differential effects of subcutaneous electrical stimulation (sqs) and transcutaneous electrical...

8
Differential Effects of Subcutaneous Electrical Stimulation (SQS) and Transcutaneous Electrical Nerve Stimulation (TENS) in Rodent Models of Chronic Neuropathic or Inflammatory Pain Louis P. Vera-Portocarrero, PhD*; Toni Cordero, PhD*; Tina Billstrom, MSOT ; Kim Swearingen, AAS ; Paul W. Wacnik, PhD*; Lisa M. Johanek, PhD* Objectives: Electrical stimulation has been used for many years for the treatment of pain. Present-day research demonstrates that stimulation targets and parameters impact the induction of specific pain-modulating mechanisms. New targets are increasingly being investigated clinically, but the scientific rationale for a particular target is often not well established. This present study compares the behavioral effects of targeting peripheral axons by electrode placement in the subcutaneous space vs. electrode placement on the surface of the skin in a rodent model. Materials and Methods: Rodent models of inflammatory and neuropathic pain were used to investigate subcutaneous electrical stimulation (SQS) vs. transcutaneous electrical nerve stimulation (TENS). Electrical parameters and relative location of the leads were held constant under each condition. Results: SQS had cumulative antihypersensitivity effects in both inflammatory and neuropathic pain rodent models, with signifi- cant inhibition of mechanical hypersensitivity observed on days 3–4 of treatment. In contrast, reduction of thermal hyperalgesia in the inflammatory model was observed during the first four days of treatment with SQS, and reduction of cold allodynia in the neuropathic pain model was seen only on the first day with SQS. TENS was effective in the inflammation model, and in agreement with previous studies, tolerance developed to the antihypersensitivity effects of TENS. With the exception of a reversal of cold hypersensitivity on day 1 of testing, TENS did not reveal significant analgesic effects in the neuropathic pain rodent model. Conclusions: The results presented show that TENS and SQS have different effects that could point to unique biologic mecha- nisms underlying the analgesic effect of each therapy. Furthermore, this study is the first to demonstrate in an animal model that SQS attenuates neuropathic and inflammatory-induced pain behaviors. Keywords: Inflammatory pain, neuropathic pain, peripheral nerve stimulation, transcutaneous electrical nerve stimulation Conflict of Interest: All authors are employees of Medtronic Inc. INTRODUCTION The advent of the gate control theory of pain (1) gave rise to the use of electricity-evoked Ab fiber stimulation to abolish pain (2). The highly myelinated large Ab fibers transmit sensory information, such as touch and vibration, to the central nervous system and can be electrically activated at a number of sites. A common site of activa- tion is the central projection of the axon within the dorsal columns. The resulting therapy of targeting the dorsal columns for the relief of chronic pain eventually became known as spinal cord stimulation (SCS) (3). Mechanisms of analgesia in the central nervous system may also be activated at the peripheral axons. The targeting of peripheral nerves has been generally termed peripheral nerve stimulation (PNS) (4,5). Typically, PNS has implied stimulation of a nerve trunk; however, sensory neurons can also be activated at their distal termi- nals. For example, transcutaneous electrical nerve stimulation (TENS) applies electrical pulses through the skin in order to stimulate the peripheral endings of the nervous system. These same Ab sensory nerve-ending branches are the likely targets of electrical stimulation electrodes placed in the subcutaneous space. The method of deliv- ering electrical stimulation to the peripheral nerve endings by place- ment of electrodes in the subcutaneous space will be called subcutaneous electrical stimulation (SQS) in this report. Because all of these stimulation modalities focus on activation of large Ab fibers, it might be assumed that they would all have similar mechanisms and thus, similar therapeutic benefits. However, Address correspondence to: Louis P. Vera-Portocarrero, PhD, Neuromodulation Research, Medtronic, Inc., 7000 Central Ave NE MS RCE470, Minneapolis, MN 55432, USA. Email: [email protected] * Neuromodulation Research, Medtronic, Inc., Minneapolis, MN, USA; and Physiological Research Laboratories, Medtronic, Inc., Minneapolis, MN, USA For more information on author guidelines, an explanation of our peer review process, and conflict of interest informed consent policies, please go to http:// www.wiley.com/bw/submit.asp?ref=1094-7159&site=1 Source of Financial support: Medtronic Inc. Neuromodulation: Technology at the Neural Interface Received: October 12, 2012 Revised: December 5, 2012 Accepted: January 3, 2013 (onlinelibrary.wiley.com) DOI: 10.1111/ner.12037 328 www.neuromodulationjournal.com Neuromodulation 2013; 16: 328–335 © 2013 Medtronic, Inc.

Upload: lisa-m

Post on 29-Mar-2017

214 views

Category:

Documents


1 download

TRANSCRIPT

Page 1: Differential Effects of Subcutaneous Electrical Stimulation (SQS) and Transcutaneous Electrical Nerve Stimulation (TENS) in Rodent Models of Chronic Neuropathic or Inflammatory Pain

Differential Effects of Subcutaneous ElectricalStimulation (SQS) and Transcutaneous ElectricalNerve Stimulation (TENS) in Rodent Models ofChronic Neuropathic or Inflammatory PainLouis P. Vera-Portocarrero, PhD*; Toni Cordero, PhD*; Tina Billstrom, MSOT†;Kim Swearingen, AAS†; Paul W. Wacnik, PhD*; Lisa M. Johanek, PhD*

Objectives: Electrical stimulation has been used for many years for the treatment of pain. Present-day research demonstrates thatstimulation targets and parameters impact the induction of specific pain-modulating mechanisms. New targets are increasinglybeing investigated clinically, but the scientific rationale for a particular target is often not well established. This present studycompares the behavioral effects of targeting peripheral axons by electrode placement in the subcutaneous space vs. electrodeplacement on the surface of the skin in a rodent model.

Materials and Methods: Rodent models of inflammatory and neuropathic pain were used to investigate subcutaneous electricalstimulation (SQS) vs. transcutaneous electrical nerve stimulation (TENS). Electrical parameters and relative location of the leadswere held constant under each condition.

Results: SQS had cumulative antihypersensitivity effects in both inflammatory and neuropathic pain rodent models, with signifi-cant inhibition of mechanical hypersensitivity observed on days 3–4 of treatment. In contrast, reduction of thermal hyperalgesiain the inflammatory model was observed during the first four days of treatment with SQS, and reduction of cold allodynia in theneuropathic pain model was seen only on the first day with SQS. TENS was effective in the inflammation model, and in agreementwith previous studies, tolerance developed to the antihypersensitivity effects of TENS. With the exception of a reversal of coldhypersensitivity on day 1 of testing, TENS did not reveal significant analgesic effects in the neuropathic pain rodent model.

Conclusions: The results presented show that TENS and SQS have different effects that could point to unique biologic mecha-nisms underlying the analgesic effect of each therapy. Furthermore, this study is the first to demonstrate in an animal model thatSQS attenuates neuropathic and inflammatory-induced pain behaviors.

Keywords: Inflammatory pain, neuropathic pain, peripheral nerve stimulation, transcutaneous electrical nerve stimulation

Conflict of Interest: All authors are employees of Medtronic Inc.

