An assessment of co-contraction in reinnervated muscle
Abstract
Aim: To investigate co-contraction in reinnervated elbow flexor muscles following a nerve transfer. Materials & methods: 12 brachial plexus injury patients who received a nerve transfer to reanimate elbow flexion were included in this study. Surface electromyography (EMG) recordings were used to quantify co-contraction during sustained and repeated isometric contractions of reinnervated and contralateral uninjured elbow flexor muscles. Reuslts: For the first time, this study reveals reinnervated muscles demonstrated a trend toward higher co-contraction ratios when compared with uninjured muscle and this is correlated with an earlier onset of muscle fatigability. Conclusion: Measurements of co-contraction should be considered within muscular function assessments to help drive improvements in motor recovery therapies.
Peripheral nerve injuries often manifest following blunt or penetrating trauma with an estimated prevalence of 2% among the general trauma population [1]. Despite significant advancements in reconstructive nerve surgery, muscle reinnervation is often seen by patients as being incomplete resulting in permanent functional disabilities [2,3].
The Medical Research Council (MRC) grading system and continuous measurements of peak volitional force (pvf) are objective measures that are widely employed by clinicians to monitor the recovery of motor function [4–6]. However, these assessments have a number of limitations. The MRC grading system has been shown to be insensitive in monitoring the recovery of muscular function with over 96% of recordings documented as MRC grade 4 [6]. Some authors suggest that these measurements of efferent muscular function are poorly associated with the lived experience of muscle reinnervation [7]. Recent studies suggest that the impairment of afferent functions of muscle such as decreased proprioception, an earlier onset of muscular fatigue and muscle pain are central themes of reinnervated muscle reported by patients [5–8].
Proprioceptive nerve fibers together with their peripheral receptors relay data about muscle force as well as length [9]. They detect and transmit somatosensory information to the CNS providing awareness of body and limb position in space [10,11]. Through complex interactions with the CNS, these afferents are able to coordinate the activity of motor pools to agonist and antagonist muscles around a joint [9,12]. This physiologic process of co-contraction is essential to the execution of stable and purposeful movements [13–15]. However, co-contraction can become pathological when there is a disproportionate increase in muscular effort against the intended movement [16]. Animal models of traumatic injury to the primary sensory afferent system have shown that dysregulation of mechanosensory circuits coordinating the actions of agonist and anatagonist motor pools leads to pathological co-contraction [17]. This is thought to impair proprioception, exacerbate muscle fatigue and reduce pvf [18–20].
In the clinical setting, surface electromyography (sEMG) recordings from agonist and antagonist muscles are widely used to quantify co-contraction. Raw EMG signals are filtered and used to calculate the root mean square (RMS) for the agonist and antagonist muscles [16,21–23]. The value from the antagonist muscle is divided by the agonist to determine the co-contraction ratio [16,21–23]. This method has been used to reliably compare co-contraction in neuromuscular pathologies with healthy controls [16,21–23]. These studies found higher co-contraction ratios around joints affected by neuromuscular pathologies compared with healthy controls [24–27] which often correlate with functional measures of disease severity [25,28]. This suggests that assessment of co-contraction may be a useful objective measure in neuromuscular pathologies. However, exploration of co-contraction in human reinnervated muscle is not documented. Addressing this issue may represent the first step toward the development of a measure that demonstrates better correlation with the lived experience of muscle reinnervation (compared with MRC grading and measurement of pvf).
The reanimation of elbow flexion is a common challenge encountered by the reconstructive surgeon. A nerve transfer (using suitable functioning donor fascicles from the ulnar and median nerves to transfer into the denervated musculocutaneous nerve) is commonly carried out to restore elbow flexion [29–31]. This represents an ideal surgical scenario to investigate co-contraction in reinnervated muscle for a number of reasons. The elbow joint can be modelled as a uniaxial pivot joint. Its function is necessary to perform activities of daily living. In healthy motor systems, protocols to assess co-contraction around this joint have been well documented, yielding reliable results [13,32–35]. Sustained and repeated isometric contractions of the elbow flexor muscles have been widely used to elicit and study co-contraction and/or muscle fatigue around the elbow joint in a variety of pathologies [13,32–35]. To the authors' knowledge, co-contraction has not been quantified in reinnervated muscles. Therefore, this study used previously published protocols [13,32–35] to investigate co-contraction in reinnervated elbow flexor muscles following nerve transfer.
