- Email: [email protected]
In-vivo function of the thumb muscles
Clinical Biomechanics
14 (1999) 141-151
In-vivo function of the thumb muscles Kenton IX. Kaufmana*, Kai-Nan An”, William J. Litchy’, William P. Cooney III”, Edmund Y. S. Chao” “Biotnechanlz Laboratory, Department of Orthopedics, Mayo Clinic/Foundation, ‘Section of E!ectromyography, Department of Neurology Mayo CliniclFoundation,
200 First Street, SW Rochester; MN 55905, USA 200 Fir.st Street, SW Rochester; MN 55905, USA
Received 10 June 1998; accepted 17 June 1998
Abstract Objective. The purpose of this study is to quantify the electrical activity of the thumb muscles responsible for the production of force in different directions of thumb movement. Design. The isometric forces and electromyographic activity generated by seven thumb muscles were measured on five normal healthy test subjects. Background. The thumb is very important for proper hand function. Presently available electromyographic studies of the thumb muscles provide only limited information. Most thumb muscles have more than one function. Additional studies are required to carefully examine and confirm the in-vivo relationship between the thumb muscle electromyogram and mechanical output. Methods. The direction and magnitude of the force vector generated at the interphalangeal joint and the relative electrical activity were obtained for eight directions of thumb action. The regions of function were defined for the abductor polhcis brevis, opponens pollicis, flexor pollicis brevis, adductor pollicis, flexor pollicis longus, extensor pollicis longus, and the abductor pollicis longus. Data was collected during voluntary isometric contraction, both before and after blocking the median nerve at the wrist. Results. The highest force production was obtained during flexion. The region of maximal muscle electrical activity varied for each muscle studied. The areas of maximal in-vivo muscle activity agreed with the moment arm data reported in the literature. The median nerve block eliminated the ability to produce force in abduction. Conclusions. This study has demonstrated that by combining electromyographic measurement and biomechanical analysis it is possible to confirm the relationship between in-vivo thumb muscle function and muscle mechanics in a novel manner. The findings of this study indicate the importance of the local anatomy in controlling the direction of force production.
Relevance When different tendon transfers are under consideration, the surgeon must take into account the importance of not only the agonistic action of a muscle but also the stabilizing and antagonistic action of that muscle. The relative electrical activity and strength contributions of specific thumb muscles measured in this study should be considered for optimal function. 0 1999 Elsevier Science Ltd. All rights reserved. Keywords: Thumb; Muscle force; Electromyographic
activity; Median nerve
1. Introduction
The thumb is very important for proper hand function. A thumb is needed for simple everyday tasks which require grasping or pinching. Experts in accident insurance value the loss of a thumb at 20-28% of total disability [l]. Loss of thumb function may also be caused by disease, e.g. rheumatoid arthritis, osteoart,hritis, carpal tunnel syndrome, or nerve injury. The loss of thumb function results in an inability to position *Corresponding
author. E-mail: [email protected]
the thumb correctly for precision type operations such as handling small objects, and results in a lack of power in opposition to the fingers [2,3]. Presently available electromyographic studies, which concern the thumb, provide only limited information. Weathersby et al. published an electromyographic analysis, but no motion tracking was employed [4]. Forrest and Basmajian studied the thenar intrinsic muscles in 25 subjects, but primarily isotonic actions of opposition, flexion and extension were utilized [S]. Their static testing was limited to clasping a wooden dowel and holding a cup. No external forces were
0268-0033/99/$ - see front matter 0 1999 Elsevier Science Ltd. All rights reserved. PII: SO268-0033(98)00058-S
142
K. R. Kaufman et aLlClinical Biomechanics 14 (1999) 141-150
measured and electromyographic (EMG) signals were not quantified in assessing the results. Ebskov and Long recorded muscle function in the unloaded thumb by using electrogoniometry and multichannel electromyography [6]. Hall and Long analyzed static pinch and power grip in a well-planned study but only tested the intrinsic muscles [7]. Various pinch and grasp actions were tried but the magnitudes of these activities were not accurately measured. No attempt was made to quantity the electromyographic activity. Close and Kidd studied synchronous recording of the thumb, index and long finger motion and the muscle action potentials [8]. They included flexion, extension, abduction, adduction and static grasp. Long et al. reported on intrinsic muscle action in power grip and precision handling [9]. Various devices have been described to measure the force exerted by the fingers or thumb. Some require movements of the fingers or the thumb [lo-141. However, the assurance of isometric conditions for the skeletal muscle being tested is an important physiologic requirement for description of muscle thumb function [l&16]. Three devices have been described to measure thumb isometric contraction forces [17-191. Yet none of these devices measure the three-dimensional force vector produced by the thumb muscles during contraction. The methods of comparing muscle activity to force are limited. Most of the muscles have more than one function. Thus, the variable function of a multidirectional muscle needs to be described in terms of its agonistic, synergistic and antagonistic contributions. Previous work has defined the kinematics of the thumb joint [20] and the types of forces in normal and pathologic hands [21,22]. Additional studies are required to carefully examine and confirm the in-vivo relationship between the thumb muscle electromyogram and the mechanical response of the thumb. The specific aims of this study were to evaluate the electrical and mechanical responses of the thumb muscles: (1) during voluntary isometric contraction; and (2) after blocking of the median nerve at the wrist.
2.2. Recording of the electromyographic activity
Intramuscular bipolar wire electrodes were used for recording muscle activity. The electrodes were inserted into the abductor pollicis brevis (APB), opponens pollicis (OPP), flexor pollicis brevis (FPB), adductor pollicis (ADP), flexor pollicis longus (FPL), extensor pollicis longus (EPL), and the abductor pollicis longus (APL). The wires were made of nickel-chrome, polyurethane insulated, 5 pm diameter wire (California Fine Wire Company, 338 South Fourth Street, Grover City, CA, USA). The fine wire electrodes were made using the technique described by Basmajian and DeLuca [23]. Both ends of the wire were stripped of insulation with an electrically heated filament. The wire was threaded through a 25 gauge, 5 cm hypodermic needle. The uninsulated section of one end of the fine wire was cut to a length of 2 mm. Under microscopic observation, a bend was made which reflected the wire to an angle of approximately 150”. This bend was placed 4 mm from the end. These electrodes were packaged and sterilized in a gas sterilizer. A TECA TE-4 (TECA Corp., 3 Campus Drive, Pleasantville, NY, USA) electromyographic system was used to amplify the muscle action potentials. The amplified signal was bandpass filtered with cut-off frequencies of 32 Hz and 3.2 KHz. Whole-wave rectification and integration was performed every 10 ms using BAK Pulsed Sample Integrators (BAK Electronics Inc., P.O. Box 87, Clarksburg, MD, USA). This device produced a positive square-wave output signal of known pulse width. The magnitude of the output signal was the integration of the input signal during the previous pulse width [24]. The integrated EMG signal was acquired with an IBM PC computer (IBM Corp., P.O. Box 1328-C Boca Raton, FL 33422) using ASYSTANT software. The integrated signal was acquired at 100 Hz/channel. All electromyographic measurements were normalized by converting them to a percent of the maximum voluntary contraction recorded during the complete testing session for each subject. 2.3. Recording of the mechanical response
2. Methods 2.1. Test subjects
A group of five normal subjects composed of three males and two females were tested. All subjects had no history of previous hand or neurologic disease. The subjects were in the age range from 20 to 40years. All subjects had occupations which involved primarily indoor office-type work. None of the subjects had leisure activities requiring excessivehand strength.
