European Journal of Applied Physiology

, Volume 113, Issue 1, pp 1–11

The influence of vibration type, frequency, body position and additional load on the neuromuscular activity during whole body vibration

Authors

    • Institute of Sport and Sport ScienceUniversity of Freiburg
  • Albert Gollhofer
    • Institute of Sport and Sport ScienceUniversity of Freiburg
  • Andreas Kramer
    • Department of Sports ScienceUniversity of Konstanz
Original Article

DOI: 10.1007/s00421-012-2402-0

Cite this article as:
Ritzmann, R., Gollhofer, A. & Kramer, A. Eur J Appl Physiol (2013) 113: 1. doi:10.1007/s00421-012-2402-0

Abstract

This study aimed to assess the influence of different whole body vibration (WBV) determinants on the electromyographic (EMG) activity during WBV in order to identify those training conditions that cause highest neuromuscular responses and therefore provide optimal training conditions. In a randomized cross-over study, the EMG activity of six leg muscles was analyzed in 18 subjects with respect to the following determinants: (1) vibration type (side-alternating vibration (SV) vs. synchronous vibration (SyV), (2) frequency (5–10–15–20–25–30 Hz), (3) knee flexion angle (10°–30°–60°), (4) stance condition (forefoot vs. normal stance) and (5) load variation (no extra load vs. additional load equal to one-third of the body weight). The results are: (1) neuromuscular activity during SV was enhanced compared to SyV (P < 0.05); (2) a progressive increase in frequency caused a progressive increase in EMG activity (P < 0.05); (3) the EMG activity was highest for the knee extensors when the knee joint was 60° flexed (P < 0.05); (4) for the plantar flexors in the forefoot stance condition (P < 0.05); and (5) additional load caused an increase in neuromuscular activation (P < 0.05). In conclusion, large variations of the EMG activation could be observed across conditions. However, with an appropriate adjustment of specific WBV determinants, high EMG activations and therefore high activation intensities could be achieved in the selected muscles. The combination of high vibration frequencies with additional load on an SV platform led to highest EMG activities. Regarding the body position, a knee flexion of 60° and forefoot stance appear to be beneficial for the knee extensors and the plantar flexors, respectively.

Keywords

ElectromyographyParametersVibrationTrainingExercise

Introduction

Scientific research with respect to whole body vibration (WBV) has increased in the last decade. However, functional and neuromuscular adaptations to WBV are not entirely understood. Since vibration training is a form of training that uses an external drive to stimulate the muscle (Rittweger 2010; Wilcock et al. 2009), this external drive (generated by the vibration device) and its adjustment (i.e., the choice of frequency and amplitude) have a big influence on the muscles’ response to WBV training (Rauch et al. 2010; Rittweger 2010). However, despite the substantial amount of WBV-related articles, there are few that address fundamental questions such as optimal frequency or body position on the training device (Abercromby et al. 2007a; Berschin and Sommer 2004; Hazell et al. 2007, 2010).

Only a few studies focused on the systematic variation of WBV determinants in order to define adequate training conditions. In the latter studies, the electromyographic (EMG) activity was analyzed regarding specific vibration determinants, i.e., frequencies (Cardinale and Lim 2003; Cochrane et al. 2009), amplitudes (Marín et al. 2009), or additional load (Hazell et al. 2010) or vibration types and body positions (Abercromby et al. 2007a). In general, the studies demonstrated that acute WBV is accompanied by higher EMG responses compared to conditions without WBV. More specific, it has been shown that neuromuscular responses increase with increasing amplitudes (Marín et al. 2009) and additional loads (Hazell et al. 2010). The effects of the vibration frequency remains unclear, as studies have failed to produce consistent results so far (Berschin and Sommer 2004; Cardinale and Lim 2003; Hazell et al. 2007). However, at least there is some indirect evidence suggesting that the neuromuscular activation is related to the level of frequency during WBV training (Ritzmann et al. 2010). As the number of vibration-induced stretch reflexes increase with ascending frequency, it might be suggested that the EMG activity is increased as well (Cochrane et al. 2009; Ritzmann et al. 2010). With respect to the body position there is only one study conducted by Abercromby et al. (2007a) focusing on different knee flexion angles during WBV. Their main finding was that WBV in a static squat position caused higher neuromuscular responses compared to several modalities of dynamic squatting.

