- Email: [email protected]
Effect of grip span on maximal grip force and fatigue of flexor digitorum superficialis
Applied Ergonomics 30 (1999) 401 } 405
E!ect of grip span on maximal grip force and fatigue of #exor digitorum super"cialis John R. Blackwell *, Kurt W. Kornatz, Edward M. Heath Exercise and Sport Science, University of San Francisco, 2130 Fulton Street, San Francisco, CA 94117-1080, USA University of Colorado at Boulder, Boulder, USA University of Texas at El Paso, Texas, USA Received 21 November 1997; accepted 23 September 1998
Abstract The aim of this study was to investigate the e!ect of grip span on isometric grip force and fatigue of the #exor digitorum super"cialis (FDS) muscle during sustained voluntary contractions at 60}65% of the maximal voluntary contraction (MVC). Eighteen subjects performed isometric, submaximal gripping contractions using a grip dynamometer at four di!erent grip span settings while the pronated forearm rested on a horizontal surface. Maximal absolute grip force and median power frequency of FDS surface electromyography (EMG) during the submaximal trials were analyzed. Fatigue of FDS, as inferred from EMG frequency shifts, did not change as a function of grip size. However, middle grip sizes allowed for greater absolute forces than the small or large size. When contractions are at 60}65% MVC and the muscle is allowed to fatigue, however, grip size may be less in#uential than when maximal absolute force is required. 1999 Elsevier Science Ltd. All rights reserved. Keywords: Grip; Fatigue; Force
1. Introduction Various occupational and sport settings require a sustained level of hand prehensile (power grip) force to maximize control and performance, and to reduce the possibilities for injuries. The hand}handle interface has been investigated in a number of applied settings to study factors such as vibration damping while chipping or grinding (Burstrom, 1990; Reynolds et al., 1984), pressure distributions of the hand with di!erent hand tool designs (Yun et al., 1992), hand-held terminals for production data acquisition (Moore et al., 1991), mathematical modeling of hand-held sporting equipment (Adali and Brannigan, 1979), hand-held gasoline engines (Cundi! and Suggs, 1973) and redesigning tools to decrease manual e!ort so that the risk for upper extremity musculoskeletal disorders is reduced (Grant and Habes, 1993). Factors to consider during these activities include the
* Corresponding author. Tel: 415 422 5988; fax: 415 422 6267; e-mail: [email protected].
absolute level of grip force required as well as the rate of fatigue of the muscles responsible for grip formation. Muscle fatigue may be de"ned as a decline in the maximal contractile force of the muscle (Vollestad, 1995), and an increase in e!ort while unable to maintain a desired level of force (Enoka, 1988). It has been shown that fatigue at the muscle, commonly called localized muscle fatigue (LMF), can be objectively measured by analyzing the median frequency (f ) of the power spectrum of the
electromyographic (EMG) signal (Bigland-Ritchie et al., 1981; Komi and Tesch, 1979; Petrofsky and Lind, 1980, Potvin, 1997). Fatiguing contractions a!ect the underlying muscle physiology in a variety of ways, and so the f changes over time during the course of a sustained
contraction. The process leading to movement, which includes motoneuron excitation, action potential propagation via the axon across the neuromuscular junction to the sarcolemma, calcium-based excitation and the eventual cycling of crossbridge (actin}myosin) attachments, leading to force generation, may be a!ected by sustained contractions (Enoka, 1988). Enoka reports that a 60 s maximal isometric contraction leads to changes in
0003-6870/99/$ - see front matter 1999 Elsevier Science Ltd. All rights reserved. PII: S 0 0 0 3 - 6 8 7 0 ( 9 8 ) 0 0 0 5 5 - 6
402
J.R. Blackwell et al. / Applied Ergonomics 30 (1999) 401}405
motoneuron excitability as well as a decrease in the rate at which action potentials are discharged. This decrease leads to lower "ring frequencies. Additionally, the power shift may occur due to motor unit synchronization and decreases in conduction velocity (Basmajian and DeLuca, 1985). Therefore, as fatigue occurs, the power of the EMG signal tends to shift toward lower frequencies. Muscles that are signi"cant contributors of power grip formation are the extrinsic muscles of the hand (Grabiner, 1989). A prime "nger #exor, and consequent grip forming muscle, is the #exor digitorum super"cialis (FDS) muscle (Gray, 1985). In fact, the fourth "nger, which is served by the FDS, contributes 25}28% of the total power grip force regardless of wrist position (Hazelton et al., 1975; Ohtsuki, 1981 a, b). The purpose of this study was to measure the e!ect of grip span, thus manipulating the length of the FDS, on absolute maximal grip force levels and fatigue rates of the FDS muscle. Absolute grip force is the amount of force measured using a hand dynamometer without reference to any other value. The fatigue rate is the speed at which the muscle fatigues while gripping at a speci"c force level, in this case, expressed as a percentage of the maximal grip force. This information could lead to improved grip handle design to allow for the highest grip force as well as the lowest rate of grip muscle fatigue.
