Older adults show higher increases in lower-limb muscle activity during whole-body vibration exercise

Author’s Accepted Manuscript Older adults show higher increases in lower-limb muscle activity during whole-body vibration exercise Karin Lienhard, Jordyn Vienneau, Sandro Nigg, Bernd Friesenbichler, Benno M. Nigg www.elsevier.com/locate/jbiomech

PII: DOI: Reference:

S0021-9290(16)31275-1 http://dx.doi.org/10.1016/j.jbiomech.2016.12.009 BM8040

To appear in: Journal of Biomechanics Accepted date: 9 December 2016 Cite this article as: Karin Lienhard, Jordyn Vienneau, Sandro Nigg, Bernd Friesenbichler and Benno M. Nigg, Older adults show higher increases in lowerlimb muscle activity during whole-body vibration exercise, Journal of Biomechanics, http://dx.doi.org/10.1016/j.jbiomech.2016.12.009 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ORIGINAL ARTICLE

Older adults show higher increases in lower-limb muscle activity during wholebody vibration exercise.

Karin Lienharda,b,c,*, Jordyn Vienneaua, Sandro Nigga, Bernd Friesenbichlera,d, Benno M. Nigga

a

Human Performance Laboratory, Faculty of Kinesiology, University of Calgary, Calgary, Alberta, Canada

b

University of Nice Sophia Antipolis, I3S, UMR7271, Sophia Antipolis, France

c

University of Nice Sophia Antipolis, LAMHESS, EA 6312, Nice, France; University of Toulon, LAMHESS, EA

6312, La Garde, France d

Human Performance Lab, Schulthess Clinic, Zürich, Switzerland

Keywords: Surface electromyography Aging Frequency Amplitude Acceleration threshold

1

Abstract The purpose of this study was to compare lower limb muscle activity during whole-body vibration (WBV) exercise between a young and an older study population. Thirty young (25.9 ± 4.3 yrs) and thirty older (64.2 ± 5.3 yrs) individuals stood on a side-alternating WBV platform while surface electromyography (sEMG) was measured for the tibialis anterior (TA), gastrocnemius medialis (GM), soleus (SOL), vastus lateralis (VL), vastus medialis (VM), and biceps femoris (BF). The WBV protocol included nine vibration settings consisting of three frequencies (6, 11, 16 Hz) x three amplitudes (0.9, 2.5, 4.0 mm), and three control trials without vibration (narrow, medium, wide stance). The vertical platform acceleration (peak values of maximal displacement from equilibrium) was quantified during each vibration exercise using an accelerometer. The outcomes of this study showed that WBV significantly increased muscle activity in both groups for most vibration conditions in the TA (averaged absolute increase: young: +3.9%, older: +18.4%), GM (young: +4.1%, older: +9.5%), VL (young: +6.3%, older: +12.6%) and VM (young: +5.4%, older: +8.0%), and for the high frequency-amplitude combinations in the SOL (young: +7.5%, older: +12.6%) and BF (young: +1.9%, older: +7.5%). The increases in sEMG activity were significantly higher in the older than the young adults for all muscles, i.e., TA (absolute difference: 13.8%, P < 0.001), GM (4.6%, P = 0.034), VL (7.6%, P = 0.001), VM (6.7%, P = 0.042), BF (6.4%, P < 0.001), except for the SOL (0.3%, P = 0.248). Finally, the vertical platform acceleration was a significant predictor of the averaged lower limb muscle activity in the young (r = 0.917, P < 0.001) and older adults (r = 0.931, P < 0.001). In conclusion, the older population showed greater increases in lower limb muscle activity during WBV exercise than their young counterparts, meaning that they might benefit more from WBV

2

exercises. Additionally, training intensity can be increased by increasing the vertical acceleration load.

Introduction Aging can lead to a decline in muscle mass and neuromuscular activity (Doherty, 2003), which can have negative effects on physical function in an elderly population, including reductions in strength (Macaluso and De Vito, 2004), balance, gait performance, and increased risk of falling (Dutta, 1997). Whole-body vibration (WBV) exercise has been recently introduced as an alternative strength training intervention, with reports of lower limb strength and power gains in young (Delecluse et al., 2003; Petit et al., 2010), as well as in older (Machado et al., 2010; Roelants et al., 2004) individuals. Strength gains following WBV training interventions may be related to increased neuromuscular activity during the exposure to vibration. While numerous studies have observed an increase in surface electromyography (sEMG) recordings during WBV in young adults (Lienhard et al., 2014; Ritzmann et al., 2013), sparse research has been conducted on older individuals (Marin et al., 2012).

