Heteronymous Ia-afferent connections in the upper limb following stroke

ABSTRACT: Synergistic arm movement patterns are common followingstroke and may arise through enhanced spinally mediated reflex connectionsbetween muscles. Our goal was to investigate the excitability of heteron-ymous Ia-afferent pathways in people with chronic stroke. Responses totendon taps of the flexor carpi radialis (FCR) muscle were recorded in FCR,biceps brachii (BB), and middle deltoid (MD) of 13 people with stroke and 13controls. Heteronymous reflexes were elicited in BB and MD in some, but notall, stroke and control subjects. The prevalence and size of the heteron-ymous responses were not significantly different between groups. Homon-ymous reflex responses in FCR were significantly larger in the stroke group.We found that the excitability of heteronymous Ia-mediated pathways fromFCR to BB and MD muscles is not enhanced following stroke, despiteexaggerated homonymous reflexes, and they are therefore not likely tocontribute to coactivation of forearm and more proximal upper limb muscles.

Muscle Nerve 41: 71–77, 2010

HETERONYMOUS Ia-AFFERENT CONNECTIONS INTHE UPPER LIMB FOLLOWING STROKE

GWYN N. LEWIS, PhD and PETER J. McNAIR, PhD

Health and Rehabilitation Research Centre, AUT University, Private Bag 92006,Auckland 1142, New Zealand

Accepted 13 May 2009

Following stroke, the formation of synergistic mus-cle activation patterns in the upper limb is one ofthe most prevalent motor dysfunctions.1 Musclesynergies present as unintended coupling or coac-tivation of additional muscles during intended acti-vation of another muscle. Such muscle couplinggives rise to stereotypical movement patterns thatdisrupt multijoint control and impair the perform-ance of everyday tasks. The most common activa-tion pattern is a flexion synergy, which incorpo-rates shoulder abduction, elbow flexion, and wristflexion/supination. This synergy is particularly dis-ruptive in reaching movements, as it severelyrestricts the functional workspace.2 Potential neu-ral mechanisms for the formation of muscle syner-gies include a lack of specificity of descending out-put from motor cortex, increased reliance onbrainstem motor pathways, or hyperexcitable reflex

connections between muscles at the segmentallevel.3

There is a strong neural basis for afferent-related pathways at the segmental level to contrib-ute to coupling of muscle activity. Pioneering workby Eccles demonstrated a vast array of heterony-mous Ia monosynaptic connections in the hind-limb of the cat.4 In humans, Ia afferents have beenshown to form heteronymous connections betweensynergist and non-synergist muscles in the upperlimb.5–11 Following stroke, changes in these spinalpathways or in their descending regulation couldcontribute to abnormal motor recruitment pat-terns.12 For example, if inhibition of these heter-onymous connections from supraspinal centers waslost or reduced following stroke, an enhanced cou-pling of neurally connected motoneuron poolscould arise.

The excitability of upper limb Ia-mediated het-eronymous pathways has not been studied exten-sively in the stroke population. The goal of thisstudy was to investigate Ia-afferent connectionsbetween muscles involved in the flexion synergy inpeople with chronic stroke. Previous studies haveshown that distal-to-proximal Ia-mediated pathwaysare more common in the human upper limb com-pared with proximal-to-distal pathways.7 Therefore,

Abbreviations: BB, biceps brachii; EMG, electromyographic; FCR,flexor carpi radialis; FMA, Fugl–Meyer assessment; MD, middle deltoid;PAD, post-activation depression; rms, root mean square

Correspondence to: G.N. Lewis; e-mail: [email protected]

VC 2009 Wiley Periodicals, Inc.Published online 18 September 2009 in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/mus.21444

Key words: hemiparesis; heteronymous pathway; Ia afferent; stretch reflex;upper limb

Upper Limb Reflexes in Stroke MUSCLE & NERVE January 2010 71

we examined reflex responses in the flexor carpiradialis (FCR), biceps brachii (BB), and middledeltoid (MD) muscles following tendon taps to theFCR. It was our hypothesis that, if heteronymousneural pathways involving FCR Ia afferents contrib-ute to synergistic activation of elbow and shouldermuscles in people with stroke, heteronymousreflex responses in the BB and MD would behyperexcitable compared with neurologically intactindividuals.