Each of the vibration determinants mentioned above has its own impact on WBV training and training adaptation (Cochrane 2011). However, until now, there has not been a systematic approach focusing on the combination of the WBV training determinants in an effort to identify conditions, which provide the highest neuromuscular responses during WBV. It is well known that during a voluntary action, the muscles’ EMG activity is related to the extent of the muscle fibers’ recruitment (Aagaard 2003; Hogrel 2003; Milner-Brown et al. 1973a, 1973b; Riley et al. 2008) and frequency (Aagaard 2003; Milner-Brown et al. 1973a). A high EMG activity results from a high number of recruited muscle fibers and high motor unit discharge frequencies (Moritani and Muro 1987) and is accompanied by high forces generated by the target muscle (Freund et al. 1975; Moritani and Muro 1987). Hence, based on these relations, the EMG activity could be used to easily determine the activation intensity of the muscle in a given set of WBV treatment at least within limits mentioned as cross talk and nonlinearity (Farina et al. 2004; Keenan et al. 2005). Within those limitations, the EMG activity can serve as an adequate parameter to estimate the muscles’ activation intensity. Designing a WBV-based training regimen with the objective to achieve a distinct type of adaptation (e.g., improved power generated by a specific muscle group) requires a thorough understanding of the influence of the training setup and the selected WBV parameters (Cochrane 2011; Rittweger 2010).

This work presents a systematic approach to indentifying conditions that provide a high neuromuscular activity and therefore high activation intensities during WBV. The purpose of this study was to assess the effect of five different vibration determinants (vibration type, frequency, knee angle, stance condition and load variation) and the influence of two factors (muscle group and gender) on the EMG activity during WBV. Based on findings in already existing literature, the following hypotheses were established: (1) we expected that side-alternating vibration causes higher neuromuscular responses than synchronous vibration and (2) a progressive increase of the EMG activity in response to a progressively ascending vibration frequency. With respect to the subjects’ body position, we focused on the static variation of the knee angle and foot position according to the findings of Abercromby et al. (2007a). We assumed that (3) the more the knee joint is flexed the higher the EMG activity in the knee extensor muscles. Accordingly, we hypothesized that (4) a plantar flexion in the ankle joint results in a higher EMG activity in the plantar flexor muscles. Furthermore, we assumed that (5) an additional load during WBV causes higher EMG responses compared to WBV without additional load.

Methods

Subjects

Based on the results of a power analysis (f = 0.4; alpha = 0.05; power = 0.9), 18 volunteers participated in this study. The participants were physically fit students at the department of sports science (8 females and 10 males, age 25 ± 4 years, weight 66 ± 15 kg, height 174 ± 10 cm) with no previous neurological irregularities or injuries to the lower extremity. All volunteers gave their written informed consent. The study was conducted according to the latest revision of the Declaration of Helsinki and was approved by the ethics committee of the University of Freiburg.

Experimental design

In order to examine the influence of five vibration-related determinants on the neuromuscular activity during WBV, a single-group, repeated measures, crossed-study design was used (Fig. 1). For that purpose, the EMG activity of six leg muscles was analyzed with respect to (1) two different vibration types (side-alternating vibration vs. synchronous vibration), (2) a progressive increase in vibration frequencies (5–10–15–20–25–30 Hz), (3) three knee flexion angles (10°–3°–60°), (4) two stance conditions (forefoot vs. normal stance) and (5) load variation (no extra load vs. an additional load equal to one-third of the respective body weight). Among the subjects, the order of the test conditions was randomized to control for confounding effects such as familiarization or fatigue.
https://static-content.springer.com/image/art%3A10.1007%2Fs00421-012-2402-0/MediaObjects/421_2012_2402_Fig1_HTML.gif
Fig. 1

Schematics of the study design. During WBV the EMG activity of six leg muscles was analyzed with respect to the determinants vibration type, vibration frequency, knee flexion angle, stance conditions and load variation

General procedure

Vibration exposure during a single trial was limited to 10 s with at least 30 s of rest in between trials. At the beginning of each measurement, baseline activity in that position without WBV was recorded. This procedure allowed an assessment of the vibration’s contribution to the neuromuscular activity. During testing, the subjects were barefoot in order to avoid any dampening effects due to different footwear. The subjects were instructed to direct their head and eyes forward and distribute their weight equally on both feet. Prior to data collection, familiarization sessions for both WBV devices were conducted to make sure that the participants were able to maintain the defined positions at higher vibration frequencies. In case of trunk movement, foot sliding, changes in body position or any other disturbances the measurements were repeated. The body position was controlled by two operators.

EMG recordings and kinematics

Bipolar Ag/AgCl surface electrodes (Ambu Blue Sensor P, Ballerup, Denmark; diameter 9 mm, center-to-center distance 34 mm) were placed over the M. soleus (SOL), the M. gastrocnemius medialis (GM), the M. tibialis anterior (TA), the M. rectus femoris (RF), the M. vastus medialis (VM) and the M. biceps femoris (BF) of the right leg. The longitudinal axes of the electrodes were in line with the presumed direction of the underlying muscle fibers. The reference electrode was placed on the patella. Interelectrode resistance was kept below 2 kΩ by means of shaving, light abrasion, and degreasing of the skin. The EMG signals were transmitted to the amplifier (band-pass filter 10 Hz–1 kHz, 1,000× amplified) via shielded cables and recorded with 4 kHz. The cables were carefully taped to the skin.

