September 23-24, 2004
Holiday Inn Select
Sponsored by the National Institute of Neurological Disorders and Stroke
Cognitive deficits in varying degrees of severity are characteristic of many chronic neurological disorders. The National Institute of Neurological Disorders and Stroke has identified a pressing need for a paradigm shift from diagnosis and descriptive analysis of neuropsychological assessment of cognitive impairments in neurological disorders. In order to improve the quality of life of neurologically-impaired patients, it is necessary to establish a systematic linkage of diagnosis with intervention procedures that lead to adaptive changes in neurological damage and dysfunction. As a first step toward that goal, the National Institute of Neurological Disorders and Stroke sponsored a workshop to promote the use of evidence-based interventions in the evaluation, treatment and assistance of patients with disorders of the brain affecting higher thought processes such as working memory, attention, and executive function.
This workshop focused on a limited set of neurological conditions where progress in the rehabilitation of higher thought processes would benefit from formal partnerships between basic cognitive neuroscientists and clinicians in assessing residual capacity within specified lesioned circuits and potential for functional return. Stroke, Traumatic Brain Injury, and Brain Tumor were identified as conditions where such collaborations would assist in deciding what inputs or interventions need to be maximized to promote restoration.
The planning group for this meeting was comprised of Dr. Emmeline Edwards (NINDS), Dr. Robert Finkelstein (NINDS), and Dr. Mary Ellen Michel (NIDA) in collaboration with a consortium of neuroscientists and rehabilitation specialists. Eight months prior to the September 2004 meeting, the planning group assembled three multidisciplinary teams of scientists comprised of neurologists, cognitive neuroscientists, psychologists, occupational therapists, pharmacologists, and functional imaging experts to establish a roadmap for accelerated progress in cognitive rehabilitation interventions for stroke, traumatic brain injury and brain tumor patients. These three research teams were charged with the following mission:
Below are the members of the cognitive rehabilitation project research teams and their primary focus for intervention approaches:
Christina A. Meyers, Ph.D., Univ. of Texas- M.D., Anderson Cancer Center
Robert Butler, Ph.D., University of Washington
Raymond K., Mulhern, Ph.D., St Jude Children's Research Hospital
Robert Knight, M.D., University of California Berkeley
Edward Shaw, M.D., Wake Forest Univ. School of Medicine
William Milberg, Ph.D., Harvard Medical School
Anna Barrett, M.D., Pennsylvania State University
Argye Hillis, M.D., John Hopkins University
Branch Coslett, M.D., University of Pennsylvania
Kenneth Heilman, M.D., University of Florida
Ian Robertson, Ph.D., Trinity College
Laurel Buxbaum, Ph.D., Moss Rehabilitation Research Institute
Donald Stuss, Ph.D. (C. Psych), Rotman Research Institute
Keith Cicerone, Ph.D., JFK-Johnson Rehabilitation Institute
Harvey Levin, Ph.D., Baylor Medical College
Jim Malec, M.D., Mayo Clinic
John Whyte, MD, Moss Rehabilitation Institute
An Advisory Panel worked with the NIH Program Directors to help guide the progress of the rehabilitation project research teams and evaluate the proposed approaches chosen by these teams. As members of this Advisory Panel, Dr. Sally Shaywitz, Yale School of Medicine; Dr. Paul Eslinger, Pennsylvania State University College of Medicine; Dr. Elkhonon Goldberg, New York University School of Medicine; Dr. Jordan Grafman, Division of Intramural Research National Institute of Neurological Disorders and Stroke; Dr. Patricia Reuter-Lorenz, University of Michigan; Dr. Stephanie Clarke, University Hospital Lausanne; Dr. Paula Tallal, Rutgers State University of New Jersey; and Dr. Leslie Gonzalez-Rothi, University of Florida provided expertise in areas such as brain plasticity, neural mechanisms of higher thought processes, cognitive neuroscience, neurology, neuropsychological and cognitive assessment, cognitive rehabilitation, and successful intervention approaches.
