Human isometric force production and electromyogram ... - Springer Link

Eur J Appl Physiol (1999) 80:52±56

Ó Springer-Verlag 1999

ORIGINAL ARTICLE

Tapani PoÈyhoÈnen á Kari L. Keskinen á Arto Hautala Jukka Savolainen á Esko MaÈlkiaÈ

Human isometric force production and electromyogram activity of knee extensor muscles in water and on dry land

Accepted: 16 January 1999

Abstract This study was designed to determine trial-totrial and day-to-day reproducibility of isometric force and electromyogram activity (EMG) of the knee extensor muscles in water and on dry land as well as to make comparisons between the two training conditions in muscle activity and force production. A group of 20 healthy subjects (12 women and 8 men) were tested three times over 2 weeks. A measurement session consisted of recordings of maximal and submaximal isometric knee extension force with simultaneous recording of surface EMG from the vastus medialis, vastus lateralis and biceps femoris muscles. To ensure identical measurement conditions the same patient elevator chair was used in both the dry and the wet environment. Intraclass correlation coecients (ICC) and coecients of variation (CV) showed high trial-to-trial (ICC=0.95±0.99, CV=3.5%±11%) and day-to-day reproducibility (ICC=0.85±0.98, CV=11%±19%) for underwater and dry land measurements of force and EMG in each muscle during maximal contractions. The day-to-day reproducibility for submaximal contractions was similar. The interesting ®nding was that underwater EMG amplitude decreased signi®cantly in each muscle during maximal (P < 0.01±P < 0.001) and submaximal contractions (P < 0.05±P < 0.001). However, the isometric force measurements showed similar values in both wet and dry conditions. The water had no disturbing e€ect on the electrodes as shown by slightly lowered

T. PoÈyhoÈnen (&) á E. MaÈlkiaÈ University of JyvaÈskylaÈ, Department of Health Sciences, PL 35, FIN 40351 K.L. Keskinen University of JyvaÈskylaÈ, Department of Biology of Physical Activity, JyvaÈskylaÈ, Finland A. Hautala á J. Savolainen Central Hospital of Kymenlaakso, Department of Physiatrics, Kotka, Finland

interelectrode resistance values, the absence of artefacts and low noise levels of the EMG signals. It was concluded that underwater force and EMG measurements are highly reproducible. The signi®cant decrease of underwater EMG could have electromechanical and / or neurophysiological explanations. Key words Hydrotherapy á Reproducibility á Isometric force á Electromyography

Introduction Despite the fact that aquatic rehabilitation and water exercises are widely practised, there have been only few biomechanical studies of hydrodynamic e€ects on normal and pathological neuromuscular systems under water. On the other hand, force measurements and electromyogram (EMG) examinations have been performed on swimmers related to their swimming techniques and the energetics of swimming. Propulsive and resistance forces have been measured in dynamic conditions during both tethered and free swimming (Karpovich 1935; Alley 1952; Keskinen et al. 1989) while Berger et al. (1995) have measured hydrodynamic forces on human hand/arm models. Ikai et al. (1964) were the ®rst to record underwater EMG signals during swimming and Lewillie (1968) has introduced techniques using telemetered EMG. Okamoto and Wolf (1980) have found excellent correlations between ®ne wire and surface electrodes in underwater EMG. In Brussels, Clarys and co-workers have published many EMG studies of swimming. Clarys et al. (1985) have made comparisons between conventional and telemetered underwater EMG and have found that signal amplitude was smaller when measured telemetrically compared to using traditional methods. They have reported also that the signal amplitude under water was decreased compared with signals measured on dry land during maximal muscle contractions. However, muscle force was not measured. In neurophysiology, Dietz et al.

53

(1989) have used underwater surface EMG measurements and a moving force platform to investigate the signi®cance of gravity in the generation of postural adjustments. The results have indicated that the function of re¯exes to stabilise human posture may depend mainly on the activity of pressure receptors in joints and in the sole of the foot in comparison to vestibulospinal and muscle proprioceptor re¯ex mechanisms. However, while in many studies the EMG and force measurements have been used to analyse swimming techniques, no work has speci®cally looked at the reproducibility of the measurements in controlled conditions under water. Similarly, no comparisons between EMG during isometric contractions in the water and on dry land have been made in swimming or in rehabilitation. In fact, aquatic rehabilitation is an area largely unexplored in terms of basic hydrodynamic conditions, and especially where biomechanical methods are involved. Isometric knee extension was selected as the test contraction because reproducibility of EMG in knee extensor muscles has been widely reported and isometric contraction can be compared between under water and dry conditions. Therefore, the present study was designed to determined the trial-to-trial and day-to-day reproducibility of EMG recordings as well as the reproducibility of isometric force measurements in water compared with reproducibility on dry land. In addition, the purpose of this study was to make comparisons between EMG in knee extensor muscles and isometric force production using a standardised design in these two di€erent environments.