Materials & methods
This study received full ethical approval (Research Ethics Committee ref.: 16/LO/0623; Integrated Research Application System [IRAS] ID: 202847). The study was designed and reported according to the Strengthening the Reporting of Observational Studies in Epidemiology guidelines for cohort studies.
The patient cohort recruited was identical to previously published data by the authors where reinnervated muscular fatigability was quantified using sEMG and force dynamometry [8]. Identical protocols to those outlined here were used in previously published work by the authors to concomitantly interrogate muscle fatigability alongside co-contraction [8]. This allowed the present study to explore the link between these two important physiological processes in reinnervated muscle.
Patients who underwent Oberlin's nerve between May 2006 and May 2012 were identified through search of the institution database. The surgical procedure involved a longitudinal neurotomy along the course of donor ulnar and median nerve. A fascicle approximately one-eighth the size of the donor nerve was identified which demonstrated predominantly wrist flexor activity from flexor carpi ulnaris/flexor carpi radialis activity upon low amplitude neurophysiological stimulation [8]. Stimulation of other fascicles was performed to ensure active wrist flexion could be achieved even once donor nerves had been harvested. Fascicles that elicited intrinsic hand function were avoided [8].
Selection criteria
Patients had to be aged greater than 18 years, communicate in fluent English, verbally engage with the experimental protocol and be at a minimum of 24 months after surgery to allow a sufficient time interval to have elapsed such that sufficient motor function could be recovered to allow engagement with the study protocol [8].
Patients less than 24 months postoperation, those with cognitive disability, verbal communication difficulties and birth-related brachial plexus injuries were all excluded.
After application of these criteria, 68 patients were identified. Written invitation letters were sent to all of these patients followed up with a second letter in 1 month if no response to the first letter was received. 12 patients subsequently attended for review and gave fully informed consent to participate in the study [8].
Experimental protocol
Experiments were performed on the reinnervated arm with the uninjured contralateral arm used as a control [8].
Co-contraction was assessed by repeatability and sustained isometric contractions of the elbow flexors based on previously published protocols [4,5,36,37]:
For repeatability, patients were asked to performed two sets of three maximal contractions. Each set was separated by a 60 s break. The peak force was defined as the maximal force exerted across the six isometric contractions;
For sustainability, participants were instructed to perform a continuous maximal isometric contraction for 1 min. Force and sEMG signals were analyzed in 10 s blocks. The mean for each time block was calculated and compared with quantify changes in the parameters of interest over time.
A four-channel sEMG microprocessor with programmable gain amplifiers (Data Log MWX8 Biometrics Ltd, Newport, UK; 3-dB bandwidth, 10–500 HZ) was used to capture sEMG signals [8]. A sampling rate of 2048 samples per second was implemented for each channel. A Biometrics SX230 (Biometrics Ltd) bipolar sensory electrode was used with an intraelectrode interval of 20 mm [8].
Skin was cleaned with alcohol before electrode placement. All electrode placements were determined in accordance with international guidelines from Surface ElectroMyoGraphy for the Non-Invasive Assessment of Muscles (SENIAM) [38]. The placement zone for the biceps brachii was standardized as one-third along a line from the acromion to the center of the antecubital fossa in line with the long axis of the arm (Figure 1A) [39]. The placement zone for the triceps brachii was standardized at halfway along a line between the posterior acromion and the olecranon (Figure 1B) [39].
Image (A) indicates the placement of the recording electrode over the elbow flexor compartment (biceps brachialis), (B) placement of the recording electrode over the elbow extensor compartment (triceps brachii) and (C) indicates the positioning of the handheld dynamometry (HHD) for force measurements during the sustained and repeated models of fatigue assessment.
Reproduced with permission from [8] © The British Editorial Society of Bone and Joint Surgery.
Electrodes were adhered to the skin using adhesive tape. The position of the electrodes was marked on the skin. This was to ensure that should the adhesive have failed the electrodes could be repositioned in identical positions before becoming loose. A reference electrode was placed on the contralateral wrist.