METHODS

Subjects. Thirteen people with hemiparesis fol-lowing stroke (Table 1) and 13 age-matched con-trol subjects (age 31–81 years, mean 64 � 15 years;2 left-hand dominant) participated in the study.Subjects with stroke were required to be at least6 months post-stroke and to display residualimpairments in upper limb function. Subjects withflexion contractures of the wrist joint wereexcluded if the wrist could not be passivelyextended to the neutral position. The functionalability of the stroke group was assessed using theupper limb Fugl-Meyer assessment13 and the Boxand Block Test14 at the beginning of the test ses-sion. All participants were required to have noother neurological conditions or any orthopedicimpairments of either upper limb. The affectedarm was tested in stroke subjects, and the rightarm was tested in all control subjects. Ethics ap-proval for the study was received from the ethicscommittee at AUT University, and informed writ-ten consent was obtained prior to participation.

Experimental Set-Up. Subjects were seated with theshoulder of the target arm flexed and abducted to30�. The forearm and hand were supported in a rigidfiberglass cast with the forearm in mid-pronation, thewrist in a neutral position, and the elbow approxi-mately 45� from full extension (see Fig. 1). An open-ing was present in the fiberglass cast over the distalmedial surface of the forearm to expose the FCR ten-don. Subjects were able to relax in this position with-out activation of the target upper limb muscles.

An electrodynamic shaker (V201; Ling DynamicSystems, UK) was used to apply taps to the FCRtendon. A 0.5-cm3 rubber stop was adhered to theend of the shaker protuberance. The shaker waspositioned so that light pressure was applied to theFCR tendon approximately 5 cm proximal to thevolar crease of the wrist joint. Single square-wavepulses (10-ms duration) were delivered to theshaker that resulted in a 1-mm movement of theshaker protuberance onto the FCR tendon.

Electromyography. Electromyographic (EMG) ac-tivity was recorded from the FCR, BB, and MDusing an AMT-8 device (Bortec Biomedical, Can-ada). Self-adhesive Ag–AgCl electrodes with a 2-cminterelectrode distance (Noraxon, Scottsdale, Ari-zona) were applied to the target muscles after shav-ing to remove hair and cleansing with alcohol.EMG data were bandwidth filtered (10–1000 HZ)and sampled at 5000 HZ using Micro1401 and Sig-nal software (CED, UK).

Protocol. Responses to FCR tendon taps wererecorded in FCR, BB, and MD during complete

Table 1. Characteristics of stroke subjects.

Subject Age Gender Months PS FMA AS BB-tested UA BB-tested AA

P1 53 M 89 42 L 66 10P2 61 M 37 54 L 49 18P3 46 M 21 16 R 75 1P4 42 F 47 55 L 58 22P5 51 F 7 31 R 67 3P6 83 M 23 7 L 52 0P7* 61 M 38 53 R 46 20P8 78 F 14 42 L 54 9P9 73 F 48 14 R 61 0P10 79 F 23 47 L 74 52P11* 84 M 8 52 R 62 37P12 68 M 38 27 R 58 0P13 72 M 84 54 R 51 36Mean (SD) 65 (14) 8 M 37 (26) 38 (17) 6 L 59 (9) 16 (17)

PS, post-stroke; FMA, upper limb Fugl-Meyer assessment (out of 66); AS, affected side; BB, Box and Block; UA, unaffected arm; AA, affected arm; M,male; F, female.*Subjects from whom homonymous reflexes could not be elicited.