The mechanism leading to increased sEMG activity during WBV may be multifaceted, however the main contributor is believed to result from stretch reflex responses, i.e. the tonic vibration reflex (TVR), induced by activation of alpha-motorneurons (Burke et al., 1976; Hagbarth and Eklund, 1966). The response to vibration exercise may be affected by age, as the inhibition of stretch reflexes (Burke et al., 1996) and the H-reflex (Butchart et al., 1993) are less pronounced in the elderly. Reduced inhibition of reflex activity is expected to result in higher increases in

3

sEMG activity during WBV in the older adults. However, this effect has only been investigated in one previous study (Carlucci et al., 2015), which indeed reported greater increases in muscle activity of the quadriceps muscles in the older (74.2 ± 6.0 yrs) compared to the young group (21.6 ± 2.4 yrs). Considering the vast use of WBV exercise for training and rehabilitation purposes in the elderly (Cardinale and Bosco, 2003), it is of upmost importance to research their neuromuscular levels during WBV, and to confirm significant increases in sEMG activity (Marin et al., 2012) and potential differences to their young counterparts (Carlucci et al., 2015).

Optimization of WBV training is often achieved by manipulating the vibration frequency and amplitude (displacement from baseline to peak). Increasing the frequency and/or amplitude has been associated with increased lower limb muscle activity in young (Lienhard et al., 2014; Pollock et al., 2010; Ritzmann et al., 2013) and older adults (Marín et al., 2012; Perchthaler et al., 2013). Thus, maximized sEMG activity has often been reported for the highest frequencyamplitude combination used. In the last years it has been suggested (Marín et al., 2012) that the vertical acceleration load may directly affect sEMG activity levels since the platform frequency (Freq) and platform amplitude (Amp) determine the sinusoidal platform acceleration (Acc) via the formula Acc = (2π*Freq)2 * Amp. Indeed, several studies have found a strong correlation between the vertical platform acceleration and the lower limb sEMG activity (Lienhard et al., 2015c, 2014; Marín et al., 2012). However, the studies using a young study population (Lienhard et al., 2015c) determined the mean amplitude of the vertical platform acceleration as a significant predictor of the lower limb muscle activity, whereas the study investigating older adults (Marín et al., 2012) lacked to identify the root mean square (RMS) of the vertical platform acceleration

4

as a significant predictor. Thus, it remains to be determined to what extent the vertical platform acceleration affects the muscle activity in older adults.

Therefore, the goal of this study was to investigate the neuromuscular activity during WBV in a young and older study population. More specifically, this study aimed to quantify the increases in sEMG activity during WBV, and to compare them between a young and an older group. Lastly, it was aimed to determine the effect of the platform acceleration on sEMG activity levels in both groups. It was hypothesized that (1) both the young and older adults would show significant increases in sEMG activity during WBV exercise, but (2) that those increases would be higher in the older group. Further it was hypothesized that (3) there would be a positive relationship between the vertical platform acceleration and the sEMG activity in both the young and older groups.

Methods Subjects Sixty adults participated in this study, consisting of 30 young adults aged between 20 and 35 years (15 males, 15 females) and 30 older adults aged between 55 and 80 years (15 males, 15 females, Table 1). The exclusion criteria for both groups included frequent use of WBV platforms, a recent lower limb injury, lung conditions, heart problems, high blood pressure, epilepsy, osteoporosis, and diabetes. The study was approved by the University of Calgary’s Conjoint Health Research Ethics Board. Prior to any measurements being acquired, written informed consent was obtained from all the participants.

5

Surface Electromyography Recordings sEMG signals were measured of selected lower limb muscles during relaxed standing with and without WBV, and during isometric maximal voluntary contractions (MVCs). Muscle activity was measured according to the SENIAM guidelines (Hermens et al., 2000) from the muscle bellies of the tibialis anterior (TA), gastrocnemius medialis (GM), soleus (SOL), vastus lateralis (VL), vastus medialis (VM) and biceps femoris (BF) of the right lower limb using bipolar surface electrodes (Ag-AgCl) with preamplifiers (Biovision, Wehrheim, Germany). The electrodes had a 10mm diameter and 22mm inter-electrode spacing. The electrodes were secured to the skin with Cover-Roll stretch tape (Beiersdorf AG, Hamburg, Germany) in order to prevent movement of the electrodes. A reference electrode was placed on the right tibial tuberosity. Prior to the placement of the electrodes, the subject’s right leg was prepared for sEMG measurements, i.e. the skin was shaved, dead skin cells were removed using abrasive tape, and the area was cleaned with an isopropyl wipe. Myoelectric signals were pre-amplified at the source (gain = 1000) and sampled at 2400 Hz.

MVC Assessment After all subjects had completed a 5-min warm-up on a stationary bike at low resistance, isometric MVC trials of the right lower limb were recorded for dorsiflexion, plantarflexion, knee extension and knee flexion using an isokinetic dynamometer (Biodex System 3, Medical Systems Inc., New York, NY, USA). The ankle contractions were performed with the ankle joint at 90° (neutral position) and the knee contractions were performed with the knee joint at 60° (0° corresponding to full extension) (Thorstensson et al., 1976); the rotational axis of the dynamometer was visually aligned to the lateral malleolus and the femoral condyle, respectively.

