Whole-body vibration versus proprioceptive training on postural control in post-menopausal osteopenic women

Gait & Posture 38 (2013) 416–420

Contents lists available at SciVerse ScienceDirect

Gait & Posture journal homepage: www.elsevier.com/locate/gaitpost

Whole-body vibration versus proprioceptive training on postural control in post-menopausal osteopenic women Nils Stolzenberg a,*, Daniel L. Belavy´ a, Rainer Rawer b, Dieter Felsenberg a a b

Centre for Muscle and Bone Research, Charite´ Universita¨tsmedizin Berlin, Hindenburgdamm 30, 12203 Berlin, Germany Novotec Medical GmbH, 75172 Pforzheim, Germany

A R T I C L E I N F O

A B S T R A C T

Article history: Received 2 May 2012 Received in revised form 11 November 2012 Accepted 3 January 2013

Background: To prevent falls in the elderly, especially those with low bone density, is it necessary to maintain muscle coordination and balance. The aim of this study was to examine the effect of classical balance training (BAL) and whole-body vibration training (VIB) on postural control in post-menopausal women with low bone density. Methods: Sixty-eight subjects began the study and 57 completed the nine-month intervention program. All subjects performed resistive exercise and were randomized to either the BAL- (N = 31) or VIB-group (N = 26). The BAL-group performed progressive balance and coordination training and the VIB-group underwent, in total, four minutes of vibration (depending on exercise; 24–26 Hz and 4–8 mm range) on the Galileo Fitness. Every month, the performance of a single leg stance task on a standard unstable surface (Posturomed) was tested. At baseline and end of the study only, single leg stance, Rombergstance, semi-tandem-stance and tandem-stance were tested on a ground reaction force platform (Leonardo). Results: The velocity of movement on the Posturomed improved by 28.3 (36.1%) (p < 0.001) in the VIBgroup and 18.5 (31.5%) (p < 0.001) in the BAL-group by the end of the nine-month intervention period, but no differences were seen between the two groups (p = 0.45). Balance tests performed on the Leonardo device did not show any significantly different responses between the two groups after nine months (p  0.09). Conclusions: Strength training combined with either proprioceptive training or whole-body vibration was associated with improvements in some, but not all, measures of postural control in postmenopausal women with low bone density. The current study could not provide evidence for a significantly different impact of whole-body vibration or balance training on postural control. ß 2013 Elsevier B.V. All rights reserved.

Keywords: Galileo whole body vibration exercise Postmenopausal osteoporosis balance exercise Leonardo Mechanography

1. Introduction In the course of age related musculoskeletal deterioration, proprioception and balance worsen, the risk of falls increases [1] and the risks of injuries increases [2]. In older women with reduced bone density, a particular concern is of bone fracture which may lead to significant costs for the individual [3] and society [4]. Balance training is considered a useful method to prevent falls [5,6]. Such training is best performed progressively to present a constant challenge for the adaptation of the proprioceptive and vestibular system [7]. Such training can also counter aging associated sarcopenia [8–10].

* Corresponding author at: Center for Muscle and Bone Research, Charite´ Universita¨tsmedizin Berlin, Hindenburgdamm 30, D-12200 Berlin, Germany. Tel.: +49 8445 3040; fax: +49 30 793 5918. E-mail address: [email protected] (N. Stolzenberg). 0966-6362/$ – see front matter ß 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.gaitpost.2013.01.002

Some authors have however suggested that whole-body vibration training can be used to improve balance and proprioception [11,12]. There is some evidence that wholebody vibration training can improve neuromuscular function [12], lumbo-pelvic proprioception [11] and evidence suggest it stimulates muscle activity via the muscle spindle system [13]. To the best of our knowledge, no prior work has evaluated the effect of balance training versus whole-body vibration exercise on postural control. In female post-menopausal subjects with reduced bone density (osteopenia/osteoporosis) we examined the influence of whole-body vibration and balance training on postural control. We hypothesized that vibration training could improve postural control in single leg stance in non-deconditioned postmenopausal women. We hypothesized further, however, that classical balance training would be more effective than whole-body vibration training in improving postural control.

