Neural-Targeted Rehabilitation Strategies to Address Neuroplasticity After Joint Injury

Across a myriad of musculoskeletal pathologies (ankle sprain, anterior cruciate ligament [ACL] injury, low back pain, and glenohumeral instability, to name a few), clinicians returning athletes to participation and daily activities face consistent frustrations as these individuals will report back to the clinic with recurrent injury, along with continued impairments that affect daily function, generating activity limitations. Reports suggest that reinjury rates may be as high as 70% following ankle sprain, 30% to 40% following ACL injury, 75% for nonspecific low back pain, and 92% for glenohumeral dislocation.^1–5  In the case of ligamentous pathology, these injuries have a strong potential to progress along a continuum that includes joint instability during functional activity and posttraumatic osteoarthritis.^6,7  This negative progression of function following an initial injury appears to occur despite current rehabilitation efforts, with many individuals experiencing subsequent reinjury weeks to months following return to sport.^3,8  Collectively, these negative long-term outcomes and high reinjury rates contribute to high degrees of disablement and decreased health-related quality of life among patients with joint injury, leading to a question of whether new therapeutic approaches are needed to curb these high reinjury rates.^9,10  The purpose of this commentary is to provide a summary of potential manners in which changes within the central nervous system (CNS) subvert rehabilitation efforts following musculoskeletal injury (with an emphasis on the knee and ankle joints) and to offer the practicing clinician neural-targeted rehabilitation strategies that may better address the alterations in nervous system function and activity, collectively known as neuroplasticity, experienced by individuals with ligament injury. Additionally, this commentary will offer insight into the emerging role of brain stimulation (specifically electrical brain stimulation) in musculoskeletal rehabilitation and its potential impacts on addressing injury-induced neuroplasticity.

A key emerging factor that aids in explaining the phenomenon surrounding injury recurrence is changes within the CNS.¹¹  Both movement-related outcome measures (eg, changes in gait function) and neurophysiologic measures (eg, electroencephalography and neural excitability) have suggested impairments in patients with knee and ankle ligament pathology that reflect reorganization of the CNS.^12–14  These neuroplastic changes seemingly modify motor planning and leave individuals vulnerable to episodes of reinjury and giving-way as they are unable to adequately stress-shield the joint. These alterations may contribute to sensations of giving-way and lead to abnormal contact forces that lead to joint degeneration.^12,13,15 

These negative CNS adaptations have largely been observed in individuals with chronic dysfunction (eg, chronic ankle instability and following ACL reconstruction), leading investigators to better understand how an initial injury progresses to a state of maladaptation.¹¹  It has been proposed that following initial injury (ie, joint sprain), acute injury symptoms such as pain and swelling at the site of injury contribute to arthrogenic muscle inhibition (AMI) at the segmental level, affecting spinal cord reflexes and making it more difficult to activate stabilizing musculature surrounding the joint.¹⁶  Subsequently, after acute symptoms subside, the persistent AMI combined with sensory deafferentation and joint laxity are associated with depression of some measures of motor cortex excitability, such as motor threshold and motor evoked potential size, albeit with some inconsistency.^17,18  The combination of decreased motor cortex excitability and AMI makes activation of stabilizing musculature more difficult, necessitating the recruitment of “extraneous” areas. For example, planning and visual areas of the cortex and the cerebellum may be required to initiate even simple movements, reflecting decreased neural efficiency (Figure 1).^19–21  Brain imaging and electrophysiologic studies have demonstrated an increased activation of planning (eg, frontal cortex and supplementary motor areas) and coordinative (eg, cerebellum and contralateral motor cortex) brain areas following ankle and knee joint injury.^22,23  Collectively, this suggests that more efficient motor pathways to activate muscles are impaired, and movement and muscle function are restored using greater amounts of cortical activation.