6

sEMG activity was recorded from the TA during dorsiflexion, from the GM and SOL during plantarflexion, from the VL and VM during knee extension and from the BF during knee flexion. All subjects performed two 5-s MVC trials in which they were instructed to produce as much force as possible and to maintain this force at a constant level. After each trial, the subjects were given a 1-min break. Prior to each contraction, the subjects were given two practice trials in order to become familiarized with the respective contraction. Peak MVC torques over 500ms windows were calculated for both trials, whereas the trial with the higher torque was retained for further analysis. The root mean square of the sEMG data (sEMGRMS) was calculated for the 500ms time period of the selected torque output from the muscle corresponding to the contraction.

(Table 1 about here)

WBV Protocol The vibration platform used in this study (TBS 100A, Total Image Fitness, Calgary, Alberta, Canada) induces sinusoidal side-alternating vibrations by rotating along the sagittal axis. The frequency setting of this platform ranges from 6 to 16 Hz, and frequencies of 6 Hz, 11 Hz, and 16 Hz were investigated. For side-alternating vibration platforms, the amplitude is dictated by the foot position, where the further from the centre line, the larger the amplitude. For the current model, the amplitude (defined as the maximal displacement from the equilibrium position) ranged from 0mm to 4mm. Three amplitude settings were used in this study: 0.9mm (corresponding to a foot position of 5.3cm from the central axis of the vibration platform), 2.5mm (corresponding to 14.3cm from the central axis of the vibration platform), and 4mm

7

(corresponding to 23.3cm from the central axis of the vibration platform). This resulted in a total of nine WBV trials (3 frequencies x 3 amplitudes). In addition, three trials without vibration were recorded in order to determine the baseline muscle activity required for quiet standing during the three stances that corresponded to the 0.9, 2.5 and 4 mm amplitudes. After familiarizing the subjects with WBV, they performed the WBV protocol with each trial consisting of 20-s quiet standing and a 1-min break in between trials. The subjects were instructed not to hold on to the handle bar of the platform and to stand with their knees bent to 20° flexion (0° corresponding to full extension). Their knee angle was monitored using a digital goniometer (SG150 Twin Axis, Biometrics Ltd, Newport, UK) secured to their left knee joint. During the trial, the subjects were provided with a visual real-time feedback of their knee joint angle using MATLAB software (version 7.13 The Mathworks, Inc., Natick, MA, USA). If the measured knee angle varied more than ± 3°, the trial was repeated. Additionally, a 3D accelerometer (ADXL 78, Analog Devices USA, encapsulated in a 20/12/5 mm plastic shell, measuring range ±70 g, frequency response of 0-400 Hz, mass: 2 g, sampled at 2400Hz) was placed on the WBV platform aligned with the third toe of the right foot. In order to calculate the averaged peak acceleration, the mean of the local maxima and minima of the vertical platform acceleration signal was computed of each acceleration signal. The averaged peak values for each WBV trial are presented in Table 2. sEMG analysis and normalization to the MVC are described in Appendix I and II, respectively.

(Table 2 about here)

8

Statistical Analysis All statistical procedures were performed using IBM SPSS software (Version 20, Chicago, IL). The Kolmogorov-Smirnov test was used to check the normality of the increases in sEMGRMS(%MVC) data for the young and older groups, which proofed normal distribution (P > 0.05) for both groups and all trials. Unpaired Student’s t-tests were computed to compare age, height, body mass, body mass index (BMI), and the MVC torques between the two groups. Paired Student’s t-tests were used to compare sEMGRMS(%MVC)

during WBV to the

sEMGRMS(%MVC) during the no-vibration trial with the corresponding stance. This means that the three tested vibration frequencies (6 Hz, 11 Hz, 16 Hz) at a given amplitude were always compared to the no vibration trial with the same stance. The significance level was adjusted for multiple comparisons, which resulted in a new ɑ = 0.017. To compare the baseline sEMGRMS(%MVC) during the no-vibration trial between the two groups, one-way repeated measures analysis of variance (ANOVA) with group (young vs. older) as between-subjects factor was performed. In order to compare the increase in sEMGRMS(%MVC) between the two groups a one-way repeated measures ANOVA with group (young vs. old) as between-subjects factor was performed. Pearson’s correlation coefficients were computed between the peak values of vertical platform acceleration (maximal displacement from equilibrium) and sEMGRMS(%MVC) of each muscle as well as the averaged lower limb sEMGRMS(%MVC). Additionally, a linear regression analysis was performed between the vertical platform acceleration and the averaged lower limb sEMGRMS(%MVC). The level of significance was set at ɑ = 0.05, and the results are presented as mean ± standard deviation (SD).

9

Results Subjects’ Characteristics Age, body mass and BMI were significantly higher in the older than the young group (P < 0.01). The maximal torque normalized to the body mass (Nm/kg) was significantly greater in the young than the older participants for all contractions, i.e. dorsiflexion, plantarflexion, knee extension, and knee flexion (P < 0.001, Table 1).