The cognitive rehabilitation research teams worked on their projects for six to eight months prior to the September 2004 meeting. Two weeks before the meeting, each team submitted to the NIH Program Directors and the Advisory Panel a comprehensive white paper detailing some potential rehabilitation approaches, the rationale and supporting background materials for these choices and a plan for implementation. The workshop was devoted to the presentation of the cognitive rehabilitation approaches developed by the Stroke, TBI and Brain Tumor teams. Intervention approaches were discussed in terms of potential outcome, feasibility and adaptation to clinical settings. On the second day, a final session was devoted to the development of an implementation plan and drafting of recommendations.
The "Cognitive Rehabilitation" workshop focused on Stroke, Traumatic Brain Injury, and Brain Tumor. This limited set of neurological conditions was chosen because these conditions offered the possibility of residual capacity within specified lesioned circuits and potential for functional return.
Presentation topics focused on the state of our knowledge on research issues such as: plasticity and integrative brain function, models of successful interventions in verbal and motor abilities, neural pathways associated with executive function, characterization of executive function in neurological diseases, factors that modulate executive function, behavioral and/or pharmacological interventions, applications of the principles of attentional processes, adaptive learning, guided recovery and mental retraining, and the use of modeling approaches and imaging as tools to predict and follow functional recovery.
III. Workshop Presentations and Discussions
III A. Brain Tumor/cognitive deterioration
Tumors of the central nervous system (CNS) are a critical concern for neuroscientists, as up to 18,000 are diagnosed in a given year. Secondary tumors originating in other sites are far more common. The majority of primary CNS tumors occur in the brain, and up to 90% are diagnosed in adults. Interventions to maximize neuroprotection and neurorecovery are being explored to preserve cognitive abilities and rehabilitate deficits. Special consideration must be given to the negative effects of medical treatments widely used to treat tumors, and mechanisms to mediate these effects must be explored.
Brain tumor survivors experience a range of deficits that depend largely on the type and location of the tumor, the momentum of growth, and the treatments provided to eliminate dangerous tissue. In addition to the cognitive deficits that result from tumor infiltration and resection procedures, patients also experience fatigue, mood disorders, sleep disturbance, and sexual dysfunction. The mechanisms responsible for these symptoms are not well understood, challenging clinicians to develop effective treatments. There is some evidence to suggest that a tumor with slower momentum may allow for greater neuroplasticity and produce fewer cognitive symptoms. Individuals at the extreme ends of the age spectrum appear to be the most vulnerable to cognitive deficits, although additional studies with appropriate controls are needed to determine the reasons for this disparity. A related issue may be that children usually experience tumors in the posterior fossa region, whereas adults are more often affected in the supratentorial regions of the brain.
Rehabilitation efforts must also address cognitive impairments that are a result of adjuvant therapies. Radiation therapy is associated with damage to the vasculature of the brain, white matter injury, and demyelization. Adverse outcomes that develop 6 months or more following radiation are referred to as late radiation-induced brain injury, and are characterized by both cognitive deficits and fatigue. Immunotherapy is another adjuvant therapy increasingly used with adult populations, despite the potential for producing significant cognitive impairment (frontal-subcortical), executive dysfunction, memory deficits, mood disturbance, and apathy. Researchers have proposed that neuroendocrine, neurotransmitter, and cytokine pathways may mediate the effects of this type of therapy. Hormonal therapies can also have profound effects on cognition, particularly memory. Neurochemical and neurotransmitter treatments are linked to membrane damage in areas remote from the original tumor location and the targeted radiation area.
Primary prevention of brain tumor has a small likelihood of success, as the etiology of brain tumors remains unknown. Instead, researchers have focused their attention on neuroprotective efforts subsequent to the primary injury. Neural stem cells, which would potentially be delivered to regions adjacent to the tumor location, may someday provide protection from progressive injury. Anti-inflammatory agents such as nonsteroidal anti-inflammatory drugs (e.g., aspirin, acetaminophen, ibuprofen, and some COX inhibitors) may be useful in treating symptoms, although more research into the antineoplastic properties of these agents is needed. Pharmacological interventions, particularly psychostimulants, have demonstrated some success in treating cognitive impairments and mood disturbance. Prophylactic behavioral rehabilitation is being explored in pediatric populations to determine if training patients in neurocognitive strategies prior to the onset of deficits will result in less disabling impairments.