INTRODUCTION

The advent of the gate control theory of pain (1) gave rise to theuse of electricity-evoked Ab fiber stimulation to abolish pain (2). Thehighly myelinated large Ab fibers transmit sensory information, suchas touch and vibration, to the central nervous system and can beelectrically activated at a number of sites. A common site of activa-tion is the central projection of the axon within the dorsal columns.The resulting therapy of targeting the dorsal columns for the relief ofchronic pain eventually became known as spinal cord stimulation(SCS) (3). Mechanisms of analgesia in the central nervous system mayalso be activated at the peripheral axons. The targeting of peripheralnerves has been generally termed peripheral nerve stimulation (PNS)(4,5). Typically, PNS has implied stimulation of a nerve trunk;however, sensory neurons can also be activated at their distal termi-nals. For example, transcutaneous electrical nerve stimulation (TENS)applies electrical pulses through the skin in order to stimulate theperipheral endings of the nervous system. These same Ab sensorynerve-ending branches are the likely targets of electrical stimulation

electrodes placed in the subcutaneous space. The method of deliv-ering electrical stimulation to the peripheral nerve endings by place-ment of electrodes in the subcutaneous space will be calledsubcutaneous electrical stimulation (SQS) in this report.

Because all of these stimulation modalities focus on activation oflarge Ab fibers, it might be assumed that they would all have similarmechanisms and thus, similar therapeutic benefits. However,

Address correspondence to: Louis P. Vera-Portocarrero, PhD, NeuromodulationResearch, Medtronic, Inc., 7000 Central Ave NE MS RCE470, Minneapolis,MN 55432, USA. Email: [email protected]

* Neuromodulation Research, Medtronic, Inc., Minneapolis, MN, USA; and† Physiological Research Laboratories, Medtronic, Inc., Minneapolis, MN, USA

For more information on author guidelines, an explanation of our peer reviewprocess, and conflict of interest informed consent policies, please go to http://www.wiley.com/bw/submit.asp?ref=1094-7159&site=1Source of Financial support: Medtronic Inc.

Neuromodulation: Technology at the Neural Interface

Received: October 12, 2012 Revised: December 5, 2012 Accepted: January 3, 2013

(onlinelibrary.wiley.com) DOI: 10.1111/ner.12037

328

www.neuromodulationjournal.com Neuromodulation 2013; 16: 328–335© 2013 Medtronic, Inc.

Page 2: Differential Effects of Subcutaneous Electrical Stimulation (SQS) and Transcutaneous Electrical Nerve Stimulation (TENS) in Rodent Models of Chronic Neuropathic or Inflammatory Pain

unique mechanistic differences may exist based on 1) the number ofafferents activated; 2) the relative proportion of non-nociceptive(Ab) and nociceptive (C and Ad) sensory fibers activated; and 3) theimpact of electrical stimulation on nonsensory or non-neuronal celltypes (e.g., keratinocytes and endothelial cells). Extensive researchhas been done to investigate the mechanisms of electro-analgesiaincluding electroacupuncture (EA), which is a therapy that com-bines the use of acupuncture points and needles that conduct elec-trical current to produce analgesic effects in human subjects (6) andexperimental animals (7–9). These studies show that there are keysimilarities between TENS and EA, mainly in the engagement of theopioid system when applying stimulation (10–13). However, somedifferences can be identified between TENS, EA, and SCS. Forexample, TENS and EA applied at the superficial skin layers mayactivate small diameter nociceptors as well as large mechanorecep-tive fibers (14–16). In contrast, SCS targeted over the dorsal columnswould primarily activate non-nociceptive Ab fibers (17,18). Dueto the anatomy in the subdermal layers of the skin, SQS likely acti-vates Ab fibers at low intensities but nociceptive fibers at higherintensities.

Mechanisms of PNS targeting a nerve trunk have been minimallystudied. The available preclinical evidence has been centered onusing conditioning stimulation with high amplitudes at C-fiberstrength (19,20) employing acute models of pain (21,22). Mecha-nisms engaged by lower stimulation at Ab fiber strength in chronicpain models have not been studied. Although preliminary researchhas been done to investigate the effects of PNS on chronic modelsof pain (23), no animal model of SQS exists, and definitive mecha-nisms for pain relief are unknown.

The present study is the first report on a rodent model of SQS inboth neuropathic and inflammatory pain models. Moreover, SQSwas compared with TENS in both persistent pain rodent models.Although similarities were anticipated, we hypothesized that SQSwould produce more robust antihyperalgesia than TENS in the neu-ropathic pain model and would show less tolerance than TENS inthe model of inflammation.

Therefore, the present study took the first step in establishingmechanisms of SQS by observing the differential effects of TENS andSQS on two rodent models of persistent pain.

METHODSAnimals and Treatments

Male Sprague Dawley rats (N = 128) weighing 250–300 g (CharlesRiver, Kingston, NY, USA) were used, and all protocols for the experi-ments were approved by the Institutional Animal Care and UseCommittee of the Physiological Research Laboratory, Medtronic,Inc. All rats were housed at a room temperature of 22°C on a 12-hourlight/dark cycle (lights on at 6:00 AM) with free access to food andwater. Rats either received a nerve injury or an injection of carrag-eenan into the calf muscle in order to induce neuropathic andinflammatory pain, respectively. Rats assigned to receive SQStherapy were also implanted with electrode leads in the subcutane-ous layers at the time of nerve injury or injection. TENS electrodeswere placed in an acute manner on the day of stimulation; therefore,there was no need of long-term implantation of TENS electrodes.

Pain Models

Spared Nerve Injury (SNI)Under isofluorane anesthesia, an incision was made along the

long axis of the left hindlimb to expose the sciatic nerve and its

three branches (sural, peroneal, and tibial nerves). A piece of sterile4-0 silk was placed around the common peroneal and tibial nerve.The suture was drawn tightly (ligated) around the common pero-neal and tibial nerves below the bifurcation. In addition, the nerveswere transected 2 mm distal to the nerve ligation, and 2–3 mm ofthe nerve from both branches was removed (24). Sham animals hadtheir nerve exposed, and suture was placed under and around thenerve; however, it was not drawn tightly, and the nerves were notcut. The muscle and skin were then sutured closed. Rats wereinjected with a solution containing 5% dextrose in 0.9% NaCl intra-peritoneally at the end of surgery and were allowed to recover fromanesthesia in a warmed and dedicated recovery area.

Muscle InflammationRats were injected with 0.2 mL of a mixture of 3% kaolin and 3%

carrageenan (Sigma-Aldrich, St Louis MO, USA) under 2–3% isofluo-rane anesthesia. A 26G needle was used for the procedure that tookabout two minutes per rat. The injection was given in the center(belly) of the gastronecmious muscle (calf ). Sham animals receivedan equal volume injection with sterile saline.

Implantation of the SQS LeadsAt the end of the SNI procedure or the carrageenan injection, a

small incision was made to the dorsal midline of the rat. A localanalgesic, Marcaine (Hospira Inc. Lake Forest IL, USA), diluted insaline was applied to the cut skin. An electrode lead (Pisces compactModel #3887, Medtronic, Inc., Minneapolis, MN, USA) was insertedat the midline incision and tunneled under the skin using an angio-catheter as a tunneling tool. The electrode was a four-contact cylin-drical lead with a length of 3 mm for each contact and a spacing of4 mm between contacts. The lead was tunneled until the electrodearray section was lying over the long axis of the lateral section of thehindlimb. The lead was left lying above the incision made either inthe muscle to access the sciatic nerves, or overlying the belly of thecalf muscle. The electrode array was secured to the underlying fasciaand muscle using sutures so that the electrodes were touching theunderlying muscle. The opposite side of the lead (the connectorside) was tunneled in the opposite direction so it could be external-ized above the shoulder blades and secured at the back of the neck.