sEMG between Groups The sEMGRMS(%MVC) between the groups during the no-vibration trials using the one-way repeated measures ANOVA showed significant main effects, with significantly higher baseline sEMG activity in the older than the young participants for all the muscles, i.e., TA (averaged absolute difference: 5.3%, P < 0.001), GM (9.1%, P < 0.001), SOL (10.7%, P = 0.001), VL (8.1%, P < 0.001), VM (2.1%, P = 0.010), and BF (1.7%, P = 0.033). Paired Student’s t-test showed that exposure to WBV significantly increased muscle activity in both groups for most WBV conditions in the TA (averaged absolute increase: young: +3.9%, older: +18.4%), GM (young: +4.1%, older: 9.5%), VL (young: +6.3%, older: 12.6%) and VM (young: +5.4%, older: 8.0%), and for the high frequency-amplitude combinations in the SOL (young: +7.5%, older: 12.6%) and BF (young: +1.9%, older: 7.5%, Figure 1). ANOVA analysis on the absolute increase in muscle activity (sEMGRMS(%MVC) during WBV - sEMGRMS(%MVC) during novibration) revealed that the older group showed significantly higher sEMGRMS(%MVC) increases than the young adults in all muscles, i.e., TA (averaged absolute difference: 13.8%, P < 0.001), GM (4.6%, P = 0.034), VL (7.6%, P = 0.001), VM (6.7%, P = 0.042), BF (6.4%, P < 0.001), with the exception of the SOL (0.3%, P = 0.248).

10

(Figure 1 about here)

Relation between Platform Acceleration and sEMG The Pearson’s correlation coefficient between the sEMGRMS(%MVC) and the vertical platform acceleration was high and significant for all the measured muscles in the young (TA: r = 0.695, P = 0.038, GM: r = 0.809, P = 0.008, SOL: r = 0.824, P = 0.006, VL: r = 0.882, P = 0.002, VM: r = 0.784, P = 0.012, BF: r = 0.951, P < 0.001) and the older adults (TA: r = 0.886, P = 0.007, GM: r = 0.944, P < 0.001, SOL: r = 0.944, P < 0.001, VL: r = 0.796, P = 0.010, VM: r = 0.804, P = 0.009, BF: r = 0.952, P < 0.001). Averaging the sEMGRMS(%MVC) of all the muscles resulted in high and significant correlations as well, with r = 0.917 (P < 0.001) in the young and r = 0.931 (P < 0.001) in the older group (Figure 2). The linear regression analysis on the averaged lower limb sEMGRMS(%MVC) showed that the vertical platform acceleration (Accv) was a significant predictor (P < 0.001) of the sEMGRMS(%MVC) both in the young and the older group. The equation (Figure 2) for the young adults was sEMGRMS(%MVC) = 7.933 + (0.148 x Accv) and the equation for the older adults was sEMGRMS(%MVC) = 18.142 + (0.279 x Accv).

(Figure 2 about here)

Discussion The outcomes of this study were that (1) the older group showed significantly higher sEMG baseline activity during the no-vibration trial than the young group for all the measured lower limb muscles. With the use of WBV, (2) sEMG activity significantly increased in both the young and older groups for most vibration conditions in the TA, GM, VL and VM, and for the high 11

frequency-amplitude combinations in the SOL and BF. However, (3) the increases in sEMG activity due to WBV were significantly higher in the older than the young adults for all muscles except for the SOL. Finally, (4) the vertical platform acceleration was a significant predictor of the lower limb sEMG activity in the young and older adults.

It is not surprising that the baseline sEMG was higher in the older adults considering their substantial deficit in lower limb muscle strength as compared to their young counterparts. Their significantly lower MVC torques for dorsiflexion, plantarflexion, knee extension, and knee flexion required them to activate their lower limb muscles to a greater proportion of their maximum in order to maintain the standing position. The probability that the older group was unable to produce a true MVC following an underestimation of their maximum sEMG values is rather low. It has been shown for numerous types of contractions that, after a familiarization period, older adults are equally able to maximally contract their muscles in order to produce a MVC as are young adults (Klass et al., 2007; Power et al., 2013). Studies have found that there were no significant differences in the percentage of a stimulated maximum value that the two age groups were able to voluntarily produce during dorsiflexion (Connelly et al., 1999; Simoneau et al., 2005), plantarflexion (Simoneau et al., 2005), and knee extension (Roos et al., 1999). Therefore it can be concluded that the differences in baseline sEMG activity between the two groups were not attributed to the normalization method, but rather to the different strength levels between the two groups.

Significant increases in muscle activity were found in both the young and the older groups for relatively low acceleration inputs, providing support for the first hypothesis that both the young

12

and older adults would show significant increases in sEMG activity during WBV exercise. It should be considered, however, that the spread of the data was greater in the older group, and that a higher number of outliers were observed in this group, which is a limitation of the present study. However, in the young group, increases in sEMG activity were significant for vertical acceleration loads of at least 1.6 m.s-2 in the TA, 3.9 m.s-2 in the GM, 5.7 m.s-2 in the VM and VL, 14.0 m.s-2 in the SOL, and 20.4 m.s-2 in the BF. Comparable or lower accelerations were required in order to induce significant sEMG increases in the older group, namely with 1.6 m.s-2 in the TA, VM and VL, and 3.9 m.s-2 in the GM and BF. The SOL differed from the other muscles in that higher accelerations were necessary to significantly increase the muscle activity in the older group (20.4 m.s-2). This observation is in line with another finding of this study, confirming the second hypothesis; significantly greater sEMG increases during WBV were found in the older than the young group for a given acceleration load in all the measured muscles except for the SOL.