Methods Subjects After gaining the approval of the Central Hospital of Kymenlaakso Ethics Committee, 12 healthy women mean age 32.3 (SD 5.5) years, height 166.0 (SD 3.1) cm, body mass 61.3 (SD 4.7) kg, body mass index (BMI) 22.2 (SD 1.4) and 8 healthy men, mean age 28.0 (SD 4.8) years, height 182.7 (SD 2.9) cm, body mass 79.5 (SD 7.5) kg, BMI 23.9 (SD 2.7) volunteers, gave their written informed consent and agreed to participate in this study. According to activity questionnaires the participants were habitually active. The subjects visited the measurements laboratory three times over 2 weeks, separated by at least 48 h. In each of the three measurement sessions the subjects performed identical force tests with simultaneous EMG recording on dry land and in the water. Of the subjects 10 were randomly selected to start the ®rst session on dry land. The treatment order was changed in the next two sessions. The subjects were asked to avoid vigorous exercise for 24 h before the experiments. Measurements were performed using the same patient elevator chair at the poolside and in the therapeutic pool (Fig. 1). The room temperature and water temperature were 26°C and 30°C during the measurement days, respectively. Experiment procedures Force measurements The subjects were carefully familiarised with the test procedure and trained to produce the maximal force output in knee extension

Fig. 1 The patient elevator chair at the poolside before each measurement session. The warming-up was standardised as follows: 10 min of cycling on a stationary cycle ergometer, at a heart rate 120±130 beats á min)1 plus 4 min of stretching exercises for the lower legs. The isometric force exerted by the dominant leg was measured in the patient elevator chair in a sitting position with knee and hip angles set at 90°. Securing straps were used to immobilise the thigh and hip while sitting in the measurement chair. A waterproof strain gauge dynamometer (range 0±200 N, sampling rate 1000 Hz) was integrated with the eightchannel EMG system (ME 4000, Mega Electronics, Kuopio, Finland). The dynamometer was attached around the ankle, 2.5 cm above the mid-point of the malleolus lateralis. The strain gauge dynamometer was calibrated before each measurement session. After two submaximal repetitions, the subjects were instructed to perform the muscle contractions (4 s) as fast and as hard as possible. Verbal encouragement was used to motivate the subjects to reach their maximums. The time between contractions was 60 s. Three repetitions were performed and the largest force was taken as the peak force contraction. After 3-min rest, the subjects were asked to hold one submaximal contraction at a force level of 150 N (women) and 250 N (men). This submaximal force for the women ranged from 30% to 48%, and for the men from 34% to 53%, of the maximal force output. Visual feedback was used only in the submaximal contractions. The tests were identical in the dry and in the water. The time interval between these two session was 10 min. Before the measurements in the water, the subjects walked about for 3 min in the pool in order to get used to the new conditions. The patient elevator chair was lowered into the pool so that the water surface was at the mid sternum level. EMG measurements The EMG of the vastus medialis, vastus lateralis and biceps femoris muscles were measured by telemeter (Clonner Electronic GmbH, Munich, Germany)) simultaneously with the force measurement. Oval shaped bipolar pregelled silver chloride surface electrodes (Medicotest N-OO-S, Denmark), width 2.1 cm and length 2.9 cm were placed between the distal motor point and the tendon longitudinally along the muscle ®bres. The interelectrode distance was 2.0 cm. To keep the interelectrode resistance low (100 dB, signal-to noise ratio 72 dB. Filtering of the raw EMG was performed with low and high pass ®lters (Butterworth type); the cut-o€ frequencies being low 20 Hz and high 500 Hz. The raw EMG signal was passed through at 12-bit A-D converter with a 1000-Hz sampling frequency and transferred to a computer for further analysis. Background noise in the ®ltered signal was less than 1 lV. The raw EMG signal was full-wave recti®ed and averaged (aEMG) with a window width of 500±1500 or 1000±2000 ms from the starting point of the force curve depending on the peak force. In submaximal contractions a 1±3 s section of the force curve was selected for further analysis. The criteria were that the force should remain steady and that it should not be within the ®rst and last 10% of the contraction time. Statistical analysis Reproducibility for each variable was determined by calculating the intraclass correlation coecient (ICC) and the intrasubject coecient of variation (CVintra). The purpose of ICC was to estimate systematic error and variance which would have been e€ected by systematic changes in the measurements. The two-way ANOVA model for SPSS 6.1 (SPSS Inc.) reliability program was used in this study. There are no universally accepted values for ICC and some authors have suggested that ICC greater than 0.80 represents good reproducibility (Baumgartner 1989). The CVintra was the root mean square error over all trials as a percentage of the mean of the observations and has been shown to estimate the magnitude of pure measurement error (Knutson et al 1994). In addition, standard error of measurement (SEM) was determined using the following equation SEM= SD ´ (1)ICC). Only results of subjects who completed the test on all 3 days were included in the analyses. Normality of the data distribution was tested using the ShapiroWilk's W-test. Accordingly, a paired two-tailed Student's t-test or the Wilcoxon Signed Ranks-test was used to analyse the di€erences between measurement conditions with signi®cance being accepted at the P < 0.05 level.