Biometrics software was utilized to process the raw EMG data and calculate the RMS of EMG data and compute the area under the RMS curve (AOC sEMG) [8]. EMG data was retrieved from both flexor and extensor muscular compartments [8]. This was used to determine the co-contraction ratio (Equation 1). This methodology for the assessment of co-contraction has been well documented in other studies [15,24,40–42]. The 60 s sustained maximal contraction was divided into 6 × 10 s blocks and average within those blocks to assess change over time from one block to the others.
Equation 1 was used to calculate the co-contraction ratio. The AOC sEMG was calculated from the EMG signals from the antagonist muscle (triceps brachii) and agonist muscle (biceps brachii).
Statistical analysis
All data is expressed as a median, mean and one standard deviation (±) unless otherwise stated.
Results
Ten male and two female patients were recruited into this study. At the time of referral, the mean age was 45.5 years (27–69 years), and 22 weeks (from 2 days to 72 weeks) was the median time interval between injury and nerve transfer surgery. All patients could flex the elbow against resistance (MRC 4 or greater). Four patients had right-sided brachial plexus injuries and eight had left-sided injuries; seven of these were on the nondominant side and five were on the dominant. All 12 patients acquired their injuries through road traffic accidents. A total of five patients had C5/6 nerve root avulsions and another two patients had avulsions of C5/6/7. Tumor of the upper trunk was found in four patients and a humeral fracture with accompanying rupture of the musculocutaneous nerve was found in one patient [8].
The following data was obtained at a mean follow-up interval of 58 months (28–100) for co-contractability of muscle: the population mean co-contraction ratio was higher in reinnervated muscle compared with uninjured muscle, 0.16 (± 0.11) compared with 0.11 (± 0.06) respectively. This differential was not statistically significant.
Six repeated isometric contractions
Co-contraction ratios were higher in reinnervated muscle across all six isometric contractions with an overall trend to decrease after the first three isometric contractions (Figure 2).
(A) Mean co-contraction ratios (error bars represent ± 1 standard deviation) for each isometric contractions of elbow flexor muscles in the reinnervated (n = 12) and uninjured contralateral arms (n = 12). (B) Tabulated co-contraction data from (A). Data presented as mean ± 1 standard deviation.
When comparing the change in co-contraction ratio between isometric contractions, there was no statistically significant change (Figure 2).
Sustained isometric contraction of 60 sec
The population mean co-contraction ratio across the 60 s sustained isometric contraction was 0.182 (±0.134) for the uninjured arm compared with 0.165 (±0.085) for the reinnervated arm. This was not a statistically significant differential.
When comparing between the time groups, there was no statistically significant difference (Figure 3). However, in the reinnervated arms there was a trend was for a higher co-contraction ratio in the first 20 s of the sustained isometric contraction (Figure 3). After this time period, the co-contraction ratio in the reinnervated arms returned to levels similar to those found in the uninjured muscle (~0.165) (Figure 3).
(A) Mean co-contraction ratios (error bars represent ± 1 standard deviation) for each time group of the sustained isometric contraction of elbow flexor muscles in the reinnervated (n = 12) and uninjured contralateral arms (n = 12). (B) Tabulated co-contraction data from (A). Data presented as mean ± 1 standard deviation.
Discussion
This study set out to quantify co-contraction in reinnervated elbow flexor muscles. The study was designed to provide a greater insight into the mechanism(s) that underpin patient-reported impairments in reinnervated muscle: an earlier onset of muscle fatigue and decreased proprioception.
Overall, the results presented in this study demonstrate a larger range of co-contraction ratios in reinnervated muscle. This is consistent with previous findings which have shown significant variation in co-contraction ratios [43]. Few studies have reported statistically significant differences when comparing co-contraction ratios in patients with neuromuscular pathologies to healthy controls [44,45]. Reasons for this may be that measurement of any statistical difference in such a widely variable process is challenging. This could be circumvented by the recruitment of larger cohorts of patients (which is challenging in rare pathologies such as brachial plexus injury). However, this could render the result susceptible to type one statistical errors. The underlying variations in co-contraction rates are largely attributable to the fact that co-contraction is determined by local, joint-specific as well as generalized, individual-specific factors [43].
