[188] MUSCLE STRENGTH AND FUNCTIONAL PERFORMANCE IN PARKINSON'S DISEASE: A PILOT STUDY
Thomas Mathews, MD; Roger M. Glaser, PhD; Thomas W.J.
Janssen, PhD; D. Drew Pringle, EdD; Kathy D.Zoerb, PAC;
Trudy L.Cortez, PAC; José W. Almeyda, BS; William P.
Couch, BS
Veterans Affairs Medical Center, Dayton, Ohio 45428;
Institute for Rehabilitation Research and Medicine, Wright
State University School of Medicine, Dayton, Ohio 45420;
Rehabilitation Institute of Ohio, Miami Valley Hospital,
Dayton, Ohio 45409. email: mathews.thomas@dayton.va.gov;
rglaser@desire.wright.edu;
tjanssen@desire.wright.edu;
dpringle@desire.wright.edu
Sponsor: Department of Veterans Affairs, VA
Rehabilitation Research and Development Service, Washington,
DC 20420
(Pilot Project #B2064-PA)
PURPOSE--The purpose of this 1-year pilot study is to gain more insight into the different characteristics of muscular strength and force production in relation to functional performance of individuals with Parkinson's disease (PD), so as to improve validity of clinical evaluations. Specific objectives are to: 1) develop an objective and reliable evaluation method of upper- and lower-body voluntary isometric and isokinetic (concentric and eccentric) strength of persons with PD; 2) determine the influence of movement velocity and contraction type on the rate and pattern of force production; 3) evaluate differences in strength parameters and fatigability between persons with PD and age-matched able-bodied (AB) controls; 4) correlate muscle strength parameters to gait performance and daily activity for those with PD; and 5) correlate scores on conventional PD clinical rating scales to the objective measures obtained for muscle strength and gait performance.
METHODOLOGY--Sixteen (8 male, 8 female) sedentary persons with PD (Hoehn and Yahr stages 1-4) will be solicited from outpatient services at the Dayton VAMC. Care will be taken to include 4 subjects (2 male, 2 female) from each of the Hoehn and Yahr stages. Additionally, 16 sedentary (8 male, 8 female) nondisabled (AB) persons, matched for age, gender, race, and body mass index will be elicited and tested to evaluate differences in muscle strength characteristics between persons with and without PD. In order to eliminate activity/exercise level as a confounding variable, only sedentary subjects will be used. This results in a total of 32 subjects; 16 PD and 16 matched controls.
Following the medical examination, all subjects with PD will be evaluated and classified by use of both the Hoehn and Yahr Scale and the Unified Parkinson's Disease Rating Scale by an experienced neurologist. Subsequently, all PD subjects will perform a sit-stand-walk test to determine gross functional performance. Within 2 weeks, the PD and corresponding AB paired subjects will report for muscle strength testing. Additionally, on a separate day, a random sample of 2 PD subjects from each of the 4 stages (8 PD total) will report for gait analysis. This gait analysis will take place within 3 days following the strength testing session. Subjects selected for the gait analysis will be fitted with a Mini-Logger, a portable physiological data recorder, which will monitor heart rate and body movements during one typical day. Subjects will be instructed on the procedures for using the Mini-Logger during the strength testing session and will take the instrument home.
PROGRESS--The initial subject recruitment criteria and demographics for age-type matching have been established. The design, development, and construction of the finger and foot tapping devices were completed. Subjects with PD have performed the specific evaluation protocols and adjustments in the timing and procedures have been made. Technicians have been trained in the data collection process and data collections will begin in the near future.
IMPLICATIONS--With the development of improved evaluation techniques, interventions that may impede PD progression or prevent and/or minimize secondary impairments could be better tested for efficacy. A detailed database of normal and abnormal values for strength and functional performance could lead to an improvement in diagnostic accuracy using the developed quantitative measures combined with clinical assessment. This more objective evaluation technique could provide a more valid starting point on which to base interventions (medication, surgery, physical training) to reduce strength deficits and improve functional performance in persons with PD.
[189] BIOCHEMICAL AND MYOELECTRIC EVENTS DURING FATIGUE
Serge H. Roy, ScD; Edward Kupa, MS; Carlo J. De Luca,
PhD; Susan C. Kandarian, PhD; L. Donald Gilmore, ABEE
NeuroMuscular Research Center, Boston University,
Boston, MA 02215; email: sroy@bu.edu
Sponsor: Boston University, Boston, MA 02215; Liberty Mutual Insurance Company, Boston, MA 02117
PURPOSE--In vitro research studies are being conducted using a rat animal model to provide a better understanding of how the biochemical events associated with fatigue influence the median frequency and conduction velocity estimates of the electromyographic (EMG) signal. This approach can accurately measure biochemical correlates of fatigue while controlling for processes that typically confound in vivo studies in humans. Our studies are currently focused on how the fiber type composition of a muscle influences the EMG signal and its spectral parameters. These results are a first step toward predicting the fiber type percentages in humans using similar surface EMG procedures.
METHODOLOGY--Whole muscles and nerves are surgically removed and placed in a test chamber in which the muscle temperature, oxygenation, and extracellular ionic fluids are maintained. EMG signals and isometric twitch and tetanic forces during elicited contractions are detected and sampled by a personal computer workstation. Muscles are then prepared for later histochemical analysis and fiber typing.
PRELIMINARY RESULTS--In the past year we have doubled the number of neuromuscular preparations studied from two different hindlimb muscles of the rat and the diaphragm muscle. Results from eight specimens of each muscle show that muscles with fast glycolytic enzyme content produce M-waves that are modified to a greater extent during fatigue than muscles with slow oxidative enzyme content. Time and amplitude scaling coefficients using wavelet analysis were significantly different for these muscles. Differences in the initial values and rate of decay of the median frequency and conduction velocity paralleled the distinct differences in the fiber type composition of these muscles. We were able to estimate the fiber type percentages of these muscles to a high degree of accuracy using just the EMG parameters.