2. Methods 2.1. Data collection Right-handed, male subjects (N"18), age"25.8 (3.6) years, height"181.4 (5.9) cm, weight"773.4 (109) N, with no history of hand, wrist or arm problems participated according to the guidelines set by the University of Texas at El Paso human subjects review board, with approved consent, and did not receive any compensation. Subjects did not have any previous history of unusual gripping activities (e,g., occupation as a professional mountain biker or using a jackhammer consistently). In order to ensure that subjects did not have unequal grip capability due to di!erent hand sizes, all subjects measured between 18}22 cm, from the wrist}hand joint line to the end of the third "nger. The average hand size was 19.3 (0.9) cm. Additionally, because body fat can a!ect the frequency of a surface EMG signal, all subjects were less than 22% body fat percent. The average body fat percentage was 15.7 (4.3)%, measured using a bioelectric impedance analyzer (Bioanalogics model ELG-3259). Normal skin preparation techniques including light abrasion and cleansing with alcohol in order that the interelectrode resistance was less than 5 k). Bipolar, on-site pre-ampli"ed (gain"35), "xed distance (2 cm), silver}silver chloride, surface EMG electrodes were used along with a Therapeutics Unlimited (model 544) multichannel
EMG processor that is a di!erential ampli"er with an input impedance of '25 M) at dc and '15 M) at 100 Hz, and a frequency response of 40}4000 Hz. Prior to the study, two cadaver arms were dissected in order to allow the investigators accurate identi"cation of the FDS muscle belly, related to anatomical landmarks. The electrode set was placed over the belly of the FDS muscle, parallel to the assumed longitudinal axis of the muscle "bers. The FDS muscle was located by asking the subject to #ex the fourth "nger against external resistance while visually observing and palpating the forearm over the contracting muscle. The subjects were not allowed to curl the "nger during this test, thereby reducing the possibility that the #exor digitorum profundus, which inserts into the distal phalanges, was a major contributor to the EMG signal. To ensure that the #exor carpi radialis muscle (a wrist #exor) was not contributing signi"cantly to the EMG signal, electrode placement was considered appropriate when the EMG signal was present during "nger #exion with a voluntary stable wrist, yet absent during wrist #exion with no voluntary "nger #exion. All signals were monitored with an oscilloscope to check for artifacts and to ensure an adequate signal-to-noise ratio. Data were digitized at 1024 Hz using a 12-bit, A/D converter (Data Translation model DT2821) along with Peak Performance Technologies Analog Sampling Module software so that one-second bursts could be analyzed using a fast Fourier transform (FFT) with Run Technologies Datapac II software. Each subject was seated comfortably with the pronated forearm resting on a horizontal table top while the "ngers gripped the dynamometer which rested on the table. Therefore, wrist #exion could not occur because the forearm and hand were against "xed resistance. It was easy to identify wrist extension because this action would lift the dynamometer o! the table. The wrist was extended 403. The position of the hand remained consistent because it has been shown that grip strength can be dependent on wrist position (Hazelton et al., 1975). Subjects were required to squeeze a hand dynamometer (Jamar model 5030J1) with the gripping handles horizontal and perpendicular to the longitudinal axis of the forearm. Grip circumference sizes were: C1 (100 mm), C2 (130 mm), C3 (160 mm), and C4 (180 mm). These sizes were chosen because they span the range of very small grips to maximal grip sizes. A "fth, larger grip size setting was not used due to the fact that it was too large for the subjects and is larger than any practical gripping situation. After a demonstration by the experimenter, subjects were required to perform two maximal voluntary contractions (MVC) for three seconds at each of four grip size settings (randomly assigned) on the dynamometer. These dynamometer readings were used for the absolute grip force values. Only 3 min of rest were given between MVC trials because they were not trials that fatigued the subject. After another demonstration
J.R. Blackwell et al. / Applied Ergonomics 30 (1999) 401}405
and con"rmation that the subject understood the task, the subject was then required to hold the dynamometer force level (provided by visual feedback) within a band of 60}65% of that setting's MVC for as long as possible. During MVC trials as well as the %MVC (submaximal) trials, verbal encouragement was given by the experimenters, especially as the force levels neared the 60% MVC level at the end of a trial. The trial was considered "nished when the level of force fell below the 60% MVC level at any time after the initiation of the trial. This was performed once for each of four grip size settings, with a 10 min rest interval between each trial. It was assumed that 10 min rest was adequate for recovery due to the fact that the f recovers within 4}5 min after a sustained
contraction (Sabbahi et al., 1979; Stulen, 1980; Petrofsky and Lind, 1980; Mills, 1982; Merletti et al., 1984). Additionally, 4}5 min is enough time to allow for the removal of lactic acid after a sustained contraction (Harris et al., 1981) and for the conduction velocity of the muscle "bers to recover (Broman, 1977), which is linearly related to f (Stulen and De Luca, 1981). Due to the ease of the
task as well as the fact that it was desirable to limit the exertion of each subject prior to data collection, no practice trials were allowed.
403
Table 1 Values of t for pairwise comparisons of MVC force levels for di!erent grip settings. Settings are: 1"100 mm, 2"130 mm, 3"160 mm, 4"180 mm in circumference Grip setting
Grip setting
a
2 3 4
1
2
3
9.05 10.31 5.81
0.90 2.25
4.35
indicates P(0.008.
2.2. Data analysis The average of the two MVC trials at one grip setting was taken as the absolute MVC force level. One second bursts of surface EMG data were analyzed using an FFT at intervals of every 5 s for the length of grip formation. The FFT output revealed a power spectrum for each second-long EMG burst and the f was identi"ed for
each burst. The f EMG versus time data for each grip
setting were plotted and the slope of a line of best "t was determined. Therefore, each subject provided four slopes and four absolute grip force values, each corresponding to the four grip span settings. Additionally, the length of time a subject was able to keep the force level within the 60}65% MVC range was measured and is referred to as time to fatigue (TF). Dependent group t-tests of signi"cance, using the Bonferroni procedure for multiple (6) comparisons (alpha"0.05/6"0.0083) were used to identify signi"cant di!erences between MVC forces at di!erent grip settings, di!erences in slopes of f versus time, as well as
to identify di!erences in TF values.