In order to control the body position, ankle angles (dorsalflexion and plantarflexion) and knee angles (knee flexion and extension) were recorded by electrogoniometers (Biometrics®, Gwent, UK). For that purpose, one goniometer was placed over the lateral epicondyle of the femur with one endplate attached to the shank and aligned to the lateral malleolus of fibula and the other to the thigh aligned to the greater trochanter. The knee flexion angle was set to zero at 180° angle between the femur and the fibula. The second goniometer was fixed at the lateral aspect of the right ankle with its ends attached parallel to the major axis of the foot in line with the fifth metatarsal and the major axis of the leg in line with the fibula. An angle of 90° between the fifth metatarsal and the fibula was defined as 90° ankle angle, whereas a plantar flexion was reflected by an ankle angle greater than 90°. To control the vibration amplitude, an accelerometer was fixed on both vibration devices. All signals were recorded with a sampling frequency of 4 kHz.

Independent variables

Vibration type

Two different WBV devices were used: a side-alternating vibration platform (Novotec Medical, Pforzheim, Germany) and a synchronous vibration platform (Power Plate Germany, Frankfurt am Main, Germany). The side-alternating vibration platform generates vibration by rotating along the sagittal axis. Thus, the positioning of the feet in relation to the axis of rotation determines the vibration amplitude. The frequency is adjustable in steps of 0.5 Hz within a range of 5–30 Hz. The synchronous vibration platform moves synchronously in vertical direction, so that the amplitude is fixed at 2 mm and does not change when a different foot position is assumed. It is adjustable in steps of 5 Hz within a range of 30–50 Hz.

For a direct comparison of the two types of vibration, the frequency was set to the only frequency the two devices had in common (i.e., 30 Hz) and the foot-to-foot distance was set to 21 cm (resulting in a amplitude of 2 mm on the side-alternating platform, making it comparable to the fixed amplitude of the synchronous vibration platform). This identical foot position was marked on both platforms.

For the evaluation of the vibration frequency, only the side-alternating vibration device was used, with a foot-to-foot distance of 42 cm, resulting in a vibration amplitude of 4 mm (Marín et al. 2009). This foot position was also marked on the side-alternating platform.

Vibration frequency

To examine the effects of vibration frequency steps of 5 Hz were selected in a range from 5 to 30 Hz (i.e., 5–10–15–20–25–30 Hz). In this setting, only the side-alternating vibration platform was used.

Knee flexion

Knee angles of 5° versus 30° versus 60° were compared to each other (0° corresponding to the fully extended knee). In order to control each of the defined knee flexion angles, the subjects’ knee joint position was adjusted by means of templates given in 5°, 30° and 60° before each measurement. Trials were repeated when a subject could not maintain the given position.

The rationale for the parameter selection was based on several studies that have shown that WBV training with a knee flexion angle of 5°–60° causes beneficial effects on jump height, lower extremity muscle torque and flexibility (Jacobs and Burns 2009; Stewart et al. 2009; Torvinen et al. 2002).

Stance condition

Standing on the forefoot (i.e., forefoot in contact with the platform and heel without contact) was compared to a normal stance (i.e., heel and forefoot in contact with the platform). In order to control the forefoot stance, a foam cube (3 × 3 × 3 cm in dimension) was fixed under the subjects’ right and left heel, in order to reduce the variance in ankle joint position within and between subjects. The subjects were instructed to keep contact with this foam cube without deforming the cube. Measurements were repeated when the defined position could not be held.

Load variation

In order to quantify the influence of load variation, control trials (without additional load) were compared to trials with an additional load equal to 1/3 of the respective body weight. One-third of the body weight as an additional load has been selected based on studies that showed that WBV training in combination with an additional load has a beneficial effect on performance and metabolic power (Cochrane et al. 2008; Garatachea et al. 2007; Rittweger et al. 2001). The additional load was applied via a standard weightlifting bar (180 cm in length and a weight of 10 kg) with weight plates attached to each side. The bar was positioned on the shoulders of the subject. According to the recommendation of Abercromby et al. (2007b), we omitted the additional load in the condition with a 5° knee flexion and a normal stance as this combination has been shown to cause excessive head acceleration. All participants were familiar with loaded squat techniques.

As the load was applied via a bar on the subjects’ shoulders the position of the hands in the load condition was different from the other conditions performed without an additional load in which the hands have been on the hips. In order to control for possible differences due to the altered hand position an additional experiment was conducted in 8 of 18 subjects. For that purpose, we compared hands on the hips (identical to the unloaded condition) to hands high (identical to the load condition) in the forefoot stance and a 5° knee angle as well as in the forefoot stance and a 60° knee angle without and with 30 Hz vibration on both devices.