The mechanical force was measured with a customdesigned three-component load sensor (NK Biotechnical Engineering Co., P.O. Box 26335, Minneapolis, MN, USA). The general design of the transducer was based on aluminum plates separated by shear webs. Strain gages were mounted on these webs to measure deflection (strain) when subjected to a force. The mechanical concept of the sensor was based on very high mechanical impedance with the least possible bending and torsional deflections. The sensor was
143
K. R. Kaufman et aLlClinical Biomechanics 14 (1999) 141-150
carefully calibrated by the manufacturer using a loading device traceable to the National Institute for Standards and Technology (NIST). The sensor had a resolution of kO.065 N with a nonlinearity of ~0.22% full scale (F.S.) and a hysteresis of ~0.33% F.S. The transducer was constructed to minimize cross-talk of the channels. This was checked for each force direction. Although cross-talk was minimal, a 3 x 3 correction matrix was used to completely eliminate the effect of cross-talk between the output channels. All force recordings were normalized by converting them to a percent of the maximum signal recorded during the complete testing session for each subject. 2.4. Testingprotocol Upon arrival, several preliminary procedures were performed. The subject was informed of the risks as well as the rationale of the study. The subject read and signed an informed consent form. A hand history was taken to ensure no underlying pathology. Next, the wire electrodes were inserted and placement was confirmed by isolated muscle function testing. The needle was withdrawn once the correct electrode position was con:firmed and the wire was anchored in the muscle. The hand was then placed in a splint which was used for stabilization of the hand. This splint ensured a repeatable position of the hand during the testing procedure. Next, a roentgenographic examination consisting of a biplanar x-ray of the thumb was performed to further ensure no underlying pathology and to obtain the orientation of the metacarpal segments. Upon satisfactory completion of this examination, the subject was readied for data collection. The subject was seated in a chair. The thumb was positioned to simulate key pinch. In this position, the hand and wrist were immobilized in the splint with the dorsal surface of the hand upward (Fig. 1). The interphalangeal (IP) joint of the thumb was positioned in the neutral position and inserted into an adjustable metal ring which was placed around the thumb at the level of the interphalangeal joint. Prior to placing the thumb in the restraining ring, it was splinted with a metal splint to prevent any motion of the IP joint. The thumb was wrapped with foam tape so that the ring might fit tightly but not with discomfort or excessivepressure on the digital nerve or blood vessels. The subject’s hand was placed in the test device. A stable hand and wrist were required for accurate force and muscle recordings. Fixation of the hand and forearm was achieved by securing the wrist in a stand with a heavy base. The stand could be varied in height and the transducer could be moved in two horizontal directions thereby allowing for three degrees-of-freedom in positional
rbm
Fig. 1. Schematic diagram of experimental test setup. The electromyographic activity of seven thumb muscles was studied using finewire eiectrodes. Simultaneously, the isometric force production was measured with a three-component load cell.
placement of the transducer with respect to the hand. In this manner, measurement of isometric muscle functions could be achieved in a comfortable position of the hand and forearm. Force recordings and electromyographic activity were collected for maximal voluntary contraction in flexion, extension, abduction, adduction, and combinations of these movements. Accordingly, data were obtained in eight positions which were 45” apart with the vertical upward position corresponding to thumb adduction. All positions were tested twice, resulting in a total of 16 recordings being obtained. Each data collection trial was started with the subject relaxed. A command was given at which time the subject attempted a maximal contraction in the predefined direction. The subject was given visual feedback of the direction of attempted motion by displaying the two force components which corresponded to flexionextension and abduction-adduction on an oscilloscope. This permitted the subject to maintain directional control of the digit accurately. The maximal contraction was held for 2 s, at which time the data collection sequence was discontinued and the subject relaxed. A 2-min rest period was observed between data collection trials. The motion testing position proceeded in a clockwise direction. Following completion of the test sequence, the subject had the median nerve blocked at the wrist using 1% lidocaine. The nerve block was confirmed with a nerve conduction study. Evidence to support that a satisfactory block had been achieved was obtained when electrical stimulation of two times threshold failed to yield any myoelectric signal distal to the block. The threshold value was obtained immediately prior to the injection of the lidocane. Once the
144
K. R. Kaufman et aLlClinical Biamechanics 14 (1999) 141-150
block had been confirmed the subject was again asked to repeat the testing sequence described above. Force and electromyographic recordings were again collected. The force and electromyographic data were averaged for each postion. The force data were normalized to the maximum force measured throughout all trials. Similarly, electromyographic data for each muscle were normalized to the maximum electromyographic data recorded for that muscle throughout all trials.
FORCE ADDUCTION
3. Results
The force potential and electromyographic activity are presented in polar plots which display the data with respect to the carpometacarpal (CMC) joint. To retain their descriptive nature, the two orthogonal force vectors generated by flexion-extension and abductionadduction are displayed in the form of a single resultant force vector. This force vector is orientated perpendicular to the long axis of the metacarpal bone. The third orthogonal component, the axial force, would tend to compress the joint and is not displayed. All of the polar graphs have the same orientation with the positive vertical position indicating thumb adduction. Thus, these graphs can be interpreted as if the thumb was being viewed along the long axis of the metacarpal bone toward the distal phalanx. The concentric circles indicate the relative effort in terms of force or electrical activity from 0 to 100%. 3.1. Normal values 3.1.1. Force production
All subjects displayed similar patterns of normal force production (Fig. 2). The highest force production was obtained during flexion. The lowest level of force was measured during extension reaching only 25-40% of the flexion value. Generally, in all other directions the force was in the range of 25-75% of maximum. 3.1.2. Electromyographic activity
The direction of the force vectors applied by the thumb during activation of the muscles gives a direction on which the magnitude of the percentage maximal voluntary contraction is plotted (Fig. 3). Thus, the outermost circle indicates 100% voluntary activation of the muscle. The FPB was most active in the range from flexion to abduction with about 50% activity or less in extension and adduction. The OPP displayed activity in all directions of motion that included abduction. The APB displayed maximal activity in the abduction and flexion-abduction directions with about 50-75% maximal activity in the extension and abduction-extension. The ADP had a much
‘..
. ..’
“...._, ‘-.._
.. . . . . .._.._.........._...
..,..‘..
n=5
ABDUCTION Fig. 2. Maximal voluntary isometric force production at the CMC joint. The data for each test subject is presented in the graph. The data for each subject was normalized to their individual maximum.
more limited region of dominant activity. The ADP was active primarily only during flexion. The APL was most active in the region which encompassed abduction and/or extension. The EPL showed the most activity during extension and in combination with abduction or adduction functions. The FPL was the most active during flexion and/or adduction functions. The response of the muscles tested was classified as either agonistic, synergistic, or antagonistic (Table 1). Agonistic muscle function consisted of electrical activity greater than 75% maximum. Synergistic activity was classified as electrical activity greater than 25% but less than 75% maximum activity. Antagonistic activity was defined as less than 25% maximum activity. 3.2. Effects of median nerve block 3.2.1. Force Production
The ability to produce any appreciable force production in the abduction direction was essentially nonexistent after the median nerve block was applied (Fig. 4). The one subject who was able to produce a force level of approximately 20% did so by contraction of the FPB (deep head), and FPL in a synergistic manner. In contrast, the ability to produce force in either flexion, adduction or extension was not seriously affected.
K. R. Kaufman et aLlClinical
3.2.2. Electromyographic
Biomechanics 14 (1999) 141-150
was no electrical activity in the OPP and the APB. Some of the other muscles do not display activity levels of 100% because the muscles were normalized to their maximal activity throughout the total testing session
activity
The myoelectric activity of the median innervated muscles (APB, OPP, superficial head of FPB) was eliminated after the median nerve block (Fig. 5). There
(8)
145
FPB
OPP
APB
ADDUCTION
ADDUCTION
ADDUCTION
1.00
IOU
1.00
0.75
0.75
0.75
0.50
0.50
0 50
n=5
n=5
(h)
AEIDUCTION
n=5 (C)
ABDUCTION
ABDUCTION
APL
EPL
ADDUCTION
ADDUCTION
ADDUCTION
1.oo
7.oo
1.oo
0.75
a75
n=4
(4
(e)
ABDUCTION
ABDUCTION
n=5 (0
ABDUCTION
FPL ADDUCTION
n=5
(d
ABDUCTION
Fig. 3. Electromyographic activity of the thumb muscles during maximal voluntary isometric contraction. The data for each test subject is presented in the graph. The data for each subject was normalized to their individual maximum. A description of the muscle abbreviations is given in Table 1.
