Secondary Logo

Journal Logo

Original Research

A Vibratory Bar for Upper Body: Feasibility and Acute Effects on EMGrms Activity

Moras, Gerard1; Rodríguez-Jiménez, Sergio1; Tous-Fajardo, Julio1; Ranz, Daniel1; Mujika, Iñigo2,3

Author Information
Journal of Strength and Conditioning Research: August 2010 - Volume 24 - Issue 8 - p 2132-2142
doi: 10.1519/JSC.0b013e3181aa3684
  • Free



Vibration training has been promoted during the last decade as an alternative and complementary method to resistance training (26,32). A number of studies have examined its effects on several physiologic systems, including the neuromuscular (38), endocrine (6), cardiovascular (37), sensory (30), circulatory (27), and bone systems (16).

Vibratory stimuli can be applied directly to the targeted tendon (24,28) or muscle belly (18,20) or indirectly through a part of the body to the target muscle(s) (34) in either the lower- (9,33) or upper-body extremities (4,8,19). In the most common and well-investigated indirect method, a subject stands on a vibration platform that transmits the vibration to all the muscles of the lower extremities (38) via the feet. In this case the vibration signal is attenuated during its transmission through soft tissues, which are capable of absorbing and dampening the vibration (15). This means that the amplitude of vibration could be very small in the target muscle and may be insufficient to elicit a training effect (34), especially at the more distant muscles. In contrast, when vibration is directly applied to the tendon or muscle belly, the attenuation of vibration by transmission is much lower (24).

The acute response of the neuromuscular system to different vibration loads may be analyzed by EMG measurements. It is well known that sustained vibration applied to a muscle or tendon can stimulate muscle spindles and elicit a tonic vibration reflex (TVR) primarily via Ia monosynaptic and polisynaptic pathways (12,28). However, as stated earlier, it is important to separate research using the direct application of a vibration signal at the surface of the muscle/tendon unit or an indirect application through platforms or dumbbells. Nevertheless, several studies have reported an increase in EMG activity in both lower- (2,5,17,38) and upper-body muscles (4,24), but to our knowledge, only 4 studies have analyzed EMG responses to different vibration frequencies. Two studies (5,17) reported significant differences between some of the vibration frequencies selected, whereas 2 other studies (13,35) observed no difference. To date, no studies using devices similar to the VB presented in this study have examined the acute effects of indirect vibration on the upper body on EMG activity.

Research evaluating the effects of vibration on the upper body is scant, and this may be because the majority of vibration-training devices for the upper extremities are not readily available for purchase (21). For example, Issurin and Tenenbaum constructed an electromotor device to produce superimposed vibration onto an isotonic cable-pulley resistance system (19). Other authors have used a vibratory dumbbell (8,29). Cochrane and Hawke (8) used a commercialized electric-powered dumbbell with a rotating axis that delivered oscillatory movements to the shoulders and arms. More recently, Poston et al. (36) used a vibration apparatus consisting of an electric motor mounted on 1 sleeve of a standard barbell to examine the acute effects on power output in the bench press. An important contribution of that study was their choice of a multijoint upper-body exercise commonly used in resistance training. It is important to emphasize that although all upper-body athletic movements involve the coordination of several joints, most studies involving vibration of an upper limb included single-joint exercises, usually the elbow (24). In addition, most of the vibration systems used in those studies were not designed to simultaneously stimulate upper-body muscles of both upper limbs. With this aim, a vibratory bar was designed to provide an indirect and simultaneous vibratory stimulus equivalent to that found in vibration platforms. However, this device would allow for a wider range of multijoint upper-body exercises in contrast with the widespread platforms, where the options are mainly limited to push-up variations. Given the reported differences between upper- and lower-body muscles in muscle quality (25), time course, and extent of training adaptations (1), more research clearly is needed to determine the effects of vibration on upper-body muscles.

Therefore, the first purpose of this study was to determine the feasibility of a purpose-designed vibratory bar (VB) to produce an intensity of vibration with a peak-to-peak amplitude, frequency, and acceleration range appropriate for vibration training. The second purpose was to assess the reliability of the acceleration measured directly on the VB. The last purpose was to assess the acute effects on electromyographic (EMG) activity during vibration while subjects held the bar in extended and flexed isometric positions during the bench press exercise. First, it was hypothesized that the indirect application of the vibration through the bar would result in a higher EMG activity of the upper-body muscles in all vibration conditions when compared with no vibration condition. Second, it was hypothesized that there would be differences on EMG activity between the different vibration conditions.

