[ RESEARCH REPORT ]
`
`ASHRAF S. GORGEY, PT, PhD¹(cid:22)(cid:22)(cid:154)(cid:22)(cid:22)(cid:57)(cid:62)(cid:72)(cid:63)(cid:73)(cid:74)(cid:69)(cid:70)(cid:62)(cid:59)(cid:72)(cid:22)(cid:58)(cid:36)(cid:22)(cid:56)(cid:66)(cid:55)(cid:57)(cid:65)(cid:34)(cid:22)PhD²
`(cid:22)(cid:57)(cid:62)(cid:72)(cid:63)(cid:73)(cid:74)(cid:69)(cid:70)(cid:62)(cid:59)(cid:72)(cid:22)(cid:70)(cid:36)(cid:22)(cid:59)(cid:66)(cid:58)(cid:59)(cid:72)(cid:34)(cid:22)PhD³(cid:22)(cid:22)(cid:154)(cid:22)(cid:22)(cid:61)(cid:55)(cid:72)(cid:79)(cid:22)(cid:55)(cid:36)(cid:22)(cid:58)(cid:75)(cid:58)(cid:66)(cid:59)(cid:79)(cid:34)(cid:22)PhD4
`
`Effects of Electrical Stimulation Parameters
`on Fatigue in Skeletal Muscle
`
`Neuromuscular electrical stimulation (NMES) is a promising
`
`tool in the rehabilitation of individuals with a limited ability
`to activate their skeletal muscles,13,35,36 as well as a method
`of strength training and short-term resistance training in
`athletic populations.26,27 During NMES application, the capacity to
`maintain performance is compromised compared to voluntary exercise,
`
`resulting in a higher rate of muscle fa-
`tigue.23 Muscle fatigue is defined as a
`
`reduction in the peak force, with contin-
`uous and repeated activation that could
`
`(cid:84)(cid:22)(cid:73)(cid:74)(cid:75)(cid:58)(cid:79)(cid:22)(cid:58)(cid:59)(cid:73)(cid:63)(cid:61)(cid:68)(cid:48) Experimental laboratory study.
`(cid:84)(cid:22)(cid:69)(cid:56)(cid:64)(cid:59)(cid:57)(cid:74)(cid:63)(cid:76)(cid:59)(cid:73)(cid:48) The primary purpose was to
`investigate the independent effects of current
`amplitude, pulse duration, and current frequency
`on muscle fatigue during neuromuscular electrical
`stimulation (NMES). A second purpose was to
`determine if the ratio of the evoked torque to the
`activated area could explain muscle fatigue.
`(cid:84)(cid:22)(cid:56)(cid:55)(cid:57)(cid:65)(cid:61)(cid:72)(cid:69)(cid:75)(cid:68)(cid:58)(cid:48) Parameters of NMES have
`been shown to differently affect the evoked torque
`and the activated area. The efficacy of NMES is
`limited by the rapid onset of muscle fatigue.
`(cid:84)(cid:22)(cid:67)(cid:59)(cid:74)(cid:62)(cid:69)(cid:58)(cid:73)(cid:22)(cid:55)(cid:68)(cid:58)(cid:22)(cid:67)(cid:59)(cid:55)(cid:73)(cid:75)(cid:72)(cid:59)(cid:73)(cid:48) Seven healthy
`participants underwent 4 NMES protocols that
`were randomly applied to the knee extensor
`muscle group. The NMES protocols were as fol-
`lows: standard protocol (Std), defined as 100-Hz,
`450-μs pulses and amplitude set to evoke 75% of
`maximal voluntary isometric torque (MVIT); short
`pulse duration protocol (SP), defined as 100-Hz,
`150-μs pulses and amplitude set to evoke 75%
`of MVIT; low-frequency protocol (LF), defined as
`25-Hz, 450-μs pulses and amplitude set to evoke
`75% of MVIT; and low-amplitude protocol (LA),
`defined as 100-Hz, 450-μs pulses and amplitude
`
`set to evoke 45% of MVIT. The peak torque was
`measured at the start and at the end of the 4
`protocols, and percent fatigue was calculated. The
`outcomes of the 4 NMES protocols on the initial
`peak torque and activated cross-sectional area
`were recalculated from a companion study to
`measure torque per active area.
`
`(cid:84)(cid:22)(cid:72)(cid:59)(cid:73)(cid:75)(cid:66)(cid:74)(cid:73)(cid:48) Decreasing frequency from 100 to 25
`Hz decreased fatigue from 76% to 39%. Decreas-
`ing the amplitude and pulse duration resulted in no
`change of muscle fatigue. Torque per active area
`accounted for 57% of the variability in percent
`fatigue between Std and LF protocols.