[190] EFFECTS OF MUSCLE FIBER SIZE ON EMG PARAMETERS
Edward Kupa, MS; Serge H. Roy, ScD; Susan C. Kandarian,
PhD; Carlo J. De Luca, PhD
NeuroMuscular Research Center, Boston University,
Boston, MA 02215; email: sroy@bu.edu
Sponsor: Boston University, Boston, MA 02215
PURPOSE--Empirical studies are needed to validate the use of surface electromyographic (EMG) techniques to noninvasively quantify muscle atrophy. We hope to achieve this goal by studying the relationship between parameters describing EMG signal characteristics and measurements of muscle fiber cross sectional area (CSA). Previous studies by others have been successful in this regard, but were limited primarily to isolated, single fibers. We have expanded upon these earlier studies by detecting EMG signals from the surface of whole muscles, a procedure which more closely approximates the typical application of surface EMG procedures for human studies.
METHODOLOGY--Isolated whole muscle sections from the rat were studied in vitro and EMG signals and muscle force were recorded from the soleus, diaphragm, and extensor digitorum longus muscles during tetanic contractions. The average muscle fiber CSA and muscle fiber type were measured afterward by standard histochemical methods.
PRELIMINARY RESULTS--Data for all three muscles were pooled to provide a sample of muscle fiber CSA covering a broad range of values. In contrast to previous findings by others using single fiber techniques, the muscle fiber CSA in our study was unrelated to either the initial median frequency or the initial conduction velocity. However, when the proportional differences in fast fiber type for these three muscle groups were accounted for by calculating a weighted measure of CSA, a significant, positively correlated relationship was observed. This finding demonstrates that the proportional area of fast fibers in a whole muscle has a strong influence on surface EMG median frequency and conduction velocity. The results suggest that surface EMG median frequency and conduction velocity measurements from whole muscle are influenced by the combined effects of muscle fiber size and type on muscle membrane depolarization and propagation.
[191] MUSCLE ADAPTATION FOLLOWING LIMB UNLOADING AND ITS INFLUENCE ON EMG PARAMETERS
Edward Kupa, MS; Serge H. Roy, ScD; Susan C. Kandarian,
PhD; Carlo J. De Luca, PhD
NeuroMuscular Research Center, Boston University,
Boston, MA 02215; email: sroy@bu.edu
Sponsor: Boston University, Boston, MA 02215
PURPOSE--Electromyographic (EMG) signal analysis may provide a noninvasive method of monitoring muscle adaptations associated with disuse. However, further information must be gained to determine how specific changes in muscle fiber characteristics associated with disuse affect the EMG signal. In this study, in vitro techniques were applied to compare the changes in muscle fiber type with alterations in the EMG signal from muscles exposed to long- and short-term muscle unloading.
METHODOLOGY--A "tail-suspension" method of limb unloading was used to prevent weightbearing of rat muscles. Animals were suspended for either 7 or 21 days to induce atrophic and fiber type changes in the muscle fibers. Neuromuscular preparations of the soleus and extensor digitorum longus muscle were surgically removed from the hindlimb and placed immediately in an in vitro oxygenated bath. EMG signals were recorded directly from the muscle membrane during tetanic contractions. A control group which did not undergo unloading was also studied. Following EMG signal detection, muscles were histochemically analyzed to determine the muscle fiber cross sectional area and type.
PRELIMINARY RESULTS--Preliminary results demonstrated that the initial median frequency of the EMG signal was significantly decreased in unloaded muscles compared to controls. There was also a marked decrease in peak to peak amplitude of the signals from unloaded muscles. These signal changes reflect the extreme reduction in muscle fiber cross-sectional area caused by hindlimb unloading. Muscle fiber type percentages in unloaded muscle were shifted toward a greater proportion of fast fiber type. Unexpectedly, this did not result in a concomitant increase in median frequency. The rate of fatigue of the muscles, as measured by the decay in median frequency during a sustained contraction, remained unchanged.
[192] EFFECTS OF INTRAMUSCULAR APONEUROTOMY AND RECOVERY ON PENNATE SKELETAL MUSCLE
Reinald Brunner, MD; Peter A. Huijing, PhD; Richard
Jaspers; Johan Pel
Dept. of Pediatric Orthopaedics, Children's Hospital
Basel, University of Basel, Switzerland; Institute of
Fundamental and Clinical Movement Sciences, Faculty of Human
Movements Sciences, Vrije Universiteit, 1081 BT Amsterdam,
The Netherlands; Institute of Biomedical Technology,
University of Twente, 7500AE Enschede, The Netherlands;
email: P_A_J_B_M_Huijing@FBW.VU.NL
Sponsor: Ciba-Geigy Jubileum Stiftung, Basel
PURPOSE--Intramuscular aponeurotomy is used clinically either to lengthen muscle or to weaken overactive muscle. Little is known about effects on muscle physiology and mechanisms of muscle recovery after such a intervention. The purpose of this project is to evaluate acute and longer term effects and mechanisms by which these effects are reached.
METHODOLOGY--The medial gastrocnemius muscle of three groups of Wistar rats (age 15 weeks) was studied. Length force characteristics and muscle geometry were determined: 1) 6 weeks after proximal aponeurotomy at 50 percent length and 3 days immobilization in maximal dorsiflexion, 2) 6 weeks after sham operation and identical immobilization, and 3) 6 weeks of no special treatment (controls). After determining length force characteristics in groups 2 and 3, proximal aponeurotomy was performed to assess its acute effects.