Material and Methods

Experimental Approach to the Problem

The use of vibratory bars to provide an indirect and simultaneous vibratory stimulus to upper-body muscles during multijoint exercises seems to be an interesting option for improving neuromuscular performance. For this purpose, a vibratory bar was designed and the acute effects on EMGrms were assessed. The testing procedure included 2 sessions separated by 4 days. During session 1 the vibration maximal acceleration (Accmax), frequency (Freq), and peak-to-peak amplitude (Ampp-p) obtained at the center (C) and at the end (E) of the VB by means of an accelerometer were assessed. Acceleration data were recorded directly on the VB in an isometric extended position (EP) during the bench press exercise with full elbow extension (joint angle of 180 degrees) in the following vibration conditions selected in the inverter (i, a device used to adjust the speed of the electromotor according to the required frequency): no vibration (0), 20, 30, 40, and 50 Hzi. In session 2, the acute effects of vibration on EMG activity while subjects held the VB were investigated. EMG activity of the triceps (long head), deltoid (anterior fibers), and pectoralis major were recorded in 2 different isometric positions (EP: full elbow extension with joint angle of 180 degrees; and flexion position (FP): elbow joint angle of 90 degrees) in the following vibration conditions: 0, 25, and 45 Hzi. For comparison, muscle activation during a maximal voluntary isometric contraction (MVIC) was also recorded.


Sixteen male physical education students (age 23.9 ± 3 years, height 180.4 ± 6.5 cm, mass 78.1 ± 8.2 kg) with previous experience in resistance training and the bench press exercise, but not necessarily in vibration training, volunteered to participate in both sessions of this study. Experience was defined as training in an appropriately supervised environment at least 2 times per week for at least 39 weeks of the year for at least 2 years. However, none of the subjects had a highly trained status. Reasons for exclusion were diabetes, epilepsy, acute hernia, joint implants, any history of severe musculoskeletal problems in the upper body, or recent operative wounds. All subjects were fully informed of the procedures and risks involved before written consent was obtained. The study was reviewed and approved by an Institutional Review Board of INEFC Barcelona for research with human subjects.

Vibratory Bar Design

The VB consisted of 2 three-phase motors (Italvibras MVSI 3/200E-S90) fixed at the bar with 2 brackets 20 cm apart (Figure 1). The outer diameter of the bar was 50 mm, and 30 mm at the C and right and left E, with a wall thickness of 2.5 mm. The total length of the bar was 1.15 m. The total mass of the VB was 15.0 kg. An inverter (Omron Sysdrive 3G3JV) with a control panel provided the speed reduction of the electromotor. The motors' speed was set at 3,000 rpm. This device was used to adjust the speed of the 3-phase motors according to the required frequency. Three-phase motors were placed near the C of the bar because pilot testing with the motors fixed at the extremes of the bar resulted in uncomfortable and high Ampp-p (around 10 mm) at E. These high amplitudes were considered out of the appropriate range for upper-body vibration training.

Figure 1
Figure 1:
Exploded view of the vibratory bar parts (1, bar; 2, brackets; 3, three-phase motors connected to an inverter to adjust the electromotor speed according to the required frequency.


A single-axis accelerometer (Ergotest Technology, Langesund, Norway) was fixed to the VB with adhesive tape. The accelerometer was used to measure the vibration acceleration directly at the right E and at C of the VB in the vertical axis (Figure 2). The size of the accelerometer was 35 mm (width) × 55 mm (height) × 15 mm (depth), and the mass was 22 g. The output of the accelerometer was connected to a MuscleLab 4000e unit (Ergotest Technology, Langesund, Norway). The sampling frequency of the acceleration signal was set at 100 Hz.

Figure 2
Figure 2:
Accelerometer at the center (C) of the bar during isometric extended position (EP).

EMG Analysis

EMG activity of triceps brachii (long head), deltoid (anterior fibers), and pectoralis major muscles were recorded with bipolar surface electrodes (Blue Sensor, Medicotest, Olstykke, Denmark) fixed longitudinally to the muscle belly in the right arm with a 25-mm interelectrode distance. The skin was shaved and cleaned with an alcohol swab following Cram and Kasman guidelines (10). The sampling frequency for the EMG signal was 1,000 Hz. Raw EMG signals were preamplified 600 times and filtered through a band-pass filter with low and high cutoff frequencies of 6 and 1,500 Hz, respectively. The filtered signal was converted to a root mean square signal using a AD536 circuit (Analog Devices, Norwood, Massachusetts) with an average constant of 100 ms. The converted signal, the root mean square of the raw signal, was then sampled at 100 Hz using a MuscleLab system (Model 4000e; Ergotest Technology, Langesund, Norway). Electromyography cables were fastened to prevent them from swinging and from motion artifacts.

Elbow Joint Angle Measurement

The joint angle was monitored by means of a goniometer (SG110, Biometrics, UK) in order to locate the elbow joint angle at 180 degrees (extended elbow) or 90 degrees.

Experimental Procedures

The first experimental session started with a standardized warm-up consisting of 5 minutes of cycling on an ergometer without resistance and a general warm-up including arm/shoulder mobilization exercises. Before starting the trials, the subjects were familiarized with the protocol and the equipment used. The Acc data were recorded directly on the VB during an isometric EP in the bench press exercise with full elbow extension (joint angle of 180 degrees) in the following vibration conditions: 0, 20, 30, 40, and 50 Hzi. The Acc was recorded 3 times in every vibration condition (5 vibration conditions × 3 trials). In the first trial the accelerometer was positioned at C of the VB (Figure 2). In the second and third trials it was positioned at E of the VB. The third trial was performed to analyze test-retest reliability at E.