Behavioral TestingAfter acclimation to a testing compartment with clear plastic

walls and a mesh bottom, mechanical withdrawal thresholds weretested by applying calibrated Von Frey filaments (0.41. 0.70, 1.20,2.00, 3.63, 5.50, 8.5, and 15.1 g) perpendicularly to the lateral aspectof the plantar side of the hind paw until the filament bent slightly. Asharp withdrawal of the hindpaw indicated a positive response. Thestimulus was incrementally increased until a positive response wasobtained, then decreased until a negative result (no paw with-drawal) was observed in order to determine a pattern of responses(the up-down method [25]). The final threshold was determinedfrom the final pattern of responses using the Dixon nonparametrictest.

For cold allodynia testing, a volume of 50 mL of acetone wasapplied onto the lateral surface of the paw using a syringe cappedwith a blunt needle. The time the rat spent presenting pain behav-iors (shaking, licking, or guarding of the paw) during a 20-secondtesting period was recorded.

For heat hyperalgesia testing, animals were tested with a beam ofradiant heat applied to the medial lower aspect of the plantar hindpaw. The maximum time of exposure to the radiant heat was 20seconds to ensure no injury to tissue. The latency in seconds that therat took to withdraw its paw from the radiant heat was recorded.Rats were tested three times with an interval of five minutes

329EFFECTS OF TENS VS. SQS IN RODENT MODELS OF PAIN

www.neuromodulationjournal.com Neuromodulation 2013; 16: 328–335© 2013 Medtronic, Inc.

Page 3: Differential Effects of Subcutaneous Electrical Stimulation (SQS) and Transcutaneous Electrical Nerve Stimulation (TENS) in Rodent Models of Chronic Neuropathic or Inflammatory Pain

between tests, and the average of the latencies obtained from thelast two tests was used for analysis and reporting.

Due to the externalized proximal end of the SQS lead, the experi-menter performing the behavior testing could not be blinded to thetreatment modality (SQS or TENS). However, the experimenter wasblind to the condition of the rodent (sham, inflammation, or nerveligation) and the treatment (0 Hz or 100 Hz).

Stimulation ParadigmRats were placed under isofluorane anesthesia (2–3%) after base-

line behavioral testing. For rats receiving SQS, the distal end of theSQS electrodes was connected using an extension kit (Model#37081-20, Medtronic, Inc.) to a commercially available externalneurostimulator (ENS, Model #37022), which was previously pro-grammed using an N’Vision Physician Programmer (Model #8840).Only the most distal pair of electrodes was used for a bipolar con-figuration (+-). Stimulation was turned on, and the amplitude wasincreased until a motor contraction of the muscles surrounding theelectrode array to determine motor threshold (MT). The parametersused for stimulation were 90% of MT, pulse width of 250 msec, andthe frequency was either 0 Hz or 100 Hz. Stimulation was given for20 min and rats were allowed to recover from anesthesia for 20 min.Rats receiving TENS had TENS electrodes placed on the shaved skinsurface overlying the gastronecmious muscle and kept in place withelectrical tape. The electrodes were connected to a TENS unit (TENS7000, LGMed supply, Cherry Hill NJ, USA). Amplitude was increaseduntil a muscle contraction was observed. This amplitude was con-sidered MT. The amplitude was set at 90% of MT, pulse width at250 msec, and a frequency of 100 Hz. Rats receiving sham (0 Hz)stimulation under anesthesia had TENS electrodes placed on theskin surface, but the TENS unit was kept off. Stimulation was givenfor 20 min, and rats were allowed to recover from anesthesia for20 min. At the end of the recovery period, rats were transportedback to the behavioral monitoring room to be tested again formechanical and thermal sensitivity. This stimulation paradigmwhere MTs were assessed at the beginning of the stimulationsession while rats were anesthetized was repeated for each day ofstimulation.

Treatment Schedule

Inflammatory Pain Model Animals (N = 64) underwent baselinebehavior testing prior to carrageenan injection. The injection andlead implantation (for SQS-treated rodents only) were performed onday 0 and subsequent behavior testing began on day 3. Rats weretested before and after electrical stimulation for five consecutivedays (Fig. 1a).

Neuropathic Pain Model Animals (N = 64) underwent baseline behav-ior testing prior to surgery. SNI and lead implantation (for SQS-treated rodents only) were performed on day 0 and subsequentbehavior testing began on day 7 after surgery. Rats were testedbefore and after electrical stimulation for five consecutive days(Fig. 1b).

Statistical AnalysisControl data were analyzed and compared with analysis of vari-

ance (ANOVA) with repeated measures and Dunnet’s test as a posthoc. Data from the experimental groups (except for data from coldtesting) were normalized as percent of preinjury baseline and wascompared using ANOVA with Student’s I-test to test for significance.Data from cold testing were normalized as a percent of the durationof the response after injury. Significance was set at p < 0.05.All analyses were done with Prism 4 statistical software (GraphPadSoftware Inc., San Diego, CA, USA).

RESULTSEffects of TENS vs. SQS on Inflammatory Pain

Rats receiving intramuscular injection of kaolin/carrageenandeveloped mechanical allodynia and thermal hyperalgesia startingon day 3 postinjection. The average mechanical withdrawal thresh-old for rats injected with carrageenan was 4.5 � 0.4 g, which wassignificantly different from the preinjury baseline value of 14.0 �0.4 g indicating a reduction to 32.0 � 10.5% of baseline (p < 0.05,Fig. 2a). Rats received stimulation with TENS or SQS for five consecu-tive days at frequencies of either 0 HZ or 100 Hz. Rats receiving TENS

a

b

Preinjury

testing

Preinjury

testing

Carrageenan injection +

SQS lead implantation

3 days

Baseline testing–stimulation 20 min-testing

Day 3Day 1 Day 2 Day 4 Day 5

SNI surgery and

SQS lead implantation

7 days

Baseline testing–stimulation 20 min-testing

Day 3Day 1 Day 5

Carrageenan

injection

for TENS rats

SNI surgery

for TENS rats Day 2 Day 4

Figure 1. a. Schematic representation of the sequence of events with the inflammatory pain model. b. The neuropathic pain model. Only rats allocated to SQSgroups received permanent lead implantation. Rats allocated to TENS received either carrageenan injection or SNI surgery alone. Baseline testing, 20 min ofstimulation, and poststimulation testing were performed on each of the five days.

330

VERA-PORTOCARRERO ET AL.

www.neuromodulationjournal.com Neuromodulation 2013; 16: 328–335© 2013 Medtronic, Inc.

