Brachial plexus injury and resting-state fMRI: Need for consensus
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.263178
Source of Support: None, Conflict of Interest: None
Keywords: Brachial plexus, functional imaging, functional MRI, injury
Brachial plexus injury is associated with significant morbidity as a result of the loss of function of the upper limb. In adults, the most common cause is road traffic accident affecting mostly young males who are in their most productive stage of life., Obstetric brachial plexus injury is a very rare type of injury seen in 0.5 − 2.0 births per 1000 live births. Most of the birth brachial plexus injury cases are due to the obstructed delivery  (shoulder dystocia, forceps delivery, etc.). Despite major advances in microsurgical techniques, tissue glues, autogenic, allogeneic, and synthetic grafts, surgical outcomes after nerve repair are nothing to boast off. There has been a constant endeavor to improve these outcomes.
Cortical plasticity has been shown to play an important role in the recovery following central nervous system damage. However, the understanding of its role in the repair of the peripheral nervous system following injury is in its infancy. A better understanding of cortical plasticity following peripheral nerve injury may provide further insight into nerve regeneration mechanisms which may help in improving outcomes after nerve injuries. Functional MRI (fMRI) [task based] has been used to assess for the development of cortical plasticity following nerve injuries. Over the past two decades, a new technique, the resting-state fMRI, has emerged as an important tool in studying various diseases of the brain. Resting-state functional magnetic resonance imaging (rs-fMRI) is a technique to measure oscillations in the blood oxygen level dependent (BOLD) signal across time, under nonarousal (resting) states. The temporal correlation of these BOLD signal fluctuations across various brain regions represents a resting-state network (RSN). Intranetwork functional connectivity (connectivity among various regions within an RSN) and internetwork functional connectivity (interaction between various RSNs) provide insights into the functional topography of healthy and diseased brain.
In recent years, there has been an increase in the use of resting-state fMRI for looking into how exactly the brain responds to such a type of injury. Resting-state scan gives a picture of various RSNs present in the brain (Biswal et al., 1992). Even when the brain is in the state of rest, it is not actually at rest and various regions in the brain are still activated. Resting-state fMRI picks the BOLD signals from these areas and helps in the quantification of the particular network. As discovered by Raichle et al., the brain is continuously in an active state with a high level of activity even when the person is not engaged in focused mental work.
In the cases of brachial plexus injury, this technique is particularly useful as patients are not capable of performing tasks because of the paralysed limb; hence, the resting-state scan is helpful to look at the changes in various RSNs. This review focuses on the literature available on resting-state fMRI and its use for the management of traumatic and birth brachial plexus injury.
Inclusion criteria: It includes all those studies which have used resting-state fMRI for interpreting cortical reorganization in patients with traumatic or birth brachial plexus injury.
Exclusion criteria: It includes studies which have not used resting-state fMRI. Some of the studies included in this review have used task-based or motor function-related paradigms; these studies were included only if a resting-state fMRI was performed in addition to a task-based fMRI.
The studies which used only task-based fMRI, transcranial magnetic stimulation studies, other neuro-modulation techniques, and animal studies were all excluded.
Studies were identified by searching PubMed using the terms 'brachial plexus', 'resting-state fMRI', 'cortical reorganization', 'pediatric', 'birth', 'adult', and 'traumatic'. The reference list of the relevant studies was also reviewed according to the suitability. The search was limited to articles published in English.
Data collected from the studies included the number of patients/controls, the type of intervention performed, the side of injury, and general information in descriptive terms. Each study was classified according to the type of intervention and the time at which resting-state fMRI scan was performed, before or after surgery. The studies were further evaluated based on the study design.
The search strategy yielded 10 suitable papers which were then analyzed. Two studies were prospective. All other studies were retrospective. The natural history of birth brachial plexus injury was the subject of 2 of these papers [2 and 7, [Table 1]]. Surgical intervention had been carried out in 7 papers [1, 3, 5, 6, 8, 9 and 10, [Table 1]].