The Acc was recorded during 15 seconds in every trial. To standardize the bar-holding position of the hands during both testing sessions, a 90-degree elbow angle was ensured by using a goniometer, thus providing the same muscle participation in all trials (SG110, Biometrics, UK). In addition, the subjects' upper arms were parallel to the ground while lying supine on the bench. In this position, the grasping places on the bar of both index fingers were marked, and then all attempts were performed using these marked places during the testing sessions. Subjects were asked to hold the VB with a pronated grip and not to modify the wrist position while maintaining a straight angle (180 degrees) at the elbow joint during the assessment (Figure 2). Subjects were encouraged to maintain a comfortable and constant grip force throughout the procedure. The vibration conditions were applied in a random order between subjects with a minimum rest of 5 minutes.

The purpose of session 2 was to determine whether vibration could acutely increase the EMGrms values of upper-body muscles. In addition, Acc produced directly at E of the VB was recorded. The same general warm-up as in session 1 was used. Subsequently, the subjects were familiarized with the assessment protocols of the session. After positioning of the electrodes and before assessment protocols were started, a MVIC test was conducted. Using the same grip distance between index fingers as in session 1, subjects performed 2 MVIC over 5 seconds. During all MVIC tests subjects were encouraged to push as hard as possible against an immovable bar during an isolated bench press at arm angles of 90 degrees. If results differed by more than 10% between trials, a third trial was performed. Arbitrary (mV) data were recorded and normalized (%) data were rescaled in the following muscles: triceps brachii (long head), deltoid (anterior fibers), and pectoralis major. Thereafter, assessment protocols were started after 10 minutes of rest. Acc at E of the VB and EMGrms were recorded during isometric EP and FP. The subjects performed both isometric positions at 0, 25, and 45 Hzi. The elbow joint angle was monitored by means of a goniometer to ensure 180 degrees or 90 degrees (Figure 3). EMGrms recordings started when the subjects assumed the correct position. Muscle activity and Acc were recorded during 5 seconds before vibration and for the 15 seconds of vibration. At 0 Hzi muscle activity was recorded during 15 seconds. The 6 trials (2 positions × 3 vibration conditions) were applied to each subject in a random order. Athletes self-selected the rest interval between trials, with a minimum of 5 minutes of rest enforced.

Figure 3
Figure 3:
Accelerometer at the end (E) of the bar during isometric flexion position (FP).

All subjects were encouraged to report any unusual symptoms (e.g., discomfort, queasiness) during vibration exposure in either session.

Data Analysis

An example of the Acc measured at C of the VB is shown in Figure 4A. Fast Fourier Transformation (FFT) analysis (block size of 256 and a Hann window) performed on the acceleration signals showed always only 1 central peak (Figure 4B). The frequency at which this peak component was present was termed “peak frequency” in this study. Because the vibration acceleration of the VB was sinusoidal, the amplitude was calculated as follows: Amp = Accmax/(2πFreq),2 where Accmax is the maximal acceleration at peak frequency (m·s−2), Amp is the maximum displacement (m) from the equilibrium position in the vertical axis, and Freq is the peak frequency (Hz). Matlab V7.01.1 (Matlab, Mathworks, Nastick, Massachusetts) programming environment was used to perform FFT and calculate the Amp. The acceleration value over a range of 10 seconds (range between 5 and 15 seconds of the registration) was computed for each trial. All values are presented as means (SD).

Figure 4
Figure 4:
A) Example of measured vibration acceleration at the center of the vibratory bar in session 1 (vibration condition of 40 Hzi). B) Example of Fast Fourier Transformation analysis performed on the acceleration signal presented in panel A.

The EMGrms value over a range of 10 seconds (range between 10 and 20 seconds of the registration for vibration conditions of 25 and 45 Hzi and between 5 and 15 s for 0 Hzi conditions) was computed for each trial in each muscle group and selected for statistical analysis. Muscle activity was expressed relative to the measured MVIC values (%MVIC).

Statistical Analyses

Vibration Maximal Acceleration, Peak Frequency and Peak-to-Peak Amplitude

Differences in Accmax, Freq, and Ampp-p (dependent variables) between C and E positions of the accelerometer for all vibration conditions are expressed as standard error of measurement (SEM) and coefficient of variation (CV). A spreadsheet that analyzes validity by linear regression proposed by Hopkins was used for calculations (Hopkins WG. Analysis of validity by linear regression (Excel spreadsheet). In: A new view of statistics. Internet Society for Sport Science, 2000). A 1-way analysis of variance (ANOVA) was used to determine whether significant differences existed between vibration conditions in C and E. Additionally, an independent sample t test was used to determine whether significant differences existed between C and E for all vibration conditions. The threshold level for significance was set at p ≤ 0.05, whereas the confidence intervals (CI) were set at 95%. The intertrial reliability of Accmax at E of the VB was calculated using the intraclass correlation coefficient (ICC).


