months, “diffusion tensor imaging … is going to be very important. We’ve been doing a lot of work with magnetoencephalography and looking at things like functional connectivity…. It’s not where the lesion is, but how it affects the connectivity in the brain.”

Higher resolution imaging will also be studied. “We’ve been doing a lot of scanning of our [TBI] patients on our new 7-Tesla magnet … and it’s just absolutely amazing, the structural images,” Dr. Manley commented.

“The other thing that I think we really need to improve upon is medical informatics,” said Dr. Manley, noting that better clinical decision-making tools are needed to assist clinicians who do not focus solely on TBI. “Most importantly, I think we need a prospective, multivariate TBI database. I’m not sure we even know what TBI is in 2009. Almost everything we do as clinicians is based on what we learned from the Traumatic Coma Data Bank, which contains patient data collected over 25 years ago,” said Dr. Manley. Prospective observational studies across the entire TBI injury spectrum will require 1,000 to 2,000 patients, rather than 20 to 30, to develop a clearer picture of TBI and help to focus future research and clinical efforts.

In March 2009, a total of 49 institutions and government agencies, including the NIH, National Institute on Disability and Rehabilitation Reasearch, Defense Centers of Excellence, Defense and Veterans Brain Injury Center, Veterans Affairs, and other stakeholders, organized a consensus conference in Washington, DC to discuss these issues. “What I hope you’ll see by the end of the year is a series of white papers that essentially outline what these common data elements are for TBI demographics, neuroimaging, biomarkers, outcome, and assessment of psychological health. The vision is to provide the tools to standardize data collection to improve TBI research and clinical care.”

SEATTLE—A new, more reliable, efficient, and valid classification system for traumatic brain injury (TBI) is under development, with the goal of driving more focused research and individualized therapy for patients with head injury, reported Geoffrey Manley, MD, PhD, at the 61st Annual Meeting of the American Academy of Neurology.

Problems With the Current Classification System
TBI classification is “still a symptoms-based classification system,” said Dr. Manley, Professor of Neurosurgery at the University of California, San Francisco, and Chief of Neurosurgery at San Francisco General Hospital.

In fact, Dr. Manley argued, “we probably had a better classification system for head injury 200 years ago than we have today.” In the beginning of the 18th century, autopsies became more routine and many disease conditions, including head injuries, started to be classified based on pathoanatomic features. At that time, TBI was classified as follows: commotio (ie, concussion), contusio (ie, bruising of the brain, contusions), and compresio (ie, compressive injuries such as subdural and epidural hematomas). Today, “we classify TBI as mild, moderate, and severe, and that’s based upon the Glasgow Coma Score (GCS),” said Dr. Manley. “I think that, pathoanatomically, we’re looking at a much more heterogeneous disease process than just mild, moderate, and severe.”

As an example of the insufficiency of classifying TBI based solely on GCS score, Dr. Manley described the case of a 56-year-old executive. The patient fell and was found to have a GCS of 15. “The residents thought he was doing fine, because the day after the injury he’s showing two fingers, he seems to be conversing just fine,” stated Dr. Manley. However, the patient had an orbitofrontal lesion, and at 12-month follow-up he was found to be impulsive, unemployed, and divorced. “So I don’t think this was a mild head injury for this guy,” said Dr. Manley. “We don’t consider this lesion mild just because this patient had a GCS of 14 or 15. We look at this kind of injury, and we say, ‘This is someone who’s going to have social dysregulation in the future.’”

Dr. Manley suggested that because the GCS does not provide information about the pathophysiologic mechanisms behind neurologic deficits, a more multidimensional classification system is needed to develop more targeted treatment and improve outcomes. Cancer, for example, is classified using a multidimensional system. “It’s really the pinnacle of disease classification. It’s a mixture of anatomy, physiology, metabolomics, immunology, and genetically defined diseases,” said Dr. Manley. “We didn’t get as far as we did with cancer by saying ‘You’ve got mild cancer,’ ‘You’ve got moderate cancer,’ and ‘You’re on death’s door.’”

