ABSTRACT
 

KEY WORDS H-reflex, potentiation
 

INTRODUCTION
 

Whole-body vibration (WBV) is being examined for use in rehabilitation and sport training (6,7,9,23,30–32). Investigators have noted improvements in response to WBV training in such physiological measures as neuromuscular performance (2), force output (10,12,22,33), flexibility (36), and hormone concentrations (2). Not all investigators, however, have noted positive benefits (9,13,14,30), particularly in the short term. Rittweger and co-workers (31) examined the effects of squatting with and without vibration (26 Hz). They concluded that WBV may enhance neuromuscular excitability. This was measured, however, using the patellar reflex, and the duration of the effect was not determined. Still, little is known about the exact physiological adaptations responsible for these changes or what is the most appropriate combination of frequency (of vibration) and amplitude.
 

The Hoffmann reflex (H-reflex) is widely established as a measure of motoneuron excitability (18,20,21,28,29,34). The H-reflex is analogous to the spinal stretch reflex induced via electrical stimulation, thus bypassing the muscle spindle. Progressive stimulation of a percutaneous mixed nerve, such as the tibial nerve, reveals the H-reflex, which appears first on the electromyographic trace (EMG) as the threshold of the Ia afferents is attained, and is followed by the muscle response (M-wave) as the œ-motoneurons (œMNs) reach threshold. The peak-to-peak amplitude of the H-reflex increases until maximum and subsequently diminishes. As more aMNs are recruited, the M-wave increases until it plateaus at M_max. The disappearance of the H-reflex as the M-wave approaches M_max results from the collision of the reflexive action potential traveling down the motoneuron (orthodromic) with the action potential traveling up the motoneuron toward the spinal cord (antidromic), which blocks the H-reflex. For a review of the H-reflex and its application to sports medicine, the reader is referred to Palmieri et al. (29).
 

The H-reflex may be normalized using a stimulus intensity that produces a percentage of M_max (29). Theoretically, the H-reflex represents a given percentage of the total motoneuron (MN) pool (i.e., a peak-to-peak amplitude equal to 30% of M_max represents 30% of the MN pool). This method permits an investigator to measure changes in theMN pool after some intervention. Thus, changes in the peak-to-peak amplitude of the H-wave after a bout ofWBVmay be used to determine the extent to which WBV may affect neuromuscular activity and indicate the degree to which fatigue, recovery, and, possibly, potentiation of the reflex response occur.
 

METHODS
 

Experimental Approach to the Problem
 

The present study is the first in a series of studies involving WBVconducted in our laboratory. Exploratory in design, this study was designed to determine the time course of changes in the H-reflex after a single 1-minute bout of WBV at a moderate frequency (40 Hz) and low amplitude (2–4 mm). In particular, it was hoped that the results would allow the investigators to determine whether random order of treatment (e.g., frequency and amplitude levels) may be assigned and, if so, how much recover time is required between treatments.
 

Subjects
 

Nineteen subjects (8 male and 11 female, mean age 19 ±1 years [range: 18–21 years]) with measurable H-reflexes were recruited from the student population at Hope College by word of mouth. No effort was made to control for training specificity, but current physical activity was self-reported on a medical history questionnaire. Subjects ranged from sedentary to NCAA Division III athletes (football, soccer, swimming, and track sprinters) and were overall heterogeneous. Subjects had no indicated neurological defects, no history of lower extremity surgery, and no lower extremity injury for 12 months before the start of the study as reported by the medical history questionnaire. Informed consent was obtained, and all procedures were approved by the Hope College human subjects review board.
 

Procedures
 

H-reflex and M-wave measurements were collected at a rate of 2000 Hz using surface EMG (MP150; Biopac Systems Inc., Santa Barbara, CA). Signals were amplified (EMG100C; Biopac Systems Inc.) from disposable, 10-mm pregelled Ag- AgCl electrodes (EL503; Biopac Systems Inc.). A stimulator module (STM100C; Biopac Systems Inc.) with a 200-V (maximum) stimulus adaptor (STMISOC; Biopac Systems Inc.) using a disc electrode (EL548; Biopac Systems Inc.) and a 4-cm dispersive pad were used to stimulate the muscle contraction.
 

Subjects reported to the Athletic Training Room at Hope College’s DeVos Fieldhouse after at least 12 hours’ rest from exercise and 12-hour abstention from alcohol, caffeine, and any medication that affects the central nervous system.With the subject lying prone, two areas were abraded with fine sandpaper and cleaned with alcohol for placement of the EMG electrodes. Surface electrodes were placed on the posterior leg, centered on the soleus muscle, approximately 2 cm distal to the medial head of gastrocnemius and spaced 2 cmapart, parallel to the muscle fibers. The ground electrode was placed on the lateral malleolus of the ipsilateral leg.
 

The stimulating electrode was fixed in the medial portion of the popliteal fossa behind the knee over the tibial branch of the sciatic nerve (Figure 1). The corresponding anode (dispersive pad) was placed on the anterior thigh superior to the patella. The stimulating electrodes were then wrapped with 2-inch PowerFlex^TM tape (Andover, MA) to maintain constant pressure on the electrodes. Subjects were subsequently positioned supine with the hands positioned at the sides, knees supported at approximately 10–15 degrees, and the feet secured against a footplate to maintain neutral foot position. Participants wore noise-canceling headphones (Phillips HN110) and listened to ocean sounds to minimize the effects of extraneous sound on the H-reflex. Subjects were asked to remain still, and the same body positioning was used throughout testing.
 

H/M recruitment curves were mapped for the soleus muscle by increasing the stimulus intensity in 0.2- to 1.0-V increments, with a 10-second rest interval between stimuli, until the M_max was obtained. Peak-to-peak amplitudes of the H-reflex and M-wave were determined for all test stimulations. Five measurements of the H-reflex were made using a stimulus sufficient to produce an H-wave 30% of M_max pre- WBV (T0), and a single measurement was made immediately post-WBV (approximately 1 minute post-WBV; T010) and every 30 seconds for 30 minutes. In addition, because previous studies have indicated a conditioning effect of the electrical stimulation on the H-response, three conditioning stimuli (approximately 100 V) were applied at onset of the recruitment curve. H-reflex measurements using the described protocol have been found to be reliable (r = 0.9953, 0.9514, and 0.9747 for H_max, M_max, and H/M ratio, respectively) (28).

 

Subjects stood with the feet shoulder-width apart and the knees flexed approximately 10 degrees on a Next Generation Power Plate (Power Plate North America, Inc., Northbrook, IL) for 1 minute with the frequency and amplitude settings at 40 Hz and 2–4 mm, respectively (Figure 2).
 
 

Statistical Analyses
 

Multiple recovery patterns were observed among the subjects. Therefore, data were separated into four patterns according to the rate at which the subject returned to baseline and whether potentiation of the H-reflex was observed. Comparisons of changes in the H-reflex over time and between groups were made using repeated-measures ANOVA. One-way ANOVA was also used to compare differences between groups and between genders at key time points where notable changes in the H-reflex occurred, i.e., T0, T010, 3-min post-WBV(T030), 15-min post-WBV (T150), and 27.5-min post-WBV (T275), and Bonferroni post hoc comparisons were made to determine where significant differences between groups occurred. In addition, the intraclass correlation coefficient was determined the five pre-WBV H-reflex measurements at the test stimulus intensity. These values were used to determine internal consistency because these were the only multiple measurements taken at a given intensity. The level of significance was set at œ= 0.05.
 