Page 4: Differential Effects of Subcutaneous Electrical Stimulation (SQS) and Transcutaneous Electrical Nerve Stimulation (TENS) in Rodent Models of Chronic Neuropathic or Inflammatory Pain

at 0 Hz demonstrated mechanical withdrawal thresholds that were25.9 � 3.4%, 38.8 � 8.5%, 28.4 � 4.6%, 35.5 � 10.2%, and 38.5 � 7%(days 1–5, respectively) of preinjury baseline (Fig. 2a). These valuesare not statistically different than the prestimulation values. Thisdemonstrates that the use of isofluorane anesthesia and applicationof sham stimulation did not have an effect on mechanical allodynia.In contrast, TENS at 100 Hz demonstrated a significant anti-allodyniceffect for the first three days of stimulation compared with pre-stimulation values. Mechanical withdrawal thresholds were 91.5 �9.5%, 77.7 � 7.5%, 83.4 � 10.9%, 76.6 � 13.1%, and 44.3 � 6.1%(days 1–5, respectively) relative to preinjury baseline (Fig. 2a).

Similarly to TENS at 0 Hz stimulation, SQS at 0 Hz did not havesignificant effects on the mechanical allodynia observed in rats withcarrageenan injection. Rats with 0 Hz SQS had mechanical with-drawal thresholds of 22.8 � 4.5%, 35.8 � 3.6%, 35.1 � 9.8, 33.4 �6.4%, and 31.6 � 6.2% (days 1–5, respectively) relative to preinjurybaseline (Fig. 3a). These values are not statistically different than theprestimulation value obtained for mechanical allodynia in the ratmodel of inflammatory pain. In contrast to TENS, SQS at 100 Hz hada significant effect starting on day 3 of treatment and lastingthrough day 5 as compared with prestimulation baselines. Mechani-cal withdrawal thresholds were 33.9 � 9.2%, 47.1 � 12.1%, 72.1 �10.4%, 68.4 � 14%, and 77.3 � 13.7% (days 1–5, respectively) rela-tive to preinjury baseline (Fig. 2a).

Rats injected with intramuscular carrageenan also developedthermal hyperalgesia starting on day 3 postinjection. The averagethermal latency of rats with carrageenan was 8.7 � 0.9 sec, whichwas significantly different from the preinjury baseline value of 14.5� 1.0 sec indicating a reduction to 59.8 � 5.3% of preinjury baseline

(p < 0.05, Fig. 2b). Rats receiving TENS at 0 Hz did not show anysignificant changes in their withdrawal latencies, compared withprestimulation values, remaining in a hyperalgesic state. The with-drawal latencies were 49.6 � 3.6%, 55.6 � 3.3%, 54.4 � 1.5%, 59.4 �4.0%, and 56.6 � 4.7% (days 1–5, respectively) relative to preinjurybaseline (Fig. 2b). TENS at 100 Hz had an antihyperalgesic effect thatreached statistical significance for the first four days and was mostpronounced on the first three days of treatment. Withdrawal laten-cies were 113.0 � 6.4%, 107.3 � 8.8%, 99.7 � 10.1%, 72.8 � 2.7%,and 68.7 � 10.5% (days 1–5, respectively) relative to preinjury base-line (Fig. 2b).

SQS at 0 Hz did not show any significant changes in withdrawallatencies to heat, compared with prestimulation baselines, remain-ing in a hyperalgesic state. The withdrawal latencies were 62.6 �2.8%, 54.4 � 3.3%, 52.3 � 4.2%, 56.1 � 5.3, and 54.1 � 4.3% (days1–5, respectively) relative to preinjury baseline (Fig. 2b). In contrast,SQS at 100 Hz reduced the heat hyperalgesia observed in rats withinflammation on the first four days of treatment (p < 0.05). Thethermal withdrawal latencies were 120.3 � 3%, 86.5 � 12.8%, 95.6� 11.9%, 80.5 � 8.7%, and 73.3 � 8.3% (days 1–5, respectively)relative to preinjury baseline (Fig. 2b).

Control rats receiving intramuscular saline injection did notdevelop mechanical or thermal hypersensitivity. On the first day oftesting, saline-injected rats had an average mechanical withdrawalthreshold of 13.7 � 1.1 g, which was not statistically different fromthe baseline value of 15.1 � 0.4 g, demonstrating the lack of effectof the act of injection itself (data not shown). Rats received stimula-

Pre-stim 1 2 3 4 50

50

100

150TENS

SQS

Sham TENS

Sham SQS

* ** *

Mechanical

* **

Days of testing

Perc

en

t o

f P

rein

jury

Baselin

e

Pre-stim 1 2 3 4 50

50

100

150TENS

SQS

Sham TENS

Sham SQS

** **

* *

Thermal (Hargreaves)

*

Days of testing

Perc

en

t o

f P

rein

jury

Baselin

e

a

b

Figure 2. a and b. Comparison of the effects of 0 Hz (sham) and 100 Hz TENSor SQS on mechanical and thermal responses of rats with inflammation. Dataare expressed as percent of preinjury baseline. *denotes p < 0.05 vs. prestimu-lation baseline.

Pre-stim 1 2 3 4 50

20

40

60

80

100TENS

SQS

Sham TENS

Sham SQS

* *

Mechanical

Days of testing

Perc

en

t o

f P

rein

jury

Baselin

e

*

Cold (acetone)

a

b

Pre-stim 1 2 3 4 50

50

100

150TENS

SQS

Sham TENS

Sham SQS

**

Days of testing

Perc

en

t o

f P

ain

Resp

on

se

Figure 3. a and b. Comparison of the effects of 0 Hz (sham) and 100 Hz TENSor SQS on mechanical and thermal responses of rats with nerve injury. Data areexpressed as percent of preinjury baseline for the mechanical responses. Forcold responses, data are expressed as a percent of the maximal pre-electricalstimulation pain response in seconds after application of acetone. *denotesp < 0.05.

331EFFECTS OF TENS VS. SQS IN RODENT MODELS OF PAIN

www.neuromodulationjournal.com Neuromodulation 2013; 16: 328–335© 2013 Medtronic, Inc.

Page 5: Differential Effects of Subcutaneous Electrical Stimulation (SQS) and Transcutaneous Electrical Nerve Stimulation (TENS) in Rodent Models of Chronic Neuropathic or Inflammatory Pain

tion with TENS or SQS for five consecutive days at frequencies of0 Hz (sham stimulation) or 100 Hz. Stimulation or sham stimulationdid not alter mechanical withdrawal thresholds in sham animals asmeasured every day for five days after the stimulation period (datanot shown). Measurement of thermal hyperalgesia revealed similarresults. Saline-injected rats had average withdrawal latency fromthe thermal stimulus of 14.3 � 1.0 s, which was not statisticallydifferent from the baseline value of 15.4 � 1.4 s, demonstrating thelack of effect of the act of injection on thermal sensory thresholds.

Effects of TENS and SQS on Neuropathic PainExperimental rats that received a nerve injury (SNI) developed

both mechanical and cold allodynia. The average mechanical with-drawal threshold of rats on day 7 post-SNI was 2.9 � 0.7 g, whichwas statistically different from the preinjury baseline value of 14.4 �0.5 g, indicating the development of mechanical allodynia (a reduc-tion to 20.1 � 3.7% of the baseline value, Fig. 3a).