Workshop Recommendations
In October 2007, the NINDS, with support from the Brain Injury Association of America, the Defense and Veterans Brain Injury Center, and the National Institute of Disability and Rehabilitation Research, convened a workshop to discuss the limitations of using the GCS for classification of TBI and the need to develop a more pathoanatomic-based classification system. The organizing committee was cochaired by Dr. Manley, and their recommendations were published in the July 2008 Journal of Neurotrauma.

In order to begin to reclassify TBI, “we really have to restructure the field from the ground up,” said Dr. Manley. “We need to define a common set of demographic data elements so that we’re all collecting the same kind of information across studies, whether they’re funded by the NIH, whether they’re funded by the Department of Defense, whether this is just a study that you’re going to get funded out of your individual funds at your own department. We should all be collecting the same demographic data.”

Dr. Manley emphasized the need for agreement on outcome measures. “We have to do more in terms of getting better outcome measures that define the disabilities that these patients have.” He also suggested that imaging requirements should be better defined. When doing CT and MRI scans, he said, the same kind of sequences should be used so that films can be compared from site to site and across different studies.

Lastly, Dr. Manley recommended defining requirements for genomics and proteomics. Setting standards for collecting, processing, and sorting is essential for investigating biomarkers, he said.

New Horizons in TBI Research
Dr. Manley noted that in the upcoming months, “diffusion tensor imaging … is going to be very important. We’ve been doing a lot of work with magnetoencephalography and looking at things like functional connectivity…. It’s not where the lesion is, but how it affects the connectivity in the brain.”

Higher resolution imaging will also be studied. “We’ve been doing a lot of scanning of our [TBI] patients on our new 7-Tesla magnet … and it’s just absolutely amazing, the structural images,” Dr. Manley commented.

“The other thing that I think we really need to improve upon is medical informatics,” said Dr. Manley, noting that better clinical decision-making tools are needed to assist clinicians who do not focus solely on TBI. “Most importantly, I think we need a prospective, multivariate TBI database. I’m not sure we even know what TBI is in 2009. Almost everything we do as clinicians is based on what we learned from the Traumatic Coma Data Bank, which contains patient data collected over 25 years ago,” said Dr. Manley. Prospective observational studies across the entire TBI injury spectrum will require 1,000 to 2,000 patients, rather than 20 to 30, to develop a clearer picture of TBI and help to focus future research and clinical efforts.

In March 2009, a total of 49 institutions and government agencies, including the NIH, National Institute on Disability and Rehabilitation Reasearch, Defense Centers of Excellence, Defense and Veterans Brain Injury Center, Veterans Affairs, and other stakeholders, organized a consensus conference in Washington, DC to discuss these issues. “What I hope you’ll see by the end of the year is a series of white papers that essentially outline what these common data elements are for TBI demographics, neuroimaging, biomarkers, outcome, and assessment of psychological health. The vision is to provide the tools to standardize data collection to improve TBI research and clinical care.”

SEATTLE—A new, more reliable, efficient, and valid classification system for traumatic brain injury (TBI) is under development, with the goal of driving more focused research and individualized therapy for patients with head injury, reported Geoffrey Manley, MD, PhD, at the 61st Annual Meeting of the American Academy of Neurology.

Problems With the Current Classification System
TBI classification is “still a symptoms-based classification system,” said Dr. Manley, Professor of Neurosurgery at the University of California, San Francisco, and Chief of Neurosurgery at San Francisco General Hospital.

In fact, Dr. Manley argued, “we probably had a better classification system for head injury 200 years ago than we have today.” In the beginning of the 18th century, autopsies became more routine and many disease conditions, including head injuries, started to be classified based on pathoanatomic features. At that time, TBI was classified as follows: commotio (ie, concussion), contusio (ie, bruising of the brain, contusions), and compresio (ie, compressive injuries such as subdural and epidural hematomas). Today, “we classify TBI as mild, moderate, and severe, and that’s based upon the Glasgow Coma Score (GCS),” said Dr. Manley. “I think that, pathoanatomically, we’re looking at a much more heterogeneous disease process than just mild, moderate, and severe.”