RESULTS
 

Four distinct recovery patterns were observed among the participants (Figure 3). All subjects displayed a significant suppression of the H-reflex during the first minute post-WBV. Recovery to baseline occurred in three of the groups, whereas the fourth group (G4) showed a nearly complete suppression (peak-to-peak amplitude <10% of M_max) for the duration of recovery. One group (G1) returned to baseline within 3 minutes, and the H-reflex subsequently increased to above baseline. Group 2 (G2) returned to and remained at baseline after 7 minutes. Group 3 (G3) was suppressed until approximately 15 minutes post-WBVand gradually increased to baseline by 30 minutes.

 

 

There was a significant effect for time for all groups (P < 0.001) and for group by time (P <0.001). There were no significant differences between genders across time (P = 0.410). There was not, however, sufficient gender representation in all groups to perform time xgroup xgender comparisons. Pre-WBV, G1 displayed a baseline H-reflex significantly higher than G3 and G4 (P = 0.014 and 0.019, respectively) but not significantly different from G2 (P = 0.07). At T010, there was no significant difference between groups on the H-reflex (P >0.34). At T030, there was a significant between-group interaction (P = 0.002). G1 was significantly higher than G3 and G4 (P >0.35), but G2 did not differ from the other groups (P >0.16). At T150 and T275, there was a significant between-group interaction (P , 0.001). At T150, G1 was higher but not significantly different than G2 (P = 0.083); G3 and G4 were similar (P >0.99); and G1 and G2 were significantly higher than G3 and G4 (P < 0.001). At T275, G1 was significantly higher than the other groups (P <0.04), and G2 and G3 were similar (P >0.99) and significantly greater than G4 (P <0.004). Group means for T0, T010, T030, T150, and T275 are reported in Table 1.
 

Intraclass correlations were performed on the five pre- WBV H-reflex measurements at the stimulus test intensity. The observed intraclass correlations value was r = 0.81.
 

DISCUSSION
 

The investigators sought to examine the recovery response of the H-reflex to a 1-minute bout of WBV exercise of moderately high intensity. The most significant finding of the present study is that the response of the H-reflex to a single bout of WBV is highly variable and may be associated with individual differences other than gender, training specificity, or conditioning status. One plausible explanation for these differences is muscle fiber type differences. The present study is limited in that the effects of training cannot be thoroughly eliminated. In addition, without information (i.e., biopsies of muscle samples) about muscle fiber content, these conclusions are speculative at best.
 

It has been reported that acute bouts of cycling or jogging exercise reduce the H-reflex (3–5,15,26,27). deVries and coworkers (15) indicated a ‘‘tranquilizer effect’’ of 20 minutes of moderate aerobic exercise on H/Mratio. These investigators examined H-M recruitment curves in the gastrocnemius of 10 volunteers (2 ‘‘elderly’’ [66 and 80 years] and 8 ‘‘younger’’ [mean age 27.3 ±4.8 years]) for which training levels and muscle composition are not reported. Suppression of the H/M ratio ranged from 6% to 44%. Motl and Dishman (26) reported an attenuated soleus H-reflex 10 minutes after moderate-intensity leg cycling (60% V_ o_2peak) but observed no change in the flexor carpi radialis H-reflex. In a similar study, Motl et al. (27) observed suppressed soleus H-waves 10 and 30 minutes after active and passive cycling. The mechanism by which this occurred is unclear. Bulbulian (3) concluded that endogenous opiods play no role in the suppression of H and M after exercise. No study was found that detailed the pattern of H-reflex response over time after fatiguing exercise.
 

The data in the present study support previous findings that there is a fatiguing effect of WBV and that, comparing intensities for research, a substantial recovery should be allowed between exercise bouts. At least some of the subjects showed a suppressed H-reflex for 30 minutes, whereas others showed some level of potentiation for this duration. In either case, not all subjects were fully recovered with the 30-minute period. Because there was a limit to the recovery time, there is no way to conclude how long a recovery is required. The minimal period of recovery may be anywhere between 3 hours (13) and 24 hours (4).
 

There were at least four clear response patterns revealed in the study. Similar patterns were observed in an earlier experiment in which the researchers studied the H-reflex for 15 minutes after two bouts of the Wingate test (1).
 

The role of training on the H-reflex has been clearly demonstrated (16,24). The differences observed in both the Wingate study (1) and the present study, however, cannot be explained by level or type of training. In the Wingate study (1), the sample size was too small to quantify significant difference in training. In the present study, observed variability was not consistent among individuals with similar training experience (e.g., similar sport team participation, endurance-trained vs. agility/power-trained). The sample in this study was heterogeneous, and no definitive analysis of training effects could be performed. Although there were similarities in the nature of training among some of the subjects, few subjects trained with the same specificity (i.e., whereas soccer goalies and football defensive backs both train for agility, the activities to accomplish this goal are not necessarily the same). Examination of individual differences should be performed using larger, homogeneous groups controlled for training specificity.
 

The soleus is composed largely of type I muscle fibers (25). The posture tested using WBVwould put stress on the soleus muscle to maintain balance, and these data confirm that the soleus was indeed fatigued by the protocol. One plausible interpretation of these data is that the multiple patterns of recovery may be a reflection of muscle fiber content variations among subjects. Postactivation potentiation may be affected by the muscle fiber characteristics (8,11,17–19, 35,37), although this hypothesis was not tested. Comparison of biopsied soleus muscle tissue with H-reflex recovery after fatiguing exercise is warranted. The postactivation response of the H-reflex may be dependent not only on fiber type and the nature of the stimulus (e.g., frequency and intensity), but also on the complex interaction between the muscle fiber fatigue characteristics and the nature of the fatiguing protocol (17,18). In addition, other muscles (e.g., flexor carpi radialis, quadriceps femoris, gastrocnemius) should be tested to compare the response of the H-reflex after vibration and other fatiguing exercise.
 

Intraclass correlations indicate that the H-reflex may have been more variable in the present study than noted by Palmieri and co-workers (28). This is likely the result of the prolonged time the subjects were required to lay motionless. H/M recruitment curves can take longer than 30 minutes, and finding a suitable testing stimulus intensity took more than 5 minutes in some individuals. The subjects were also asked to lie still for an additional 30 minutes post-WBV. Maintaining a stable level of alertness (i.e., keeping the subjects awake, focused, and in the same postural position) proved impossible. In addition, the rhythmicity of stimulus application may have caused some subjects to anticipate the stimulation, thereby affecting the H-reflex. Nevertheless, the frequency of measurements permitted the investigators to note observable patterns in the recovery H-reflex.
 

PRACTICAL APPLICATIONS
 

H-reflex is a viable measure ofMN excitability. After fatiguing exercise, this variable is notably suppressed. The rate at which the H-reflex recovers after an acute bout of WBVwas studied in the present study. The data indicate that the pattern of recovery is variable and may not be attributable to gender or type of training. Recovery from a single bout may be longer for some individuals. Therefore, when comparing the acute effects of WBV at varying intensities, investigators should consider designs that include subjects who are tested on multiple days rather than using designs that, although randomly ordered, may not permit adequate recovery of the motor unit (i.e., utilize short-duration recovery between tests).
 