Rats that received TENS at 0 Hz had mechanical withdrawalthresholds of 29.3 � 6.1%, 23.8 � 6.1%, 19.7 � 5.7%, 21.7 � 8.1%,and 24.2 � 5.1% (days 1–5, respectively) relative to preinjury base-line, indicating the lack of effects due to anesthesia alone. Ratstreated with TENS at 100 Hz had mechanical withdrawal thresholdsof 40.5 � 13.7%, 48.5 � 13.8%, 46.5 � 14.8%, 45.3 � 13.9%, and 52.1� 15.8% (days 1–5, respectively) relative to preinjury baseline. Therewas a trend for increased withdrawal thresholds with TENS com-pared with prestimulation values, but they did not reach statisticalsignificance (Fig. 3a).

Rats treated with SQS at 0 Hz had mechanical withdrawal thresh-olds of 35.7 � 10.3%, 22.5 � 7.5%, 20.7 � 5.1%, 33.5 � 11.3%, and29.3 � 6.4% (days 1–5, respectively) relative to preinjury baseline,indicating no effect throughout the observation period. In contrast,SQS at 100 Hz had significant effect starting on day 3 of treatment.The mechanical withdrawal thresholds were 42.4 � 11.9%, 44.2 �11.7%, 58.5 � 11.6%, 79.2 � 9.0%, and 81.6 � 9.2% (days 1–5,respectively) relative to preinjury baseline (Fig. 3a).

Rats with SNI also developed cold allodynia indicated by anincreased duration of pain behaviors after the application ofacetone to the injured paw. On day 7 after SNI, rats had duration ofresponse of 12.1 � 2.3 sec, which was statistically significant com-pared with the preinjury baseline value of 0.2 � 0.2 sec (p < 0.05).Rats receiving TENS at 0 Hz did not show any significant changes incold allodynia. The durations of the responses were 86.3 � 19.9%,73.3 � 20.1%, 114.9 � 13.8%, 118.6 � 20.9%, and 117.7 � 18.0%(day 1–5, respectively) relative to the postinjury baseline. Ratsreceiving 100 Hz had durations of responses of 31.8 � 13.7%, 96.4 �19.2%, 86.1 � 14.9%, 82.4 � 20.2%, and 73.9 � 24.2% (days 1–5,respectively) relative to postinjury baseline, demonstrating a signifi-cant effect only on day 1 (p < 0.05, Fig. 3b).

Rats receiving SQS at 0 Hz had durations of responses of 90.7 �23.7%, 84.8 � 25.4%, 84.8 � 17.9%, 73.7 � 21.8%, and 91.6 � 22.7%(days 1–5, respectively) relative to postinjury baseline, demonstrat-ing no effects of sham SQS. Rats receiving SQS at 100 Hz had dura-tions of responses of 37.1 � 19.1%, 79.7 � 23.8%, 51.9 � 19.9%, 92.7� 23.8%, and 66.8 � 21.5% (days 1–5, respectively) relative topostinjury baseline, demonstrating a significant effect (p < 0.05)only on day 1 (Fig. 3b).

Control rats receiving a sham nerve injury did not developmechanical or cold allodynia. On the first day of testing, sham-injured rats had an average mechanical withdrawal threshold of13.3 � 1.2 g, which was not statistically different from the presur-gery value of 14.9 � 0.3 g, demonstrating that the sham nerve

injury procedure did not have an effect on mechanical sensitivity.Sham-injured rats received stimulation with TENS or SQS for fiveconsecutive days at frequencies of 0 Hz (sham stimulation) or100 Hz. Both had no effects on the mechanical withdrawal thresh-olds as measured every day after the stimulation period (data notshown). Measurement of cold sensitivity yielded similar results.Sham-injured rats had an average duration of response to coldacetone of 1.1 � 0.7 s, which was not statistically different from thepresurgery baseline value of 0.3 � 0.2 s, demonstrating the lack ofeffect of the sham injury procedure on cold sensitivity. Rats receivedstimulation with TENS or SQS for five consecutive days at frequen-cies of 0 Hz or 100 Hz. Stimulation did not have any effects on theduration of response to cold acetone measured each day after theend of the stimulation period (data not shown).

DISCUSSION

This study is the first to demonstrate antihyperalgesic effects ofSQS in rodent models of neuropathic and inflammatory pain. More-over, this research sets the foundation for investigating the mecha-nisms of SQS by comparing the outcomes to a well-establishedmodel of TENS. The effectiveness of SQS or TENS was investigated inrat models of neuropathic or inflammatory pain. In the inflamma-tory model, TENS reduced mechanical hypersensitivity withdecreasing efficiency (i.e., tolerance) over the five days of testing. Incontrast, the effect of SQS on mechanical hypersensitivity increasedover the five testing days. In the inflammation model, SQS wassimilar to TENS in that tolerance developed in the thermal test. Inthe neuropathic pain model, TENS differed from SQS in havingminimal reversal of SNI-induced mechanical hypersensitivity. SQSreduced SNI-induced mechanical hypersensitivity starting on day 3that increased in efficacy to day 5. TENS and SQS were similar in theneuropathic pain model in their ability to reduce cold hypersensi-tivity on day 1. These results show that TENS and SQS have differen-tial effects on hypersensitivity assessed via a mechanical assay butsimilar effects on thermal-induced hypersensitivity. The effect ofSQS vs. TENS in the mechanical assay in both pain models suggeststhat different biologic mechanisms underlie its influence on thistype of injury-evoked hypersensitivity.

The results are also in agreement with similar studies that haveshown the development of tolerance to the analgesic effects ofTENS in inflammatory pain models (26–28). The development oftolerance is thought to be linked to the engagement of opioidmechanisms. Indeed, TENS applied with low (4 Hz) and high(100 Hz) frequencies activates spinal m and d opioid receptormechanisms, respectively (11). Previous studies also provide indi-rect evidence of endogenous opioid release as both high and lowfrequency TENS increase b-endorphin concentrations in the cere-brospinal fluid (29–31). Interestingly, this effect is not unique toTENS, as EA also produces release of opioids (12,13). The involve-ment of opioids in these two types of neurostimulation therapieshas also been demonstrated in human studies (32,33).

In contrast to TENS, SQS had a different effect on the mechanicalhypersensitivity induced by inflammation. This difference may bein part due to the difference in the position of the electrodes rela-tive to the skin. The peak SQS effect was not observed until the lastday of the stimulation period. This “ramping-up” pattern of SQS is instark contrast to the “ramping-down,” tolerance pattern apparent inthe TENS model. The reason for this differential effect might relateto the types or numbers of fibers that are activated by TENS vs.SQS. Along with activation of large Ab fibers, TENS can also activate

332

VERA-PORTOCARRERO ET AL.

www.neuromodulationjournal.com Neuromodulation 2013; 16: 328–335© 2013 Medtronic, Inc.

Page 6: Differential Effects of Subcutaneous Electrical Stimulation (SQS) and Transcutaneous Electrical Nerve Stimulation (TENS) in Rodent Models of Chronic Neuropathic or Inflammatory Pain

smaller Ad fibers (34). Because TENS electrodes are placed on top ofthe skin, the electrical field might reach the more superficial Adfibers (35), giving rise to release of opioids (36,37). In contrast, SQSleads placed in the subcutaneous space are in closer proximity tothe large Ab fibers, which may lead to their preferential activation.In other electrical stimulation therapies that primarily activate largeAb fibers, such as SCS, the release of nonopioid neurotransmittershas been documented (38–41). These neurotransmitters contributeto the analgesic effect of SCS. It would be reasonable to hypoth-esize that SQS primarily activates Ab fibers, resulting in the releaseof a neurotransmitter profile that differs from TENS. Different neu-rotransmitter profiles may be apparent through the contrastingeffects on mechanical allodynia. Indeed, a previous study hasrevealed that SQS and TENS might be recruiting different numberof fibers (42).