Comparison of muscle activity during WBV between an older and a young population has, to the best of our knowledge, only been compared in one previous study (Carlucci et al., 2015). Similar to the present study, sEMG activity during WBV was found to be higher in the older than the young group. While the older individuals showed higher muscle activity levels ranging from 17 to 24% in the study by Carlucci et al. (2015), lower absolute differences were found in the present study (4 – 16%). This inconsistency could be attributed to several parameters that were dissimilar between the two studies, such as the used vibration platform type (vertical vs. sidealternating vibration), the acceleration load (~31 - 57 m.s-2 vs. 1.6 – 49.3 m.s-2), and the sEMG filtering regime (no-filter vs. linear interpolation).

13

It is well known that when vibrations are directly applied to the muscle or the tendon, preactivation of the muscles increases the sensitivity of the spindles with concurrent increases in responses similar to those found in the TVR (Burke et al., 1976; Eklund and Hagbarth, 1966). Recently, this observation has been confirmed for WBV stimuli (Zaidell et al., 2013). Since the baseline sEMG was higher in the older individuals, their muscles were pre-activated to a higher extent. This may have increased their muscle spindle sensitivity which could explain their higher neuromuscular responses during the vibration than in their young counterparts.

Another mechanism that could be responsible for this outcome is muscular damping. When resonance occurs, increased muscle activity is expected to reduce resonance in order to reduce soft tissue vibrations (Wakeling et al., 2002). The natural frequencies of the lower limb soft tissues (11 – 14 Hz) are within the range of the tested vibration frequencies (11 – 16 Hz), which supports the occurrence of muscular damping. Interestingly, the SOL was the only muscle that did not depict significantly greater sEMG increases during WBV in the older group. It has been shown that exposure to WBV leads to reduced recruitment threshold of high-threshold (type II) motor units (Pollock et al., 2012). Furthermore, it has been shown that with aging, motor unit recruitment thresholds tend to decline (Erim et al., 1999). Considering that the muscles that showed higher sEMG increases in the older group have a relatively high proportion of type II fibers (37 to 73%) (Johnson et al., 1973), these two findings support the higher increases in muscle activity that were found in most lower limb muscles in the older group during WBV. In contrast, the SOL is predominantly (~86%) composed of type I muscle fibers (Johnson et al., 1973). Thus, the SOL presents a lower proportion of high-threshold motor units than the other

14

measured lower limb muscles, which may explain why the older group did not exhibit higher sEMG increases during WBV than the young group in the SOL.

In line with previous observations (Lienhard et al., 2015c, 2014; Marín et al., 2012), the vertical platform acceleration was positively related to the muscle activity during WBV in the young and older individuals, which confirms the third hypothesis. Furthermore, the vertical platform acceleration was found to be a significant predictor of the sEMG activity, which can explain why previous studies found the highest increases in muscle activity using the highest frequencyamplitude combinations (Lienhard et al., 2014; Marín et al., 2012; Perchthaler et al., 2013; Pollock et al., 2010; Ritzmann et al., 2013). Exposure to WBV may be related to bodily hazards (Rittweger, 2010) and maximal acceleration limits have yet to be determined to prevent injury specific to WBV training, since available standards (ISO 2631-1:1997) are not applicable (Friesenbichler et al., 2014; Pel et al., 2009; Rubin et al., 2003). The guidelines developed by the International Organization for Standardization (ISO 2631-1:1997) address vibrations transmitted through the rear while sitting (for example among bus drivers), rather than through the feet. At this point in time, the health implications of absorbing vibrations while standing are not known, so the ISO guidelines cannot be applied to WBV. Therefore, it has been suggested (Lienhard et al., 2015c) that an acceleration threshold be defined, i.e. the minimal acceleration load that is required to significantly increase lower limb muscle activity during WBV, in order to minimize potential detrimental effects. The outcomes of this study showed that the minimal acceleration thresholds to increase lower limb muscle activity varied quite substantially between muscles in the young (TA: 1.6 m.s-2; GM, VL, VM: 5.7 m.s-2; SOL: 14.0 m.s-2; BF: 20.4 m.s-2) and older adults (TA, VM, VL: 1.6 m.s-2; GM, BF: 3.9 m.s-2; SOL: 20.4 m.s-2). However, averaged across

15

all muscles, the young group showed a higher acceleration threshold (8.6 m.s-2) than the older group (5.5 m.s-2). Thus it can be concluded that lower acceleration inputs are required for older individuals to achieve neuromuscular changes to WBV exercise.

Conclusion This study found that older adults might benefit more than young adults from WBV exercise, as the older group showed greater increases in lower limb muscle activity during WBV than the young population. Older individuals are advised to exercise on a WBV platform using vertical platform accelerations of 5.5 m.s-2 and higher, whereas young individuals required at least 8.6 m.s-2 to stimulate their lower limb muscles. Training intensity can be increased by increasing the vertical acceleration load, as the vertical platform acceleration was a significant predictor of the lower-body sEMG activity in both the young and older individuals.