Data processing

For each of the recorded muscles in each condition, EMG signals with a length of 3 s were rectified, integrated and time normalized (iEMG [mVs]). This kind of data processing was selected as the integrated EMG signal has shown to be a reliable predictor for the level of neuromuscular activation (Kellis and Katis 2008), i.e., the extent of the motor unit recruitment and discharge frequency (Aagaard 2003; Moritani 2002; Moritani and Muro 1987). Subsequently, the vibration gain was calculated in order to allow for a comparison between subjects and to quantify the vibration’s contribution to the neuromuscular activity: in case of the independent variables vibration type and frequency, the iEMG during WBV was normalized to the corresponding stance position without vibration. For the independent variables load variation, knee flexion and stance condition, the iEMG was normalized to the standard stance condition (standing on the forefoot with a knee angle of 5° without additional load).

Mean ankle and knee angles were calculated for each subject and each of the WBV conditions. The vibration amplitude was calculated indirectly by means of an accelerometer signal (Fig. 2) via the formula amplitudecalc (m) = 2 × a (m/s2)/(2 × Π × Frequency)2 (1/s2), where a is defined as the peak acceleration and Π is pi (see e.g., Rittweger et al. 2001).
https://static-content.springer.com/image/art%3A10.1007%2Fs00421-012-2402-0/MediaObjects/421_2012_2402_Fig2_HTML.gif
Fig. 2

EMG data, joint angles and kinematics of one subject during 10 Hz (5 vibration cycles), 20 Hz (10 vibration cycles) and 30 Hz (15 vibration cycles) WBV. All signals were recorded in the forefoot stance condition with 10° knee flexion without additional load. The top shows the EMG data of the gastrocnemius medialis (GM), tibialis anterior (TA) and rectus femoris muscles (RF). Note the frequency-dependent increase in EMG activity from 10 to 20 to 30 Hz. In the middle, the frequency-dependent oscillation of the electrogoniometer signals within a range of approximately 6° reflecting the dampening effects of the ankle and knee joint. The curves at the bottom illustrate the platform acceleration and the vibration amplitude

Statistics

We used an analysis of variance (ANOVA) to analyze the data. The dependent variables in all statistical tests were iEMG values measured from the muscles SOL, GM, TA, RF, VM, and BF. The independent variables were vibration type, vibration frequency, foot position, knee angle, and load variation. Muscle group was included as a within-subject factor to detect differences between the recorded thigh and calf muscles and gender as a between-subject factor. In order to test hypothesis (1) and to detect interaction effects between the independent variables, a repeated measures ANOVA [type (2) × stance (2) × knee angle (3) load variation (2)] with Bonferroni corrected post hoc tests was used. To test hypotheses (2)–(5) and to detect interaction effects between the independent variables, a repeated measures ANOVA [frequency (7) × stance (2) × knee angle (3) load variation (2)] with Bonferroni corrected post hoc tests was used. In addition, to verify the kinematic data, i.e., vibration amplitudes of 2 and 4 mm, knee flexions of 5°, 30° and 60°, forefoot and normal stance, Student’s t tests were used to reveal differences within a given specification. Further, Student’s t tests were used to test for differences in the hand position.

The level of significance was set to 0.05. All analyses were executed using SPSS 17.0 (SPSS Inc., Chicago, IL, USA). Group data are presented as means ± SD unless otherwise stated.

Results

In Table 1, mean values of the vibration amplitude, knee and ankle angles for the different vibration condition are displayed. No significant differences could be observed for the knee and ankle angles across conditions. Although the subjects were instructed to maintain a knee angle of 5° (30° and 60°, respectively), the analysis of the goniometer signals revealed a mean flexion angle of 10° (30° and 60°, respectively). The vibration amplitude was significantly smaller (−5 %) during vertical compared to side-alternating vibration (P < 0.05).
Table 1

Mean values of all participants of the vibration amplitude (amplitudecalc), knee flexion and ankle angles for the different vibration conditions

Vibration type

Synchronous

Side-alternating

Load variation

No load

Load

No load

Load

No load

Load

Amplitude (mm)

2

2

2

2

4

4

Amplitudecalc (mm)

1.90 ± 0.06*

1.86 ± 0.08*

1.98 ± 0.04

1.98 ± 0.08

3.96 ± 0.06

3.96 ± 0.10

Knee flexion

 5° (°)

9 ± 7

11 ± 6

12 ± 4

11 ± 7

7 ± 3

8 ± 6

 30° (°)

27 ± 6

29 ± 10

32 ± 4

33 ± 6

31 ± 8

29 ± 5

 60° (°)

61 ± 4

57 ± 6

57 ± 5

59 ± 8

65 ± 5

63 ± 3

Stance condition

 Forefoot stance (°)

101 ± 12

98 ± 9

97 ± 13

103 ± 14

96 ± 11

95 ± 7

 Normal stance (°)