Results The present data showed high trial-to-trial and dayto-day reproducibility during maximal contractions (Table 1) as well as high reproducibility for submaximal Table 1 Trial-to-trial (three repetitions daily) and day-to-day (best repetition daily) reliability of maximal isometric force and averaged electromyogram activity in three muscles demonstrated by intraclass correlation coecients (ICC), intrasubject coecient of Trial-to-trial reliability

Force VM VL BF

Fig. 2 The relationship between maximal force on dry land and in the water during the 2nd measurement day

contractions (ICC=0.90±0.97, CV=12.7%±17.4%) in each of the test conditions. The force production was found to be similar among three measurement sessions in dry land and in water during maximal contractions. Figure 2 demonstrates the scatter of the plots and the regression line between the force output of the subject group during the second measurement session in the two conditions. A statistically signi®cant di€erence was found between aEMG amplitudes in maximal (P < 0.01±P < 0.001) and submaximal (P < 0.05± P < 0.001) contractions in dry and wet conditions, dry land values demonstrating the highest values in each muscle. Figures 3 and 4 give absolute values (lV) of aEMG activity of the vastus medialis and biceps femoris muscles in maximal and submaximal contractions. The decrease of aEMG amplitudes in water was (11%±17%) in the vastus medialis and vastus lateralis muscles and the decrease was 17%±25% in the biceps femoris muscle.

Discussion The main results of the present experiment indicated that the isometric force and EMG activity of knee variation (CV) and standard error of measurement (SEM), (n = 20). VM Vastus medialis, VL vastus laterialis, BF biceps femoris muscles Day-to-day reliability

Condition

ICC day 1

ICC day 2

ICC day 3

CV (%)

ICC

CV (%)

SEM

Dry land Water Dry land Water Dry land Water Dry land Water

0.99 0.99 0.99 0.99 0.99 0.99 0.95 0.97

0.99 0.99 0.98 0.98 0.99 0.98 0.96 0.97

0.99 0.99 0.98 0.99 0.99 0.99 0.98 0.96

3.5±6.0 3.5±4.8 7.5±10.9 10.4±11.2 7.7±11.1 7.0±9.3 9.0±9.99 8.8±10.5

0.98 0.96 0.95 0.94 0.97 0.97 0.90 0.85

11.9 11.3 11.9 11.3 13.7 15.0 14.3 19.2

16.9 23.1 13.9 16.1 18.8 18.4 2.6 2.4

N N lV lV lV lV lV lV

55

Fig. 3 Averaged electromyograms of vastus medialis (VM) and biceps femoris (BF) muscles in maximal contractions on dry land and in the water for 3 measurement days (MAXVMD, MAXVMW, MAXBFD, MAXBFW, respectively).* P < 0.05, ** P < 0.01, *** P < 0.001

