Twitch interpolation of the elbow flexor muscles at high forces

Twitch interpolation of the elbow flexor muscles at high forces
of 11
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  ABSTRACT: We investigated factors affecting maximal voluntary torqueand the assessment of the level of voluntary drive in the elbow flexormuscles. First, the effective compliance of the system was tested by usingsingle, paired, and trains of four stimuli to measure voluntary activation. Athigh voluntary torques the responses to all these stimuli were identical,suggesting that single stimuli are adequate for estimating voluntary drive.Second, the contribution of torque from synergist elbow flexor muscles wasassessed. In attempted maximal voluntary contractions (MVCs), the volun-tary activation of brachioradialis (median 91.5%, range 68.9–100%) waslower than for biceps brachii (median 99.1%, range 78.5–100%;  P   < 0.01).This suggests extra torque may be generated by brachioradialis during el-bow flexion, beyond the torque where biceps brachii is maximally activated.Finally, lengthening of the elbow flexors occurred during MVCs, due to slightshoulder movements. This would allow force to increase independently of anincrease in voluntary drive.  © 1998 John Wiley & Sons, Inc.  Muscle Nerve   21: 318–328, 1998 Key words: maximal voluntary contraction; muscle contraction; voluntarymovement TWITCH INTERPOLATION OF THEELBOW FLEXOR MUSCLES ATHIGH FORCES GABRIELLE M. ALLEN, BSc(Hons), 1,3 DAVID K. M C KENZIE, MBBS, PhD, 2,3 andSIMON C. GANDEVIA, MD, DSc 1,3 1 Department of Clinical Neurophysiology, The Prince Henry and Prince of WalesHospitals, Sydney, Australia 2 Department of Respiratory Medicine, The Prince Henry and Prince of WalesHospitals, Sydney, Australia 3 Prince of Wales Medical Research Institute, University of New South Wales, HighStreet, Randwick NSW 2031, Sydney, Australia  Received 13 March 1997; accepted 31 August 1997  C an the maximal force from a muscle group beachieved using voluntaryand reflex drives during anattempted maximal voluntary contraction? Biglandand Lippold 7 and Merton 23 showed independentlythat tetanic stimulation of the ulnar nerve produceda similar force for adduction of the thumb as did anattempted maximal voluntary contraction. Merton 23 extended his findings for adductor pollicis by show-ing that interpolation of a single electrical stimulusto the ulnar nerve during a maximal voluntaryeffortadded no detectable force. Thistechnique, known as‘‘twitch interpolation’’ appeared to confirm that allrelevant motoneuronswere recruited and producingmaximal force.However, fewmuscles can be studied in isolationwith the twitch interpolation technique. It is also dif-ficult to apply to a muscle group because the indi-vidual synergist muscles rarely share the same inner-vation, or if they do, antagonist muscles may also bestimulated. For example, the ulnar nerve innervatesantagonists to most of the intrinsic muscles of thehand, and the common peroneal nerve innervates atleast one plantar flexor of the ankle in addition to alldorsiflexors.Since Merton’s experiment, twitch interpolationhas been applied to a variety of muscles. Well-motivated subjects can achieve near complete volun-tary activation during attempted maximal contrac-tions of most muscle groups during attemptedmaximal contractions, with the exception of theankle plantar flexors (e.g., see Refs. 1, 4, 6, 12, 13,and 25). The technique has been used in the clinicalsetting by a number of investigators. It allows themaximal force of a muscle to be measured accurately *Correspondence to: Dr. Simon C. GandeviaContract grant sponsor: National Health and Medical Research Council ofAustraliaCCC 0148-639X/98/030318-11 © 1998 John Wiley & Sons, Inc. 318  Voluntary Drive to the Elbow Flexors MUSCLE & NERVE March 1998  and therefore provides a quantitative method formonitoring a patient’s strength (e.g., Refs. 2, 4, and24). The technique has the advantage of allowingthe central drive to the muscle to be determined, inaddition to the peripheral force-generating capacityof the muscle.However, the accuracy of estimates of voluntarydrive obtained using this technique are dependenton the exact methods employed. One limitation withtwitch interpolation is that it is often not used withsufficient sensitivityto detect failures in the abilitytoattain the maximal evocable force (see Fig. 2 in Ref.15). With sensitive analysis of the force incrementsadded by interpolated stimuli to limb muscles, it isclear that complete or maximal voluntary activationis the exception rather than the rule during at-tempted maximal voluntary isometric contractions,even in well-motivated subjects using feedback of force (e.g., see Refs. 13, 19, and 22; see also Ref. 25).When subjects were studied on five separate occa-sions the median maximal voluntary activation of their elbow flexors was 98%, and in only 25% of attempted maximal efforts were they able to achieve100% voluntary activation. 