Interestingly, this study suggests that there may also be mecha-nistic similarities between SQS and TENS. For example, the antihy-peralgesic effect of both TENS and SQS stimulation declined overthe five days of behavioral testing with heat in the inflammatorymodel. This finding could speak to a common tolerance, or loss ofantihyperalgesic effect, that works primarily through pathwaystransmitting thermal sensory information. In the neuropathic painmodel, both TENS and SQS reversed hypersensitivity in the coldassay on day 1, but in subsequent days, became ineffective. Thisresult again suggests that there are some similarities between TENSand SQS depending on the sensory modality studied.

The results obtained in this study with SQS on the neuropathicpain model mimic the findings obtained with rodent models of SCSusing neuropathic pain rodent models (18,43). These studies alsofound a “ramping up” of the effects of SCS more than four days ofstimulation with day 4 SCS showing the greatest effect. These simi-larities speak to the possibility that SQS and SCS target similarneural substrates and/or involve similar mechanisms that mediatepain-relieving effects.

The effects of TENS on the neuropathic pain model were not asstriking as those of SQS. There was a trend of mechanical allodyniareduction across the five days of treatment, but it did not reach asignificant difference. Previous human studies (44–46) and animalstudies (47) have suggested that TENS might not be as effective forneuropathic pain conditions as for inflammatory pain conditions.Nonetheless, there are also studies in humans that show that TENScould still be effective in neuropathic pain (48–50). In addition, astudy in a rodent model of neuropathic pain also demonstrated thatapplication of low amplitude 100 Hz TENS attenuated mechanicalhyperalgesia but not thermal hyperalgesia when applied contralat-eral to a nerve injury (51). The disagreement between these studiesmost probably arises from the wide array of parameters used forstimulation and the location of the TENS electrodes (18,52–54). Inour study, we used the same parameters on the neuropathic painmodel that were employed for the inflammatory pain model. Thelocation of the electrodes was chosen based on previous studieswhere the electrodes were placed close to the original injury site(10,11,25–27). It is possible that a different location or differentparameters might have been more effective for addressing painbehaviors in the neuropathic pain model.

The effects ofTENS are also in contrast to previous results obtainedwith SCS where SCS produces a “ramping up” of effects over thecourse of three days (18) or more (43). As stated above, this differencein behavioral results might stem from the different areas of thenervous system that each therapy is targeting. SCS targets the dorsalcolumn fibers, which are composed mostly of large Ab fibers. TENS istargeting the peripheral nervous system, activating nerve endings in

the skin that consist of large Ab fibers and also smaller Ad and C-fibers(35). TENS might be activating, besides Ab fibers to produce pares-thesia, also the smaller fibers that eventually produced release ofopioids in the spinal cord (36,37). This can lead to the toleranceeffects observed in this study and previous work (26–28). In thepresent study, SQS had some similarities with TENS in the thermalassays, but in the mechanical assays the effects of SQS were more inline with SCS. Another source of differences observed in this studyand previous studies with SCS is that electrical stimulation in theperiphery (the skin) might bring into account other peripheralmodulators. For example, EA can produce release of opioids in theperiphery (55) that could contribute to differential behavioral effects.

In the present study, stimulation (TENS or SQS) was given whilerats were anesthetized as previously described in TENS studies (25–27) and in order to equalize the mode of administration of eachtherapy. To control for possible effects of anesthesia in the behav-ioral outcomes, there were two sets of controls. Sham SNI or shaminflammation rats (“normal”) did not show any change in behaviorafter 20 min of isofluorane anesthesia. Moreover, rats with SNI orinflammation that received “sham” stimulation (0 Hz) still demon-strated pain behaviors suggesting that isofluorane anesthesia didnot act as an analgesic. Furthermore, we were careful in placing theelectrodes for SQS and TENS in identical positions in both models toattain a fair comparison. A previous study (56) described the sizeand placement of TENS electrodes on the surface of the skin over-lying the inflamed calf muscle. We used the same size of electrodesand replicated this placement in our study. We also placed the SQSelectrodes in the same location but under the skin. In nerve injurymodels, previous studies have demonstrated that rats with SNI arehypersensitive to compression of the hindlimb muscles (43,57). Itwas assumed that nerve injury not only produces hypersensitivity ofthe plantar aspect of the paw but also the nerve territories in thehindlimb; therefore, SQS and TENS electrodes were placed in thesame hindlimb area.

CONCLUSION

The findings in this study provide the first evidence that electricalstimulation of nerve endings in the subcutaneous space inducemechanisms that counteract both inflammatory and neuropathicpain. This conclusion is further supported by the replication in ourstudy of previously published findings of TENS in similar pain animalmodels (25–27).

In any animal model, there are limits to how well the conditionstranslate to a clinical setting. For example, electrical stimulation wasprovided in an assumed area of pain, whereas the behavior testingwas done in a location remote from both the injury site and stimu-lation site. It is not clear whether the efficacy of SQS would changeif lead location changes or if behavioral testing could be performedcloser to the stimulation site. The site chosen for SQS in this studywas based on an established TENS target site in rodent models.Therefore, the implant technique and location was determined tobest align with reported placement of TENS electrodes. Further-more, the skin of the rodent at the site of SQS implant is quite thinand does not exactly model clinical electrode placement betweenthe dermis and a layer of subcutaneous fat. In addition, only onestimulation parameter paradigm (100 Hz, 250 msec, 90% of MT) wasinvestigated in this study. This paradigm was also chosen based onestablished TENS models. However, electrical stimulation therapiesappear to activate different neural mechanisms when different set-tings, such as frequency, are used (12,13,25). Future research will

333EFFECTS OF TENS VS. SQS IN RODENT MODELS OF PAIN

www.neuromodulationjournal.com Neuromodulation 2013; 16: 328–335© 2013 Medtronic, Inc.

Page 7: Differential Effects of Subcutaneous Electrical Stimulation (SQS) and Transcutaneous Electrical Nerve Stimulation (TENS) in Rodent Models of Chronic Neuropathic or Inflammatory Pain

build upon this SQS model to further investigate mechanisms, opti-mize electrode placement, and investigate optimal stimulationparameters.

Authorship Statements

Dr. Vera-Portocarrero designed and conducted the study, includ-ing data analysis, data interpretation, and writing of the manuscript.Drs. Cordero, Wacnik, Johanek, and Ms. Billstrom assisted withexperimental design and provided input for manuscript prepara-tion. Ms. Swearingen collected all data and assisted with experimen-tal design and manuscript preparation.

