Plantar foot pressure is a risk factor for the development of foot ulcers in persons with diabetes mellitus. PURPOSE: The objective of this study was to examine the effects of local vibrations on plantar skin blood flow (SBF) responses during weight-bearing standing. Wavelet analysis of plantar SBF was used to analyze microvascular regulation in response to standing with and without local vibrations. METHODS: Fifteen (15) healthy participants (26.5 ± 5.7 years; 4 male and 11 female) received a local vibration intervention (35 Hz, 1 mm, 2 g vibration) and a sham vibration to the skin of the right first metatarsal head during 10-minute standing. Laser Doppler flowmetry was used to measure SBF before and after 10 minutes of standing. SBF after standing was expressed as a ratio of SBF before standing to minimize blood flow variations. The use of wavelet analysis allowed the authors to examine the frequency bands corresponding to the physiological controls in the vibration and sham areas of the foot, including metabolic (0.0095–0.02 Hz), neurogenic (0.02–0.05 Hz), myogenic (0.05–0.15 Hz), respiratory (0.15–0.4 Hz), and cardiac (0.4–2.0 Hz) regulations. RESULTS: Plantar SBF ratio changes in the vibration protocol (1.83 ± 0.27) were significantly higher compared with the sham protocol (0.97 ± 0.08) (P < .01). SBF before and after the 35 Hz vibrations were 41.96 ± 14.02 perfusion units (range 6.68–208.9 perfusion units) and 61.16 ± 14.74 perfusion units (range 7.76–155.37 perfusion units), respectively. SBF before and after the sham vibration were 37.32 ± 9.29 perfusion units (range 5.74–120.44 perfusion units) and 33.97 ± 8.11 perfusion units (range 6.95–108.44 perfusion units), respectively. Wavelet analysis of SBF oscillations showed a significant difference in all regulations: metabolic (P < .05), neurogenic (P < .05), myogenic (P < .05), respiratory (P < .05), and cardiac (P < .05) in response to 35 Hz local vibrations compared with the sham vibration. CONCLUSIONS: Local vibrations (35 Hz frequency, 1 mm amplitude) to the plantar tissues during 10 minutes of weight-bearing standing resulted in a significant increase in after-standing plantar SBF compared with sham vibration. The control mechanisms contributing to this increase in SBF were metabolic, neurogenic, myogenic, respiratory, and cardiac regulations. These findings confirm results of preclinical studies and support the need for additional research to examine the potential protective effects of local vibration to decrease the risk of plantar ulcers. 

The Centers for Disease Control and Prevention indicated that diabetes was already the seventh leading cause of death in the United States in 2015.¹ Diabetic foot ulcers (DFUs) are a major complication of diabetes. It was estimated that an annual incidence rate of DFUs was around 2%, and the lifetime incidence rate was reported between 15% and 25%. DFUs can lead to amputation of the lower extremity, premature death, and expensive health care costs.^2–5

The pathogeneses of DFUs are complicated and include diabetic neuropathy, microvascular dysfunction, foot deformity, and high plantar pressure.^3,6 Standing, a weight-bearing activity, is an essential position during activities of daily living.^7,8 Weight-bearing activities result in plantar pressure over the metatarsal head and heel of the foot. If the plantar pressure is beyond the threshold that the plantar tissue can withstand, this pressure may cause plantar tissue ischemia and even DFUs, especially in people with diabetes and peripheral neuropathy because they cannot sense discomfort and pain in the foot.^9,10 The average peak plantar pressure during standing is about 140.5 kPa in healthy adults; as has been shown in preclinical studies similar to our study design, the peak plantar pressure of people with diabetes is 1.5 to 2.3 times more than people who do not have diabetes.^11,12 The normal peak systolic blood pressure is about 17 kPa, which is much lower than peak plantar pressure during standing. This implies that plantar skin blood flow (SBF) could easily be occluded during prolonged standing.^11,13 Studies showed that a duration of 2 minutes was the minimum time of occlusion to induce a hyperemic response in the plantar foot in healthy people.¹⁴ It was demonstrated in preclinical studies that people with diabetes spent 13.5% of their time in the standing position, which was twice as much as time spent walking (6.1%).⁷ However, the literature focuses on walking-related research rather than standing in people with diabetes. Thus, there is a significant need to study the role of the standing position in the risk of DFUs for people with diabetes. The prolonged plantar stress during standing could result in plantar skin ischemia. Therefore, interventions that can reduce the effects of weight-bearing load on the plantar foot during standing may reduce the risk of DFUs.

