3. Dry Needling_Spinal Cord Mechanism.pdf

tme

o

cbpdtsin

ascm

1098

ORIGINAL ARTICLE

Spinal Cord Mechanism Involving the Remote Effects of DryNeedling on the Irritability of Myofascial Trigger Spots inRabbit Skeletal Muscle
Yueh-Ling Hsieh, PT, PhD,* Li-Wei Chou, MD, MSc,* Yie-San Joe, MD, Chang-Zern Hong, MD
pt

nsst

a

A

lal

ABSTRACT. Hsieh Y-L, Chou L-W, Joe Y-S, Hong C-Z.Spinal cord mechanism involving the remote effects of dryneedling on the irritability of myofascial trigger spots in rabbitskeletal muscle. Arch Phys Med Rehabil 2011;92:1098-105.

Objective: To elucidate the neural mechanisms underlyinghe remote effects produced by dry needling rabbit skeletal

uscle myofascial trigger spots (MTrSs) via analyses of theirndplate noise (EPN) recordings.

Design: Experimental animal controlled trial.Setting: An animal laboratory of a university.Animals: Male New Zealand rabbits (N�96) (body weight,

2.5–3.0kg; age, 16–20wk).Intervention: Animals received no intervention for neural

interruption in group I, transection of the tibial nerve in groupII, transection of L5 and L6 spinal cord in group III, andtransection of the T1 and T2 spinal cord in group IV. Eachgroup was further divided into 4 subgroups: animals receivedipsilateral dry needling, contralateral dry needling, ipsilateralsham needling, or contralateral sham needling of gastrocne-mius MTrSs.

Main Outcome Measures: EPN amplitudes of biceps fem-ris (BF) MTrSs.Results: BF MTrS mean EPN amplitudes significantly in-

reased (P�.05) initially after gastrocnemius verum needlingut reduced to a level significantly lower (P�.05) than thereneedling level in groups I and IV with ipsilateral dry nee-ling or contralateral dry needling, and in group II with con-ralateral dry needling (but not ipsilateral dry needling). Noignificant EPN amplitude changes were observed in BF MTrSn group III or in the control animals receiving superficialeedling (sham).Conclusion: This remote effect of dry needling depends on

n intact afferent pathway from the stimulating site to thepinal cord and a normal spinal cord function at the levelsorresponding to the innervation of the proximally affecteduscle.

From the Department of Physical Therapy and Graduate Institute of RehabilitationScience, Taichung (Hsieh, Chou), and School of Chinese Medicine, College ofChinese Medicine, China Medical University, Taichung (Chou); Department ofPhysical Medicine and Rehabilitation, China Medical University Hospital, Taichung(Chou); Department of Physical Medicine and Rehabilitation, Cheng Ching Hospital,Taichung (Joe); College of Life Science, National Chung Hsing University, Taichung(Joe); Department of Physical Therapy, Hungkuang University, Taichung (Hong),Taiwan.

*Hsieh and Chou contributed equally to this work.Supported by a grant from National Science Council (NSC 99-2314-B241-001) and

China Medical University (CMU97-188), Taiwan.No commercial party having a direct financial interest in the results of the research

supporting this article has or will confer a benefit on the authors or on any organi-zation with which the authors are associated.

Reprint requests to Chang-Zern Hong, MD, Department of Physical Therapy,Hungkuang University, 34 Chung-Chie Road, Shalu, Taichung, Taiwan, e-mail:[email protected].

0003-9993/11/9207-00001$36.00/0doi:10.1016/j.apmr.2010.11.018

Arch Phys Med Rehabil Vol 92, July 2011

Key Words: Electromyography; Myofascial trigger spot;Needle stick; Neural pathway; Rehabilitation.

© 2011 by the American Congress of RehabilitationMedicine

MYOFASCIAL PAIN IS one of the most common sourcesof musculoskeletal pain and has as its hallmark the

resence of taut bands and small hyperirritable regions referredo as MTrPs.1,2 Clinically, a given MTrP has a characteristic

referred pain pattern and may be associated with an LTR inresponse to snapping palpation.2 In the MTrP region, electro-myographic activity of EPN can be recorded, and both preva-lence and amplitude of EPN can be used as indicators to assessthe irritability of MTrP.2-5

Dry needling has a well established role in the treatment ofmyofascial pain.6-9 Clinical studies have demonstrated that dryeedling MTrPs at the pain site suppresses their activity, re-ulting in pain reduction, but also that those MTrPs can beuppressed through dry needling distant MTrPs that are ana-omically proximal or distal to the site of clinical pain.8-12

Acupuncture studies have demonstrated the Traditional Chi-nese Medicine principle that pain conditions can be improved bystimulating acupuncture points distant from the site of pain.4,13-14

The effects of acupuncture may also spread to the contralateralside.15 Studies of acupuncture needle stimulation in anesthetizedanimals have identified a wide variety of reflex responses inremote modification of various organ functions.16 However, itsunderlying neuronal control mechanism remains unclear.

The purpose of this study is to confirm the remote effects of dryneedling on trigger points and to elucidate the neural mechanismsunderlying the remote effects produced by dry needling MTrSs(equivalent to human MTrPs) in the gastrocnemius muscle17,18 vianalyses of EPN recordings from the BF.