`
`(cid:84)(cid:22)(cid:57)(cid:69)(cid:68)(cid:57)(cid:66)(cid:75)(cid:73)(cid:63)(cid:69)(cid:68)(cid:73)(cid:48) Altering the amplitude of the
`current and pulse duration does not appear to
`influence the percent fatigue in NMES. Lowering
`the stimulation frequency results in less fatigue, by
`possibly reducing the evoked torque relative to the
`activated muscle area. J Orthop Sports Phys Ther
`2009;39(9):684-692. doi:10.2519/jospt.2009.3045
`
`(cid:84)(cid:22)(cid:65)(cid:59)(cid:79)(cid:22)(cid:77)(cid:69)(cid:72)(cid:58)(cid:73)(cid:48) amplitude, frequency, NMES,
`pulse duration
`
`impair functional or therapeutic goals.14,15
`Muscle fatigue could result from either
`increasing the metabolic cost of muscular
`contractions or from the pattern of motor
`units recruitment during stimulation.
`Measuring the peak torque or torque-
`time integral (TTI) has been used as an
`index to reflect the metabolic cost of
`the stimulated muscle,5,29-31 because the
`force-generating capacity is a function
`of the number of cross-bridges between
`actin and myosin myofilaments which
`are directly related to ATP hydrolysis.32,33
`Additionally, the initial peak torque has
`been correlated to fatigue resulting from
`NMES.29 Moreover, during stimulation,
`muscle fiber recruitment patterns vary
`from the well-known size principle re-
`cruitment that occurs during voluntary
`contractions.20,21 Evidence suggests that
`muscle recruitment during NMES oc-
`curs in a random order, likely depending
`on the position of the stimulating elec-
`trodes,1,18,21,24 and that motor units are
`activated in a synchronous and repeated
`manner.1,8,24 This pattern of motor unit
`activation may lead to greater fatigue by
`preventing the cycling of motor unit acti-
`vation that is thought to occur during sub-
`maximal voluntary muscle actions.1,8,24
`NMES parameters (eg, current am-
`plitude, frequency, and pulse duration)
`are known to play a critical role in torque
`production during repeated contrac-
`
`1 Research Physical Therapist, Spinal Cord Injury and Disorders Center, Hunter Holmes McGuire VAMC, Richmond, VA. 2 Graduate student (at time of study), Department of
`Kinesiology, The University of Georgia, Athens, GA; Assistant Professor, Department of Kinesiology, Georgia College and State University, Milledgeville, GA. 3 Graduate student
`(at time of study), Department of Kinesiology, The University of Georgia, Athens, GA; Postdoctoral Research Fellow, Vanderbilt University, Nashville, TN. 4 Distinguished Research
`Professor (deceased), Department of Kinesiology, The University of Georgia, Athens, GA. The Institutional Review Board of The University of Georgia approved the current study.
`This study was supported by NIH grants to Gary A. Dudley (HD39679 and HD39676S2). Address correspondence to Dr Ashraf S. Gorgey, Department of Veterans Affairs, Hunter
`Holmes McGuire Medical Center, Spinal Cord Injury and Disorders Service, 1201 Broad Rock Boulevard, Richmond, VA 23249. E-mail: ashraf.gorgey@va.gov
`
`684 | september 2009 | volume 39 | number 9 | journal of orthopaedic & sports physical therapy
`
` Journal of Orthopaedic & Sports Physical Therapy®
`
` Downloaded from www.jospt.org at on August 3, 2021. For personal use only. No other uses without permission.
`
` Copyright © 2009 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
`
`LUMENIS EX1014
`Page 1
`
`

`

`tions.1,8,16,17 It is generally accepted that
`increasing current amplitude, frequency,
`and pulse duration will increase evoked
`torque; however, the independent ef-
`fects of these parameters on motor units
`recruitment are less appreciated. Pre-
`vious studies have clearly established
`that increasing current amplitude leads
`to increased torque production via the
`activation of additional motor units.1,17
`Binder-Macleod et al8 showed that in-
`creasing the current amplitude results
`in steep rise of the torque, followed by
`a plateau at a high level of stimulation.
`Increasing pulse duration has also been
`shown to increase the evoked torque by
`possibly increasing motor unit activa-
`tion.17 A pulse duration of 450 μs elicited
`22% and 55% greater torque output com-
`pared to pulse durations of 250 and 150
`μs, respectively.16,17 However, increasing
`the frequency of NMES has been shown
`to increase evoked torque by increasing
`the torque per active muscle area of skel-
`etal muscle.9,17
`Because increasing the frequency and
`pulse duration increase the evoked torque
`per unit of activated muscle,9,17 it causes
`an increased energy demand that can-
`not be supplied by the muscle and thus
`leads to muscular fatigue.4,28 These find-
`ings may illustrate that fatigue during
`NMES is not necessarily related to peak
`muscle torque production, but may in
`fact be related to the metabolic demand
`placed upon each activated motor unit.