3. Results Results indicated that signi"cant di!erences in absolute grip force existed between all grip sizes except the tests for size C2 versus C3 and the test for C2 versus C4 (Table 1). An inverted-U relationship between grip size and absolute grip force was evident, as lowest levels
Fig. 1. MVC absolute grip force as a function of dynamometer grip size setting. Error bars indicate standard deviations.
of force were exhibited for the smallest and largest settings, 1 and 4, and the greatest forces were found at the middle settings, 2 and 3 (Fig. 1). Fig. 2 illustrates the drop in f for all grip settings
across all subjects. Negative f slopes across all subjects
were exhibited for all four grip settings, however, pairwise comparisons for all possible pairs using dependent group t-tests revealed that none of the slopes were statistically signi"cant from each other. This suggests that the rates of fatigue while holding the dynamometer at the same relative force values were similar for all grip settings. The time to fatigue data are presented in Fig.3. Pairwise comparisons for all possible pairs resulted in no signi"cant di!erences between grip settings, thus supporting the lack of signi"cant di!erences in f slope
values.
404
J.R. Blackwell et al. / Applied Ergonomics 30 (1999) 401}405
Fig. 2. Median FDS EMG power spectrum frequency (f ) over time
during submaximal (60}65% MVC), isometric gripping contraction across all subjects for four di!erent grip settings.
Fig. 3. Time to fatigue across all subjects for four di!erent grip settings.
the TF data for the di!erent grip sizes illustrate this fact, and con"rms earlier work by Petrofsky et al. (1980), who used time as the dependent measure of endurance while squeezing a grip dynamometer. However, our data also illustrate that the absolute maximal value di!er between grip sizes. This allows the relative values to also di!er among grip sizes. For example, if a task requires 100 units of force and the maximal value capable of being produced by the muscle is 100 units, the relative value is 100%. If, however, the maximal value capable by the muscle is 200 units, the relative value is 50%. The latter case would allow for less fatigue. This is the case when the muscle is near resting length, allowing for maximal crossbridge attachments, rather than when the muscle is very short (i.e., small grip) or very long (i.e., large grip) because the number of attachments decreases and the relative force value is less at resting length. Our results illustrate that the two middle grip sizes produced greater absolute forces than the smallest or largest sizes. This also con"rms results of previous studies (Aulicino and Dupuy, 1990; Fransson and Winkel, 1991; Pheasant and O'Neill, 1975; Petrofsky et al., 1980). To take advantage of this information, each muscle contributing to grip formation should be at its optimal length when creating a grip. This could be accomplished by varying grip sizes according to hand sizes. Further, to make the grip size optimal for a speci"c person, grip handles should be designed keeping in mind the di!erent muscle lengths related to di!erent "ngers. For example, a tool or sport implement handle that is uniform in circumference does not allow for the di!erent muscle lengths of di!erent "nger #exor muscles. That is, for a particular cylinder of consistent diameter, the "fth "nger may be extended to produce a longer relative muscle length with minimal crossbridge attachments while the middle "nger might be at a length to allow for a muscle length with maximal crossbridge attachments. Therefore, grip handles should be designed that vary in size between "nger positions. A tapered cylinder would allow for the grip size that facilitates the greatest absolute force to be used for all "ngers. In real-world settings, force requirements are not necessarily set according to an individual's ability, rather the force required is dependent on an external device (e.g., the torque produced by a tennis ball hitting a tennis racket, or the force required to stabilize a pneumatic chipping tool). In other words, the requirement is absolute, not relative, and agrees with the idea that fatigability varies in proportion to the absolute force, as predicted by the force-fatigability relationship (Enoka and Stuart, 1992).
4. Discussion If a speci"c percentage of the maximal value is needed, the fatigue rates at di!erent grip sizes should be similar as long as the relative amounts of force are constant, regardless of muscle length. The similarity of the f slopes and
5. Conclusion Rate of FDS fatigue at 60}65% MVC is independent of muscle length. There is an optimal grip span for
J.R. Blackwell et al. / Applied Ergonomics 30 (1999) 401}405
producing MVC absolute grip force, and the optimal span is closely related to the middle of the length span of the "nger #exors. However, it has also been shown that handle shape and size can a!ect grip forces when di!erent tasks (e.g., pushing, pulling, wrist #exion/extension) are taken into consideration (Cochran and Riley, 1986), complicating the issue. Therefore, further research is needed related to the interaction of grip handle design and di!erent tasks.
Acknowledgement This project was supported by a grant from Gripping Solutions, Inc., El Paso, TX.