METHODS

nimalsThe experiments were performed on adult male New Zea-

and rabbits (age, 16–20wk; body weight, 2.5–3.0kg). Eachnimal was housed and cared for following the ethical guide-ines of the International Association for Study of Pain in

List of Abbreviations

ANOVA analysis of varianceBF biceps femorisDNIC diffuse noxious inhibitory controlEPN endplate noiseLTR local twitch responseMTrP myofascial trigger point (human)MTrS myofascial trigger spot (rabbit)
SEM standard error of mean
mailto:[email protected]

rdwd(ggaoaf

A

aalws

o

alate

1099NEURAL MECHANISM FOR REMOTE EFFECT OF DRY NEEDLING, Hsieh

animals.19 Effort was made to minimize discomfort and toreduce the number of animals used. All animal experimentswere conducted with the procedure approved by the AnimalCare and Use Committee of China Medical University, Tai-chung, Taiwan, in accordance with the Guidelines for AnimalExperimentation.

Ninety-six rabbits were divided randomly into 4 groups(fig 1) based on the procedure performed. Group I (n�24)animals received no surgical intervention (intact neural path-way), group II animals (n�24) received transection of the tibialnerve in the electrophysiologically investigated side (peripheralsensory pathway), group III animals (n�24) received transec-tion of the L5-6 spinal cord (BF innervation level), and groupIV animals (n�24) received transection of T1–T2 spinal cord(suprasegment of BF innervation). For the EPN amplitudevariable, a sample size of 24 subjects in each group wassufficient to give statistical power of 97.06% with a signifi-cance level of P less than .05. Animals in each group wereandomly divided further into 4 subgroups based on the con-ition of treatment of gastrocnemius: experimental animalsith ipsilateral dry needling (n�8) or contralateral dry nee-ling (n�8) and control animals with ipsilateral sham needlingn�4) or contralateral sham needling (n�4) on the MTrS of theastrocnemius. Fewer animals were studied in the controlroup because no significant changes were observed in allnimals treated with sham needling. Regarding the assignmentf groups or subgroups, animals were selected from the firstvailable litter, and subsequently from the next litter, and soorth according to the sequence in a random table.

nimal PreparationBefore anesthesia, the most tender spots (ie, MTrS) of BF

nd gastrocnemius were identified by finger pinching. Thenimal responded to pinch stimulation with withdrawal of theower limb, turning its head, screaming, and so forth, onlyhen the most painful spot was pinched, and this most tender

pot was confirmed as the MTrS.2,17,18,20-23 These painful

Fig 1. Study flow diagram. Contra, contr

regions were marked on the skin with an indelible marker and

were designated for electrophysiologic assessment or dry nee-dling. The animals were anesthetized with 2% isoflurane inoxygen flow for induction followed by a 0.5% maintenancedose.24 Body temperature was monitored by a thermistor probef a thermometera in the rectum and maintained at approxi-

mately 37.5°C using a body temperature control system con-sisting of thermostatically regulated direct current heating padand an infrared lamp. The hind limbs of anesthetized rabbitswere shaved and cleaned with povidone-iodine solution. Theskin of the lateral thigh in 1 randomly selected side was incisedto expose the BF, which served as an EPN recording site. Themarked spot region in the BF muscle was grasped between 2fingers from behind the muscle and the muscle palpated bygently rubbing (rolling) it between the fingers to discover a tautband. A taut band felt like a clearly delineated rope of musclefibers and was roughly 2 to 3mm or more in diameter. Thefibers of the taut band were unmistakably firmer in consistencythan the surrounding muscle.

Needling of Gastrocnemius MuscleAll needling procedures were performed by the same

investigator who was blind to the group assignment regard-ing the surgical intervention on neural pathway. Dry nee-dling stimulation was performed with a disposable 30Gacupuncture needle (300�m in diameter, 3.7cm in length)b atthe ipsilateral or contralateral gastrocnemius (fig 2). The tech-nique of dry needling was similar to that suggested by Hongand colleagues18,20,25-27 with multiple needle insertions to elicitrabbit-LTRs as much as possible. For needling in MTrS ofgastrocnemius, the needle was first inserted through the skinperpendicularly at the center of the marked spot and advancedslowly and gently into the muscle until the needle tip touchedthe bone surface to estimate the thickness of the muscle. Theneedle was withdrawn back to the subcutaneous layer andrapidly moved in and out for insertion of multiple sites indifferent directions (in a cone shape with the center at the initialneedle insertion of a perpendicular direction; the angle of the

ral; GAS, gastrocnemius; Ipsi, ipsilateral.

cone margin was about 20°). For each needle insertion, the


sgha(wesrpEnaTt

rwoa1mcMm

D

dSinmhwtTpspe

1100 NEURAL MECHANISM FOR REMOTE EFFECT OF DRY NEEDLING, Hsieh

needle was advanced into the depth near the bone surface.Simultaneous needle rotation was performed to facilitate fastin-and-out needle movement as suggested by Chou et al4 inorder to elicit as many LTRs as possible. For sham needling,the needle was inserted into the subcutaneous layer of themarked MTrS region at a depth approximately 1 to 2mm fromthe skin surface. After insertion, the needle stayed there with-out further movement.