`Thus torque per active area, rather than
`torque production, could provide a better
`indicator of the metabolic demand dur-
`ing stimulation.
`The independent effects of these 3
`parameters on muscular fatigue are still
`controversial. Conflicting results exist on
`the role of current amplitude on muscle
`fatigue, with 1 study demonstrating an
`increase fatigue with increasing am-
`plitude8 and others demonstrating no
`change in fatigue with increasing current
`amplitude.1,34 Increasing the frequency of
`pulses has been shown to accelerate mus-
`cle fatigue.7,25 For example, a stimulus at
`a frequency of 85 Hz has been shown to
`
`cause more fatigue compared to 25 Hz10
`because of the high metabolic cost asso-
`ciated with stimulation at 85 Hz.29,30 Yet
`another study showed that this general
`rule is debatable when settings of 80 Hz
`and 100 Hz demonstrated no significant
`difference in muscular fatigue.31 Com-
`pared to the influence of current am-
`plitude and frequency, the role of pulse
`duration on muscle fatigue is even less
`well established.
`The primary purpose of this study was
`to examine the independent effects of
`current amplitude, frequency, and pulse
`duration on muscle fatigue after altering
`the evoked torque and muscle recruit-
`ment. To accomplish this purpose, the
`current amplitude was increased from
`that needed to evoke 45% of maximal
`voluntary isometric torque (MVIT) to
`that needed to evoke 75% of MVIT, pulse
`duration was increased from 150 to 450
`μs, and the frequency was increased from
`25 to 100 Hz. A second purpose was to
`examine the relationship between the
`evoked torque adjusted to the activated
`area and muscle fatigue. The rationale
`was based on the hypothesis that alter-
`ing the NMES parameters to increase
`the initial peak torque relative to the ac-
`tivated area would lead to a concomitant
`increase in muscle fatigue.
`
`(cid:67)(cid:59)(cid:74)(cid:62)(cid:69)(cid:58)(cid:73)
`
`(cid:74)his study used data collected,
`
`but not analyzed, in an earlier study
`of the effects of NMES on specific
`tension (ie, the evoked torque relative
`to the activated area). Because of the
`inherent difficulties in determining the
`physiological variables (pennation angle,
`moment arm, and fiber length) used to
`estimate specific tension for the knee ex-
`tensor muscle group, we have separated
`these data into 2 sets to address different
`research questions using the same NMES
`protocols.
`
`(cid:73)(cid:107)(cid:88)(cid:96)(cid:91)(cid:89)(cid:106)(cid:105)
`Seven healthy participants (6 males and 1
`female) were recruited from the universi-
`
`ty community. None had a history of knee
`or hip pathological conditions. They were
`(mean (cid:22) SD) 28 (cid:22) 4 years old, weighed
`68 (cid:22) 9 kg, and were 173 (cid:22) 9 cm tall. They
`all had previous experience with similar
`research protocol to address different re-
`search questions. The associated benefits
`and risks of participating in the study
`were explained to each subject, and each
`subject signed a written informed con-
`sent. The Institutional Review Board of
`The University of Georgia approved the
`protocol for this study.
`
`(cid:70)(cid:104)(cid:101)(cid:89)(cid:91)(cid:90)(cid:107)(cid:104)(cid:91)
`Familiarization Session One week prior
`to data collection, subjects participated in
`a 30-minute practice session to acquaint
`themselves with the NMES protocols. In
`this session, each subject was asked to
`perform 3 trials of maximum voluntary
`isometric knee extension efforts for both
`lower extremities. The highest trial for
`each lower extremity was considered the
`MVIT effort. To demonstrate tolerance of
`the 4 NMES protocols, each knee exten-
`sor muscle group was assigned 2 protocols
`and was then stimulated. Each stimu-
`lation protocol delivered 30 isometric
`contractions to the knee extensor muscle
`group. The procedure was performed to
`determine if all participants could toler-
`ate stimulation at 75% of their MVIT.
`Maximum
`Voluntary
`Isometric
`Torque MVIT of the left and right knee
`extensors were determined for each
`participant, as described previously.1,34
`The participant sat on a custom-built
`chair, with a hip angle of 110° and the
`knee secured at approximately 60° of
`flexion. The shin of the lower leg was
`firmly secured to a rigid lever arm with
`an inelastic strap, to ensure that the knee
`extensors could perform only isomet-
`ric actions. A lever arm was established
`by placing a load cell perpendicular to,
`and 33 cm away from, the axis of rota-
`tion of the lever arm. The participant was
`asked to contract the knee extensors as
`fast and forcefully as possible, while ver-
`bal encouragement was provided. Trials
`were repeated if the difference between
`
`journal of orthopaedic & sports physical therapy | volume 39 | number 9 | september 2009 | 685
`
` Journal of Orthopaedic & Sports Physical Therapy®
`
` Downloaded from www.jospt.org at on August 3, 2021. For personal use only. No other uses without permission.