Although addressing impairments such as strength or range-of-motion deficits can be easily appreciated, maladaptive neuroplasticity and decreases in neural efficiency do not offer a consistent clinical impairment that can be easily targeted or even assessed outside of a research lab. This may explain the frustrating gap experienced by clinicians and patients alike, where reinjury occurs and residual impairments remain even after laborious rehabilitation efforts and despite “successful” return-to-activity testing. For instance, rehabilitative exercises allow for a large degree of task-related focus in a controlled environment, whereas injuries are more likely to occur in chaotic, unconstrained environments. Muscle activation and patient function appear typical in these controlled environments (ie, clinical settings); however, movement patterns and coordination likely degrade on return to unconstrained play. Although clinicians may attempt to replicate these environments and establish better return-to-play criteria that include dual-task and cognitive loading or approach return to play in a gradual manner, it may not address the underlying maladaptive neuroplasticity.^19,24,25 

Following a joint injury, current best practices in injury rehabilitation suggest impairment-based rehabilitation as the optimal standard of care for these patients.^26,27  This process dictates that patients are assessed for specific impairments (eg, pain, strength, and balance), treated for that specific impairment, and then reassessed until function is typical compared with preinjury values or with an uninjured limb. Intuitively, this is a strong approach to injury rehabilitation as it attempts to minimize any patient functional impairments by the time they are discharged from care. However, under the scope of maladaptive neuroplasticity, correction of many of these impairments may be achieved using atypical neural activation. Unfortunately, these atypical neural activation patterns that result in successful performance in the controlled environment of a formal rehabilitation setting may prove insufficient in the uncontrolled real world. As an increased challenge, maladaptive neuroplasticity does not seem to offer a consistent clinical impairment that can be easily targeted, although research is ongoing into assessment techniques that can identify these atypical pathways in the clinical setting.²⁴  However, this does not prevent a clinician from considering how to address neuroplasticity throughout the rehabilitation process. To optimize the translation of the effects of impairment-based rehabilitation beyond the controlled setting of the clinic, traditional rehabilitation activities may need to be paired with neural-targeted rehabilitation strategies that mediate maladaptive neuroplastic changes.

With the onset of many maladaptive changes being tied to sensory aberrations (eg, pain, swelling, deafferentation), clinicians and researchers have incorporated an array of sensory-targeted interventions to improve function and reduce injury rates among these patients. Proposed interventions have included massage and insoles aimed at improving cutaneous receptor function, joint mobilizations to activate joint mechanoreceptors (eg, Pacinian corpuscles, Ruffini endings), stretching and vibration protocols to target the muscle spindle, and perturbation training that targets the Golgi tendon organ.^28,29  Each of these therapies is tied to positive clinical outcomes in patients, although with some inconsistencies, particularly when it comes to which therapies might be most effective.³⁰  Increases in neural excitability have been observed with sensory-specific interventions, including focal joint cooling, transcutaneous electrical nerve stimulation, and taping interventions, whereas those targeting deeper tissues (eg, massage and joint mobilizations) have not generated similar results.³¹ 

An alternative approach toward addressing neural function is to target the output of the system, which is motor function, via motor disinhibitory interventions. Although a variety of strengthening exercises (eg, strengthening with resistance bands, body weight) would be effective in improving motor function and clinical performance, the pathways by which some exercises achieve this potentially encourage recruitment of extraneous brain areas, decreasing neural efficiency by involving more activation from the cerebellum, planning, visual, and contralateral regions of the brain. A strategic option for clinicians is to purposefully choose interventions that have the potential to reverse AMI and cortical inhibition that may be impairing a patient’s ability to activate stabilizing musculature, thereby diminishing the need to recruit additional cortical resources.^31,32  Current evidence suggests that optimal motor disinhibitory interventions encourage high levels of muscle activation, which is theorized to increase the neural awareness of that muscle. For example, eccentric-focused strengthening allows for the muscle to withstand higher loads than those performed isotonically, challenging the nervous system to activate the muscle in a manner that is sufficient to withstand injurious loads.^33,34  Alternately, plyometric training involves high power production that encourages rapid recruitment of motor units.³⁵  Electrical stimulation can also work to achieve this goal, with neuromuscular electrical stimulation shown to aid in disinhibition when increased to the highest level tolerable by the patient.³⁶  Collectively, evidence suggests that each of these interventions is capable of improving muscle function; however, the evidence that it modifies neural excitability or changes cortical activation is far more limited, leading to caution when interpreting these findings.