K. R. Kaufman et al,/Clinical Biomechanics 14 (1999) 141-150
f46
Table 1 Classification of isometric thumb muscle activity (percent of maximum) Motion
FPB
OPP
APB
ADD
Adduction Adductionlflexion Flexion Flexion/abduction Abduction Abduction/extension Extension Extension/adduction
NW2
NW) W’l)
W3) W4) NU3) W3) GW-4 S(74) S(43) NW
NW)
W07)
SW)
NO91 GW WW
GW) St581 St311
W32)
NW) NW
S(47) W'4)
S(55)
G(W
NW W4) W'l) NW) NW)
FPL
EPL
APL
S(51) W) ~(42) St511 V9) G@5) G@7)
St311
GW)
NW ~(23) S(53) G(77) G(91)
w38) w39
G = agonist (i.e. > 75% maximum activity); S = synergist (i.e. 25-75% maximum activity); N = antagonist (i.e. ~25% maximum activity). FPB, flexor pollicis brevis; OPP, opponens pollicis; APB, abductor pollicis brevis; ADP, adductor pollicis; FPL, flexor pollicis longus; EPL, _extensor pollicis longus; APL, abductor pollicis longus.
and had produced higher activity levels prior to the nerve block. The FPB (deep head) displayed activity at levels up to 75% of maximum, but over a substantially reduced range than before the block. The level of activity of the ADP was not reduced after the block, but its direction of maximal activity was shifted towards more adduction and less flexion. The FPL continued to display maximal activity in flexion and adduction. The APL was most active during extension and extensionadduction. The EPL was recruited minimally by four subjects, while one subject utilized the EPL at levels above 50% in all directions.
4.1.1. Force production
ADDUCTION
‘..,
_..’ ,_.’
‘.... .-......
// ‘....
_....._.__._...
The in-vivo study of thumb function is difficult because of the complex anatomy of the joints and muscles. A custom-designed force transducer was utilized to measure the three-dimensional force vector produced by the thumb musculature during isometric contraction. This instrument provided not only the instantaneous magnitude of strength, but also the resultant direction of the force vector produced during evaluation. The associated myoelectric activity of four intrinsic and three extrinsic muscles which produce these forces was also measured in vivo in this study. 4.1. Normal values
FORCE
“...
4. Discussion
..‘...
n=4
ABDUCTION Fig. 4. Maximal voluntary isometric force production at the CMC joint after blocking the median nerve at the wrist. The data for each test subject was normalized to their individual maximum before the nerve block and is presented in the graph.
The force production may be explained based on the muscle mechanics, physiological characteristics and activation levels. All muscles exert a force proportional to the physiological cross-sectional area [25,26]. The physiological cross-sectional area (PCSA) of the thumb musculature has been measured [27-291. The strength of a muscle is also related to the moment arm of the muscle. The movements of the thumb at the CMC are both flexion-extension and abduction-adduction [20]. The thumb motion can be thought of as following around the surface of a cone, with its apex proximal to the CMC joint and its base marked out by the moving thumbnail. The pulp of the thumb faces the axis of this cone. Most muscles attached to the thumb have a vector component of their force available for both flexion-extension and abduction-adduction [30]. The spatial position of the thumb controls both the magnitude and direction of the moment arm. Based on the orientation of the subject’s thumbs in this study, as determined from the bi-planar roentgenograms, it is possible to calculate the moment arms of the muscles examined in this study using the data of Ou [30]. The product of the muscle PCSA with its moment arm at a
K. R. Kaufman et al.iClinical
Biomechanics I4 (1999) 141-150
particular joint orientation is defined as the joint moment potentiall. This parameter helps to assessthe contribution of all muscles at a joint in producing flexion-extension and abduction-adduction moments. This definition has a different meaning than Fick’s
147
muscle work capacity, since he considered the muscle’s maximum cross-sectional area and excursion length [26]. The length of the vector represents the relative magnitude of moment potential for each muscle controlling the joint. The thumb moment potential for
FPB
OPP
ADDUCTION
ADDUCTION
ADDUCTION
1.04
1.00
1 .oo
0.75
0.75
075
0.50
050 0.25
i5 cr5 m
6 .z 2
h
G m
a
E
n=4 (a)
ABUUCTION
n=4 (b)
ABDUCTION
ABDUCTION
(4
ADD
APL
EPL
ACDUCTION
ADDUCTION
ADDUCTION
1 00
1.00
0 75
0.75 0.50
n=4 (4
ll=2 @)
ABDUCTION
ABDUCTION
(0
FPL ADDUCTION
0.75
n=3 (9)
ABDUCTION
Fig. 5. Electromyographic activity of the thumb muscles after blocking the median nerve at the wrist. The data for each test subject was normalized to their individual maximum before the nerve block and is presented in the graph. A description of the muscle abbreviations is given in Table 1.
148
K. R. Kaufman et aLlClinical Biomechanics 14 (1999) 141-150
ADDUCTION 8
EXrLFNslON ... ... . .. .. .... ..
ADP
FLExtoN .. . .. .. . .. .. . .. .-
ABDUCTiON
Fig. 6. Moment potential of the thumb muscles at the CMC joint calculated from data in Ou [30]. The moment potential is defined as the product of the muscle moment arm and physiological crosssectional area. A description of the muscle abbreviations is given in Table 1.
the CMC joint (Fig. 6) demonstrates that the moment potential varies with the direction of movement. These theoretical concepts agree with the experimental measurements of force production made in this study (Fig. 2). 4.1.2. Electromyographic activity
A potentially accurate method of describing the function of a muscle is to give the direction of the resultant force produced by the muscle and the location of this force relative to the joint. EMG studies have the advantage that the muscle function can be described with respect to the spatial relationship of external force production and intrinsic muscle contraction. A muscle’s actions may be subdivided into agonistic, antagonistic and synergistic activity. Synergistic activity results in stabilizing forces about a joint. The advantage of doing in-vivo studies is that the true physiological effect of synergists and antagonists can be studied. In this study, the experimental variable was the differing direction of the force production in relation to the direction of muscle pull. Eight positions of force production were utilized. In one of these positions a selected muscle will be at its maximum mechanical advantage and thus will be a major contributor to the force produced. In an adjacent position, the direction of pull for this same muscle will become oblique to the direction of force production. Thus, the muscle will function more as a stabilizer of the joint. In subsequent positions, the ability of the muscle to contribute directly to the force production will be less as the muscle functions as an antagonist. Since the only experimental variable is the direction of force production in relation to the direction of muscle pull, the
variation in muscle electrical activity is an indication of the muscle’s in-vivo function. The direction of muscle function based on the moment arm potential (Fig. 6) correlated well with the regions of predominant muscle activity (Fig. 3). A muscle’s function can be conceptualized as a symmetrical hemicircle over which it has a functional role as an agonist, synergist, or antagonist ]31]. The primary direction of agonistic activity covers a quadrant of motion centered on the direction of peak muscle activity. This region of agonistic activity is collaborated by the joint moment potential displayed in Fig. 6. Each muscle has a region of stabilizing function which extends 90” in both directions from the agonist direction. Activity changes as the agonist function changes toward stabilization. In the synergistic region, the muscle functions at about 50% of maximal activity level. Muscle function then further decreases to a fraction of maximum as the muscle assumesan antagonist role. In the direct antagonist position, there may still be activity. It is apparent that muscles exert lower but appreciable joint stabilizing forces when they are not directly producing the dominant motion. This activity is manifested in the synergistic and antagonist quadrant. Possibly at joints where ligaments or bony surfaces bear the brunt of stabilizing activity, such stabilizing effects of muscle might be negligible. The data obtained in this study agree with those of Weathersby et al. [4], Forrest and Basmajian [5], Close and Kidd [8], Long et al. [9], and Cooney et al. [32]. Replacing the thumb muscles with muscles that closely approximate functions will provide the most stable form of thumb strength and mobility. When different tendon transfers are under consideration, one must take into account the importance of not only the agonistic action of a muscle but also the stabilizing effect of that muscle as well as the importance of antagonistic muscle action. The data in this study documents the functions of the thumb muscles for providing the actions which will result in overall balance in the thumb. Nevertheless, additional considerations of muscle architecture should also be taken into consideration in tendon transfer surgery [28,33]. 4.2. Effect of median nerve block 4.2.1. Force production
According to the usual textbook description, the motor fibers of the median nerve supplies the abductor pollis brevis, opponens pollicis, radial half (superficial head) of the flexor pollicis brevis, and the lumbricales to the index and middle fingers [34]. The median nerve therefore governs abduction and opposition movements of the thumb and, broadly speaking, may be regarded as the most important nerve of the hand. The ulnar nerve innervates all the other intrinsic hand
K. R. Kaufman et al.iClinical
ADDUCTION t
FLEXION
EXTENSIQN C.” ...-........................