These data may also indicate that a physiological difference other than training occurs among subjects. This, however, was not quantified in the present study. Future investigation should consider factors such as training specificity and muscle fiber type that might contribute to the differing H-reflex response. In addition, the effect of WBV on specific performance measures (e.g., flexibility, vertical jump, force output, balance) should be addressed and interpreted with the understanding that there may be considerable variability among individuals, and sample size should be adjusted accordingly.
 

All authored papers and editorial news and comments, opinions, findings, conclusions or recommendations in JBMR WebFirst papers are those of the author(s) and do not necessarily reflect the views of the JBMR and its publisher, nor does their publication imply any endorsement. No responsibility is assumed, and responsibility is hereby disclaimed, by the American Society for Bone and Mineral Research and the Journal of Bone and Mineral Research for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of methods, products, instructions or ideas presented in these articles. Independent verification of diagnosis and drug dosages should be made. Discussions, views and recommendations as to medical procedures, choice of drugs and drug dosages are the responsibility of the authors. JBMR WebFirst papers have been peerreviewed, however the articles have not gone through the copyediting process. Papers will not appear in JBMR style and format until the final print and online version is available. The WebFirst publication date is the official date of publication for each paper. There will be minor changes made to the WebFirst paper in the copyediting process, however no scientific content will be changed. The final paper published in the print Journal and on JBMR Online will not change in scientific content, only in presentation and to adhere to JBMR style.
 

Shake, rattle and roll have now entered the osteoporosis quietness with two papers on the clinical effect of whole body vibration in this issue of JBMR. Besides, we can already see these products at scientific meetings. To what extent does the present knowledge address the potential for such approaches in the prevention of fractures

Good vibrations
 

Clinical studies have been based on animal experiments that have shown a positive effect on bone strength and mass of various forms of loading. The basis of these experiments is the concept that trabecular bone adapts to its mechanical environment – Wolff’s law. Support for these experiments has also come from epidemiological findings that greater physical activity – mechanical stimulation is associated with greater bone mass and, in some studies, fewer fractures.
 

The question of the ideal form of stimulation has been addressed in animal studies. High frequency (30 Hz), low magnitude (200 μstrain) signals stimulated large increases in cortical bone in turkeys. However higher amplitude and lower frequency was not anabolic in that model. In a longer term study in sheep over one year, daily 20 minutes sessions of high frequency mechanical stimulation of sheep produced a 35% increase in bone density. This kind of vibration may also affect the sarcopenia that occurs at the same time as bone loss with aging. Other animal studies have shown similar results. Low magnitude mechanical loading became osteogenic when rest is inserted between each load cycle. Effects of loading frequency on mechanically induced bone formation and periosteal osteogenesis suggested a complex interaction between extracellular fluid forces and cellular mechanics in mechanotransduction, best predicted by a mathematical model that assumed: 1) bone cells are activated by fluid shear stresses and 2) that stiffness of the bone cells and the extracellular matrix near the cells increases at higher loading frequencies because of viscoelasticity. These animal experiments have formed the scientific basis for studies in humans.
 

Shakin´ all over
 

In humans extremely low level, high frequency mechanical accelerations have been shown to be readily transmitted into the lower appendicular and axial skeleton of the standing individual. In a recent study, 21 male and 35 female volunteers (age d 19-38 years) were randomly assigned to a vibration or control group. Individuals stood on a vibration platform that was either stationery or oscillated in an ascending order from 25 to 45 Hz, corresponding to maximum vertical accelerations from 2g to 8g intervention for 4 minutes/day, 3-5 times per week over an 8-month period. Although there was no effect on bone mass or serum markers or other performance and balance tests, there was an increase in vertical jump height in the vibration group.
 

In this issue, Verschueren et al report on a 6-month study of whole body vibration in older women with respect to hip density, muscle strength and postural control. The 70 volunteer women, aged 60-70 years, who were healthy and had a BMD T score >-2, were randomized to a control group with no organized training; resistance training knee extensor by dynamic leg press and leg extension exercises or whole body vibration, where the subjects performed the same exercises for 20 minutes per day on a vibration platform that had a vibration frequency of 35-40 Hz and peak acceleration of 2.3-5.1 g. The vibration training improved the isometric and dynamic muscle strength by 15 and 16%, respectively and increased BMD by 0.93%. No hip BMD changed was observed in the resistance training or the age-matched controls. Serum markers of bone turnover did not change in any groups. The authors concluded that whole body vibration training might be a feasible and effective way to modify well recognized risk factors for falls and fractures in elderly women.
 

Also in this issue, Rubin et al, report another trial for 1 year in 70 healthy women, 3-8 years post menopause (mean age 57 years). Those randomized to the vibration platform were exposed to a peak vertebral acceleration of 0.2 g at a frequency of 30 Hz. Compliance was not good as in many other exercise (and pharmacological) interventions and the intention-to-treat analysis did not show an effect. In an analysis limited to those in the highest quartile of compliance (86% compliant), vibration subjects gained 0.04% in femoral neck BMD while placebo subjects lost 2.13% over the year in the femoral neck. The corresponding figures for the lumbar spine were +0.1 and –1.6%. Interestingly, the lower body weight (<65 Kg) women experienced the greatest benefit, 3.4% increase in the highest compliance group and 2.7% in the mean compliance group. The authors conclude that these preliminary results indicate a potential for a non-invasive mechanical mediated intervention for osteoporosis that is perhaps more effective in lighter women who are at greatest need of intervention.

How shall interpret these trials? First of all there are differences in age. The study by Torvinen al had younger and perhaps healthier participants. The greater benefit in lighter individuals in the Rubin study could explain some of these differences. Although there were differences in study duration these overlapped and do not seem to explain any of the differences reported. The Torvinen study used a short exposure period (4 minutes) for each treatment and somewhat greater loads although at similar frequency.(7) None of the studies showed any differences in bone turnover markers but there were observable differences in muscle strength e.g. jump height. These possibilities require examination in further studies with respect to study sample age and weight as well as vibration exposure and amplitude.
 

Any side effects?
 

Vibration of the human body has been proposed from epidemiological studies to cause back pain. However no such major side effects were reported from these studies and whole body vibration exercise has been proposed for treatment of chronic low back pain.(10) Another possible safety aspect is that the displacement could be large enough for the patient to fall but this was not reported in these studies.

What could be the biological mechanisms of this whole body vibration?
 

The vibration is sufficiently low to be unappreciable by the participants so it seems unlikely to be a direct effect of the mechanical strain. It could be an indirect effect through amplifying of signals by intramedullary pressure or through fluid flow in the bone tissue. For the neuromuscular or muscular effects, stimulation of the skeletal muscular pump has also been proposed to affect circulatory flows and flow through the bone tissue. However these potential mechanisms are still to be fully studied.
 

How do these effects compare with published studies on pharmacological interventions?
 

Leaving aside any potential muscle or balance effects, the net benefit versus placebo ranged from 1.55% to 2.2% and up to 3.4% in those of lower weight and best compliance. These effects over 1 year are difficult to compare with pharmacological studies over 2-3 years but in a bisphosphonate study with a 1-year endpoint, the difference from placebo was 2.4%. This comparison might suggest somewhat similar benefit provided good compliance could be achieved. However it is recognised that change in BMD cannot be easily translated to fracture reduction. Thus the burning question is what fracture reduction could be achieved with whole body vibration.
 

Future
 

A tantalizing possibility is that there could an interaction between whole body vibration and pharmacological treatment. Could whole body vibration enhance the effect of an anabolic agent or an antiresorptive? In one study in rat tail, there was a synergistic effect of PTH and mechanical stimulation on trabecular bone formation. It remains to be seen whether similar interactions could be seen in humans, where no major effect on bone turnover from whole body vibration has been observed. A further harder development in the future might be shock wave treatment, in animals shown to be positive to bone mass in fractured limbs.
 

What's Shakin'?
 

What are the requirements to bring this equipment to the market place? Vibration platforms are regarded as ‘devices’ and not a pharmaceutical intervention and hence are subject to different regulatory criteria for safety and efficacy. Hence for considerations of clinical application, it is important to determine what kinds of data are needed to support vibration as a valid and rational treatment option. Should BMD change be sufficient or should we require fracture reduction data ? Is analysis by compliance reasonable in light of our judgements about other randomized placebo controlled trials where ITT is and should remain the gold standard? Although vibration platforms seem to be relatively safe, it will be important to establish their anti-fracture and BMD efficacy as well as their safety in larger and more adequately powered randomized double blinded controlled trials.
 

ABSTRACT.
 

KEY WORDS. strength, power, vertical jump, force-velocity, ballerina
 

INTRODUCTION
 

Ballet is considered mainly as an aesthetic activity (17, 22). Recent findings, however, have shown dance to be a physically demanding exercise mode (17). Due to its highly complex, multidirectional movement requirements, dance may be considered as a high-intensity intermittent exercise (26)
 

Body composition, joint mobility, cardiovascular fitness, and muscular strength have been reported as limiting factors in ballet performance (19). Dance requires quick bursts of energy interspersed with low-intensity activities (9), and strength level is an important component of ballet performance (19, 27). Additionally, strength has been suggested as an effective strategy for injury prevention in ballet (16, 22). Nevertheless, dancers see themselves as artists rather than athletes and only rarely undertake structured physical conditioning, particularly strength training (16, 17, 19, 27, 32, 33).
 

The few studies that have addressed the issue of strength training in ballet reported significant improvements from pretraining strength levels, without any detrimental effect on actual and perceived physical appearance (19, 27).
 

Despite the reported encouraging effect of strength training on ballet performance, the protocols that have been suggested are time-consuming and require moderate to high skill to be performed. Furthermore, they necessitate training facilities that are very often difficult to implement in a ballet academy due to economic or logistical reasons. Consequently, these protocols, usually involving free-weight exercises or weight-training machines, may have little applicability to the ballet population.
 

Recently, whole body vibrations (WBV) have been proposed as a training intervention to develop strength and flexibility (7, 15). The major documented benefit of WBV is explosive strength performance, usually considered as vertical jump height (2, 12, 29–31). This result is of particular interest, because vertical jump performance is considered to be related to ballet performance (19, 27). Additionally, WBV training protocols require only a limited amount of time and very limited, if any, specific skills to be performed (7).
 

The aim of this study was to examine the effects of a short-term WBV training intervention on dynamic strength and vertical jump performance in a selected population of elite-level classical ballerinas. It was hypothesized that WBV intervention may result in explosive strength and dynamic strength improvements in professional ballerinas if provided as a supplement to the usual dance training routine.
 

METHODS
 

Experimental Approach to the Problem
 

The strength training protocols that used ballerinas as subjects considered weight-machine and free-weight exercises (19, 27). Furthermore, the strength training protocols involved progressive loads exceeding 70% of 1 repetition maximum (1RM) during 3–4 exercises in 5–6 sets with up to 8 repetitions each (19, 27). Although these training protocols are widely used by athletes of different sports, they require proper facilities, good weight-training apparatus, and professional supervision (10). Additionally, those successful protocols require extensive resistance training over 50–90 minutes (19), an amount of time that may be intolerable given the often-busy dance practice routines of professional dancers. Moreover, strength training is felt by ballerinas to have potentially detrimental effect on actual and perceived physical appearance (19, 27).
 

With respect to this, WBV may grant several advantages (7). In fact, WBV has been reported to exert a positive effect on both explosive and maximal strength, requiring only a small amount of time per session (2, 3, 12, 30, 31). Furthermore, the WBV training exercise needs very limited familiarization and supervision (12, 30, 31) and can be performed without frequenting well-equipped weight training facilities. Additionally, there is no need to progressively adjust training loads with each testing session, being time of treatment an effective way to implement increments into training stimulus (30, 31). Finally, unlike the reported successful strength protocols (19, 27), WBV does not involve exercising to failure, a condition that may exacerbate the likelihood of a pernicious condition of overreaching that, over a long training period, may lead to an overtraining status (18).
 

This study used a randomized, fully controlled experimental design to examine the effect of WBV on a group of classical ballerinas. Ballerinas’ countermovement vertical jump height (CMJ) and maximal strength and power at selected loads (50, 70, and 100 kg) were considered as dependent variables in this study. These variables had not been addressed previously in training studies despite being considered as limiting factors in ballet performance (17, 19, 27).
 

Recent research has shown that dance training routines provide few stimuli for physical performance improvement, suggesting the need for supplemental conditioning (17, 19, 27, 32, 33). However, usual ballet training routines involve plyometric-type exercises (17, 27) that may improve jumping ability as a synergistic effect of ballet practice. In this regard, Newton et al. (21) showed that if proper training stimuli are provided, vertical jump performance may be improved further in female athletes who are already highly trained in the jump. In this study we used WBV as a training stimulus for inducing improvement in vertical jump and dynamic strength, because such training has been shown to enhance this neuromuscular performance domain (12, 23).
 

The working hypothesis was that in non–strengthtrained but potentially highly trained individuals who jump, WBV may induce significant improvements in various aspects of functional strength performance using nonspecific movements (2, 7, 20).
 

 

Subjects
 

Twenty-two well-trained ballerinas (age, 21.25 ±1.25 years; body mass, 50.8 ±3.7 kg; height, 165.7 ±5.6 cm), full-time students of the National Ballet School, volunteered for the present study. At the time of the investigation, all of the subjects had at least 8 years of dance experience and no previous history of fractures or musclebone injuries. In order to be included in the study, participants had to possess official medical clearance at the beginning of the season, according to the law. None of the subjects were active smokers or suffered nutritional disturbances, and no medication or drug that would have been expected to affect physical performance was being taken by the ballerinas during the course of this investigation. Furthermore, none of the ballerinas had been involved in strength or explosive-power training prior to the commencement of the training study.

 

 

Ballerinas were assigned randomly into an experimental group (E, n =11) or control group (C, n =11). Throughout the 8-week training intervention, the E and C groups undertook the same amount of ballet training (5 session per week), equated in terms of intensity and volume. Ballet training was performed 5 times a week (Monday–Friday), each session lasting 60–90 minutes. The typical training session consisted of barre technical exercises, center choreography exercises, and ballet jump exercises.
 

Unlike the C group, the E group also undertook a specifically designed WBV training 3 times per week. No other form of conditioning, apart from those considered for this study, were allowed or were performed by participants during the study duration.
 

Full advisement was given to the volunteers regarding the possible risks and discomfort that might be associated with the testing and training procedures used in this study. Prior to the training study and after a detailed written and verbal explanation as to the nature of the procedure involved in this study, all the subjects gave their written informed consent. The study was approved by the Institutional Review Board of the Italian Society of Sport Science. Table 1 presents the physical characteristics of the subjects.
 

Testing Procedures
 

Testing procedures were administered at the beginning and the end of the training intervention using the same order. Posttraining assessments were accomplished 3 days after the last training session to avoid the acute effects of WBV and to allow recovery in both groups. Prior to the commencement of the training study, subjects became accustomed to the testing and training procedures with 2 familiarization sessions that took place during the preceding week. Familiarization procedures were performed on an individual basis until subjects appeared to be fully accustomed to the testing and training procedures involved with this study. The first 2 authors of this study supervised the familiarization sessions.
 

Before each testing session, the ballerinas’ body mass was measured to the nearest 0.5 kg (Seca Beam Balance 710, Hamburg, Germany) with subjects lightly dressed and barefooted, and standing height was measured to nearest 0.5 cm (Seca Stadiometer 208). After body mass and body height were measured, ballerinas undertook a standardized general warm-up protocol that preceded the jumping and mechanical power measurements. The general warm-up comprised stationary cycling (5 minutes) on a computerized cycle ergometer (Technogym, Gambettola, Italy) followed by 5 minutes of static stretching of quadriceps and calf muscles.
 

Countermovement Jump Height Testing
 

Countermovement jump height was assessed with subjects performing maximal vertical jumps (with hands on hips) on a switch mat connected to a computer (Ergojump; Boscosystem, Rieti, Italy), according to the procedures reported by Bosco et al. (6). Prior to testing, each subject performed 3 submaximal CMJs with 1 minute of recovery between repetitions, followed by 3 maximal CMJs paced with similar recovery time. During CMJs, knee flexion was standardized, allowing subject to bend their knee at approximately 90°. According to Bosco et al. (6), the rise of center of mass after the takeoff (h) was measured applying the following ballistic formula:
 

h = tf² g 8^-1,
 

where g is the gravity pool constant (9.81 m·s^-2). The best h was used for statistical analysis. The reproducibility of the CMJ testing procedure here used has been reported to be high
 

Mechanical Power Measurements
 

Following CMJ testing, the mechanical power of subjects’ knee extensors was assessed using a horizontal leg-press machine (Technogym, Gambettola, Italy). After a brief, specific warm-up of approximately 10 minutes, consisting of 3 sets with increasing loads (40, 60, and 70 kg with 10, 6, and 3 repetitions, respectively) separated by 2–3 minutes of passive recovery, each subject performed maximal dynamic leg-press exercises with 50, 70, and 100 kg until exhaustion. During leg-press exercises, knee flexion was standardized at 90° and only the repetitions that matched the selected angular displacement were used for calculations. Knee flexion was monitored using an electronic goniometer (Muscle-Lab, Ergotest Technology, Langesund, Norway) worn by subjects on their nondominant leg and interfaced with a laptop computer. In order to improve measurement reliability, horizontal leg-press seat position was modified according to subjects’ lower limb length and was recorded for posttraining testing reconsideration. Magnitude and rate of load displacement were assessed with a computerized system (Muscle-Lab, Ergotest Technology, Langesund, Norway) according to Bosco and colleagues’ (1) procedures. As suggested by Bosco et al. (1), average knee-extensor power was considered as a variable representing lower limb mechanical power.
 

Subjects performed 3 trials of each load and the best measures were considered for calculations.
 

The reliability of this testing procedures and measurements reported as an intraclass correlation coefficient has been reported to be r =0.95
 

Vibration Treatment Procedures
 

Subjects were exposed to vertical sinusoidal WBVs using the Nemes LC device (Boscosystem, Rieti, Italy). Vibration frequency was set at 30 Hz according to Cardinale et al. (8) (5-cm displacement; magnitude =5 g) to maximize WBV training effects. During each training session, the subjects in the E group were exposed to WBV treatment before their usual ballet lesson with C group. Whole body vibration treatment consisted of five 40-second repetitions separated by 60 seconds of passive recovery and was provided 3 times a week (Monday, Wednesday, and Friday). The total training intervention comprised 24 WBV sessions over 8 weeks. Whole body vibrations were applied with the subject standing on the vibration platform in a half-squat position (approximately 100°) with feet and knee rotated externally (i.e., demi-plie´ position). In order to avoid bruises, E group ballerinas wore dancertype shoes throughout the WBV session.
 

Data are shown as mean and standard deviation. Paired t-tests were used for within-group comparisons. The level of significance was set at p <=0.05. Unpaired t-tests with Bonferroni adjustment were used for between-group comparisons, with a resulting p value of 0.025. In order to assess meaningfulness of pre- to posttraining changes, the effect size (ES) was calculated according to Thomas et al. (28). Effect sizes of 0.8 or greater, around 0.5, and 0.2 or less were considered as large, moderate, and small, respectively (28).
 

RESULTS
 

The E and C groups differed at pretraining only for average velocity (AV) and average power (AP) during the 50-kg condition (p >0.05). The E group showed significant CMJ performance improvement after training (p < 0.001) (Table 2). Effect size for CMJ was medium (0.67).
 

Results of the mechanical variables of interest are showed in Table 3. The E group showed significant (p < 0.05–0.001) pre- to posttraining improvements in all the mechanical parameters (average force, AP, and AV) for the 70- and 100-kg conditions. Effect sizes ranged from medium to large (0.33–1.8). In the 50-kg condition, only AP and AV showed pre- to postintervention improvements in the E group (0.89 and 1.00, respectively). These results shifted the force-velocity and power-velocity curves to the right (Figure 1). No significant change in any of the mechanical variables was observed in the C group.
 

DISCUSSION
 

This is the first training study that investigated the effects of WBV on vertical jump and lower limb dynamic muscular strength and power in elite-level ballerinas.
 

The main finding of this investigation is that WBV training positively affected vertical jump performance in well-trained elite ballerinas. This result is of particular interest, because vertical jump performance is considered to be related to ballet performance (19, 27). The pre- to postintervention improvement in CMJ performance found in this study is similar (6.3 vs. 7.6%) to that reported by Delecluse et al. (12) for age-matched, untrained female subjects submitted to WBV treatment. However, in the Delecluse et al. (12) study, the WBV treatment was undertaken for 12 weeks using higher WBV frequencies (35–40 vs. 30 Hz) but with a similar vibration platform displacement (5 mm) and magnitude (5 g). Furthermore, WBV training loads were adjusted (vibration frequency, repetition number, and duration) over the intervention duration in order to provide progressive overloads (12).

 

 

The few studies that addressed the issue of strength training in ballerinas reported significant improvement in isometric and isokinetic strength (19, 27). Although those studies analyzed the effect of free-weight and weight-training machines on ballet performance using repeated-jumping ballet simulation protocols, they did not study CMJ performance. In this regard, Delecluse et al. (12) reported no CMJ improvement in a population of untrained young female subjects using strength-training protocols similar to those used for ballerina training studies (19, 27).
 

Providing 40-Hz vibrations during squat training (6– 10RM), Rønnestad (24) reported significant improvement in CMJ (9%) and squat 1RM performance (32.4%) as a consequence of a 5-week WBV training in male resistance- trained subjects. Interestingly, the control group, which comprised resistance-trained individuals who performed the same strength training protocol without added vibration stimuli, reported no significant improvements in CMJ performance.
 

In light of these considerations, it may be argued that WBV or explosive strength training (14) should be considered for ballerinas in order to improve vertical leap performance. However, transfer of CMJ height improvements to dance performance should be assessed with sound performance analysis.
 

In this study, the WBV training determined significant improvement of AP. The major effects of WBV were observed at the 100-kg load with an average AP improvement of 25% (ES =0.62). This AP improvement was due mainly to a concomitant AV increment of 26% (ES =1.0).
 

As a consequence of WBV training, a general shift of force and power profiles to the right was observed (see Figure 1). Training-induced improvements (i.e., shift to the right) of force-power profiles (11, 13) may be caused by several factors such as motor unit synchronization, cocontraction of synergistic muscle, or increased inhibition of the antagonist muscles (25). From this we can infer that WBV training probably was effective in causing an alteration of the neural traffic that regulates muscle stiffness (2, 7). Interestingly, WBV intervention induced force-power profile improvements similar to those reported in studies involving heavy-load strength training (11, 13).
 

According to Bosco et al. (4), stimuli duration seems to be important in inducing neuromuscular adaptation. In the present study, the total duration of WBV intervention was short (only 24 minutes), with a constant gravitational perturbation of 5 g. This means that in order to induce similar training loads using leg-press or half-squat exercises, 150 repetitions with extra loads of approximately 3 times body mass should be provided over the same training period (4) twice a week.
 

The mechanisms mediating the effects of WBV on the neuromuscular system are not completely understood (7, 20). However, current knowledge of WBV effects determinism are considered to be related to neuromuscular, hormonal mechanisms, or both (5, 7). In the present study, no neural extra excitatory inflow or hormonal changes have been demonstrated, because neither electromyographic recordings nor blood samples were collected. Nevertheless, enhancement of mechanical power in leg press performance and the improvement in the CMJ height strongly suggest that a neural adaptation occurred in response to the vibration training in the subjects of the experimental group.
 

These findings are of particular interest because maximal anaerobic power is highly involved in ballet performance (17). However, as suggested for CMJ, the actual effects of AP and AV improvements on ballet performance should be investigated with proper performance analysis.
 

The successful WBV protocol used in this training study was devised specifically to be as nonintrusive as possible to the usual dance routine. However, use of this training protocol obtained very limited improvements in dynamic strength in the E group at the selected external loads (1–3%, p <0.05). It can be speculated that if periodized and progressive WBV training programs (12) are used, further improvements in strength, explosive strength, and CMJ performance could be achieved.

 

PRACTICAL APPLICATIONS
 

According to previous research (18, 19, 27), this study reported that ballet practice is ineffective in promoting significant strength improvements in well-trained ballerinas.
 

Whole body vibration training may be considered as an effective and safe training strategy for improving muscular power in well-trained ballerinas. In this regard, a 30-Hz vibration frequency should be used for inducing muscular adaptation throughout the muscular strengthpower spectrum. Whole body vibration stimuli should be provided in the form of 3–5 repetitions of 40 seconds. Recovery time between repetitions should be no less than 60 seconds.
 

In this study, no significant (p >0.05) variations in body mass were found to result from the WBV treatment. Furthermore, the WBV group reported a perceived increase in physical appearance and technical efficacy during ballet performance (a multi-response questionnaire was submitted at the end of the training intervention). Whole body vibration training studies that used similar or even higher training loads for a longer duration also reported no variation in body mass (12, 30, 31). Although this study did not provide detailed information about the variations in body composition that may have occurred as consequence of WBV, it can be suggested that WBV might increase muscular performance without significantly affecting body appearance. Finally, it is difficult to say if increasing dynamic and explosive muscular strength can increase ballet performance, but it can help the dancers be safer and prevent muscle injures related to their highly intense plyometric activity.
 

Abstract

Objective:
The aim of this study was to investigate the effectiveness of a 6-week traditional exercise program with supplementary whole-body vibration (WBV) in improving health status, physical functioning, and main symptoms of fibromyalgia (FM) in women with FM.

Methods:
Thirty-six (36) women with FM (mean ± standard error of the mean age 55.97 ±1.55) were randomized into 3 treatment groups: exercise and vibration (EVG), exercise (EG), and control (CG). Exercise therapy, consisting of aerobic activities, stretching, and relaxation techniques, was performed twice a week (90 min/day). Following each exercise session, the EVG underwent a protocol with WBV, whereas the EG performed the same protocol without vibratory stimulus. The Fibromyalgia Impact Questionnaire (FIQ) was administered at baseline and 6 weeks following the initiation of the treatments. Estimates of pain, fatigue, stiffness, and depression were also reported using the visual analogue scale.

Results:
A significant 3 x2 (group xtime)-repeated measures analysis of variance interaction was found for pain (p =0.018) and fatigue (p =0.002) but not for FIQ (p =0.069), stiffness (p =0.142), or depression (p = 0.654). Pain and fatigue scores were significantly reduced from baseline in the EVG, but not in the EG or CG. In addition, the EVG showed significantly lower pain and fatigue scores at week 6 compared to the CG, whereas no significant differences were found between the EG and CG (p >0.05).

Conclusion:
Results suggest that a 6-week traditional exercise program with supplementary WBV safely reduces pain and fatigue, whereas exercise alone fails to induce improvements.

Introduction

Fibromyalgia (FM) syndrome is considered a chronic rheumatic condition of unknown etiology characterized by widespread noninflammatory musculoskeletal pain with tenderness on palpation in a minimum of 11 of the 18 tender points for at least 3 months.1 Numerous symptoms may also be associated with this syndrome, including fatigue, anxiety, depression, nonrestorative sleep, muscular stiffness, or irritable bowel syndrome.

The management of FM is based on symptomatic multidisciplinary treatment through pharmacologic and nonpharmacologic strategies. Among all nonpharmacologic treatments, exercise, cognitive behavioral therapy, and education have the strongest evidence for efficacy.2 Numerous researchers support the benefits of aerobic exercise,2–7 strength training,2,6,8 health education,3,9 and relaxation techniques7 on health status, physical functioning, and main symptoms of FM.

Whole-body vibration (WBV) is a mode of exercise that has recently been utilized to improve muscle strength, bone density, and balance in healthy adults10 and aging populations.11,12 In this therapy, the subject performs exercises on a platform that generates vertical vibrations with a frequency and amplitude of 20–50 Hz and 2.0–10.5 mm, respectively.

The rationale for investigating the effects of WBV in patients with FM is based on the following evidence. First, vibratory stimulation has been shown to induce pain relief in acute and chronic pain.13 However, this was a local vibration applied to the forearm, face, and skull rather than a WBV, effects that have not been studied in patients with FM. Second, WBV has been shown to increase strength in untrained14 and postmenopausal women.15 Third, some studies have reported increased growth hormone secretion with WBV.16,17 It is possible that this WBV-induced endocrine effect could enhance the effectiveness of the exercise therapy.18 Therefore, it is reasonable to expect that patients with FM could benefit from the salutary effects of WBV by improving health status, physical functioning, and main symptoms of FM. However, the effects of WBV on patients with FM have not been previously studied.

The purpose of this study was to investigate the effectiveness of a 6-week traditional exercise program with supplementary WBV in improving health status, physical functioning, and main symptoms of FM in women with FM. It was hypothesized that women with FM undergoing the traditional exercise program with supplementary WBV would improve health status, physical functioning, and main symptoms of FM more so than women undergoing exercise only. We considered pain, fatigue, stiffness, and depression as the main symptoms of FM.

Patients and Methods

Subjects

Participants were recruited by referral from family physicians and through public announcements distributed in local associations of FM in Barcelona (Spain). One hundred and four (104) women were interested in the study, and those with diagnosis of FM, according to the American College of Rheumatology criteria,1 for at least 3 years were considered for the study. Women were excluded if they had any orthopedic limitation, cardiovascular, pulmonary, or metabolic disease that would preclude exercise, or when participating in any other study (Fig. 1). Written informed consent was obtained from each subject prior to participation in the study according to procedures approved by the Committee on Biomedical Ethics of the Jordi Gol Gurina Foundation (Spain).

Study Design

A 2-factor mixed experimental design was employed in this study. Women were randomized into 3 treatment groups: exercise and vibration (EVG), exercise (EG), and control (CG) (Fig. 1). Measurements were taken at baseline and at 6 weeks following the initiation of the treatments. To minimize the residual effects of the last bout of exercise, the sixthweek questionnaire was administered 48 hours following the last session of the treatment. Prior to the initiation of the study, an individualized interview was performed with each patient in order to obtain the following information: (1) complete medical history, (2) current medications, (3) physical activity habits, (4) preferred types of exercises, and (5) socioeconomic status. In addition, patients were instructed to report any changes in medication regimens.

Self-report Health Status, Physical Functioning, and Main Symptoms of FM

Health status, physical functioning, and main symptoms of FM were assessed using the validated Spanish version of the Fibromyalgia Impact Questionnaire (FIQ).19 The FIQ is a valid, reliable, and sensitive tool19–22 widely used in research and clinical settings. This questionnaire measures physical function (activities of daily living), work, and general well-being. It contains 100-mm visual analogue scales (VAS) for pain, sleep, fatigue, stiffness, anxiety, and depression.19,20 An overall FIQ score is obtained by normalization of physical function, work, and well-being into a 0–10 scale, which are added to the total scores of VAS. We chose to singly document VAS scores for pain, fatigue, stiffness, and depression, as these are among the most common symptoms related to FM.25 For all our dependent variables, a higher score indicates a greater level of difficulty or illness. Questionnaires were administered to all patients on the same day, time, and by the same investigator to ensure equal instructions and thus minimize potential confounding variables. In addition, administration and analysis of the questionnaires were performed by an investigator who was blind to the treatment groups.

 

Both EVG and EG underwent the same traditional exercise program twice per week for a total of 6 weeks. Following each traditional exercise session, EVG and EG were separated into 2 different rooms. EVG underwent a protocol with WBV, whereas the EG performed the same protocol without vibratory stimulus. The vibratory apparatus in the EG was turned on yet did not produce vibrations, which is a strategy that has been utilized by other authors.13 We informed both EVG and EG that they would receive a perceptible and imperceptible vibratory stimulus, respectively, thus maintaining the potential of a placebo effect consistent in both groups. Women in the CG performed neither the traditional exercise program nor the protocol with WBV. All 3 groups were instructed to continue with their pharmacologic care.2 All exercise sessions were conducted by the same instructor, who had experience in working with FM. The protocol with WBV was supervised by experienced investigators. The 2 per week frequency of sessions was adopted to avoid the risk of exacerbating symptoms and to ensure adherence.

Although our traditional exercise program was similar to that from other studies,3–5,7,9 the inclusion of WBV in patients with FM is the unique aspect of this investigation. Six (6) weeks was chosen as the duration of our intervention because, based on previous research, this is sufficient time for WBV-induced adaptations to occur26–28 while inadequate duration for a traditional exercise program to improve pain, fatigue, stiffness, and depression in patients with FM.3,7 This approach allowed us to investigate the impact of WBV in enhancing the effectiveness of a traditional exercise program.

Traditional Exercise Program

The traditional exercise program was designed using the recommendations of Jones and Clark.29 The program consisted of 15 minutes of a warmup, 30 minutes of aerobic exercise, 25 minutes of stretching exercises, and 20 minutes of relaxation. Major emphasis was given to aerobic exercise because this is the most beneficial type of exercise for the population investigated.2 Aerobic activities consisted of 30 minutes of walking on flat ground at an intensity of 65%–85% of the theoretical maximal heart rate (220 – age (years)). The intensity was monitored by instructing the patients to take their pulse rate during the exercises. For adaptation purposes, during the first week, target heart rate was set at the lower range. Games and dances with low impact were presented at weeks 3 and 5 to avoid monotony. Patients danced salsa music for 30 minutes, performing two repetitions of 15 minutes each with a recovery of 3 minutes in between. Patients performed dancing tasks with a partner, whereas dancing games were performed in groups of 3–10 patients. Stretching exercises consisted of 5 different static wholebody stretches. Patients performed 5 repetitions of each whole-body stretch, holding the position for 30 seconds with a recovery of 30 seconds in between. Stretches involved (a) hamstrings, calves, Achilles tendons, back, and shoulders (downward-facing dog stretch); (b) gluteals, lower back, upper back, shoulders, arms, and chest (lying spinal twist stretch); (c) hamstrings, calves, shoulders, chest, and arms (forward-bend shoulder stretch); (d) Achilles tendons, calves, hamstrings, adductors, cervical spine, shoulders, and forearms (lying supine with open and straight legs with 90° of hip flexion and 100° of shoulder abduction with extended wrists); and (e) Achilles tendons, calves, hamstrings, back, and shoulders (sit-and-reach stretch). Stretches were individualized to teach each patient to locate her stop point and avoid overstretching. Relaxation exercises included diaphragmatic respiration, progressive muscular relaxation, contraction–relaxation, and imagery techniques. The total duration (over 12 sessions) of aerobic exercise, stretching, and relaxation was 9 hours, 6 hours, and 4 hours, respectively.

Whole-Body Vibration

The protocol, given to both EVG and EG, consisted of static and dynamic tasks while standing on a WBV platform (PowerPlate, ® Power Plate International B.V., Badhoevendorp, The Netherlands). Tasks mainly involved lower extremities and included (a) static squat at 100° of knee flexion; (b) dynamic squat between 90° and 130° of knee flexion; (c) maintained ankle plantar-flexion with legs in extension; (d) flexo-extension of the right leg between 100° and 130° of knee flexion; (e) flexo-extension of the left leg between 100° and 130° of knee flexion; and (f) squat at 100° of knee flexion shifting the body weight from 1 leg to the other. For all tasks, subjects held onto the supporting bar. The 6 exercises (30 seconds each) were repeated 6 times with a recovery of 3 minutes between repetitions. For adaptation purposes, only tasks (a), (b), and (c) (repeated 3 times) were performed during the first 2 sessions.

For the EVG, the WBV intensity was kept constant at 30 Hz of frequency and 2 mm of amplitude; whereas for the EG, the apparatus did not produce vibrations. The intensity of vibration was chosen based on previous literature. Thirty (30) Hz has been shown to induce maximal muscular electrical activity. 30 Lower frequencies (i.e., 20 Hz) were not used because they evoke muscular relaxation, whereas higher frequencies (i.e., 50 Hz) were not employed because they can generate severe soreness in untrained individuals.10,31 The duration of WBV was 4.5 minutes per session for the first 2 sessions, and 18 minutes for the remaining 10 sessions. Thus, the total duration after completion of 6 weeks (12 sessions) was 189 minutes.

Statistical Analysis

Descriptive statistics were used to summarize the demographic characteristics of the subjects and one-way analysis of variance (ANOVA) was used to compare the demographic variables among the 3 groups. To test the effects of the treat- ments on FIQ, pain, fatigue, stiffness, and depression, a 3 x 2 (group xtime) mixed-design repeated-measures ANOVA was performed for each dependent variable. Simple main effects were evaluated if a significant interaction was elicited. Tukey’s Honestly Significant Differences (HSD) procedure was used when a significant F-ratio with more than 1 degree of freedom was found. All data are presented as mean ± standard error of the mean). For all statistical tests, the (alpha) level was set at 0.05. Statistical analyses were performed with SPSS v.15.0. (SPSS, Inc. Chicago, IL).

 

Results

Of 104 women, 36 were recruited for this study, and randomly and evenly distributed to 3 treatment groups (Fig. 1). One (1) subject from the EVG and 2 from the CG did not complete the 6-week questionnaire due to a no-show on testing day. Since repeated-measures analysis only allows for inclusion of those subjects who completed all measurements, those subjects with missing values were excluded from the analysis. The demographics of the subjects in each group are summarized in Table 1. There were no differences (p >0.05) among the groups for any of the demographic variables.

Both experimental groups, EVG and EG, adhered to the treatments 93% and 92%, respectively (100% adherence =all sessions attended). None of the subjects dropped out of the study. All patients reported they maintained the same medication prescription throughout the study (Table 2). This program neither exacerbated FM-related symptoms nor resulted in musculoskeletal injuries; however, 1 patient exhibited a mild anxiety attack on the first session of WBV. This patient responded normally for the remainder of the sessions.

Figures 2 and 3 illustrate the change in FIQ and pain, fatigue, stiffness, and depression, respectively, across time and treatment groups. As shown, significant 3 x2 (group xtime) repeated-measures ANOVA interactions were found for pain (F(2,30) =4.60; p =0.018) and fatigue (F(2,30) = 7.50; p = 0.002) but not for FIQ (F(2,30) = 2.93; p = 0.069), stiffness (F(2,30) = 2.09; p =0.142), or depression (F(2,30) = 0.43; p = 0.654). When examining the simple main effects of the significant interactions, pain and fatigue scores were significantly reduced from baseline in the EVG, but not in the EG or CG (Fig. 3). In addition, pain and fatigue scores after 6 weeks of treatment were significantly lower in the EVG compared to the EG and CG; the differences between the EG and CG, however, were not statistically different (p >0.05) (Fig. 3).

 

Discussion

The purpose of this study was to investigate the effectiveness of a 6-week traditional exercise program with supplementary WBV in improving health status, physical functioning, and main symptoms of FM. Although WBV has been shown to produce positive results in a variety of populations, this is the first study to investigate the effects of WBV on women with FM. We found that a 6-week traditional exercise program with supplementary WBV safely reduced pain and fatigue, whereas the effect of exercise alone was not evident in any of the parameters.

The characteristics of our population are similar to these reported in other FM studies. The mean baseline FIQ score in our patients is slightly higher than that reported in several FM studies, but similar to others.6,7 Importantly, our 0% dropout and 92%–93% treatment adherence rates are highly unusual in interventions including patients with FM, and distinguish this investigation from other FM studies.Because subjects were not reimbursed, this high adherence rate may be suggestive of patient’s satisfaction with the intervention.

Our results clearly suggest the positive effects of the traditional exercise program plus supplementary WBV intervention on pain and fatigue, whereas the effect of exercise alone is not apparent in any of the variables. The results from the EG are consistent with those obtained by Gowans et al.,4 who found no improvements in estimates of pain, fatigue, stiffness, and depression using FIQ after 6 weeks of a similar exercise program plus education therapy. Similarly, Wiggers et al.7 showed no improvements in pain, fatigue, sleep, and depression at week 7 of aerobic exercise therapy, and King et al.9 reported no changes in FIQ scores after 12 weeks of a comparable exercise therapy. In contrast, Redondo et al.24 found improvements in FIQ and fatigue scores (but not in pain, stiffness, and depression) following an 8-week program consisting of cycling, strength, and stretching exercises (5 sessions per week). Our findings for EG do not differ from those of other studies utilizing similar intervention duration, exercise protocols, and the same assessment techniques.

Although differences between EVG and EG in pain and fatigue were not statistically different, the significant difference between EVG and CG, but not between EG and CG, could illustrate that exercise plus supplementary WBV is more effective in improving pain and fatigue than exercise alone. Furthermore, the EVG, but not the EG or the CG, exhibited an improvement in pain and fatigue from baseline to week 6.

 

There are some limitations to this study. The small sample size can be considered a limitation of the study; however, the presence of statistically significant interactions for pain and fatigue with a large effect size are indicative of a meaningful treatment effect. Nevertheless, it is likely that with inclusion of more subjects, a statistically significant interaction for FIQ could have been found. Moreover, in terms of the experimental design, the absence of a fourth treatment group consisting of WBV without exercise impeded the examination of the WBV effect solely; thus, it is unknown to what extent an intervention exclusively based on WBV elicits improvements in health status, physical functioning, and main symptoms of FM. However, WBV alone was not considered as a single therapy, but instead in combination with exercise, because the salutary effects of exercise in this population are well recognized and should not be neglected. This study was attempted to improve the efficacy of exercise as opposed to devising a potential substitute for exercise. Nevertheless, whether WBV strengthens the effect of the exercise therapy, produces improvements independently, or both, needs further investigation.

The significance of this study is threefold. First, we showed that a traditional exercise program with supplementary WBV safely reduces pain and fatigue, the most important symptoms in patients with FM, whereas exercise alone fails to induce significant changes. Second, the results of WBV are observed after only 6 weeks of intervention. Because patients with FM typically report an insufficient available time to invest with an exercise therapy, or complain about not attaining positive results in a quick manner, the WBV may be considered a novel strategy in the management of FM. Third, the present study opens a new line of investigation; further research should be focused on exploring the mechanisms by which WBV produces these positive results, and determining other potential benefits associated with WBV (i.e., muscle strength, hormonal changes, sleep quality). Also, a longer study with greater sample size and objective measures appears timely. Since the present study did not describe the long-term effects or the rate of change, further investigations with multiple follow-ups are also warranted.