To determine the effect of vibration conditions (0, 25, and 45 Hzi) on EMGrms (dependent variable) during the isometric EP and FP, a 2-factor ANOVA (2 [isometric positions] × 3 [vibration conditions]) with repeated measures on the subjects was used. For all analysis, statistical significance was set at p ≤ 0.05. The variables were tested by the Bonferroni honestly significantly difference (HSD) or the Games-Howell test in case the Levene's test was p > 0.05 or p ≤ 0.05, respectively. SPSS 13.0 software was used for all statistical analysis.


Vibration Maximal Acceleration, Peak Frequency, and Peak-To-Peak Amplitude (Session 1)

As can be seen in Table 1, significant differences were found between the Accmax measured at C and E at all vibration conditions selected in the inverter. Significant differences were also found (p < 0.05) for Accmax measures among all vibration conditions at C and E, except between 40 Hzi and 50 Hzi, at E. Although an increase of the Freqi increased the Accmax, these increases of Accmax were smaller as Freqi was higher. In fact, Accmax measured at 50 Hzi was smaller than that measured at 40 Hzi (96.4 m·s−2 vs. 92.5 m·s−2, respectively). The intertrial reliability (ICC) of measurement at E ranged from 0.68 to 0.91 (Table 2).

Maximal acceleration (Accmax) measured directly at the center (C) and at the end (E) of the vibratory bar (VB). Values are means (SD), n = 16.
Intertrial reliability (ICC) of the maximal acceleration (Accmax) measured at the end (E) of the vibratory bar (VB). Values are means (SD), n = 16.

No significant differences were found between the vibration Freq measured at C and E for any of the conditions (Table 3). Significant differences were found (p < 0.001) for Freq measures among all vibration conditions (20, 30, 40, and 50 Hzi) at C and E. However, Freq obtained at 40 Hzi and 50 Hzi resulted in smaller differences than the rest of conditions. All Freq obtained at C and E were significantly different (p < 0.001) to the Freqi. Thus, the variation obtained between the Freqi 20, 30, 40, and 50 Hzi and the Freq obtained at C and E were 6.8%, 11.6%, 25.5%, and 38.4% and 7.4%, 11.9%, 26.0%, and 38.6%, respectively.

Frequency (Freq) measured directly at the center (C) and at the end (E) of the vibratory bar (VB). Values are means (SD), n = 16.

The Ampp-p values measured ranged from 4.62 mm to 6.10 mm (Table 4). Significant differences were found between the Ampp-p measured at C and E at all vibration conditions. Significant differences were also found (p < 0.05) for Ampp-p measures among all vibration conditions except between 20 Hzi and 30 Hzi and between 40 Hzi and 50 Hzi, at E. Statistical analysis showed that the position of the accelerometer (C or E) had a significant effect on the Accmax and Ampp-p.

Peak-to-peak amplitude (Ampp-p) measured directly at the center (C) and at the end (E) of the vibratory bar (VB). Values are means (SD), n = 16.

None of the subjects reported any unusual symptoms (e.g., discomfort, queasiness) during vibration exposure at either session.

Muscle EMGrms Activity (Session 2)

As can be seen in Figure 5, triceps brachii (long head) EMGrms (%MVIC) activity during EP and FP was always significantly higher at 25 Hzi and 45 Hzi compared with 0 Hzi.

Figure 5
Figure 5:
Electromyography root-mean-square (%MVIC [maximal voluntary isometric contraction]) in the triceps brachii muscle (long head), deltoid muscle (anterior fibers), and pectoralis major muscle during A, isometric extended position (EP), and B, isometric flexion position (FP), for all vibration conditions (0, 25 and 45 Hzi). *p < 0.05; †p < 0.01. n = 16.

Deltoid (anterior fibers) EMGrms (%MVIC) activity during EP and FP was always significantly higher at 45 Hzi compared with 0 Hzi, but only during EP at 25 Hzi (Figure 5).

Pectoralis major EMGrms (%MVIC) activity during EP and FP was always significantly higher at 45 Hzi compared with 0 Hzi but only at 25 Hzi during EP.

No significant differences were found in any of the muscles between 45 Hzi and 25 Hzi, neither at EP nor at FP.

The Freq obtained at E for vibration conditions of 25 Hzi and 45 Hzi were 22.98 (0.3) Hz and 30.84 (0.4) Hz, respectively. As expected, these Freq at E were significantly different. The Accmax and the Ampp-p obtained for both vibration conditions ranged from 61.3 (5.3) m·s−2 to 111.1 (6.2) m·s−2 and 6.18 (0.8) mm to 6.01 (0.4) mm, respectively. Whereas the Accmax were significantly different, the Ampp-p obtained were not significantly different between them.


This study determined the feasibility of a purpose-designed VB to produce an intensity of vibration with an Ampp-p, Freq, and Accmax range appropriate for vibration training. Thereafter, the acute effects of different vibration frequencies on EMG activity using the VB were examined. The VB is capable of simultaneously stimulating upper-body muscles of both limbs via the hands during a multijoint upper-body exercise. To our knowledge, this is the first study to address this topic.