Acknowledgements The vibration platform was provided by Total Image Fitness, Calgary, Alberta, Canada. The study was financially supported by the Natural Sciences and Engineering Research Council of Canada (NSERC) and Total Image Fitness, Calgary, Alberta, Canada. Financial support for travel related expenses was obtained from the Fondation Partenariale DreamIT, France.

Conflict of interest The authors declare that they have no conflict of interest.

16

References Burke, D., Hagbarth, K.E., Löfstedt, L., Wallin, B.G., 1976. The responses of human muscle spindle endings to vibration during isometric contraction. J. Physiol. 261, 695–711. Burke, J.R., Schutten, M.C., Koceja, D.M., Kamen, G., 1996. Age-dependent effects of muscle vibration and the Jendrassik maneuver on the patellar tendon reflex response. Arch. Phys. Med. Rehabil. 77, 600–604. Butchart, P., Farquhar, R., Part, N.J., Roberts, R.C., 1993. The effect of age and voluntary contraction on presynaptic inhibition of soleus muscle Ia afferent terminals in man. Exp. Physiol. 78, 235–242. Cardinale, M., Bosco, C., 2003. The use of vibration as an exercise intervention. Exerc. Sport Sci. Rev. 31, 3–7. Carlucci, F., Orlando, G., Haxhi, J., Laudani, L., Giombini, A., Macaluso, A., Pigozzi, F., Sacchetti, M., 2015. Older Age Is Associated with Lower Optimal Vibration Frequency in Lower-Limb Muscles During Whole-Body Vibration. Am. J. Phys. Med. Rehabil. Assoc. Acad. Physiatr. 94, 522–529. Connelly, D.M., Rice, C.L., Roos, M.R., Vandervoort, A.A., 1999. Motor unit firing rates and contractile properties in tibialis anterior of young and old men. J. Appl. Physiol. 87, 843– 52. Delecluse, C., Roelants, M., Verschueren, S., 2003. Strength increase after whole-body vibration compared with resistance training. Med. Sci. Sports Exerc. 35, 1033–1041. Doherty, T.J., 2003. Invited review: Aging and sarcopenia. J. Appl. Physiol. Bethesda Md 1985 95, 1717–1727. Dutta, C., 1997. Significance of sarcopenia in the elderly. J. Nutr. 127, 992S–993S. Eklund, G., Hagbarth, K.E., 1966. Normal variability of tonic vibration reflexes in man. Exp. Neurol. 16, 80–92. Erim, Z., Beg, M.F., Burke, D.T., Luca, C.J. De, Pascoe, M.A., Holmes, M.R., Enoka, R.M., Laine, C.M., Bailey, E.F., Dalton, B.H., Power, G.A., Vandervoort, A.A., Rice, C.L., Luca, C.J.D.E., 1999. Effects of Aging on Motor-Unit Control Properties. J. Neurophysiol. 82, 2081–2091. Friesenbichler, B., Lienhard, K., Vienneau, J., Nigg, B.M., 2014. Vibration transmission to lower extremity soft tissues during whole-body vibration. J. Biomech. 47, 2858–2862. Hagbarth, K.E., Eklund, G., 1966. Tonic vibration reflexes (TVR) in spasticity. Brain Res. 2, 201–203. Hermens, H.J., Freriks, B., Disselhorst-Klug, C., Rau, G., 2000. Development of recommendations for SEMG sensors and sensor placement procedures. J. Electromyogr. Kinesiol. 10, 361–374. International Organization for Standardization (1997) ISO 2631-1:1997. Mechanical Vibration and Shock-Evaluation of Human Exposure to Whole-Body Vibration, Part 1, General Requirements. Johnson, M. a, Polgar, J., Weightman, D., Appleton, D., 1973. Data on the distribution of fibre types in thirty-six human muscles. An autopsy study. J. Neurol. Sci. 18, 111–129. Klass, M., Baudry, S., Duchateau, J., 2007. Voluntary activation during maximal contraction with advancing age: A brief review. Eur. J. Appl. Physiol. 100, 543–51. Lienhard, K., Cabasson, A., Meste, O., Colson, S.S., 2015a. sEMG during Whole-Body Vibration Contains Motion Artifacts and Reflex Activity. J. Sports Sci. Med. 14, 54–61. 17