Transection OperationsTransection of tibial nerve. During anesthesia for the an-

imals in group II, the incision was made over the posterioraspect of 1 thigh ipsilaterally to the EPN recording side. Underthe operating microscope, the sciatic nerve was exposed, andthe tibial nerve isolated and transected at the site about 1cmfrom its insertion into the gastrocnemius.

Transection of spinal cord. After completing laminectomyand making a slit in the dorsal portion of the dura mater, thecord was transected by a knife and then aspirated by suctionat about 2mm caudal and rostral to the level of transection, atL5-6 levels of the spinal cord for animals in group III, or atT1–T2 levels of the spinal cord for group IV animals. Gelfoamwas placed into the empty vertebral column to seal the emptyvertebral cavity and reduce bleeding. In a previous study20 andin our preliminary data, about 2½ hours after surgery, therabbits would have almost completely recovered from spinalshock. The animal would then be ready for the needling study.

Recording of Endplate Noise

Electromyography setting. For EPN assessment, a 2-chan-

Fig 2. Sites of EPN recording, dry needling for all animals, and areareceiving surgical transection for animals in groups II, III, and IV.

nel digital electromyogram machinec and monopolar needle d


electrodes (37-mm disposable Teflon-coated model) were used.The gain was set at 20�V per division for recordings from bothchannels. Low-cut frequency filter was set at 100Hz and thehigh-cut at 1000Hz. Sweep speed was 10 milliseconds perdivision. The search needle for EPN recording was insertedinto the MTrS region and connected to the first channel of theelectromyogram machine. The control needle was inserted intothe nontaut band region near the MTrS in the same muscle andconnected to the second channel. A common reference needleelectrode for each channel was placed on the incised skin andconnected to both channels via a y-connector.

Search for endplate noise. This procedure was performedby an investigator who was blind to the group assignment. Thesearch needle was inserted into the MTrS region in a directionparallel to the muscle fibers at an angle of approximately 60° tothe surface of the muscle. After initial insertion just short of thedepth of the MTrS or to a comparable depth in the case ofcontrol sites, the needle was advanced very slowly with simul-taneous slow rotation to prevent it from grabbing and releasingthe tissue suddenly to advance in a large jump. Each advancewas of minimal distance (�1mm). When the needle ap-proached an active locus (EPN locus), the continuous distantelectric activity—EPN—could be heard. As soon as EPN withamplitude higher than 10�V could be recorded, the examinertopped advancing the needle but minimally moved the needleently in different directions, trying to obtain the EPN withighest amplitude. If this was impossible, the needle wasdvanced to another site until an EPN with optimal amplitudeusually higher than 30�V) could be recorded. Then the needleas fixed in place (carefully and firmly taped on the skin) to

nsure that this EPN could run continuously on the recordingcreen with constant amplitudes. Continuous EPN tracing wasecorded throughout the entire course of the experiment androvided the opportunity for continuous visual observation ofPN changes on the electromyogram screen. If the EPN couldot be sustained, the searching needle would be moved tonother site until a satisfactory EPN tracing could be obtained.he entire EPN tracing found in MTrS of BF was recorded for

he analysis of amplitude changes.Measurement of the amplitude of endplate noise. Five

andomly selected samples of EPN recordings (10ms each)ere taken before, during, and 3 minutes after the completionf the needling treatment for all groups; before and 30 minutesfter surgery for group II animals; and every 30 minutes up to20 minutes after surgery for group III and IV animals. Theean amplitude of EPN of 5 random samples was analyzed and

alculated through the embedded software in the Neuro-EMG-icro equipment and was recorded as the value for a certaineasurement point for each animal.

ata AnalysisData of EPN amplitudes in different measurement points for

ifferent groups or subgroups were expressed as the mean �EM for further statistical analysis. The Shapiro-Wilk normal-

ty test was conducted to determine whether the data fit aormal distribution prior to subsequent analyses and showed alleasures of EPN amplitude were normally distributed. Tests of

omogeneity or baseline balance on covariates including bodyeight, age, and anesthesia condition were measured and found

o be equivalent before the needling treatment in all animals.he differences in EPN amplitude across measurementoints in each group were carried out using repeated-mea-ures ANOVA and later further analyzed by a Bonferroniost hoc method. The differences in EPN amplitudes withinach of the subgroups (ipsilateral dry needling, contralateral
ry needling, ipsilateral sham needling, contralateral sham

ER

dgnt

n

sigdfeltsoestsnsiidp

np

ET

l2teaPtnAidbsdpcde(

EL

sl


needling) and across measurement points (before, during,after needling) were analyzed using 2-way ANOVA (side �time) followed by a Bonferroni post hoc analysis for eachgroup. The differences in EPN amplitude within measure-ment point (before, during, after needling) across subgroups(ipsilateral dry needling, contralateral dry needling, ipsilat-eral sham needling, contralateral sham needling) were testedby paired t test. A P value less than .05 was considered to bestatistically significant. All data were analyzed using SPSSversion 10.0 for Windows.d

RESULTS

ffects of Dry Needling of Distal MTrS in Intactabbits (Group I)The serial alterations of the mean EPN amplitude before,

uring, and after dry needling at ipsilateral and contralateralastrocnemius for group I are demonstrated in figure 3. Beforeeedling treatment, there was no significant difference amonghe 4 subgroups (2-way ANOVA; F�.10; P�.05).