In nerve injured limbs, repeated isometric contractions demonstrate a trend toward higher co-contraction ratios in comparison to uninjured limbs. Co-contraction is deployed in healthy motor systems to improve function in novel or challenging environments (i.e., carrying a tray of drinks on a train or learning to ride a bike). It can be considered pathologic if it hinders function (by reducing pvf) and/or exacerbates fatigue [15,21]. Identical experimental protocols used to quantify the earlier onset of fatigue in reinnervated elbow flexor muscles found that reinnervated muscles demonstrated an earlier onset of fatigue as well as a lower pvf (approximately threefold lower) compared with uninjured arms. When considering the results from both studies it is observed that there is a link between lower pvf/earlier onset of fatigue and increased co-contraction ratios in reinnervated arms. This finding concurs with animal models of nerve injury. It is known that trauma to muscle proprioceptors induces a number of changes in the periphery and the CNS [9,12]. Some muscle proprioceptors do not regenerate beyond the injury site and some may lose sensitivity to stretch [46–49]. In addition, muscle stretch signalling is impaired by around 50% of the afferents innervating inappropriate targets rather than muscle spindles [46,47,49]. More recently, it has been shown that interrupted synaptic transmission in the spinal cord also contributes to an impaired muscle stretch response [50–53]. Reduced afferent feedback from the nerve injured muscle is thought to lead to an adaptive increase in proprioceptive feedback from the antagonist muscle [54,55]. This dysregulation of mechanosensory circuits in the spinal cord leads to higher co-contraction ratios in nerve-injured limbs [17,54–56].
In both the repeated and sustained isometric elbow flexor contractions, reinnervated arms demonstrated higher co-contraction ratios in the first three repetitive contractions and first 20 s of the sustained contraction. Co-contraction then decreased to levels similar to that of the uninjured arm. This concurs with findings in other studies of neuromuscular pathologies which have explored co-contraction [25,26,45]. A number of these studies found that co-contraction increased during dynamic movement and then reduced with subsequent repetition of the same exercise [25,26,45]. We postulate that in order to compensate for the earlier onset of fatigue in reinnervated muscle, co-contraction is reduced in an attempt to sustain force over time. This may be attributable to a process mediated in the CNS [9,12]; as the biceps muscle fatigues, there may be suppression of synaptic transmission to the motor pool controlling the triceps (antagonist muscle) leading to a reduction in the co-contraction ratio. This hypothesis is supported by the fact that changes in co-contraction preceded changes in force output which were concurrently measured and reported in a recent study by the authors. This should be further explored in animal models to investigate the interactions of mechanosensory reflexes in reinnervated muscle and how these changes under experimental conditions can induce muscle fatigue.
It is recognized that this study has a number of limitations that should be addressed in future studies. First, out of total of 68 eligible participants, only 12 attended for review (17%). This low rate of follow-up is a frequent feature of the trauma population [57,58]. This study was executed at a national referral center therefore many of the patients had to travel long distances to partake in the study. A 2-year point for follow-up was selected to allow for maximal muscle reinnervation and central plasticity assessment [29,59]. However, this chronic time point postoperatively may have contributed to low follow-up rates.
Second, it has been previously reported that patients who have undergone an Oberlin nerve transfer may co-contract the flexor-pronator mass in order to augment elbow flexion [59]. This phenomenon is known as the ‘Steindler effect’ and manifests in this patient demographic because patients initially use wrist flexion to drive elbow flexion [59]. While the study protocol was designed to bias elbow flexion to the biceps brachialis muscles, some patients found it challenging to isolate elbow flexion and wrist flexion. Future research should use sEMG to study activation of the wrist flexor mass while also instructing patients to extend the wrist and fingers while flexing the elbow. In addition, there is some evidence to suggest that co-contraction around the elbow joint can be more reliably assessed if the EMG signals are normalized by their respective maximal voluntary contraction obtained during maximal coactivation prior to dividing the antagonistic muscle activity by the mean between the agonistic and antagonistic muscle activations [60]. Future studies should consider this.
There was substantial variation in the following variables in this patient cohort: patient age, time elapsed between injury and surgery as well as time between surgery and follow-up.
