Neuromodulatory Interventions for Traumatic Brain Injury.

Abstract

THE IDEA that noninvasive neuromodulation could be used for traumatic brain injury (TBI) rehabilitation originated, largely, from evidence that transcranial magnetic stimulation (TMS) improves motor function and mood regulation.1–4 These seminal findings suggested to many rehabilitation researchers that TMS could be used to treat functional impairments for many neurologic conditions,5 including TBI.6 While the idea of using neuromodulation for therapeutic purposes evolved from TMS studies, the growing interest in examining the merits of neuromodulation for TBI rehabilitation is related, in part, to the growing number of noninvasive devices (eg, intermittent theta burst stimulation,7 transcranial direct8 and alternating9 current stimulation, numerous vestibular re-training devices, and deep-brain ultrasound10) that make it more feasible to deliver brain stimulation.11 Further contributing to growing interest is the integration and translation of knowledge across the basic and clinical sciences (eg, neuroscience characterizing the brain circuitry driving sensorimotor functions, cognition, and behavior; neuroplasticity characterizing reorganization and alterations of circuitry; neurophysiology characterizing the neural injuries; biomedical engineering; and clinical specialties of medical rehabilitation, neurology, and psychiatry), ultimately providing an empirical basis to study the merits of using of neuromodulation in TBI rehabilitation. Technological advancements, coupled with convergence of knowledge across disciplines, highlight the importance of understanding the potential role for neuromodulatory interventions in TBI rehabilitation. NEUROMODULATION WITHIN THE CONTEXT OF TBI REHABILITATION Neuromodulation, as defined by the International Neuromodulation Society (INS),12 is the “alteration of nerve activity through targeted delivery of a stimulus to modulate abnormal neural pathway behavior caused by the disease process.” “Alteration of nerve activity” indicates that neuromodulation may be mediated by stimuli of varying intensities both below and above the firing threshold of a neural target. A sub-threshold stimulus can lead to depolarization of the membrane, but the magnitude of the depolarization does not reach the critical level to initiate an action potential in an excitable cell. In contrast, neurostimulation at or above the firing threshold initiates an action potential that may mediate activity or signal transmission between neurons, causing downstream effects within and between neural pathways. It is easy to see that initiating an action potential, using neuromodulation, plays a critical role in facilitating recovery after TBI. When considering the large body of work indicating that stimuli of varying strengths are needed to engage the mechanisms of meta-plasticity,13 that support relearning and skill restoration, it also becomes evident that subthreshold stimuli are equally important to TBI rehabilitation and recovery. Applying the INS definition to TBI rehabilitation may be too restrictive, in that the definition specifies that the stimulation target is the “abnormal neural pathway.” Limiting neuromodulation to the abnormal pathway is contrary to a growing understanding of how neuroplasticity supports recovery after acquired brain injury. Specifically, evidence suggests that network-level reorganization after acquired brain injury allows more intact neural pathways to compensate for less functional pathways. Right lateralized network reorganization, for example, is related to language recovery after stroke.14,15 Furthermore, a growing number of advanced imaging studies indicate that neuronal changes occurring within and between functional and dysfunctional pathways also support TBI recovery. Alterations in network communications (ie, firing synchrony) and structural neural integrity (eg, white fiber tracts) of more and less intact pathways, for example, are related to recovery of functional skills.16–20 When considering this evidence coupled with the knowledge that neuromodulation can enable a dynamic regulation of neuronal circuits,21,22 neuromodulatory interventions for TBI should not be restricted to targeting of abnormal pathways. For TBI rehabilitation, neuromodulation has the potential to reconfigure neural networks into different functional circuits, improve functioning and structural integrity of viable networks, engage dormant networks, and possibly create new neural connections. The INS definition is also problematic for TBI rehabilitation, as it does not reflect the very limited evidence regarding the safety of neuromodulation of any neural pathways after TBI. This is concerning given long-standing evidence that persons with TBI, without neuromodulation, are at an elevated risk for seizures.23 There are empirically based safety guidelines for TMS,24 but even for TMS there remain very few studies reporting safety for TBI. Considering the limited evidence, this special issue includes a report of safety findings when repetitive TMS (rTMS) is provided, alone and when combined with pharmacological neurostimulants, to persons remaining in states of disordered consciousness after TBI (see Kletzel et al25). A scoping review is also included (see Oberman et al26), in part, because it provides a summary of rTMS safety evidence for mild TBI. While these reports address knowledge gaps, there continues to be limited safety evidence specific to TBI. Thus, safety of neuromodulation of any neuropathway after TBI is largely unknown. PROMISING CONTRIBUTIONS OF NEUROMODULATION TO RECOVERY AFTER TBI Neuromodulation is therapeutically promising because it has the potential to facilitate and/or initiate recovery of somatosensory and higher order, cortically based skills after TBI. This