The mean amplitudes of EPN before, during, and aftereedling were 18.20�0.70�V, 27.71�0.47�V, and 13.15�

0.59�V, respectively, in the ipsilateral dry needling subgroupand 17.96�0.69�V, 24.66�1.47�V, and 14.01�0.86�V, re-pectively, in the contralateral dry needling subgroup. In thepsilateral dry needling and contralateral dry needling sub-roups, the amplitudes at different times were significantlyifferent (repeated-measures ANOVA; F�45.99 and P�.05or ipsilateral dry needling; F�113.98 and P�.05 for contralat-ral dry needling). Compared with the data in the preneedlingevel, the EPN amplitudes were significantly increased duringhe dry needling (Bonferroni post hoc test; P�.05) and thenignificantly decreased to a much lower level after completionf the needling treatment (Bonferroni post hoc test; P�.05) forither ipsilateral dry needling or contralateral dry needlingubgroup as shown in figure 3. However, these serial altera-ions of EPN amplitudes were not found in the comparableubgroup with ipsilateral sham needling or contralateral shameedling (repeated-measures ANOVA; P�.05). There wereignificant differences in EPN amplitudes recorded either dur-ng or after needling between the ipsilateral dry needling andpsilateral sham needling or between the contralateral dry nee-ling and contralateral sham needling subgroups (Bonferroniost hoc test; P�.05), but not between the ipsilateral dry

Fig 3. A series of changes in the EPN amplitude measured atgastrocnemius (GAS) in group I. (A) Time course of EPN amplitude.
howed significant differences among the 4 subgroups. *P<.05; showedevel in subgroups with dry needling at ipsilateral and contralateral GAS
eedling and contralateral dry needling subgroups (Bonferroniost hoc test; P�.05).

ffects of Dry Needling of Distal MTrS in Rabbits Withibial Nerve Transection (Group II)The serial alterations of the mean EPN amplitude � SEM

throughout the entire experiment in group II are demonstratedin figure 4. The mean amplitude of EPN had no significantchanges before, during, and after ipsilateral tibial nerve tran-section (ie, gastrocnemius denervation) (repeated-measuresANOVA; F�.06; P�.05). Before the needling treatment, therewas no significant difference in EPN amplitude among thesubgroups treated with dry or sham needling on the ipsilateralor contralateral side (2-way ANOVA; F�.68; P�.05).

The mean amplitudes � SEM of EPN recorded from BFbefore, during, and after needling were 16.72�0.34�V,16.64�0.37�V, and 15.46�0.50�V, respectively, in the ipsi-ateral dry needling subgroup and were 16.90�0.38�V,1.63�0.91�V, and 12.40�0.36�V, respectively, in the con-ralateral dry needling subgroup. There were significant differ-nces in EPN amplitudes among those recorded before, during,nd after needling (repeated-measures ANOVA; F�80.77;�.05) in the contralateral dry needling subgroup (similar to

he changes in group I; Bonferroni post hoc test; P�.05) butot in the ipsilateral dry needling subgroup (repeated-measuresNOVA; F�2.89; P�.05). There were significant differences

n mean EPN amplitudes between the contralateral dry nee-ling and comparable contralateral sham needling subgroups,ut not between the ipsilateral dry needling and ipsilateralham needling subgroups (Bonferroni post hoc test; P�.05)uring (Bonferroni post hoc test; P�.05) and after (Bonferroniost hoc test; P�.05) needling. Moreover, there were signifi-ant differences in the magnitude of EPN amplitude or time-ependent alterations of EPN amplitude between the contralat-ral dry needling and contralateral sham needling subgroupsBonferroni post hoc test; P�.05).

ffects of Dry Needling of Distal MTrS in Rabbits Withumbar Cord Transection (Group III)The serial alterations of the mean EPN amplitude � SEM

throughout the entire experiment for group III are shown infigure 5. There were significant differences in EPN amplitudeamong those recorded before, immediately after, and 30, 60,90, and 120 minutes after L5-6 transection (repeated-measures

of BF before, during, and after dry needling manipulation atample recordings of EPN responses in 2 rabbits of group I. †P<.05;

MTrS(B) S

the significant differences compared with the values at preneedling.


(m

a1

cFpsPd

ebndr

ET

tfiatrpsF

b

l2t

diffeoups


ANOVA; F�29.81; P�.05). During the 2-hour period aftertransection, the mean EPN amplitudes were significantly lowerthan the pretransected levels (Bonferroni post hoc test; all wereP �.05 at 30, 60, 90, and 120 minutes). There was no signif-icant difference among the 4 subgroups treated with dry orsham needling at the ipsilateral or contralateral side regardlessof the time before (2-way ANOVA; F�.23; P�.05), duringF�1.45; P�.05), or after (F�1.72; P�.05) needling treat-ents.The mean amplitudes � SEM of EPN before, during, and

fter needling were 11.56�0.36�V, 11.47�0.43�V, and1.28�0.47�V, respectively, in the ipsilateral dry needling

subgroup and 11.67�0.45�V, 12.32�0.46�V, and 12.33�0.46�V, respectively, in the contralateral dry needling sub-group. There was no significant difference in EPN amplitudeamong those recorded before, during, and after ipsilateraldry needling (repeated-measures ANOVA; F�.63; P�.05),ontralateral dry needling (repeated-measures ANOVA;�1.17; P�.05), ipsilateral sham needling treatment (re-eated-measures ANOVA; F�.23; P�.05), or contralateralham needling (repeated-measures ANOVA; F�.52;�.05). There were no significant differences in each time-ependent alteration of EPN amplitude between the ipsilat-

dry needling

Before During

Tibial nerve transection

Pre-surgery level

EP

N A

mpl

itude

(µV

)

