Effects of Whole Body Vibration on Muscle Strength and Quality of Life in Health Elderly : A Meta-Analysis

Introduction: The literature presents different findings about the vibration training efficacy on muscle performance, even using protocols with similar parameters. Objective: The purpose of this systematic review was to investigate the effects of whole body vibration (WBV) on strength and quality of life in health elderly people, presenting a meta-analisys. Methods: PubMed, CINAHL, SciELO, LILACS and PEDro databases were systematically searched for studies that used WBV in healthy elderly. These searches were supplemented with material identified in references and a qualitative and quantitative analysis was performed to summarize the findings. The search was performed by two independent researchers with a third was selected to solve problems of search disagreement, data collection, and quality score. Results: Nine studies with strength outcome and two studies with quality of life outcome were identified, with sample ranging 21 to 220 elderly, all studies had control groups performing exercises or guidelines. Some studies have shown significant improvements in muscle strength, muscle power, vertical jump height, timed get up and go test and quality of life. Conclusion: The meta-analysis of the findings in these studies shows that WBV could benefit * MFP: Doctoral student, e-mail: mairapessoa@yahoo.com.br DCB: PhD, e-mail: daniellacunha@hotmail.com RBS: MS, e-mail: rafaelabsa@hotmail.com HCMS: Doctoral student, e-mail: helgamuniz@yahoo.com.br HKBF: Doctoral student, e-mail: helen.fisio@uol.com.br ADA: PhD, e-mail: armeledornelas@yahoo.com Pessoa MF, Brandão DC, Sá RB, Souza HCM, Fuzari HKB, Andrade AD. 172 Fisioter Mov. 2017 Dec;30(Suppl 1):S171-81 health elderly, increasing muscle strength and improving the quality of life mainly in functional capacity. The number of publications found in the databanks searched is small, with limitations in design of protocols with a weakness to the interpretation of the findings, suggesting the need of investigation with WBV with well-designed protocols and controlled parameters into the effects of WBV training in elderly people.


Introduction
Vibration is a fast and oscillatory movement (1).It was first used therapeutically in the Soviet Union, in the prevention of hypotonia in cosmonauts.Initially used in segmental form, it evolved to the vibration of the whole body with the use of machines (2), with overall effect.
The Whole Body Vibration (WBV) is an alternative for resistance exercises (3 -5), since stimulates muscle receptors through the vibratory tonic reflex (4,6).This reflex is triggered by the oscillation of muscles and tendons, which causes small and fast changes in the muscle-tendon unit length.These changes are then detected by muscle spindles, which try to avoid muscle stretching by a reflex muscle contraction (6).
The high frequency and low amplitude vibration increases the gravitational force by changes in acceleration (7) benefiting bone morphology and promoting muscle toning (8).This mode of vibration can be used by groups with reduced tonus and mobility, as the elderly (7).
The aging process triggers a muscle, joint, and bone degradation process (9,10), reducing muscle mass and strength up to 40% after the 40th life year (11).Although resistance exercises is the treatment of choice for the reduction of sarcopenia and increase of strength, it is considered a relatively aggressive training for the elderly, due to its wide range of motion and the risk of fractures and strains (12).WBV reduces the risks of weight-lifting training (13), although its effects are still poorly documented and little assessment has been done on the impact on the elderly's quality of life.
Systematic reviews about the specific effects of vibration on bone density (14 -16), balance, strength, Fisioter Mov.2017;30(Suppl 1):S171-82 walk ability and functional mobility of the elderly are found in the literature (14 -17).Although the latter has made a summarized assessment on muscle performance, it does not present a meta-analysis.The literature presents different findings about the vibration training efficacy on muscle performance, even using protocols with similar parameters (14 -17).
The purpose of this systematic review was to investigate the effects of WBV on muscle strength and quality of life on healthy elderly people, presenting a meta-analysis.

Methods
Randomized or quasi-randomized clinical trials that assessing the vibrating platform effects on muscle strength or quality of life, in healthy aged 65 or over, both sexes were included, without linguistic or date restrictions.The search was performed by two independent researchers between December 15 th to December to March 10 th with pre-determined keywords and word crossing.The data were compared between researches, observing material disparity, inclusion/exclusion disagreement, and duplicity of studies.A third researcher was selected to solve problems of search disagreement, data collection, and score quality.