Fig. 4 Averaged electromyogram values of vastus medialis (VM) and biceps femoris (BF) in submaximal contractions on dry land and in the water for 3 measurement days (SUBVMD, SUBVMW, SUBBFD, SUBBFW, respectively). * P < 0.05, ** P < 0.01, *** P < 0.001

extensor muscles showed high trial-to-trial and dayto-day reproducibility. Water and dry land conditions were found to be similar in reproducibility coecients. The major ®nding was that EMG amplitudes measured in the water decreased when compared with the dry land measurements. However, the force output showed no di€erences between the wet and dry conditions. According to published literature, the reproducibility coecients of isometric knee extension force have seemed to be good or excellent (e.g. Viitasalo et al. 1980; Bemben et al. 1992; Christ et al. 1994). In numerous studies, the reliability coecients of EMG measurements have been determined for various types of muscles and contractions (e.g. Komi and Buskirk 1970; Yang

and Winter 1983; Gollhofer et al. 1990). In EMG reproducibility studies of isometric knee extension, Viitasalo and Komi (1975) have reported within-day values of rectus femoris muscle ranging from 0.77 to 0.92 and day-to-day reproducibility ranging from 0.34 to 0.88. Viitasalo et al. (1980) have reported high withinday reproducibility of integrated (iEMG) measurements (ICC=0.98, CV=6.9%) for isometric contractions of quadriceps muscle. According to Heinonen et al. (1994) the day-to-day reproducibility (ICC) of knee extensors of sedentary women has ranged from 0.30 to 0.70 and CV from 4.5% to 14%. Sleivert and Wenger (1994) have reported day-to-day ICC values of knee extensor iEMG ranging from 0.80 to 0.86. Howard and Enoka (1991) have found that the trial-to-trial variability of the EMG of vastus lateralis muscle between three maximal knee extensions is considerable, which may re¯ect intertrial variations in neutral drive, whereas the force output remained constant. The results of the present study were similar: the trial-to-trial CV ranged from 3.5% to 6.0% in maximal force output, while CV in EMG measurements ranged from 7.5% to 11.5%. However, comparisons among di€erent reproducibility studies are dicult to interpret because of di€erences in statistical methods. The ICC addresses agreement in evaluating reproducibility because it is a€ected by systematic changes in the measurements. Intrasubject CV demonstrates more the precision of the measurements instead of reproducibility and, therefore, it has been reported that CV can be used for reference to other studies in this ®eld. (Knutson et al. 1994). The EMG showed signi®cantly decreased amplitudes when measured in the 3-day period during maximal isometric contractions in vastus muscles and biceps femoris muscles in water (11%±17% and 17%±25%, respectively). The amplitudes decreased less in submaximal contractions of each muscle (9%±17%). The force output remained constant in maximal knee extensions. It is assumed that there could be two explanations for the decreased amplitude of the EMG in the water. The ®rst could be the electromechanical factors caused by the e€ect of water on the electrode attachment, wires or detected signal. Clarys et al. (1985) have stated that despite varying the tapings and plastic protection on the electrodes, the detectable electrical output of human muscle was decreased in water. The ®ndings of the present study were in agreement with those of Clarys et al. (1985), who, however, have not measured muscle force or tried to explain their ®ndings. In this study the force output was recorded simultaneously with the EMG measurements. It was assumed in the present study that water had no disturbing e€ects on the attachments of the electrodes as shown by the before and after measurements of interelectrode resistance and the high underwater coecients of reproducibility. The resistance values decreased after the measurement sessions. Theoretically, it can be assumed that volume conductivity in the tissues is higher in water and the voltage di€erence between the two electrodes becomes

56

smaller, which is also a€ected by the reduced interelectrode resistance values. Sweat on the skin under the electrodes is a possible source of artefact when measuring EMG from a subject in warm water. Because perspiration is about 99% water, the EMG amplitude might be decreased as the amount of perspiration increases. Using the same experiment design in a pilot study, the skin temperatures (forehead, upper arm, lower arm, chest, back, thigh, skin and calf) of two subjects were recorded on dry land and in water (27°C) during a 9-min period (Yellow Springs Instrument, YSI 400 series). The mean of the eight local skin temperatures showed values lower by 1.5° in the water. Therefore, it is suggested that the temperature di€erence between two conditions has no e€ect on the EMG signal. A second explanation could be neurophysiological. The e€ects of weightlessness on the neuromuscular system, especially on the muscle spindles and proprioceptive systems during maximal or submaximal voluntary contractions are relatively unknown. Dietz et al. (1989) have found that the activation of pressure receptors controls re¯ex and proprioceptive mechanisms and Avela et al. (1994) have reported that during unexpected gravity conditions muscle spindle activity might be reduced. The measurement of re¯exes and the use of indwelling electrodes would be appropriate in order to evaluate a single motor unit in water in future studies. In conclusion, the high trial-to-trial and day-to-day reproducibility and similarity of CV and SEM values between dry and wet conditions demonstrated the usefulness of underwater force and EMG measurements in rehabilitation and in swimming. The decrease of EMG amplitude in the water was marked in maximal and submaximal contractions, while the force output showed similar values. Whether this signi®cant decrease in amplitude was caused by electromechanical factors in the water and/or neurophysiological phenomena needs further study. Acknowledgement This study was supported by the Finnish Academy and Ministry of Education in the form of a TULES Graduate School Scholarship.