The results of the present investigation showed that Ampp-p (from 4.6 mm to 6.1 mm), Freq (from 18.5 to 30.8 Hz), and Accmax (from 37.4 to 96.4 m·s−2) ranges found in the VB were inside those found in the literature for vibration platforms. Indeed, studies have reported vibrations across a range of amplitudes from <1 mm to 10 mm, frequencies from 15 to 60 Hz, and accelerations that can reach 15 g (1 g = 9.81 m·s−2) (7, 21). In addition, these results are in line with vibration characteristics used in previous studies for upper-body muscles (4,8,29). However, in some of these studies maximal acceleration is not specified (26) or is inconsistent (4); moreover, it is not clear if the vibration amplitude refers to the peak-to-peak amplitude. According to Lorenzen et al. (23) vibration exercise research should report the peak-to-peak amplitude or displacement (mm), frequency (Hz), and maximal acceleration (m·s−2) for the standardization of terminology. In this regard, we are far from knowing if the same vibration parameters could be equally effective in the long term for both upper- and lower-body limbs. In fact, it has been reported that upper-body skeletal muscle adaptations to strength training are greater and occur earlier compared to those from the lower extremity (1). Furthermore, a greater muscle quality (strength per unit of muscle mass) has been found in the arm in contrast with the leg in both genders throughout the adult life span (25).

We also found an acceptable intertrial reliability in the Accmax at E (range from 0.68 to 0.91). E was chosen to be the nearest point to the hands grip on the bar. The lowest ICC values were found at 30 and 40 Hzi (0.68 and 0.74, respectively). The variations of ICC values could be explained by the fact that the amplitude of the acceleration is highly influenced by grip force. It must also be taken into account that the activation of the tonic vibratory reflex could also affect the grip force. Although subjects were asked to maintain a constant grip force during protocols, this variable could not be thoroughly controlled. Accmax was significantly different between E and C of the VB, being generally higher at E. These differences could be explained by the fact that vibration Ampp-p was also significantly different between C and E, being generally higher at E without significant differences in vibration Freq. This justification is based on the formula used in data analysis (Amp = Accmax/(2πFreq)2). Another important consideration about this VB is that selecting frequencies of 40 Hzi or 50 Hzi does not result in major changes of vibration Accmax and Freq on the bar. Thus, the VB designed for this study achieved its higher mechanical output in Acc and Freq around 45 Hzi.

In contrast with the results of Accmax and Ampp-p, no significant differences in Freq were found between C and E. However, Freqi were higher than those recorded on the VB. Precisely, the differences between 20, 30, 40, and 50 Hzi, and the Freq obtained at E were 7.4%, 11.9%, 26.0%, and 38.6%, respectively. Thus, the higher Freq value that this VB can provide was around 30 Hz, which was achieved selecting 40, 45, or 50 Hzi. Perhaps, this could be explained from the interaction between the 2 motors and the vibration transmission through the brackets. Therefore, to analyze the acute effects on muscle activity using different vibration frequencies provided by this VB, it was necessary to take into account the real vibration frequency produced along the VB and more concretely at the nearest place of transmission with the hands grip. These findings highlight the need for future investigations to verify if the vibration frequency and amplitude provided by the manufacturers of commercially available devices are obtained at the nearest place in contact with the body. In this respect, a recent study (27) verified with an accelerometer movements in the X, Y, and Z directions to prove that mass location or quantity did not alter acceleration. Poston et al. (36) have used a vibration apparatus similar to the VB. However, the vibration parameters provided by the authors (frequency 30 Hz; amplitude 1 mm) do not allow knowledge of the real vibration amplitude produced on the bar. Taking into account our results, the vibration acceleration and amplitude should be different along the barbell. In addition, considering that only 1 motor was mounted on one end of the barbell, it could be expected that vibration parameters on the opposite end were different.

According to the results obtained during session 1, two representative vibration frequencies were chosen to analyze their acute effects on muscle activity (EMGrms). Vibration frequencies of 25Hzi and 45Hzi were selected with the aim of obtaining 2 Freq close to 20 Hz and 30 Hz that were different enough from each other. Thus, Freq obtained at E at 25 Hzi and 45 Hzi were 23 Hz and 30.8 Hz, respectively, which were significantly different. First, we hypothesized that the indirect application of the vibration through the bar would result in higher EMG activity of the upper-body muscles in vibration conditions when compared with no vibration condition. Our data support the first hypothesis because EMGrms for all vibration conditions and isometric positions were significantly higher in all muscles compared with no vibration, except during 25Hzi for the deltoid and pectoralis, in FP. In addition, the highest EMGrms was found at 45 Hzi except for the pectoralis muscle in EP, suggesting this Freqi (Freq at E of 30.8 Hz) was the one eliciting the highest reflex response during this study. These findings are in accordance with various studies that found that using vibration platforms at frequencies between 30 and 50 Hz elicited the maximal EMG response of lower-body muscles compared with the same exercises without vibration (5,17,38).