Sensory-targeted interventions and motor disinhibitory interventions are well-thought interventions that aim to restore neural function after joint injury; yet there appears to be a shortfall when considering their effectiveness in restoring long-term patient function and reversing maladaptive neuroplasticity. We propose that these interventions should be a foundation for neural-targeted rehabilitation, as increased effort must be taken when translating the benefits from these common therapeutic tools to appropriate neural activation. There must be a consideration of integrative therapies that will act to cement changes from aforementioned sensory and motor interventions and transition them into appropriate activation (Figure 2).

In clinical settings, traditionally, sensorimotor integration is considered a goal of interventions such as balance and perturbation training, where individuals must use sensory information from the joint to create appropriate motor responses. Balance training does in fact contribute to the automaticity of balance, but it does little to curb the neuroplasticity that has been associated with reinjury.³⁷  Instead, we propose integrative therapies that are aimed at optimizing and automating the manner in which the CNS interprets sensory function and executes a motor response, specifically by decreasing the resource utilization necessary for these pathways.

Integrative therapies may take several forms, but 2 general categories may include motor learning strategies and brain stimulation. Motor learning strategies as tools to improve movement automaticity have existed for decades but have recently been highlighted in application to individuals with joint injuries, most notably in a recent review by Gokeller et al.³⁸  Highlighted in the review is the use of implicit learning, external focus, differential learning, and self-controlled learning strategies to improve neural function.³⁸  Implicit learning emphasizes minimalization of explicit knowledge of movement execution during the learning of a motor task. For instance, as a patient completes an exercise, offering feedback of the task being done “good” or “bad” (or “too fast” or “too slow”) is preferred versus correcting the movement itself (eg, “bend your knee more” or “activate your calf”), as it offers patients the opportunity to discover their own movement solutions.³⁹  External focus highlights the use of cues outside of one’s body in performing a task. For example, rather than instructing patients to keep knees over toes or contract a specific muscle, use environmental cues (eg, aligning to the room or using a biofeedback device) to achieve a similar goal.^40,41  Differential learning emphasizes using the aspects of dynamic systems theory to encourage flexible movement strategies by modifying the task, environment, and individual to challenge motion in a manner that forces individuals to develop affordances.^41,42  Increasing task complexity in a rehabilitation setting may support developing more efficient neural pathways for simplified movement. Last, self-controlled learning emphasizes giving patients the option to select components of their daily treatment plan within a series of options from their clinician.^43,44 

The above are brief explanations and examples, with these techniques detailed in the aforementioned review,³⁸  but there are certain features to highlight in their role as integrative therapies. Each of the above strategies is easily incorporated with current rehabilitation strategies, particularly motor interventions emphasizing disinhibition. For example, during eccentric-focused exercise, patients can be encouraged to implicitly learn the pacing of the exercise through minimized feedback (eg, too fast/too slow) or provided with external feedback to maintain proper joint alignment. However, although the use of these motor learning strategies is well established as aiding in the retention of movement strategies in patients with joint injury, its relationship to correcting neuroplasticity is largely theoretical in these populations. Further, as the use of these strategies is based around motor learning, it is unclear how they could be incorporated with sensory strategies.

An alternate approach to integrative therapy lies in noninvasive brain stimulation techniques. These techniques include transcranial direct current stimulation (tDCS), repetitive transcranial magnetic stimulation, and transcranial alternating current stimulation and share the common goal of aiming to modify neural plasticity and facilitate motor learning. Specifically, these techniques use energy (eg, electricity, magnetic, light) to modify membrane plasticity of intracortical neurons, such that Hebbian plasticity (ie, “neurons that fire together wire together”) is facilitated.⁴⁵  Thus, when combined with an appropriate task, such as a rehabilitative exercise (eg, strengthening, gait training), acquisition of the task may be faster and retention may be improved through faster and more durable synaptic plasticity, reflecting a degree of improved sensorimotor integration.