85 ± 6

82 ± 11

80 ± 9

79 ± 7

81 ± 9

79 ± 12

Values represent MW ± SD. The first column contains the specification according to the predefined vibration determinants. The three rows at the top display the vibration conditions. Note that the vibration amplitudecalc was calculated by means of the acceleration signal. No significant differences could be observed for the knee and ankle angles across conditions, whereas the amplitudecalc differed significantly during vertical compared to side-alternating vibration. Although the knee flexions have been predefined as 5°, 30° and 60°; the analysis revealed mean values of 10°, 30° and 60°

* indicates a significant difference (p < 0.05)

Vibration type

Figure 3 illustrates the effect of the two different vibration types—side-alternating and synchronous vibration—on the EMG activity during WBV. In all recorded muscles (SOL +67 %, GM +91 %, TA +70 %, BF +74 %, VM +69 % and RF +66 %), the EMG activity was significantly enhanced during side-alternating vibration compared to synchronous vibration (P < 0.05). Furthermore, for both vibration types the EMG activity was higher during WBV compared to the control condition without WBV (P < 0.05, Fig. 3).
https://static-content.springer.com/image/art%3A10.1007%2Fs00421-012-2402-0/MediaObjects/421_2012_2402_Fig3_HTML.gif
Fig. 3

Changes in neuromuscular activity during WBV in response to two different vibration types: black bar symbolizes side-alternating vibration and white bar synchronous vibration. The EMG activity during WBV was normalized to the corresponding stance condition without WBV. As a reference, the horizontal dashed line marks the stance condition without WBV. In all recorded muscles, the EMG activity was enhanced during side-alternating compared to synchronous vibration. Note that between the vibration types, the only comparable frequency was 30 Hz and the only comparable amplitude was 2 mm. Data are presented as mean values ± SD of all subjects and conditions. Asterisk indicates a significant difference (P < 0.05)

Vibration frequency

The average EMG activities of all recorded muscles were significantly affected by the vibration frequency (P < 0.05). Figure 4 illustrates the progressive increase in the EMG activity as the vibration increases from 5 to 30 Hz. The frequency-dependent increase in EMG activity was highly pronounced in the shank muscles SOL, TA, and GM proximally located from the vibration device and slightly less pronounced in the distally located thigh muscles BF, RF and VL (Fig. 4).
https://static-content.springer.com/image/art%3A10.1007%2Fs00421-012-2402-0/MediaObjects/421_2012_2402_Fig4_HTML.gif
Fig. 4

Changes in neuromuscular activation due to progressively enhanced vibration frequencies on the side-alternating vibration platform: the higher the vibration frequency, the higher the EMG activity in all recorded muscles. As a reference, the horizontal dashed line marks the stance condition without WBV. The EMG activity during WBV was normalized to the corresponding stance condition without WBV. Note that pooled data from all the conditions except the vibration type are presented as mean values ± SD of all subjects

Knee flexion

Increasing the knee flexion from 10° to 30° to 60° resulted in increasing EMG activity in the knee extensor muscles (RF: +27 % P < 0.05; +30 % P < 0.05 and VM: +29 % P < 0.05; +44 % P < 0.05) and TA (+24 % P < 0.05; +39 % P < 0.05; Fig. 5a). In contrast, the EMG activity of the plantar flexors (GM: −14 % P < 0.05; −16 % P < 0.05 and SOL: −9 % P < 0.05; −15 % P < 0.05) was significantly reduced (Fig. 5a). The BF (−1 % P = 0.74; −10 % P = 0.39) remained unaffected in response to changes in knee flexion. These results show that the muscles were inversely affected by different knee angles: the knee extensors and TA showed highest EMG responses when the knee was flexed at 60°, whereas the plantar flexors showed highest EMG responses at a 10° angle, i.e., nearly extended knee joint (see Figs. 5a, 6).
https://static-content.springer.com/image/art%3A10.1007%2Fs00421-012-2402-0/MediaObjects/421_2012_2402_Fig5_HTML.gif
Fig. 5

Differences in joint torque in the given knee joint position (10°, 30° and 60°) and changes in neuromuscular activation during WBV with respect to different knee and ankle joint positions. a Illustrates the changes in EMG activity due to progressively flexed knee angles from 10° (nearly extended) to 30° to 60°. Note that with an increased knee flexion the residual knee joint torque (Tk) increased in response to an increase of the lever arm (Lk). b Shows the changes in EMG activity during forefoot stance compared to the normal stance, where the heel was in contact with the vibration platform. Note that in all conditions, the subjects maintained a static body position. Data are presented as mean values ± SD of all subjects and conditions. Asterisk indicates a significant difference (P < 0.05); mechanical parameters are abbreviated as follows: center of gravity (COG), axis of rotation (R), weight force vector (Fw)

https://static-content.springer.com/image/art%3A10.1007%2Fs00421-012-2402-0/MediaObjects/421_2012_2402_Fig6_HTML.gif
Fig. 6