1 A second limitation of the twitch interpolationtechnique is that when the interpolated stimulus isapplied to only one muscle of a synergist group, thedegree of voluntaryactivation of the other synergists(i.e., those not stimulated, but that still contribute totorque) is not assessed. The influence of synergistson the variability in voluntary activation and forceproduction is unknown. Depending on the musclegroup and its innervation, the possible unintend-ed stimulation of antagonists is also not usuallyconsidered.At submaximal forces, there is an inverse rela-tionship between the force evoked by an interpo-lated stimulus to a nerve innervating a muscle (or tointramuscular nerve fibers) and the voluntary forceproduced by the muscle group. 4,5,14 However, athigh voluntary torques this relationship is nonlinearfor many muscle groups including dorsiflexors of the ankle, 4 diaphragm, 5,14 and biceps brachii. 9 Thus,voluntary activation appears to be nearly maximalwhen voluntary force is less than maximal. As theexact methodologyis important for the accuracyandinterpretation of the results, we investigated threefactors which may contribute to this observation.First, we examined whether the compliance of the system affected the assessment of voluntarydriveat these high levels of voluntary torque, by compar-ing results obtained with single, paired, or a train of four stimuli. Second, we investigated voluntary acti-vation of a synergist muscle to determine the level of its recruitment relative to that of biceps brachii dur-ing elbowflexion. Finally, we investigated whether ornot small changes in muscle length occur during thequasi-isometric contractions which may influencevoluntary force independently of voluntary drive. METHODS Studies were performed on several occasions on 9subjects (5 males, 4 females) aged 21–40 years. Sub- jects were free of neuromuscular disease and pro-vided informed consent. All procedures were ap-proved by the institutional ethics committee. Force Recording.  Subjects were seated comfortablywith the right forearm fixed to a vertical isometricmyograph just proximal to the wrist. The wrist wasattached to the myograph through a felt-lined plasticmolding, and fastened with Velcro (approximately2.5 cm in width) or in some experiments by a foam-padded canvas strap fastened with Velcro (approxi-mately 5 cm in width). The elbow was flexed at 90degrees and the forearm supinated. Contractileforce was measured in different experiments usingeither a bridge circuit based on four strain-gaugescemented to the myograph or by a sensitive loadcell. 1 Both systems were linear within the ranges of torque encountered. The load cell was linear to 2kN, and the strain gauge was linear to greater than500 N. Responses are expressed as torque about theelbow (Nm). Motor Point Stimulation.  Electrical stimuli were de-livered to intramuscular nerve fibers of the testedmuscle group (either over biceps brachii or brachio-radialis) via a cathode located over the motor pointof the muscle. The anode was a similar electrodepositioned over the distal tendon.For biceps brachii/ brachialis the cathode was lo-cated over biceps brachii (midway between the an-terior edge of deltoid and the proximal elbowcreasewith the elbowflexed to 90 degrees), with the anodeon the skin over the bicipital tendon. Similartwitches were obtainable with the cathode locatedover the musculocutaneous nerve just medial to theanterior deltoid at the axilla, with an anode 70 mmdistal. For stimulation of brachioradialis, a cathode(15 mm wide, 10 mm long) waslocated over brachio-radialis at the point of lowest threshold for stimula-tion of the muscle. The anode was a similar elec-trode positioned 6–7 cm distal along brachioradialis(10 mm wide, 10 mm long). The motor point waslocated first using a roving cathode, and was usuallylocated about 7 cm distal to the elbow crease.Electrodes consisted of aluminium foil (10 mm Voluntary Drive to the Elbow Flexors MUSCLE & NERVE March 1998  319  by 10 mm) covered in gauze and soaked in salineand conductive paste. Stimuli were rectangularpulses of 100-µs duration from a constant-voltagesource (Devices 3072, 0–400 V) or constant-currentstimulator (modified Digitimer DS7, 0–1000 mA).Intensity was increased until twitch amplitude was10–20% supramaximal. At the usual intensity (120–180 V, 100–200 mA) there was no detectable activa-tion of distal remote muscles innervated by the me-dian or radial nerve trunks. Such activation wasdetected bythe observation of finger flexion, musclepalpation, and a distortion in the shape of the twitchprofile. Twitch Interpolation.  