is possible because the effects of neuromodulation include induced or enhanced plasticity, as indicated by alterations of signal transmission between neurons. Initiating or enhancing signaling ultimately modulates the activity between neurons that comprise the neural pathways supporting sensorimotor functions, cognition, and behavior. Neuromodulatory effects can also include engagement of activity-dependent neural mechanisms, termed “metaplasticity.”13 Engaging these mechanisms is necessary to regulate neural repair and to enable relearning (ie, via acquisition of long-term potentiation [LTP] and depression [LTD]-like mechanisms).27 Providing rehabilitation training exercises during the time of induced plasticity, for example, is thought to engage activity-dependent neural mechanisms in targeted networks. Informed selection of training exercises is key to targeting the networks known to support the behavioral functions identified for remediation. Engaging these mechanisms enables neural adaptation in the targeted networks to modify the structure and function of neurons, ultimately tuning synapses and networks, to guide recovery of the targeted behavioral functions. Considering that the goal of TBI neurorehabilitation is to modulate dysfunctional as well as functional neural pathways to create the neural changes necessary for sustained recovery of functional skills, these profound effects of neuromodulation are important contributions to TBI rehabilitation. FOCAL AND DIFFUSE NEUROMODULATION OF VARYING STRENGTHS Inducing and enhancing plasticity and engaging mechanisms of metaplasticity can now be achieved by selecting from a wide variety of neuromodulatory techniques including noninvasive devices, pharmacological agents, and rehabilitation training interventions. For research and clinical use, particularly after noninvasive devices receive approval by the US FDA for clinical use in specific populations, it is important to understand how to select the technique(s) aligned with each research question or each patient's rehabilitation goals. Currently, there are 2 parameters, focality and strength of stimuli, that provide a starting point to inform these research and clinical decisions. Neuromodulatory techniques can be used, independently and in combination, according to the strength and focality of stimulation needed to elicit the desired neurobehavioral effect. Strength will determine whether the neuromodulatory intervention produces sub- or suprathreshold stimulation. For noninvasive devices designed to alter electrical signaling, strength of stimuli is characterized by traditional engineering parameters (eg, voltage).28 The radius, or focality, of stimulation will determine how precise a brain region can be targeted. Focal stimulation is thought to alter the activity of a specific brain region, whereas diffuse stimulation is thought to alter the activity of broad neural circuitry. Most noninvasive neuromodulatory devices are capable of providing stimuli spanning a range from focal to diffuse, but an example of a device thought to strictly provide focal stimulation involves the use of ultrasound.10 As a fairly new application, the mechanism of how the ultrasonic sound waves interact with cells to alter brain activity is unknown, and once the sound waves penetrate the neural tissue the strength of the stimulation is also unclear. It is thought, however, that ultrasound can provide stimulation to a specific brain region, and the safety and merits of targeting the thalamus with low-intensity pulses are currently being examined for severe TBI.29 Repetitive transcranial magnetic stimulation (rTMS) is one of the most widely studied noninvasive neuromodulatory interventions, in part, because it can be used to provide stimulation, at multiple sites and because the stimuli can range from focal to diffuse as well as from sub- to suprathreshold stimuli. Focality of the stimulation is determined according to the coils used (eg, circular, figure of 8, and cone) whereas stimuli strength is determined by pulse frequency,26,30,31 pulse intensity, and temporal patterns of pulses.30 These 3 parameters can be manipulated to provide more rapid or slower activity than would be physiologically normal, thereby inducing greater or slower neuronal depolarization. Repetitive TMS is also of interest to researchers because it can also induce secondary activations in other regions or networks outside of the radius of stimulation. Specifically, rTMS-induced activations are maximal on the brain surface closest to the stimulating coil and rapidly dampen toward deeper brain areas.32 Yet, activations spread from the surface to other brain regions in a white matter connectivity-dependent manner.22 The flexibility of stimuli that can be provided with rTMS, in terms of focality and strength, allows researchers and clinicians to address the needs of TBI rehabilitation where patients present with heterogeneous neurologic sequelae and a vast range of TBI-specific impairments. For example, we are currently completing 2 double-blind placebo-controlled randomized clinical trials examining different types of rTMS. One trial is examining effects of our previously published rTMS protocol6,33,34 for persons remaining in states of disordered consciousness after TBI, and the other is examining the effects of a type of rTMS, intermittent theta burst stimulation, on attention impairments persisting after mild TBI. Considering the alignment between rTMS capabilities and the heterogeneous presentations of TBI as well as the growing body of evidence for rTMS safety and efficacy with TBI, this special issue includes a scoping review (see Oberman et al26) of the safety and empirical basis for using rTMS to treat mild TBI-related neurocognitive and neuropsychiatric symptoms. The focality and strength of