Biomechanics I4 (1999) 141-150
149
analysis it is possible to study the mechanism of thumb function in a novel manner. For a joint with multiple degrees of freedom, the force production and the muscle activity will be determined by the underlying mechanical and physiological principles. This type of investigation might be valuable before transferring thumb muscle tendons, or in the development of electronically monitored hand prostheses.
MEDIAN-ELBOW
Acknowledgements
RADIAL > I
MEDIAN-WRIST
This study was supported in part by NIH grant AR 17172 and the Bush Foundation.
i A&WlJCTlON
Fig. 7. Moment potential of thumb muscles innervated by the radial, median, and ulnar nerves.
muscles. From a functional point of view, the ulnar nerve is next most important because it governs movements of all the small muscles of the hand except those innervated by the median nerve. An interesting observation may be made if the contribution of muscles are grouped according to innervation. The moment potential of this grouping (Fig. 7) depicts the combined magnitude and direction of moment which would be produced as a result of activating a particular nerve. Activation of the median nerve at the wrist provides the ability to produce force in the lower right-hand quadrant of the moment potential graph. In converse, if the medial nerve is blocked at the wrist, the ability to produce force in the lower right-hand quadrant of the moment potential graph is reduced. Referring to Fig. 4, this observation was confirmed experirnentally in this study. 4.2.2. Electromyographic activity Overlapping innervation of the flexor pollicis brevis by the ulnar nerve is common [35,36]. The ulnar nerve and the lateral head innervate the medial head of the FPB by the median nerve. Depending on the amount of cross-innervation, movement of the thumb out of the plane of the palm into opposition may be possible by means of muscles innervated by the ulnar nerve. This motion also can occur through anomalous slips of the abductor pollicis longus innervated by the radial nerve [34]. The data in this study confirm the dual innervation of the FPB. Prior to the nerve block, the FPB was active: in both flexion and abduction movements. After the nerve block, the FPB was active in flexion. Thus, the lateral head of the FPB has primarily an abduction function while the medial head of the FPB has primarily a flexion function. In summary, this study has shown that by combining electromyographic measurement and biomechanical
References [l] California Labor Code, Section No. 4660, Sacramento CA, 1992. [2] Keller J, Frost J, Silberberg N, Iverson I, Cummings R. Hand strength and dexterity (norms for clinical use). Am J Occup Therapy 1971;25:77-83. [3] Landsmeer JMF. Power grip and precision handling. Am Rheum Dis 1962;21:164-169. [4] Weathersby HT, Sutton LR, Kniseu HL. The kinesiology of muscles of the thumb; an electromyographic study. Arch Phys Med 1963;44:321-326. [5] Forrest WJ, Basmajian JV. Function of human thenar and hypothenar muscles. J Bone Joint Surg 1965;47A:1585-1594. [h] Ebskov B, Long C. A method of electromyographic kinesiology of the thumb. Arch Phys Med Rehabil 1967;48:78-84. [7] Hall EA, Long C. Intrinsic hand muscles in power grip: An electromyographic study. Electromyography 1968;8:397-421. [8] Close JR, Kidd CC. The functions of the muscles of the thumb, the index and long fingers. J Bone Joint Surg 1969;51A: 1601-1620. [9] Long C, Conrad PW, Hall EA, Fowler SL. Intrinsic-extrinsic muscle control of the hand in power grip and precision handling. J Bone Joint Surg 1970;52A:853-867. [lo] Azar I, Wu W, Tumdorf H. “Twitch-box”: A new device to monitor neuromuscular blockade. Med lnstrum 1978;12:92-93. [ll] Baraka A. Monitoring of neuromuscular transmission in anesthetized man by a bulb-transducer assembly. Anesth Analg 1973;52:36-38. [12] Brunner EA, Badola RP. A simple muscle-twitch monitor. Anesthesiology 1969;31:466-467. [13] Karis JP, Burton LW, Karis JH. A quantitative neuromuscular blockade monitor. Anesth Analg 1980;59:308-310. [14] Waits LF, Lebowitz M, Dillion JB. A means of recording force of thumb adduction. Anesthesiology 1968;29:1054-1055. [15] Ali HH, Savarese JJ. Monitoring of neuromuscular function. Anesthesiology 1976;45:216-249. [16] Katz RL. Comparison of electrical and mechanical recording of spontaneous and evoked muscle activity. Anesthesiology 1965;26:204-211. [17] An KN, Chao EYS, Askew LT. Hand strength measurement instruments. Arch Phys Med Rehabil 1980;61:366-368. [18] Kroemer KHE, Gienapp EM. Hand-held device to measure finger (thumb) strength. J Appl Physiol 1970;29:526-527. [19] Stanec A, Stanec G. The adductor pollicis monitor Apparatus and method for the quantitative measurement of the isometric contraction of the adductor pollicis muscle. Anesth Analg 1983;62:602-605.
150
K. R. Kaufman et al./Clinical Biomechanics 14 (1999) 141-150
[20] Cooney WP III, Lucca MJ, Chao EYS, Linscheid RL. The kinesiology of the thumb trapeziometacarpal joint. J Bone Joint Surg 1981;63-A:1371-1381. [21] An KN, Cooney WP, Chao EYS, Askew W, Daube JR. Determination of forces in extensor pollicis longus and flexor pollis longus of the thumb. J Appl Physiol 1983;54:714-719. [22] Cooney WP III, Chao EYS. Biomechanical analysis of static forces in the thumb during hand function. J Bone Joint Surg 1977;59-A:(1):27-36. [23] Basmajian JV, DeLuca CJ. Muscles Alive: Their functions revealed by electromyography. Baltimore: Williams and Wilkins, 198559-60. [24] Bak MJ, Loeb GE. A pulsed integrator for EMG analysis. Electroenceph Clin Neurophys 1979;47:738-741. [25] Brand PW, Beach RB, Thompson DE. Relative tension and potential excursion of muscles in the forearm and hand. J Hand Surg 1981;6:209-219. [26] Fick A. Statische berachtung der muskulature des oberschenkels. Z Rationelle Med 1850;9:94-106. [27] Law C-L, Berglund LJ, Cooney WP, An KN. Biomechanical study of tendon excursion of thumb muscles related to interphalangeal and metacarpophalangeal joint motion. The Kaohsiung J Med Sci 1989;5:(12):649-662. [28] Lieber RL, Jacobson MD, Babak MF, Abrams RA, Botte MJ. Architecture of selected muscles of the forearm: Anatomy and
[29] [30]
[31]
[32]
[33]
[34] [35]
[36]
implications for tendon transfer. J Hand Surg l992;17A:787-798. Linscheid RL, An KN, Gross RM. Quantitative analysis of intrinsic muscles of the hand. Clin Anat 1991;4:265-284. Ou CA. The biomechanics of the carpometacarpal joint of the thumb. PhD Dissertation, Department of Mechanical Engineering, Louisiana State University, 1979. Dempster WT, Finerty JC. Relative activity of wrist moving muscles in static support of the wrist joint: An electromyographic study. Am J Physiol 1947;150:596-606. Cooney WP III, An KN, Daube JR, Askew LJ. Electromyographic analysis of the thumb: A study of isometric force in pinch and grasp. J Hand Surg 1985;10A:202-210. Zajac FE. How musculotendon architecture and joint geometry affect the capacity of muscles to move and exert force on objects: A review with application and forearm tendon transfer design. Commentary. J Hand Surg 1992;17A:799-804. Kaplan EB. Functional and Surgical Anatomy of the Hand. Second Edition. Philadelphia: JB Lippincott, 1965:103-236. Forrest WJ. Motor innervation of human thenar and hypothenar muscles in 25 hands: A study combining electromyography and percutaneous nerve stimulation. Can J Surg 1967;10:196-199. Rowntree T. Anomalous innervation of the hand muscles. J Bone Joint Surg 1949;31B:505-510.