PRELIMIMARY RESULTS--Acutely, aponeurotomy (group 2 and 3) caused a scaled length-force curve at approximately 55 percent (i.e., force decreased but length-force curves normalized for force showed only minor changes). This is compatible with the observation that the distal part of the muscle, despite being active, contributes only little to the muscle force.
After recovery for group 1, the length force curve shifted to higher muscle lengths (i.e. became longer) but optimum force recovered to values similar to controls. The severed ends of the aponeurosis were connected with scar tissue. Muscle geometry showed extensive alterations compared to controls.
IMPLICATIONS--Despite acute effect of decreased muscle force, aponeurotomy is an appropriate method for lengthening short muscles. Full recovery of maximal muscle force can be expected, but on the ascending limb force may be decreased because of shifts of the length-force curve to higher lengths. This fact needs to be taken into account when muscle weakening is a major indication for surgery.
[193] MOTOR UNIT CONTROL PROPERTIES DURING SUSTAINED CONSTANT-FORCE ISOMETRIC CONTRACTIONS
Carlo J. De Luca, PhD; Patrick J. Foley, MS; Zeynep
Erim, PhD; Markus Khouri, MS
NeuroMuscular Research Center, Boston University,
Boston, MA 02215; email: cjd@bu.edu
Sponsor: Liberty Mutual Insurance Company, Boston, MA 02117
PURPOSE--There are contradictory reports in literature regarding the firing behavior of motor units during sustained isometric contractions. Increasing, decreasing, and haphazard behavior in motor unit firings have all been reported. The purpose of this study was to characterize firing patterns of motor units in isometric isotonic contractions lasting up to 15 s and to investigate the mechanisms responsible for such patterns. This insight regarding the normal control of nonimpaired motor units in isometric isotonic contractions will provide a basis for repairing dysfunctional muscles.
METHODOLOGY--A total of 122 contractions at 30, 50, and 80 percent of maximal voluntary contraction (MVC) of the tibialis anterior (TA) and first dorsal interosseus (FDI) muscles level were analyzed. Mean firing rates were calculated and analyzed as a function of the recruitment threshold of the motor unit and the force level of the contraction.
PRELIMINARY RESULTS--In the region where the force output was kept constant, mean firing rates of motor units showed a continual decrease. Motor units with higher recruitment thresholds decreased their firing rates faster than motor units with lower thresholds. The rate of decrease in firing rates was also found to be proportional to the force level of the contraction, the rate of decrease increasing with increasing force level. The case of 80 percent MVC contractions of the FDI presented an exception to this rule, possibly due to fatigue being more predominant in this case. In investigating the mechanisms allowing for the maintaining of constant force levels while motor unit firing rates decrease, three possibilities were hypothesized: recruitment of motor units to compensate for a decrease in the force contributions of motor units decreasing their firing rates; compensatory activity in the agonist or antagonist muscle groups; and twitch potentiation which would increase the force contribution of a motor unit per firing and hence result in forces comparable to initial values even when the firing rates decreased.
No recruitment was observed after the targeted force level was initially reached in any of the contractions, ruling out the possibility that recruitment was responsible for maintaining of a constant force level. Wrist extensor and flexor muscles were employed to investigate if compensatory action, such as a decrease in antagonist activity or an increase in agonist activity, could explain the sustained force output. However, no such activity was observed. In the absence of recruitment and complementing agonist/antagonist activity, twitch potentiation was accepted as the only plausible mechanism that could require the firing rates to decrease during sustained contractions.
[194] CONTROL OF MUSCLE FIBERS: HOW DOES A MUSCLE REGULATE FORCE?
Zeynep Erim, PhD; Carlo J. De Luca, PhD; Kiyoshi Mineo,
PhD
NeuroMuscular Research Center, Boston University,
Boston, MA 02215; Keio Medical School, Tokyo 160 Japan;
email: erim@bu.edu
Sponsor: Liberty Mutual Insurance Company, Boston, MA 02117
PURPOSE--Insight into the architecture of the motoneuron pool and the inherent characteristics of individual motor units was gained from our previous studies. In the present study we aimed to complement this knowledge with an understanding of the signals that drive the motor units, given the formerly established pool architecture and input/output properties. This work is intended to increase our knowledge of the normal control of motor units so that better procedures may be established to diagnose and rehabilitate persons with neuromuscular deficiencies.
METHODOLOGY--Myoelectric signals were detected from the tibialis anterior muscle of six subjects with a special quadrifilar needle electrode while the subjects generated isometric forces at 20, 50, 80 or 100 percent of maximal voluntary contraction. These signals were later analyzed by our Precision Decomposition Technique.
PRELIMINARY RESULTS--The data revealed that at a given force level, the average and standard deviation of the instantaneous firing rate both decreased as a function of the recruitment threshold of the motor unit. However, the coefficient of variation parameter, defined as the standard deviation normalized by the mean, was not significantly different among the motor units active in a contraction, irrespective of their recruitment thresholds. This observation led to the suggestion that the coefficient of variation of the firing rate may be the parameter which the system controls and maintains constant. It was observed that the coefficient of variation increased as the force level increased, lending further support to the view that the instantaneous firing rate coefficient of variation may be a defining parameter in the control of motor units. The increased coefficient of variation with increasing force levels, combined with our previous observations of high cross correlation among the firing activities of motor units, which further increased at higher force levels, were interpreted as an indication that the variability in common drive increases at higher force levels; and that common drive increases faster relative to the unshared noise component as higher muscle forces are produced.