stimulation provided by behaviorally based interventions are largely unknown, but the strength of stimuli is presumed to be subthreshold and interventions are designed with the intent of targeting networks known to support specific behavioral functions. While there is very little known about the nature of central nervous system stimulation, there is growing body of evidence suggesting that behaviorally based interventions modulate brain activity. For TBI rehabilitation these interventions are important to consider, in part, because these interventions are thought to provide persistent subthreshold stimulation to maintain a neuron in either a depolarized or hyperpolarized state, thereby either priming or inhibiting its response to external stimuli. Cognitive training exercises after moderate to severe TBI, for example, are based on this idea, and a recent study by Han and colleagues35 demonstrated that cognitive training resulted in global network changes. They tested the idea that cognitive training could be used to repeatedly reference or engage the executive control network that would, in turn, modulate its intrinsic relationships with other networks. We also reported a study of repeated exposure to familiar stories, for persons in states of disordered consciousness, based on the idea of providing persistent subthreshold stimuli.36 This study also demonstrated that prolonged repeated exposure to familiar autobiographical stories increased neural responsivity in right hemisphere language homologs37 and altered resting-state functional and structural connectivity of networks important to recovery from severe TBI.38 Thus, although little is known about the strength and focality of behaviorally based neuromodulatory interventions, they have clinical effects, and evidence is emerging that these clinical gains are related to induced changes in neural activity. PAIRING NEUROMODULATORY INTERVENTIONS The evidence, integrated across disciplines, suggests that pairing neuromodulatory interventions that provide stimulation of different focality and strength is a plausible approach to optimizing the effects of each neuromodulatory intervention individually. Specifically, stimulation of varying strengths could be provided simultaneously or in strategic sequences to elicit subthreshold stimuli or action potentials, thereby inducing neural plasticity and/or prolonging the induced plasticity to either (i) generate persistent forms of synaptic plasticity, such as LTP and LTD that are used to learn new information or relearn lost skills,27 or (ii) engage mechanisms of metaplasticity13 that serve to constrain or regulate the induced plasticity and can be leveraged to support recovery of specific functional skills. It follows that, based on selection of neural targets, strategic pairing of interventions could provide synergistic modulation, via different neural mechanisms, ultimately increasing the likelihood of generating LTP or LTD mechanisms and/or engaging mechanisms of metaplasticity, both of which are important to TBI rehabilitation. Considering the potential therapeutic implications, the effect of pairing different types of neuromodulation is important to study. While this idea is just starting to be investigated, emerging evidence supports the merits of this idea. For example, the provision of transcranial direct current stimulation (tDCS) prior to starting attention training exercises to, presumably, prime the brain's ability to engage in the training exercise improves outcomes.39–41 It is thought that the role of tDCS is to standardize the brain state, thereby providing a neural environment for optimal engagement in training exercises.42 Considering that this is a nascent area of neuromodulatory research, this special issue includes a report (see Bender Pape et al43) of a study that examined using rTMS and amantadine for severe TBI. For persons living in chronic states of disordered consciousness, either rTMS or amantadine was provided prior to the pairing of rTMS and amantadine. The idea being examined was whether or not the provision of rTMS or amantadine could each modulate activity of distinct networks that, via different mechanisms, modulate dopamine.21,44,45 Considering the role that dopamine plays in TBI recovery,46 the merits of combining these treatments was examined in terms of changes in neurobehavioral function and resting-state functional connectivity of neural networks important to recover from severe TBI. Although we are learning more about the efficacy of multiple forms of neuromodulatory interventions for TBI rehabilitation, our strides forward raise many questions about how to optimally pair neuromodulatory interventions to address the needs of individual patients. DEFINING NEUROMODULATION FOR TBI REHABILITATION In the context of TBI, we offer a definition of neuromodulation based on current empirical evidence integrated across disciplines. We offer this definition to provide a basis for future discussion. Specifically, we propose defining neuromodulation, in the context of TBI rehabilitation, as “the alteration of nerve activity through targeted delivery of stimulation provided to modulate dysfunctional as well as functional neural pathways to support neural repair and neural alterations necessary for sustained recovery of functional skills valued by the patient.” It is our hope that this definition will evolve over time by linking evidence, as it emerges, across disciplines. To further inform this definition, research is needed across disciplines to understand how neuromodulation contributes to functional recovery after TBI. RESEARCH PRIORITIES Considering the heterogeneity of TBI etiology, pathology, comorbid conditions, and recovery trajectories, it is important to consider how neuromodulatory research can help us realize the dream of precision