Elsevier

Life Sciences

Volume 288, 1 January 2022, 120161
Life Sciences

Review article
From symmetry to chaos and back: Understanding and imaging the mechanisms of neural repair after stroke

https://doi.org/10.1016/j.lfs.2021.120161Get rights and content

Abstract

Neuroscience has made strides in recent years allowing us insight into the workings of the brain – from the molecular to the regional anatomy. These insights have given researchers an advantage in seeking novel therapies for neurological disorders, specifically stroke. Yet despite these discoveries, many aspects of stroke remain poorly understood – specifically post-stroke recovery. This review article seeks to outline cutting-edge neuroimaging technologies, and the current level of understanding of neurological repair after stroke, with the main focus on the mechanism of axonal sprouting. Neuronal connectivity has varying levels of complexity that allow neuronal networks to process information and give rise to our day-to-day functioning. As stroke causes the death of groups of regional neurons, it is likely that the reestablishment of function seen in some stroke patients is related to shifting patterns of functional connectivity. This paper touches on the timeline and limits on the amount of functional recovery, as well as the differences in organization of neuronal networks in a healthy versus post stroke brain. Finally, we discuss how the previously mentioned methods of imaging are critical in understanding the mechanisms of functional recovery. The mechanism of axonal sprouting and its theorized different types are explained, along with potential ways of imaging them in rodents. The hope is that, with a better understanding of the mechanisms underlying brain recovery, researchers can apply this knowledge to better help stroke patients and be of use in clinical settings.

Introduction

Stroke is the leading cause of death and disability in the Western world, with a mortality rate even higher than cancer [1]. In 2017, there were 6.2 million fatal cases of stroke and 132.1 million cumulative years lost due to disability after stroke [2]. This is a significant portion of the population that is left with detrimental long-term health issues due to stroke. Clearly, stroke merits significant research in all regards – from pathophysiology to treatment – and one of the most promising areas of inquiry is in post-stroke neural repair. This review article attempts to shed light on the still unanswered question of how and why certain patients gain functional recovery following stroke while others do not.

Strokes can be subdivided into two main types: ischemic and hemorrhagic. In ischemic stroke, a blood vessel, almost always an artery, becomes blocked and impairs blood flow to the brain causing the cells to die within minutes. Ischemic strokes are either thrombotic strokes, in which the blood clot develops in the blood vessels in the brain, or embolic strokes, in which a blood clot forms in another part of the body (usually the heart) that breaks loose and travels to the brain. Hemorrhagic strokes, on the other hand, are caused by sudden bleeding into the brain parenchyma when an artery that supplies the brain ruptures. The leaked blood causes the brain to swell and irritates the delicate neural tissue, putting pressure on it that can cause further damage. Hemorrhagic strokes are further divided into two types as well; intracerebral - from the vessels in the brain, and subarachnoid- bleeding between the brain and the meninges in the subarachnoid space [3].

Since the majority of functional recovery studies and mechanisms revolve around rehabilitation after ischemic stroke, this subset will be the focus of this paper. Furthermore, because ischemic strokes make up approximately 87% of all strokes [3], they are most commonly encountered in a clinical setting. Historically, stroke treatment was preventative and at best palliative. However, the t-PA studies of the mid 90s excitingly allowed for a subset of patients who met the strict clinical parameters to receive brain-saving treatment [4]. In these studies, researchers investigated the efficacy of intravenous recombinant tissue plasminogen activator (t-PA) in ischemic stroke through a two part double blind trial. The researchers observed a greater improvement in function in the group treated with t-PA 3 h after onset of stroke than the control at three months post-stroke [5]. These studies showed that immediate treatment for acute ischemic stroke is critical because it minimizes the long term effects of a stroke – namely rapid, irreversible neuronal death – and prevents death and disability. The rapid restoration of blood flow to ischemic brain tissue brings much-needed oxygen to neurons, which rely almost entirely on oxidative phosphorylation to meet their high energy demands. Thus, the more time passes without treatment, the more neurons die, and the greater the chance is of paralysis and permanent disability [6].

Clinical interventions, such as speech, physical, and occupational therapy, despite not guaranteeing a full regain of function, have proven to be quite effective alternatives for assisting in recovery. For example, an important study [7] showed that extensive physical therapy helped patients who had a persistent disability recover motor function, even though they began this treatment a year or more after stroke. Researchers administered an intensive physical therapy program for 12 weeks to 39 participants which tested three modes of rehabilitation: motor learning, electrical stimulation, and robotics assisted rehabilitation. This raises the question: how is the brain repairing these lost connections? The large number of patients in this study that showed functional recovery necessitate that there is some form of neural reorganization underlying it. It is likely, therefore, that a better understanding of these mechanisms can assist health care providers in improving this recovery rate [7].