Surgical intervention and resting-state fMRI in brachial plexus injury
Out of all the studies involving any surgical intervention and resting-state scan, the study carried by Bhat et al., had the maximum number of patients (35 cases) with brachial plexus injury. The study was divided into two groups based on the improvement of muscle power. Group 1 with 14 patients (postoperatively) had improvement in muscle power, whereas Group 2 with 16 patients (postoperatively) had no improvement. The average time between injury and surgery was 6 months. In total, 5 cases (postoperatively) were not included because of the longer duration between surgery and the MRI. All 35 cases underwent preoperative fMRI scan as well. The resting-state scan before surgical intervention revealed a lower connectivity in the sensorimotor network (SMN) and salience network (SN), whereas an increased connectivity was seen in the default mode network (DMN). The resting scan after the surgery revealed an increased connectivity of the left supplementary motor cortex in SMN and the medial frontal gyrus in SN in patients of Group 1. No changes were seen in patients of Group 2. Although there was no direct impact on the brain, still cortical changes were seen after the surgery, which points out to the reorganization that occurs in the RSNs.
The side of injury was the central theme of the study done by Yun-Dong Shen et al. The authors used the amplitude of low-frequency fluctuations for comparing if there was any cortical reorganization due to the side of the injury. The authors reported that cortical reorganization is definite and depends on the side of injury as well as the handedness of the patient.
Involvement of the primary motor areas (M1) in brachial plexus injury was observed by Vargas et al. The authors assumed that the correlation between far away voxels would lose strength in the primary motor cortex M1. The results were based on the assumption that the correlation values were higher for closer voxels and lower for far away voxels. This essentially means that, within the primary motor cortex, the functional connectivity is reduced due to the injury.
Use of task-based fMRI to derive resting-state functional connectivity was used by Liu et al. and Qiu et al. Resting-state fMRI was performed on brachial plexus patients and task-based fMRI was carried out on healthy controls, respectively. In total, 4 regions of interest (ROI), 2 each in the primary motor area (M1) and supplementary motor areas (SMAs) were extracted from scans performed on healthy controls (task based). For the study mentioned by Qiu et al., SMAs were extracted from the task-based scan from controls. These ROIs were then used to check the interhemispheric functional connectivity across brachial plexus patients. The results revealed reduced functional connectivity between two M1 areas present in each hemisphere. No correlation was seen in SMA areas. The connectivity in the brachial plexus injury group was reduced. Decreased voxel wise FC between the SMAs and multiple brain regions that are closely associated with information integration or motor processing were seen in the brachial plexus injury group.
The study of brachial plexus and resting-state fMRI has forced researchers to come up with new ways to see the functional connectivity in brachial plexus patients, which Wang et al., have achieved by using the graph-based analysis of resting-state fMRI. The focus of the paper is relatively based on brain networks rather than focusing on functional reorganisation. The results showed that there are no global cerebral resting-state functional network changes after nerve repair. The authors, however, were able to see the effectiveness of small world characteristics which incorporated the local activity within the brain. The authors argued that since the sample size is very less, it would be interesting to see how global cerebral RSN responds to brachial plexus injury. The authors of this study have used the technique of correlations between various networks based on the shortest path and clustering. They were able to observe the changes in small world environments such as nearby networks, but overall global resting-state activity remained nonsignificant.
The study carried out by Yu et al., had an average interval of 18 − 48 months between surgery and resting-state fMRI, making it the only study with the longest time gap [Table 2]. In total, 12 patients were included in the study and they underwent contralateral C7 transfer (CC7) before the resting-state scan was done. Seed-based approach was used to define the regions of interest (ROIs) for hand motor regions. Interhemispheric functional connectivity in the primary motor cortex, M1 areas, was increased in patients with brachial plexus injury. As discussed earlier in one of the study [Table 1], the interhemispheric functional connectivity decreases in M1 areas in brachial plexus injury, but the duration between the surgery and resting-state fMRI scan was not mentioned in either of the studies.