N. Stolzenberg et al. / Gait & Posture 38 (2013) 416–420

417

The subjects for the study were recruited via the Siemens Betriebskrankenkasse, Centre for Muscle and Bone Research (Charite´ Universita¨tsmedizin Berlin) and the Immanuel Krankenhaus Berlin. From 101 women who were screened for the study, 68 subjects were included, from which 57 subjects finished the study (Table 1). Subjects were required to complete each training session and to attend at least 75% of each training sessions and to not miss more than four trainings sessions in a row. Drop-outs (N = 8 in the VIB group, N = 3 in the BAL group) were due to injury or illness not associated with the study. 2.2. Interventions Each subject attended two training appointments per week. Group training was performed with a maximum of 10 subjects per session. The lead author performed all exercise sessions with the subjects. At the start of each training session a short warm up of 15 min of cycle ergometry was performed. This was followed by a one-set training of minimum 10 to maximum 20 repetitions resistance exercise. Loading levels were progressed if the subject could achieve 20 or more repetitions. Standard gym equipment was used to perform the following exercises: bilateral leg press in lying, bilateral hips abduction in sitting, bilateral hip adduction in sitting, hip extension in standing (left and right), trunk flexion in sitting, trunk extension in sitting, trunk rotation in sitting, lat-pull downs in sitting, and seated cable pull in standing. The resistive exercise component lasted approximately 30 min. Following this exercise, subjects performed either balance training (BAL) or whole-body vibration (VIB). The BAL-group performed a progressive proprioceptive and balance program exercises each week by the use of a variety of exercises, equipment (such as staves and balls) and/or different surfaces (foam mats, pillows, wobble-boards) to increase task difficulty. The following exercises were progressed in difficulty over the 9month intervention period:

Fig. 1. Subject flow. SBK, Siemens Betriebskrankenkasse; IKH, Immanuel Krankenhaus; CBF, Charite´ Campus Benjamin Franklin; DXA, dual X-ray absorptiometry screening. Exclusion after DXA-scanning: 24 candidates had a Tscore less than 3.0 SD, 7 greater than 2.0 SD and DXA data from 2 candidates were not of adequate quality to make an accurate assessment of bone density.

2. Methods 2.1. Study design, sample size estimate and subjects The current study was a non-pharmacological, randomized, controlled study which was approved by the ethical commission of the Charite´ Universita¨tsmedizin Berlin and the subjects gave their informed written consent. Sixty-eight otherwise healthy post-menopausal women with reduced bone density took part in a nine month intervention randomized into either a whole-body vibration (VIB) or balance (BAL) training group (Fig. 1). The primary outcome measure was velocity (millimeters per second) of stance surface movement during single-leg-stance on the Posturomed. The remaining secondary outcome measures were performed before and after the training intervention. The inclusion criteria were: a minimum of eight years postmenopausal, total hip or lumbar spine (L1–L4) T-score from 2.0 to 3.0 SD on dual energy X-ray absorptiometry (DXA; Lunar Prodigy Advance, GE Medical Systems, Wisconsin, USA) and able to walk without any aids. Exclusion criteria were: whole-body vibration exercise, balance training or resistive exercise in the last six month, any metal implants, disturbance of the vestibular system, prior experience with the testing apparatus, bone fractures within the last year, neuromuscular and neurological diseases. Based upon an estimated effect size of 0.8 standard deviations between the two groups for the primary outcome measure, and assuming a power of 0.8 and an alpha-level of 0.05 (two-sided), it was estimated that 26 subjects were necessary for each subject group. Assuming a drop-out rate of 10%, we aimed to recruit a minimum of 60 subjects in total for randomization into each subject group.