Behavioral interventions have been understudied in brain tumor patients, despite the cost- effective nature of the treatment and the potential to maximize patients' level of functioning and quality of life. The heterogeneity of the clinical population, particularly in tumor type, progression, and treatment, make the development of practice guidelines difficult. Prognosis must be considered in the development of treatment goals. For patients with more advanced conditions and poorer prognosis, compensatory interventions (e.g., environmental modifications, prosthetic devices) may be the most beneficial in increasing quality of life and social interaction. For patients with tumors that are responsive to treatment, long term goals (e.g., graduating from high school) may be appropriate targets for treatment.
Strategies adopted from related disciplines have been applied to this population, including metacognitive strategies (special education, educational psychology), attention process training (brain injury rehabilitation), cognitive reframing of stressors, monitoring internal dialogue, and developing realistic and positive self-statements (cognitive-behavioral approaches in clinical psychology). Other approaches include exercise, relaxation therapy, and self-hypnosis to alleviate non-cognitive symptoms such as pain, nausea, and fatigue. Environmental modifications in school settings, work settings, or in the home may effectively increase patients' ability to function independently. Additional interventions may be needed to address ecological disturbances (e.g., sleep disturbance, poor socialization, inadequate nutrition) as these may also affect cognitive performance.
Discussions and Recommendations
To develop aggressive and effective cognitive rehabilitation interventions for brain tumor survivors
The prevalence of CNS tumors, particularly primary brain tumors, and the paucity of research into effective cognitive rehabilitation strategies warrants attention to this area. As the treatment of malignancies continues to improve, resulting in more survivors of this condition, the need for cognitive interventions in addition to other adjuvant therapy becomes even more critical. The most effective treatment regimen may consist of 1) neuroprotective efforts to prevent further progressive injury due to infiltration or adjuvant therapy, 2) adjuvant therapy, 3) environmental modifications and supports, and 4) cognitive-behavioral interventions.
To maximize neuroprotective strategies by identifying and minimizing the negative effects of adjuvant therapies, particularly radiation and chemotherapy
A clearer understanding is needed of the multiple compartments affected by radiation and chemotherapy. Certain areas of the brain may be more vulnerable to damage with these treatments (e.g., the sub-ventricular zones in rodent models). Although once thought that radiation only affected the vasculature of the brain, more recent studies suggest that the entire brain is affected and injury due to this treatment can be observed outside of the radiated area. As neurogenesis is ultimately affected, this must be taken into consideration in designing rehabilitation strategies. Rather than lowering dosages, which may be ineffective in treating the tumor, prophylactic interventions, including cognitive remediation training may begin prior to adjuvant therapies and before impairments occur. In children, this practice may advance their cognitive development to a subsequent stage so that treatment will not prevent these skills from developing later (e.g., children with posterior fossa tumors who never achieve developmentally appropriate reading skills).
To increase our understanding of the reorganization potential of the brain
This area of research must increase our current understanding of neural plasticity as it relates to brain tumor patients. In order to develop effective interventions, research must identify the critical mechanisms responsible for recovery, the temporal course of these processes, and mediators of this sequence. Studies must engage populations across the lifespan, as recovery may vary in course and be affected by different mechanisms. Genetic vulnerability to infiltration and treatment effects may also play a role in recovery. Neuroimaging may be an effective tool to determine the effects of interventions, both cost (e.g., neurotoxicity) and benefit (e.g., neurorehabilitation). Researchers must identify the most effective toolbox of imaging techniques illustrate spontaneous and facilitated recovery of function, and consider non-invasive, less expensive methods for assessing time-based changes with interventions (e.g., evoked potentials).
