The effects of whole body vibration in patients with type 2 diabetes: a systematic review and meta-analysis of randomized controlled trials

Background: Whole body vibration (WBV) has been used to increase physical activity levels in patients with type 2 diabetes mellitus (T2DM). Objective: To carry out a systematic review of the effects of WBV on the glycemic control, cardiovascular risk factors, and physical and functional capacity of patients with T2DM. Method: MEDLINE, LILACS, PEDro, and Cochrane Central Register of Controlled Trials were searched up to June 1st, 2015. Randomized controlled trials investigating the effects of WBV, compared to control or other intervention, on blood glucose levels, blood and physical cardiovascular risk factors, and physical and functional capacity in adult individuals with T2DM. Two independent reviewers extracted the data regarding authors, year of publication, number of participants, gender, age, WBV parameters and description of intervention, type of comparison, and mean and standard deviation of pre and post assessments. Results: Out of 585 potentially eligible articles, two studies (reported in four manuscripts) were considered eligible. WBV interventions provided a significant reduction of 25.7 ml/dl (95% CI:-45.3 to -6.1; I2: 19%) in 12 hours fasting blood glucose compared with no intervention. Improvements in glycated hemoglobin, cardiovascular risk factors, and physical and functional capacity were found only at 12 weeks after WBV intervention in comparison with no intervention. Conclusion: WBV combined with exercise seems to improve glycemic control slightly in patients with T2DM in an exposure-dependent way. Large and well-designed trials are still needed to establish the efficacy and understand whether the effects were attributed to vibration, exercise, or a combination of both.


Introduction
Physical activity plays an important role in prevention and control of type 2 diabetes mellitus (T2DM) and its related complications 1 . Both aerobic and resistance training improve insulin action and can assist with the management of blood glucose levels, lipids, blood pressure, cardiovascular risk, mortality, and quality of life; however, exercise must be undertaken regularly for continued benefits 1,2 . Nevertheless, most of people with T2DM are not active, mirroring the inertia of a lifetime of habits and motivational barriers such as lack of interest, lack of time, and depression 3 . In addition, physical disabilities and perceived discomfort when exercising are challenges to adherence to physical activity 3,4 .
Physical therapists are able to help people plan an individualized exercise program in order to maintain good blood glucose and achieve optimal weight 5 . To help people with diabetes improve their quality of life, physical therapists can intervene with physical treatment techniques such as manual or manipulative treatments, electrophysical agents, and mechanical agents 5,6 .
Among the alternatives aimed to increase overall physical activity, whole body vibration (WBV) has been shown to be a new effective option in healthy subjects and individuals with several health conditions 7 . It is assumed that vibration activates muscle spindles and evokes muscle contractions induced by a complex spinal and supraspinal neurophysiological mechanism known as tonic vibration reflex, allowing muscular activity enhancement even in static positions 8 .
Some systematic reviews 9-14 summarized the effects of WBV in some outcomes of specific populations as follows: improvements in bone mineral density in postmenopausal women 9 ; leg muscle strength 10 and balance improvement in older individuals 11 ; balance, gait, and proprioception improvement in individuals with neurological conditions such as Parkinson's disease, multiple sclerosis, and stroke 12 ; pain intensity decrease and physical function enhancement in individuals with knee osteoarthritis 13 ; and functional exercise capacity and quality of life improvement in people with chronic obstructive pulmonary disease 14 . Furthermore, WBV requires significantly less time than conventional training and, therefore, reached a satisfactory compliance in previously inactive patients 11 .
Nevertheless, the effects of WBV in patients with T2DM were infrequently reported through a case report 15 and acute 16 , crossover 17 , or pilot 18,19 studies. In the last years, few randomized controlled trials were performed 20,21 with conflicting results. To summarize the current evidence, we aimed carry out a systematic review of the effects of WBV intervention on the blood glucose levels, blood and physical cardiovascular risk factors, and physical and functional capacity of individuals with T2DM in comparison with a control or other intervention group.

Method
This systematic review was performed in accordance with the Cochrane Handbook for Systematic Reviews of Interventions 22 and the recommendations of the Brazilian Journal of Physical Therapy tutorial 23 . The protocol of this systematic review was prospectively registered at PROSPERO under the identification CRD42014010495 and can be assessed online 24 .