References Adali, S., Brannigan, M., 1979. Mathematical modeling of hand-held sporting equipment. J. Interdisciplinary Modeling and Simulation 2(4), 337}356. Aucilino, P.L., DuPuy, T.E., 1990. Clinical examination of the hand. In: Hunter, J., Schneider, L.H., Mackin, E.J., Callahan, A.D. (Eds.), Rehabilitation of the Hand: Surgery and Therapy. C.V. Maximal Co., St. Louis. Basmajian, J.V., DeLuca, C.J., 1985. Muscles Alive, Their Functions Revealed by Electromyography, 5th ed. Williams & Wilkins, Baltimore, MD. Bigland-Ritchie, B., Donovan, E.F., Roussos, C.S., 1981. Conduction velocity and EMG power spectrum changes in fatigue of sustained maximal e!orts. J. Appl. Physiol. 51, 1300}1305. Broman, H., 1977. An investigation on the in#uence of a sustained contraction on the succession of action potentials from a single motor unit. Electromyogr. Clin. Neurophysiol. 17, 341}358. Burstrom, L., 1990. Measurement of the mechanical energy absorption in the hand and arm whilst using vibrating tools. J. Low Frequency Noise Vib. 9(1), 1}14. Cochran, D.J. Riley, M.W., 1986. The e!ects of handle shape and size on exerted forces. Human Factors 28(3), 253}265. Cundi!, J.S., Suggs, C.W., 1973. Analytical model for rigid body vibration of a hand-held gasoline engine. ASAE Annual Meeting. Enoka, R.M., 1988. Neuromechanical Basis of Kinesiology. Human Kinetics Books, Champaign, II. Enoka, R.M., Stuart, D.G., 1992. Neurobiology of muscle fatigue. J. Physiol. 17, 1631}1648. Fransson, C., Winkel, J., 1991. Hand strength: the in#uence of grip span and grip type. Ergonomics 34(7), 881}892. Grabiner, M.D., 1989. The wrist and hand. In: Rasch, P.J., (Ed.), Kinesiology and Applied Anatomy, 7th ed., Lea & Febiger, Philadelphia, PA, pp. 151}168. Grant, K.A., Habes, D.J., 1993. E!ectiveness of a handle #ange for reducing manual e!ort during hand tool use. Int. J. Ind. Ergonomics 12(30), 199}207.
405
Gray, H., 1985. In: Clemente, C.D., (Ed.,) Anatomy of the Human Body. Clemente, C.D., Lea & Febiger, Philadelphia. Harris, R.C., Hultman, E., Sahlin, E., 1981. Glycolytic intermediates in human muscle o!er isometric contraction. Pfugers Arch 389, 277282. Hazelton, F.T., Smidt, G.L., Flatt, A.D., Stephens, R.I., 1975. The in#uence of wrist position on the force produced by the "nger #exors. J. Biomechanics 8,301}306. Komi, P.V., Tesch, P., 1979. EMG frequency spectrum, muscle structure, and fatigue during dynamic contractions in man. European J. Appl. Physiol. 42, 4}50. Merletti, R., Sabbahi, M.A., De Luca, C.J., 1984. Median frequency of the myoelectric signal: e!ects of ischemia and cooling. European J. App. Physiol. Mills, K.S., 1982. Power spectral analysis of electromyogram and compound muscle action potential during muscle fatigue and recovery. J. Physiol. 362, 401}409. Moore, R.D., Dennison, J.A., Greenhill, K.L., Sisk, L.B., 1991. Generic hand-held terminal for production data acquisition. Proc. Society of Petroleum Engineers Production Operations. pp. 829}836. Ohtsuki, T., 1981a. Inhibition of individual "ngers during grip strength exertion. Ergonomics 24 (21). Ohtsuki, T., 1981b. Decrease in grip strength induced by simultaneous bilateral exertion with reference to "nger strength. Ergonomics 24(37). Pheasant, S., O'Neill, D., 1975. Performance in gripping and turning-A study in hand/handle e!ectiveness. Appl. Ergonomics 6,205}208. Petrofsky, J.S., Lind, A.R., 1980. Requency analysis of surface electromyogram during sustained isometric contraction. European J. Appl. Physiol. 43, 173}182. Petrofsky, J.S., Williams, C., Kamen, G., Lind, A., 1980. The e!ect of handgrip span on isometric exercise performance. Ergonomics 23 (12), 1129}1135. Potvin, J.R., 1997. E!ects of muscle kinematics on surface EMG amplitude and frequency during fatiguing dynamic contractions. J. Appl. Physiol. 82(1), 144}151. Reynolds, D.D., Basel, R., Wasserman, D.E., Taylor, W., 1984. A study of hand vibration on chipping and grinding operators. J. Sound Vibr. 95(4), 479}497. Sabbahi, M.A., DeLuca, C.J., Powers, W.R., 1979. The e!ect of ischemia, cooling and local anesthesia on the median frequency of the myoelectric signal. Proc. 4th Cong. Int. Soc. of Electrophysiological Kinesiology. pp. 94}95. Stulen, F.B., 1980. A technique to monitor localized muscular fatigue using frequency domain analysis for the myoelectric signal. Ph.D.Dissertation, Massachusetts Institute of Technology, Cambridge, MA. Stulen, F.B., DeLuca, C.J., 1981. Frequency parameters of the myoelectric signal as a measure of muscle conduction velocity. IEEE Trans Biomed Engng 28, 515}523. Vollestad, N.K., 1995. Metabolic correlates of fatigue from di!erent types of exercise in man. In:Gandevia, S.C., Enoka, R.M., McComas, A.J., Stuart, D.G., Thomas, C.K., (Ed.), Fatigue; Neural and Muscular Mechanisms. Adv.Exp. Med. Biol. series, 384, 185}194, Plenum Press, New York. Yun, M.H., Kotani, K., Ellis, D., 1992. Using force sensitive resistors to evaluate hand tool grip design. Proc. 36th Annual Meeting of the Human Factors Society. Vol. 1. pp. 806}810.