The concept of preventive intervention, that is, first exposing tissues to a non-lethal dose of a form of stress (eg, mechanical or oxidative) to protect the tissues against a subsequent lethal dose of the stress, has been successfully applied to prevent soft tissue injury in various conditions.^15–19 Preventive interventions include cooling, ultrasound, electrical stimulation, and vibrations. Silvia demonstrated that preconditioning with low-intensity ultrasound improved tissue perfusion in ischemic limbs of rabbits.¹⁵ The author indicated that such a protective response was mainly due to increased shear stress caused by ultrasound. Solis et al¹⁶ applied constant pressure (38% of the body weight of each rat for 2 hours) to the quadriceps muscle of rats with  intermittent electrical stimulation (10–40 mA, 250 µs, 50 pulses/s, 10s stimulus bout) at 5-minute or 10-minute intervals. The results showed that a combination of intermittent electrical stimulation significantly reduced the deep tissue injuries of the muscles during compression.¹⁶ Wong et al¹⁹ applied compressions at 100 mm Hg for 6 hours on the biceps femoris muscles with intermittent vibrations (35 Hz, 0.25 g acceleration for 10-minute intervention and 30-minute rest) in senescence-accelerated mice (physiologically relevant aged animal model) and a control group (healthy mice). The authors demonstrated that the application of intermittent vibrations during compression reduced the compression-induced muscle breakdown. Their results suggested that intermittent vibrations could serve as a protective intervention to reduce detrimental effects of prolonged compression on muscle tissues. Di Carlo et al¹⁷ also suggested that an exposure to mechanical vibration (60 Hz, 1 g, 20 minutes) could protect embryos from oxidative stress, hypoxia, and ultraviolet light. Jan et al^18,20 demonstrated that local cooling (decreasing skin temperature by 10˚C) of sacral skin reduced skin ischemia caused by a load (60 mm Hg for 20 minutes) in people with spinal cord injury. These studies have shown that cooling, ultrasound, electrical stimulation, and vibrations could be used as preconditioning methods to reduce tissue injuries caused by compressions.^15,21 However, the majority of these studies were conducted in animals.^15–17,19 It is unclear whether these potentially preventive interventions could be reproduced in humans, especially during activities of daily living (eg, standing and walking).

Research studies have demonstrated that appropriate vibrations could improve SBF, postural balance, muscle strength, and neuromuscular activity.^22–26 Low-frequency (< 50 Hz) whole-body vibrations have been demonstrated to improve peripheral cardiovascular functions.^27–29 Maloney-Hinds et al²⁴ demonstrated that both 30 Hz and 50 Hz local vibrations (5–6 mm amplitude, a peak acceleration of 7 g) resulted in an increase in SBF of the arm in healthy people. Stewart et al²⁵ indicated that low-amplitude plantar vibrations (frequency at 45 Hz, amplitude of 0.2 g) increased peripheral and systemic blood flow in perimenopausal women. Arashi et al³⁰ suggested that the use of the local vibrations at 47 Hz could increase blood flow and improve wound healing. To the authors’ knowledge, there are currently no studies exploring whether local vibrations can improve plantar SBF during weight-bearing activities such as standing.

The physiological mechanism associated with vibration interventions may be explained by reactive hyperemia, a sudden increase in blood flow after a prolonged compression.³¹ Reactive hyperemia is a protective response to an ischemic stimulus. The level of ischemia of tissue could be reflected as a degree of reactive hyperemia.^18,32 When the ischemic stimulus is smaller, the reactive hyperemia response would be correspondingly smaller.¹⁸ The mechanical stress acting on the plantar foot during weight-bearing standing reduces plantar SBF. It is unclear whether local vibrations applied to the plantar foot could reduce the level of ischemia during weight-bearing standing. However, by comparisons of reactive hyperemic responses to standing and standing with local vibrations, the effect of local vibrations on weight-bearing tissue could be investigated.