As an example of the insufficiency of classifying TBI based solely on GCS score, Dr. Manley described the case of a 56-year-old executive. The patient fell and was found to have a GCS of 15. “The residents thought he was doing fine, because the day after the injury he’s showing two fingers, he seems to be conversing just fine,” stated Dr. Manley. However, the patient had an orbitofrontal lesion, and at 12-month follow-up he was found to be impulsive, unemployed, and divorced. “So I don’t think this was a mild head injury for this guy,” said Dr. Manley. “We don’t consider this lesion mild just because this patient had a GCS of 14 or 15. We look at this kind of injury, and we say, ‘This is someone who’s going to have social dysregulation in the future.’”

Dr. Manley suggested that because the GCS does not provide information about the pathophysiologic mechanisms behind neurologic deficits, a more multidimensional classification system is needed to develop more targeted treatment and improve outcomes. Cancer, for example, is classified using a multidimensional system. “It’s really the pinnacle of disease classification. It’s a mixture of anatomy, physiology, metabolomics, immunology, and genetically defined diseases,” said Dr. Manley. “We didn’t get as far as we did with cancer by saying ‘You’ve got mild cancer,’ ‘You’ve got moderate cancer,’ and ‘You’re on death’s door.’”

Workshop Recommendations
In October 2007, the NINDS, with support from the Brain Injury Association of America, the Defense and Veterans Brain Injury Center, and the National Institute of Disability and Rehabilitation Research, convened a workshop to discuss the limitations of using the GCS for classification of TBI and the need to develop a more pathoanatomic-based classification system. The organizing committee was cochaired by Dr. Manley, and their recommendations were published in the July 2008 Journal of Neurotrauma.

In order to begin to reclassify TBI, “we really have to restructure the field from the ground up,” said Dr. Manley. “We need to define a common set of demographic data elements so that we’re all collecting the same kind of information across studies, whether they’re funded by the NIH, whether they’re funded by the Department of Defense, whether this is just a study that you’re going to get funded out of your individual funds at your own department. We should all be collecting the same demographic data.”

Dr. Manley emphasized the need for agreement on outcome measures. “We have to do more in terms of getting better outcome measures that define the disabilities that these patients have.” He also suggested that imaging requirements should be better defined. When doing CT and MRI scans, he said, the same kind of sequences should be used so that films can be compared from site to site and across different studies.

Lastly, Dr. Manley recommended defining requirements for genomics and proteomics. Setting standards for collecting, processing, and sorting is essential for investigating biomarkers, he said.

New Horizons in TBI Research
Dr. Manley noted that in the upcoming months, “diffusion tensor imaging … is going to be very important. We’ve been doing a lot of work with magnetoencephalography and looking at things like functional connectivity…. It’s not where the lesion is, but how it affects the connectivity in the brain.”

Higher resolution imaging will also be studied. “We’ve been doing a lot of scanning of our [TBI] patients on our new 7-Tesla magnet … and it’s just absolutely amazing, the structural images,” Dr. Manley commented.

“The other thing that I think we really need to improve upon is medical informatics,” said Dr. Manley, noting that better clinical decision-making tools are needed to assist clinicians who do not focus solely on TBI. “Most importantly, I think we need a prospective, multivariate TBI database. I’m not sure we even know what TBI is in 2009. Almost everything we do as clinicians is based on what we learned from the Traumatic Coma Data Bank, which contains patient data collected over 25 years ago,” said Dr. Manley. Prospective observational studies across the entire TBI injury spectrum will require 1,000 to 2,000 patients, rather than 20 to 30, to develop a clearer picture of TBI and help to focus future research and clinical efforts.

In March 2009, a total of 49 institutions and government agencies, including the NIH, National Institute on Disability and Rehabilitation Reasearch, Defense Centers of Excellence, Defense and Veterans Brain Injury Center, Veterans Affairs, and other stakeholders, organized a consensus conference in Washington, DC to discuss these issues. “What I hope you’ll see by the end of the year is a series of white papers that essentially outline what these common data elements are for TBI demographics, neuroimaging, biomarkers, outcome, and assessment of psychological health. The vision is to provide the tools to standardize data collection to improve TBI research and clinical care.”