E!ect of grip span on maximal grip force and fatigue of #exor digitorum super"cialis John R. Blackwell *, Kurt W. Kornatz, Edward M. Heath Exercise and Sport Science, University of San Francisco, 2130 Fulton Street, San Francisco, CA 94117-1080, USA University of Colorado at Boulder, Boulder, USA University of Texas at El Paso, Texas, USA Received 21 November 1997; accepted 23 September 1998
Abstract The aim of this study was to investigate the e!ect of grip span on isometric grip force and fatigue of the #exor digitorum super"cialis (FDS) muscle during sustained voluntary contractions at 60}65% of the maximal voluntary contraction (MVC). Eighteen subjects performed isometric, submaximal gripping contractions using a grip dynamometer at four di!erent grip span settings while the pronated forearm rested on a horizontal surface. Maximal absolute grip force and median power frequency of FDS surface electromyography (EMG) during the submaximal trials were analyzed. Fatigue of FDS, as inferred from EMG frequency shifts, did not change as a function of grip size. However, middle grip sizes allowed for greater absolute forces than the small or large size. When contractions are at 60}65% MVC and the muscle is allowed to fatigue, however, grip size may be less in#uential than when maximal absolute force is required. 1999 Elsevier Science Ltd. All rights reserved. Keywords: Grip; Fatigue; Force
1. Introduction Various occupational and sport settings require a sustained level of hand prehensile (power grip) force to maximize control and performance, and to reduce the possibilities for injuries. The hand}handle interface has been investigated in a number of applied settings to study factors such as vibration damping while chipping or grinding (Burstrom, 1990; Reynolds et al., 1984), pressure distributions of the hand with di!erent hand tool designs (Yun et al., 1992), hand-held terminals for production data acquisition (Moore et al., 1991), mathematical modeling of hand-held sporting equipment (Adali and Brannigan, 1979), hand-held gasoline engines (Cundi! and Suggs, 1973) and redesigning tools to decrease manual e!ort so that the risk for upper extremity musculoskeletal disorders is reduced (Grant and Habes, 1993). Factors to consider during these activities include the
* Corresponding author. Tel: 415 422 5988; fax: 415 422 6267; e-mail: [email protected].
absolute level of grip force required as well as the rate of fatigue of the muscles responsible for grip formation. Muscle fatigue may be de"ned as a decline in the maximal contractile force of the muscle (Vollestad, 1995), and an increase in e!ort while unable to maintain a desired level of force (Enoka, 1988). It has been shown that fatigue at the muscle, commonly called localized muscle fatigue (LMF), can be objectively measured by analyzing the median frequency (f ) of the power spectrum of the
electromyographic (EMG) signal (Bigland-Ritchie et al., 1981; Komi and Tesch, 1979; Petrofsky and Lind, 1980, Potvin, 1997). Fatiguing contractions a!ect the underlying muscle physiology in a variety of ways, and so the f changes over time during the course of a sustained
contraction. The process leading to movement, which includes motoneuron excitation, action potential propagation via the axon across the neuromuscular junction to the sarcolemma, calcium-based excitation and the eventual cycling of crossbridge (actin}myosin) attachments, leading to force generation, may be a!ected by sustained contractions (Enoka, 1988). Enoka reports that a 60 s maximal isometric contraction leads to changes in
0003-6870/99/$ - see front matter 1999 Elsevier Science Ltd. All rights reserved. PII: S 0 0 0 3 - 6 8 7 0 ( 9 8 ) 0 0 0 5 5 - 6
402
J.R. Blackwell et al. / Applied Ergonomics 30 (1999) 401}405
motoneuron excitability as well as a decrease in the rate at which action potentials are discharged. This decrease leads to lower "ring frequencies. Additionally, the power shift may occur due to motor unit synchronization and decreases in conduction velocity (Basmajian and DeLuca, 1985). Therefore, as fatigue occurs, the power of the EMG signal tends to shift toward lower frequencies. Muscles that are signi"cant contributors of power grip formation are the extrinsic muscles of the hand (Grabiner, 1989). A prime "nger #exor, and consequent grip forming muscle, is the #exor digitorum super"cialis (FDS) muscle (Gray, 1985). In fact, the fourth "nger, which is served by the FDS, contributes 25}28% of the total power grip force regardless of wrist position (Hazelton et al., 1975; Ohtsuki, 1981 a, b). The purpose of this study was to measure the e!ect of grip span, thus manipulating the length of the FDS, on absolute maximal grip force levels and fatigue rates of the FDS muscle. Absolute grip force is the amount of force measured using a hand dynamometer without reference to any other value. The fatigue rate is the speed at which the muscle fatigues while gripping at a speci"c force level, in this case, expressed as a percentage of the maximal grip force. This information could lead to improved grip handle design to allow for the highest grip force as well as the lowest rate of grip muscle fatigue.