*

*

Needling manipulation at GA

35

28

21

14

7

0 30 min

after surgery

A

Ipsilateral dry needling Contralateral dry needling Ipsilateral sham-needling Contralateral sham-needling

† †

Fig 4. A series of changes in the EPN amplitude measured at MTrS oand after dry needling manipulation at gastrocnemius (GAS) in groresponses in 2 rabbits from group II. †P<.05; showed significantdifferences compared with the values at preneedling level in subgr

Fig 5. A series of changes in the EPN amplitude measured at MTrSwell as before, during, and after dry needling manipulation at gastro
recordings of EPN responses in 2 rabbits from group III. #P<.05; showedlevels.

ral dry needling and ipsilateral sham needling subgroups,etween the ipsilateral dry needling and contralateral dryeedling subgroups, or between the contralateral dry nee-ling and contralateral sham needling subgroups (Bonfer-oni post hoc test; all P�.05).

ffects of Dry Needling of Distal MTrS in Rabbits Withhoracic Cord Transection (Group IV)The serial alterations of the mean EPN amplitude � SEM

hroughout the entire experiment for group IV are shown ingure 6. The EPN amplitude recorded at 30 and 60 minutesfter T1-2 transection was significantly reduced compared withhe pretransection level (Bonferroni post hoc test; P�.05) butecovered to pretransection level after 90 minutes (Bonferroniost hoc tests; all P�.05) and normalized to be similar for allubgroups by 120 minutes after transaction (2-way ANOVA;�.09; P�.05).The mean amplitudes � SEM of EPN recorded from BF

efore, during, and after needling were 18.17�0.36�V,26.88�0.43�V, and 15.74�0.26�V, respectively, in the ipsi-ateral dry needling subgroup and were 18.28�0.45�V,7.72�0.47�V, and 16.20�0.22�V, respectively, in the con-ralateral dry needling subgroup. There were significant differ-

er

B20µV

30 min after transection

During needling

After needling

Needling at ipsilateral GAS Needling at contralateral GAS

(ms) (ms)

Pre-surgery

0 2 4 6 8 10 0 2 4 6 8 10

Before needling

before and after tibial nerve transection surgery and before, during. (A) Time course of EPN amplitude. (B) Sample recordings of EPNrences among the 4 subgroups. *P<.05; showed the significantwith dry needling at ipsilateral and contralateral GAS.

before and 30 to 120 minutes after lumbar transection surgery asius (GAS) in group III. (A) Time course of EPN amplitude. (B) Sample

Aft

S

f BFup II

of BFcnem

significant differences compared with the values at pretransected

t

el inalues


ences in EPN amplitudes among those recorded before, during,and after ipsilateral dry needling (repeated-measures ANOVA;F�264.29; P�.05) or contralateral dry needling (repeated-measures ANOVA; F�243.11; P�.05). The mean EPN am-plitudes were significantly increased during ipsilateral dry nee-dling or contralateral dry needling (Bonferroni post hoc test;P�.05). After cessation of needling, they reduced signifi-cantly to a level less than that before needling (Bonferronipost hoc test; P�.05), but there were no significant differ-ences in these changes between the ipsilateral dry needling

Fig 6. A series of changes in the EPN amplitude measured at MTrSwell as before, during, and after dry needling manipulation at gastrocrecordings of EPN responses in 2 rabbits from group IV. †P<.05; shsignificant differences compared with the values at preneedling lev#P<.05; showed significant differences when compared with the v

Fig 7. Schematic drawing of the proposed neural mechanisms for re(1) Strong irritation to nociceptors in the MTrS by dry needling at gasensory neuron) probably in the MTrS circuit. (3) Ascending projectdorsal horn). (4) Impulse via interneuron to L5-6 motoneuron (aneuromuscular junction in the BF. (6) Increase in EPN amplitude. (7)increase recurrent inhibition on firing rate. (8) Suppression of effere
excitability of inhibitory interneuron can also be influenced by descendinpool and efferent as well as the irritability of MTrS at BF.
and contralateral dry needling subgroups (Bonferroni posthoc test; P�.05). These serial alterations of EPN amplitudeswere not found in either the ipsilateral sham needling (re-peated-measures ANOVA; F�1.02; P�.05) or contralateralsham needling subgroup (repeated-measures ANOVA;F�2.02; P�.05).