Twitch interpolation is per-formed by delivering a supramaximal electricalstimulus to the nerve trunk or intramuscular nervefibers innervating the test muscle during a voluntarycontraction. Responses evoked by this stimulus are‘‘interpolated’’ twitches, which are superimposed onthe level of voluntarytorque at the moment of stimu-lation.In these experiments, torque was sampled at 200Hz during each voluntary effort. Above a set thresh-old [approximately 10 Nm below each subject’smaximal voluntarycontraction (MVC)] and after thepeak force, the sampling rate increased to 1 kHz toenable high resolution of any responses evoked bythe stimuli. A sample-and-hold amplifier, which ze-roed the raw torque and amplified it by 10, was trig-gered 25 ms prior to the delivery of the stimuli. 16 The amplified output was filtered at 500 Hz andoccupied a 12-bit analog-to-digital (A-D) converter(4096 A-D numbers). Therefore small evoked twitchresponses superimposed on top of the maximal vol-untary force could be reliably detected. 1,16 Twitchesas small as 1% of the amplitude of the twitch evokedbythe same stimulus in the relaxed muscle could bedetected. 1 These small responses had a time-to-peak that was greater than 5 ms and occurred at the ap-propriate time after the deliveryof the stimulus (i.e.,after the electromechanical delay).The interpolated stimuli were occasionally deliv-ered during a rapidly changing background force.Responses to approximately 5% of stimuli were un-able to be measured accuratelyand were eliminated.Customized software allowed traces with rapid in-creases or decreases in torque prior to the stimulusto be corrected. The correction worked by extrapo-lating the slope of the prestimulus torque beyondthe point of the stimulus. Then this poststimulus ex-trapolated torque was subtracted from the actualpoststimulus torque to give the ‘‘corrected’’ evokedresponse (asif the trace wasrotated). In 1 subject thetime period over which the prestimulus slope wasmeasured was varied from 25 to 800 ms. However,there were no differences in the mean evoked re-sponses after correction, during 19 MVCs for thesevarious times over which slope was determined.Hence, in the present studies, only 25 ms of thetorque data prior to the stimulus was routinely re-viewed at high gain.Voluntary torque and the responses to stimula-tion were measured by computer on-line and alsomonitored with a digital oscilloscope. For voluntarycontractions with interpolated stimuli the computerstored and displayed the peak voluntary torque, thetorque at the time when the stimulus was delivered,and the absolute increase in torque produced bythestimulus (i.e., superimposed twitch amplitude) to-gether with its latency. Twitch amplitude and time-to-peak tension were calculated for interpolated andcontrol twitches. Data were stored for reanalysis andall twitches, including those superimposed on volun-tary contractions, were reviewed off-line to check measurements using cursors and customized soft-ware. Calculation of Voluntary Activation.  To calculatethe level of voluntary activation during a contrac-tion, any increment in torque evoked during thecontraction was expressed as a fraction of the ampli-tude of the mean response evoked by the samestimulus in the relaxed muscle (‘‘control’’ evokedresponse). The level of drive was then quantified asa percentage: [1 − evoked torque (superimposed onvoluntary torque)/ (mean control evoked response)× 100]. 1,5,21 Protocol.  Three experiments were performed onthree separate days. On each daysubjects performeda series of MVCs of the elbow flexors. Each MVClasted 2–3 s and was separated from the subsequentmaximal effort by2–3 min. The latter interval variedbetween subjects and was chosen to minimize anyreduction in force due to peripheral muscle fatigue.Supramaximal stimuli were delivered over the testmuscle during each maximal effort at the peak forceto determine the level of voluntary activation.About 5 safter each MVC, when the elbowflexorswere relaxed, a second supramaximal stimulus wasdelivered over the test muscle. This evoked a ‘‘con-trol’’ twitch of the stimulated muscle which was po-tentiated by the preceding MVC. The amplitude of the control twitch was monitored to check that noperipheral fatigue had occurred. Subjects received a 320  Voluntary Drive to the Elbow Flexors MUSCLE & NERVE March 1998  visual display of force, and loud verbal encourage-ment accompanied all attempted ‘‘maximal’’ efforts. Voluntary Activation Assessed Using Single, Paired,and a Train of FourStimuli.  