[195] SYNCHRONOUS BEHAVIOR OF MOTOR UNIT FIRINGS
Hans P. Batra, MS; Zeynep Erim, PhD; Carlo J. De Luca,
PhD
NeuroMuscular Research Center, Boston University,
Boston, MA 02215; email: erim@bu.edu
Sponsor: Liberty Mutual Insurance Company, Boston, MA 02117
PURPOSE--Synchronization of motor unit firings at low levels has been observed in humans, yet it remains unclear whether the synchronous motor unit activity serves a physiological purpose. It has been suggested that synchronization of motor units is used to increase force output or maintain the force level of contraction. Understanding the role of motor unit synchronization will lead to a fuller understanding of the strategies employed by the central nervous system in controlling motor units.
METHODOLOGY--A strict criterion to determine if motor unit pairs exhibit an interdependence that is significantly above random behavior was applied to motor unit pairs recorded from the first dorsal interosseous and from the tibialis anterior muscle. The Synch Index (indicating strength of synchronization) and Motor Unit Synch (indicating the percentage, within a contraction, of motor unit pairs exhibiting synchronization) were computed for synchronized motor unit pairs. Changes in synchronization were characterized as a function of the contraction force level, time of contraction, and motor unit recruitment threshold.
PRELIMINARY RESULTS--Individual subjects exhibited no significant difference in Synch Index at different force levels or between the first dorsal interosseous and tibialis anterior muscles, although inter-subject variability of the Synch Index was large. In addition, the temporal location of synchronous motor unit firings did not exhibit any consistent pattern over the time course of contractions. Furthermore, the recruitment thresholds of the motor units did not influence the strength of synchronization. These findings show that there is no systematic behavior to motor unit synchronization and suggest that the observed synchronization has no purpose or function, but is most likely a byproduct of the physiological control system.
[196] DEVELOPMENT OF TEST PROTOCOLS TO ASSESS THE BEHAVIOR OF BACK MUSCLES
Lars Oddsson, DrMedSci; Johann E. Giphart, MS; Serge H.
Roy, ScD; Carlo J. De Luca, PhD
NeuroMuscular Research Center, Boston University,
Boston, MA 02215; email: loddsson@acs.bu.edu
Sponsor: Liberty Mutual Insurance Company, Boston, MA 02117; Boston University, Boston, MA 02215
PURPOSE--Most test protocols currently used to evaluate the function of back muscles are either based on the concept of measuring muscular strength or on fatiguing the muscles during high level contractions. Subjects in acute pain at the time of testing have problems complying with such procedures, due to their inability to perform a maximal exertion. We are currently in the process of investigating complementing protocols based on the concept of load sharing to assess the function of back muscles. We predict this will allow us to more easily investigate muscular function in acute phases of pain, thus improving specificity of diagnosis and allowing for better and earlier individualized treatment.
METHODOLOGY--The behavior of electromyographic (EMG) spectral and amplitude parameters has been investigated during different types of low level force contractions. Spectral parameters, including median frequency (MF) and amplitude (RMS) of EMG signals were monitored during isometric trunk extension using the Back Analysis System (BAS). Two kinds of tasks, namely step-wise incrementing "staircase" contractions for short durations (20-30 s) at nonfatiguing force levels, and low level long-duration contractions at constant force levels, were studied.
PRELIMINARY RESULTS--During staircase contractions, control subjects showed a balanced increase in the RMS level of the lumbar back muscles which was highly correlated to the developed force. Patients, however, displayed inconsistent increases in RMS levels with increasing force indicating a skewed load sharing pattern in the patient population. During long duration contractions, control subjects displayed different phases of EMG parameter behavior. These included initial increases in RMS as MF decreased; a constant RMS level while MF continued to decrease; and toward the end of the contraction, a decreasing RMS level while MF stabilized at a low level. RMS decrease was most marked in the multifidus muscle. Patients displayed a much more rigid behavior. Their EMG parameters appeared to stay more constant until the subject ended the task due to pain or discomfort. The results suggest that test protocols based on the concept of load sharing may be used to assess information related to the normal function of lumbar back muscles.
[197] ACTIVATION OF NECK MUSCLES DURING A FORCE CONTROL TASK
David A. Gabriel, PhD; Joseph Y. Matsumoto, MD; Dudley
H. Davis, MD; Bradford L. Currier, MD; Kai-Nan An,
PhD
Movement Disorders Laboratory, Department of Neurology
and Biomechanics Laboratory, Department of Orthopedics, Mayo
Clinic and Mayo Foundation, Rochester, Minnesota 55901;
email: dgabriel@mayo.edu
Sponsor: The Mayo Clinic/Mayo Foundation, Rochester, MN 55909
PURPOSE--The primary purpose of this study is to map the vectors of electromyographic (EMG) activity for the neck musculature throughout the polar range of head motion and determine the agonist, synergist, and antagonist function of each muscle under consideration. As a secondary problem, this research will then determine whether the magnitude and direction of the greatest EMG activity corresponds with the muscle's joint moment potential (JMP), defined as the product of its moment arm and physiological cross-sectional area (PCSA) obtained from fresh-frozen cadaveric specimens. Baseline data on normal subjects will be of practical benefit to those interested in establishing what muscles are involved in the various presentations of spasmodic torticollis.
METHODOLOGY--The application of force by the head will be perpendicular to the long axis of the neck through the use of a modified halo attached to a testing chair. Subjects (N=18) produce a force on the halo which will appear on a computer screen as a vector with a magnitude and direction that is related to tension at the load cell. There will be 12 targets, 1 placed on the circumference of a virtual circle every 30°. The tip of the vector will elongate in the desired direction until a force corresponding to 50 percent of maximum for that particular direction has been obtained. The basic plan is to give 5 days of testing. The first two pre-test sessions will be used to practice exerting a maximal isometric contraction in each of the 12 desired directions. The third pre-test session will then be used to practice exerting a force which is 50 percent of maximum for each of the 12 targets. Collection of EMG activity will then occur on the last 2 days of testing. The order of presentation of the direction of force application will be randomized.