Search Strategy
The research was performed in the PubMed databases via Medline, Scientific Electronic Library Online (SciELO), Latin American and Caribbean Health Sciences Literature (LILACS), Cumulative Index to Nurse and Allied Health Literature (CINAHL) and Physiotherapy Evidence Database (PEDro), with adjustments in each database, including the following MeSH descriptors: (i) "Muscle Strength", (ii) "Aged" and (iii) "Quality of life"."Whole-body vibration" or variations, like "Whole body vibration" or "WBV", does not appear in MeSH and it was inserted as keyword in search.
The strategy used the following variations according to the databases: (muscle strength OR strength OR force OR força OR fuerza) AND (aged OR elderly OR idosos OR ancianos) AND (quality of life OR qualidade de vida OR calidad de vida) AND (whole-body vibration OR whole body vibration OR WBV OR vibração de corpo inteiro OR vibraciones de cuerpo).

Inclusion Criteria for Publication Selection
Controlled and randomized or quasi-randomized clinical trials were analyzed.The potentially eligible studies were assessed by title and abstract, observing strength by dynamometry or quality of life outcome by validated scales.
The WBV training was defined as global sinusoidal vibrations in any axis, non-stochastic, without restrictions on frequency, amplitude, magnitude and dosage (18).The control groups could exercise freely or be oriented.Were excluded studies with subjects diagnosed with any pathology, that using medication for increase muscle strength or follow-up studies.

Qualification of Studies
The qualification of the studies used on The Cochrane Collaboration Reviewers' Handbook, version 5.1.0(19) that evaluates the risk assessment as high, low, or unclear bias, according to the methodological descriptions in each study.The assessed domains in papers were: selection (sequence random generation and allocation concealment), implementation (blinding of participants and evaluators), detection (blinding of each outcome), attrition (assessment of incomplete data), and data reporting (selective information).For judgment, the non-citation of process was considered as high risk, citation without clarification as obscure risk, and operation, description citation as low risk.The scores were independently measured by the researchers and compared.

Data Extraction and Analysis
The data extraction was independently done by the researchers, containing: title, author, year, number of participants, eligibility criteria, group characteristics, exclusions, intervention, and measurement of results.The data was summarized in tables and compared, being combined in a meta-analysis of fixed effect evaluation, after application of the Q Cochran Test (19) for heterogeneity.

Qualitative Synthesis
From the 1912 potentially eligible titles, 1893 were excluded for not filling the inclusion criteria.From the remaining 19, after abstract or text reading, seven were excluded for duplicates, different outcomes, follow-up studies and presence of neuromuscular disease.One more study, Bogaerts, 2007, was excluded for using same sample as Bogaerts, 2009.From the remaining, nine studies verified outcome strength and two verified quality of life outcome (Figure 1).

Strength Outcome
Among the nine studies, four were randomized with method description, four were called randomized but with no description and one was considered quasi-randomized.All of them had active or passive control group.Three studies were self-called blind, with only one describing the method.The studies were performed in Germany (20), Australia (21), Belgium (4, 22 -24), Brazil (25), Spain (26), and Portugal (27).The population was only female (20,22,23,26,27), only male (4), or both genders (21,24,25).The sample sizes ranged from 16 to 220 people and the intervention length was from 1 1/2 to 12 months (Figure 2).
The WBV protocol varied in frequency and dosage.The interventions occurred twice a week, being in some studies three times a week (4,21,22,23,27).Vibration amplitudes ranged from 02 to 10 mm.Strength was measured with an isokinetic dynamometer and in two studies also with a handgrip (20,24).Two studies did not present increase in strength outcome (24,27).The remaining studies obtained significant increases in relation to the control groups, regardless these being active or passive.Among the 09 studies, 04 were randomized with method description, 04 were called randomized but with no description and 01 was considered quasi-randomized 2 All confidence intervals touch the nullity line 3 The two selected studies were classified as contrlled and randomized, although none of them have described the allocation method 4 It was not randon sequence generations 5 There was no blinding of participants, personnal, outcome assessment 6

Incomplet outcome data
-GRADE of the studies.
None of the studies reported adaptation difficulties or adverse effects related to WBV.The dropouts were due to external causes such as holidays or address change, liver cancer, breast surgery or placing of knee prosthesis.Most of the studies had a pre-treatment series for demonstration and adaptation to the WBV.