It is important to explore the effects of local vibrations on plantar SBF during weight-bearing standing in healthy people first, before performing the experiment on people with diabetes, so that the safety of these local vibration interventions can be established.¹⁴ The aims of the current study were to 1) examine the effect of low-intensity local vibrations on plantar SBF during weight-bearing standing and 2) investigate the control mechanisms of plantar SBF in response to local vibrations. 

Participants. This was a before-and-after, repeated-measures clinical study. A recruitment flyer was used to enlist students at the University of Illinois at Urbana-Champaign, Champaign, IL. The authors conducted this pilot study to explore whether these vibrations could affect plantar SBF responses. The study was conducted between September and November 2019. All participants were healthy and free from skin disease, cardiovascular disease, neuropathy, orthopedic disease, and other diseases. Those who were exposed to chronic vibration stimulus (arm-hand vibration syndrome) and those who were routinely exposed to whole body vibrations were excluded. The detailed procedures of the experiment were introduced to all recruited participants. The participants signed the informed consent documents before the experiment. The collected data were coded for data analyses. The code for each participant’s name was not available to the research team. This study (no. 20322) was approved by the Institutional Review Board of the University of Illinois at Urbana-Champaign. 

Instrumentation. Local vibrations (2 g peak acceleration, 1 mm amplitude) at 35 Hz were applied with a vibrator. The vibrator consisted of a voice coil motor, a controller, and a power supply. The voice coil motor (YLM40-20; JDStek, CA) was chosen because of its structural stability with high positioning resolutions, fast acceleration/deceleration, and high-speed capacity. The voice coil motor could be controlled through applying voltage. The position sensor and controller (MS 15 TTLx20; RSF Elektronik, Tarsdorf, Austria) was used to control the voice coil motor with the resolution at 0.5 µm (1 mm for 2000 counts). The radius of the vibrator head was 1 cm. 

The majority of studies in the literature applied low-intensity and low-frequency (below 50 Hz) local vibrations to participants and induced positive results such as improved SBF. It was demonstrated that intermittent vibrations at 35 Hz could protect aged muscle from prolonged mechanical compression in mice.^19,24,25,30 According to the findings of these studies, we selected 35 Hz vibrations as an intervention in the current study. A sham control (0 Hz vibration) was used in this study. 

To ensure equal weight-bearing between the right and left foot, each subject stood on the F-scan plantar pressure system (Tekscan; South Boston, MA). The real-time plantar pressure distributions were displayed on the computer screen to guide the user to place equal weight on both feet. Each F-scan sensor contained 960 sensing pixels, and the size of each pixel was 5.08 mm × 5.08 mm.

The SBF under the first metatarsal head of the right foot was measured noninvasively and continuously by laser Doppler flowmetry (LDF) (PeriFlux 5001; Perimed, Las Vegas, NV). The LDF was calibrated and warmed for 30 minutes before measuring SBF.

Procedures. All participants relaxed for at least 30 minutes in the laboratory to acclimate to the temperature of the room, which was 24 ± 2°C. During this period, participants were asked to complete the demographic and medical history form. The information was used to determine the eligibility of the participant. For example, if a participant checked a cardiovascular disease under medical history, he or she would be excluded from this study.

Participant height, weight, and blood pressure were measured. Participants were randomly assigned to 1 of 2 protocols (ie, standing with local vibrations and standing with sham vibration). The authors created 16 cards, including 8 vibration-first cards and 8 sham-first cards. Each time, researchers picked a card for the participant before the experiement. The LDF probe was taped to the first metatarsal head while the participant was sitting on a chair with hip and knee at 90-degree flexion and ankle in a neutral position. Participants were standing during the vibration procedure. Participants were informed that 2 types of vibrations were being tested in this study, but they were not aware which protocol was being applied. The experimental procedures were as follows: 10-minute baseline SBF measurement, 10-minute vibration intervention (35 Hz or sham control), and 10-minute recovery SBF. A 30-minute washout period was allowed between the 2 protocols (35 Hz vibration and sham control).