`
` Copyright © 2009 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
`
`LUMENIS EX1014
`Page 2
`
`

`

`[ RESEARCH REPORT ]
`
`(cid:74)(cid:55)(cid:56)(cid:66)(cid:59)(cid:22)(cid:39)
`
`Summary of the 4 NMES Protocols
`and Their Outcomes*
`
`the peaks of 2 separate trials was greater
`than 5%. The load cell, interfaced with a
`personal computer, was used to measure
`knee extension torque expressed in Nm.
`All force data were corrected for gravity
`and then saved for future analysis.
`After determining the MVIT, each
`subject was asked to assess his/her abil-
`ity to tolerate NMES. A Theratouch 4.7
`NMES unit (Rich-Mar Corporation, Ino-
`la, OK) was used. The current amplitude
`required to elicit 75% of the MVIT for
`each lower extremity was determined by
`delivering 1-second trains of progressive-
`ly greater amplitude at a frequency of 100
`Hz, with a 450-μs pulse duration. At least
`1 minute separated each train. All partici-
`pants were asked to completely relax, and
`the current was progressively increased.
`Three to 4 trials per participant were per-
`formed to determine the amplitude of the
`current in milliamps (mA).
`NMES was applied to the knee exten-
`sor muscle group via large (8 (cid:19) 10-cm)
`surface electrodes (Uni-Patch Inc, Wa-
`basha, MN), as done previously.1,13,34 One
`electrode was placed on the skin 2 to 3
`cm above the superior aspect of the patel-
`la, over the vastus medialis muscle, and
`the other lateral to and 30 cm above the
`patella, over the vastus lateralis muscle.
`The anatomical location of each pair of
`electrodes was marked with a permanent
`marker to ensure similar positioning in
`subsequent protocols.
`NMES Protocols The current amplitude
`was adjusted until a torque equivalent
`to 75% of the MVIT was evoked, using
`the nonfatiguing trains. The current am-
`plitudes were determined for the right,
`followed by the left, knee extensors.17
`Next, 1 of 4 NMES protocols was ran-
`domly applied to the knee extensors: 2
`protocols were applied to the right lower
`extremity and the other 2 applied to the
`left. The protocols were as follows: (1) a
`standard protocol (Std) of 100-Hz fre-
`quency, 450-μs pulse duration, and a cur-
`rent amplitude set to evoke 75% MVIT;
`(2) a short pulse duration protocol (SP)
`of 100-Hz frequency, 150-μs pulse du-
`ration, and a current amplitude set to
`
`(cid:22)
`(cid:70)(cid:104)(cid:101)(cid:106)(cid:101)(cid:89)(cid:101)(cid:98)(cid:22)
`Std
`SP
`LF
`LA
`
`(cid:60)(cid:104)(cid:91)(cid:103)(cid:107)(cid:91)(cid:100)(cid:89)(cid:111)(cid:22)
`(cid:30)(cid:62)(cid:112)(cid:31)(cid:22)
`100
`100
`25†
`100
`
`(cid:70)(cid:107)(cid:98)(cid:105)(cid:91)(cid:22)(cid:58)(cid:107)(cid:104)(cid:87)(cid:106)(cid:95)(cid:101)(cid:100)(cid:22)
`(cid:30)μ(cid:105)(cid:31)(cid:22)
`450
`150†
`450
`450
`
`(cid:74)(cid:101)(cid:104)(cid:103)(cid:107)(cid:91)(cid:22)(cid:102)(cid:91)(cid:104)(cid:22)(cid:55)(cid:89)(cid:106)(cid:95)(cid:108)(cid:91)
`(cid:55)(cid:89)(cid:106)(cid:95)(cid:108)(cid:87)(cid:106)(cid:91)(cid:90)(cid:22)
`(cid:74)(cid:101)(cid:104)(cid:103)(cid:107)(cid:91)(cid:22)
`(cid:55)(cid:99)(cid:102)(cid:98)(cid:95)(cid:106)(cid:107)(cid:90)(cid:91)(cid:22)
`(cid:57)(cid:73)(cid:55)(cid:22)(cid:30)(cid:89)(cid:99)2(cid:31)(cid:22)
`(cid:30)(cid:68)(cid:99)(cid:31)(cid:22)
`(cid:30)(cid:99)(cid:55)(cid:31)(cid:22)
`(cid:55)(cid:104)(cid:91)(cid:87)(cid:22)(cid:30)(cid:68)(cid:99)(cid:37)(cid:89)(cid:99)2(cid:31)
`5.7 (cid:22) 1.2
`30 (cid:22) 7
`166 (cid:22) 41
`74 (cid:22) 18
`76 (cid:22) 16
`4.3 (cid:22) 1.3‡
`18 (cid:22) 10‡
`78 (cid:22) 40‡
`3.9 (cid:22) 0.9‡
`36 (cid:22) 8
`137 (cid:22) 30‡
`72 (cid:22) 18
`5.8 (cid:22) 2.1
`22 (cid:22) 12‡
`109 (cid:22) 35‡
`56 (cid:22) 13†
`Abbreviations: CSA, cross-sectional area; LA, low-amplitude protocol; LF, low-frequency protocol;
`NMES, neuromuscular electrical stimulation; SP, short pulse duration protocol; Std, standard protocol.