Forces will be recorded by a 6 degree-of-freedom load-cell mounted on top of the modified halo. Two intramuscular wires in bipolar configuration will be used to record the EMG activity of the following muscles on the right and left sides of the neck: 1) the sternocleidomastoid; 2) the superior fibers of the trapezius; 3) the splenius capitis; 4) the semispinalis capitis; 5) the scalenus medius; and 6) the levator scapulae. The criterion measures will be force, and the magnitude and direction of greatest activity (EMGi) for each muscle (i) as measured by the root-mean-square EMG amplitude. The inner product will then be used to determine the relationship between the vectors EMGi and JMPi in a correlational sense.
PROGRESS--The testing device and data collection programs have been completed. The electrode configuration has been validated. Two pilot dissections have taken place on fresh-frozen cadaveric specimens, where the physiological cross-sectional areas of the muscles listed above have been obtained.
FUTURE PLANS--Data collection on a normal and a spasmodic torticollis population is planned through spring 1997. The cadaveric work is in progress with collection of moment arm data begun.
[198] MUSCLE FIBER DAMAGE DUE TO ECCENTRIC CONTRACTIONS
Richard L. Lieber; D.K. Mishra, MD; Robert Benz, MD;
Tina Patel; J. Fridén, MD, PhD
Departments of Orthopaedics and Bioengineering,
Biomedical Sciences Graduate Group, University of
California; VA Medical Center, San Diego, CA 92161; email:
rlieber@ucsd.edu
Sponsor: National Institute of Arthritis and Musculoskeletal Disorders, Skin Diseases Muscle Biology Program
PURPOSE--The purpose of these studies was to determine the timing of the loss in cytoskeletal desmin that was previously observed. Since cytoskeletal loss can result in myofibrillar disruption, we felt that understanding its timing would provide insights into the damage mechanism.
METHODOLOGY--Rabbit extensor digitorum longus (EDL) and tibialis anterior (TA) muscles were examined 5 min or 15 min after eccentric exercise and 1 hour or 1 day after 30 min of an eccentric exercise protocol (n=16 rabbits). Muscles were cyclically activated at 40 Hz for 400 ms (approximately 50 percent Po) and allowed to relax for 600 ms over the treatment period. This stimulation frequency (40 Hz) was chosen based on motor unit studies, which demonstrated that at this frequency force decline was due to fatigue of the muscle fibers themselves and not the neuromuscular junction or motor nerve. The frequency was also chosen to produce the force level observed during moderate intensity exercise. We also eccentrically exercised the slow soleus muscle and measured contractile and structural parameters.
RESULTS--The earliest change noted was a significant loss of desmin labeling in 2.5±0.63 percent of the rabbit extensor digitorum longus muscle (EDL) muscle fibers (p<0.005) 5 min after initiation of eccentric exercise. Some loss of tibialis anterior (TA) fiber desmin was also apparent at this time period (0.24±0.19 percent), but the magnitude was not significantly different from zero (p>0.2). Fifteen min after initiation of exercise, desmin loss was more pronounced, increasing to 7.4±1.4 percent and 4.6±1.0 percent in the EDL and TA, respectively (p<0.005). Finally, 1 day after 30 min of eccentric exercise the percentage of fibers without desmin staining rose to 23.4±3.7 percent and 7.7±2.4 percent in the EDL and TA, respectively (p<0.001). Loss of desmin staining occurred in the absence of contractile or metabolic protein disruption. Increased staining intensity of the intrasarcomeric cytoskeletal protein titin and an inability to exclude plasma fibronectin was also observed in most but not all fibers which had lost desmin staining. Desmin disruption thus represents a very early structural manifestation of muscle injury during eccentric contraction. Soleus muscles showed no such abnormalities. Soleus muscle stiffness decreased by only 30 percent over the 30 min treatment period while isometric force decreased by 85 percent. These data indicate that, while soleus muscles decreased their force generating capability significantly, there were a number of cross-bridges still attached that were not generating force.
[199] LIGAMENTO-MUSCULAR PROTECTIVE REFLEX IN THE KNEE, SHOULDER, ANKLE, AND ELBOW
Moshe Solomonow, PhD; Robert D. D'Ambrosia, MD; Carlos
A. Guanche, MD
Bioengineering Laboratory, Department of Orthopaedic
Surgery, Louisiana State University Medical Center, New
Orleans, LA 70112
Sponsor: National Science Foundation, Arlington, VA 22330
PURPOSE--The ligaments are considered to be the primary restraints of a joint, keeping the bones aligned in their natural kinematic state throughout the joint's range of motion. A growing pool of evidence shows that the musculature significantly contributes toward joint stability as well as protection of the ligaments. Our early work with the anterior cruciate ligament (ACL) shows the variety of mechanoreceptors present in this ligament, and also that a reflex arc exists from these mechanoreceptors to the muscles crossing the knee. Strain applied directly to the ACL resulted in reflex contraction of the hamstrings in animal models and in humans. We investigate whether such a protective reflex exists in other joints.
METHODOLOGY--We undertook several new studies to determine if such a ligamento-muscular reflex arc exists in other joints while stimulating articular nerve branches emerging from the ligaments. To date we have found that the protective reflex exists in the shoulder, elbow, and ankle, in addition to the knee.
RESULTS--Therefore, a synergisitic relationship probably exists between the ligaments and muscles of every joint to ensure preservation of the tissue, prevention of damage, and proper kinematic alignment of the joint's bones when various internal and external disturbing loads are applied. During surgery, the neural integrity of the ligament-joint should be preserved as much as possible in order to avoid joint arthropathy, and then utilized to design the appropriate postsurgical therapy.