TBI rehabilitation. To match a person with TBI with the neuromodulatory intervention or combination of interventions that will optimize their recovery, there is a need for empirically based guidelines. Empirically based guidance is needed to enable informed choices on how to use neuromodulation to support sustained recovery of functional skills after TBI. To develop the empirical basis for these guidelines, future research is needed to explicate the contributions of varying strengths and focality of stimulation and to identify other parameters that can be leveraged in neuromodulation. The matching of neuromodulatory interventions to a person with TBI and their specific symptoms or functional needs will also require future research to identify the constellation of unique characteristics of persons with TBI driving responses to neuromodulatory interventions. Considering that neuropathology and recovery trajectories are heterogeneous across persons with TBI,4,5 responses to neuromodulatory treatments are likely to be similarly heterogenous. As highlighted in a review in this special issue (see Phillips et al47), for example, responsiveness to neuromodulation will likely be influenced by biological sex. This review indicates that, currently, we have a very limited understanding of how biological sex influences responsiveness to neuromodulation. Other personal factors such as age, genetics,48 and complexity of one's environment49 are also likely to influence responsiveness. To give each patient the best chance at an optimal recovery, explicating the role of these person-specific factors to treatment responsiveness is important for future TBI neuromodulation studies. When considering the primary and subsequent injury processes leading to heterogeneous outcomes for TBI and that the basis of precision neurorehabilitation is tailoring of interventions to person-specific factors influencing treatment responsiveness,5 it is imperative that future neuromodulatory research also address the need for precision neural targeting. Herrold and colleagues50 report in this special issue on their integration of advanced imaging and clinical outcome data to develop, test, and implement techniques ultimately enabling individualized neural targeting for rTMS. By utilizing advanced neuroimaging coupled with neuronavigated rTMS and integrating this data with clinical outcome data, Herrold and colleagues demonstrate that it is possible to provide rTMS targeting normal neural pathology and each patient's uniquely aberrant pathology. Importantly, this report demonstrates that the site of normal or abnormal pathology can also be selected for neuromodulation with rTMS according to each patient's specific symptoms and functional impairments. These advancements set the stage for future research to determine extent that this enhanced precision relates to improved clinical efficacy, ultimately informing researchers how to further optimize precision of neural targeting for neuromodulatory interventions. The role and timescale of altered brain activity in the recovering brain is poorly understood, and research in this area will also further develop precision neuromodulation for TBI rehabilitation. One critical knowledge gap, for example, is knowing the optimal time, during recovery trajectory, to provide neuromodulation that will elicit optimal responsiveness. Equally important is advancing knowledge regarding how the brain states change in stability during recovery and if a more versus less stable neural environment supports optimal responsiveness. Addressing these research questions would also likely address another knowledge gap, the possibility that neuromodulation can lead to maladaptive plasticity causing harm or worsening a condition. The paucity of knowledge on the likelihood of maladaptive plasticity underscores the limited evidence of safety. There is clearly a need for safety studies, specific to TBI, that will enable development of safety profiles for differing types of neuromodulatory interventions. In summary, the knowledge gaps for the use of neuromodulation in TBI rehabilitation are vast. While vast, future research addressing the knowledge gaps identified here, and in the articles included in this special issue, will advance the capability to safely provide precision TBI rehabilitation where neuromodulatory interventions are strategically combined to optimize functional recovery after TBI. CONCLUSION In this special issue of the Journal of Head Trauma Rehabilitation, we share with readers some of the latest advancements in neuromodulation specific to TBI, while providing the framework to further our understanding of how and why functional skills are likely improved. While neuromodulatory interventions can play a critical role in functional recovery for those with TBI, the heterogenous nature of TBI means that clinical implementation of neuromodulation will require understanding, at the individual and group levels, of how, when, and where to alter brain activity to support sustained recovery of sensory and higher order functions. To address barriers to future clinical adoption,51,52 future research must also advance understanding of individual response variability, relative to average group responses, to neuromodulatory interventions. Only after research in these critical areas, will we be able to develop empirically based treatment algorithms for use in daily TBI rehabilitation. Given the potential role for neuromodulation in TBI rehabilitation, this issue is dedicated to research advancing our understanding of the contributions of noninvasive neuromodulatory interventions to TBI rehabilitation and recovery. Theresa L. Bender Pape, DrPH, MA, CCC-SLP/L, FACRM Editor Amy A. Herrold, PhD Ann Guernon, PhD, MS, CCC-SLP/L Alexandra Aaronson, MD Joshua M. Rosenow, MD, FACS Co-editors