DISCUSSIONTo our knowledge, the present study is the first animal study

o investigate the neural mechanism of the remote effects of dry

before and 30 to 120 minutes after thoracic transection surgery asius (GAS) in group IV. (A) Time course of EPN amplitude. (B) Sampled significant differences among the 4 subgroups. *P<.05; showedsubgroups with dry needling at ipsilateral and contralateral GAS.at pretransected levels.

effect on proximal MTrS in response to dry needling at distal MTrS.nemius. (2) Afferent input from gastrocnemius to dorsal horn (L6-S2

upper (L5-6) sensory neurons probably in another MTrS circuit (inor horn) corresponding to BF. (5) Increase in efferent output tongly activated motoneuron also activates inhibitory interneuron totput from motoneuron. (9) Depression of EPN amplitude. (10) The

of BFnemowe

motestrocion tonteriStront ou

g inputs, thereby altering the overall excitability of the motoneuron


tcn

sp

pM


needling. In this study, we found that an intact afferent nervefrom the remote stimulation site and normal spinal cord seg-ments corresponding to the innervation of the affected proxi-mal muscle are essential for the remote effect from eitheripsilateral or contralateral stimulation.

The dry needling used in this study is a technique of MTrPinjection with multiple high-speed needle insertions into differentsensitive loci in an MTrP region suggested by Hong.25-27 High-speed needling can provide high-pressure stimulation to thesensitive loci in the MTrP region to elicit LTRs. It is essentialto elicit LTRs during needling of an MTrP in order to obtainimmediate and complete pain relief.6,25-29 Dry needling at theMTrS was effective in diminishing spontaneous electric activ-ity (ie, EPN) of MTrS of rabbit skeletal muscle if LTRs wereelicited.30 After several LTRs had been elicited by the needlingof an MTrS of rabbit skeletal muscle, no more LTRs could beelicited from the same region,18 and the irritability of the MTrScould be suppressed.30 Needling-elicited LTRs are involuntarydischarges of muscle fiber mediated through the nervous sys-tem and integrated at the spinal cord level.18,20 Therefore, it isimportant to apply this needling technique to achieve the bestneedling effect or remote needling effect for the study on theneural mechanism.

Changes in the EPN amplitude in the MTrS were foundduring and after dry needling at the distal MTrSs in animalswith intact neural circuits (group I). It appears that eitheripsilateral dry needling or contralateral dry needling to thedistal MTrS could initially increase the irritability of the prox-imal MTrS, followed by a suppression effect after cessation ofneedling. Fernandez-Carnero et al31 also found an increase inspontaneous electric activity at an MTrP region during a per-sistent noxious stimulation at another distant MTrP, followedby a suppression of electrophysiologic irritability after cessa-tion of needling. These findings strongly support clinical ob-servations related to the interaction between one MTrP andanother MTrP located in the region of the referred pain (re-ferred zone) of that MTrP.2,6,9-12,26-29

In the study on group II rabbits with ipsilateral denervationof gastrocnemius, the remote effect disappeared after ipsilateraldry needling but persisted after contralateral dry needling.These results demonstrated the importance of an intact afferentpathway to the spinal cord in the remote modulation of EPNamplitudes. After destruction of spinal cord corresponding tothe level of BF (group III), the remote effect disappeared aftereither ipsilateral dry needling or contralateral dry needlingtreatment. This finding suggested the existence of intraspinalconnections between the gastrocnemius afferents and BF spinalinterneurons. The partial suppression of the EPN amplitudeafter spinal cord transection at L5-6 is possibly related to theinfluence of the spinal shock. After interruption of upper motorneuron and suprasensory connections (group IV), the remoteeffect persisted either with ipsilateral dry needling or contralat-eral dry needling but was smaller than that in intact animals(group I). It may imply the possible influences from supraspi-nal centers, such as descending pain inhibitory systems. Theseinfluences are anticipated to be minimal after transection athigher spinal levels. Loss of inputs to this system wouldweaken the inhibition on the pain level,32 which could lead tohe EPN amplitude being less suppressed. Therefore, the re-ruitment of the DNIC system may be also elicited by dryeedling treatment on regions remote to the stimulation site.The neural pathway for the remote effect appears to be a

pinal reflex, probably similar to that mediating the referredain26,29 and local twitch response.18,20,26,29 Hong6,29,33 has

hypothesized a corresponding “MTrP circuit” for each MTrP
that can modulate the pain, referred pain, and local twitch

response elicited by stimulating the MTrP. The neural connec-tion in the spinal cord responsible for this remote effectivenessis probably similar to that for the referred pain.26,29