72 Upper Limb Reflexes in Stroke MUSCLE & NERVE January 2010

muscle relaxation and during activation of one ofthe target muscles. When activation was required,a visual display of EMG activity in the specifiedmuscle was provided, and subjects were asked toactivate the muscle at a minimal level (5–10% ofmaximum activation) that could be sustained con-sistently. For each condition, 40 tendon taps wereapplied at 5-second intervals. Tendon taps werefirst applied in the relaxed condition. The remain-ing three muscle activation conditions were com-pleted in a random order. Rest breaks were pro-vided as needed to reduce the influence of fatigue.

Data Processing. Any recordings that did not havethe appropriate level of muscle activation prior tothe tendon tap were removed before further analy-sis. In each condition, the remaining responseswere averaged. Rectified EMG was averaged in par-allel to determine pre-stimulus levels of contrac-tion. Response amplitude was determined as themaximum peak-to-peak amplitude of the non-recti-

fied averaged response in a window from 20–60 msfollowing the tendon tap. Response amplitude wasexpressed relative to the root mean square (rms)of a 30-ms window of pre-stimulus EMG. To deter-mine if a reflex response was elicited in the post-stimulus EMG, an automated procedure was usedto determine deflections in the response windowof >3 standard deviations (SD) of the pre-stimulusEMG. If detected, a response was deemed present,and the onset latency recorded.

Statistical Analysis. The number of subjects inwhom homonymous and heteronymous responseswere elicited was compared between stroke andcontrol groups using Pearson’s chi-square distribu-tion. Normalized response amplitude, response la-tency, and level of pre-stimulus EMG were com-pared between groups using Student’s t-tests. Alevel of significance of a ¼ 0.05 was adopted.Results are presented as mean � SD.

RESULTS

Homonymous and heteronymous stretch reflexresponses were elicited in some, but not all, strokeand control subjects. Example individual responsesobtained during muscle pre-activation are shownin Figure 1B. In the individual with stroke (left), aclear reflex response was elicited in FCR, BB, andMD following the FCR tendon tap. In the controlparticipant (right), a smaller response was evidentin FCR and BB, but a response in MD was not eli-cited. Group results are presented in what follows.

Heteronymous Reflexes in BB and MD. There wereno heteronymous responses elicited in BB or MDin any of the subjects while the muscles were atrest. Heteronymous responses were seen in BB in9 stroke subjects and 6 control subjects when BBwas pre-activated. This distribution was not signifi-cantly different between the two groups (Pearson’schi-square P ¼ 0.3). Across all subjects, the averagenormalized response amplitude was 3.3 � 2.3 inthe stroke group and 3.1 � 1.4 in controls (Fig. 2,left). These findings were not significantly differ-ent (P ¼ 0.4). The onset latencies of the responses(stroke: 32.1 � 6.2 ms; control: 30.5 � 5.7 ms) alsowere not different between the stroke and controlgroups (P ¼ 0.6). In MD, heteronymous responseswere seen in 5 stroke and 3 control subjects (Pear-son’s chi-square, P ¼ 0.7). The average normalizedresponse amplitude was 1.7 � 0.4 in the strokegroup and 1.2 � 0.3 in the control group. Again,this finding indicated no significant difference

FIGURE 1. (A) Experimental set-up. The subject’s forearm was

enclosed in a rigid fiberglass cast that was secured to the desk.

Tendon taps were applied to the flexor carpi radialis (FCR)

muscle through an opening in the cast over the volar surface of

the wrist. (B) Example responses from individual stroke (left)

and control (right) subjects during muscle pre-activation. Traces

are an average of approximately 40 tendon taps to the FCR

muscle, as indicated by the arrow. Heteronymous responses

are elicited in the biceps brachii (BB) and middle deltoid (MD)

muscles of the stroke subject, and in the BB of the control sub-

ject. Note the larger homonymous stretch reflex response in the

FCR of the stroke subject compared with the control. [Color fig-

ure can be viewed in the online issue, which is available at

www.interscience.wiley.com.]


between groups (P ¼ 0.9). The onset latencies ofthe MD responses (stroke: 29.0 � 3.3 ms; control:32.3 � 12.3 ms) were similar, but were not com-pared statistically due to the small number of sub-jects showing responses.