Quality of Life Outcome
The two selected studies were classified as controlled and randomized, although none of them have described the allocation method.The studies were performed in Belgium (28) and Australia (29).Both studies had male and female participants.The first lasted for 06 weeks and had 42 individuals (28), while the second lasted for 12 weeks and had 73 individuals (29).
The WBV protocol varied in the studies regarding the frequency dosage.In the first study (28), both groups performed exercises for stretching, gait and balance, transfer, and resistance for the lower limbs.In addition, the treatment group performed WBV three times a week, with four series alternating one minute of vibration and 90 seconds of rest.In the first and third series, the frequency was 10 Hz with amplitude of 03 mm.In the second and fourth series, the frequency was 26 HZ with amplitude of 07 mm.In the second study (29), the subjects were randomized for performing zero, one, two or three weekly sessions in a WBV prototype constructed by the researchers, with amplitude of 0,5 mm and frequency of 15 to 25Hz.
In both studies, the quality of life assessment was performed with The Short Form Health Survey (SF 36), which works with eight domains: vitality, physical functioning, bodily pain, general health perceptions, physical role functioning, emotional role functioning, social role functioning and mental health.Bruyere et al. (28) shows gains in the eight health domains, compared with the control group and besides evaluate balance with the Tinetti Test and motor ability with the Timed Up & Go Test.Furness and Maschette (29) compares the scale values to the control group, presenting improvement in vitality and emotional role functioning with WBV three times a week, and increased physical functioning with WBV once a week.It also assesses neuromuscular performance with the same tests and with the 5-Chair Stands Test.
None of the studies reported incompatibility or adverse effects directly related to vibration.However, two subjects quit treatment in the WBV group due to a tingling sensation in the lower limbs (28).Series of WBV adaptation or demonstration are not reported.The results present clinical homogeneity, similar patients, identical investigation question, the same intervention and result measurement, with heterogeneity verified by Chi 2 = 3,09, which is considered methodological heterogeneity.Therefore, it was done a meta-analysis in order to verify the strength effect.The selection criterion for metaanalysis was training chronicity.Previous reviews mention as chronic training periods of three or more months (14,15), and only such studies were included in the meta-analysis.These studies included 324 individuals, 157 using WBV as treatment and 167 participating actively or as control (Figure 3).Heterogeneity: Chi 2 = 3.09, df = 5 (P = 0.69); I 2 = 0% Test overall effct: Z = 2.12 (P = 0.03) Figure 3 -Forest Plot studies of meta-analysis for muscle strength outcome.

Quality of Life Outcome
Both studies presented the same intervention, in the same group, with result measurement done in identical form, using the SF 36 for measuring quality of life.In the quantitative evaluation, 30 individuals were evaluated for the control group and 32 for the WBV group.As the questionnaire outcomes are eight, eight evaluations were performed (Figure 4).

Discussion
Despite the low methodological quality of the studies, the quantitative analysis suggests that WBV training is effective for the improvement of strength in the elderly, with good statistic power, with chances of being an alternative for resistance training.Some methodological considerations should be recognized.This study is the sum of single results of randomized and quasi-randomized clinical trials and reflects their quality.In this sense, it should be noted the low methodological quality of the selected clinical trials, most of them lacking sample size calculation, concealment description or randomization method or blinding method in the outcome analysis.
For the strength outcome, the evaluation was performed with isokinetic dynamometry, considered an evaluation method that generates objective data, high sensibility, reproducibility and specificity for the strength and balance measurement of muscle groups.Furthermore, it is a low cost and fast implementation method (30 -32).The isokinetic strength evaluation is considered the best method for determining the functional pattern of muscle strength and balance, that way the subject performs maximal or submaximal contraction, which adapts itself to the device resistance (33 -35).
The estimate following the statistic combination of the clinical trials pointed to the presence of benefits for the strength outcome, according to the graph, although in all studies the confidence interval has touched the no effect line (Figure 3).
In a quantitative analysis, a couple of studies (24,27) do not show increase in strength outcome with pvalue < 0,05.Despite having the best methodological quality, Bautmans's study (24) was performed in the shortest time, just 1.5 months, while the others ranged from 02 to 12 months.The WBV is an unspecific training operating in a mechanism that includes slow physiological adaptations, which is the case of strength improvement.Therefore, for an effective outcome, it should last slightly more than the specific training, which occurs around 1.5 to 2 months.As a specific protocol for WBV is not available yet, studies with any training time, acute or chronic, were included, but only the latter ones were part of the meta-analysis, assuming that the strength outcome, if present, would occur only after a chronic period of training.
Another possible variant in this study outcome may have occurred due to pre-existing differences between the intervention and control groups.(36 -38).
Still in this study (27), comparing the walking and WBV groups, it was observed an improvement in the muscle strength associated to daily activities (walk, sit or climb stairs) essential for postural stability (33), while the WBV program improved the jump explosive strength associated to a fast muscle contraction, important to fall prevention in the elderly.
The other studies (4,20,21,22,23,25,26) present an increase in the strength outcome.These findings reinforce the theory that WBV is an option which produces adaptive results similar to resistance training (3,6) leading to improvement of strength.In theory, the WBV improves neuromuscular efficiency by fast activation of response circuits to changes in muscle position and promotion of direct action upon contraction, in such a way that it would increase voluntary movement efficiency (39 -41).
In quality of life, although the means have shown significant increases, at least in functional capacity, the graphs for each outcome do not make these results clear.In Bruyere's study (28) the differences between the means and standard deviations are big, what interferes in the final graph result.In the second study, Furness and Maschette (29) results had smaller standard deviations, however, their sample had only ten subjects training three times a week.The small number of individuals and the big standard deviation do not permit the diamond visualization in the metaanalysis (Figure 4).
Vibration proved to be an exercise well accepted by the elderly and with no adverse effects.The low to moderate frequency vibration safety is normatized by ISO 2631-1, 1997 (18).Although a training protocol has not been created yet, all studies were performed according to safety parameters concerning exposition time and frequency of the device.