14 (1999) 141-151
In-vivo function of the thumb muscles Kenton IX. Kaufmana*, Kai-Nan An”, William J. Litchy’, William P. Cooney III”, Edmund Y. S. Chao” “Biotnechanlz Laboratory, Department of Orthopedics, Mayo Clinic/Foundation, ‘Section of E!ectromyography, Department of Neurology Mayo CliniclFoundation,
200 First Street, SW Rochester; MN 55905, USA 200 Fir.st Street, SW Rochester; MN 55905, USA
Received 10 June 1998; accepted 17 June 1998
Abstract Objective. The purpose of this study is to quantify the electrical activity of the thumb muscles responsible for the production of force in different directions of thumb movement. Design. The isometric forces and electromyographic activity generated by seven thumb muscles were measured on five normal healthy test subjects. Background. The thumb is very important for proper hand function. Presently available electromyographic studies of the thumb muscles provide only limited information. Most thumb muscles have more than one function. Additional studies are required to carefully examine and confirm the in-vivo relationship between the thumb muscle electromyogram and mechanical output. Methods. The direction and magnitude of the force vector generated at the interphalangeal joint and the relative electrical activity were obtained for eight directions of thumb action. The regions of function were defined for the abductor polhcis brevis, opponens pollicis, flexor pollicis brevis, adductor pollicis, flexor pollicis longus, extensor pollicis longus, and the abductor pollicis longus. Data was collected during voluntary isometric contraction, both before and after blocking the median nerve at the wrist. Results. The highest force production was obtained during flexion. The region of maximal muscle electrical activity varied for each muscle studied. The areas of maximal in-vivo muscle activity agreed with the moment arm data reported in the literature. The median nerve block eliminated the ability to produce force in abduction. Conclusions. This study has demonstrated that by combining electromyographic measurement and biomechanical analysis it is possible to confirm the relationship between in-vivo thumb muscle function and muscle mechanics in a novel manner. The findings of this study indicate the importance of the local anatomy in controlling the direction of force production.
Relevance When different tendon transfers are under consideration, the surgeon must take into account the importance of not only the agonistic action of a muscle but also the stabilizing and antagonistic action of that muscle. The relative electrical activity and strength contributions of specific thumb muscles measured in this study should be considered for optimal function. 0 1999 Elsevier Science Ltd. All rights reserved. Keywords: Thumb; Muscle force; Electromyographic
activity; Median nerve
1. Introduction
The thumb is very important for proper hand function. A thumb is needed for simple everyday tasks which require grasping or pinching. Experts in accident insurance value the loss of a thumb at 20-28% of total disability [l]. Loss of thumb function may also be caused by disease, e.g. rheumatoid arthritis, osteoart,hritis, carpal tunnel syndrome, or nerve injury. The loss of thumb function results in an inability to position *Corresponding
author. E-mail: [email protected]
the thumb correctly for precision type operations such as handling small objects, and results in a lack of power in opposition to the fingers [2,3]. Presently available electromyographic studies, which concern the thumb, provide only limited information. Weathersby et al. published an electromyographic analysis, but no motion tracking was employed [4]. Forrest and Basmajian studied the thenar intrinsic muscles in 25 subjects, but primarily isotonic actions of opposition, flexion and extension were utilized [S]. Their static testing was limited to clasping a wooden dowel and holding a cup. No external forces were
0268-0033/99/$ - see front matter 0 1999 Elsevier Science Ltd. All rights reserved. PII: SO268-0033(98)00058-S
142
K. R. Kaufman et aLlClinical Biomechanics 14 (1999) 141-150
measured and electromyographic (EMG) signals were not quantified in assessing the results. Ebskov and Long recorded muscle function in the unloaded thumb by using electrogoniometry and multichannel electromyography [6]. Hall and Long analyzed static pinch and power grip in a well-planned study but only tested the intrinsic muscles [7]. Various pinch and grasp actions were tried but the magnitudes of these activities were not accurately measured. No attempt was made to quantity the electromyographic activity. Close and Kidd studied synchronous recording of the thumb, index and long finger motion and the muscle action potentials [8]. They included flexion, extension, abduction, adduction and static grasp. Long et al. reported on intrinsic muscle action in power grip and precision handling [9]. Various devices have been described to measure the force exerted by the fingers or thumb. Some require movements of the fingers or the thumb [lo-141. However, the assurance of isometric conditions for the skeletal muscle being tested is an important physiologic requirement for description of muscle thumb function [l&16]. Three devices have been described to measure thumb isometric contraction forces [17-191. Yet none of these devices measure the three-dimensional force vector produced by the thumb muscles during contraction. The methods of comparing muscle activity to force are limited. Most of the muscles have more than one function. Thus, the variable function of a multidirectional muscle needs to be described in terms of its agonistic, synergistic and antagonistic contributions. Previous work has defined the kinematics of the thumb joint [20] and the types of forces in normal and pathologic hands [21,22]. Additional studies are required to carefully examine and confirm the in-vivo relationship between the thumb muscle electromyogram and the mechanical response of the thumb. The specific aims of this study were to evaluate the electrical and mechanical responses of the thumb muscles: (1) during voluntary isometric contraction; and (2) after blocking of the median nerve at the wrist.
2.2. Recording of the electromyographic activity
Intramuscular bipolar wire electrodes were used for recording muscle activity. The electrodes were inserted into the abductor pollicis brevis (APB), opponens pollicis (OPP), flexor pollicis brevis (FPB), adductor pollicis (ADP), flexor pollicis longus (FPL), extensor pollicis longus (EPL), and the abductor pollicis longus (APL). The wires were made of nickel-chrome, polyurethane insulated, 5 pm diameter wire (California Fine Wire Company, 338 South Fourth Street, Grover City, CA, USA). The fine wire electrodes were made using the technique described by Basmajian and DeLuca [23]. Both ends of the wire were stripped of insulation with an electrically heated filament. The wire was threaded through a 25 gauge, 5 cm hypodermic needle. The uninsulated section of one end of the fine wire was cut to a length of 2 mm. Under microscopic observation, a bend was made which reflected the wire to an angle of approximately 150”. This bend was placed 4 mm from the end. These electrodes were packaged and sterilized in a gas sterilizer. A TECA TE-4 (TECA Corp., 3 Campus Drive, Pleasantville, NY, USA) electromyographic system was used to amplify the muscle action potentials. The amplified signal was bandpass filtered with cut-off frequencies of 32 Hz and 3.2 KHz. Whole-wave rectification and integration was performed every 10 ms using BAK Pulsed Sample Integrators (BAK Electronics Inc., P.O. Box 87, Clarksburg, MD, USA). This device produced a positive square-wave output signal of known pulse width. The magnitude of the output signal was the integration of the input signal during the previous pulse width [24]. The integrated EMG signal was acquired with an IBM PC computer (IBM Corp., P.O. Box 1328-C Boca Raton, FL 33422) using ASYSTANT software. The integrated signal was acquired at 100 Hz/channel. All electromyographic measurements were normalized by converting them to a percent of the maximum voluntary contraction recorded during the complete testing session for each subject. 2.3. Recording of the mechanical response
2. Methods 2.1. Test subjects
A group of five normal subjects composed of three males and two females were tested. All subjects had no history of previous hand or neurologic disease. The subjects were in the age range from 20 to 40years. All subjects had occupations which involved primarily indoor office-type work. None of the subjects had leisure activities requiring excessivehand strength.