2. Methods 2.1. Data collection Right-handed, male subjects (N"18), age"25.8 (3.6) years, height"181.4 (5.9) cm, weight"773.4 (109) N, with no history of hand, wrist or arm problems participated according to the guidelines set by the University of Texas at El Paso human subjects review board, with approved consent, and did not receive any compensation. Subjects did not have any previous history of unusual gripping activities (e,g., occupation as a professional mountain biker or using a jackhammer consistently). In order to ensure that subjects did not have unequal grip capability due to di!erent hand sizes, all subjects measured between 18}22 cm, from the wrist}hand joint line to the end of the third "nger. The average hand size was 19.3 (0.9) cm. Additionally, because body fat can a!ect the frequency of a surface EMG signal, all subjects were less than 22% body fat percent. The average body fat percentage was 15.7 (4.3)%, measured using a bioelectric impedance analyzer (Bioanalogics model ELG-3259). Normal skin preparation techniques including light abrasion and cleansing with alcohol in order that the interelectrode resistance was less than 5 k). Bipolar, on-site pre-ampli"ed (gain"35), "xed distance (2 cm), silver}silver chloride, surface EMG electrodes were used along with a Therapeutics Unlimited (model 544) multichannel
EMG processor that is a di!erential ampli"er with an input impedance of '25 M) at dc and '15 M) at 100 Hz, and a frequency response of 40}4000 Hz. Prior to the study, two cadaver arms were dissected in order to allow the investigators accurate identi"cation of the FDS muscle belly, related to anatomical landmarks. The electrode set was placed over the belly of the FDS muscle, parallel to the assumed longitudinal axis of the muscle "bers. The FDS muscle was located by asking the subject to #ex the fourth "nger against external resistance while visually observing and palpating the forearm over the contracting muscle. The subjects were not allowed to curl the "nger during this test, thereby reducing the possibility that the #exor digitorum profundus, which inserts into the distal phalanges, was a major contributor to the EMG signal. To ensure that the #exor carpi radialis muscle (a wrist #exor) was not contributing signi"cantly to the EMG signal, electrode placement was considered appropriate when the EMG signal was present during "nger #exion with a voluntary stable wrist, yet absent during wrist #exion with no voluntary "nger #exion. All signals were monitored with an oscilloscope to check for artifacts and to ensure an adequate signal-to-noise ratio. Data were digitized at 1024 Hz using a 12-bit, A/D converter (Data Translation model DT2821) along with Peak Performance Technologies Analog Sampling Module software so that one-second bursts could be analyzed using a fast Fourier transform (FFT) with Run Technologies Datapac II software. Each subject was seated comfortably with the pronated forearm resting on a horizontal table top while the "ngers gripped the dynamometer which rested on the table. Therefore, wrist #exion could not occur because the forearm and hand were against "xed resistance. It was easy to identify wrist extension because this action would lift the dynamometer o! the table. The wrist was extended 403. The position of the hand remained consistent because it has been shown that grip strength can be dependent on wrist position (Hazelton et al., 1975). Subjects were required to squeeze a hand dynamometer (Jamar model 5030J1) with the gripping handles horizontal and perpendicular to the longitudinal axis of the forearm. Grip circumference sizes were: C1 (100 mm), C2 (130 mm), C3 (160 mm), and C4 (180 mm). These sizes were chosen because they span the range of very small grips to maximal grip sizes. A "fth, larger grip size setting was not used due to the fact that it was too large for the subjects and is larger than any practical gripping situation. After a demonstration by the experimenter, subjects were required to perform two maximal voluntary contractions (MVC) for three seconds at each of four grip size settings (randomly assigned) on the dynamometer. These dynamometer readings were used for the absolute grip force values. Only 3 min of rest were given between MVC trials because they were not trials that fatigued the subject. After another demonstration
J.R. Blackwell et al. / Applied Ergonomics 30 (1999) 401}405
and con"rmation that the subject understood the task, the subject was then required to hold the dynamometer force level (provided by visual feedback) within a band of 60}65% of that setting's MVC for as long as possible. During MVC trials as well as the %MVC (submaximal) trials, verbal encouragement was given by the experimenters, especially as the force levels neared the 60% MVC level at the end of a trial. The trial was considered "nished when the level of force fell below the 60% MVC level at any time after the initiation of the trial. This was performed once for each of four grip size settings, with a 10 min rest interval between each trial. It was assumed that 10 min rest was adequate for recovery due to the fact that the f recovers within 4}5 min after a sustained
contraction (Sabbahi et al., 1979; Stulen, 1980; Petrofsky and Lind, 1980; Mills, 1982; Merletti et al., 1984). Additionally, 4}5 min is enough time to allow for the removal of lactic acid after a sustained contraction (Harris et al., 1981) and for the conduction velocity of the muscle "bers to recover (Broman, 1977), which is linearly related to f (Stulen and De Luca, 1981). Due to the ease of the
task as well as the fact that it was desirable to limit the exertion of each subject prior to data collection, no practice trials were allowed.
403
Table 1 Values of t for pairwise comparisons of MVC force levels for di!erent grip settings. Settings are: 1"100 mm, 2"130 mm, 3"160 mm, 4"180 mm in circumference Grip setting
Grip setting
a
2 3 4
1
2
3
9.05 10.31 5.81
0.90 2.25
4.35
indicates P(0.008.
2.2. Data analysis The average of the two MVC trials at one grip setting was taken as the absolute MVC force level. One second bursts of surface EMG data were analyzed using an FFT at intervals of every 5 s for the length of grip formation. The FFT output revealed a power spectrum for each second-long EMG burst and the f was identi"ed for
each burst. The f EMG versus time data for each grip
setting were plotted and the slope of a line of best "t was determined. Therefore, each subject provided four slopes and four absolute grip force values, each corresponding to the four grip span settings. Additionally, the length of time a subject was able to keep the force level within the 60}65% MVC range was measured and is referred to as time to fatigue (TF). Dependent group t-tests of signi"cance, using the Bonferroni procedure for multiple (6) comparisons (alpha"0.05/6"0.0083) were used to identify signi"cant di!erences between MVC forces at di!erent grip settings, di!erences in slopes of f versus time, as well as
to identify di!erences in TF values.