[200] SURFACE AND WIRE EMG CROSSTALK IN NEIGHBORING MUSCLES
Moshe Solomonow, PhD; Richard V. Baratta, PhD; Bing-He
Zhou, EE; Robert D. D'Ambrosia, MD
Bioengineering Laboratory, Department of Orthopaedic
Surgery, Louisiana State University Medical Center, New
Orleans, LA 70112
Sponsor: National Science Foundation, Arlington, VA 22330
PURPOSE--EMG crosstalk between neighboring muscles presents a long-standing controversy without final conclusion as to its existence and, if so, to the mechanism underlying the process. We set an experimental protocol to clearly delineate whether cross talk exists and under what circumstances.
METHODOLOGY--We undertook an investigation in animal model in which the muscle nerves of the soleus, medial, and lateral gastrocnemius and tibialis anterior muscles were supramaximally stimulated while surface and wire EMG from all the muscles were recorded. Later, the muscle nerves were all cut, except for the nerve to the medial gastrocnemius, which was supramaximally stimulated while surface and wire EMG from all the muscles were recorded. Any EMG in the muscles which their nerves were cut was considered pure and true crosstalk.
RESULTS--It was shown in a cat model that the extent of the crosstalk in the soleus, tibialis anterior, and the lateral gastrocnemius was less than 4-5 percent during maximal stimulation of the medial gastrocnemius when recording with surface electrodes. When recording with wire electrodes, the crosstalk was limited to 1-2 percent of maximal EMG activity in the corresponding muscles.
In several animals with a confirmed fat layer over the muscles, the crosstalk values in the inactive muscles ranged between 16-32 percent, suggesting that adipose tissue is responsible for inducing significant crosstalk in surface recordings of the EMG. There was no impact on the wire recordings.
It was concluded that, when using electrodes appropriately sized to the muscle dimensions and correctly placed, the amount of crosstalk in surface and wire recordings is negligible. However, when recording surface EMG from muscles covered with adipose tissue, the EMG will be contaminated with significant crosstalk and may lead to false conclusions.
[201] THREE-DIMENSIONAL DESCRIPTION OF MUSCLE PROPERTIES
Moshe Solomonow, PhD; Richard V. Baratta, PhD
Bioengineering Laboratory, Department of Orthopaedic
Surgery, Louisiana State University Medical Center, New
Orleans, LA 70112
Sponsor: National Science Foundation, Arlington, VA 22330
PURPOSE--Traditional descriptive models of muscle properties are limited to the length-tension relationship obtained from isometric contractions, and from force-velocity curves obtained from isotonic contraction. Unfortunately, combining both curves to yield the force-length-velocity relationships is invalid, as data obtained in isometric contractions could differ by as much as 50 percent from those of shortening contraction. This problem was solved by describing length-force and velocity in a load-moving contraction.
METHODOLOGY--We have set up a protocol by which a muscle was stimulated electrically through its nerve, allowing shortening while applying displacing loads of different size. The initial length of the muscle when loaded and the final length after supramaximal stimulation provided a set of points with which the passive and total length-load (force) ratios were plotted as a function of load values. The shortening curve of the muscle during stimulation was recorded and differentiated mathematically to yield its velocity at any given length and for any given load, thereby allowing the construction of a load-length-velocity curve.
RESULTS--The initial results provide the first valid description of load-length-velocity of an isolated muscle, which could be used in large scale modeling of human movement, neuroprosthesis design, and general understanding of muscle functions.
[202] SIMULATION OF EMG SIGNALS ELECTRICALLY EVOKED IN THE HUMAN BICEPS MUSCLE
Roberto Merletti, PhD; Edward Kupa, MS Serge H. Roy,
ScD; Elena Avignone, Italian Laurea; Piero Guglielminotti,
Italian Laurea
NeuroMuscular Research Center, Boston University,
Boston, MA 02215; Politecnico di Torino, Torino, Italy;
email: sroy@bu.edu
Sponsor: North Atlantic Treaty Organization, Brussels, Belgium; the Chamber of Commerce, Torino, Italy
PURPOSE--Many clinically important anatomical and physiological parameters of muscle cannot be measured directly. However, these unaccessible parameters can be estimated through the use of model equations. A mathematical model was developed that represents surface electromyographic (EMG) signals as a summation of contributions from single fibers of individual motor units (MU). The model was used to simulate single and double differential to study the influence of signal shape on the relative changes of median frequency and conduction velocity (CV) during fatigue.
METHODOLOGY--A 16-channel, active electrode array, EMG amplifier with stimulation artifact suppression, and neuromuscular stimulator were used to detect maximal M-waves from the human biceps brachii muscle. The muscle's main motor point was stimulated monopolarly with the EMG array placed distally to the motor point. For model simulations of the M-wave, estimates of the size and location of the neuromuscular junction, termination zones, muscle fibers, CV, and other motor unit parameters were specified. Surface monopolar and bipolar signals were computed for 12 of the channels, which spanned the entire distal aspect of the muscle.
PRELIMINARY RESULTS--Three "equivalent motor units" with different geometry, innervation zones, and CV were simulated to obtain a good match with signals from the beginning of the contraction. The same set of motor units allowed excellent simulation of the signals from the end of the contraction, with only CV being changed in order to match the morphology of all detected signals. This finding shows that, during sustained contractions, individual motor units change their CV by different amounts and not necessarily in the same direction. This phenomenon explains changes of signal shape that cause the relative change of EMG median frequency to be much greater than that of CV.