Four subjectsperformedthe first experiment. This was designed to investigatethe influence of different stimuli on the sensitivityof the technique and to determine whether the serieselastic component (compliance) of the testing sys-tem would influence the relationship between vol-untary and evoked torque. In particular, responsesevoked at high voluntary forces were analyzed to de-termine whether multiple stimuli were more sensi-tive for detecting slight failures in voluntary activa-tion than single supramaximal stimuli. With pairedstimuli, the force evoked at rest is approximatelydoubled, and therefore the signal-to-noise ratio isimproved for the detection of small evoked changesin twitch forces at high voluntarytorques. The trainsof four stimuli further improve the signal-to-noiseratio. If the series elastic element (or measurementerror) were an important influence, its effect shouldbe reduced by multiple stimuli.Initially, all subjectsperformed 3 MVCsseparatedby1 min. Then subjects performed 30 brief maximalcontractions, each followed by a stimulus to the re-laxed muscle (‘‘control’’ twitch or response), andthen a submaximal effort. The submaximal contrac-tion was performed approximately10 s after the con-trol response and was a predetermined percentageof the maximal voluntary force, usually 25%, 50%,75%, 90%, or 100% MVC (see Fig. 1A). The differ-ent levels of contraction were presented in randomorder, and an interpolated supramaximal stimuluswas delivered during the effort, 1–2 s after the onsetof the contraction, when the required force level wasreached.In this experiment, either a single stimulus,paired stimuli (interstimulus interval of 10 ms), or atrain of four stimuli (interstimulus interval 20 ms, 50Hz) were delivered over the muscle in a set of three.The first was during a MVC, the second was in therelaxed muscle, and the third was during the subse-quent submaximal effort (either 25%, 50%, 75%, or90% MVC). Both the type of stimuli and the level of submaximal contraction were randomlyassigned be-tween each set. This experiment enabled a similarprotocol to be repeated with different stimuli on thesame day, using the same electrodes. At least 2 minrest was allowed between each set of three stimuli(i.e., stimuli delivered during MVC, in the relaxedmuscle, and in the subsequent voluntaryeffort at thespecified percentage of MVC). Voluntary Activation of Brachioradialis during ElbowFlexion.  The second experiment was performed todetermine whether elbow flexor muscles other thanbiceps brachii contribute greater amounts of torqueat high levels of elbowflexor torque (see also Ref. 8).Initially, intramuscular electromyographic (EMG)electrodes were inserted into the four main elbowflexors to determine whether slight variations in theneural drive between muscles could be detected.However, the high frequency content of the EMGmeant that the signals were insufficientlysensitive todetect small differences in neural drive. Instead, we FIGURE 1. (A)  Protocol. Subject initially performed a maximalvoluntary contraction (MVC) during which a supramaximal elec-trical stimulus (at arrow) was delivered over the motor point ofbiceps brachii at the peak force. Approximately 5 s after thisMVC, a second stimulus of the same intensity was delivered overthe relaxed muscle to evoke a ‘‘control’’ twitch. Then approxi-mately 10 s after the control twitch the subjects performed asubmaximal effort (25, 50, 75, 90, or 100% MVC, chosen ran-domly) during which a stimulus was given at the target force.Each set of contractions was separated by 2–3 min.  (B)  Re-sponses evoked by supramaximal stimuli (at arrow) over bicepsbrachii at different levels of voluntary torque (offset to zero). Asvoluntary torque increases, the response evoked by the stimulusdecreases, and the time to peak of the evoked response short-ens. The largest response (control twitch) was evoked from thecompletely relaxed muscle when there is zero voluntary torque.In this example, no response is evoked by the stimulus at maxi-mal voluntary torque.Voluntary Drive to the Elbow Flexors MUSCLE & NERVE March 1998  321  used twitch interpolation to compare the level of voluntaryactivation of biceps brachii with a synergistelbow flexor, brachioradialis, during attemptedmaximal contractions.In this experiment, single electrical stimuli weredelivered over the elbow flexors innervated by thedistal radial nerve (brachioradialis and extensorcarpi radialis longus). Separate studies using micro-neurographic techniques to stimulate within the ra-dial nerve fascicles innervating extensor