Acute cardiovascular stress induced by shoulder vibratory exercise of different amplitudes

Abstract

BACKGROUND:

Vibration exercise has been investigated to enhance muscle activation, however, the effect of different amplitude vibratory exercises on cardiovascular stress is less understood.

OBJECTIVE:

Our study aims to explore the acute effect of shoulder vibratory exercises with different postures and amplitudes on the cardiovascular response in healthy adults.

METHODS:

Using a repeated measures randomized design, 36 subjects performed three different sessions with FLEXI-BAR exercise (FBE): (1) zero-amplitude, (2) small-amplitude, (3) large-amplitude. Each session included three different shoulder positions: 45-, 90- and 180-degree flexion. Heart rate variability (HRV), heart rate (HR) and rating of perceived exertion (RPE) were monitored continuously, while systolic blood pressure (SBP), diastolic blood pressure (DBP) and rate-pressure product (RPP) were measured before and after each exercise session.

RESULTS:

Compared with zero-amplitude, both small- and large-amplitude FBE protocols induced higher SBP. By contrast, DBP decreased with small- and large-amplitude. The RPP immediately after the exercise session were higher than at baseline. For high frequency, low frequency of HRV and HR there was a main effect of amplitude.

CONCLUSION:

Small- and large-amplitude FBE increased significantly SBP, RPE, HRV, HR and induced lower DBP, but the changes were modest, suggesting that FBE impose no extra threats to cardiovascular stress.

Keywords

FLEXI-BAR vibration exercise heart rate blood pressure

1. Introduction

Vibration exercise is currently enjoying popularity as an alternative exercise modality for enhancing muscle strength, power and flexibility, with its mechanical vibration with low frequency (below 50 Hz) and low intensity transmitting the power to the whole body [1, 2, 3]. It has been suggested as an attractive and efficient complement to traditional forms of exercise for athletes, as well as elderly persons and patient populations [4]. Vibration exercise for the shoulder can for example be performed with a pole known as FLEXI-BAR or the Body-Blade [5].

The FLEXI-BAR ${}^{\@setsize{\scriptsize}{8pt}{\viipt}{\@viipt}\textregistered}$ Standard is a training instrument that provides vibration stimulation with the purpose of increasing muscle activity [6] and enchancing muscle strength [7]. The underlying mechanism for enhancing neuromuscular performance may be tonic vibration reflex, motor neuron excitability and muscle hyperatrophy [7]. Vibrations are created when a shaking action is maintained, triggering a passive reflex response from the deep muscles, allowing agonist and antagonist muscles to work together in response to these vibrations, demanding stability of the trunk and the proximal region of the arms. During vibration training, the position and posture of the participants, the amplitude and orientation of the FLEXI-BAR coordinately determine which muscle groups are being activated as well as the level of activation [6]. Previous studies have verified its benefit including improvement of balance and activation of muscles [5], especially for abdominal and shoulder muscles [8]. Such activation of large muscle groups increase the oxygen demand, thus, potential change of heart rate (HR) and blood pressure (BP) may occur [9]. However, during the intensive exercise, a high HR has the potential to decrease left ventricular filling time possibly resulting in a reduced cardiac output [10]. In addition, vigorous activity can even acutely and transiently increase the risk of sudden cardiac death and acute myocardial infarction in susceptible persons [11]. Based on previous research no universal recommendations can be provided for optimizing vibration exercise protocols in terms of efficiency and cardiovascular safety [6, 12]. Up to date, no studies have measured the cardiovascular response to different types of FLEXI-BAR exercises (FBE) workouts.

Therefore, examining the effects of FBE on acute cardiovascular responses is important for two reasons. First, for safety recommendations it is necessary to determine the level of cardiovascular stress experienced when engaging in FBE. Second, before subjecting heart patients to FBE in different intensities, determining the response in healthy individuals and dose-response relationship is ethically important. The objective of this study was to determine the acute effect of different FBE protocols on the BP, rate-pressure product (RPP), HR, heart rate variability (HRV), and rating of perceived exertion (RPE) among young healthy adults. It was hypothesized that the FBE intensity, postures performed, and their interactions would significantly influence the acute cardiovascular response. We hypothesized that larger intensity and larger degree of shoulder flexion would induce higher cardiovasular stress.

2. Methods

2.1 Experimental approach to the problem

A repeated measures randomized design was performed to determine the cardiovascular response to different FBE conditions.

Each participant underwent three different FBE intensities: (a) zero-amplitude (0 cm), (b) small-amplitude (5 cm) and (c) large-amplitude (20 cm) [3, 7, 13]. All participants performed one sessions only with one intensity FBE and they finished three different sessions in different day, with a minimum of one rest day in between. During each session, the participants were instructed to perform FBE in three different postures with the shoulder flexed in the following positions: (1) 180 ${}^{\circ}$ ; (2) 90 ${}^{\circ}$ ; (3) 45 ${}^{\circ}$ (Fig. 1). The trunk was kept in neutral position while sitting in a chair with no back support, keeping both elbows 30 ${}^{\circ}$ flexed, hips horizontally 45 ${}^{\circ}$ abducted and foot resting on the floor. A similar posture has been used in previous FBE trials [8]. To avoid an order effect, the sequence of the nine FBE was randomized using Excel to simulate the throwing of a dice by the logical formula $=$ INT (6*RAND () $+$ 1) eventually resulting in a randomized exercise sequence.