Second, we hypothesized that there will be differences on EMG activity between vibration conditions. Our data did not support this hypothesis; no significant differences were found in any of the muscles between 23 Hz and 30.8 Hz, neither at EP nor at FP. This is in agreement with results reported by Fratini et al. (13) and Moras et al. (35) but in contrast with Cardinale and Lim (5) and Hazell et al. (17), who found differences between frequencies while subjects stand on a platform during a static semisquat. A possible explanation for this discordance could be the different vibration devices and frequencies selected and perhaps more important the highly individual response to vibratory stimuli (3,11). Cardinale and Lim found significant differences on EMGrms activity of vastus lateralis between 30 and 50 Hz and 40 and 50 Hz but not between 30 and 40 Hz. However, in this study the normalization of the EMG signal was not conducted and data published in microvolt values must be used with caution because they are influenced by the individual signal detection condition (22). However, Hazell et al. analyzed normalized EMG data to find significant differences in vastus lateralis between 25 to 30 Hz and 40 to 45 Hz and between 25 Hz and 40 to 45 Hz in biceps femoris. However, despite the inconsistent results found in the literature about whether a given frequency will increase muscle activity more than any other frequency, a recent study found that an 8-week vibration training program including individualized frequencies improved jumping performance more than a fixed frequency program (11).

Another possible explanation for the absence of differences between frequencies in our study is the potential presence of motion artifacts on EMGrms signal as suggested by Abercromby et al. (2) and Fratini et al. (13). Variability of motion artifacts intensity has been proposed as one of the reasons for the inhomogeneous findings reported in the literature (31). Although motion artifacts could indeed lead to an overestimation of EMG power, we consider that such artifacts would not affect the conclusions from this study. First, the vibration-induced increase in EMGrms signal is a consistent finding in the literature (4,17,38). Second, Fratini et al. (13) found no significant differences with respect to vibration frequency either for the filtered or the unfiltered EMG signal.

Using the direct vibration method, differences in upper-body muscles between vibration and no vibration conditions also have been found (24). However, it should be pointed out that all the parameters of this vibration application were different from those used in our study. Although various studies have evaluated the acute effects of indirect vibration on several measures of upper-body muscular performance (8,29,36), few of them have analyzed EMG activity during vibration stimulus (4,39). Bosco et al. (4) performed an investigation evaluating EMG from the biceps brachii muscle during the vibration treatment and a significant enhancement was found. They used a 5-minute vibration protocol at 30 Hz (6-mm amplitude) with the forearm fixed in a semiflexed position and the hand grasping a purpose-built vibration handle with a load of 2.8 kg. Although the method of vibration production, the apparatus and subject placement, the duration of vibration exposure, and the gripping characteristics used by Bosco et al. were different compared with our study, the results are similar.

In the present study, the vibration effect was dependent on the distance between the muscle and the hands grip. For example, in EP (Figure 5A) the relative vibration-induced increase in activity of the triceps brachii was higher compared with the more distant muscles from the hands grip (deltoid and pectoralis). Other authors have reported similar findings in the upper and lower body (17,38). Hazell et al. (17) analyzed EMG activity in the biceps brachii and triceps brachii while standing on a vertically oscillating vibration platform. They found that a vertically based floor vibration stimulus up to 45 Hz and 4 mm results in relatively small increases in upper-body muscle EMG activity relative to lower-body muscles for unloaded static and dynamic biceps curl. They argued that to provide sufficient upper arm muscle benefits using a vertically oscillating platform, direct hand contact with the platform might be required (i.e., push-ups or triceps dips).

It has been suggested that the position of the body during the exposure to vibration and the degree of muscle contraction can affect the biologic response to vibration (15). The EMG response in the shoulder muscles during exposure to vibration has been shown to vary according to the position of the arm and the level of muscle contraction (39). Rubin et al. (40) analyzed transmissibility of low-level, high-frequency, ground-based vibration to the hip and spine of 5 volunteers while standing erect, relaxed, and with knees bent (20 degrees of knee flexion). They found that with knees bent, the transmissibility decreased to much less than 50% in the hip, yet remained at approximately 60% in the spine. In the present experiment, this phenomenon could explain the fact that no differences were found between vibration at 25 Hzi and no vibration for deltoid and pectoralis muscles occurred during isometric FP. It could be argued that at 90 degrees of elbow flexion, the transmissibility of mechanical signals in more distant muscles from the hands grip should be lower. Thus, we can suggest that during FP the frequency should be higher than 25 Hzi to induce a significant EMG activity increase in deltoid and pectoralis muscles.

The use of different vibratory bars designed to perform isometric or dynamic exercises is intriguing. Further research is needed to determine the effects of modifying mass, motors location, the length and diameter of the vibratory bar, and the application of other vibration frequencies. Moreover, studies on the effects of long-term training programs on different strength and power measures are warranted. However, a special consideration must be given to the potential adverse effects of long-term exposure to hand-arm vibration. This can cause a variety of disorders collectively known as hand-arm vibration syndrome that has been extensively researched in occupational medicine literature (14). Nevertheless, none of the subjects reported any potential symptoms during vibration exposure at either session.