 Romberg, tandem and single-leg stance were performed on surfaces of varying degrees of instability and at varying degrees of difficulty: firm mat, soft mat, wobble-board, air-pillows, with and without shoes, with eyes open or with eyes closed.  Softballs, tennis balls, staves, elastic bands were used for coordination training involving throwing, catching, passing around the body, passing underneath a leg, flipping between hands. The exercises were performed either alone, or in the case of throwing and advanced balance exercises, with a partner. The balance training lasted approximately 15 min. At the same time, subjects of the VIB-group trained three consecutive times on the Galileo Fitness (Novotec, Pforzheim, Germany). Amplitude of vibration began at 2 mm in the first week and was progressed to 4 mm within four weeks. Three different exercises were performed: 1. Standing for 1.5 min with lightly bended knees and hips with a straight back. Vibration frequency beginning at 22 Hz and progressed to 24 Hz after one to two weeks. 2. Continuous squatting from erect standing to 908 knee flexion (2 s down, 2 s up) for 1.5 min. Vibration frequency beginning at 22 Hz and progressed to 24 Hz. 3. One minute of continuous stance in 908 knee flexion. Vibration frequency set to 26 Hz for the duration of the study. The vibration training lasted a total of 4 min, with one minute break between sets. 2.3. Primary outcome measure: balance tests on an unstable surface (Posturomed) The Posturomed 202 (60  60 cm; Haider Bioswing, Pullenreuth, Germany, www.bioswing.de) was used in conjunction with the CMS 10 (Zebris Medizintechnik, Isny, Germany) to measure the movement of an unstable surface (the Posturomed) via ultrasound sensors (the CMS 10). A similar measurement approach has been used in previous work [14]. Markers were positioned on the side of the Posturomed at a distance of 15 cm from the middle of the platform. The sensor of the CMS10 was orientated at a 608 angle to the markers on the Posturomed. The degree of instability of the platform can be adjusted into three levels by the use of

Table 1 Subject anthropometric characteristics. Group

VIB BAL

N

26 31

Age (years)

65.9 (4.5) 67.3 (3.7)

Height (cm)

161.1 (5.2) 161.4 (5.4)

Weight (kg)

Body mass index (BMI)

Baseline

End month 9

Baseline

End month 9

62.8 (9.8) 64.0 (6.9)

62.8 (10.1) 64.4(7.7)

24.0 (4.0) 24.4 (2.7)

24.1 (3.8) 24.6 (3.1)

Values are mean (SD); VIB: resistive exercise with whole-body vibration group; BAL: resistive exercise with balance training group. There were neither significant differences between groups at baseline nor within or between group changes over time (p > 0.5)

418

N. Stolzenberg et al. / Gait & Posture 38 (2013) 416–420

two diagonally orientated brakes. Pilot trials showed that use of an intermediate level of instability (one of the diagonal brakes loosened, the other fastened) was optimal for the current population. The test was performed five times for single-leg-stance on each side. All subjects wore standard gym shoes. The left leg was tested first and then the right. The eyes remained open for all tests. The subject was then instructed to take three steps on the spot and then the non-stance knee was lifted to the level just below the hip. The beginning of the measurement was registered digitally and began immediately at the start of single leg support. The test was stopped if the subject put their nonstance foot down, touched the surrounding railing with any part of their body. The test ended either with the loss of testing posture or after the test was conducted for five seconds. The test was only repeated in instances where the subject was distracted by unexpected external stimuli during the test. The path-length of surface movement and duration of movement until test cessation were measured. From this data, the average velocity (in millimeters per second) of movement during each test was calculated as the main outcome parameter. Previous work has indicated sway velocity to be an indicator of postural stability [15]. Testing was performed prior to the start of the intervention program and then every month for a total of ten testing sessions per subject. Software provided by the manufacturer (WinData V.2.22.13; Zebris) was used for recording and storing the data for further analysis. Since the Posturomed surface is unstable, an improvement of subject performance (balance) is expressed by their ability to reduce the movement of the testing surface during the testing period. Hence, decrease in the movement velocity is considered an improvement in this test. 2.4. Further balance tests on a stable surface (Leonardo) Testing was performed on a ground reaction force platform (Leonardo Mechanograph GRFP, Novotec GmbH, Pforzheim, Germany). The subject stood in relaxed stance on the testing platform with their arms at their side. After a tone signal, the subject moved into the test position, and the each of the following tests was conducted for ten seconds in the following order: Romberg-stance, semitandem-stance, tandem-stance and single-leg-stance (both the dominant and nondominant legs). Each test with eyes open was followed by same test with eyes closed. Each test was conducted once. If the subject lost their balance or put their non-stance foot down, the test was repeated only once more. All subjects wore standard gym shoes. Software provided by the manufacturer (Leonardo Mechanography Research Edition v4.2.b01.06c) was used for recording and storage of data and for subsequent calculation of the variables of interest. From each test the velocity of movement of the center of pressure (in mm/s) and area of an ellipse (in cm2) surrounding the movement of the center of pressure were used in further analysis. 2.5. Statistical analyses Linear mixed-effects models [16] were used to model time-point (Posturomed: baseline and every month until end month 9; Leonardo: baseline and end month 9) and subject-group main effects and their interaction. Allowances for heterogeneity of variance, such as due to group and/or time-point were implemented when necessary. Additional models with baseline values as a covariate, subject weight or body mass index as a covariate, or analysis of data expressed as percentage change compared to baseline did not influence the outcomes. Changes in weight or BMI did not correlate to changes in the outcome variables. Analysis of the data coding the legs as left and right, rather than dominant and non-dominant, resulted in similar findings. Where no significant between-group differences were seen, secondary analyses also considered the effects over time with both groups pooled. An alphalevel of 0.05 was taken for statistical significance. The ‘‘nlme’’ package was used for linear mixed-effects modeling in the ‘‘R’’ statistical environment (version 2.10.1, www.r-project.org). Unless otherwise specified, results are presented as mean (SD).