Conclusion
Despite the low methodological quality of the studies, the quantitative analysis suggests that WBV training is effective for the improvement of strength in the elderly, with good statistic power, with chances of being an alternative for resistance training, once there is no need of active contraction.Its use is technically easy and shows positive and fast results, being one more alternative to the therapeutic arsenal that can slow down the decrease of strength in the elderly.Regarding quality of life, the WBV was effective only in improving physical functioning.The recommendation is that there is evidence for WBV application in the studied outcomes, although further studies are needed to evaluate other effects such as cardiopulmonary and hormonal functions, because being a global training, could promote changes in the physiology of several systems.

ScreeningFigure 1 -
Figure 1 -Search and selection of studies for systematic review in accordance with PRISMA.

FisioterFigure 4 -
Figure 4-Forest Plot studies of meta-analysis for quality of life outcome, eight domains.
Should Whole body vibration vs control groups, performing exercises or guidelines be used for strength and quality of life in health eldely people?
Fisioter Mov.2017;30(Suppl 1):S171-82 Question: Absolute Muscle strength (follow-up mean 6 to 48 weeks; Measured with: Isokinetic dynamometer/handgrip; Range of scores: 0-15; Better indicated by higher values) 9 IMPORTANT Physical function outcome in elderly (follow-up 6 to 12 weeks; Measured with: Short Form Health Survey (SF 36); Range of scores: 0-100; Better indicated by higher values) 2 CRITICAL Social function outcome in elderly (follow-up 6 to 12 weeks; Measured with: Short Form Health Survey (SF36); Range of score: 0-100; Better indicated by higher values) 2 CRITICAL Role physical outcome in elderly (follow-up 6 to 12 weeks; Measured with: Short Form Health Survey (SF36); Range of score: 0-100; Better indicated by higher values) 2 CRITICAL Role emotional outcome in elderly (follow-up 6 to 12 weeks; Measured with: Short Form Health Survey (SF36); Range of score: 0-100; Better indicated by higher

Table 1 -
Methodological assessment of studies selected for the strength and quality of life outcomes according to Cochrane Collaboration Reviewer's Handbook, 5.1.0 The strength outcome varies widely, being the WBV group initial strength 270.0 ± 203.8 Newtons, against 375.2 ± 253.8 Newtons in the control group.Even considering the standard deviation values, there is a big difference in the basal values, what could explain the final difference in the outcomes of the two groups.Raimundo et al. (27) also does not increase of strength in the groups (vibration/walks) according to the isokinetic dynamometry, although it notes significant positive values when measuring stand up speed, 4 meter walking speed and vertical jump height in relation to the pre-test.This means that although the gains have not been identified in numerical terms, functional capacity has improved.Besides, exercises like walks and low frequency vibration result in electromyographic activity of low to moderate amplitude, what can justify the lack of results in the isokinetic strength, once it requires high electromyographic activity