How to Cite this Article:Vera-Portocarrero L.P., Cordero T., Billstrom T.,Swearingen K., Wacnik P.W., Johanek L.M. 2013. Differ-ential Effects of Subcutaneous Electrical Stimulation(SQS) and Transcutaneous Electrical Nerve Stimulation(TENS) in Rodent Models of Chronic Neuropathic orInflammatory Pain.Neuromodulation 2013; 16: 328–335

REFERENCES

1. Melzack R, Wall PD. Pain mechanisms: a new theory. Science 1965;150:971–979.2. Wall PD, Sweet WH. Temporal abolition of pain in man. Science 1967;155:108–

109.3. Shealy CN, Mortimer JT, Reswick JB.Electrical inhibition of pain by stimulation of the

dorsal columns: preliminary clinical report. Anesth Analg 1967;46:489–491.4. Picaza JA, Hunter SE, Cannon BW. Pain suppression by peripheral nerve stimulation.

Appl Neurophysiol 1977;40:223–234.5. Nashold BS, Goldner L, Mullen JB, Bright DS. Long-term pain control by direct

peripheral nerve stimulation. J Bone Joint Surg 1982;64:1–10.6. Takakura N, Ogawa H, Iijima S et al. Effect of acupuncture at the Hoku point of

vibration-induced finger flexion reflex in man: comparison between press needletechnique, electroacupuncture and in-situ technique. Am J Chin Med 1995;23:313–318.

7. Han J-S, Zhou Z, Xuan Y. Acupuncture has an analgesic effect in rabbits. Pain1983;15:83–91.

8. Lao LX, Zhang RX, Zhang G, Wang X, Berman BM, Ren K. A parametric study ofelectroacupuncture on persistent hyperalgesia and FOS protein expression in rats.Brain Res 2004;1020:18–29.

9. Kim JH, Min BI, Schmidt D, Lee HJ, Park DS. The difference between electroacupunc-ture only and electroacupuncture with manipulation on analgesia in rats. NeurosciLett 2000;279:149–152.

10. Kalra A, Urban MO, Sluka KA. Blockade of opioid receptors in rostral ventral medullaprevents antihyperalgesia produced by transcutaneous electrical nerve stimulation(TENS). J Pharmacol Exp Ther 2001;298:257–263.

11. Sluka KA, Deacon M, Stibal A, Strissel S, Terpstra A. Spinal blockade of opioid recep-tors prevents the analgesia produced by TENS in arthritic rats. J Pharmcol Exp1999;289:840–846.

12. Fei H, Xie GX, Han JS. Low and high frequency electroacupuncture stimulationrelease [Met5]enkephalin and dynorphin A in rat spinal cord. Sci Bull China1987;32:1496–1501.

13. Han JS. Acupuncture: neuropeptide release produced by electrical stimulation ofdifferent frequencies. Trends Neurosci 2003;26:17–22.

14. Levine JD, Gormley J, Fields HL. Observations on the analgesic effects of needleacupuncture. Pain 1976;2:149–159.

15. Nardone A, Schieppati M. Influences of transcutaneous electrical stimulation ofcutaneous and mixed nerves on subcortical and cortical somatosensory evokedpotentials. Electroencephalogr Clin Neurophysiol 1989;74:24–35.

16. Kagitani F, Uchida S, Hotta H. Afferent nerve fibers and acupuncture. Auton Neurosci2010;157:2–8.

17. Buonocore M, Bonezzi C, Barolat G. Neurophysiological evidence of antidromicactivation of large myelinated fibers in lower limbs during spinal cord stimulation.Spine 2008;4:E90–E93.

18. Yang F, Carteret AF, Wacnik PW et al. Bipolar spinal cord stimulation attenuatesmechanical hypersensitivity at an intensity that activates a small portion of A-fiberafferents in spinal nerve-injured rats. Neuroscience 2011;199:470–480.

19. Chung JM, Fang ZR, Hori Y, Lee KH, Willis WD. Prolonged inhibition of primatespinothalamic tract cells by peripheral nerve stimulation. Pain 1984;19:259–275.

20. Hanai F. Effect of electrical stimulation of peripheral nerves on neuropathic pain.Spine 2000;25:1886–1892.

21. Jorum E. The analgesic effect of peripheral nerve stimulation in various tests ofnociception in rats. Acta Physiol Scand 1988;133:754–760.

22. Nam TS, Song SH, Kim YH, Baik EJ, Paik KS. Effect of peripheral nerve stimulation onthe dorsal horn cell activity in cats with cutaneous inflammation. Yonsei Med J1992;33:109–120.

23. Rosellini W, Casavant R, Engineer N et al. Wireless peripheral nerve stimulationincreases pain threshold in two neuropathic rat models. Exp Neurol 2012;235:621–626.

24. Decosterd I, Woolf CJ. Spared nerve injury: an animal model of persistent peripheralneuropathic pain. Pain 2000;87:149–158.

25. Chaplan SR, Bach FW, Pogrel JW, Chung JM, Yaksh TL. Quantitative assessment oftactile allodynia in the rat paw. J Neurosci Methods 1994;53:55–63.

26. Chandran P, Sluka KA. Development of opioid tolerance with repeated transcuta-neous electrical nerve stimulation administration. Pain 2003;102:195–201.

27. DeSantana JM, Santana-Filho VJ, Sluka KA. Modulation between high- and low-frequency transcutaneous electrical nerve stimulation delays the development ofanalgesic tolerance in arthritic rats. Arch Phys Med Rehabil 2008;89:754–760.

28. Hingne PM, Sluka KA. Blockade of NMDA receptors prevents analgesic tolerance torepeated transcutaneous electrical nerve stimulation (TENS) in rats. J Pain2008;9:217–225.

29. Salar G, Job I, Minigrino S, Bosio A, Trabucchi M. Effect of transcutaneous electro-therapy on CSF beta-endorphin content in patients without pain problems. Pain1981;10:208–227.

30. Hughes G, Lichstein P, Whitlock D, Harker C.Response of plasma beta-endorphins totranscutaneous electrical nerve stimulation in healthy subjects. Phys Ther 1984;64:1062–1066.

31. Almay BGL, Johansson F, Knorring L, Sakurada T, Terenius L. Long-term high fre-quency transcutaneous electrical nerve stimulation (hi-TNS) in chronic pain. Clini-cal response and effects of CST endorphins, mono-amine metabolites, substanceP-like immunoreactivity (SPLI) and pain measures. J Psychosom Res 1985;29:247–257.

32. Han JS, Chen XH, Sun SL et al. Effect of low- and high-frequency TENS onmetenkephalin-Arg-Phe and dynorphin A immunoreactivity in human lumbarcerebrospinal fluid. Pain 1991;47:295–298.

33. Liebano RE, Rakel B, Vance CGT, Walsh DM, Sluka KA. An investigation of the devel-opment of analgesic tolerance to TENS in humans. Pain 2011;152:335–342.

34. Lee KH, Chung JM, Willis WD. Inhibition of primate spinothalamic tract cells by TENS.J Neurosurg 1985;62:276–287.

35. Tschachler E, Reinisch CM, Mayer C, Paiha K, Lassmann H, Weninger W. Sheet prepa-rations expose the dermal nerve plexus of human skin and render the dermal nerveend organ accessible to extensive analysis. J Invest Dermatol 2004;122:177–182.

36. Lao L, Song B, Chen W, Marvizon JCG. Noxious mechanical stimulation evokes thesegmental release of opioid peptides that induce m-opioid receptor internalizationin the presence of peptidase inhibitors. Brain Res 2008;1197:85–93.