3. Results 3.1. Participants No significant differences were seen between the two groups for any of the parameters measured (p  0.22) at baseline (Fig. 2; Table 2). VIB-subjects completed 91.1(6.4%) of all scheduled training sessions and the BAL-subjects completed 90.5 (7.5%). 3.2. Balance testing on an unstable surface: posturomed Significant improvements in this test were seen in both groups over the course of the study (Fig. 2), though there was no difference between the groups (group  time-point: p = 0.45). By the end of the study, the velocity of movement during the Posturomed-test reduced by 28.3(36.1%) (p < 0.001) in the VIB-group and 18.5 (31.5%) (p < 0.001) in the BAL-group.

Fig. 2. Single leg stance on the posturomed with eyes open. Values are mean (SD) in mm/s on each testing day from the beginning of the study (time-point 1) and for the following nine-months. BAL: resistive exercise with balance training group. VIB: resistive exercise with whole-body vibration group. Values have been averaged across dominant and non-dominant legs. *p < 05; yp < 0.01; zp < 0.001 and indicate significance of difference compared to baseline testing. Reductions in the movement velocity are considered as an improvement (see Section 2). There was no significant difference in the response of the two groups over the course of the study.

3.3. Balance testing on a stable surface: Leonardo Mechanography Single-leg-stance testing on the Leonardo showed no significant differences between the groups (group  time-point: p  0.09) and no significant changes after the 9-month intervention period within each group (p  0.053; Table 2). Pooling the data across all tests (eyes open, eyes closed, dominant leg, non-dominant leg) yielded similar results. No differential effect of the training programs was seen on the area of center of pressure movement (group  time-point: p  0.074; Table 2). Although significant changes were seen within the VIB-group for Romberg-stance eyes closed (p = 0.039; Table 2) and in the BALgroup for tandem-stance eyes open (p = 0.011; Table 2) and semitandem-stance eyes closed (p = 0.020), no significant differences were seen between the groups for any of these remaining tests (group  time-point: p  0.24). No significant changes in movement area of the center of pressure were seen between groups (Table 2). 3.4. Correlation analyses Comparison between the single leg stance (eyes open) tests on the two devices showed that although the velocity of center of pressure movement did correlate moderately between the two testing approaches at both baseline (r = 0.48, p = 0.0002) and the end of the study (r = 0.45, p = 0.0002), the changes in performance of each test were unrelated (r = 0.04, p = 0.76). 4. Discussion The hypothesis of the current study was that balance training would have a greater positive effect on postural control in singleleg-stance than whole-body vibration training. Postural control, as measured by the velocity of movement on the Posturomed device, improved in both groups over the nine-month intervention period. However, no significant difference was seen between the two groups. Testing on a stable surface (Leonardo) showed few significant changes after the intervention period. Contrary to our hypothesis, the data of the current study imply that a whole-body vibration program in conjunction with resistive exercise, can improve postural control to a similar extent as