Interaction effects: alterations in EMG activity in response to load variation and changing foot and knee positions: on the x-axis the knee angles (10°, 30° and 60°), the foot positions (forefoot stance and normal stance) and the load variation (in gray without load and in white with an additional load equal to 1/3 of the subject’s body weight) are illustrated. The y-axis illustrates the EMG activity. The EMG activity during WBV was normalized to the standard stance condition (without WBV on the forefoot with a knee angle of 10° without load). All vibration frequencies were pooled and data are presented as means of all subjects

Stance condition

Forefoot stance in comparison to normal stance resulted in significantly increased EMG activity only for the plantar flexor muscles SOL and GM (for details see Fig. 5b). In contrast, the EMG activity of the knee extensor muscles (RF and VM) and TA showed a significant decreased EMG activity during forefoot stance compared to normal stance (see Figs. 5b, 6). The BF (P = 0.10) remained unaffected in response to changes in the stance condition (Figs. 5b, 6).

Load variation

Figure 7 illustrates the effect of an additional load on the EMG activity during WBV. An additional load caused an increase in EMG activity in SOL (+11 % P < 0.05), GM (+14 % P < 0.05), RF (+21 % P < 0.05), VM (+15 % P < 0.05) and TA (+13 % P < 0.05). The changes in BF were not significant (+17 % P = 0.31). Detailed results are illustrated in Fig. 6, which shows the changes in EMG activity due to an additional load and with respect to the different knee angles and foot positions. Note that no differences could be observed due to altered hand positions (i.e., hands high vs. hands on the hips) with respect to the knee and ankle joint position as well as in the EMG activity.
https://static-content.springer.com/image/art%3A10.1007%2Fs00421-012-2402-0/MediaObjects/421_2012_2402_Fig7_HTML.gif
Fig. 7

Changes in neuromuscular activity in response to load variation during WBV. Black bar symbolizes the EMG activity during WBV with an additional load and white bar symbolizes the EMG activity during WBV without additional load. In all recorded muscles, an additional load caused an increase in EMG activity. Data are normalized to the conditions established without additional load and data are presented as mean values ± SD of all subjects and conditions. Asterisk indicates a significant difference (P < 0.05)

Interactions

Interaction effects are illustrated in Fig. 6. Significant interaction effects were observed between the variables stance and vibration frequency [stance × frequency] with respect to the EMG activity in SOL (P < 0.05), GM (P < 0.05) and TA (P < 0.05): for the shank muscles, the effect of the vibration frequency on the EMG activity was more pronounced during the forefoot stance than during the normal stance (where the heel is in contact with the platform; see Fig. 5).

Furthermore, the variable load significantly interacted with the variable stance [load variation × stance] in Sol (P < 0.05), GM (P < 0.05) and TA (P < 0.05): in the forefoot stance condition the additional load had a significantly greater effect on the EMG activity than in the normal stance condition (see Fig. 6).

For RF (P < 0.05) and GM (P < 0.05), the knee angle significantly modified the influence of the vibration frequency [knee angle × frequency]; however, the muscles were inversely affected: the more the knee was extended the greater the effects of vibration frequency on the EMG activity in GM were. In contrast, in RF the effect of vibration frequency on the EMG activity was more pronounced when the knee joint was flexed.

For RF (P < 0.05) and VM (P < 0.05), the knee angle significantly modified the effect of additional load [knee angle × load]: the more the knee was flexed the greater the effects of the additional load.

The variable vibration type did significantly interact with the knee angle [knee angle × type] in VM (P < 0.05): during side-alternating vibration, changing the knee angle from 10° to 30° to 60° caused higher EMG responses compared to synchronous vibration.

Moreover, the factor muscle group significantly interacted with the variables frequency [muscle group × frequency] (P < 0.05), knee angle [muscle group × knee angle] (P < 0.05) and stance [muscle group × stance] (P < 0.05). No statistical differences could be observed with respect to the factor gender.

Discussion

The purpose of this study was to assess the effect of five different vibration determinants on the EMG activity during WBV. The main findings were that (1) during side-alternating vibration the neuromuscular activity was enhanced compared to synchronous vibration. Furthermore, (2) the neuromuscular activation during WBV was shown to be closely related to the frequency of vibration: the higher the frequency, the higher the EMG activity. The body position was demonstrated to affect the neuromuscular activation during WBV exposure: (3) the EMG activity for the knee extensors was highest when the knee joint was 60° flexed and (4) the EMG activity for the plantar flexors was highest in the forefoot stance condition. Moreover, (5) an additional load caused an enhancement in EMG activity.