References Alley LE (1952) An analysis of water resistance and propulsion in swimming the crawl stroke. Res 23:253±270 Avela J, Santos PM, KyroÈlaÈinen H, Komi PV (1994) E€ects of di€erent simulated gravity conditions on neuromuscular control in drop jump exercises. Aviat Space Environ Med April, 65:301±308

Baumgartner TA (1989) Norm-referenced measurement: reliability. In: Safrit MJ, Wood TM (eds) Measurement concepts in physical education and exercise science. Human Kinetics, Champaign, III., pp 62±65 Bemben MG, Massey BH, Boileau RA (1992) Reliability of isometric force-time curve parameters for men aged 20 to 79 years. J Appl Sport Sci Res 6:158±164 Berger MAM, de Groot G, Hollander AP (1995) Hydrodynamic and lift forces on human hand models. J. Biomechanics 28:125± 133 Christ CB, Slaugter MH, Stillman RJ (1994) Reliability of selected parametric of isometric muscle function associated with testing 3 days ´ 3 trials in women. J Strength Condit Res 8:65±71 Clarys JP, Robeaux R, Delbeke G (1985) Telemetrical versus conventional EMG in air and water. In: Winter D, Norman R, Hayes R, Patla A (eds) Biomechanics, vol. IX. Human Kinetics, Champaign, III. pp 286±294 Dietz V, Horstmann GA, Trippel M, Gollhofer A (1989) Human postural re¯exes and gravity ± an under water simulation. Neurosci Lett 108:350±355 Gollhofer A, Horstman GA, Schmidbleicher D, SchoÈntahl D (1990) Reproductivity of electromyographic patterns in stretchshorterning type contractions. Eur J Appl Physiol 60:7±14 Heinonen A, SievaÈnen H, Viitasalo J, Pasanen M, Oja P, Vuori I (1994) Reproducibility of computer measurement of maximal isometric force and electromyography in sedentary middle-aged women. Eur J Appl Physiol 68:310±314 Howard JD, Enoka RM (1991) Maximum bilateral contractions are modi®ed by neutrally mediated interlimb e€ects. J Appl Physiol 70:306±316 Ikai M, Ishii K, Miyashita M (1964) An electromyographic study of swimming. Jap Res J Phys Educ 7:55±87 Karpovich PV (1935) Analysis of the propelling forces in the crawl stroke. Res Q 6:49±56 Keskinen KL, Tilli LJ, Komi PV (1989) Maximum velocity swimming. Interrelationships of stroking characteristics, force production and antropometric variables. Scand J Sports Sci 11:87±92 Knutson LM, Soderberg GL, Ballantyne BT, Clarke WR (1994) A study of various normalization procedures for within day electromyographic data. J Electromyogr Kinesiol 4:47±59 Komi PV, Buskirk ER (1970) Reproducibility of electromyographic measurements with inserted wire electrodes and surface electrodes. Electromyography 4:357±367 Lewillie L (1968) Telemetrical analysis of the electromyograph. In: Wartenweiler J, Jokl E, Hebbclinck M (eds) Biomechanics I. Karger Verlag: Basel, pp 147±148 Okamoto T, Wolf SL (1980) Underwater recording of electromyographic activity using ®ne-wire electrodes. In:Terauds J, Bedingfeld EW (eds) Swimming III. University Press, Baltimore, pp 160±166 Sleivert GG, Wenger HA (1994) Reliability of measuring isometric and isokinetic peak torque, rate of torque development, integrated electromyography, and tibial nerve conduction velocity. Arch Phys Med Rehabil 75:1315±1321 Viitasalo JT, Komi PV (1975) Signal characteristics of EMG with special reference to reproducibility of measurements. Acta Physiol Scand 93:531±539 Viitasalo JT, Saukkonen S, Komi PV (1980) Reproducibility of measurements of selected neuromuscular performance variables in man. Electromyogr Clin Neurophysiol 20:487±501 Yang JF, Winter D (1983) Electromyography reliability in maximal and submaximal isometric contractions. Arch Phys Med Rehabil 64:417±420