`* Values, except those of frequency and pulse duration, are mean (cid:22) SD.
`† Different from the Std protocol.
`‡ Significantly different from Std (P(cid:12).05).
`
`A
`
`B
`
`50 μs
`
`200 μs
`
`Phase duration
`
`Pulse duration
`
`200 μs
`
`50 μs
`
`50 μs
`
`50 μs
`
`Phase duration
`
`Pulse duration
`
`(cid:60)(cid:63)(cid:61)(cid:75)(cid:72)(cid:59)(cid:22)(cid:39)(cid:36)(cid:22)Illustration of the durations (450 and 150 μs) of the symmetrical biphasic pulses used for the Std, LF,
`LA (450 μs), and SP (150 μs) protocols. The phase duration was 200 and 50 μs for the 450 and 150 μs pulses,
`respectively.
`
`evoke 75% MVIT; (3) a low-frequency
`protocol (LF) of 25-Hz frequency, 450-
`μs pulse duration, and a current ampli-
`tude set to evoke 75% MVIT; and (4) a
`low-amplitude protocol (LA) of 100-Hz
`frequency, 450-μs pulse duration, and
`at a current that evoked the average of
`the initial torques of SP and LF, as there
`was no available consensus on the lowest
`amplitude that should be used to stimu-
`late the knee extensors ((cid:74)(cid:55)(cid:56)(cid:66)(cid:59)(cid:22)(cid:39)). Rectan-
`gular symmetrical biphasic pulses were
`used for the 4 protocols ((cid:60)(cid:63)(cid:61)(cid:75)(cid:72)(cid:59)(cid:22)(cid:39)). Thirty
`3-second contractions were evoked over a
`
`3-minute period for each protocol (work-
`to-rest cycle of 3 seconds on and 3 sec-
`onds off ).17 The administration order of
`the 4 protocols was randomized to each
`participant and to both knee extensors.
`At least 120 minutes separated 2 subse-
`quent protocols to ensure muscle fatigue
`recovery. Before starting a new protocol,
`the recovery of force was examined by ap-
`plying Std for 1 second and performing
`a MVIT. The recovery force and MVIT
`had to be within 1% of the initial test-
`ing to proceed to the next protocol. Pilot
`work suggested that recovery time should
`
`686 | september 2009 | volume 39 | number 9 | journal of orthopaedic & sports physical therapy
`
` Journal of Orthopaedic & Sports Physical Therapy®
`
` Downloaded from www.jospt.org at on August 3, 2021. For personal use only. No other uses without permission.
`
` Copyright © 2009 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
`
`LUMENIS EX1014
`Page 3
`
`

`

`(cid:74)(cid:55)(cid:56)(cid:66)(cid:59)(cid:22)(cid:40)
`
`Knee Extensor Torque for Number of
`Contractions by the 4 Stimulation Protocols*
`
`(cid:22)
`Std
`SP
`LF
`LA
`
`(cid:57)(cid:101)(cid:100)(cid:106)(cid:104)(cid:87)(cid:89)(cid:106)(cid:95)(cid:101)(cid:100)(cid:22)(cid:41)(cid:38)
`(cid:57)(cid:101)(cid:100)(cid:106)(cid:104)(cid:87)(cid:89)(cid:106)(cid:95)(cid:101)(cid:100)(cid:22)(cid:40)(cid:39)(cid:22)
`(cid:57)(cid:101)(cid:100)(cid:106)(cid:104)(cid:87)(cid:89)(cid:106)(cid:95)(cid:101)(cid:100)(cid:22)(cid:39)(cid:39)(cid:22)
`(cid:57)(cid:101)(cid:100)(cid:106)(cid:104)(cid:87)(cid:89)(cid:106)(cid:95)(cid:101)(cid:100)(cid:22)(cid:39)(cid:22)
`40 (cid:22) 18†
`51 (cid:22) 17†
`64 (cid:22) 19†
`166 (cid:22) 41
`25 (cid:22) 14†
`31 (cid:22) 27†
`36 (cid:22) 29†
`78 (cid:22) 40
`83 (cid:22) 28†
`88 (cid:22) 33†
`105 (cid:22) 36†
`137 (cid:22) 30
`37 (cid:22) 20†
`39 (cid:22) 26†
`49 (cid:22) 29†
`109 (cid:22) 35
`Abbreviations: LA, low-amplitude protocol; LF, low-frequency protocol; SP, short pulse duration proto-
`col; Std, standard protocol.