N. Stolzenberg et al. / Gait & Posture 38 (2013) 416–420

419

Table 2 Center of pressure (CoP) movement on Leonardo testing. Eyes

Group

CoP movement area (cm2)

CoP movement velocity (mm/s) Baseline

End month 9

Percentage change

Baseline

End month 9

Romberg-stance VIB Open BAL VIB Closed BAL

19.1 17.8 31.9 30.8

(8.2) (5.7) (15.5) (15.3)

19.4 17.9 37.1 33.6

(7.9) (4.2) (12.5)* (9.8)

Semi-tandem-stance Open VIB BAL Closed VIB BAL

23.5 24.5 39.6 40.5

(7.1) (8.1) (20.3) (18.8)

24.5 25.1 41.6 46.3

(6.5) (6.1) (10.7) (13.5)*

Tandem-stance VIB Open BAL Closed VIB BAL

48.4 47.5 81.0 85.4

(25.3) (14.6) (58.9) (40.6)

51.3 52.4 85.3 84.0

(22.5) (10.3)* (58.1) (32.1)

+6.0 +10.3 +5.3 1.6

(0.4) (0.2)* (0.7) (0.4)

3.8 3.7 17.7 13.8

(0.9) (0.7) (2.9) (1.9)

3.3 4.0 11.2 11.5

Single-leg-stance (dominant leg) Open VIB 77.4 BAL 77.5 Closed VIB 131.6 BAL 142.1

(32.8) (27.1) (56.2) (40.0)

84.7 74.3 151.0 135.4

(26.5) (17.7) (73.4) (86.9)

+9.4 4.2 +14.7 4.7

(0.3) (0.2) (0.5) (0.6)

6.7 6.9 21.2 25.2

(1.0) (0.9) (1.5) (1.5)

Single-leg-stance (dominant leg) Open VIB 81.4 BAL 75.3 Closed VIB 153.0 BAL 146.6

(31.5) (26.5) (63.0) (53.4)

85.3 78.0 131.1 136.8

(27.1) (22.4) (56.2)* (54.7)

+4.8 +3.7 14.4 6.7

(0.3) (0.3) (0.4)* (0.4)

6.8 5.9 30.6 33.0

(1.0) (0.7) (1.8) (2.3)

+1.7 +0.1 +16.2 +9.2

(0.4) (0.2) (0.3)* (0.3)

+4.3 (0.3) +2.5 (0.2) +5.2 (0.3) +1.43 (0.3)*

2.1 2.0 4.3 4.0

(0.5) (0.4) (0.8) (0.7)

2.46 (0.5) 2.52 (0.5) 6.0 (0.9) 5.0 (0.8)

2.0 1.7 4.6 3.9

(0.5) (0.4) (0.7) (0.7)

Percentage change 7.5 12.2 +8.3 1.6

(33.6) (31.3) (34.3) (33.0)

7.8 9.1 6.2 +18.0

(36.3) (33.0) (33.5) (34.0)

(0.9) (0.7)y (2.8) (2.0)

12.3 +6.3 36.9 17.2

(49.5) (36.9)y (85.1) (59.1)

7.0 4.8 40.0 26.9

(1.1) (0.8) (2.8)* (2.3)

+3.6 30.3 +88.6 +6.7

(42.7) (35.7) (43.7)* (44.4)

6.1 6.7 22.0 28.0

(1.0) (0.7) (1.7)y (2.4)

10.5 +12.9 28.0 15.1

(41.6) (28.7) (37.1)y (45.1)

2.27 (0.5) 2.29 (0.5) 5.7 (0.8) 5.9 (0.8)

Values are mean (SD); VIB: resistive exercise with whole-body vibration group; BAL: resistive exercise with balance training group. * Significance of change within each group indicated by: p < 0.05 y Significance of change within each group indicated by: p < 0.01.