Barriers and proposed solutions
III B. Stroke/unilateral neglect
The neglect syndrome is common following stroke, particularly right hemisphere lesions. Neglect is less frequent and typically less severe with left hemisphere strokes. The syndrome is characterized by a failure to respond to stimuli in external space that cannot be explained by a dysfunction of the motor or sensory systems. Many different subtypes of neglect have been proposed, including neglect of an area in space in relation to the affected individual versus neglect of a proportion of objects in space. Researchers continue to be challenged with categorizing these subtypes, which are not mutually exclusive, either by performance-based factor analysis or identifying the underlying neuroanatomical correlates. Successful identification of these subtypes may have important treatment implications in the future, as subtypes may be more amenable to distinctly different interventions.
Current behavioral measures have not been successfully linked to neuroanatomical correlates and may be prone to experimental bias in assessment. Ideally, these measures would be sensitive to distinguish true cases of neglect from other comorbid disturbances and fluctuations in arousal. Assessments should include a clinical assessment of the conditions that are producing the most disability, in other words, the extent of the impairment in quantifiable terms as well as the clinical impact on a patient's functioning. Additionally, the assessment tools should be able to distinguish between neglect and other related, comorbid conditions that are inhibiting functioning. For example, drastically reduced arousal may be causing more impairment for a patient than a mild case of neglect. Appropriate outcome measures must be identified or developed in order to conduct clinical trials with this population.
Imaging may be used more effectively with as a gauge of progress in treatment only if there is a base of functional imaging of non-injured controls available for comparison. This will allow researchers to distinguish between normal physiological changes and those associated with recovery of function. These discoveries need to be linked to appropriate treatment strategies, and not be considered an end in themselves.
The development of effective cognitive rehabilitation strategies for neglect/stroke requires an evaluation of the effects of specific and replicable interventions, the duration of such effects, and an understanding of the mechanism responsible for producing functional changes (which may vary, depending on clinical sample). Outcome studies that assess functioning within one hour following treatment do not provide insight into the clinical relevance of the interventions or if "booster" sessions are needed to maintain effects.
There may be value in studying interventions delivered during both the acute and chronic stages of recovery. Although the level of cognitive functioning may progress rapidly during the acute phase, treatment in clinical settings often begins during this time and would benefit from guidelines to maximize patient recovery. Further studies may indicate that different mechanisms are responsible for recovery during this phase as opposed to the chronic phase, in that natural recovery may be facilitated rather than determined solely by interventions.
The pursuit of clinical trials with this population may be premature, as treatments for stroke and neglect are still in a state of innovation. Prior to initiating trials in this area, appropriate outcome measures are needed and a crosswalk between animal and human models of neglect must be established. Interventions that are based on animal models of neglect may not have a smooth transition into human clinical trials, as animals do not appear to develop the same neglect patterns as humans. For example, primates experience an equal amount of neglect regardless of hemisphere of injury, and cases are typically milder and transient than humans.
Discussions and Recommendations
Several areas of research were identified as promising and recommended for further study. Treatment approaches focused on the dopaminergic and noradrenergic neurotransmitter systems have shown some efficacy in treating neglect. Additional studies with other systems, particularly norepinephrine, may prove to be beneficial as well. There is also some evidence to suggest that the effects of pharmacological treatment may be enhanced by behavioral interventions, although additional studies are needed. Related disorders of attention and arousal may exacerbate neglect, and should be investigated for their role in this disorder. Treatments to encourage environmental re-mapping (e.g., virtual reality) as well as physiological treatments (e.g., transcranial magnetic stimulation) have shown some efficacy in existing studies, and both areas were identified as being worthy of further study. Stimulus deprivation, such as monocular patching, is commonly used in animal models but may have a negative or even instigative effect in certain patients. These methods should be used with caution and warrant further investigation.