No intervention and resting-state fMRI in brachial plexus injury
Studies involving no surgical intervention have focused on SMN and its activity in response to the injury. Kislay et al., included 14 cases of neonatal brachial plexus injury (NBPP). A decreased activation in the SMN was seen. Chen et al., reported that cortical maps of sensorimotor motor areas associated with hand and arm function contralaterally had a weaker correlation as compared to the ipsilateral side. However, Anguelova et al., reported no changes in resting-state activation in 16 NBPP adults as compared to normal controls, highlighting the fact that changes in RSN are a dynamic process, and patients included in studies involving resting-state fMRI need to be followed up. Thus, these studies need to be of a long term duration in nature.
Resting-state fMRI is a powerful tool which can provide some insight into the mechanism of cortical plasticity and repair following peripheral nerve injury. RSNs are a network of areas in different regions of the brain which have been associated with certain functions. They are picked up on fMRI when the subject is scanned at rest, with the patient's eyes closed and his/her mind allowed to wander. Various RSNs have been identified. Most of them are involved in cognition, memory, behavior, and salience identification. RSNs such as DMN, SMN, fronto-parietal network, and right attention network are most frequently studied networks in the case of brachial plexus injury.
We conducted this review to look at all the outcomes of resting-state fMRI in brachial plexus injury. There is conclusive evidence to suggest that patients affected by brachial plexus injury tend to undergo cortical reorganization. There is reduced functional connectivity in various RSNs and interhemispheric communications. Changes are seen following surgical intervention, and differences are seen between patients who have improved after surgery and those who have not. All studies published so far have focused on the RSNs at a given point in time (after or before surgery). Longitudinal studies involving multiple resting-state MRIs over a long time period are lacking. These studies may help in studying the progressive changes and reorganizations occurrring in the brain after injury and repair. One of the studies did use a new approach of studying the changes in RSN instead of looking into cortical reorganisation, but the limited number of patients included in the study cannot help in generalising the result to a larger population [Table 1].
The timing of performing the rest-fMRI may be crucial in picking up vital information, as RSNs are dynamic and may change with time. Unfortunately, in the few studies published in literature, the authors are far from being anywhere near to a commonly agreed-upon timing. Some studies have patients with NBPP, and the first resting-state scan was performed within 2 years to as long as 33 years. The authors of the study found no changes in RSNs. The focus shifts to the fact that the resting-state scan was performed much later in the age; and, if there were any changes in the RSNs, the authors had no way to determine this as a result of partial or complete recovery. The time duration is very critical in the studies like these.
In our opinion, focussing on the timing in the resting network rather than changes in them will provide more information about the progress of cortical reorganization in the cases of BPI. This, in turn, will help in the timing of intervention in order to optimize outcome and prognosis., The current literature is focused on reporting changes in different RSNs in response to the injury and the cortical reorganisation that follows, as affected limb loses the corresponding interpretation areas in the brain. We suggest that longitudinal studies be performed so that the effects of cortical reorganisation are studied in more details.
The current literature has enough evidence to prove, based on fMRI studies, that cortical reorganisation occurs in brachial plexus injury. It is interesting to note that RSNs usually linked to cognitive and higher mental functions are altered following a functionally distant injury and recovery.
Resting fMRI is an essential tool in understanding cortical reorganization not only in central nervous system diseases but also in peripheral nerve injuries. Whether the changes are a result of injury and recovery, or whether they play a role in the recovery of functions is not clear. This may be answered by the conduction of longitudinal studies focusing on resting fMRI, carried out over several years, right from the time of injury to years following partial or complete recovery.
Financial support and sponsorship
There is no financial support for this paper. A longitudinal study has been initiated at NIMHANS with 2nd and 4th authors as the principal investigator (PI) and co-PI, respectively, with support from Department of Science and Technology (DST).
Conflicts of interest
There are no conflicts of interest.
[Table 1], [Table 2]