Among these tools, tDCS has been implemented in musculoskeletal rehabilitation among patients with ankle sprain, chronic ankle instability, and ACL injury. The use of tDCS is new in the field of sports medicine, with most prior clinical implementation of this technology used in psychiatric disorders and in those with neurological impairment (eg, Parkinson disease and stroke). In patients with Parkinson disease and stroke, tDCS has been found to improve neural excitability and motor function, particularly when paired with physical therapy.^46,47  Recently, research has emerged using tDCS in the treatment of lateral ankle sprains and chronic ankle instability and in patients following ACL reconstruction, offering promising results.

Transcranial direct current stimulation uses the application of a direct current (0.5 to 2 mA, up to 4 mA) to initiate physiologic changes, similar in technology to iontophoresis. Electrical current is administered through sponge-like electrodes (anode and cathode) that are soaked with saline and placed over the patient’s scalp. The current is then delivered between 5 and 30 minutes, with participants receiving additional treatment (eg, other modalities, performing therapeutic exercise) during and/or following the stimulation.⁴⁸  Transcranial direct current stimulation is able to modify cortical excitability for as long as 1 hour following the end of stimulation.

The primary goal of tDCS, as well as other forms of brain stimulation, is to modify brain plasticity such that motor learning and the synaptic connections associated with that learning are facilitated (or inhibited) throughout therapeutic efforts. Although tDCS may enhance traditional clinical outcome measures (eg, manage pain, improve muscle activation), its use will also modify synaptic plasticity over both individual and repeated sessions. The effects of tDCS are largely dependent on the placement of electrodes in terms of location and anode and cathode orientation. Anodal tDCS is typically used to facilitate synaptic plasticity, making areas more likely to improve activation, whereas cathodal tDCS is used to inhibit synaptic plasticity. Cathodal tDCS is uncommon in motor applications but is sometimes seen in pain-related research, with the cathode placed over frontal or somatosensory areas to inhibit pain processing.⁴⁹  More commonly, the faciliatory effects of anodal tDCS are used to generate improvements in motor execution and motor planning. To improve motor execution, an anode may be placed over the primary motor cortex, whereas improvement in motor planning may occur with anode placement over the dorsolateral prefrontal cortex.⁵⁰  These techniques can be combined in bihemispheric montages that can upregulate one region while downregulating the other (eg, anode over the impaired motor cortex, cathode over the unimpaired motor cortex; Figure 3).⁵⁰ 

To date, 5 published studies have described the effects of tDCS in individuals with musculoskeletal injury, with 3 addressing individuals with ankle injury and 2 performed in a subset of patients following ACL injury (Table 1).^51–55  Bruce et al and Ma et al both used a 4-week intervention of anodal tDCS over the motor cortex paired with a motor task (eccentric exercise or short-foot exercise) in patients with chronic ankle instability and compared the results with those from a group receiving sham stimulation.^51,52  Collectively, these studies observed improvements in balance, proprioception, neural excitability, and patient-reported function in the tDCS group beyond that of the sham group, cementing the link between cortical changes and patient functional status. A third study used a different electrode montage aiming to cathodally stimulate the somatosensory cortex, succeeding in decreasing pain and improving range of motion in patients with subacute ankle sprains compared with a sham group over 5 days.⁵³  Rush et al investigated the effects of anodal tDCS over the motor cortex on quadriceps function after ACL rehabilitation; however, the single-session intervention integrated with a low cortical demand task of walking did not succeed in generating changes.⁵⁴  Last, Tohiridad et al demonstrated improved postural control and lower extremity power in ACL-deficient patients following a 2-week intervention and at 1-month follow-up.⁵⁵ 