Lienhard, K., Cabasson, A., Meste, O., Colson, S.S., 2015b. Comparison of sEMG processing methods during whole-body vibration exercise. J. Electromyogr. Kinesiol. Off. J. Int. Soc. Electrophysiol. Kinesiol. Lienhard, K., Cabasson, A., Meste, O., Colson, S.S., 2014. Determination of the optimal parameters maximizing muscle activity of the lower limbs during vertical synchronous whole-body vibration. Eur. J. Appl. Physiol. Lienhard, K., Vienneau, J., Nigg, S., Meste, O., Colson, S.S., Nigg, B.M., 2015c. Relationship Between Lower Limb Muscle Activity and Platform Acceleration During Whole-Body Vibration Exercise. J. Strength Cond. Res. Natl. Strength Cond. Assoc. 29, 2844–2853. Macaluso, A., De Vito, G., 2004. Muscle strength, power and adaptations to resistance training in older people. Eur. J. Appl. Physiol. 91, 450–472. Machado, A., García-López, D., González-Gallego, J., Garatachea, N., 2010. Whole-body vibration training increases muscle strength and mass in older women: a randomizedcontrolled trial. Scand. J. Med. Sci. Sports 20, 200–207. Marín, P.J., Herrero, A.J., García-López, D., Rhea, M.R., López-Chicharro, J., GonzálezGallego, J., Garatachea, N., 2012. Acute effects of whole-body vibration on neuromuscular responses in older individuals: implications for prescription of vibratory stimulation. J. Strength Cond. Res. Natl. Strength Cond. Assoc. 26, 232–239. Marin, P.J., Santos-Lozano, A., Santin-Medeiros, F., Vicente-Rodriguez, G., Casajús, J., Hazell, T., Garatachea, N., 2012. Whole-body vibration increases upper and lower body muscle activity in older adults: Potential use of vibration accessories. J. Electromyogr. Kinesiol. 22, 456–62. Pel, J.J.M., Bagheri, J., van Dam, L.M., van den Berg-Emons, H.J.G., Horemans, H.L.D., Stam, H.J., van der Steen, J., 2009. Platform accelerations of three different whole-body vibration devices and the transmission of vertical vibrations to the lower limbs. Med. Eng. Phys. 31, 937–944. Perchthaler, D., Horstmann, T., Grau, S., 2013. Variations in neuromuscular activity of thigh muscles during whole-body vibration in consideration of different biomechanical variables. J. Sports Sci. Med. 12, 439–446. Petit, P.-D., Pensini, M., Tessaro, J., Desnuelle, C., Legros, P., Colson, S.S., 2010. Optimal whole-body vibration settings for muscle strength and power enhancement in human knee extensors. J. Electromyogr. Kinesiol. Off. J. Int. Soc. Electrophysiol. Kinesiol. 20, 1186–1195. Pollock, R.D., Woledge, R.C., Martin, F.C., Newham, D.J., 2012. Effects of whole body vibration on motor unit recruitment and threshold. J. Appl. Physiol. Bethesda Md 1985 112, 388–395. Pollock, R.D., Woledge, R.C., Mills, K.R., Martin, F.C., Newham, D.J., 2010. Muscle activity and acceleration during whole body vibration: effect of frequency and amplitude. Clin. Biomech. Bristol Avon 25, 840–846. Power, G.A., Dalton, B.H., Rice, C.L., 2013. Human neuromuscular structure and function in old age: A brief review. J. Sport Health Sci. 1–12. Rittweger, J., 2010. Vibration as an exercise modality: how it may work, and what its potential might be. Eur. J. Appl. Physiol. 108, 877–904. Ritzmann, R., Gollhofer, A., Kramer, A., 2013. The influence of vibration type, frequency, body position and additional load on the neuromuscular activity during whole body vibration. Eur. J. Appl. Physiol. 113, 1–11. 18

Ritzmann, R., Kramer, A., Gruber, M., Gollhofer, A., Taube, W., 2010. EMG activity during whole body vibration: motion artifacts or stretch reflexes? Eur. J. Appl. Physiol. 110, 143–151. Roelants, M., Delecluse, C., Verschueren, S.M., 2004. Whole-body-vibration training increases knee-extension strength and speed of movement in older women. J. Am. Geriatr. Soc. 52, 901–908. Roos, M.R., Rice, C.L., Connelly, D.M., Vandervoort, A.A., 1999. Quadriceps muscle strength, contractile properties, and motor unit firing rates in young and old men. Muscle Nerve 22, 1094–103. Rubin, C., Pope, M., Fritton, J.C., Magnusson, M., Hansson, T., McLeod, K., 2003. Transmissibility of 15-hertz to 35-hertz vibrations to the human hip and lumbar spine: determining the physiologic feasibility of delivering low-level anabolic mechanical stimuli to skeletal regions at greatest risk of fracture because of osteoporosis. Spine 28, 2621–2627. Sebik, O., Karacan, I., Cidem, M., Türker, K.S., 2013. Rectification of SEMG as a tool to demonstrate synchronous motor unit activity during vibration. J. Electromyogr. Kinesiol. Off. J. Int. Soc. Electrophysiol. Kinesiol. 23, 275–284. doi:10.1016/j.jelekin.2012.09.009 Simoneau, E., Martin, A., Van Hoecke, J., 2005. Muscular performances at the ankle joint in young and elderly men. J. Gerontol. Biol. Sci. 60, 439–47. Thorstensson, A., Grimby, G., Karlsson, J., 1976. Force-velocity relations and fiber composition in human knee extensor muscles. J. Appl. Physiol. 40, 12–16. von Tscharner, V., Schwameder, H., 2001. Science and skiing II: Filtering of force variables in skiing by specified wavelet analysis. Hamburg. Wakeling, J.M., Nigg, B.M., Rozitis, A.I., 2002. Muscle activity damps the soft tissue resonance that occurs in response to pulsed and continuous vibrations. J. Appl. Physiol. Bethesda Md 1985 93, 1093–1103. Zaidell, L.N., Mileva, K.N., Sumners, D.P., Bowtell, J.L., 2013. Experimental evidence of the tonic vibration reflex during whole-body vibration of the loaded and unloaded leg. PloS One 8, e85247.