37. Chen W, Song B, Marvison JCG. Inhibition of opioid release in the rat spinal cord bya2c adrenergic receptors. Neuropharmacology 2008;54:944–953.

38. Cui JG, O’Connor WT, Ungerstedt U, Linderoth B, Meyerson BA. Spinal cord stimula-tion attenuates augmented dorsal horn release of excitatory amino acids inmononeuropathy via a GABAergic mechanism. Pain 1997;73:87–95.

39. Song Z, Ultenius C, Meyerson BA, Linderoth B. Pain relief by spinal cord stimulationinvolves serotonergic mechanisms: an experimental study in a rat model of monon-europathy. Pain 2009;147:241–248.

40. Cui JG, Meyerson BA, Sollevi A, Linderoth B. Effect of spinal cord stimulation ontactile hypersensitivity in mononeuropathic rats is potentiated by simultaneousGABAb and adenosine receptor activation. Neurosci Lett 1998;247:183–186.

41. Schechtmann G, Song Z, Ultenius C, Meterson BA, Linderoth B. Cholinergic mecha-nisms involved in the pain relieving effect of spinal cord stimulation in a model ofneuropathy. Pain 2008;139:136–145.

42. Yang F, Chung CY, Wacnik PW et al. Electrical stimulation at distinct peripheral sitesin spinal nerve injured leads to different afferent activation profiles. Neurosci Lett2011;505:52–57.

43. Maeda Y, Wacnick PW, Sluka KA. Low frequencies, but not high frequencies ofbi-polar spinal cord stimulation reduce cutaneous and muscle hyperalgesiainduced by nerve injury. Pain 2008;138:143–152.

44. Lampl C, Kreczi T, Klingler D. Transcutaneous electrical nerve stimulation in thetreatment of chronic pain: predictive factors and evaluation of method. Clin J Pain1998;14:134–142.

45. Norrbrink C. Transcutaneous electrical nerve stimulation for treatment of spinalcord injury neuropathic pain. J Rehabil Dev 2009;46:85–93.

46. Gossrau G,Wahner M,Kuschke M et al.Microcurrent transcutaneous electrical nervestimulation in painful diabetic neuropathy: a randomized placebo-controlled study.Pain Med 2011;12:953–960.

47. Somers DL, Clemente R. The relationship between dorsal horn neurotransmittercontent and allodynia in neuropathic rats treated with high-frequency transcuta-neous electrical nerve stimulation. Arch Phys Med Rehabil 2003;84:1575–1583.

48. Cheing GL, Luk ML. Transcutaneous electrical nerve stimulation for neuropathicpain. J Hand Surg Br 2005;30:50–55.

49. Jin DM, Xu Y, Geng DF, Yan TB. Effect of transcutaneous electrical nerve stimulationon symptomatic diabetic peripheral neuropathy: a meta-analysis of randomizedcontrolled trials. Diabetes Res Clin Pract 2010;89:10–15.

334

VERA-PORTOCARRERO ET AL.

www.neuromodulationjournal.com Neuromodulation 2013; 16: 328–335© 2013 Medtronic, Inc.

Page 8: Differential Effects of Subcutaneous Electrical Stimulation (SQS) and Transcutaneous Electrical Nerve Stimulation (TENS) in Rodent Models of Chronic Neuropathic or Inflammatory Pain

50. Yameen F, Shahbaz NN, Hasan Y, Fauz R, Abdullah M. Efficacy of transcutaneouselectrical nerve stimulation and its different modes in patients with trigeminalneuralgia. J Pak Med Assoc 2011;61:437–439.

51. Somers DL, Clemente FR. Contralateral high or a combination of high- and low-frequency transcutaneous electrical nerve stimulation reduces mechanical allo-dynia and alters dorsal horn neurotransmitter content in neuropathic rats. J Pain2009;10:221–229.

52. Leem JW, Park ES, Paik KS. Electrophysiological evidence for the antinociceptiveeffect of transcutaneous electrical nerve stimulation on mechanically evokedresponsiveness of dorsal horn neurons in neuropathic rats. Neurosci Lett1995;192:197–200.

53. Nam TS, Choi Y, Yeon DS, Leem JW, Paik KS. Differential antinociceptive effect oftranscutaneous electrical stimulation on pain behavior sensitive or insensitive tophentolamine in neuropathic rats. Neurosci Lett 2001;301:17–20.

54. Somers DL, Clemente FR. Transcutaneous electrical nerve stimulation for the man-agement of neuropathic pain: the effects of frequency and electrode position onprevention of allodynia in a rat model of complex regional pain syndrome type II.Phys Ther 2006;86:698–709.

55. Zhang GG, Yu C, Lee W, Lao L, Ren K, Berman BM. Involvement of peripheral opioidmechanisms in electroacupuncture analgesia. Explore 2005;1:365–371.

56. Ainsworth L, Budelier K, Clinesmith M et al. Transcutaneous electrical nerve stimu-lation (TENS) reduce chronic hyperalgesia induced by muscle inflammation. Pain2006;120:182–187.

57. Skyba DA, Radhakrishnan R, Sluka KA. Characterization of a method for measuringprimary hyperalgesia of deep somatic tissue. J Pain 2005;6:41–47.

COMMENTS

This manuscript reports on the effects of two procedures for peri-pheral nerve stimulation to modulate nociceptive behavior in twodifferent rodent models for pain, based either on nerve lesion (monon-europathy) or local inflammation.

The effects of both types of peripheral stimulation (TENS and SQS)were, to a certain extent, comparable and were moderate to minimalwhen compared to those of central (spinal cord) stimulation. Basedon previous studies, the fiber groups in this stimulation are madeof a major component of Aa/b with a minor involvement of Ad group.It would be interesting to search for a peripheral mechanism underly-ing the observed effects, such as opioids (kindly see recent work byAwad H, et al. (2012) Endogenous Opioids in Wound-Site Neutrophilsof Sternotomy Patients. PLoS ONE 7(10): e47569. doi:10.1371/journal.pone.0047569.

It might also be interesting to compare the observed results ofelectrical stimulation of peripheral nerves or terminals to those ofsensory activation reported by various groups in the 1980’s (Per-tovaara,A. 1979; Lundberg,T, 1983–85 and others).

Nayef Saade, PhDBeirut, Lebanon

***This manuscript deals with the question of the mechanism of SQS inreversing mechanical and thermal sensitization in nerve injury (SNI)and inflammatory muscle models, compared with the known mecha-nism of TENS. The use of electrostimulation as a treatment modality fordifferent types of chronic pain is increasing in popularity. Furthermore,the strong point of the paper is that SQS may be more effective,because the stimulation leads are closer to the nerve terminal.

Karina Sato, PhDIowa City, IA, USA

***This work shows some data related to possible mechanisms of subcu-taneous and TENS and importantly how these might also differ. This iswholly thematic to the journal. Priority is only lowered because insightremains limited.

Patrick Dougherty, PhD, BAHouston, TX, USA

Comments not included in the Early View version of this paper.

335EFFECTS OF TENS VS. SQS IN RODENT MODELS OF PAIN

www.neuromodulationjournal.com Neuromodulation 2013; 16: 328–335© 2013 Medtronic, Inc.