[203] SIMULATION OF EVOKED EMG SIGNALS FROM IN VITRO PREPARATIONS
Edward Kupa, MS; Roberto Merletti, PhD; Serge H. Roy,
ScD; Elena Avignone, Italian Laurea; Piero Guglielminotti,
Italian Laurea; Loredana Lo Conte, Italian Laurea
NeuroMuscular Research Center, Boston University,
Boston, MA 02215; Politecnico di Torino, Torino, Italy;
email: sroy@bu.edu
Sponsor: North Atlantic Treaty Organization, Brussels, Belgium; the Chamber of Commerce, Torino, Italy
PURPOSE--The electromyographic (EMG) signal detected from the skin overlying a muscle is the result of numerous electrophysiological processes that cannot be experimentally measured simultaneously. Mathematical models can simulate the surface detected EMG signal on the basis of numerous electrophysiological factors, but solutions are only theoretical estimates needing empirical verification. The utilization of in vitro techniques, in combination with mathematical model simulations, may therefore provide a greater understanding of the relationship between the surface EMG signal and muscle physiological characteristics than either technique can provide alone.
METHODOLOGY--Signals were recorded from isolated neuromuscular preparations of the soleus and diaphragm muscles of the rat. Muscles were placed in an isothermal, oxygenated bath in contact with an EMG electrode array spanning the entire muscle length. Electrical stimulation of the nerve elicited M-waves which were detected in both monopolar and bipolar configurations. Measurements were made on the muscles to obtain geometrical information for input into the model. The model was used to simulate signals obtained from the in vitro preparation.
PRELIMINARY RESULTS--The model provided a close approximation of the actual signals recorded in vitro when multiple motor units with slightly different characteristics were simulated. There was a very narrow set of model input parameters that fit the experimental sequence of M-waves. Preliminary results to date suggest that further iterative studies combining the in vitro technique with the muscle model may lead to noninvasive estimation techniques for assessing the geometrical and functional properties of superficial muscles.
[204] LOW-LEVEL MUSCLE ACTIVITY AS A RISK FACTOR IN THE DEVELOPMENT OF CUMULATIVE TRAUMA DISORDERS
Rolf H. Westgaard, PhD; Carlo J. De Luca, PhD; Markus
Khouri, BS
NeuroMuscular Research Center, Boston University,
Boston, MA 02215; email: westgaard@oral.unit.no
Sponsor: Norwegian Research Council; Liberty Mutual Insurance Company, Boston, MA 02217
PURPOSE--Pain syndromes in the shoulder and neck region may develop in occupational situations with a low level of muscle activation. Previous studies, combining epidemiological methods with recording of the electromyographic (EMG) signal from the trapezius muscle by surface electrodes, have suggested that prolonged activity in low-threshold motor units can be causally related to the development of shoulder pain syndromes. The present series of experiments aim to obtain direct evidence for this hypothesis by utilizing the motor unit decomposition technique.
METHODOLOGY--Experimental situations known to be stressful to many individuals, or mimicking occupational work situations where some workers develop shoulder and neck pain, are created in the laboratory. A combination of surface EMG and single motor unit recordings is carried out to relate overall activation of the trapezius muscle with the activity pattern of single motor units, activated at low activity levels. The analysis of the EMG recordings will look for the following evidence, considered to support the above hypothesis: 1) signs of fatigue in motor units, active for extended periods of time; 2) higher firing rates or more regular activity patterns of motor units in experimental situations known to be stressful for the individual or mimicking work situations with a high risk of musculoskeletal pain syndromes developing; and 3) similar changes to 2), but observing normally pain-free subjects in situations with discomfort developing, or observing subjects with pain syndromes in the shoulder and neck.
PRELIMINARY RESULTS--The Precision Decomposition Technique, normally used in experiments of 30-s duration, has been modified to be used in experiments that last 1 hour or longer. Also, considerable time and effort has been spent to determine an intramuscular electrode configuration that is sufficiently sensitive to detect motor units at activation levels of 1-2 percent of maximal muscle activation, and at the same time sufficiently discriminating to reliably identify a motor unit when many motor units are active. Currently, experimentation is continuing to look for evidence as detailed above.
[205] MYOELECTRIC DATA COMPRESSION USING ADPCM
D.F. Lovely, BSc, PhD, PEng
Institute of Biomedical Engineering, University of New
Brunswick, Fredericton, NB E3B 5A3 Canada email:
lovely@unb.ca
Sponsor: UNB Research Fund
PURPOSE--The purpose of this project is to determine the nature of the errors associated with employing adaptive differential pulse code modulation (ADPCM) to the compression of myoelectric data.
METHODOLOGY--The myoelectric signal, obtained either by surface or needle electrodes, is used in many areas of clinical research and diagnostics. The conventional method of storing such information is in digitized form on a computer. However, the bandwidth of the signal and the required resolution results in large memory requirements. In this work, ADPCM has been investigated as a method of reducing the memory requirements for myoelectric data storage.
Differential PCM is a technique whereby the difference between successive samples is coded, thus reducing the required dynamic range of the coder. The disadvantage of such a scheme is that slope overload can result if the input signal amplitude changes suddenly. ADPCM overcomes this drawback by using a predictive algorithm to modify the quantization width of the coder. Thus the resolution of the coding scheme varies due to the statistics of the data.
PROGRESS--Recent research concentrated on the nature of the errors, in addition to the suitability of the technique for the compression of the myoelectric signal from dynamic contractions.
Six subjects were used in the study. Myoelectric data was collected from the biceps while subjects were asked to maintain 10, 20, 30, 40, and 50 percent of maximum voluntary contraction (MVC) levels with the forearm restrained. The order of the contraction levels was randomized and adequate rest periods were given to ensure no fatigue effects were taking place. Both direct data and compressed/expanded data were collected simultaneously and compared on the basis of percent residual difference (PRD). A similar experiment was also conducted for a dynamic contraction where the forearm was allowed to move through a fixed arc. Again simultaneous collection of raw data and compressed/expanded data was carried out.