The question of spontaneous or elicited functional post-stroke recovery is inherently complex. If ischemic stroke impairs or destroys anatomically segregated clusters of neurons directly proximal to the occluded vessel, how can the function directly controlled by those neurons return in a subset of patients? The key to answering this question, and others like it, is to consider that clusters of neurons are spatially organized within the cortex and subcortex and, although responsible for governing somatic and cognitive functions within their micro-network, do not act in isolation. Rather than being independent islands, each responsible for doing its own job, these clusters of cells are structurally and functionally connected with each other, and throughout the brain, thus giving rise to intricate, complex networks of cells. We contend that brain connectivity across multiple regions and varying levels of complexity of scale allows neuronal networks to process information and give rise to baseline brain function. Thus, it is likely that shifting patterns of neuronal functional connectivity – that is, changes in the network – allow for the regain of function seen in some stroke patients who recover what they had lost.

Functional connectivity is the key term here, and is defined as the temporal dependency of neuronal activation patterns between anatomically separated brain regions [8]. This web of interconnectivity between brain regions that exhibit correlated patterns of synchronous activity across time, allows us to consider the brain and its anatomy as a network. To begin with, any network consists of two components: the structural regions, or nodes, and the links that connect them. This notion of structural islands and the bridges connecting them is certainly not new to the field of mathematics. The mathematical network model approach to science was born from a thought experiment conducted by Leonhard Euler which subsequently produced a field known as graph theory. Graph theory is, at its core, about mathematically modelling pairwise relationships between objects. As such, it is fruitful for neuroscience – especially as we shift from paradigms that look at isolated brain regions, to whole brain network connectivity. Thus, by applying graph theory, networks can be composed of nodes and edges, where, in the brain, the nodes represent neurons or anatomic brain regions, and the edges are the axonal connections between them. Thus, in the area of neuronal architecture, any network can be defined as structurally distinct regions of the brain that exhibit activation patterns that are temporally correlated. These activation patterns can be measured by using an fMRI scan, and do necessarily have direct connections that can be seen by the naked eye [9]. We therefore contend that understanding how the brain functions as an interconnected network of neuronal circuits is essential to studying and stimulating repair after injury [10]. Furthermore, after stroke, we argue that structural damage that impairs specific nodes within a network may affect all the nodes within that network. Additionally, nodes outside of the affected network are not affected, although they may step in to take over functioning from the affected nodes, which may be a mechanism of post-stroke recovery.

In exploring these ideas further, we begin with a John Hopkins study by Zeiler et al., in which researchers explored these mechanisms of connectivity by teaching mice a special way to get their food involving the primary motor cortex [11]. The mice were trained to perform a skilled prehension task to an asymptotic level of performance, which involved the mouse reaching its forelimb through the slit created in the cage, grabbing the food pellet and eating it without knocking it off its resting space, dropping it, or losing control of the pellet. The group then induced mild strokes which damaged the primary motor cortex. As expected, the mice could no longer reach their food in the same way they were able to prior to the stroke. Mice were then subjected to a week of retraining which began 48 h after the stroke, and led to the mice performing the task just as well as they did pre-stroke. Surprisingly, the authors found that the primary motor cortex had not yet recovered; rather it was the medial premotor cortex that took over lost function. Once strokes were induced in the medial premotor cortex, that function was also lost. Once again, however, the mice were able to be retrained and though it was not clear which part of the brain took over, the function was recovered. Interestingly, the control group were trained to perform this special reach for their food and were given a stroke in the medial premotor cortex without any prior stroke, and were not affected. In other words, the medial premotor stroke did not impair their ability to perform the task, which suggested that in healthy animals, this region of the brain is not responsible for such activity but can take over when needed from the affected primary motor cortex.

Cao et al. explored this same phenomenon in humans. A set of six patients were instructed to do a task with each of their hands while being scanned by fMRI. The patients ranged from the age of 7 to 22 and had all experienced unilateral brain damage in the perinatal period. When the patients were scanned while using their unaffected hand, the fMRI only showed activity on the opposite hemisphere, however, when using the affected hand, researchers saw that both hemispheres showed heightened activity in the scan [12].