The mechanical force was measured with a customdesigned three-component load sensor (NK Biotechnical Engineering Co., P.O. Box 26335, Minneapolis, MN, USA). The general design of the transducer was based on aluminum plates separated by shear webs. Strain gages were mounted on these webs to measure deflection (strain) when subjected to a force. The mechanical concept of the sensor was based on very high mechanical impedance with the least possible bending and torsional deflections. The sensor was
143
K. R. Kaufman et aLlClinical Biomechanics 14 (1999) 141-150
carefully calibrated by the manufacturer using a loading device traceable to the National Institute for Standards and Technology (NIST). The sensor had a resolution of kO.065 N with a nonlinearity of ~0.22% full scale (F.S.) and a hysteresis of ~0.33% F.S. The transducer was constructed to minimize cross-talk of the channels. This was checked for each force direction. Although cross-talk was minimal, a 3 x 3 correction matrix was used to completely eliminate the effect of cross-talk between the output channels. All force recordings were normalized by converting them to a percent of the maximum signal recorded during the complete testing session for each subject. 2.4. Testingprotocol Upon arrival, several preliminary procedures were performed. The subject was informed of the risks as well as the rationale of the study. The subject read and signed an informed consent form. A hand history was taken to ensure no underlying pathology. Next, the wire electrodes were inserted and placement was confirmed by isolated muscle function testing. The needle was withdrawn once the correct electrode position was con:firmed and the wire was anchored in the muscle. The hand was then placed in a splint which was used for stabilization of the hand. This splint ensured a repeatable position of the hand during the testing procedure. Next, a roentgenographic examination consisting of a biplanar x-ray of the thumb was performed to further ensure no underlying pathology and to obtain the orientation of the metacarpal segments. Upon satisfactory completion of this examination, the subject was readied for data collection. The subject was seated in a chair. The thumb was positioned to simulate key pinch. In this position, the hand and wrist were immobilized in the splint with the dorsal surface of the hand upward (Fig. 1). The interphalangeal (IP) joint of the thumb was positioned in the neutral position and inserted into an adjustable metal ring which was placed around the thumb at the level of the interphalangeal joint. Prior to placing the thumb in the restraining ring, it was splinted with a metal splint to prevent any motion of the IP joint. The thumb was wrapped with foam tape so that the ring might fit tightly but not with discomfort or excessivepressure on the digital nerve or blood vessels. The subject’s hand was placed in the test device. A stable hand and wrist were required for accurate force and muscle recordings. Fixation of the hand and forearm was achieved by securing the wrist in a stand with a heavy base. The stand could be varied in height and the transducer could be moved in two horizontal directions thereby allowing for three degrees-of-freedom in positional
rbm
Fig. 1. Schematic diagram of experimental test setup. The electromyographic activity of seven thumb muscles was studied using finewire eiectrodes. Simultaneously, the isometric force production was measured with a three-component load cell.
placement of the transducer with respect to the hand. In this manner, measurement of isometric muscle functions could be achieved in a comfortable position of the hand and forearm. Force recordings and electromyographic activity were collected for maximal voluntary contraction in flexion, extension, abduction, adduction, and combinations of these movements. Accordingly, data were obtained in eight positions which were 45” apart with the vertical upward position corresponding to thumb adduction. All positions were tested twice, resulting in a total of 16 recordings being obtained. Each data collection trial was started with the subject relaxed. A command was given at which time the subject attempted a maximal contraction in the predefined direction. The subject was given visual feedback of the direction of attempted motion by displaying the two force components which corresponded to flexionextension and abduction-adduction on an oscilloscope. This permitted the subject to maintain directional control of the digit accurately. The maximal contraction was held for 2 s, at which time the data collection sequence was discontinued and the subject relaxed. A 2-min rest period was observed between data collection trials. The motion testing position proceeded in a clockwise direction. Following completion of the test sequence, the subject had the median nerve blocked at the wrist using 1% lidocaine. The nerve block was confirmed with a nerve conduction study. Evidence to support that a satisfactory block had been achieved was obtained when electrical stimulation of two times threshold failed to yield any myoelectric signal distal to the block. The threshold value was obtained immediately prior to the injection of the lidocane. Once the
144
K. R. Kaufman et aLlClinical Biamechanics 14 (1999) 141-150
block had been confirmed the subject was again asked to repeat the testing sequence described above. Force and electromyographic recordings were again collected. The force and electromyographic data were averaged for each postion. The force data were normalized to the maximum force measured throughout all trials. Similarly, electromyographic data for each muscle were normalized to the maximum electromyographic data recorded for that muscle throughout all trials.
FORCE ADDUCTION
3. Results
The force potential and electromyographic activity are presented in polar plots which display the data with respect to the carpometacarpal (CMC) joint. To retain their descriptive nature, the two orthogonal force vectors generated by flexion-extension and abductionadduction are displayed in the form of a single resultant force vector. This force vector is orientated perpendicular to the long axis of the metacarpal bone. The third orthogonal component, the axial force, would tend to compress the joint and is not displayed. All of the polar graphs have the same orientation with the positive vertical position indicating thumb adduction. Thus, these graphs can be interpreted as if the thumb was being viewed along the long axis of the metacarpal bone toward the distal phalanx. The concentric circles indicate the relative effort in terms of force or electrical activity from 0 to 100%. 3.1. Normal values 3.1.1. Force production
All subjects displayed similar patterns of normal force production (Fig. 2). The highest force production was obtained during flexion. The lowest level of force was measured during extension reaching only 25-40% of the flexion value. Generally, in all other directions the force was in the range of 25-75% of maximum. 3.1.2. Electromyographic activity
The direction of the force vectors applied by the thumb during activation of the muscles gives a direction on which the magnitude of the percentage maximal voluntary contraction is plotted (Fig. 3). Thus, the outermost circle indicates 100% voluntary activation of the muscle. The FPB was most active in the range from flexion to abduction with about 50% activity or less in extension and adduction. The OPP displayed activity in all directions of motion that included abduction. The APB displayed maximal activity in the abduction and flexion-abduction directions with about 50-75% maximal activity in the extension and abduction-extension. The ADP had a much
‘..
. ..’
“...._, ‘-.._
.. . . . . .._.._.........._...
..,..‘..
n=5
ABDUCTION Fig. 2. Maximal voluntary isometric force production at the CMC joint. The data for each test subject is presented in the graph. The data for each subject was normalized to their individual maximum.
more limited region of dominant activity. The ADP was active primarily only during flexion. The APL was most active in the region which encompassed abduction and/or extension. The EPL showed the most activity during extension and in combination with abduction or adduction functions. The FPL was the most active during flexion and/or adduction functions. The response of the muscles tested was classified as either agonistic, synergistic, or antagonistic (Table 1). Agonistic muscle function consisted of electrical activity greater than 75% maximum. Synergistic activity was classified as electrical activity greater than 25% but less than 75% maximum activity. Antagonistic activity was defined as less than 25% maximum activity. 3.2. Effects of median nerve block 3.2.1. Force Production
The ability to produce any appreciable force production in the abduction direction was essentially nonexistent after the median nerve block was applied (Fig. 4). The one subject who was able to produce a force level of approximately 20% did so by contraction of the FPB (deep head), and FPL in a synergistic manner. In contrast, the ability to produce force in either flexion, adduction or extension was not seriously affected.
K. R. Kaufman et aLlClinical
3.2.2. Electromyographic
Biomechanics 14 (1999) 141-150
was no electrical activity in the OPP and the APB. Some of the other muscles do not display activity levels of 100% because the muscles were normalized to their maximal activity throughout the total testing session
activity
The myoelectric activity of the median innervated muscles (APB, OPP, superficial head of FPB) was eliminated after the median nerve block (Fig. 5). There
(8)
145
FPB
OPP
APB
ADDUCTION
ADDUCTION
ADDUCTION
1.00
IOU
1.00
0.75
0.75
0.75
0.50
0.50
0 50
n=5
n=5
(h)
AEIDUCTION
n=5 (C)
ABDUCTION
ABDUCTION
APL
EPL
ADDUCTION
ADDUCTION
ADDUCTION
1.oo
7.oo
1.oo
0.75
a75
n=4
(4
(e)
ABDUCTION
ABDUCTION
n=5 (0
ABDUCTION
FPL ADDUCTION
n=5
(d
ABDUCTION
Fig. 3. Electromyographic activity of the thumb muscles during maximal voluntary isometric contraction. The data for each test subject is presented in the graph. The data for each subject was normalized to their individual maximum. A description of the muscle abbreviations is given in Table 1.