RESULTS--Initial results show very little variation in errors as reported earlier for sustained isometric contractions. At present, the data from the dynamic study is being analyzed.
[206] SKELETAL MUSCLE LENGTH-FORCE CHARACTERISTICS DURING MAXIMAL AND SUBMAXIMAL ACTIVATION
Peter A. Huijing, PhD; Guus C. Baan; Drs. Boris Roszek;
Drs. Peter Bosch
Institute of Fundamental and Clinical Movement Sciences,
Faculty of Human Movements Sciences, Vrije Universiteit,
1081 BT Amsterdam, The Netherlands; Institute of Biomedical
Technology, University of Twente, 7500AE Enschede, The
Netherlands; email: P_A_J_B_M_Huijing@FBW.VU.NL
Sponsor: Vrije Universiteit, 1081 BT Amsterdam, The Netherlands; Institute of Biomedical Technology, University of Twente, 7500AE Enschede, The Netherlands
PURPOSE--The purpose of this project was to make a link between results of experiments studying muscle properties under maximal activation and in vivo properties of muscle which are rarely characterized by maximal activation. For that purpose several conditions of submaximal activation were imposed on in situ muscles for a systematic analysis.
METHODOLOGY--In addition to variables of muscle geometry (i.e., fiber length, aponeurosis length, and fiber and aponeurosis angles), the number of sarcomeres in series within fibers and filament length parameters were determined. Submaximal activity was induced in two ways: 1) in fully recruited muscle, stimulation frequencies between 100 and 15 Hz were imposed for a constant frequency or decreasing frequencies protocol, and 2) in partially recruited muscle, stimulation current was lowered to a bipolar electrode on the nerve so that only a part of the motor units remained active.
RESULTS--Length-force characteristics of submaximally active rat medial gastrocnemius muscle differ very substantially from that of the maximally active muscle: with all motor units active, force decreased at lower stimulation frequencies and muscle optimum length shifted to higher muscle lengths. If the low stimulation frequencies followed higher ones (history effects), the decrease in force was smaller (i.e., potentiation occurred). The shift of optimum length was also smaller in this case.
In a sustained isometric contraction length force characteristics are time dependent due to summation of highly nonlinear effects of potentiation and fatigue. The net results are highly dependent on stimulation frequency. With fewer motor units active, the optimum length (i.e., the length at which maximal active force was generated) was found to occur at either higher or lower muscle lengths at submaximal currents.
IMPLICATIONS--Length-force and force velocity properties of submaximally active muscle are likely to be very much dependent on the degree of activation. This allows the central nervous system an extra dimension of control in the execution of movements. In effect, it creates the need for a control which may referred to as intramuscular coordination. For functional electrostimulation (FES) these findings will have important implications as well for modelling of muscle and movement.
[207] A COMPARISON OF TWO KNEE-SCORING QUESTIONNAIRES ADMINISTERED TO A NORMAL ATHLETIC POPULATION
Kevin A. Thomas, PhD; Carlos A. Guanche, MD; Anna
Demirdjian, BS; Scott G. Petrie, MD
Bioengineering Laboratory, Department of Orthopaedic
Surgery, Louisiana State University Medical Center, New
Orleans, LA 70112; email: kthoma@nomvs.lsumc.edu
Sponsor: None listed
PURPOSE--The Noyes and Lysholm knee scoring questionnaires are commonly used for knee ligament surgery follow-up to assess surgical outcomes. These questionnaires were developed based on knees with pre-existing pathology and have not been standardized to normal knees. In this study, both scores were administered to normal subjects in order to determine whether the resultant score would correlate with the arbitrated normal score of 100.
METHODOLOGY--A total of 246 athletes (150 male, 96 female) were screened. Any subject reporting a history of injury or surgery to either knee was excluded from the study. Any pre-existing knee pathology, either self-reported or clinically diagnosed, also excluded the knee from the study.
PROGRESS--A total of 492 knees were evaluated, 418 knees (253 from male subjects, 165 from female subjects) qualified for statistical analysis.
RESULTS--The female athletes produced significantly lower scores than the male athletes on both questionnaires (Noyes: female mean=97.82, male mean=99.10; Lysholm: female mean 97.16, male mean 99.10). Analysis of the subsections of each score revealed that women produced significantly lower scores than the males in the pain category of both questionnaires in the stair-climbing and running categories of the Noyes questionnaire, and in the squatting and instability categories of the Lysholm questionnaire. The males scored lower than the females in the walking category of the Noyes questionnaire, but these differences were not significant. When comparing the mean values in each category to that of the maximum value for that category, the males scored significantly lower than the maximum value in all of the Noyes categories except walking, stair climbing, and running; and the females scored significantly lower than the maximum value in all categories except walking. Similarly, for the Lysholm questionnaire, the males scored significantly lower than the maximum in all categories except support and stair-climbing; and the females scored significantly lower than the maximum in all categories except limb and thigh atrophy. An analysis also was performed comparing the total scores and the scores in each category for right versus left knees. For the Noyes questionnaire, there was a significant difference for the male subjects in the pain category, and for the female subjects in the jumping/twisting category. There were no significant differences in total score or any categories of the Lysholm questionnaire when comparing right versus left knees for either male or female subjects. This study exemplifies the need for more accurate instruments in the evaluation of knee surgical outcomes. Modification of the subsections of the existing questionnaires that seem to induce error is one option. The development of a tool that will increase the validity of reported scores without modifying existing scores is thus proposed.
Go to TOP.
Last revised Wed 05/26/1999