Data analysis. To understand the mechanisms responsible for the recovery SBF response after standing and standing combined with local vibrations (35 Hz), the authors performed wavelet analysis of the SBF data.^33,34 The following calculation procedures were used.

For an SBF signal, the continuous wavelet transform is defined as: 

(Eq. 1)

where Ψ is the mother wavelet function,   t is time, and s is the scale related to the central frequency of Ψ_s,t. In this study, the Morlet wavelet was used as the mother wavelet function, defined as:

(Eq. 2)

By choosing  ω₀=2π, s is related to the central frequency of Ψ_s,t , f  by s = 1/f . The mean wavelet amplitude of a characteristic frequency, denoted as A[f₁, f₂], was defined as the mean absolute value of the wavelet coefficients over time and over the frequency interval [f₁, f₂]. In the present study, because the signals lasted 10 minutes, the authors did not take the frequency component (0.005–0.0095 Hz) into account. Only 5 characteristic frequencies were investigated: metabolic (0.0095–0.02 Hz), neurogenic (0.02–0.05 Hz), myogenic (0.05–0.15 Hz), respiratory (0.15–0.4 Hz), and cardiac (0.4–2.0 Hz). Their relative amplitudes were computed as:

(Eq. 3)

Wavelet analysis was performed using Matlab³⁴ (MathWorks, Natick, MA). 

Mean baseline SBF and recovery SBF were determined as an average over a 10-minute previbration and postvibration period, respectively. The SBF ratio was calculated to determine the changes of plantar SBF after each standing protocol to minimize the variations of SBF.^34,35 The two-sample t test was used to compare the SBF ratio and ratio of wavelet amplitudes of the 2 protocols. All data were expressed as means ± SD. Confidence interval was 95%, and P < .05 was considered as statistical significance. The statistical analyses were performed using SPSS Data Analysis Software (Version 25, Chicago, IL).

Fifteen (15) participants were recruited into the study. The average age was 26.5 ± 5.7  years. Average (SD) height, weight, and body mass index were 1.68 ± 0.09 m, 63.7 ± 11.1 kg, and 22.3 ± 3.1 kg/m², respectively. Their average (SD) diastolic blood pressure, systolic blood pressure, and heart rate were 67.4 ± 7.9 mm Hg, 101.0 ± 5.7 mm Hg, and 67.6 ± 4.0 beats/min, respectively. During standing, the mean load of the right foot of each participant was 55% of the weight.

Typical examples of SBF responses of a participant are shown in Figure 1. SBF of the sham control showed no difference between before and after 10 minutes of standing. SBF of post-35 Hz vibration demonstrated about a 2-fold increase versus pre-35 Hz vibration. 

After vibrations, the SBF ratio of the 35 Hz vibration (1.83 ± 0.27) was significantly higher compared to the sham vibration of all participants (0.97 ± 0.08) (P < .01) (Figure 2). SBF before the 35 Hz vibrations was 41.96 ± 14.02 perfusion units (range 6.68–208.9 perfusion units), and SBF after the 35Hz vibrations was 61.16 ± 14.74 perfusion units (range 7.76–155.37 perfusion units). SBF before the sham vibration was 37.32 ± 9.29 perfusion units (range 5.74–120.44 perfusion units), and SBF after sham vibration was 33.97 ± 8.11 perfusion units (range 6.95–108.44 perfusion units).  

Wavelet analysis of SBF data revealed a distinct difference in all 5 regulations (metabolic, neurogenic, myogenic, respiratory, and cardiac) in response to 35 Hz vibration in all participants (Figure 3). For the metabolic control, the postvibration average values for the 35 Hz protocol (2.14 ± 0.35 au) were significantly higher than the sham control (1.06 ± 0.16 au, P < .01) (Figure 3). For the neurogenic control, the 35 Hz protocol ratio (2.0 ± 0.33 au) was significantly greater than the sham control (1.08 ± 0.14 au) (P < .05). For the myogenic control, there was also a significant difference between the local vibration ratio (2.01 ± 0.31 au) and the sham control ratio (1.05 ± 0.16 au) (P < .05) (Figure 3).