Noninvasive brain stimulation-based technologies and tDCS present a high degree of potential for improving function in patients with musculoskeletal injury given their ability to enhance integratory changes in conjunction with motor disinhibitory and sensory-targeted interventions while being commercially available, relatively inexpensive, and reusable. However, there are several barriers and precautions before they can be broadly integrated in rehabilitative practice. First and foremost, the availability of tDCS stimulators is largely tied to general wellness uses (eg, decreasing fatigue, modifying attention) that limit manufacturers’ obligations to meet Food and Drug Administration medical device regulatory standards. Importantly, the Food and Drug Administration does not regulate the practice of medicine, as that is the role of state and federal professional and licensing boards. Therefore, Food and Drug Administration approval (or lack thereof) for using tDCS for a condition does not restrict clinician judgement regarding application. As such, similar to other emerging practices (eg, dry needling, high-level laser therapy), clinicians should consider appropriate statutes governing their practice as an athletic trainer regarding the potential therapeutic use of stimulators to perform tDCS when treating joint injuries or other disorders.⁵⁶  Further, there are several contraindications and precautions for these devices, including risk in patients susceptible to seizures and the potential for adverse dermatological effects, with those latter precautions being similar to those clinicians implement for other direct current-based interventions (eg, iontophoresis).⁵⁷  Last, there is debate regarding optimal targets for these interventions, with questions existing as to what areas of the brain should be targeted (eg, motor cortex versus prefrontal areas) and the ability to target the deeper lower extremity representation within the cortex.⁵⁰  These barriers include questions regarding optimal electrode placement and a history of research in limited populations that may not translate to musculoskeletal patients.

To this point, we have postulated that the correction of maladaptive neuroplasticity is necessary to optimize clinical outcomes and improve patient function in a sustained manner and that this may be achieved through a combination of interventions addressing sensory and motor impairments in conjunction with integrative interventions. There is one additional key consideration in restoring function among these patients. Neuroplasticity has also been correlated with kinesiophobic changes that may impact an individual’s readiness to return to play and partake in fear-avoidance behaviors that further limit patient function.^58–60  The multifactorial relationship between kinesiophobia, decreased physical performance, and neuroplasticity is new and exists in a chicken-or-the-egg paradox. Is an individual fearful of activity because of an innate awareness of their physical deficits, or have they adopted maladaptive activation and movement strategies in response to this perception of fear? Although cause and effect may not be clear, the relationship between kinesiophobia and potentially maladaptive neuroplastic changes has been demonstrated in both injured and healthy individuals.^58–60  Therefore, strategically addressing cognitive and affective factors such as kinesiophobia and fear of reinjury is an important concomitant treatment when considering neural-targeted rehabilitation strategies.⁶¹ 

Multiple strategies have been proposed to address adverse psychosocial responses to musculoskeletal injuries (Table 2). Specifically, the dynamic biopsychosocial model has been proposed as one lens through which to consider these responses and develop therapeutic options.^62,63  Behavioral interventions, including imagery, guided relaxation, and mindfulness, have been proposed as potential interventions to address cognitive and affective responses to injury.^62,64  Alternately, goal setting and reprioritization and readjustment may also be relevant strategies for creating realistic, patient-centered expectations and avoiding negative affective responses when unrealistic or disconnected expectations are not achieved.⁶⁵  Graded exposure therapy and thought stoppage techniques can also be incorporated as behavioral therapies to enhance musculoskeletal outcomes.^64,66  Guided imagery and virtual reality specifically have been shown to decrease kinesiophobia and improve clinical outcomes when combined with a therapeutic exercise intervention, and exposure therapy has demonstrated sustained changes in neural activation, specifically to fear response-associated areas such as the amygdala and limbic regions, among individuals with phobias.^67–69  For example, among patients who underwent ACL reconstruction, Lebanon et al observed significantly greater increases in vastus medialis activation following 12 sessions of kinesthetic mental imagery in which participants completed 2 sets of 10, 10-second imaged repetitions of maximal isometric quadriceps muscle contractions in addition to traditional rehabilitation than a control group that did not complete imagery exercises.⁷⁰  Also in patients following ACL reconstruction, Cupal and Brewer observed higher knee extensor limb symmetry index, lower pain, and lower reinjury anxiety in individuals who, in addition to standard rehabilitation, participated in a 10-session relaxation and guided imagery intervention where each imagery session centered on current rehabilitation goals (eg, visualization of physiologic healing during session 1 and peak physical performance during session 10) than both the control and placebo groups.⁶⁷  The use of multimodal interventions that incorporate both physical and cognitive interventions to address kinesiophobia and injury-related fear has been endorsed by 2 recent systematic reviews and has been proposed as a strategy for inducing adaptive neuroplasticity in the corticolimbic regions.^58,71,72 