Vibration type

The vibration type extensively affects the neuromuscular activation: during side-alternating vibration the EMG activity was significantly greater than during synchronous vibration in all recorded muscles and conditions. As Marín et al. (2009) have shown that neuromuscular responses increase with increasing amplitudes, it can be deduced that the 5 % difference in vibration amplitude between the vibration devices could have had a considerable impact (Table 1). However, based on the literature some other aspects should be considered as well (Abercromby et al. 2007b; Pel et al. 2009; Pollock et al. 2010). The comparison between side-alternating and synchronous vibration devices revealed differences with respect to the mechanical induction of vibration stimuli to the human body (Abercromby et al. 2007b; Pel et al. 2009; Pollock et al. 2010) and those differences might be associated with substantial consequences for the neuromuscular system (Abercromby et al. 2007b; Pel et al. 2009). Firstly, it was demonstrated that side-alternating vibration generated acceleration magnitudes that were twice as high as during synchronous vibration (Pel et al. 2009). Secondly, the dampening effect of the ankle joint was shown to be smaller during side-alternating vibration than synchronous vibration (Pel et al. 2009). Both aspects indicate that the mechanical induction of the muscle and tendon structures of the lower extremities is extensively more pronounced during side-alternating vibration than during synchronous vibration. Thus, the enhanced transmission of the vibration stimulus during side-alternating vibration might have contributed to the 66–91 % higher EMG responses obtained during side-alternating vibration. However, this conclusion is based on vibration exposure of 30 Hz, which was the only comparative condition between both vibration types. For training purposes, it can be concluded that during side-alternating vibration the mechanical stimulation and as a consequence the muscles’ activation intensity is higher than during synchronous vibration. Based on the positive relation between the muscles’ EMG activity and its motor units recruitment and discharge frequency (Aagaard 2003), side-alternating vibration training might be more efficient compared to synchronous vibration training.

Vibration frequency

The EMG activity increased in response to a progressive increase in vibration frequency in all recorded muscles. These observations are in line with the findings of Berschin and Sommer (2004) and Pollock et al. (2010); they documented a linearly increased EMG activity as a function of the vibration frequency. As the vibration energy is dissipated by the ankle and knee joints as well as the calf muscles, the proximity of a muscle to the vibration stimulus might also have affected the magnitude of neuromuscular responses to WBV exposure (Abercromby et al. 2007b). In the present study, the frequency-induced increase in EMG activity was dependent on the muscles’ anatomical location - proximal or distal from the vibration platform. Accordingly, the proximally located shank muscles (SOL, GM, TA) were more affected by vibration frequency than the distally located thigh muscles (BF, VM, RF; see Fig. 4) and the slope of the EMG activity in response to increased frequencies was also steeper. Therefore, the cushioning effects caused by the angle and knee joint during WBV compromizes the transmission of the vibration stimulus from the proximal to the distal ends with respect to the vibration platform (Bressel et al. 2010; Pel et al. 2009; Pollock et al. 2010).

From a functional point of view, the general increase in neuromuscular activation in response to an increase in vibration frequency leads to an enhanced co-activation of lower extremity extensor and flexor muscles (Pollock et al. 2010; Roelants et al. 2004). This co-activation is considered to have a positive effect on joint stabilization associated with postural control strategies during WBV (Berschin and Sommer 2004; Pollock et al. 2010; Roelants et al. 2004). Based on the data presented by Cochrane et al. (2009) and Ritzmann et al. (2010) it can be assumed that during WBV the muscle tendon units are stretched in every WBV cycle (Cochrane et al. 2009), these stretches induce a frequency-dependent activation of the muscle spindles and thus elicit stretch reflex responses detectable in the EMG signal (Ritzmann et al. 2010). In consequence, it can be assumed that the increases in EMG activity due to enhanced vibration frequencies were caused by the higher number of stretch reflex responses.

Cardinale and Lim (2003) showed that EMG response using WBV frequencies above 30 Hz in the half squat position were lower. In summary, both findings indicate that high vibration frequencies of 30 Hz are efficient for a WBV-based training regimen aiming to achieve high activation intensities in the muscles of the lower extremities.

Knee flexion

The modulation of the knee joint flexion substantially influenced the EMG activation as well. For the knee extensors, neuromuscular activation was highest when the knee joint was flexed at an angle of 60° (Fig. 5a). This finding might be attributed to the residual joint torque (Tk,), which is related to the length of the lever arm (Lk) in a given knee joint position (see Fig. 5a). According to the muscles’ anatomical functions, a progressive increase of knee flexion might enhance the voluntary activation of RF and VM in order to generate higher forces that compensate for increased joint torques (Kooistra et al. 2006; Pincivero et al. 2004). In addition, the size of the stretch reflex amplitude is related to the amount of voluntary background activation (Bedingham and Tatton 1984). Based on both aspects it might be supposed that larger knee angles would result in a greater muscle stretch in the knee extensor muscles in each vibration cycle and that the increased Ia-afferent stimulation would result in greater neuromuscular responses (Abercromby et al. 2007b). For WBV training purposes aiming to achieve high neuromuscular demands in the knee extensor muscles, a flexed knee joint position is advisable in any given WBV training setup. In contrast, it should be considered that in more extended knee positions, the neuromuscular activation of the plantar flexors is increased (Fig. 5). Therefore, for the plantar flexors a knee extension rather than a flexion is recommendable. In conclusion, in order to achieve high EMG activations and consequently high activation intensities in specific muscles, an appropriate adjustment of knee flexion dependent on the targeted muscle group is necessary.