resistive exercise combined with balance training. Recent studies have shown that whole-body vibration can improve balance in early geriatric rehabilitation [17] and after unilateral stroke [18]. Similarly, balance training [19] and Tai-Chi [20] can improve postural control in elderly women with low bone density. As such, the current study supports the idea that both whole-body vibration exercise and proprioceptive training, both in conjunction with resistive exercise, can improve balance. This could suggest greater time effectiveness of whole-body vibration exercise, but this would need to be tested in further work. An interesting finding of the current study was that changes in single leg stance postural control on an unstable testing surface (Posturomed) were not reflected in tests on a stable testing surface (Leonardo). The performance of the two tasks was moderately correlated at baseline and nine months later but the changes over time were not correlated. Improvements in balance on the unstable testing surface (Posturomed) were, in our view, reflected by decreases in movement velocity. Based on current literature, it is unclear whether increases in center of pressure movement velocity and area on the stable testing surface (Leonardo) are adaptive or maladaptive. On the one hand it is possible that the body, due to otherwise sub-optimal postural control, has to quickly react to the external and internal perturbations applied to it. This would consequently generate a higher movement velocity and may reflect poorer postural performance. On the other hand, a more adaptive control of posture could act in a feed-forward manner to utilize more joints to maintain balance rather than, for example, just using an ankle-strategy and reacting. This process could also result in a greater center of pressure movement velocity and area. It is important to discuss some of the limitations of the current study. Aside from low bone density, the subjects included in the study were otherwise in good physical condition. It is possible that

subjects with poorer balance at the start of the study may have shown more improvement or a differential effect of the exercise programmes. Given the results of the current study and also its methodology, it is not clear to what extent the resistive exercise protocol influenced the changes in postural parameters. Clinical trials have provided evidence suggesting that resistive exercise can improve postural control in elderly individuals [21,22]. Resistive exercise is considered an important component of interventions to improve balance and prevent falls in the elderly [23]. Furthermore, whilst we consider it unlikely that a once-monthly test on the Posturomed led to a selective ‘‘training effect’’ on this testing device, given the current methodology, we cannot rule out this possibility. In conclusion, the current study showed in a study of postmenopausal osteopenic women that, contrary to our expectations, a similar effect of whole-body vibration and balance training, in conjunction with resistive exercise. These effects were seen in some, but not all, aspects of postural control testing. Funding No financial support was received for the conduct of this project. The companies Zebris (CMS 10 system), Haider-Bioswing (Posturomed system), and Novotec Medical (Leonardo Mechanography system) loaned equipment for use in the current project. Acknowledgements The subjects who participated in the study are thanked for their involvement. We also thank the staff from the radiology department (Charite´ Campus Benjamin Franklin) and the Center of Muscle and Bone Research for their assistance. The SportGesundheitspark Berlin e.V. provided training rooms for the

420

N. Stolzenberg et al. / Gait & Posture 38 (2013) 416–420

current study and also assisted with the medical screening. We also acknowledge the assistance of the Immanuel-Krankenhaus Berlin, particularly Drs. Semler and Hellmich, as well as the SBK Siemens Krankenkasse for assistance with subject recruitment. The lead author thanks his family for their support during the course of his PhD. Conflict of interest statement Rainer Rawer is an employee of Novotec Medical, Dieter Felsenberg acts as an unpaid consultant to Novotec Medical for the exploitation of the study’s results. All other authors have no conflict of interest.

[11]

[12]

[13]

[14]

[15]

References [1] Lord SR, Clark RD, Webster IW. Postural stability and associated physiological factors in a population of aged persons. Journal of Gerontology 1991;46:M69– 76. [2] Kannus P, Parkkari J, Koskinen S, Niemi S, Palvanen M, Jarvinen M, et al. Fallinduced injuries and deaths among older adults. Journal of the American Medical Association 1999;281:1895–9. [3] Magaziner J, Simonsick EM, Kashner TM, Hebel JR, Kenzora JE. Predictors of functional recovery one year following hospital discharge for hip fracture: a prospective study. Journal of Gerontology 1990;45:M101–7. [4] Autier P, Haentjens P, Bentin J, Baillon JM, Grivegnee AR, Closon MC, et al. Costs induced by hip fractures: a prospective controlled study in Belgium. Belgian Hip Fracture Study Group. Osteoporosis International 2000;11:373–80. [5] Perrin PP, Gauchard GC, Perrot C, Jeandel C. Effects of physical and sporting activities on balance control in elderly people. British Journal of Sports Medicine 1999;33:121–6. [6] Chang JT, Morton SC, Rubenstein LZ, Mojica WA, Maglione M, Suttorp MJ, et al. Interventions for the prevention of falls in older adults: systematic review and meta-analysis of randomised clinical trials. BMJ 2004;328:680. [7] Hilberg T, Herbsleb M, Gabriel HH, Jeschke D, Schramm W. Proprioception and isometric muscular strength in haemophilic subjects. Haemophilia 2001;7:582–8. [8] Campbell AJ, Borrie MJ, Spears GF. Risk factors for falls in a community-based prospective study of people 70 years and older. Journal of Gerontology 1989;44:M112–7. [9] Guralnik JM, Ferrucci L, Simonsick EM, Salive ME, Wallace RB. Lower-extremity function in persons over the age of 70 years as a predictor of subsequent disability. New England Journal of Medicine 1995;332:556–61. [10] Messier SP, Royer TD, Craven TE, O’Toole ML, Burns R, Ettinger Jr WH. Longterm exercise and its effect on balance in older, osteoarthritic adults: results