`* Knee extensor torque (Nm) was measured at the beginning of each minute (contractions 1, 11, 21) and
`for the final contraction (30). Values are mean (cid:22) SD Nm.
`† Significantly different from the initial contraction (P(cid:12).05).
`
`active area was calculated by dividing the
`highest torque (Nm) achieved for each
`NMES protocol by the total activated
`skeletal muscle area (cm2).9,16,17 The torque
`and the activated CSA values in response
`to the 4 NMES protocols were previously
`measured and published.17 Considering
`the clinical purpose of the current study,
`we recalculated these values and present-
`ed them as torque relative to the activated
`CSA (Nm/cm2): torque per active area =
`peak torque of the first contraction/the
`activated knee extensor CSA. The activat-
`ed CSA was measured using T2 magnetic
`resonance imaging (MRI).
`Magnetic Resonance Imaging Standard
`spin echo images of the thighs were col-
`lected using a Signa 1.5-T superconducting
`magnet (General Electric Company, Mil-
`waukee, WI) ((cid:60)(cid:63)(cid:61)(cid:75)(cid:72)(cid:59)(cid:22)(cid:41)). After 30 minutes of
`lying down supine to avoid body fluid shift,
`subjects were positioned within the mag-
`net using the whole body coil. Transaxial
`images were obtained before NMES, and
`the participant was then moved out of the
`magnet to a separate room to perform the
`NMES protocols. After each NMES pro-
`tocol, the subject was asked to walk to the
`MRI unit without bearing weight on the
`stimulated lower extremity so as to repeat
`the imaging within 3 minutes after end-
`ing the electrical stimulation. The total
`time of the scan was around 4 minutes
`and 40 seconds. The scout view time and
`subsequent imaging adjustments (mean
`(cid:22) SD, 2 minutes (cid:22) 23 seconds) made the
`total imaging time almost 7 minutes. The
`transaxial T2 images (TR/TE = 2000/30,
`
`(cid:60)(cid:63)(cid:61)(cid:75)(cid:72)(cid:59)(cid:22)(cid:41)(cid:36)(cid:22)Representative anatomically matched
`axial T2 magnetic resonance images from the
`mid-thigh region of the knee extensor muscle group
`before (A) and immediately after (B) stimulation with
`the Std protocol. Letters R and L denote the right and
`left thighs, respectively. Note the activation on the R
`side immediately after stimulation.
`
`60) were 1 cm thick and 1 cm apart. They
`had a 40-cm field of view, with a 256 (cid:19)
`256 matrix size, and the number of excita-
`tions was 1. Fourteen to 18 slices for each
`subject were analyzed for the knee exten-
`sors, beginning with the first slice contain-
`ing the 4 heads of the quadriceps femoris
`muscle group, and continued distally until
`the slice just before the proximal pole of
`the patella. Images were analyzed and T2
`values calculated with the NIH Image 1.62
`software.16,17
`
`(cid:58)(cid:87)(cid:106)(cid:87)(cid:22)(cid:55)(cid:100)(cid:87)(cid:98)(cid:111)(cid:105)(cid:95)(cid:105)
`Data were analyzed using a 2-way (proto-
`cols by contractions), repeated-measures
`
`200
`
`160
`
`120
`
`80
`
`40
`
`0
`
`Torque (Nm)
`
`First Contraction
`
`Last Contraction
`
`Std
`LF
`
`SP
`LA
`
`(cid:60)(cid:63)(cid:61)(cid:75)(cid:72)(cid:59)(cid:22)(cid:40)(cid:36)(cid:22)Force traces of the 3-second first and last
`contractions of electrically evoked torque for the 4
`stimulation protocols. For the last contraction, Std
`and LF were only labeled for the purpose of clarity.
`Abbreviations: LA, low-amplitude protocol; LF, low-
`frequency protocol; SP, short pulse duration protocol;
`Std, standard protocol.
`
`not exceed 10 minutes after any of the 4
`protocols. Therefore, the 2-hour interval
`provided between the 2 protocols was
`enough to ensure full recovery of force of
`the same knee extensors.
`Peak Torque Peak isometric torque was
`reported as the average torque over a
`500-millisecond window. The window
`began after the contraction rose above
`baseline and recorded torque from 250
`to 750 milliseconds ((cid:60)(cid:63)(cid:61)(cid:75)(cid:72)(cid:59)(cid:22)(cid:40)). The peak
`torque was reported at the beginning of
`each minute (contractions 1, 11, 21) of the
`3-minute session and for the final con-
`traction (contraction 30) ((cid:74)(cid:55)(cid:56)(cid:66)(cid:59)(cid:22) (cid:40)). The
`fatigue index was measured and reflects
`the difference between the torques of the
`initial and final contractions divided by
`the torque of the initial contraction2,3:
`percent fatigue = ([torque of the first con-
`traction – torque of the last contraction]/
`torque of the first contraction) (cid:19) 100.