Figure 1.

Shoulder vibratory exercise in three different postures: a) shoulder flexed 180 ${}^{\circ}$ ; b) shoulder flexed 90 ${}^{\circ}$ ; c) shoulder flexed 45 ${}^{\circ}$ .

2.2 Subjects

The sample size calculation was based on a small effect size [represented by f score in analysis of variance (ANOVA)]. Based on the G*Power 3.1.9.2 ANOVA: Repeated measures, within factors was set, with 3 amplitudes and 3 postures, assuming the effect size: f $=$ 0.25, with an alpha of 0.05, power ( $1-\beta$ ) $=$ 0.8, and nonsphericity correction $\in$ of 1 a minimum of 30 participants would be required. Considering a possible loss of up to 20%, 36 participants were included.

Participants were recruited at the Sun Yat-sen University from January to February 2018. The inclusion criteria were right-handed, young healthy adults (age from 20 to 39 years) who voluntarily joined this study. The exclusion criteria were upper extremity deficits, metabolic syndrome and cardiac dysfunction within the last year, and inability to perform FBE continuously for one minute. All participants signed an informed-consent document approved by the Office for Research Ethics of the Sixth Affiliated Hospital of Sun Yat-Sen University (SYSU). On a separate day before the experiment, participants were trained and learned to shake the FLEXI-BAR for at least one minute at each amplitude and specified position, and were taught to be familiarized with the experiment procedure and method. Before session, the participants were prohibitied to drink coffee or tea, smoke and so on.

2.3 Procedures

All experiments were conducted in an assessment room. The FLEXI-BAR ${}^{\@setsize{\scriptsize}{8pt}{\viipt}{\@viipt}\textregistered}$ Standard (Togu, Germany) was used for all experiments. Shaking the FLEXI-BAR could cause the rubber handle to vibrate with 5-Hz [12]. The amplitude, which represent the FBE intensity [14], was measured as the distance of the caudal movement. A custom-made stable rack as a visual feedback with adjustable 10 cm or 40 cm length bars respectively, were used to guide the amplitude to ensure consistent performance of FBE. This method had been relatively standardized and its reliability has been tested before commencement of the experiments. The test leader gave constantly verbal feedback to participants if the amplitude came out of tune. Each session started with a practice and warm-up trial.

Before starting, individuals were instructed sit and rest for 5 minutes. The operator then measured baseline BP, HR and RPE. The individual rested around 2 minutes after these measurements meanwhile HR and HRV were recorded (Fig. 2).

Figure 2.

The scheme of experimental session.

In the pilot study, we found that most participants felt extremely fatigued when performing large-amplitude FBE for more than one minute. Although the Task Force statement suggests a minimal of 2-minute recording to quantify the LF components of HRV while 1-minute to address the HF component, for recording and analysis of HRV at least one minute is required [15]. Thus, to balance the collection of data and the individual’s ability to maintain the required FBE, the duration of 1 minute was chosen in this study. After each exercise, subjects were asked to remain seated and rest until the HR resumed to baseline values before initiation of the next exercise.

The outcomes were diastolic and systolic blood pressure (DBP and SBP, in mmHg), heart rate (b $\cdot$ min ${}^{-1}$ ), rate-pressure product, heart rate variability, and the Borg scale for the rating of perceived exertion. Participants wore a HR and HRV monitor (Polar ${}^{\@setsize{\scriptsize}{8pt}{\viipt}{\@viipt}\textregistered}$ RS800CX, Finland) [16] throughout all FBE sessions. The HRV results was based on during each the 1-minute exercise. The last 10-second averaged HR of each 1-minute exercise period [17] recorded by Polar ${}^{\@setsize{\scriptsize}{8pt}{\viipt}{\@viipt}\textregistered}$ RS800CX was defined as the final HR, and these values were applied to analyzed the cardiovascular responses during exercise. The values of the BP (in mmHg) and HR (in b $\cdot$ min ${}^{-1}$ , pre- and post-session values) were recorded by a electronic sphygmomanometer (Tensoval comfort, Germany) [18] before the first exercise and immediately after the third exercise, and these values were used to calculated the RPP as: (HR $\times$ SBP)/100. An additional device (WristOx2TM, Model 3150, USA) was applied to monitor the HR until the HR resume to the baseline level. The Borg scale for RPE was used. The RPE analysis was based on the highest value documented in each FBE that participants gave at the 20 ${}^{\text{th}}$ , 40 ${}^{\text{th}}$ and 60 ${}^{\text{th}}$ second during the exercise.

Before starting the following session, the participants were subjectively evaluated to determine whether they had recovered from the previous FBE session. In case of ongoing fatigue or soreness, the second and third sessions were rescheduled.