3. Results Results indicated that signi"cant di!erences in absolute grip force existed between all grip sizes except the tests for size C2 versus C3 and the test for C2 versus C4 (Table 1). An inverted-U relationship between grip size and absolute grip force was evident, as lowest levels
Fig. 1. MVC absolute grip force as a function of dynamometer grip size setting. Error bars indicate standard deviations.
of force were exhibited for the smallest and largest settings, 1 and 4, and the greatest forces were found at the middle settings, 2 and 3 (Fig. 1). Fig. 2 illustrates the drop in f for all grip settings
across all subjects. Negative f slopes across all subjects
were exhibited for all four grip settings, however, pairwise comparisons for all possible pairs using dependent group t-tests revealed that none of the slopes were statistically signi"cant from each other. This suggests that the rates of fatigue while holding the dynamometer at the same relative force values were similar for all grip settings. The time to fatigue data are presented in Fig.3. Pairwise comparisons for all possible pairs resulted in no signi"cant di!erences between grip settings, thus supporting the lack of signi"cant di!erences in f slope
values.
404
J.R. Blackwell et al. / Applied Ergonomics 30 (1999) 401}405
Fig. 2. Median FDS EMG power spectrum frequency (f ) over time
during submaximal (60}65% MVC), isometric gripping contraction across all subjects for four di!erent grip settings.
Fig. 3. Time to fatigue across all subjects for four di!erent grip settings.
the TF data for the di!erent grip sizes illustrate this fact, and con"rms earlier work by Petrofsky et al. (1980), who used time as the dependent measure of endurance while squeezing a grip dynamometer. However, our data also illustrate that the absolute maximal value di!er between grip sizes. This allows the relative values to also di!er among grip sizes. For example, if a task requires 100 units of force and the maximal value capable of being produced by the muscle is 100 units, the relative value is 100%. If, however, the maximal value capable by the muscle is 200 units, the relative value is 50%. The latter case would allow for less fatigue. This is the case when the muscle is near resting length, allowing for maximal crossbridge attachments, rather than when the muscle is very short (i.e., small grip) or very long (i.e., large grip) because the number of attachments decreases and the relative force value is less at resting length. Our results illustrate that the two middle grip sizes produced greater absolute forces than the smallest or largest sizes. This also con"rms results of previous studies (Aulicino and Dupuy, 1990; Fransson and Winkel, 1991; Pheasant and O'Neill, 1975; Petrofsky et al., 1980). To take advantage of this information, each muscle contributing to grip formation should be at its optimal length when creating a grip. This could be accomplished by varying grip sizes according to hand sizes. Further, to make the grip size optimal for a speci"c person, grip handles should be designed keeping in mind the di!erent muscle lengths related to di!erent "ngers. For example, a tool or sport implement handle that is uniform in circumference does not allow for the di!erent muscle lengths of di!erent "nger #exor muscles. That is, for a particular cylinder of consistent diameter, the "fth "nger may be extended to produce a longer relative muscle length with minimal crossbridge attachments while the middle "nger might be at a length to allow for a muscle length with maximal crossbridge attachments. Therefore, grip handles should be designed that vary in size between "nger positions. A tapered cylinder would allow for the grip size that facilitates the greatest absolute force to be used for all "ngers. In real-world settings, force requirements are not necessarily set according to an individual's ability, rather the force required is dependent on an external device (e.g., the torque produced by a tennis ball hitting a tennis racket, or the force required to stabilize a pneumatic chipping tool). In other words, the requirement is absolute, not relative, and agrees with the idea that fatigability varies in proportion to the absolute force, as predicted by the force-fatigability relationship (Enoka and Stuart, 1992).
4. Discussion If a speci"c percentage of the maximal value is needed, the fatigue rates at di!erent grip sizes should be similar as long as the relative amounts of force are constant, regardless of muscle length. The similarity of the f slopes and
5. Conclusion Rate of FDS fatigue at 60}65% MVC is independent of muscle length. There is an optimal grip span for
J.R. Blackwell et al. / Applied Ergonomics 30 (1999) 401}405
producing MVC absolute grip force, and the optimal span is closely related to the middle of the length span of the "nger #exors. However, it has also been shown that handle shape and size can a!ect grip forces when di!erent tasks (e.g., pushing, pulling, wrist #exion/extension) are taken into consideration (Cochran and Riley, 1986), complicating the issue. Therefore, further research is needed related to the interaction of grip handle design and di!erent tasks.
Acknowledgement This project was supported by a grant from Gripping Solutions, Inc., El Paso, TX.