`Torque-Time Integral (TTI) The TTI of
`the first contraction was measured and
`was used as an index for the force gener-
`ated during the 3-second isometric con-
`tractions of the 4 NMES protocols.29,30
`The TTI of the first contraction was ad-
`justed to the activated cross-sectional
`area (CSA) to determine its possible role
`in muscle fatigue.
`Torque per Active Area The torque per
`
`journal of orthopaedic & sports physical therapy | volume 39 | number 9 | september 2009 | 687
`
` Journal of Orthopaedic & Sports Physical Therapy®
`
` Downloaded from www.jospt.org at on August 3, 2021. For personal use only. No other uses without permission.
`
` Copyright © 2009 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
`
`LUMENIS EX1014
`Page 4
`
`

`

`[ RESEARCH REPORT ]
`
`*
`
`*
`
`*
`
`600
`
`500
`
`400
`
`300
`
`200
`
`100
`
`0
`
`Torque-Time Integral (Nm·s)
`
`Std
`
`SP
`
`LF
`
`LA
`
`NMES Protocols
`
`(cid:60)(cid:63)(cid:61)(cid:75)(cid:72)(cid:59)(cid:22)(cid:42)(cid:36)(cid:22)Torque-time integral for the first contraction for the 4 NMES protocols. *Significantly different from
`Std. Values are mean (cid:22) SD. Abbreviations: LA, low-amplitude protocol; LF, low-frequency protocol; NMES,
`neuromuscular electrical stimulation; SP, short pulse duration protocol; Std, standard protocol.
`
`*†
`
`*†
`
`*†
`
`1.2
`
`1
`
`0.8
`
`0.6
`
`0.4
`
`0.2
`
`0
`
`Torque Normalizied to Initial Contraction
`
`0
`
`10
`
`20
`
`30
`
`Contraction Number
`
`Std
`
`SP
`
`LF
`
`LA
`
`(cid:60)(cid:63)(cid:61)(cid:75)(cid:72)(cid:59)(cid:22)(cid:43)(cid:36)(cid:22)Torque for each contraction was normalized to the initial contraction. Values are mean (cid:22) SD. *LF
`was different from Std, SP, and LA (P(cid:12).01). †Decline in torque over repeated contractions for Std, SP, LF, and
`LA (P(cid:12).0001). Abbreviations: LA, low-amplitude protocol; LF, low-frequency protocol; SP, short pulse duration
`protocol; Std, standard protocol.
`
`ber interaction (F9,54 = 13.2, P(cid:12).0001)
`was observed, with differences between
`contractions 11, 21, and 30 for both the
`Std and LF protocols (P(cid:12).02), suggest-
`
`ing that lowering the frequency could
`enhance performance over repeated con-
`tractions. No differences in the decline of
`peak torque over repeated contractions
`
`analysis of variance (ANOVA) to examine
`the effects of the 4 NMES protocols on
`muscle fatigue. The independent vari-
`ables were the protocols (Std, SP, LF,
`LA), the contraction numbers were 1, 11,
`21, and 30, and the dependent variable
`was peak torque. If there was an inter-
`action, alpha level was adjusted for pair-
`wise comparison using the Bonferroni
`correction. A 1-way ANOVA was per-
`formed to compare the difference in TTI
`of the 4 NMES protocols. Simple linear
`regression was used to examine the rela-
`tionship between the selected variables
`(percent fatigue and torque per active
`CSA). Statistical difference was set at a
`level of P(cid:12).05, and values were presented
`as means (cid:22) SD.
`
`(cid:72)(cid:59)(cid:73)(cid:75)(cid:66)(cid:74)(cid:73)
`
`(cid:74)he mean (cid:22) SD current ampli-
`
`tudes for Std, SP, LF, and LA pro-
`tocols were 74 (cid:22) 18, 76 (cid:22) 16, 72 (cid:22)
`18, and 56 (cid:22) 13 mA, respectively ((cid:74)(cid:55)(cid:56)(cid:66)(cid:59)(cid:22)(cid:39)).
`The Std, SP, LF, and LA protocols evoked
`mean (cid:22) SD percents of MVIT of 74% (cid:22)
`3%, 31% (cid:22) 12%, 60% (cid:22) 8%, and 45% (cid:22)
`9%, respectively ((cid:60)(cid:63)(cid:61)(cid:75)(cid:72)(cid:59)(cid:22)(cid:40)). The influence
`of the 4 NMES protocols on the evoked
`torque, activated area, and torque per
`active area are summarized in (cid:74)(cid:55)(cid:56)(cid:66)(cid:59)(cid:22) (cid:39).