The findings of this study suggested that after 10 minutes of standing (sham vibration), the plantar SBF perfusion did not change, whereas standing combined with 35 Hz vibration for 10 minutes significantly increased plantar SBF. This study demonstrated that local vibrations improved plantar SBF during weight-bearing standing. Wavelet analysis showed that plantar SBF in response to vibrations during standing was attributed to the metabolic control (0.0095–0.02 Hz), neurogenic control (0.02–0.05 Hz), myogenic control (0.05–0.15 Hz), respiratory control (0.15–0.4 Hz), and cardiac control frequencies (0.4–2 Hz).³⁶ The results supported the authors’ hypotheses that 35 Hz vibration during standing could improve plantar skin microcirculation during standing. 

The SBF increased after the 35 Hz vibration during standing while the perfusion value of SBF after the sham vibration did not change, which indicated that the local vibrations could act as a protective method to reduce negative effects such as skin ischemia caused by weight-bearing stress. Previous studies demonstrated that local vibrations could improve peripheral circulation,^24,25 but there has been no study exploring the plantar SBF response on mechanical compression such as in standing. The present study demonstrated that local vibrations could also increase plantar SBF under prolonged compression, which might serve as a protective way to reduce the risk of DFUs. 

Some related animal studies applied vibration and showed that vibration could protect tissues from mechanical damage. Wong et al¹⁹ suggested that the intermittent vibration (35 Hz, 0.25 g acceleration for 10-minute intervention and 30-minute rest) during prolonged compression could reduce the muscle breakdown in aged mice, as indicated by the decreased number of interstitial nuclei; the study also indicated that vibration during compression could curb compression-induced oxidative damage. Di Carlo et al¹⁷ indicated that mechanical vibration (60 Hz, 1 g, 20 minutes) under hypoxia stress and other external stimulation could induce a stress response, such as increased dopamine, neurotensin, and substance P, and suggested that the mechanical vibration could serve as a preconditioning and protective method against oxidative stress. The current study confirmed these protective effects of vibration during stress in a human study by showing that the local vibrations at 35 Hz induced protection against weight-bearing stress, as indicated by the increased SBF after local vibrations during standing. Ennis et al²¹ suggested that mechanotransduction plays a significant role in tissue and wound healing; that is, energy-based modalities could create a local pulsatile phenomenon to stimulate normal blood flow and reestablish the microenvironment. Vibration is an energy-based modality, and under weight-bearing stress, through the use of energy transfer, it might induce off-and-on pressure to counter the ischemic effects of standing.

The results of wavelet analysis suggested that an increase in plantar SBF after local vibrations during stress was correlated with metabolic, neurogenic, and myogenic regulations. The metabolic control might be attributed to the increased release of nitric oxide, and many in vivo studies have demonstrated that vibration could induce laminar shear forces and produce more nitric oxide, and then cause vasodilation.^31,37 The neurogenic control was also one of the SBF control mechanisms in this study. Tzen et al³⁸ applied 2 rounds of 10-minute vibration (30 Hz, peak acceleration 0.4 g) to the right foot of research participants and used short-time Fourier transform to analyze the control mechanisms of vibration-induced SBF response. These authors also indicated that the neurogenic control mechanism played a role in the regulation of SBF and suggested that vibration could activate more ganglionic nerve and autonomic neurons.³⁸ Because the plantar foot is glabrous, there are many mechanoreceptors that could receive vibration stimulation, and it was hypothesized that high-frequency vibration stimulation (above 30 Hz) could activate predominantly fast-adapting cutaneous afferents and provide feedback for efferent autonomic regulation.³⁹ The myogenic regulation also controlled the SBF response on local vibrations during standing. The vascular smooth muscles could maintain intravascular pressure and induce vasoconstriction, which may be a local mechanism to counterbalance stress and was associated with changes in blood vessel transmural pressure. It has been hypothesized that vibration is beneficial to maintain vasomotor tone when skin is under stress.^33,38