Finally, although this study was informed by studies that characterized patient-reported impairments in reinnervated muscle, the present study was purely objective. Future work should incorporate subjective outcome measures to better understand the many facets of reinnervated muscle.
Conclusion
In summary, the assessment of motor recovery must be considered beyond pvf. This study has characterized and quantified co-contraction in reinnervated muscle. The earlier onset of fatigue in reinnervated muscle is at least partially attributable to higher co-contraction. These metrics should be adopted into clinical assessments to enable a more meaningful comparison to be made between differing treatment options for motor recovery.
Future perspective
Clinical translation of therapies developed in the laboratory requires the development of outcome measures which correlate with the human experience of muscle reinnervation. For the first time, this study shows co-contraction is an important feature related to patient reported impairments in reinnervated muscular function. Incorporation of co-contraction measurements into the design of future clinical trials of regenerative therapies will be important to ensure they can capture a meaningful response.
Co-contraction describes co-ordination of agonist and antagonist muscles around a joint.
Accurate and precise movements around a joint are highly dependent on appropriate co-contraction around a joint.
For the first time, this study shows that co-contraction becomes imbalanced in reinnervated muscle correlating with an earlier onset of muscle fatigue.
Author contributions
M Wilcox analyzed the data and wrote the manuscript. H Brown performed experiments and edited the manuscript. K Johnson assisted with experiments. M Sinisi reviewed the manuscript. T Quick conceived the project and edited the manuscript.
Acknowledgments
The authors would like to acknowledge funding from the Royal National Orthopaedic Hospital Charitable Trust (T Quick), a University College London Graduate Research Scholarship (M Wilcox) and the England Golf Trust (M Wilcox).
Presentations
This work was presented in part at American Society for Peripheral Nerve 2018, CA, USA and the Narakas Brachial Plexus Symposium 2018, Leiden, The Netherlands.
Financial disclosure
This work was supported by funding from the Royal National Orthopaedic Hospital Charitable Trust (T Quick), a University College London Graduate Research Scholarship (M Wilcox) and the England Golf Trust (M Wilcox). The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.
Competing interests disclosure
The authors have no competing interests or relevant affiliations with any organization or entity with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, stock ownership or options and expert testimony.
Writing disclosure
No writing assistance was utilized in the production of this manuscript.
Ethical conduct of research
The authors state that they have obtained appropriate institutional review board approval or have followed the principles outlined in the Declaration of Helsinki for all human or animal experimental investigations. In addition, for investigations involving human subjects, informed consent has been obtained from the participants involved.
Open access
This work is licensed under the Attribution-NonCommercial-NoDerivatives 4.0 Unported License. To view a copy of this license, visit http://creativecommons.org/licenses/by-nc-nd/4.0/
Papers of special note have been highlighted as: • of interest; •• of considerable interest
References
- 1. . Analysis of the Peripheral Nerve Repair Market in the in the US (2012). www.semanticscholar.org/paper/ANALYSIS-OF-THE-PERIPHERAL-NERVE-REPAIR-MARKET-IN-Brattain/cba129fbda15780ea0abd09da3da97de04f9c6b5
- 2. . Advances in peripheral nerve regeneration. Nat. Rev. Neurol. 9(12), 668–676 (2013).
- 3. . Pathophysiology of surgical nerve disorders. Winn HR (Ed.). (2011).
- 4. . A quantitative assessment of the functional recovery of flexion of the elbow after nerve transfer in patients with a brachial plexus injury. Bone Joint J. 98-B(11), 1517–1520 (2016).
- 5. . Fatigue analysis of human reinnervated muscle after microsurgical nerve repair. Clin. Orthop. Relat. Res. 31(334), 144–149 (1997).
- 6. . Critical reappraisal of Medical Research Council muscle testing for elbow flexion. J. Hand Surg. Am. 32(2), 149–153 (2007). • Shows how Medical Research Council grading for the recovery of muscular function following nerve injury is insensitive to the highly slow and incremental process of nerve regeneration.
- 7. . The lived experience of motor recovery of elbow flexion following Oberlin nerve transfer: a qualitative analysis. Hand Ther. 23(4), 130–138 (2018). •• This paper identifies co-contraction as a key patient descriptor of muscle reinnervation.