`The 1-way ANOVA revealed a significant
`difference in the TTI among the 4 pro-
`tocols (P(cid:12).0001). TTI was significantly
`higher for the Std protocol compared to
`the SP (P(cid:12).0001), LF (P(cid:12).035), and LA
`(P(cid:12).0001) protocols ((cid:60)(cid:63)(cid:61)(cid:75)(cid:72)(cid:59)(cid:22)(cid:42)). After ad-
`justing for the activated CSA, mean (cid:22) SD
`TTIs were 16 (cid:22) 4 and 10 (cid:22) 4 Nm·s/cm2
`for the Std and LF protocols, respectively
`(P = .014).
`(cid:60)(cid:63)(cid:61)(cid:75)(cid:72)(cid:59)(cid:73)(cid:22)(cid:40) and 5 illustrate the decline in
`the evoked torque for the 4 NMES pro-
`tocols. For all 4 protocols, there was a
`significant reduction in torque from the
`initial contraction (F3,18 = 12, P(cid:12).009).
`The LF protocol resulted in less fatigue
`when compared to the other 3 protocols
`(mean (cid:22) SD percent MVIT, 39% (cid:22) 19%
`versus 76% (cid:22) 10%; F1,6 = 85.2; P(cid:12).001). A
`significant protocol-by-contraction num-
`
`688 | september 2009 | volume 39 | number 9 | journal of orthopaedic & sports physical therapy
`
` Journal of Orthopaedic & Sports Physical Therapy®
`
` Downloaded from www.jospt.org at on August 3, 2021. For personal use only. No other uses without permission.
`
` Copyright © 2009 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.
`
`LUMENIS EX1014
`Page 5
`
`

`

`100
`
`80
`
`60
`
`40
`
`20
`
`0
`
`Percent Fatigue
`
`0
`
`2
`
`4
`
`6
`
`8
`
`Torque per Active Area (Nm/cm2)
`
`(cid:60)(cid:63)(cid:61)(cid:75)(cid:72)(cid:59)(cid:22)(cid:44)(cid:36)(cid:22)Percent fatigue versus torque per active area for Std (orange circle) and LF (blue circle) NMES protocols
`(n = 7 per protocol). Percent fatigue = 13.25 (torque per active area) – 6.086 (r2 = 0.57, P = .002).
`
`er NMES parameters constant modestly
`increased fatigue.8 This suggests that as
`the current amplitude increases, more
`fast-twitch motor units are recruited,
`resulting in greater fatigue due to their
`higher metabolic demand in comparison
`to slow-twitch motor units. We found no
`such response in this study, because the
`extent of fatigue was independent of the
`current amplitude. Our results suggest
`that as the current amplitude is increased,
`fast- and slow-twitch motor units are ran-
`domly recruited.21,24 Our recent findings
`are in agreement with previous findings
`from our laboratory. Adams et al1 showed
`that increasing the current amplitude
`from that required to evoke 25% to 75%
`of MVIT did not alter fatigue. Slade et al34
`also showed that moderate versus high-
`amplitude protocols resulted in similar
`fatigue. Their findings could be explained
`by the fact that current amplitude does
`not affect specific tension. Increasing the
`current amplitude increased the evoked
`torque, which was associated with in-
`crease in the recruited muscle area and
`maintained the metabolic demand per
`activated motor units.17
`
`(cid:70)(cid:107)(cid:98)(cid:105)(cid:91)(cid:22)(cid:58)(cid:107)(cid:104)(cid:87)(cid:106)(cid:95)(cid:101)(cid:100)
`Compared to the Std protocol, a pulse
`duration of 150 μs showed no difference
`in skeletal muscle fatigue during repeat-
`
`ed stimulation. Currently, the indepen-
`dent effect of pulse duration on muscle
`fatigue is not clear, yet it has been shown
`that pulse duration modulation has less
`effect on muscle fatigue than frequency
`modulation.22 Previously, the effect of the
`product of the frequency and pulse dura-
`tion was tested on muscle fatigue. After
`matching the initial peak torques, a com-
`bination of 20 Hz and 500 μs produced
`less fatigue compared to 50 Hz and 200
`μs.19 In our study, we observed puzzling
`effects of pulse duration on torque per
`active area and muscle fatigue. As re-
`vealed in (cid:74)(cid:55)(cid:56)(cid:66)(cid:59)(cid:22)(cid:39) and (cid:60)(cid:63)(cid:61)(cid:75)(cid:72)(cid:59)(cid:22)(cid:43), long pulse
`duration modestly increased torque per
`active area but not muscle fatigue, which
`runs contrary to the main hypothesis.
`This effect may have occurred because
`both pulses (150 and 450 μs) were ap-
`plied at 100 Hz; this means that the im-
`pact