Although the direct relationship between affect-based interventions and the correction of maladaptive neuroplasticity is limited and warrants subsequent investigation, the relationship between self-reported psychosocial outcomes and self-reported function and performance outcomes, including return to activity, is well established.⁷³  Based on the relationship between self-reported function and kinesiophobia, it has been proposed that one intervention for kinesiophobia may be serial evaluation of self-perceived function via validated and reliable patient-reported outcome measures, followed by addressing of self-identified functional deficits throughout rehabilitation.⁷⁴  Similarly, alterations in psychosocial well-being can be evaluated via domain-specific outcome measures such as the Tampa Scale for Kinesiophobia or the Fear Avoidance Beliefs Questionnaire or disease-specific measures such as the ACL Return-to-Sport after Injury scale.⁷⁵  However, it is critical to pair the use of such measures with direct patient follow-up and discussion to both bolster therapeutic alliance and address potential measure limitations based on the limited populations in which they have been developed and validated. Although these tools can serve as a measurement similar in part to how range of motion and strength are continually evaluated, the use of these over time can be used to not only adjust treatment strategies but also improve patient confidence and increase the probability of successful return to activity.

Within the context of neural-targeted rehabilitation, identifying and addressing cognitive and affective impairments is arguably the least investigated, particularly in concert with motor, sensory, and integrative treatment options. However, considering the potential interventions to address this domain, their use is potentially the easiest to incorporate with the smallest barrier of required equipment or expertise.

The evidence suggesting that maladaptive changes in the brain negatively affect patients with joint injury and contribute to persistent reinjury has only continued to grow over the past decade. Given the improved physiological understanding of how these changes progress, it is time for clinicians to critically consider the manner in which their current therapeutic efforts may positively or negatively be affecting patients’ long-term function. Based on the currently available evidence in the fast-growing subfield of sports medicine emphasizing neuromodulatory rehabilitation, we recommend a multilevel approach toward correcting these neuroplastic changes that incorporates sensory, motor, integrative, and affective components in an intentional manner. Many of these strategies are readily available and affordable to incorporate into clinical practice. Sensory and motor strategies such as joint cooling, transcutaneous electrical nerve stimulation, or neuromuscular electrical stimulation have been proposed for several decades in the treatment of ACL injuries and ankle sprains to reduce AMI.^31,34,36  More recently, affective strategies such as imagery have demonstrated positive effects on muscle activation and more broadly on the kinesiophobia that has been strongly linked to poor clinical outcomes.⁷⁰  To best leverage the positive benefits of these established interventions, which each may alter some component of neural function, an integrative strategy should be used. Although motor learning approaches, such as implicit feedback, are currently the most clinically accessible integrative strategies, the body of evidence supporting the use of noninvasive brain stimulation following joint injury and other conditions associated with maladaptive neuroplasticity is rapidly growing. It is theorized that integrative approaches may be essential to transferring the functional success observed in clinical settings with current impairment-based rehabilitation approaches to the real world, ultimately resulting in higher rates of return to activity and lower rates of reinjury by increasing neural efficiency and reducing neural inhibition. Future directions for research in this area should continue to concurrently assess both patient-reported and neural function following joint injury, allowing for the establishment of a direct link between neural function and patient-reported function. Additionally, these recommendations are based around ankle and knee ligament injury, leaving much to be investigated in other models of injury, including those affecting the spine and upper extremity.