K. R. Kaufman et al,/Clinical Biomechanics 14 (1999) 141-150
f46
Table 1 Classification of isometric thumb muscle activity (percent of maximum) Motion
FPB
OPP
APB
ADD
Adduction Adductionlflexion Flexion Flexion/abduction Abduction Abduction/extension Extension Extension/adduction
NW2
NW) W’l)
W3) W4) NU3) W3) GW-4 S(74) S(43) NW
NW)
W07)
SW)
NO91 GW WW
GW) St581 St311
W32)
NW) NW
S(47) W'4)
S(55)
G(W
NW W4) W'l) NW) NW)
FPL
EPL
APL
S(51) W) ~(42) St511 V9) G@5) G@7)
St311
GW)
NW ~(23) S(53) G(77) G(91)
w38) w39
G = agonist (i.e. > 75% maximum activity); S = synergist (i.e. 25-75% maximum activity); N = antagonist (i.e. ~25% maximum activity). FPB, flexor pollicis brevis; OPP, opponens pollicis; APB, abductor pollicis brevis; ADP, adductor pollicis; FPL, flexor pollicis longus; EPL, _extensor pollicis longus; APL, abductor pollicis longus.
and had produced higher activity levels prior to the nerve block. The FPB (deep head) displayed activity at levels up to 75% of maximum, but over a substantially reduced range than before the block. The level of activity of the ADP was not reduced after the block, but its direction of maximal activity was shifted towards more adduction and less flexion. The FPL continued to display maximal activity in flexion and adduction. The APL was most active during extension and extensionadduction. The EPL was recruited minimally by four subjects, while one subject utilized the EPL at levels above 50% in all directions.
4.1.1. Force production
ADDUCTION
‘..,
_..’ ,_.’
‘.... .-......
// ‘....
_....._.__._...
The in-vivo study of thumb function is difficult because of the complex anatomy of the joints and muscles. A custom-designed force transducer was utilized to measure the three-dimensional force vector produced by the thumb musculature during isometric contraction. This instrument provided not only the instantaneous magnitude of strength, but also the resultant direction of the force vector produced during evaluation. The associated myoelectric activity of four intrinsic and three extrinsic muscles which produce these forces was also measured in vivo in this study. 4.1. Normal values
FORCE
“...
4. Discussion
..‘...
n=4
ABDUCTION Fig. 4. Maximal voluntary isometric force production at the CMC joint after blocking the median nerve at the wrist. The data for each test subject was normalized to their individual maximum before the nerve block and is presented in the graph.
The force production may be explained based on the muscle mechanics, physiological characteristics and activation levels. All muscles exert a force proportional to the physiological cross-sectional area [25,26]. The physiological cross-sectional area (PCSA) of the thumb musculature has been measured [27-291. The strength of a muscle is also related to the moment arm of the muscle. The movements of the thumb at the CMC are both flexion-extension and abduction-adduction [20]. The thumb motion can be thought of as following around the surface of a cone, with its apex proximal to the CMC joint and its base marked out by the moving thumbnail. The pulp of the thumb faces the axis of this cone. Most muscles attached to the thumb have a vector component of their force available for both flexion-extension and abduction-adduction [30]. The spatial position of the thumb controls both the magnitude and direction of the moment arm. Based on the orientation of the subject’s thumbs in this study, as determined from the bi-planar roentgenograms, it is possible to calculate the moment arms of the muscles examined in this study using the data of Ou [30]. The product of the muscle PCSA with its moment arm at a
K. R. Kaufman et al.iClinical
Biomechanics I4 (1999) 141-150
particular joint orientation is defined as the joint moment potentiall. This parameter helps to assessthe contribution of all muscles at a joint in producing flexion-extension and abduction-adduction moments. This definition has a different meaning than Fick’s
147
muscle work capacity, since he considered the muscle’s maximum cross-sectional area and excursion length [26]. The length of the vector represents the relative magnitude of moment potential for each muscle controlling the joint. The thumb moment potential for
FPB
OPP
ADDUCTION
ADDUCTION
ADDUCTION
1.04
1.00
1 .oo
0.75
0.75
075
0.50
050 0.25
i5 cr5 m
6 .z 2
h
G m
a
E
n=4 (a)
ABUUCTION
n=4 (b)
ABDUCTION
ABDUCTION
(4
ADD
APL
EPL
ACDUCTION
ADDUCTION
ADDUCTION
1 00
1.00
0 75
0.75 0.50
n=4 (4
ll=2 @)
ABDUCTION
ABDUCTION
(0
FPL ADDUCTION
0.75
n=3 (9)
ABDUCTION
Fig. 5. Electromyographic activity of the thumb muscles after blocking the median nerve at the wrist. The data for each test subject was normalized to their individual maximum before the nerve block and is presented in the graph. A description of the muscle abbreviations is given in Table 1.
148
K. R. Kaufman et aLlClinical Biomechanics 14 (1999) 141-150
ADDUCTION 8
EXrLFNslON ... ... . .. .. .... ..
ADP
FLExtoN .. . .. .. . .. .. . .. .-
ABDUCTiON
Fig. 6. Moment potential of the thumb muscles at the CMC joint calculated from data in Ou [30]. The moment potential is defined as the product of the muscle moment arm and physiological crosssectional area. A description of the muscle abbreviations is given in Table 1.
the CMC joint (Fig. 6) demonstrates that the moment potential varies with the direction of movement. These theoretical concepts agree with the experimental measurements of force production made in this study (Fig. 2). 4.1.2. Electromyographic activity
A potentially accurate method of describing the function of a muscle is to give the direction of the resultant force produced by the muscle and the location of this force relative to the joint. EMG studies have the advantage that the muscle function can be described with respect to the spatial relationship of external force production and intrinsic muscle contraction. A muscle’s actions may be subdivided into agonistic, antagonistic and synergistic activity. Synergistic activity results in stabilizing forces about a joint. The advantage of doing in-vivo studies is that the true physiological effect of synergists and antagonists can be studied. In this study, the experimental variable was the differing direction of the force production in relation to the direction of muscle pull. Eight positions of force production were utilized. In one of these positions a selected muscle will be at its maximum mechanical advantage and thus will be a major contributor to the force produced. In an adjacent position, the direction of pull for this same muscle will become oblique to the direction of force production. Thus, the muscle will function more as a stabilizer of the joint. In subsequent positions, the ability of the muscle to contribute directly to the force production will be less as the muscle functions as an antagonist. Since the only experimental variable is the direction of force production in relation to the direction of muscle pull, the
variation in muscle electrical activity is an indication of the muscle’s in-vivo function. The direction of muscle function based on the moment arm potential (Fig. 6) correlated well with the regions of predominant muscle activity (Fig. 3). A muscle’s function can be conceptualized as a symmetrical hemicircle over which it has a functional role as an agonist, synergist, or antagonist ]31]. The primary direction of agonistic activity covers a quadrant of motion centered on the direction of peak muscle activity. This region of agonistic activity is collaborated by the joint moment potential displayed in Fig. 6. Each muscle has a region of stabilizing function which extends 90” in both directions from the agonist direction. Activity changes as the agonist function changes toward stabilization. In the synergistic region, the muscle functions at about 50% of maximal activity level. Muscle function then further decreases to a fraction of maximum as the muscle assumesan antagonist role. In the direct antagonist position, there may still be activity. It is apparent that muscles exert lower but appreciable joint stabilizing forces when they are not directly producing the dominant motion. This activity is manifested in the synergistic and antagonist quadrant. Possibly at joints where ligaments or bony surfaces bear the brunt of stabilizing activity, such stabilizing effects of muscle might be negligible. The data obtained in this study agree with those of Weathersby et al. [4], Forrest and Basmajian [5], Close and Kidd [8], Long et al. [9], and Cooney et al. [32]. Replacing the thumb muscles with muscles that closely approximate functions will provide the most stable form of thumb strength and mobility. When different tendon transfers are under consideration, one must take into account the importance of not only the agonistic action of a muscle but also the stabilizing effect of that muscle as well as the importance of antagonistic muscle action. The data in this study documents the functions of the thumb muscles for providing the actions which will result in overall balance in the thumb. Nevertheless, additional considerations of muscle architecture should also be taken into consideration in tendon transfer surgery [28,33]. 4.2. Effect of median nerve block 4.2.1. Force production
According to the usual textbook description, the motor fibers of the median nerve supplies the abductor pollis brevis, opponens pollicis, radial half (superficial head) of the flexor pollicis brevis, and the lumbricales to the index and middle fingers [34]. The median nerve therefore governs abduction and opposition movements of the thumb and, broadly speaking, may be regarded as the most important nerve of the hand. The ulnar nerve innervates all the other intrinsic hand
K. R. Kaufman et al.iClinical
ADDUCTION t
FLEXION
EXTENSIQN C.” ...-........................