[16] [17]

[18]

[19]

[20]

[21]

[22]

[23]

from the Fitness, Arthritis, and Seniors Trial (FAST). Journal of the American Geriatrics Society 2000;48:131–8. Fontana TL, Richardson CA, Stanton WR. The effect of weight-bearing exercise with low frequency, whole body vibration on lumbosacral proprioception: a pilot study on normal subjects. Australian Journal of Physiotherapy 2005;51:259–63. Rittweger J. Vibration as an exercise modality: how it may work, and what its potential might be. European Journal of Applied Physiology 2010;108:877– 904. Cochrane DJ, Loram ID, Stannard SR, Rittweger J. Changes in joint angle, muscle-tendon complex length, muscle contractile tissue displacement, and modulation of EMG activity during acute whole-body vibration. Muscle and Nerve 2009;40:420–9. Muller O, Gunther M, Krauss I, Horstmann T. Physical characterization of the therapeutic device Posturomed as a measuring device – presentation of a procedure to characterize balancing ability. Biomedizinische Technik Biomedical Engineering2004;49:56–60. Chiari L, Cappello A, Lenzi D, Della Croce U. An improved technique for the extraction of stochastic parameters from stabilograms. Gait & Posture 2000;12:225–34. Pinheiro J, Bates D. Mixed-effects models in S and S-PLUS. Berlin: Springer; 2000. Merkert J, Butz S, Nieczaj R, Steinhagen-Thiessen E, Eckardt R. Combined whole body vibration and balance training using vibrosphere (R): improvement of trunk stability, muscle tone, and postural control in stroke patients during early geriatric rehabilitation. Zeitschrift fur Gerontologie und Geriatrie 2011;44:256–61. van Nes IJ, Geurts AC, Hendricks HT, Duysens J. Short-term effects of wholebody vibration on postural control in unilateral chronic stroke patients: preliminary evidence. American Journal of Physical Medicine and Rehabilitation 2004;83:867–73. Burke TN, Franca FJ, Ferreira de Meneses SR, Cardoso VI, Marques AP. Postural control in elderly persons with osteoporosis: efficacy of an intervention program to improve balance and muscle strength: a randomized controlled trial. American Journal of Physical Medicine and Rehabilitation 2010;89:549– 56. Wayne PM, Buring JE, Davis RB, Connors EM, Bonato P, Patritti B, et al. Tai Chi for osteopenic women: design and rationale of a pragmatic randomized controlled trial. BMC Musculoskeletal Disorders 2010;11:40. Krebs DE, Scarborough DM, McGibbon CA. Functional vs. strength training in disabled elderly outpatients. American Journal of Physical Medicine & Rehabilitation 2007;86:93–103. Ryushi T, Kumagai K, Hayase H, Abe T, Shibuya K, Ono A. Effect of resistive knee extension training on postural control measures in middle aged and elderly persons. Journal of Physiological Anthropology and Applied Human Science 2000;19:143–9. Campbell AJ, Robertson MC, Gardner MM, Norton RN, Tilyard MW, Buchner DM. Randomised controlled trial of a general practice programme of home based exercise to prevent falls in elderly women. BMJ 1997;315:1065–9.