In conclusion, the results of this study showed that the vibration parameters of Accmax, Freq, and Ampp-p provided by the VB were inside the ranges reported in the scientific literature and seem appropriate for upper-body vibration training. Another finding of this study is the increase in muscle activation in both vibration conditions (25 Hzi and 45 Hzi) compared with no vibration, except during 25Hzi for deltoid and pectoralis muscle, in FP. Thus, the vibration of 45Hzi resulted in the highest vibration effect, but no significant differences were found between vibration frequencies. During the FP vibration frequency should be higher than 25 Hzi to induce significant increases in EMG activity in deltoid and pectoralis muscles.

Practical Applications

A vibratory bar was built to stimulate indirectly and simultaneously upper-body muscles via the hands during a multijoint upper-body exercise. It was found that the use of the VB system can result in a significant increase in upper-body muscle activity compared with the same exercise without vibration. Therefore, strength and conditioning professionals should consider the use of a vibratory bar as an effective method for neuromuscular training, being a more suitable system to simultaneously stimulate upper-body muscles than performing push-ups on a vibration platform. Moreover, gripping a bar allows a better and more comfortable posture for the wrists and makes it easier to perform a greater variety of exercises.

Results from this study suggest that considerable differences could be expected between the real vibration parameters and those provided by the manufacturer. Moreover, it is important to take into consideration the real vibration frequency and amplitude produced along these devices and more concretely at the nearest place of transmission with the feet or hands placement.

Given the little research available about the effects of vibrations on the upper body and the different time course and extent of the training responses in this body area, a call for professionals involved in the health, athletic, and therapeutic sectors is made to determine the potential effects of training programs using a vibratory bar system.


This research was supported by the Commission for Universities and Research of the Department of Innovation, Universities and Enterprises of the Catalan Government, and the European Social Fund. We disclose professional relationships with companies or manufacturers that will benefit from the results of the present study, and we state that the results of the present study do not constitute endorsement of the product by the authors or the NSCA.