The initial increase in EPN with remote dry needling fol-lowed by a reduction in EPN after local twitch responses areelicited indicates inactivation of the remote MTrS. Strongstimulation from continuous dry needling of an MTrS canactivate the sensitized nociceptors and generate strong im-pulses propagating to the spinal cord to activate the corre-sponding motoneurons (including those in the same segmentcorresponding to the needling muscle and other segments cor-responding to the remote muscles) to fire reflexively, therebycausing increased EPN amplitude in MTrSs not only at theneedling muscle but also at other remote muscles. Thesestrongly activated motoneurons are also controlled by recurrentinhibitions. As the firing rate of motoneurons increases, theamount of recurrent inhibitions will also increase, subsequentlylimiting and suppressing the firing rate of the efferents. In thisway, these impulses elicited by dry needling eventually breakthe vicious cycle of the neural circuits (ie, MTrP circuits29,33)responsible for MTrSs through spatial and temporal interac-tions in the spinal cord. Thus the EPN amplitude is suppressedafter dry needling. Possibly there are certain neural connectionsamong the inhibitory interneuron and descending pain controlsystem in the spinal cord that can modulate the irritability ofMTrPs when a remote painful dry needling stimulation isapplied (fig 7). The initial increase in EPN is consistent withsuppression of the DNIC system, and the subsequent reductionin EPN is consistent with activation or enhancement of theDNIC system. Therefore, the physiologic basis for the remoteeffects of dry needling may be related to an inactivation ofMTrS29,33 and activation of DNIC34,35 induced by noxiousstimulation applied at the painful region (such as trigger pointneedling) or at a remote site (such as in remote dry needling).This is probably the mechanism of remote pain control by dryneedling, which is similar to hyperstimulation analgesia inacupuncture.32,36

Study LimitationsThe difficulty in confirming the correlation between the

alterations of EPN amplitude and pain intensity in rabbits maybe criticized. However, a conclusion based on the humanstudy4 may be reasonably applied to rabbits because there arelenty of similarities between the human MTrP and rabbitTrS.18,26,28,29 Lack of follow-up assessments for the long-

term remote effect is another deficiency of this study. However,we rarely see the long-term effects of dry needling if theunderlying pathology of MTrP activation is not treated appro-priately.6,26,27,28,33 Another problem is that our sham needling(similar to superficial dry needling) may not be appropriate asa control. Fortunately, we see no significant changes aftertreatment with sham needling. However, this could be relatedto the small sample size. In our clinical practice, we haveobserved much less effectiveness of superficial needling thanthat of deep dry needling with our multiple quick insertiontechnique. In addition, considering the individual differences inthe motoneuron excitability and the supraspinal control ofspinal inhibitory interneurons, based on electrophysiologicstudy, we are unable to distinguish the relative contributionfrom each inhibitory mechanism for motoneuronal excitabilityto the changes in MTrP irritability caused by remote dryneedling. All these factors should be taken into consideration
for data interpretation.

ucmmec


CONCLUSIONSWe have demonstrated that an intact afferent from the stim-

lating site to the spinal cord and a normal function of spinalord corresponding to the innervation of the remotely affecteduscles are essential for this remote effectiveness. This studyay help in the understanding of the mechanism for beneficial

ffects of dry needling at remote MTrPs for myofascial painontrol.

Acknowledgments: We thank Ms Pin-Wen Tu, BS, and Pei-Ling Chang, BS, for their skillful technical assistance with the elec-tromyographic recordings and data analyses.

References1. Simons DG. Review of enigmatic MTrPs as a common cause of

enigmatic musculoskeletal pain and dysfunction. J ElectromyogrKinesiol 2004;14:95-107.

2. Simons DG, Travell JG, Simons LS. Travell & Simons’s myofas-cial pain and dysfunction: the trigger point manual. Vol 1. 2nd ed.Baltimore: Williams & Wilkins; 1999.

3. Kuan TS, Hsieh YL, Chen SM, Chen JT, Yen WC, Hong CZ. Themyofascial trigger point region: correlation between the degree ofirritability and the prevalence of endplate noise. Am J Phys MedRehabil 2007;86:183-9.

4. Chou LW, Hsieh YL, Kao MJ, Hong CZ. Remote influences ofacupuncture on the pain intensity and the amplitude changes ofendplate noise in the myofascial trigger point of the upper trape-zius muscle. Arch Phys Med Rehabil 2009;90:905-12.

5. Simons DG, Hong CZ, Simons LS. Endplate potentials are com-mon to midfiber myofascial trigger points. Am J Phys Med Re-habil 2002;81:212-22.

6. Hong CZ. Treatment of myofascial pain syndrome. Curr PainHeadache Rep 2006;10:345-9.

7. Fernandez-Carnero J, La Touche R, Ortega-Santiago R, et al.Short-term effects of dry needling of active myofascial triggerpoints in the masseter muscle in patients with temporomandibulardisorders. J Orofac Pain 2010;24:106-12.

8. Srbely JZ, Dickey JP, Lee D, Lowerison M. Dry needle stimula-tion of myofascial trigger points evokes segmental anti-nociceptive effects. J Rehabil Med 2010;42:463-8.

9. Hsieh YL, Kao MJ, Kuan TS, Chen SM, Chen JT, Hong CZ. Dryneedling to a key myofascial trigger point may reduce the irrita-bility of satellite MTrPs. Am J Phys Med Rehabil 2007;86:397-403.

10. Tseng CL, Kao MJ, Chou LW, Hong CZ. Injection of remotemyofascial trigger points for pain control: a case report. TaiwanJournal Phys Med Rehabil 2008;36:53-8.

11. Tsai CT, Hsieh LF, Kuan TS, Kao MJ, Chou LW, Hong CZ.Remote effects of dry needling on the irritability of the myofascialtrigger point in the upper trapezius muscle. Am J Phys MedRehabil 2010;89:133-40.

12. Hong CZ, Simons DG. Response to treatment for pectoralis minormyofascial pain syndrome after whiplash. J Musculoskelet Pain1993;1:89-131.