References Adali, S., Brannigan, M., 1979. Mathematical modeling of hand-held sporting equipment. J. Interdisciplinary Modeling and Simulation 2(4), 337}356. Aucilino, P.L., DuPuy, T.E., 1990. Clinical examination of the hand. In: Hunter, J., Schneider, L.H., Mackin, E.J., Callahan, A.D. (Eds.), Rehabilitation of the Hand: Surgery and Therapy. C.V. Maximal Co., St. Louis. Basmajian, J.V., DeLuca, C.J., 1985. Muscles Alive, Their Functions Revealed by Electromyography, 5th ed. Williams & Wilkins, Baltimore, MD. Bigland-Ritchie, B., Donovan, E.F., Roussos, C.S., 1981. Conduction velocity and EMG power spectrum changes in fatigue of sustained maximal e!orts. J. Appl. Physiol. 51, 1300}1305. Broman, H., 1977. An investigation on the in#uence of a sustained contraction on the succession of action potentials from a single motor unit. Electromyogr. Clin. Neurophysiol. 17, 341}358. Burstrom, L., 1990. Measurement of the mechanical energy absorption in the hand and arm whilst using vibrating tools. J. Low Frequency Noise Vib. 9(1), 1}14. Cochran, D.J. Riley, M.W., 1986. The e!ects of handle shape and size on exerted forces. Human Factors 28(3), 253}265. Cundi!, J.S., Suggs, C.W., 1973. Analytical model for rigid body vibration of a hand-held gasoline engine. ASAE Annual Meeting. Enoka, R.M., 1988. Neuromechanical Basis of Kinesiology. Human Kinetics Books, Champaign, II. Enoka, R.M., Stuart, D.G., 1992. Neurobiology of muscle fatigue. J. Physiol. 17, 1631}1648. Fransson, C., Winkel, J., 1991. Hand strength: the in#uence of grip span and grip type. Ergonomics 34(7), 881}892. Grabiner, M.D., 1989. The wrist and hand. In: Rasch, P.J., (Ed.), Kinesiology and Applied Anatomy, 7th ed., Lea & Febiger, Philadelphia, PA, pp. 151}168. Grant, K.A., Habes, D.J., 1993. E!ectiveness of a handle #ange for reducing manual e!ort during hand tool use. Int. J. Ind. Ergonomics 12(30), 199}207.
405
Gray, H., 1985. In: Clemente, C.D., (Ed.,) Anatomy of the Human Body. Clemente, C.D., Lea & Febiger, Philadelphia. Harris, R.C., Hultman, E., Sahlin, E., 1981. Glycolytic intermediates in human muscle o!er isometric contraction. Pfugers Arch 389, 277282. Hazelton, F.T., Smidt, G.L., Flatt, A.D., Stephens, R.I., 1975. The in#uence of wrist position on the force produced by the "nger #exors. J. Biomechanics 8,301}306. Komi, P.V., Tesch, P., 1979. EMG frequency spectrum, muscle structure, and fatigue during dynamic contractions in man. European J. Appl. Physiol. 42, 4}50. Merletti, R., Sabbahi, M.A., De Luca, C.J., 1984. Median frequency of the myoelectric signal: e!ects of ischemia and cooling. European J. App. Physiol. Mills, K.S., 1982. Power spectral analysis of electromyogram and compound muscle action potential during muscle fatigue and recovery. J. Physiol. 362, 401}409. Moore, R.D., Dennison, J.A., Greenhill, K.L., Sisk, L.B., 1991. Generic hand-held terminal for production data acquisition. Proc. Society of Petroleum Engineers Production Operations. pp. 829}836. Ohtsuki, T., 1981a. Inhibition of individual "ngers during grip strength exertion. Ergonomics 24 (21). Ohtsuki, T., 1981b. Decrease in grip strength induced by simultaneous bilateral exertion with reference to "nger strength. Ergonomics 24(37). Pheasant, S., O'Neill, D., 1975. Performance in gripping and turning-A study in hand/handle e!ectiveness. Appl. Ergonomics 6,205}208. Petrofsky, J.S., Lind, A.R., 1980. Requency analysis of surface electromyogram during sustained isometric contraction. European J. Appl. Physiol. 43, 173}182. Petrofsky, J.S., Williams, C., Kamen, G., Lind, A., 1980. The e!ect of handgrip span on isometric exercise performance. Ergonomics 23 (12), 1129}1135. Potvin, J.R., 1997. E!ects of muscle kinematics on surface EMG amplitude and frequency during fatiguing dynamic contractions. J. Appl. Physiol. 82(1), 144}151. Reynolds, D.D., Basel, R., Wasserman, D.E., Taylor, W., 1984. A study of hand vibration on chipping and grinding operators. J. Sound Vibr. 95(4), 479}497. Sabbahi, M.A., DeLuca, C.J., Powers, W.R., 1979. The e!ect of ischemia, cooling and local anesthesia on the median frequency of the myoelectric signal. Proc. 4th Cong. Int. Soc. of Electrophysiological Kinesiology. pp. 94}95. Stulen, F.B., 1980. A technique to monitor localized muscular fatigue using frequency domain analysis for the myoelectric signal. Ph.D.Dissertation, Massachusetts Institute of Technology, Cambridge, MA. Stulen, F.B., DeLuca, C.J., 1981. Frequency parameters of the myoelectric signal as a measure of muscle conduction velocity. IEEE Trans Biomed Engng 28, 515}523. Vollestad, N.K., 1995. Metabolic correlates of fatigue from di!erent types of exercise in man. In:Gandevia, S.C., Enoka, R.M., McComas, A.J., Stuart, D.G., Thomas, C.K., (Ed.), Fatigue; Neural and Muscular Mechanisms. Adv.Exp. Med. Biol. series, 384, 185}194, Plenum Press, New York. Yun, M.H., Kotani, K., Ellis, D., 1992. Using force sensitive resistors to evaluate hand tool grip design. Proc. 36th Annual Meeting of the Human Factors Society. Vol. 1. pp. 806}810.