Stance condition

The plantar flexors are extensively more activated in the forefoot stance condition compared to the normal stance (with the heels in contact with the platform). In contrast, the knee extensors and TA are more activated in the normal stance condition, whereas BF remained unaffected. The distinct increase in EMG activity of the plantar flexors in the forefoot stance condition might be attributed to the muscles’ anatomical functions and thus, to the level of neuromuscular activation. The residual ankle joint torque is higher in the forefoot stance condition compared to the normal stance. Thus, the voluntary activation of SOL and GM might be enhanced in order to generate a higher force that compensates for the increase in angle joint torque (Nolan and Kerrigan 2003; Sasagawa et al. 2009). Furthermore, the size of the stretch reflex amplitude is related to the amount of voluntary background activation (Bedingham and Tatton 1984). Thus, based on both aspects it can be proposed that in the forefoot stance condition the vibration-induced stretch reflexes are enhanced during WBV compared to the normal stance due to an increase in voluntary activation. For WBV training purposes, a forefoot stance is recommendable if the main focus is on the plantar flexor muscles. A normal stance condition should be preferred when the emphasis is on the knee extensors, TA or BF. One limitation of the study could be that the back- and forwards shifting of the weight towards the forefoot or towards the heel has not been controlled. Although we instructed the subjects to distribute their weight equally and we supervised the trunk movement visually, we cannot exclude slight changes in weight shifts possibly influencing the data in the normal stance condition.

Load variation

The current study provides evidence for load-dependencies during WBV training: In all recorded muscles, an additional load caused an increase in EMG activity during WBV (Fig. 7). This indicates that the pre-loading of certain muscles during WBV exercise enhances the activation of those muscles according to the suggestions of Rittweger et al. (2003); they observed an increase in metabolic power in response to an additional load. Moreover, the present result is well in line with the findings of Hazell et al. (2010) who documented an enhancement in EMG activity during WBV due to an additional load in a dynamic squatting condition. For WBV training purposes this data suggests that the training effectiveness of WBV exposure can be boosted by means of external loading. One limitation of the present study is that in contrast to the knee and the ankle, the hip angle has not been controlled. The hip angle is an important ingredient in static squat and might be modified in the loaded compared to the unloaded condition.

Interaction effects

There are two main aspects that might be derived from the observed interaction effects. First and most importantly, the effects of the selected WBV determinants can be enhanced by each other. Thus, the most relevant parameter seems to be the body position consisting of knee angle and stance condition. As the variable knee angle interacted with the variables vibration type, frequency and load variation, it is concluded that independent of the level of voluntary activation, the more the knee joint is flexed the more the effect of those determinants is enhanced for the knee extensor muscles. Furthermore, the variable foot position interacted with the variables vibration frequency and load variation for the plantar flexors. Hence, the forefoot stance condition significantly augmented the effect of the vibration frequency and an additional load for the knee extensors, whereas normal stance had the same effect for TA. In summary, those interaction effects indicate that the body position is the target parameter, which predominantly modifies the effect of the other determinants. Second, the between-subject factor muscle group (thigh vs. calf muscle) interacted with the variables frequency, knee angle and stance condition. Based on these interactions it can be proposed that vibration training recommendations should be specified with respect to the muscle group and their anatomical location, i.e., the proximity to the vibration device.

In contrast to recently published studies demonstrating gender-specific differences in response to WBV (Sañudo et al. 2011; Merriman et al. 2011) the present study revealed no gender-effect. The contradictory results could be based on the discrepancy between neuromuscular properties and functional aspects. The studies so far conducted to evaluate the effect of gender have focused on physical performance (Merriman et al. 2011) and knee stability (Sañudo et al. 2011) within a given functional setting. In contrast, in the present study the activation intensity was evaluated in a given WBV treatment.

Conclusion

The objective of the current study was to evaluate the influence of vibration type, vibration frequency, additional load, knee angle and stance condition on the neuromuscular activity of the lower extremity muscles during WBV. The rationale of the study was to define specific recommendations for WBV training purposes.

The combination of high vibration frequencies of 30 Hz and an additional load on a side-alternating vibration platform is associated with the highest EMG activity during WBV exposure. Therefore, this combination is supposed to be the most efficient for WBV training purposes. Regarding the body position, a knee flexion of 60° and normal stance conditions with the heels in contact with the platform seem to be most beneficial for the knee extensor muscles. In contrast, forefoot stance and a nearly extended knee position is recommended for a specific training of the plantar flexor muscles.

Acknowledgments

This study was funded by the Federal Institute for Sports Science (BISp AZ 070608/10).

Copyright information

© Springer-Verlag 2012