Biomechanics I4 (1999) 141-150
149
analysis it is possible to study the mechanism of thumb function in a novel manner. For a joint with multiple degrees of freedom, the force production and the muscle activity will be determined by the underlying mechanical and physiological principles. This type of investigation might be valuable before transferring thumb muscle tendons, or in the development of electronically monitored hand prostheses.
MEDIAN-ELBOW
Acknowledgements
RADIAL > I
MEDIAN-WRIST
This study was supported in part by NIH grant AR 17172 and the Bush Foundation.
i A&WlJCTlON
Fig. 7. Moment potential of thumb muscles innervated by the radial, median, and ulnar nerves.
muscles. From a functional point of view, the ulnar nerve is next most important because it governs movements of all the small muscles of the hand except those innervated by the median nerve. An interesting observation may be made if the contribution of muscles are grouped according to innervation. The moment potential of this grouping (Fig. 7) depicts the combined magnitude and direction of moment which would be produced as a result of activating a particular nerve. Activation of the median nerve at the wrist provides the ability to produce force in the lower right-hand quadrant of the moment potential graph. In converse, if the medial nerve is blocked at the wrist, the ability to produce force in the lower right-hand quadrant of the moment potential graph is reduced. Referring to Fig. 4, this observation was confirmed experirnentally in this study. 4.2.2. Electromyographic activity Overlapping innervation of the flexor pollicis brevis by the ulnar nerve is common [35,36]. The ulnar nerve and the lateral head innervate the medial head of the FPB by the median nerve. Depending on the amount of cross-innervation, movement of the thumb out of the plane of the palm into opposition may be possible by means of muscles innervated by the ulnar nerve. This motion also can occur through anomalous slips of the abductor pollicis longus innervated by the radial nerve [34]. The data in this study confirm the dual innervation of the FPB. Prior to the nerve block, the FPB was active: in both flexion and abduction movements. After the nerve block, the FPB was active in flexion. Thus, the lateral head of the FPB has primarily an abduction function while the medial head of the FPB has primarily a flexion function. In summary, this study has shown that by combining electromyographic measurement and biomechanical
References [l] California Labor Code, Section No. 4660, Sacramento CA, 1992. [2] Keller J, Frost J, Silberberg N, Iverson I, Cummings R. Hand strength and dexterity (norms for clinical use). Am J Occup Therapy 1971;25:77-83. [3] Landsmeer JMF. Power grip and precision handling. Am Rheum Dis 1962;21:164-169. [4] Weathersby HT, Sutton LR, Kniseu HL. The kinesiology of muscles of the thumb; an electromyographic study. Arch Phys Med 1963;44:321-326. [5] Forrest WJ, Basmajian JV. Function of human thenar and hypothenar muscles. J Bone Joint Surg 1965;47A:1585-1594. [h] Ebskov B, Long C. A method of electromyographic kinesiology of the thumb. Arch Phys Med Rehabil 1967;48:78-84. [7] Hall EA, Long C. Intrinsic hand muscles in power grip: An electromyographic study. Electromyography 1968;8:397-421. [8] Close JR, Kidd CC. The functions of the muscles of the thumb, the index and long fingers. J Bone Joint Surg 1969;51A: 1601-1620. [9] Long C, Conrad PW, Hall EA, Fowler SL. Intrinsic-extrinsic muscle control of the hand in power grip and precision handling. J Bone Joint Surg 1970;52A:853-867. [lo] Azar I, Wu W, Tumdorf H. “Twitch-box”: A new device to monitor neuromuscular blockade. Med lnstrum 1978;12:92-93. [ll] Baraka A. Monitoring of neuromuscular transmission in anesthetized man by a bulb-transducer assembly. Anesth Analg 1973;52:36-38. [12] Brunner EA, Badola RP. A simple muscle-twitch monitor. Anesthesiology 1969;31:466-467. [13] Karis JP, Burton LW, Karis JH. A quantitative neuromuscular blockade monitor. Anesth Analg 1980;59:308-310. [14] Waits LF, Lebowitz M, Dillion JB. A means of recording force of thumb adduction. Anesthesiology 1968;29:1054-1055. [15] Ali HH, Savarese JJ. Monitoring of neuromuscular function. Anesthesiology 1976;45:216-249. [16] Katz RL. Comparison of electrical and mechanical recording of spontaneous and evoked muscle activity. Anesthesiology 1965;26:204-211. [17] An KN, Chao EYS, Askew LT. Hand strength measurement instruments. Arch Phys Med Rehabil 1980;61:366-368. [18] Kroemer KHE, Gienapp EM. Hand-held device to measure finger (thumb) strength. J Appl Physiol 1970;29:526-527. [19] Stanec A, Stanec G. The adductor pollicis monitor Apparatus and method for the quantitative measurement of the isometric contraction of the adductor pollicis muscle. Anesth Analg 1983;62:602-605.
150
K. R. Kaufman et al./Clinical Biomechanics 14 (1999) 141-150
[20] Cooney WP III, Lucca MJ, Chao EYS, Linscheid RL. The kinesiology of the thumb trapeziometacarpal joint. J Bone Joint Surg 1981;63-A:1371-1381. [21] An KN, Cooney WP, Chao EYS, Askew W, Daube JR. Determination of forces in extensor pollicis longus and flexor pollis longus of the thumb. J Appl Physiol 1983;54:714-719. [22] Cooney WP III, Chao EYS. Biomechanical analysis of static forces in the thumb during hand function. J Bone Joint Surg 1977;59-A:(1):27-36. [23] Basmajian JV, DeLuca CJ. Muscles Alive: Their functions revealed by electromyography. Baltimore: Williams and Wilkins, 198559-60. [24] Bak MJ, Loeb GE. A pulsed integrator for EMG analysis. Electroenceph Clin Neurophys 1979;47:738-741. [25] Brand PW, Beach RB, Thompson DE. Relative tension and potential excursion of muscles in the forearm and hand. J Hand Surg 1981;6:209-219. [26] Fick A. Statische berachtung der muskulature des oberschenkels. Z Rationelle Med 1850;9:94-106. [27] Law C-L, Berglund LJ, Cooney WP, An KN. Biomechanical study of tendon excursion of thumb muscles related to interphalangeal and metacarpophalangeal joint motion. The Kaohsiung J Med Sci 1989;5:(12):649-662. [28] Lieber RL, Jacobson MD, Babak MF, Abrams RA, Botte MJ. Architecture of selected muscles of the forearm: Anatomy and
[29] [30]
[31]
[32]
[33]
[34] [35]
[36]
implications for tendon transfer. J Hand Surg l992;17A:787-798. Linscheid RL, An KN, Gross RM. Quantitative analysis of intrinsic muscles of the hand. Clin Anat 1991;4:265-284. Ou CA. The biomechanics of the carpometacarpal joint of the thumb. PhD Dissertation, Department of Mechanical Engineering, Louisiana State University, 1979. Dempster WT, Finerty JC. Relative activity of wrist moving muscles in static support of the wrist joint: An electromyographic study. Am J Physiol 1947;150:596-606. Cooney WP III, An KN, Daube JR, Askew LJ. Electromyographic analysis of the thumb: A study of isometric force in pinch and grasp. J Hand Surg 1985;10A:202-210. Zajac FE. How musculotendon architecture and joint geometry affect the capacity of muscles to move and exert force on objects: A review with application and forearm tendon transfer design. Commentary. J Hand Surg 1992;17A:799-804. Kaplan EB. Functional and Surgical Anatomy of the Hand. Second Edition. Philadelphia: JB Lippincott, 1965:103-236. Forrest WJ. Motor innervation of human thenar and hypothenar muscles in 25 hands: A study combining electromyography and percutaneous nerve stimulation. Can J Surg 1967;10:196-199. Rowntree T. Anomalous innervation of the hand muscles. J Bone Joint Surg 1949;31B:505-510.