To determine whether there was a relationshipbetween the presence of heteronymous reflexresponses and functional recovery in the strokegroup, we correlated the size of BB and MD responsesusing Fugl-Meyer assessment (FMA) scores. The cor-relation was not significant for either BB (Spearman’srho ¼ 0.33, P ¼ 0.3) or MD (Spearman’s rho ¼ 0.12,P ¼ 0.7) responses.

For both BB and MD, the pre-stimulus EMGrms was significantly higher in the control group(both P < 0.05). Across both groups, we correlatedthe level of pre-stimulus EMG rms with normalizedresponse amplitude. Correlations did not reach sig-nificance levels for BB (Spearman’s rho ¼ 0.32, P¼ 0.1) or MD (Spearman’s rho ¼ 0.37, P ¼ 0.06),indicating that there were no strong relationshipsbetween pre-stimulus EMG and normalizedresponse amplitude.

Homonymous Reflexes in FCR. Reflex responses inFCR were evident in 10 of 13 stroke subjects and 9of 13 control subjects while FCR was relaxed. Thisdistribution was not significant between groups (P¼ 0.7). When FCR was pre-activated, responseswere evident in FCR of 11 stroke and 11 controlsubjects (P ¼ 1.0). In contrast to the heterony-mous responses, homonymous reflex responses inFCR were larger in the stroke group (Fig. 2, right).Normalized response amplitude was significantlylarger in the stroke subjects when FCR was relaxed(P ¼ 0.03) as well as when activated (P ¼ 0.02). Av-erage reflex response onset latency in the relaxedFCR was 30.8 � 5.7 ms in the stroke subjects and33.2 � 2.7 ms in the control subjects. This differ-ence was not significant (P ¼ 0.3). Average reflexresponse onset latency in the pre-activated FCRwas 32.4 � 4.8 ms in the stroke subjects and 33.3� 4.3 ms in the control subjects. Again, there wasno significant difference between groups (P ¼0.7).

Analysis of the pre-stimulus EMG rms in FCRrevealed a similar level of background muscle acti-vation in the stroke and control groups, both inthe relaxed (P ¼ 0.2) and activated conditions (P¼ 0.8). Correlations using Spearman’s rhorevealed no significant relationships between pre-stimulus EMG rms and the normalized responsesize (relaxed: r ¼ �0.1, P ¼ 0.6; pre-activated: r ¼�0.33, P ¼ 0.1).

DISCUSSION

Our results support previous studies that showedheteronymous reflex pathways from forearm toelbow and shoulder muscles in the upperlimb.5,15,16 We have expanded on the knowledgeof these pathways by investigating their prevalencein a population with post-stroke hemiparesis.Against our hypothesis, we found that the excitabil-ity of heteronymous reflex responses was not dif-ferent between stroke and control subjects. This isin spite of the fact that homonymous reflexresponses in FCR were substantially larger in thestroke group. Thus, our results provide no evi-dence showing that hyperexcitable Ia heterony-mous connections from the forearm contribute toupper limb synergistic muscle activation patternsin chronic stroke.