Data sources and searches
Comprehensive literature searches were performed on the following electronic databases (from inception to June 1 st , 2015): MEDLINE (accessed by PubMed), LILACS, Physiotherapy Evidence Database (PEDro), and Cochrane Central Register of Controlled Trials (Cochrane CENTRAL). The search terms included 'Whole body vibration', 'Diabetes' MeSH and synonyms, and a string of terms to optimize randomized controlled trial searches on PubMed 25 . In order to improve sensitivity, outcomes were not included in the search strategy. The references list of the articles identified in these searches were used as an additional source to identify other potentially eligible trials. The search strategy used on PubMed database can be fully assessed online 26 .
Randomized controlled trials were considered eligible if they addressed the effects of WBV on blood glucose levels, blood and physical cardiovascular risk factors, and physical and functional capacity in adult patients with T2DM, with a minimum of four weeks intervention and at least a control group not performing WBV. We considered as the primary outcome blood glucose levels, assessed by 12-hours fasting blood glucose  or glycated hemoglobin (HbA1c). The secondary outcomes were blood and physical cardiovascular risk factors (blood cholesterol and triglycerides, atherogenic index, body mass index, body composition, weight, waist circumference, waist to hip ratio, blood pressure, or heart hate) and physical and functional capacity (maximal oxygen uptake, six-minute walk test (6MWT) distance, muscle strength, or static and dynamic postural balance). The exclusion criteria were studies that included individuals with stated diabetic complications (e.g. diabetic peripheral neuropathy, retinopathy, or nephropathy) and studies with an unreliable description of WBV.

Study selection
Two independent reviewers screened the titles and abstracts of all studies identified through the search strategies. A standard screening checklist based on the eligibility criteria was used for each study. Studies that did not meet the eligibility criteria, according to titles or abstracts, were excluded. The two independent reviewers retrieved full text versions of the remaining studies for a second review. There were no disagreements between reviewers.

Data extraction and quality assessment
Two reviewers independently extracted the data from the eligible studies by using a standardized data extraction form. The following data were extracted: authors; year of publication; number of individuals analyzed; gender; age; parameters of WBV and description of intervention; type of comparison; mean and standard deviation of pre and post assessments of each outcome available. From articles referred to the same participants, the article with the larger sample was considered and the article with the smaller sample was excluded if outcome measurements were duplicated. There were no disagreements between reviewers. HbA1c and 12-h FBG mean and standard deviation values were not available in one published study 14 , but the authors informed these estimates by email.
The studies were assessed regarding methodological quality and statistical reporting using the PEDro scale 27 . When methodological quality assessment was not available on the PEDro database, two reviewers performed the ratings using the Brazilian Portuguese version of the PEDro scale 27 items. In addition, the quality of each article was evaluated based on the recommendation of the International Society of Musculoskeletal and Neuronal Interactions (ISMNI) 28 for reporting WBV intervention studies, consisting of 13 minimal reporting items about the WBV parameters and participant positioning. The instruments were rated independently by two reviewers. There were no disagreements between reviewers.

Data synthesis and analysis
After data extraction, if the outcome values could not be transformed into a common numeric scale for quantitative synthesis, a descriptive synthesis was performed. For quantitative synthesis, pooled-effect estimates were obtained by comparing the change from baseline to study end for intervention and control group. The procedures for estimation of missing data 22 were performed to obtain the standard deviation difference. Results were presented as weighted mean difference (WMD) with their respective 95% confidence intervals (CI). Meta-analysis was performed using the random effects model. The statistical heterogeneity among studies was assessed using Cochran's Q test and the inconsistency I 2 test, in which values above 25% and 50% were considered as indicatives of moderate and high heterogeneity, respectively. Sensitivity analysis was not possible given the number of available studies, therefore when I 2 >25%, meta-analysis was not considered. A p value lower than 0.05 was considered statistically significant. All analyses were conducted using Review Manager, version 5.2.