1. Abe, T, DeHoyos, DV, Pollock, ML, and Garzarella, L. Time course for strength and muscle thickness changes following upper and lower body resistance training in men and women. Eur J Appl Physiol 81: 174-180, 2000.
2. Abercromby, AF, Amonette, WE, Layne, CS, McFarlin, BK, Hinman, MR, and Paloski, WH. Variation in neuromuscular responses during acute whole-body vibration exercise. Med Sci Sports Exerc 39: 1642-1650, 2007.
3. Bongiovanni, LG, Hagbarth, KE, and Stjernberg, L. Prolonged muscle vibration reducing motor output in maximal voluntary contractions in man. J Physiol 423: 15-26, 1990.
4. Bosco, C, Cardinale, M, and Tsarpela, O. Influence of vibration on mechanical power and electromyogram activity in human arm flexor muscles. Eur J Appl Physiol 79: 306-311, 1999.
5. Cardinale, M and Lim, J. Electromyography activity of vastus lateralis muscle during whole-body vibrations of different frequencies. J Strength Cond Res 17: 621-624, 2003.
6. Cardinale, M, Soiza, RL, Leiper, JB, Gibson, A, and Primrose, WR. Hormonal responses to a single session of whole body vibration exercise in elderly individuals. Br J Sports Med 2008 Apr 15. [Epub ahead of print].
7. Cardinale, M and Wakeling, J. Whole body vibration exercise: Are vibrations good for you? Brit J Sports Med 39: 585-589, 2005.
8. Cochrane, DJ and Hawke, EJ. Effects of acute upper-body vibration on strength and power variables in climbers. J Strength Cond Res 21: 527-531, 2007.
9. Cormie, P, Deane, RS, Triplett, NT, and McBride, JM. Acute effects of whole-body vibration on muscle activity, strength, and power. J Strength Cond Res 20: 257-261, 2006.
10. Cram, JR and Kasman, GS. Introduction to surface electromyography. Gaithersburg, MD: Aspen, 1998.
11. Di Giminiani, R, Tihanyi, J, Safar, S, and Scrimaglio, R. The effects of vibration on explosive and reactive strength when applying individualized vibration frequencies. J Sports Sci 27: 169-177, 2009.
12. Eklund, G and Hagbarth, KE. Normal variability of tonic vibration reflexes in man. Exp Neurol 16: 80-92, 1966.
13. Fratini, A, Cesarelli, M, Bifulco, P, and Romano, M. Relevance of motion artifact in electromyography recordings during vibration treatment. J Electromyogr Kinesiol 19(4): 710-718, 2009. [Epub 2008 May 20].
14. Friden, J. Vibration damage to the hand: clinical presentation, prognosis and length and severity of vibration required. J Hand Surg Br 26: 471-474, 2001.
15. Griffin, MJ. Handbook of Human Vibration. London: Academic Press, 1996.
16. Gusi, N, Raimundo, A, and Leal, A. Low-frequency vibratory exercise reduces the risk of bone fracture more than walking: A randomized controlled trial. BMC Musculoskelet Disord 7: 92, 2006.
17. Hazell, TJ, Jakobi, JM, and Kenno, KA. The effects of whole-body vibration on upper- and lower-body EMG during static and dynamic contractions. Appl Physiol Nutr Metab 32: 1156-1163, 2007.
18. Humphries, B, Warman, G, Purton, J, Doyle, TLA, and Dugan, E. The influence of vibration on muscle activation and rate of force development during maximal isometric contractions. J Sports Sci Med 3: 16-22, 2004.
19. Issurin, VB and Tenenbaum, G. Acute and residual effects of vibratory stimulation on explosive strength in elite and amateur athletes. J Sports Sci 17: 177-182, 1999.
20. Jackson, S and Turner, D. Prolonged muscle vibration reduces maximal voluntary knee extension performance in both the ipsilateral and the contralateral limb in man. Eur J Appl Physiol 88: 380-386, 2003.
21. Jordan, MJ, Norris, SR, Smith, DJ, and Herzog, W. Vibration training: An overview of the area, training consequences, and future considerations. J Strength Cond Res 19: 459-466, 2005.
22. Konrad, P. The ABC of EMG: Practical Introduction to Kinesiological Electromyography. Scottdale, AZ: Noraxon USA, Inc, 2005.
23. Lorenzen, C, Maschette, W, Koh, M, and Wilson, C. Inconsistent use of terminology in whole body vibration exercise research. J Sci Med Sport 2008 Aug 31. [Epub ahead of print].
24. Luo, J, McNamara, BP, and Moran, K. A portable vibrator for muscle performance enhancement by means of direct muscle tendon stimulation. Med Eng Phys 27: 513-522, 2005.
25. Lynch, NA, Metter, EJ, Lindle, RS, Fozard, JL, Tobin, JD, Roy, TA, Fleg, JL, and Hurley, BF. Muscle quality. I. Age-associated differences between arm and leg muscle groups. J Appl Physiol 86: 188-194, 1999.
26. Mahieu, NN, Witvrouw, E, Van de Voorde, D, Michilsens, D, Arbyn, V, and Van den Broecke, W. Improving strength and postural control in young skiers: whole-body vibration versus equivalent resistance training. J Athl Train 41: 286-293, 2006.
27. Maloney-Hinds, C, Petrofsky, JS, and Zimmerman, G. The effect of 30 Hz vs. 50 Hz passive vibration and duration of vibration on skin blood flow in the arm. Med Sci Monit 14: CR112-CR116, 2008.
28. Martin, BJ and Park, H-S. Analysis of the tonic vibration reflex: Influence of vibration variables on motor unit synchronization and fatigue. Eur J App Phys 75: 504-511, 1997.
29. McBride, JM, Porcari, JP, and Scheunke, MD. Effect of vibration during fatiguing resistance exercise on subsequent muscle activity during maximal voluntary isometric contractions. J Strength Cond Res 18: 777-781, 2004.
30. Melnyk, M, Kofler, B, Faist, M, Hodapp, M, and Gollhofer, A. Effect of a whole-body vibration session on knee stability. Int J Sports Med 29 (10): 839-844, 2008.
31. Mester, J, Kleinoder, H, and Yue, Z. Vibration training: Benefits and risks. J Biomech 39: 1056-1065, 2006.
32. Mester, J, Spitzenpfeil, P, and Yue, ZY. Vibration loads: Potential for strength and power development. In: Strength and Power in Sport. Komi, PV, ed. Oxford, UK: Blackwell, 2002. pp. 488-501.
33. Mileva, KN, Naleem, AA, Biswas, SK, Marwood, S, and Bowtell, JL. Acute effects of a vibration-like stimulus during knee extension exercise. Med Sci Sports Exerc 38: 1317-1328, 2006.
34. Moran, K, McNamara, B, and Luo, J. Effect of vibration training in maximal effort (70% 1RM) dynamic bicep curls. Med Sci Sports Exerc 39: 526-533, 2007.
35. Moras, G, Tous, J, Muñoz, CJ, Padullés, JM, and Vallejo, L. Electromyographic response during whole-body vibrations of different frequencies with progressive external loads. Lecturas: EF y Deportes. Revista Digital año10, 2006.
36. Poston, B, Holcomb, WR, Guadagnoli, MA, and Linn, LL. The acute effects of mechanical vibration on power output in the bench press. J Strength Cond Res 21: 199-203, 2007.
37. Rittweger, J, Ehrig, J, Just, K, Mutschelknauss, M, Kirsch, KA, and Felsenberg, D. Oxygen uptake in whole-body vibration exercise: Influence of vibration frequency, amplitude, and external load. Int J Sports Med 23: 428-432, 2002.
38. Roelants, M, Verschueren, SMP, Delecluse, C, Levin, O, and Stijnen, V. Whole-body-vibration-induced increase in leg muscle activity during different squat exercises. J Strength Cond Res 20: 124-129, 2006.
39. Rohmert, W, Wos, H, Norlander, S, and Helbig, R. Effects of vibration on arm and shoulder muscles in three body postures. Eur J Appl Physiol Occup Physiol 59: 243-248, 1989.
40. Rubin, C, Fritton, JC, Pope, M, Magnusson, M, Hansson, T, and McLeod, K. 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, 2003.

vibration exercise; neuromuscular; posture; strength; reflex

© 2010 National Strength and Conditioning Association