Heteronymous Reflex Responses. Although smallerthan homonymous responses in FCR, clear heter-onymous reflex responses were elicited in BB and,

FIGURE 2. Group averages of response size for heteronymous

and homonymous reflex responses following tendon taps to the

flexor carpi radialis (FCR). Response size is shown normalized to

background muscle activation. BB, biceps brachii; MD, middle del-

toid. Error bars represent 1 standard error of the mean. *P < 0.05.


to a lesser extent, MD muscles of several partici-pants in the stroke and control groups. Previousstudies in healthy populations have shown modula-tion of shoulder15,16 and elbow6,17 muscle moto-neuron pools following presumed activation ofgroup I afferents from forearm muscles. Thesepathways are likely to be involved in functionalcoordination of multijoint tasks, such as reach-ing,8,16,17 or in stabilization of the proximal armduring wrist and hand movement.5 We assessedthe strength of these pathways in a stroke popula-tion to determine if there was any evidence ofaltered excitability that may contribute to synergis-tic activation of forearm, elbow, and shouldermuscles. Heteronymous reflex responses were notmore prevalent or more excitable in our strokesubjects compared with controls. Therefore, wefound no evidence of heightened heteronymous Iaconnections between forearm and elbow/shouldermuscles in our stroke population. Miller and col-leagues10 examined heteronymous reflex responsesin BB following median-nerve stimulation at theelbow. Response amplitude in BB displayed a lin-ear relationship with background muscle contrac-tion levels. Therefore, the higher level of pre-stim-ulus EMG in the control group in our study isunlikely to have influenced our results, given thatresponse amplitude was normalized to backgroundactivation. This is also supported by the lack of sig-nificant correlations between pre-stimulus EMGand normalized response size.

Muscle activation synergies are a robust charac-teristic in stroke survivors, so a heterogeneous sub-ject population with a wide range of impairmentlevels was included in our study. Targeting strokesurvivors with evidence of strong upper limb flexorsynergies may have altered our results. However, inour small sample we found no relationshipbetween FMA scores and the size of heteronymousresponses in BB and MD. Previous studies in strokeand healthy populations have also reported thatheteronymous reflexes are not evident in all sub-jects.9,12 These findings suggest that heteronymousreflex connections may have relatively limited func-tional contributions to multijoint control. It shouldbe taken into consideration that heteronymousreflex pathways examined in this and other studieshave been tested largely during static conditionsand that reflex pathway excitability is likely to beenhanced during more functional activationcontexts.18

Given the use of a tendon tap stimulus in ourstudy, it is probable that the heteronymous reflexresponses arose through activation of FCR Ia

afferents. Earlier studies suggested a monosynap-tic Ia pathway from forearm to elbow muscles inman.6,9,10 The latency of responses in BB and MDin our study is more in line with the laterresponses reported by McClelland and col-leagues,9 which were approximately 15 ms delayedfrom presumed monosynaptic connections. It ispossible that substantial temporal summation ofthe afferent volleys was required to activate BBand MD motoneurons in our study. In support ofthis conjecture, we were not able to elicit any het-eronymous responses while BB or MD were atrest, a finding similar to that in a previouswork.10 Alternatively, a polysynaptic pathwayinvolving proprioceptive interneurons5,19,20 or asupraspinal component of the reflex pathway maybe possible.12,20

Homonymous Reflex Responses. As expected, thehomonymous stretch reflex response in FCR wasenhanced in the stroke group. Spasticity is a cardi-nal symptom following lesions of the central nerv-ous system, and exaggerated velocity-dependentstretch reflexes are a requisite component of spas-ticity.21 Although the number of subjects in whomhomonymous responses were elicited was not dif-ferent between groups, the size of the reflexresponse was significantly larger in the strokegroup, both at rest and during activation of FCR.We did not find evidence of spasticity in all of ourstroke subjects. Seven of the 13 subjects with strokedemonstrated a reflex amplitude at rest that wasmore than twice the average amplitude of the con-trol subjects. This proportion agrees with estimatesof spasticity in the chronic stroke population.22

Previously, it was reported that stretch reflex am-plitude during tonic muscle activation is not exag-gerated in stroke subjects compared with healthycontrols.23–25 This conflicts with our finding of asubstantially enhanced reflex amplitude duringFCR activation in our stroke group. Muscle activa-tion levels were matched between the two groupsin our study, yet the stretch reflex response was sig-nificantly facilitated both in raw and normalizedvalues. This outcome may have been influenced byour use of a more direct stimulation of the tendonto elicit the stretch reflex response, as comparedwith joint perturbations used in previous studies.

Loss of descending regulation from the reticu-lospinal tract is likely to contribute to spasticity fol-lowing supraspinal lesions.26 Our finding of exag-gerated homonymous but not heteronymousreflexes in the stroke group suggests that the


enhanced homonymous FCR stretch reflex was notdue to a larger Ia volley. This supports the conten-tion that it is the processing of Ia input that isabnormal in spasticity, rather than control andactivation of Ia receptors.27,28 The precise segmen-tal pathways implicated in the development ofspasticity following stroke are yet to be defined.Reduced post-activation depression of Ia afferentshas been routinely reported in stroke studies.29,30

Rossi-Durand and colleagues31 reported that post-activation depression was substantially shorter (<1second) in FCR muscle compared with the morecommonly tested soleus muscle in the lower limb.In addition, post-activation depression was foundto be markedly attenuated when it was tested in anactive muscle.32–34 Therefore, it post-activationdepression is unlikely to fully explain theenhanced reflex responses in the stroke group inour study, as the tendon tap stimuli were 5 secondsapart. In the upper limb, impaired pre-synaptic in-hibition of FCR Ia afferents has been reported inindividuals with stroke29,35,36 and could accountfor the selective facilitation of homonymous butnot heteronymous responses. Finally, followingstroke, a greater proportion of the descendingdrive from the cortex to the upper limb involvesthe disynaptic propriospinal pathway.37 Enhancedexcitability of pre-motoneurons in this pathwaymay contribute to the exaggerated stretch reflexresponse in this population.

In conclusion, heteronymous reflex responseswere elicited in BB and MD of healthy and post-stroke individuals following tendon taps to FCR.These responses were likely to have arisen throughan Ia-mediated non-monosynaptic pathway. Theheteronymous pathway from FCR to BB and MDwas not hyperexcitable in people with chronic post-stroke hemiparesis. Therefore, we have provided noevidence to support a contribution of Ia pathwaysfrom the forearm to the formation of upper limbmuscle activation synergies. Instead, we suggest thatsynergistic activation of upper limb muscles follow-ing stroke may arise through abnormal excitabilityof spinal reflex pathways mediated by other affer-ents, such as group II,38,39 alterations in the balanceof activation of brainstem pathways,40 or a reduceddifferentiation of descending commands from themotor cortex.41,42 Finally, our results indicate thatthe homonymous stretch reflex response in FCR isexaggerated in stroke subjects while at rest and dur-ing muscle activation.

This study was funded by the Foundation for Research, Scienceand Technology, New Zeland.

REFERENCES

1. Brunnstrom S. Movement therapy in hemiplegia. New York:Harper and Row; 1970.

2. Ellis MD, Sukal T, DeMott T, Dewald JPA. Augmenting clini-cal evaluation of hemiparetic arm movement with a labora-tory-based quantitative measurement of kinematics as afunction of limb loading. Neurorehabil Neural Repair 2008;22:321–329.

3. Dewald JPA, Pope PS, Given JD, Buchanan TS, Rymer WZ.Abnormal muscle coactivation patterns during isometric tor-que generation at the elbow and shoulder in hemipareticsubjects. Brain 1995;118:495–510.

4. Eccles JC, Eccles RM, Lundberg A. The convergence ofmono-synaptic excitatory afferents on to many different spe-cies of motoneurones. J Physiol 1957;137:22–50.

5. Alexander CM, Harrison PJ. Reflex connections from fore-arm and hand afferents to shoulder girdle muscles inhumans. Exp Brain Res 2003;148:277–282.

6. Cavallari P, Katz R. Pattern of projections of Group Ia affer-ents from forearm muscles to motoneurones supplyingbiceps and triceps muscles in man. Exp Brain Res 1989;78:465–478.