carpi radia-lis longus confirmed that this muscle contributes toelbow flexion in the posture used here. Responseswere compared to those obtained with stimuli overthe usual biceps brachii site.In 5 subjects up to 42 attempted maximal flex-ions of the elbowwere performed, each separated byat least 2 min. During each MVC, a supramaximalstimulus was delivered to either the brachioradialisor the biceps brachii, in random order. Approxi-mately5 s after each maximal effort, the same stimu-lus was delivered over the relaxed muscle. This wasto evoke a control response for that site of stimula-tion which was used for the calculation of voluntaryactivation (see above). Changes in Muscle Length during Attempted ElbowFlexion.  Small changes in length of the elbowflexormuscle fibers may occur as a result of slight adjust-ments of the position of the shoulder during theattempted maximal contractions. Lengthening orshortening of the elbowflexor muscles during quasi-isometric contractions would place them at differentpositions on the length–tension and force–velocityrelationship curves, which may be more or less ad-vantageous. This modifies force produced duringthe contraction independentlyof a change in volun-tary activation. The possibility that this mechanismmight account for some of the variation in voluntarytorque at high voluntary forces was investigated inthe third experiment in 4 subjects.Changes in length of the elbowflexors from con-traction to contraction and within a contractionwere measured indirectly by monitoring verticalmovement of the shoulder with linearized mag-netometers (Micronta U.S.A.). One magnetometerwas positioned in a foam shoulder pad taped to theskin, and the reference was fixed 15 cm above theshoulder on a rigid frame. As the forearm was fixedin a vertical position, with the upper arm horizontal,any vertical shoulder motion must alter elbow angleand hence the overall lengths of the elbow flexors.Given the rigid wrist coupling and lowcompliance of the myograph, a downward movement of the shoul-der represented lengthening of elbowflexors. Shoul-der movement and voluntary torque were recordedon disk during each contraction from a starting po-sition at rest. Data were stored for off-line analysis(1401 interface; Cambridge Electronic Design, Cam-bridge, U.K.).Each subject performed 15–20 MVCs duringwhich electrical stimuli were delivered over bicepsbrachii. Each MVC was followed by a ‘‘control’’twitch evoked in the relaxed muscle, and a submaxi-mal contraction (as for the first experiment, seeabove). Contractions were analyzed to determinewhether an initial shortening or lengthening of theshoulder occurred. The position of the shoulder wasmeasured at the time of stimulation and comparedwith the initial position of the shoulder when themuscle was relaxed.To estimate the shortening of muscle fibers thatwould correspond with the vertical movement of theshoulder, the change in length of brachialis, which isa major flexor of the elbow, 3 was calculated usingsimple trigonometry. These calculations were basedon the assumptions that: (a) the distance from theaxis of rotation of the elbowto the point of insertionof the brachialis muscle at the tuberosityon the ulnawas constant between subjects at 45 mm, (b) thesrcin of the brachialis muscle was 1/ 3 of the mea-sured distance from the axis of rotation at the elbowto the acromion, (c) the position of the elbow axisdid not change during MVCs, and (d) the brachialismuscle length was the same as the linear distancefrom its srcin to its insertion. Terminology and Statistics.  Throughout the textthe term maximal voluntary contraction (MVC) de-notes the maximal force produced when a subjectreceived force feedback and verbal encouragement,and when the subject felt he/ she had achieved amaximal effort. This force is often less than the larg-est possible force that may be obtained from themuscle, and the ‘‘true’’ maximal force is defined asthat achieved when no superimposed twitch couldbe discerned (i.e., the maximal force that the muscleis capable of generating).Relationships between voluntary and evoked el-bow flexor torque and between shoulder displace-ment and voluntary torque were determined usingSpearman’s rank correlation, as the data were non-parametric. Differences in the evoked responses toone, two, and four stimuli at high voluntary torqueswere assessed by the Friedman’s test. The differencebetween voluntary activation of biceps brachii andthat of the brachioradialis muscle was compared byusing the Wilcoxon matched-pairs signed-ranks test. 322  Voluntary Drive to the Elbow Flexors MUSCLE & NERVE March 1998
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks