Effects of whole-body vibration exercise on the endocrine system
Power Plate Studies
ABSTRACT.
 
Whole-body vibration is reported to increase muscle performance, bone mineral density and stimulate the secretion of lipolytic and protein anabolic hormones, such as GH and testosterone, that might be used for the treatment of obesity. To date, as no controlled trial has examined the effects of vibration exercise on the human endocrine system, we performed a randomized controlled study, to establish whether the circulating concentrations of glucose and hormones (insulin, glucagon, cortisol, epinephrine, norepinephrine, GH, IGF-1, free and total testosterone) are affected by vibration in 10 healthy men [age 39±3, body mass index (BMI) of 23.5±0.5 kg/m2, mean±SEM]. Volunteers were studied on two occasions before and after standing for 25 min on a ground plate in the absence (control) or in the presence (vibration) of 30 Hz whole body vibration. Vibration slightly reduced plasma glucose (30 min: vibration 4.59±0.21, control 4.74±0.22 mM, p=0.049) and increased plasma norepinephrine concentrations (60 min: vibration 1.29±0.18, control 1.01±0.07 nM, p=0.038), but did not change the circulating concentrations of other hormones. These results demonstrate that vibration exercise transiently reduces plasma glucose, possibly by increasing glucose utilization by contracting muscles. Since hormonal responses, with the exception of norepinephrine, are not affected by acute vibration exposure, this type of exercise is not expected to reduce fat mass in obese subjects. (J. Endocrinol. Invest. 27: ??-??, 2004)
 
INTRODUCTION
 
Muscle contraction induced by low amplitude, high frequency mechanical stimulation is reported to increase muscle strength and performance (1-4), as well as bone density (5-7). Whole body mechanical vibration has a tonic excitatory influence on muscles eliciting a response named “tonic vibration reflex” (4). This reflex involves activation of muscle spindles, mediation of the neural signals by 1a afferents, and activation of the muscle fibers via large _-motor neurones (4). Potentially, repeated muscle contractions might exert endocrine and/or metabolic effects. Vibration has been reported to affect the endocrine system augmenting the circulating concentrations of GH and testosterone and reducing those of cortisol (8). At present, use of whole-body vibration is confined to the training regimens of athletes. As far as the potential clinical applications of the technique are concerned, the few controlled studies published in literature focus on its beneficial effects on osteoporosis (5-7). However, the paper by Bosco et al. (8) in showing that whole-body vibration increases serum GH and testosterone and reduces cortisol concentrations, suggests a potential use of vibration exercise for obesity treatment. Visceral obesity is associated with deficient GH and testosterone (9-12) and probably also with excessive cortisol secretion (13), so that hypercortisolism and hyposomatotropism might both contribute to visceral fat deposition (9, 13). Furthermore, lipolysis of visceral obese men is particularly sensitive to treatment with very low doses of recombinant human GH (14). Thus, the concomitant stimulation of two protein anabolic hormones, such as GH and testosterone (15), and the inhibition of cortisol secretion should reduce fat and augment lean body mass in obese subjects. Vibration exercise might be a safe, low-cost way of inducing these hormonal changes with the advantage of being more easily accepted by overweight individuals than conventional physical exercise. To date, as no controlled trial has examined the acute effects of whole-body vibration on the human endocrine system, we performed a randomized controlled study to establish whether circulating concentrations of glucose and several hormones (insulin, glucagon, cortisol, epinephrine, norepinephrine, GH, IGF-1, and testosterone) are affected by vibration exercise in healthy men, using the vibration protocol published by Bosco et al. (8).
 
MATERIALS AND METHODS
 
Subjects
 
After the study was approved by the Local Ethical Committee, informed written consent was obtained from 10 adult men in good health as determined by medical history, physical examination and laboratory evaluation. Their age (mean±SEM) was 39±3 yr (range 25-50 yr) and their body mass index (BMI) was 23.5±0.5 (range 21-26 kg/m2).
 
Design of the study
 
All subjects were studied, in random order, on two different occasions: vibration or control at 2- to 4-day intervals. On each occasion, volunteers were admitted to the Clinical Research Center of our Department at ~ 07:30 h, after an overnight fast. An antecubital vein was cannulated with a 21-gauge plastic catheter needle and kept patent by 0.9% NaCl infusion. The volunteers relaxed in the sitting position and at ~ 08:00 h 15 ml of blood were drawn for measurements of basal circulating concentrations of glucose and hormones (insulin, glucagon, GH, IGF-1, cortisol, epinephrine, norepinephrine and free and total testosterone). After the basal blood sampling, in the control study, subjects stood with the knees slightly bent (~ 70°) for 25 min on the ground platform of the vibration device (NEMES 30 L, KB Ergotest, Mikkeli, Finland). Their hands were placed on the rigid lever arms. In the vibration study, volunteers in the same position on the vibration device, were exposed to 10 vibration series of 1 min duration with 1 min rest between each treatment and with 5 min rest after the first 5 series (total 25 min). The frequency of the vibrations was set at 30 Hz (displacement±4 mm; acceleration 17 g); all subjects wore thin-soled gymnastic-type shoes. In both studies, after 25 min standing on the platform the subjects sat down and 15 ml of blood were withdrawn for measurements of glucose and hormones 5 min after standing (i.e. at 30 min from baseline) and again 30 min later (i.e. at 60 min from baseline).
 
Analytical methods
 
Plasma concentrations of glucose were determined using a Beckman glucose analyzer (Beckman Instruments, Palo Alto, CA). Serum concentrations of insulin (Technogenetics, Milan, Italy), GH (Biodata, Ares Serono, Norwell, MA), IGF-1 (acid alcohol extraction, Nichols Institute Diagnostics, S. Juan Capistrano, CA), free testosterone (Biochem Immuno System, Bologna, Italy) and the plasma concentrations of glucagon (DRG International Inc., USA) were measured using commercial immunoradiometric assays. Serum concentrations of cortisol and total testosterone were determined by enhanced chemiluminescence using Ortho-Clinical Diagnostics kits (Johnson & Johnson, New Brunswick, NJ). The plasma concentrations of catecholamines were measured by high performance liquid chromatography (HPLC) (16).
 
Statistical analysis
 
Statistical analyses were performed using repeated measure analysis of variance with treatment (vibration vs no vibration) and time as within factors and corrections for non-sphericity according to Huynh-Feldt epsilon. Where significant differences in mean responses were found, Fisher’s LSD Multiple-Comparison Test was applied. Data are presented as mean±SEM, p<0.05 was considered statistically significant. All analyses were run using Statistica 4.5 (StatSoft, Inc. 1993, OK).
 
RESULTS
 
The results are reported in Table 1. In the vibration and control studies, plasma glucose concentration was similar at baseline. In the vibration study, it decreased at 30 min (p=0.049 vs control) and returned to normal values at 60 min. The reduction in plasma glucose concentration occurred in the absence of significant changes in the circulating concentrations of serum insulin and plasma glucagon.
 
During both the control and the vibration studies, serum cortisol concentration significantly decreased (p<0.05) at 30 and 60 min compared with baseline, without differences between the two studies. In both control and vibration studies, plasma norepinephrine concentration significantly increased at 30 min (p<0.05 vs baseline). In the vibration study, plasma concentration of norepinephrine at 60 min was higher than the 60 min control level (p=0.038). Vibration had no significant effect on serum concentrations of GH, IGF-1, free testosterone, total testosterone and plasma epinephrine. Although at 30 min circulating serum GH concentrations showed a trend to increase more in the vibration (7.4±3.6 μU/ml) than in the control study (5.2±2.5 μU/ml), these changes did not reach statistical significance (p=0.216).
 
DISCUSSION
 
The results of this study demonstrate that acute exposure to whole-body vibration in healthy men induced transient changes in plasma glucose and norepinephrine concentrations, but did not activate the pituitary-adrenal-gonadal axis.
 
As serum insulin concentrations were not affected by vibration exercise, hepatic glucose production appeared to remain unchanged. The transient drop in plasma glucose concentration at the end of the vibration session was probably due to increased uptake of circulating glucose by contracting muscles. Our hypothesis is supported by several studies showing that exercise leads to GLUT 4 translocation from the microsomes to the plasma membranes of contracting muscles via direct (exercise-mediated) and indirect (insulin-mediated) pathways (17). Rittweger et al. (18) have shown that exhaustive wholebody vibration increases carbohydrate oxidation because it significantly shifted the respiratory quotient from 0.82 to 0.90. Enhanced blood flow to muscle, induced by vibration exercise (19), might increase glucose uptake. Data from indirect calorimetry (18) and our results suggest that carbohydrates are the predominant energy source for contracting muscles during vibration exercise. Further research is warranted in this field. Our data were obtained in lean healthy subjects. Possibly, other effects following vibration treatment may be seen in obese or Type 2 diabetic individuals. From a clinical point of view, it would be interesting to establish whether repeated exposure to vibration exercise might increase insulin- sensitivity in subjects with insulin-resistance and/ or Type 2 diabetes mellitus.
 
Effects of vibration exercise on plasma concentrations of glucose, epinephrine, norepinephrine and glucagon and on serum concentrations of insulin, cortisol, GH, IGF-1, free and total testosterone in 10 healthy men.
In our study, vibration significantly affected the plasma norepinephrine concentration. In both control and vibration studies plasma norepinephrine concentration, unlike the plasma epinephrine concentration, significantly increased at 30 min compared with baseline. This increase was expected because standing is a classic stimulus for norepinephrine release by peripheral sympathetic neurons (20). After 30 min of rest in the sitting position, plasma norepinephrine concentration returned to baseline (p=NS 60 vs 0 min) in the control study, but remained higher than baseline and in controls after vibration, suggesting that exercise resulted in protracted peripheral sympathetic pathway activation. In healthy humans, exposed to exhaustive vibration exercise, transient rises in heart rate and systolic blood pressure and a drop in diastolic blood pressure have been observed (18). These hemodynamic changes cannot be explained by the isolated norepinephrine response observed in our study. They are probably the result of full activation of the adrenergic system triggered by exhaustive stimulation of muscle contraction by vibration.
 
In the present study we did not observe significant effects of vibration on the circulating concentrations of cortisol, testosterone and GH, thus diverging from the results of Bosco et al. (8) who, using a similar vibration protocol, reported that serum cortisol significantly decreased and serum testosterone and GH concentrations significantly increased in 14 male volunteers after vibration. The effects of vibration on hormonal responses were compared to baseline and the effects of variables such as time or postural changes could not be assessed as the study was not controlled (8). In our study we also observed serum cortisol levels were lower than baseline after vibration but a similar decrease in the control study shows that vibration did not inhibit cortisol secretion, indicating that fall in cortisol reproduced the hormonal circadian rhythm. Bosco et al. (8) reported significant increases in serum total testosterone (p<0.026) from a baseline value of 22.7±6.6 (SD) to 24.3±6.6 nM after vibration and in serum GH (p<0.014) from a basal value of 6.2±16.2 (SD) to 28.6±29.6 (SD). We were unable to confirm the modest effect of vibration on testosterone because no significant effect of time or intervention on either free or total testosterone concentrations was observed in our study. In our vibration study serum GH increased from the basal 1.7±0.6 to a peak of 7.4±3.6 μU/ml at 30 min, but the comparison with the 30 min point of the control study (5.2±2.5 μU/ml) did not reach statistical significance (p=0.216). This finding suggests that part of the effect of vibration on GH in Bosco’s study (8) was caused by other variables such as pulsatile GH secretion and/or postural changes. A Type 2 error may be present in our study considering the limited number of volunteers and the high standard deviation due to erratic GH secretion. However, the modest difference between the intervention and the control study (2.2 μU/ml) and the related standard deviation required a sample size of 113 subjects to achieve 80% power and to rule out a Type 2 statistical error. However, it must be underlined that since the natural GH peaks usually last for only a few minutes, our 30 min interval sampling does not offer the possibility to discover transient and small differences. In any case, the effects of acute vibration exercise on GH secretion are not encouraging because they are small and certainly not comparable to the GH peaks induced by conventional exercise (21). Thus, we do not expect repeated sessions of vibration exercise to be of benefit to the body composition of overweight individuals. Vibration sessions might be useful to increase the strength of lower limb muscles (1-4) and, possibly, to reduce plasma glucose concentrations by increasing glucose oxidation and disposal by contracting muscles.
 
Very promising results have been reported about the potential application of whole body vibration for treatment of osteoporosis. Low intensity vibration generated by a ground platform prevented osteoporosis in ovariectomized rats (5) and 1 yr of treatment (20 min daily sessions) increased by 34% the bone density of trabecular bone in the proximal femur of adult sheep (6). Intervention studies in post-menopausal women (22) and in children with cerebral palsy (23) have been initiated. The results of our study show that acute whole body vibration does not activate production of anabolic bone hormones such as GH, testosterone or IGF-1 suggesting that the endocrine system does not mediate the positive effects of vibration on bone density that could be merely the result of mechanical stimulation. However, a long-term study would be needed addressing possible endocrine alterations under similar conditions as those producing beneficial bone effects. In fact, it is possible that long-term, repeated vibration and muscle contraction could directly or indirectly (e.g. via improved insulin sensitivity and altered body composition) affect the levels of these and other hormones.