While these results from Zeiler and Cao were exciting, motor recovery in humans needed to be further explored. The most accepted measure for recovery of body function and structure is the change in the Fugl-Meyer Assessment of upper extremity (UE-FM) scores [13], as well as other measures more focused on activities, such as the Chedoke Arm and Hand Activity Inventory (CAHAI) which look at outcome measures such as improvement of impaired limbs. Motor recovery after stroke can be unpredictable and recovery seems to level off after a few months [14], which is why it is widely believed that there exists a “critical window” for functional recovery, in which a patient is more responsive to treatment. It is hypothesized that this window is within the first 3 to 6 months poststroke, where there is likely to be heightened plasticity; however, the range of this time period is still unclear [15], [16].

Thus, Ballester et al. [17] attempted to determine if there is in fact an upper limit to the possible amount of recovery. The dynamics of post-stroke recovery and the sensitivity to treatment were investigated from 11 rehabilitation pilot studies, and the researchers observed that there is improvement in function over all stages post-stroke. They hypothesized that a specific gradient of recovery faded out exponentially and reached asymptotic levels after one year and a half post-stroke. To test this hypothesis, the group examined the temporal structure of motor recovery using individual patient data from a homogeneous sample of 209 individuals. The patients all had middle cerebral artery ischemic strokes or intracerebral hemorrhagic strokes in which mild to moderate upper limb hemiparesis resulted and patients were all between ages 45 and 85, and had no significant cognitive impairment. The data were divided into 17 sets based on specific characteristics. The primary outcome considered in their analysis was the UE motor impairment and activity after therapy, as measured by the standardized clinical scales: UE-FM and CAHAI scales [17]. Based on these, significant gains in body function and structure were seen after treatment in all patients. At the chronic and late chronic stage, the therapy showed overall effectiveness in facilitating improvements. There was a dependency between the number of days post-stroke before the start of the RGS therapy (a virtual reality based training protocol using a Rehabilitation Gaming System), and the improvements in motor function. The subacute group that trained with RGS showed a significant improvement during the 3 month period at the end of treatment, while the acute groups showed interindividual variability and nonsignificant gains. Only the early chronic group (6–18 months) who used RGS showed higher recovery rates than the late chronic group (>18 months). This reveals a long-lasting gradient of sensitivity to treatment that remains visible across the first 18 months post-stroke. This effect was not present in the follow-up measures, a period in which no therapy was administered. The researchers detected a smooth decrease in the sensitivity to treatment which forms the critical window for recovery that extends beyond 12 months post-stroke. What this means, is that post-stroke recovery occurs rapidly early on and then fades over time. This shows the need to provide both early rehabilitation therapy and therapy in the chronic stages post stroke as recovery never ceases completely; that is, there is always a possibility of improvement. This is the first time that such an extended critical period of recovery is reported, and it needs further research. While this shows that the decrease in recovery occurs much more gradually than initially thought, most professionals believe earlier is still better. A recent study by Dromerick et al. emphasizes that there is an optimal time in which recovery is greatest [18]. In this study, elements of animal motor training paradigms were applied to humans, which also served to validate comparisons made between animal and human studies. Patients underwent task-specific motor therapy; the acute group started at less than 30 days, the subacute group at 2–3 months post-stroke, and the chronic group at over 6 months. Results showed significant improvement in the subacute group, decent improvement in the acute group, and no significant improvement in the chronic group. The researchers observed greater recovery in patients that received intensive motor therapy in the first 60–90 days poststroke. This provides strong evidence for a critical period in human brain recovery [18]. While Zeiler's findings provided the where of functional recovery, these studies expanded on those findings through a clinical applied study with stroke patients, helping us understand when functional recovery takes place. A picture now begins to come together of the complexity of functional recovery, the next missing piece being how this process unfolds.

Section snippets

Neuroimaging fundamentals

It is important to have a basic understanding of the various methods of neuroimaging, in order to develop our understanding of neural repair and reorganization. As well as its benefit in treatment, and the improvement of patient outcome, neuroimaging can provide insight into the more subtle functions of the brain, and the relationships between different regions. In the moments immediately following a stroke, neuroimaging can provide valuable information such as the lesion location, structural

Mechanisms of functional recovery

Having seen the usefulness of a graph theory approach to brain recovery, and having explored some of the imaging modalities that may allow researchers and clinicians to see network changes in real time as the brain recovers, we now turn our attention to possible mechanisms of recovery.