Barriers and proposed solutions
III C. Traumatic brain injury (TBI)/executive function
The emphasis of this team is on the frontal lobes and the immediate frontal systems, since the pathology of TBI is most highly related to these brain regions. Traumatic brain injury produces primary brain damage of two broad types: diffuse axonal injury (DAI) and focal cortical contusions (FCC). Diffuse axonal injury is a microscopic shearing injury of axons and small blood vessels. Although the changes are diffuse, much DAI pathology is located deep in the frontal white matter. Such petechial hemorrhages may also involve subcortical structures with critical frontal projections, such as the ventral tegmental area of the midbrain and the anterior or medial thalamus. Functional imaging supports the frontal lobe effects of DAI; prefrontal hypometabolism alone correlated with executive, behavioral and memory dysfunction in a resting positron emission tomography (PET) study of patients with DAI. Disruption of ascending catecholaminergic inputs to prefrontal cortex by DAI could possibly explain the reported prefrontal hypometabolism.
Focal cortical contusions are caused either by a direct blow to the skull transmitted to the brain, or by powerful inertial forces (acceleration/deceleration) causing the brain to be abraded by adjacent skull and dura. Focal cortical contusions are primarily confined to the basal frontal and anterobasal temporal regions. Large deep hemorrhages from acceleratory disruption of subcortical penetrating vessels may involve subcortical white and grey matter (e.g., caudate circuits). Secondary damage to frontal systems after focal injury may result from delayed neuronal injury (as occurs after diffuse injury), herniation syndromes (especially frontal transfalcine herniation that may compromise medial frontal lobes and anterior cerebral artery perfusion) and hypoxic-ischemic injury (including anterior cerebral artery and middle cerebral artery anterior borderzone ischemia from systemic hypotension).
The pathology of TBI most often involves several brain areas and often results in syndromes that have been termed abnormalities of executive function. The definitions of executive function used here are more precise, involve highly complex behaviors, follow anatomy and evolutionary development, and are related to the physiological changes seen after TBI.
The team proposed a schema that divides what has been loosely termed "executive functions" into four more clearly defined and circumscribed domains: (1) executive cognitive functions, (2) behavioral self-regulatory functions, (3) activation regulating functions, and (4) metacognitive processes. A fundamental concern for future research in this area is operationalizing and further refining increasingly precise definitions of these functions.
Executive cognitive functions: In this workshop, we defined executive functions as high-level cognitive functions that are involved in the control and direction (planning, monitoring, activating, switching, inhibiting) of lower-level, more automatic functions. It is also important to recognize that, because of the interconnectivity between the lateral frontal and posterior regions, diffuse pathology can cause executive dysfunction.
Behavioral self-regulatory functions: These are the functions more associated with the ventral prefrontal cortex (VPFC). The ventral (medial) frontal region is part of the paleocortical trend emerging from the caudal orbitofrontal (olfactory) cortex, closely connected with limbic nuclei involved in emotional processing, including the acquisition and reversal of stimulus-reward associations. Because of this involvement in reward processing, a key role for the VPFC is behavioral self-regulation in situations where cognitive analysis, habit, or environmental cues are not sufficient to determine the most adaptive response.
Activation regulating functions: Activation plays a key role in self-regulation. More limited medial pathology results in disorders of activation and drive, clinically known as apathy or abulia. Activation regulating functions warrant further specific study.
Meta-cognitive processes: The frontal poles (possibly more particularly on the right), are most recently evolved. They appear to bridge self-regulatory and executive cognitive functions, because of their unique position to integrate the higher-level cognitive, executive cognitive functions, and emotional or drive-related inputs. The frontal poles are involved in the recently studied meta-cognitive aspects of human nature: personality, social cognition, autonoetic consciousness, and self-awareness.
In summary, the team chose to define executive function with four domains that can be approached within the context of diagnosis and treatment of conditions seen after TBI. Such anatomical/behavioral definitions are essential in understanding the sequelae of TBI, and the development of potential rehabilitation. Although there are finer distinctions within each of the above categories, they have yet to be studied in the context of TBI.
It was clear from the discussion of the complexities of executive function that it is necessary to bridge the gap between the state-of-the-art paradigms developed by cognitive neuroscientists and the more dated neuropsychological tests and rehabilitation packages used by clinicians to establish baselines and gauge progress in treatment. To that end, it is imperative to involve basic neuroscientists in clinical instrument design.