Description of studies
The search strategy yielded 585 articles. From these, eight 16,[19][20][21][29][30][31][32] were considered as potentially relevant and retrieved for a detailed analysis. After full-text reading, four articles were excluded. As three articles 21,31,32 referred to the same original study (clinical trial register: ACTRN12613000021774), they were considered as a single study. From this, two studies reporting outcomes on four different articles 20,21,31,32 were included in this systematic review.   shows the flow diagram of the studies and Table 1 summarizes their characteristics. A total of 70 participants with T2DM were assessed. The year of publication of the included studies ranged from 2011 to 2014. Both the studies included individuals with T2DM diagnosis and excluded those with established diabetes complications and HbA1c>10% or fasting blood glucose >250 ml/dl. Age ranged from adult to elderly classification and only males were included by Behboudi et al. 20 while the other study 21,31,32 included both genders. All of the studies randomly allocated the individuals to a control group without additional intervention, keeping normal daily activities and medical instructions. In addition, Behboudi et al. 20 randomly allocated individuals to a third group that performed an increasing aerobic exercise (AE) program only.
Regarding WBV intervention, both studies 20,21,31,32 applied an intermittent exposure to WBV and acceleration and frequency parameters were very similar. Studies kept peak acceleration between 1 and 2 g (units of gravity; 1g=1 m.s -2 ). In Behboudi et al. 20 , the peak acceleration was influenced mainly by higher vibration frequencies and lower amplitude, but in Sañudo et al. 21 and Del Pozo-Cruz et al. 31,32 , higher amplitude and lower vibration frequencies determined the peak acceleration.
Both the studies proposed a thrice-weekly intervention of WBV with total session duration increasing progressively from 12 (8-16) to 14 (16)(17)(18)(19)(20)(21)(22)(23)(24) minutes. All of the studies designed protocols in which individuals stood on the vibrating platform in a 100 to 110° squat position (considering total knee extension as 180°) and the vibratory stimulus was not isolated. Behboudi et al. 20 proposed WBV in addition to an increasing AE program (WBV+AE) with a follow-up after eight weeks. The study reported by Sañudo et al. 21 and Del Pozo-Cruz et al. 31,32 proposed a protocol of lower and upper limb exercises performed on the vibrating platform with a follow-up after 12 weeks.
No adverse effects were reported in any of the studies. Loss of follow-up occurred only in the assessment after 12 weeks 21,31,32 , in which five participants from the control group dropped out because of lack of interest. Six participants from the intervention group dropped out because of lack of time (five participants) and change of home address (one participant). Participants attended more than 75% of the sessions in both trials 20,21,31,32 .
Overall, the methodological quality assessed by the PEDro scale was low to moderate ( Table 2). Table 3 shows the quality of each article based on the recommendation of the ISMNI 28 for reporting WBV intervention studies.

Blood glucose levels
For 12-h FBG, meta-analysis was performed and included data of two trials 20,32 with a total of 59 patients (29 of which were on WBV). The comparison groups did not perform any intervention. There was an improvement in 12-h FBG by reduction in 25.7 ml/dl (95% CI: -45.3 to -6.1; I 2 : 19%), favoring WBV intervention (Figure 2A). There was no additional effect (p=0.09) of WBV to an eight-week increasing AE program regarding 12-h FBG, but both the groups (WBV+AE and AE only) presented significantly lower 12-h FBG levels (p=0.02) than the control group 20 .
Regarding HbA1c, a meta-analysis was not considered given an I 2 of 80% between studies. After the 12-week program of upper and lower limb exercises performed on the vibrating platform, participants in the intervention group exhibited significantly lower levels of HbA1c (p=0.002) at the time of follow-up when compared to the control group, with a mean difference of −0.55% (95% CI: −0.15 to −0.76) 32 . The eight-week WBV+AE program was not sufficient to promote a significant difference in HbA1c levels compared to the control group. Furthermore, there was no additional effect of WBV on the eight-week AE program as no significant difference in HbA1c levels was found between WBV+AE and AE only. Both intervention groups did not differ significantly from controls.

Blood and physical cardiovascular risk factors
Regarding secondary outcomes, a meta-analysis was only possible for Body Mass Index (BMI). Data of two studies 20,31 with a total of 59 patients (29 of which were on WBV) were included and comparison groups did not perform any intervention. A non-significant decrease of 0.67 Kg.cm -2 (95% CI:-2.21 to 0.87; I 2 : 8%) in BMI was observed ( Figure 2B).
After the 12-week program of upper and lower exercises performed on the vibrating platform, a significant decrease (p<0.050) was found in cholesterol, triglycerides, atherogenic index 32 , weight, waist circumference, waist-to-hip ratio, and body fat percentage 21 compared to the control group. However, no statistically significant changes were detected for high-density lipoprotein (LDL), low-density lipoprotein (LDL), or LDL/HDL 32 . After the eight-week WBV+AE program, no significant differences in body fat percentage were found compared to the control group or compared to the AE group 20 .

Physical and functional capacity
Improvements (p<0.05) were found in the 6MWT distance and muscle strength assessed by the 30-second Sit-to-Stand (30s-STS) test after the 12-week WBV program with upper and lower limb exercises compared with the control group. Regarding static balance, the same comparison showed a significant decrease in center of pressure excursions with eyes closed (feet apart and together), but TUG time did not improve significantly 32 . Although maximal oxygen uptake increased significantly (p=0.01) after the eight-week WBV+AE and AE only programs, WBV had no additional effect on AE (p=0.3) 20 .