The focus of many recent studies has been on the ways in which the brain changes its structure during recovery. For example, Wang et al. [29] in their analysis of the brain after stroke show that connectivity was

Imaging axonal sprouting

Spotting clear evidence of axonal sprouting has proven difficult thus far. There is no truly unique pattern of connections after stroke that can be identified easily or be obviously seen as a characteristic or biomarker of axonal sprouting. For example, it is not clear that the immunohistochemical staining of an axonal protein after stroke means there are more axons- “…it may be protein half-life, axonal transport, epitope presentation or other aspects of immunohistochemical staining.” [42] The

Conclusion

This paper provides an overview of the most recent advances in the field of neuroimaging and neurological repair after stroke. Various imaging technologies were discussed, such as MRI and its various subsets. In regards to stroke recovery, DTI has proven extremely effective in determining nerve fibre tracks and neural connections, but further research and development is required when multiple nerve fibres overlap. Resting state fMRI is of great use in locating and assessing the initial infarct

CRediT authorship contribution statement

Caroline Alionte: Investigation, Writing- Original Draft; Writing – Review & Editing Christian Notte: Investigation, Writing – Original Draft; Writing – Review & Editing Christos D. Strubakos Conceptualization, Writing – Original Draft, Writing – Review & Editing, Supervision.

References (61)

  • R.V. Krishnamurthi et al.

    Global, regional and country-specific burden of ischaemic stroke, intracerebral haemorrhage and subarachnoid haemorrhage: a systematic analysis of the global burden of disease study 2017

    Neuroepidemiology

    (2020)
  • J.W. Chung et al.

    Trial of ORG 10172 in acute stroke treatment (toast) classification and vascular territory of ischemic stroke lesions diagnosed by diffusion-weighted imaging

    J. Am. Heart Assoc.

    (2014)
  • C.D. Strubakos et al.

    Non-invasive treatment with near-infrared light: a novel mechanisms-based strategy that evokes sustained reduction in brain injury after stroke

    J. Cereb. Blood Flow Metab.

    (2020)
  • Tissue plasminogen activator for acute ischemic stroke

    N. Engl. J. Med.

    (1995)
  • Y. Xian et al.

    The need for speed

    Circ. Cardiovasc. Qual. Outcomes

    (2018)
  • Svetlana Pundik et al.

    Recovery of post stroke proximal arm function, driven by complex neuroplastic bilateral brain activation patterns and predicted by baseline motor dysfunction severity

    Front. Hum. Neurosci.

    (2015)
  • M.P. Heuvel et al.

    Exploring the brain network: a review on resting-state fMRI functional connectivity

    Eur. Neuropsychopharmacol.

    (2010)
  • L. Meng et al.

    Brain network analysis and classification based on convolutional neural network

    Front. Comput. Neurosci.

    (2018)
  • S.R. Zeiler et al.

    Medial premotor cortex shows a reduction in inhibitory markers and mediates recovery in a mouse model of focal stroke

    Stroke

    (2013)
  • Y. Cao et al.

    Functional magnetic resonance studies of the reorganization of the human hand sensorimotor area after unilateral brain injury in the perinatal period

    Proc. Natl. Acad. Sci.

    (1994)
  • G. Kwakkel et al.

    Standardized measurement of sensorimotor recovery in stroke trials: consensus-based core recommendations from the stroke recovery and rehabilitation roundtable

    Int. J. Stroke

    (2017)
  • G. Kwakkel et al.

    Understanding the pattern of functional recovery after stroke: facts and theories

    Restor. Neurol. Neurosci.

    (2004)
  • B.R. Ballester et al.

    A critical time window for recovery extends beyond one-year post-stroke

    J. Neurophysiol.

    (2019)
  • A.W. Dromerick et al.

    Critical period after stroke study (CPASS): a phase II clinical trial testing an optimal time for motor recovery after stroke in humans

    Proc. Natl. Acad. Sci. U. S. A.

    (2021)
  • C. Sorg et al.

    Selective changes of resting-state networks in individuals at risk for Alzheimer's disease

    Proc. Natl. Acad. Sci.

    (2007)
  • A. Anand et al.

    Reciprocal effects of antidepressant treatment on activity and connectivity of the mood regulating circuit: an fMRI study

    J. Neuropsychiatry Clin. Neurosci.

    (2007)
  • M. Liang et al.

    Widespread functional disconnectivity in schizophrenia with resting-state functional magnetic resonance imaging

    NeuroReport

    (2006)
  • Y. Lv et al.

    Identifying the perfusion deficit in acute stroke with resting-state functional magnetic resonance imaging

    Ann. Neurol.

    (2013)
  • I.R. König et al.

    Predicting long-term outcome after acute ischemic stroke

    Stroke

    (2008)
  • W. Heiss et al.

    Imaging for prediction of functional outcome and assessment of recovery in ischemic stroke

    Stroke

    (2014)
  • Cited by (0)

    1

    Co-first authors.

    2

    Senior author.

    View full text