- 8. . An assessment of fatigability following nerve transfer to reinnervate elbow flexor muscles. Bone Joint J. 101-B(7), 867–871 (2019).
- 9. . Complex impairment of IA muscle proprioceptors following traumatic or neurotoxic injury. J. Anat. 227(2), 221–230 (2015).
- 10. . The proprioceptive senses: their roles in signaling body shape, body position and movement, and muscle force. Physiol. Rev. 92(4), 1651–1697 (2012).
- 11. . Kinesthesia: the role of muscle receptors. Muscle Nerve 34(5), 545–558 (2006).
- 12. . Central suppression of regenerated proprioceptive afferents. J. Neurosci. 25(19), 4733–4742 (2005).
- 13. . Relationship between cocontraction, movement kinematics and phasic muscle activity in single-joint arm movement. Exp. Brain Res. 140(2), 171–181 (2001).
- 14. . The role of cocontraction in the impairment of movement accuracy with fatigue. Exp. Brain Res. 185, 151–156 (2008).
- 15. . Role of cocontraction in arm movement accuracy. J. Neurophysiol. 89(5), 2396–2405 (2003).
- 16. Abnormal motor unit synchronization of antagonist muscles underlies pathological co-contraction in upper limb dystonia. Brain 121(Pt 5), 801–814 (1998).
- 17. . Dysregulation of mechanosensory circuits coordinating the actions of antagonist motor pools following peripheral nerve injury and muscle reinnervation. Exp. Neurol. 318, 124–134 (2019).
- 18. . The effect of antagonist muscle sensory input on force regulation. PLOS ONE 10(7), e0133561 (2015).
- 19. . The role of cocontraction in the impairment of movement accuracy with fatigue. Exp. Brain Res. 185(1), 151–156 (2008).
- 20. . Influence of fatigue on EMG/force ratio and cocontraction in cycling. Med. Sci. Sports Exerc. 32(4), 839–843 (2000).
- 21. . Relationship between cocontraction, movement kinematics and phasic muscle activity in single-joint arm movement. Exp. Brain Res. 140(2), 171–181 (2001).
- 22. Short- and long-term changes in joint co-contraction associated with motor learning as revealed from surface EMG. J. Neurophysiol. 88(2), 991–1004 (2002).
- 23. . Comparison of electromyography and joint moment as indicators of co-contraction. J. Electromyogr. Kinesiol. 22(4), 607–611 (2012).
- 24. Functional implications of muscle co-contraction during gait in advanced age. Gait Posture 53, 110–114 (2017).
- 25. . Reductions in co-contraction following neuromuscular re-education in people with knee osteoarthritis. BMC Musculoskelet. Disord. 17(1), 372 (2016).
- 26. . Modeling and simulating the neuromuscular mechanisms regulating ankle and knee joint stiffness during human locomotion. J. Neurophysiol. 114(4), 2509–2527 (2015).
- 27. . Effects of experimental muscle pain on muscle activity and co-ordination during static and dynamic motor function. Electroencephalogr. Clin. Neurophysiol. 105(2), 156–164 (1997).
- 28. Lower limb antagonist muscle co-activation and its relationship with gait parameters in cerebellar ataxia. Cerebellum 13(2), 226–236 (2014).
- 29. . Results of reinnervation of the biceps and brachialis muscles with a double fascicular transfer for elbow flexion. J. Hand Surg. Am. 30(5), 978–985 (2005).
- 30. . Nerve transfer to biceps muscle using a part of ulnar nerve for C5-C6 avulsion of the brachial plexus: anatomical study and report of four cases. J. Hand Surg. Am. 19(2), 232–237 (1994).
- 31. . Nerve transfer to biceps muscle using a part of the ulnar nerve in brachial plexus injury (upper arm type): a report of 32 cases. J. Hand Surg. Am. 23(4), 711–716 (1998).
- 32. . Surface electromyograms of agonist and antagonist muscles during force development of maximal isometric exercises – effects of instruction. Eur. J. Appl. Physiol. 89(1), 79–84 (2003).
- 33. . Muscle activation of the elbow flexor and extensor muscles during self-resistance exercises: comparison of unilateral maximal cocontraction and bilateral self-resistance. J. Strength Cond. Res. 26(9), 2468–2477 (2012).