Summary of evidence
It seems that the 12-week progressive intervention with WBV and exercise was sufficient for a statistically significant, but slight improvement in the 12-h FBG and HbA1c of individuals with T2DM, in comparison with no intervention. Furthermore, the eight-week intervention improved 12-h FBG, but did not improve HbA1c.
Because erythrocytes are freely permeable to glucose, the level of HbA1c in a blood sample provides a glycemic history of the previous 120 days, the average erythrocyte lifespan 33 . It is possible that a period of eight weeks was not enough to reach modifications in blood glucose profile, as no significant alterations were found in the WBV+AE or AE only programs.
The effect size for HbA1c improvement discovered after the 12-week progressive intervention with WBV and exercise was close to the one found after aerobic or resistance training reported previously in two meta-analyses 34,35 . Although the vibratory stimulation was not isolated from exercises in the proposed interventions, session duration was considerably lower in the WBV studies (8 to 24 minutes) than in the aerobic or resistance training studies (40 to 75 minutes) 34,35 . This fact corroborates other studies that found similar results in WBV application compared conventional intervention, but in a shorter time of exposure [36][37][38] .
The meta-analysis for BMI found no significant decrease after the WBV interventions. According to Cochrane 39 , although WBV has gained popularity as a modality for weight loss, it does not have the ability to generate large energy expenditure to substitute conventional aerobic exercise. However, it had positive effects on blood flow 32,40 that could indirectly improve associated diseases such as hypertension. In fact, this could be seen in some of the blood and physical markers of cardiovascular risk (cholesterol, triglycerides, atherogenic index, body weight, waist circumference, and waist-to-hip ratio) that improved after 12 weeks of progressive intervention with WBV combined with exercises 32 .
It seems that an eight-week WBV intervention was not enough to reach significant improvements in the aerobic capacity 20 of patients with T2DM. In contrast, the 12-week progressive intervention with WBV and exercise improved aerobic capacity measured by the 6MWT distance, with similar values to those found in a multi-center study on fitness among healthy elderly subjects 41 . The same improvement was found in lower limb strength measured through the 30s-STS. It is possible that the time of exposure in patients with T2DM must be greater than that required for the non-diabetic population. For example, a previous meta-analysis found a significant beneficial effect of WBV on lower limb strength of elderly subjects with a treatment effect comparable to other forms of active exercises (e.g. resistance training) within six to 10 weeks 10 .

Limitations and conclusions
This is the first systematic review to synthesize the effects of these outcomes in individuals with T2DM after WBV interventions. Analysis from data extraction of this systematic review was limited by the small number of available trials and duplicated articles. Furthermore, results from this systematic review must be interpreted with caution as most of the trials have some methodological limitations such as lack of concealed allocation and intention-to-treat analysis. Regarding the minimal items required for WBV intervention reproducibility, clear reporting is still necessary of the type of vibration, whether amplitude displacement was peak-to-peak, the peak acceleration, whether and how accuracy of vibration parameter were assessed, whether and how skidding of the feet were avoided, what was the rationale for choosing specific vibration parameters, whether and what support devices were used during vibration exposure and whether the type of footwear was controlled. Failing to report those items impairs protocol reproducibility as well as protocol comparison 28 .
Despite the slight beneficial effect of WBV intervention on glycemic control, a paramount outcome for T2DM management, caution is required in extrapolating this result to practice. First, a significant reduction in glycemic values was found in comparison with no intervention and WBV was not investigated alone, but in addition to exercise. Similar caution must be taken regarding blood markers and functional capacity. Even if WBV parameters were very similar between studies, the combined exercises differed between studies and follow-up was also distinct, which may have influenced pooled effects and heterogeneity. It is necessary to highlight that these implications should only be considered for patients with T2DM without reported complications or contraindications for WBV exposure as well as glycemic profile <10% for HbA1C or <250 mg/dl for 12-h FBG. Furthermore, it seems that effectiveness of WBV is exposure-related as the 12-week intervention presented the better results.
Similarly to other studies that used WBV as an intervention in sedentary or elderly individuals 10 , there was good adherence and compliance in the 8-week and 12-week follow-up assessments. There was similar loss of follow-up in the intervention and control groups in the 12-week WBV program related to personal reasons 21,31,32 . Adverse effects, such as hypoglycemia, discomfort, and musculoskeletal injuries, are highly reported in studies performing exercise interventions 34 , however they were not reported in the studies included in this systematic review 20,21,31,32 .
WBV performed close to the parameters presented in the primary studies and combined with low-level exercises seems to be a safe, feasible, and less time-consuming intervention to help improve the glycemic control, cardiovascular risk markers, and functional capacity of individuals with T2DM in an exposure-dependent way compared to no intervention. However, given the methodological weaknesses of the primary studies and the heterogeneous protocols, confidence is limited on the decreasing effect of WBV on 12-h FBG. Further well-designed trials are still required to strengthen the current evidence and clarify whether the effect should be attributed to vibration, exercise, or the combination of both.