13. Rho SW, Choi GS, Ko EJ, et al. Molecular changes in remotetissues induced by electro-acupuncture stimulation at acupointST36. Mol Cells 2008;25:178-83.

14. Carlsson C. Acupuncture mechanisms for clinically relevant long-term effects-reconsideration and a hypothesis. Acupunct Med2002;20:82-99.

15. Miura K, Ohara T, Zeredo JL, Okada Y, Toda K, Sumikawa K.Effects of traditional “Juci” (contralateral acupuncture) on oro-fascial nociceptive behavior in the rat. J Anesth 2007;21:31-6.

16. Sato A, Sato Y, Uchida S. Reflex modulation of visceral functionsby acupuncture-like stimulation in anesthetized rats. Int Congr Ser
2002;1238:111-23.
17. Chen KH, Hong CZ, Kuo FC, Hsu HC, Hsieh YL. Electrophysi-ologic effects of a therapeutic laser on myofascial trigger spots ofrabbit skeletal muscles. Am J Phys Med Rehabil 2008;87:1006-14.

18. Hong CZ, Torigoe Y. Electrophysiologic characteristics of local-ized twitch responses in responsive bands of rabbit skeletal musclefibers. J Musculoskelet Pain 1994;2:17-43.

19. Zimmermann M. Ethical guidelines for investigations of experi-mental pain in conscious animals. Pain 1983;16:109-10.

20. Hong CZ, Torigoe Y, Yu J. The localized twitch responses inresponsive bands of rabbit skeletal muscle fibers are related to thereflexes at spinal cord level. J Musculoskelet Pain 1995;3:15-33.

21. Hong CZ, Yu J. Spontaneous electrical activity of rabbit triggerspot after transection of spinal cord and peripheral nerve. J Mus-culoskelet Pain 1998;6:45-58.

22. Kuan TS, Chen JT, Chen SM, Chien CH, Hong CZ. Effect ofbotulinum toxin on endplate noise in myofascial trigger spots ofrabbit skeletal muscle. Am J Phys Med Rehabil 2002;81:512-20.

23. Simons DG, Hong CZ, Simons LS. Prevalence of spontaneouselectrical activity at trigger spots and at control sites in rabbitskeletal muscle. J Musculoskelet Pain 1995;3:35-48.

24. Wood PL. Animal models in analgesic testing. In: Kuhar MJ,Pasternak GW, editors. Central nervous system pharmacology:analgesics: neurochemical, behavioral and clinical perspective.New York: Raven Pr; 1984. p 175-94.

25. Hong CZ. Lidocaine injection versus dry needling to myofascialtrigger point: the importance of the local twitch response. Am JPhys Med Rehabil 1994;73:256-63.

26. Hong CZ. Myofascial trigger points: pathophysiology and corre-lation with acupuncture points. Acupunct Med 2000;18:41-7.

27. Hong CZ. Consideration and recommendation of myofascial trig-ger point injection. J Musculoskelet Pain 1994;2:29-59.

28. Hong CZ, Simons DG. Pathophysiologic and electrophysiologicmechanisms of myofascial trigger points. Arch Phys Med Rehabil1998;79:863-72.

29. Hong CZ. Research on myofascial pain syndrome. Crit Rev PhysRehabil Med 2008;20:343-66.

30. Chen JT, Chung KC, Hou CR, Kuan TS, Chen SM, Hong CZ.Inhibitory effect of dry needling on the spontaneous electricalactivity recorded from myofascial trigger spots of rabbit skeletalmuscle. Am J Phys Med Rehabil 2001;80:729-35.

31. Fernandez-Carnero J, Ge HY, Kimura Y, Fernandez-de-Las-PenasC, Arendt-Nielsen L. Increased spontaneous electrical activity at alatent myofascial trigger point after nociceptive stimulation ofanother latent trigger point. Clin J Pain 2010;26:138-43.

32. Melzack R. Myofascial trigger points: relation to acupuncture andmechanisms of pain. Arch Phys Med Rehabil 1981;62:114-7.

33. Hong CZ. Myofascial pain therapy. J Musculoskelet Pain 2004;12:37-43.

34. Reinert A, Treede R, Bromm B. The pain inhibiting pain effect: anelectrophysiological study in humans. Brain Res 2000;862:103-10.

35. Murase K, Kawakita K. Diffuse noxious inhibitory controls inanti-nociception produced by acupuncture and moxibustion ontrigeminal caudalis neurons in rats. Jpn J Physiol 2000;50:133-40.

36. Zhao ZQ. Neural mechanism underlying acupuncture analgesia.Prog Neurobiol 2008;85:355-75.

Suppliersa. Physitemp Instruments, Inc, 154 Huron Ave, Clifton, NJ 07013.b. Yu-Kuang Industrial Co, Ltd, 5F-6, No. 20, Lane 609, Sec 5,

Chung-shing Rd, San-Chung City, Taipei, Taiwan.c. Neuro-EMG-Micro, Neurosoft, 5, Voronin Str, Ivanovo, Russia.
d. SPSS Inc, 233 S Wacker Dr, 11th Fl, Chicago, IL 60606.

Documents

3. Dry Needling_Spinal Cord Mechanism.pdf