- 34. . Cocontraction of the elbow muscles during combined tasks of pronation-flexion and supination-flexion. Electromyogr. Clin. Neurophysiol. 31(8), 483–488 (1991).
- 35. . Role of cocontraction in arm movement accuracy. J. Neurophysiol. 89(5), 2396–2405 (2003).
- 36. . Electrical manifestations of muscle fatigue during concentric and eccentric isokinetic knee flexion–extension movements. IEEE Trans. Biomed. Eng. 53(7), 1309–1316 (2006).
- 37. . Electromyographic models to assess muscle fatigue. J. Electromyogr. Kinesiol. 22(4), 501–512 (2012).
- 38. . Development of recommendations for SEMG sensors and sensor placement procedures. J. Electromyogr. Kinesiol. 10(5), 361–374 (2000).
- 39. European recommendations for surface electromyography: results of the SENIAM project. Roessingh Research and Development, The Netherlands (1999).
- 40. Cocontraction measured with short-range stiffness was higher in obstetric brachial plexus lesions patients compared to healthy subjects. J. Biomech. 63, 192–196 (2017).
- 41. . Electromyography exposes heterogeneity in muscle co-contraction following stroke. Front. Neurol. 8, 699–699 (2017).
- 42. . Muscle activation profiles and co-activation of quadriceps and hamstring muscles around knee joint in Indian primary osteoarthritis knee patients. J. Clin. Diagn Res. 11(5), RC09–RC14 (2017).
- 43. . Muscle coactivation: a generalized or localized motor control strategy? Muscle Nerve 48(4), 578–585 (2013).
- 44. Abnormal motor unit synchronization of antagonist muscles underlies pathological co-contraction in upper limb dystonia. Brain 121(5), 801–814 (1998).
- 45. . Co-activation: its association with weakness and specific neurological pathology. J. Neuroeng. Rehabil. 3, 26 (2006).
- 46. . Specificities of afferents reinnervating cat muscle spindles after nerve section. J. Physiol. 408, 345–372 (1989).
- 47. . On the specificity of sensory reinnervation of cat skeletal muscle. J. Physiol. 375, 587–609 (1986).
- 48. . Muscle receptors in the cross-reinnervated soleus muscle of the cat. J. Physiol. 331, 367–383 (1982).
- 49. . Regeneration of afferent and efferent fibres to muscle spindles after nerve injury in adults cats. J. Physiol. 260(2), 253–266 (1976).
- 50. . Factors that control the efficacy of group Ia synapses in alpha-motoneurons. J. Physiol. (Paris) 83(3), 133–140 (1988).
- 51. . Handbook of Physiology. the Nervous System American Physiological Society, USA, 423–507 (1981).
- 52. . Normal distribution of VGLUT1 synapses on spinal motoneuron dendrites and their reorganization after nerve injury. J. Neurosci. 34(10), 3475–3492 (2014).
- 53. . Permanent central synaptic disconnection of proprioceptors after nerve injury and regeneration. I. Loss of VGLUT1/IA synapses on motoneurons. J. Neurophysiol. 106(5), 2450–2470 (2011).
- 54. . Neural contributions to muscle fatigue: from the brain to the muscle and back again. Med. Sci. Sports Exerc. 48(11), 2294–2306 (2016).
- 55. Recovery of proprioceptive feedback from nerve crush. J. Physiol. 589(Pt 20), 4935–4947 (2011).
- 56. . Central suppression of regenerated proprioceptive afferents. J. Neurosci. 25(19), 4733–4742 (2005).
- 57. . Outpatient follow-up after traumatic injury: challenges and opportunities. J. Emerg. Trauma Shock 7(4), 256–260 (2014).
- 58. . Follow-up disparities after trauma: a real problem for outcomes research. Am. J. Surg. 199(3), 348–352; discussion 353 (2010).
- 59. . Donor activation focused rehabilitation approach: maximizing outcomes after nerve transfers. Hand Clinics 32(2), 263–277 (2016).
- 60. . A simple test of muscle coactivation estimation using electromyography. Braz. J. Med. Biol. Res. 45(10), 977–981 (2012).