7. Cavallari P, Katz R, Penicaud A. Pattern of projections ofGroup Ia afferents from elbow muscles to motoneuronessupplying wrist muscles in man. Exp Brain Res 1992;91:311–319.

8. Gielen CC, Ramaekers L, van Zuylen EJ. Long-latencystretch reflexes as co-ordinated functional responses inman. J Physiol 1988;407:275–292.

9. McClelland VM, Miller S, Eyre JA. Short latency heterony-mous excitatory and inhibitory reflexes between antagonistand heteronymous muscles of the human shoulder andupper limb. Brain Res 2001;899:82–93.

10. Miller TA, Mogyoros I, Burke D. Homonymous and heter-onymous monosynaptic reflexes in biceps brachii. MuscleNerve 1995;18:585–592.

11. Smeets JB, Erkelens CJ. Dependence of autogenic and het-erogenic stretch reflexes on pre-load activity in the humanarm. J Physiol 1991;440:455–465.

12. Sangani SG, Starsky AJ, McGuire JR, Schmit BD. Multijointreflexes of the stroke arm: neural coupling of the elbowand shoulder. Muscle Nerve 2007;36:694–703.

13. Fugl-Meyer AR, Jaasko L, Leyman I, Olsson S, Steglind S.The post-stroke hemiplegic patient. I. A method for evalua-tion of physical performance. Scand J Rehabil Med 1975;7:13–31.

14. Mathiowetz V, Volland G, Kashman N, Weber K. Adultnorms for the Box and Block Test of manual dexterity. AmJ Occup Ther 1985;39:386–391.

15. Creange A, Faist M, Katz R, Penicaud A. Distribution of het-eronymous Ia facilitation and recurrent inhibition in thehuman deltoid motor nucleus. Exp Brain Res 1992;90:620–624.

16. Roberts LV, Stinear CM, Lewis GN, Byblow WD. Task-de-pendent modulation of propriospinal inputs to humanshoulder. J Neurophysiol 2008;100:2109–2114.

17. Miyasaka T, Naito A, Shindo M, Kobayashi S, Hayashi M,Shinozaki K, et al. Modulation of brachioradialis moto-neuron excitabilities by group I fibers of the median nervein humans. Tohuku J Exper Med 2007;212:115–131.

18. Perreault EJ, Chen K, Trumbower RD, Lewis GN. Interac-tions with compliant loads alter stretch reflex gains but notintermuscular coordination. J Neurophysiol 2008;99:2101–2113.

19. Pierrot-Deseilligny E. Transmission of the cortical commandfor human voluntary movement through cervical pro-priospinal premotoneurons. Prog Neurobiol 1996;48:489–517.

20. Matthews PBC. The human stretch reflex and the motorcortex. Trends Neurosci 1991;14:87–91.


21. Lance JW. The control of muscle tone, reflexes, andmovement:Robert Wartenberg lecture. Neurology 1980;30:1303–1313.

22. Thilmann AF, Fellows SJ, Ross HF. Pathological changes in spas-tic muscle reflexes evoked by passive stretch or tendon taps. In:Thilmann AF, Burke DJ, Rymer WZ, editors. Spasticity: mecha-nisms andmanagement. Heidelberg: Springer; 1993. p 239–250.

23. Burne JA, Carleton VL, O’Dwyer N. The spasticity paradox:movement disorder or disorder of resting limbs? J NeurolNeurosurg Psychiatry 2005;76:47–54.

24. Ibrahim IK, Berger W, Trippel M, Dietz V. Stretch-inducedelectromyographic activity and torque in spastic elbowmuscles. Brain 1993;116:971–989.

25. Powers RK, Marder-Meyer J, Rymer WZ. Quantitative rela-tions between hypertonia and stretch reflex threshold inspastic hemiparesis. Ann Neurol 1988;23:115–124.