Functional imaging could be essential in revealing correlations between the nature and location of brain injury and behavior, as well as comparisons with "normal" brain anatomy and behaviors. Many powerful new techniques, including spectroscopy and diffusion tensor imaging, lend themselves to the study of systems of transmitters and white matter; both systems are highly involved in TBI and executive function. Most of the neuroimaging of TBI conducted by cognitive neuroscientists is taking place in the absence of converging evidence of brain differences among groups and without being informed by decades of neuropsychology research and practice. Potential new tasks and the theories they inspire are not being extended to the perturbed brain, where they could have an important impact on the development of strategies to re-establish impaired functions. Moreover, this kind of cross-integration is necessary for the development of new neuropsychological assessment tools. Convergence among methods and disciplines would be particularly useful in TBI and executive function.
Discussions and Recommendations
To utilize knowledge of CNS plasticity in the development of novel interventions
Two disconnects were realized during the course of this discussion. The first disconnect exists between animal models of injury and recovery and their applicability to human survivors of TBI. Although animal research has provided insights into the relationship between injury location and related deficits, the heterogeneity of human brain insults and comparatively complex cognitive abilities leave many questions unanswered by existing animal research. The second disconnect is the chasm between basic neuroscience research and interventions currently being utilized by neuropsychologists for treatment of TBI. Just as clinicians do not rely on current neuroscience research to provide cognitive rehabilitation, so do basic neuroscience researchers operate independently of the knowledge gained from decades of neuropsychological practice.
A collaborative workshop was recommended to bring together basic neuroscience researchers and practicing neuropsychologists. This session would invite neuroscientists to discuss clinical instrument design and encourage the use of functional imaging to identify the neural correlates of recovery and spared function as well as alternate neural routes by which recovery may be achieved. Studies with non-injured controls could determine the appropriateness of distinctions made in current taxonomy, including the four executive function domains outlined for the current session. An additional focus of the proposed workshop would be that clinicians be exposed to the most up-to-date neuroscience research and discuss translation of these advances.
To promote the development and implementation of evidence-based practices
There is a need to determine the clinical significance of progress achieved in rehabilitation settings. Currently, there is a paucity of research in the area of transfer effects of rehabilitation gains in natural settings. This area is of critical importance, as even effective interventions may be of little value if they do not result in increased functional independence outside of controlled environments.
Several considerations were raised related to the development of effective cognitive rehabilitation regimens. There is substantial evidence to suggest that gender differences exist in response to neurological insults, and therefore different mechanisms of recovery may be active in men and women. Intact executive function may be necessary for the rehabilitation of other areas of cognition, such as memory, attention, concentration, and language. Particularly in patients with moderate to severe injuries, impaired executive function negates training efforts in the use of internal and external compensatory strategies. The temporal significance of designing rehabilitation strategies may also be critical in outcomes, although little is known about the effectiveness of interventions during the acute phase of recovery. Assessment is difficult at this point because deficits may be masked in highly restrictive environments and training may be futile as natural recovery is occurring at such a rapid pace.
To link diagnosis with interventions
Careful assessment and diagnosis may be the most critical factor in developing an effective rehabilitation plan following TBI. Research to date does not establish a model method for making a diagnosis based on injury parameters versus behavioral presentation of deficits. Complicating this issue is that seemingly identical injuries may produce distinctly different symptom patterns across individuals, and similar deficit patterns may be attributed to remarkably different injuries. Diffuse injuries are likely to require different interventions than focal injuries to achieve recovery of function.
Current assessment methods for executive function are reasonably sensitive for detecting deficits, but may lack adequate specificity as to the nature of the deficit and fail to provide clues as to the underlying neural cause. Assessment tools should be reliable and valid independent of the evaluator's level of expertise and so could be administered by a less senior health care professional. Diagnosis that is informed by an anatomical understanding of a particular injury, gained through imaging, can increase clinicians' ability to develop an appropriate intervention. However, this is dependent on current understanding of neural correlates of cognitive dysfunction and the physiological course of recovery.
Barriers and proposed solutions
IV. Summary and Future Directions
Last Modified August 5, 2005