19

Figure Captions

Figure 1. Root mean square of the normalized surface electromyography (sEMGRMS(%MVC)) during side-alternating whole-body vibration (WBV) in the young and older adults and for all the assessed conditions, which consisted of three frequencies (6, 11, 16 Hz) combined with three amplitudes (0.9, 2.5, 4.0 mm). The asterisks indicate significant increases in sEMGRMS(%MVC) during WBV (*P < 0.017, **P < 0.01, ***P < 0.001) compared to the corresponding novibration trial (cross shaded).

Figure 2. Scatter plots of the combined lower limb EMGRMS(%MVC) and the vertical platform acceleration (Accv). Each point represents the averaged EMGRMS(%MVC) of all muscles and participants for a given WBV trial. There was a significant correlation between the vertical platform acceleration and the EMGRMS(%MVC) of the young adults (r = 0.917, P < 0.001) and the EMGRMS(%MVC) of the older adults (r = 0.931, P < 0.001).

20

Appendix I

Surface Electromyography Analysis sEMG analysis was performed using MATLAB software. The sEMG data were first clipped so that only seconds 7 to 17 of each trial remained for further analysis. The clipping of the sEMG signal was important to delete potential error signals originating from e.g. the onset of the platform or the reflex activity. Then, the raw sEMG data were filtered using a band-pass wavelet filter with a low cut-off frequency of 5 Hz and high cut-off frequency of 300 Hz (von Tscharner and Schwameder, 2001). As sEMG signals recorded during WBV contain motion artifacts (Lienhard et al., 2015a; Sebik et al., 2013), spectral linear interpolation was applied to all sEMG recordings (Lienhard et al., 2015b). First, the sEMG signals were transformed into the Power Spectral Density (PSD) with the help of the Welch method, using Hamming windows with the length of L = 1024. Then, the excessive peaks were located in the sEMG spectrum at the vibration frequency fP and its multiple harmonics (2*fP, 3*fP, ...) and were removed by connecting the fp adjacent left and right PSD amplitudes. After the linear interpolation of the sEMG signal, the sEMGRMS (average value for the whole time series) was calculated directly within the PSD, as spectral linear interpolation in the PSD does not allow a time-reversal transformation.

21

Appendix II

Surface Electromyography Normalization sEMGRMS during WBV and the no-vibration trials were calculated and expressed as a percentage value of the maximum sEMGRMS of the respective muscle obtained during the MVC trials using the formula

. Thus, the sEMGRMS results

presented in the following are normalized to the MVC and represent the percentage of muscle activity in respect to the sEMGRMS required for a maximal contraction. Additionally, normalized sEMGRMS during the no-vibration trials were subtracted from the corresponding vibration trials using the formula . This resulted in nine values (3 amplitudes x 3 frequencies), which reflected the absolute increase in sEMG activity during WBV relative to the no-vibration trials. This, in turn, enabled comparisons across different platform amplitudes, as the amplitudes resulted from different standing positions, as well as across the two study populations.

22

23

Table 1. Demographic, anthropometric and strength characteristics (torque/mass) of the young and older participants (mean ± SD). Young

Older

(N = 30)

(N = 30)

Age (years)

25.9 ± 4.3

64.2 ± 5.3*

Height (m)

1.71 ± 0.09

1.69 ± 0.12

Mass (kg)

69.5 ± 11.2

80.0 ± 15.7*

BMI (kg/m²)

23.5 ± 2.4

27.8 ± 3.5*

Dorsiflexion (Nm/kg)

0.54 ± 0.08

0.38 ± 0.11†

Plantarflexion (Nm/kg)

1.87 ± 0.49

1.13 ± 0.35†

Knee extension (Nm/kg)

2.52 ± 0.57

1.61 ± 0.43†

Knee flexion (Nm/kg)

1.33 ± 0.32

0.77 ± 0.25†

BMI, Body Mass Index; Anthropometric characteristics: *Young < Older (P < 0.01), Strength characteristics: †Older < Young (P < 0.001).

24

Table 2. Peak values (mean ± SD) of the vertical platform acceleration in m.s-2 for the nine vibration trials consisting of three platform amplitudes (0.9, 2.5, 4.0 mm) combined with three vibration frequencies (6, 11, 16 Hz). Amplitude

0.9 mm

2.5 mm

4.0 mm

Frequency

6 Hz

11 Hz

16 Hz

6 Hz

11 Hz

16 Hz

6 Hz

11 Hz

16 Hz

Peak Acceleration (m.s-2)

1.6 ± 0.3

5.7 ± 0.8

12.6 ± 1.4

3.9 ± 0.2

14.0 ± 0.9

32.1 ± 2.3

5.9 ± 0.2

20.4 ± 1.3

49.3 ± 4.8

25