26. Priori A, Cogiamanian F, Mrakic-Sposta S.Pathophysiologyof spasticity. Neurol Sci 2006;27(suppl):s307–309.

27. Wilson LR, Gandevia SC, Inglis JT, Gracies JM, Burke D.Muscle spindle activity in the affected upper limb after aunilateral stroke. Brain 1999;122:2079–2088.

28. Wilson LR, Gracies JM, Burke D, Gandevia SC. Evidence forfusimotor drive in stroke patients based on muscle spindlethixotropy. Neurosci Lett 1999;264:109–112.

29. Aymard C, Katz R, Lafitte C, Lo E, Penicaud A, Pradat-DiehlP, et al. Presynaptic inhibition and homosynaptic depres-sion. A comparison between lower and upper limbs in nor-mal human subjects and patients with hemiplegia. Brain2000;123:1688–1702.

30. Masakado Y, Kagamihara Y, Takahashi O, Akaboshi K, Mur-aoka Y, Ushiba J. Post-activation depression of the soleus H-reflex in stroke patients. Electromyogr Clin Neurophysiol2005;45:115–122.

31. Rossi-Durand C, Jones KE, Adams S, Bawa P. Comparison ofthe depression of H-reflexes following previous activation inupper and lower limb muscles in human subjects. Exp BrainRes 1999;126:117–127.

32. Burke D, Adams RW, Skuse NF. The effects of voluntarycontraction on the H reflex of human limb muscles. Brain1989;112:417–433.

33. McNulty PA, Jankelowitz SK, Wiendels TM, Burke D. Postac-tivation depression of the soleus H reflex measured usingthreshold tracking. J Neurophysiol 2008;100:3275–3284.

34. Stein RB, Estabrooks KL, McGie S, Roth MJ, Jones KE.Quantifying the effects of voluntary contraction and inter-stimulus interval on the human soleus H-reflex. Exp BrainRes 2007;182:309–319.

35. Artieda J, Quesada P, Obeso JA. Reciprocal inhibitionbetween forearm muscles in spastic hemiplegia. Neurology1991;41:286–289.

36. Nakashima K, Rothwell JC, Day BL, Thompson PD, Shan-non K, Marsden CD. Reciprocal inhibition between forearmmuscles in patients with writer’s cramp and other occupa-tional cramps, symptomatic hemidystonia and hemiparesisdue to stroke. Brain 1989;112:681–697.

37. Mazevet D, Meunier S, Pradat-Diehl P, Marchand-Pauvert V,Pierrot-Deseilligny E. Changes in propriospinally mediatedexcitation of upper limb motoneurons in stroke patients.Brain 2003;126:988–1000.

38. Marque P, Simonetta-Moreau M, Maupas E, Roques CF.Facilitation of transmission in heteronymous group II path-ways in spastic hemiplagic patients. J Neurol Neurosurg Psy-chiatry 2001;70:36–42.

39. Maupas E, Marque P, Roques CF, Simonetta-Moreau M.Modulation of the transmission in group II heteronymouspathways by tizanidine in spastic hemiplegic patients. J Neu-rol Neurosurg Psychiatry 2004;75:130–135.

40. Ellis MD, Acosta AM, Yao J, Dewald JPA. Position-dependenttorque coupling and associated muscle activation in thehemiparetic upper extremity. Exp Brain Res 2007;176:594–602.

41. Cicinelli P, Traversa R, Rossini PM. Post-stroke reorganisa-tion of brain motor output to the hand: a 2–4 month fol-low-up with focal magnetic transcranial stimulation.Electroencephalogr Clin Neurophysiol 1997;105:438–450.

42. Traversa R, Cicinelli P, Bassi A, Rossini PM, Bernardi G.Mapping of motor cortical reorganization after stroke. Abrain stimulation study with focal magnetic pulses. Stroke1997;28:110–117.


Documents

Heteronymous Ia-afferent connections in the upper limb following stroke