The findings of this study indicate that local vibrations on the plantar foot could theoretically serve as a promising protective method to reduce the ischemic effects caused by weight-bearing stress on plantar skin in persons with diabetes and other patients at risk for foot ulcers. The stress caused by longtime, static standing is prone to produce plantar skin ischemia in persons with diabetes, which might be correlated to the characteristics of diabetic plantar skin. Observational studies have shown that because the plantar soft tissue of patients with diabetes is stiffer and less elastic, the cushion property and protective sensation were also damaged; in addition, the peak plantar pressure was higher in patients with diabetes than in healthy people.^8,12 

The application of vibration during standing might serve as a protective intervention to reduce plantar ischemia associated with weight-bearing activities. Until now there have been many potential protective interventions, such as cooling, ultrasound, and electrical stimulation, to protect soft tissue against mechanical stresses, but these potential inteventions have not been tested. However, local vibration may be more suitable, convenient, and safe for application to the plantar skin of patients with diabetes during standing due to the features of diabetic skin.^8,12 In addition, plantar vibration stimulations have been successfully applied in many preclinical studies of a small sample size of research participants. Madhavan et al⁴⁰ found that plantar vibrations (30–60 Hz frequencies, 0.2 g acceleration) could enhance type IIA muscle fiber activity in the legs. Kang et al⁴¹ applied mechanical stimulations with a pressure of 3.76 N/cm², 35-second intervention, to the arch of the plantar foot in persons with diabetic peripheral neuropathy; after a 4-week intervention of daily applications, the neuropathic symptoms, gait, and balance ability were improved. However, further study is required to explore whether local vibration could serve as a protective intervention during weight-bearing activity, such as standing, in persons with diabetes.

There are limitations to the current study. First, the number of males (n = 4) was not equal to the number of females (n = 11). Previous studies have indicated that sex may affect SBF regulations. However, the authors used the repeated measures design in this study, which could minimize the influence of sex on SBF responses. Second, SBF during standing with local vibrations (35 Hz vibration and sham vibration) was not measured. This was due to a limitation of LDF that prevents a simultaneous measurement of SBF under vibrations at the same site. Third, it is unclear whether an increase in SBF after standing and local vibration reduces plantar ischemia during standing. Finally, more research is required to determine whether this intervention affects the risk of foot ulcer development.

The authors conducted a controlled study to examine the effects of local vibrations on plantar SBF responses during weight-bearing standing among 15 healthy participants. Local vibrations (35 Hz frequency, 1 mm amplitude) to the weight-bearing plantar tissues during standing resulted in a significant increase between previbration and postvibration values in plantar SBF compared with the sham vibration. The control mechanisms contributing to this increase in SBF were metabolic, neurogenic, and myogenic regulations, especially metabolic endothelial control. The findings of this study confirm the findings of animal studies and support the application of local vibration to weight-bearing plantar tissues to reduce ischemia associated with standing and other weight-bearing activities.

Ting Zhu and Yana Wang were visiting students in the International Graduate Mentors Program at the University of Illinois at Urbana-Champaign when this study was conducted. 

Ms. Zhu is a graduate student in the Department of Kinesiology and Community Health, University of Illinois at Urbana-Champaign, Champaign, IL. Ms. Y. Wang is a graduate student in the Department of Kinesiology and Community Health, University of Illinois at Urbana-Champaign, Champaign, IL, and Shanghai First Rehabilitation Hospital, Shanghai, China. Dr. X. Wang is a professor in the Department of Kinesiology and Community Health, University of Illinois at Urbana-Champaign, Champaign, IL. Dr. Liao is a professor in the Department of Biomedical Engineering, Xi’an Technological University, Xi’an, China. Dr. Liu is a professor in the School of Kinesiology, Shanghai University of Sport, Shanghai, China. Dr. Jan is an associate professor, Department of Kinesiology and Community Health, University of Illinois at Urbana-Champaign, Champaign, IL. Send all correspondence to: Dr. Yih-Kuen Jan, Rehabilitation Engineering Lab, University of Illinois at Urbana-Champaign, 1206 South Fourth Street, 211N Huff Hall MC-588, Champaign, IL 61820; tel: +1-217-300-7253; fax: +1-217-333-2766; email: yjan@illinois.edu
TZ and YW contributed equally to this work. 

None disclosed.