2.4 Statistical analysis

IBM SPSS Statistics software was used to statistically analyze the original data (Version 20.0, IBM, Armonk, NY, USA). The basic information was described as the mean and standard deviation. Paired $t$ -test was applied to compare the duration of the washout period which was measured in days between session 3 and session 2 with that between session 2 and session 1. Data were subjected to an ANOVA with two repeated measures: Intensity of FBE (zero-, small-, large-amplitude); and Time (before and after session FBE). This method estimated the difference in mean SBP, DBP, and RPP between each pre- and post-FBE sessions and corresponding different intensities. The intensity $\times$ time interaction effect determined whether changes occurring between each pre- and post-session were coherent among the three FBE intensities.

Another repeated measures ANOVA [(within-subject factors: 1. intensity [zero-, small-, large-amplitude FBE); and 2. Exercise posture (180 ${}^{\circ}$ , 90 ${}^{\circ}$ , 45 ${}^{\circ}$ shoulder flexion)] were used to assess the mean heart rate, high frequency (HF) and low frequency (LF) of HRV between exposure to three different FBE amplitudes and among the three exercises posture. The mean HF and LF of HRV during every 1-minute exercise were extracted directly from ProTrainer 5 software designed for Polar RS800CX and analysized through the frequency-domain in Kubios HRV (ver. 3.0.2) software. In order to correct artifacts from a corrupted RR interval series, the threshold (medium) in Artifact correction, the 60-seconds window length in Samples for analysis, and smoothness priors method and 1000 Lambda value in Remove trend components were chosen. We investigated the studentized residual and use the Shapiro-Wilk and Kolmogorov-Smirnov test combined with skewness and kurtosis to check the normality for related variables. The posture $\times$ intensity interaction effect decided whether the variations in HR, LF and HF provoked by FBE were posture-dependent or intensity-dependent. If the sphericity assumption was satisfied, we choose Greenhouse-Geisser epsilon adjustment to move on. Contrast analysis with Bonferroni’s adjustment was realized if we found significant results.

Other analyses were used to determine whether the cardiovascular changes were associated with the basic values. The after-session values subtracted the basic values forming the within-session change score, which represent BP and RPP data. As to HR and HRV data, the change score (before and after each session) was calculated by subtracting the baseline values from the after-exercise value. We performed Pearson’s product moment correlations to define the level of association between the change value and the basic score for all variables, respectively.

We set $P\leqslant$ 0.05 as the level of significance, and alpha during post-hoc analysis was regulated based on the number of comparisons made. Randomization was used to minimize order effects, but we did not formally investigate order effects associated with either to the amplitude or to the posture.

3. Results

3.1 Characteristics of participants

A total of 36 participants were enrolled in this study (16 men and 20 women; mean age: 22.54 $\pm$ 2.53 years; mean BMI: 20.79 $\pm$ 2.68 kg/m ${}^{2}$ ), 35 of them completed all experiments except that one subject dropped out at his final session (Fig. 3). One subject withdrew from the study because of unstable blood pressure and was diagnosed with hypertension after being referred to the doctor. None of participants withdrew from the sessions due to fatigue or any other discomfort. None of the participants reported adverse effects or requested to discontinue the exercises during any of FBE.

Table 1
Baseline SBP, DBP, HR and HRV data of study participants

Amplitude	$n$	SBP/DBP	HR	HRV LF	HRV HF
Zero	35	113.26 (12.54)/73.63 (6.87)	77.60 (9.56)	58.67 (14.28)	31.66 (14.38)
Small	33	112.85 (11.50)/73.21 (6.02)	78.88 (11.91)	54.46 (15.29)	34.89 (16.12)
Large	35	114.31 (12.20)/73.80 (7.58)	77.80 (12.38)	57.45 (12.62)	32.48 (14.24)

${}^{*}$ Mean (SD) presented for continues variables. $n$ : number count; SD: standard deviation; HR $=$ heart rate; HRV $=$ Heart rate variability; LF of HRV $=$ Low frequency for heart rate variability; HF of HRV $=$ High frequency for heart rate variability.

Figure 3.

The flow chart of participants and path of testing.

Figure 4.

a) Changes of the systolic blood pressure after vibration with different amplitudes; b) Changes of the diastolic blood pressure after vibration with different amplitudes; c) Changes of the rate-pressure product after vibration with different amplitudes.

Figure 5.

a) Effect of FBE amplitude on low frequency for heart rate variability; b) Effect of FBE amplitude on high frequency for heart rate variability.

Figure 6.

a) Effect of FBE amplitude on heart rate; b) Effect of FBE shoulder posture on heart rate.

3.2 Washout period

The correlation of washout period among all three sessions was interesting because the phase of session 1 and session 2 (mean 1.21 days, SD $=$ 0.42 Days, range $=$ 1–3 Days) was almost similar with the duration between session 2 and session 3 (mean $=$ 1.27 days, SD $=$ 0.81 days, range $=$ 1–4 Days) ( $P=$ 0.56). No any carryover effects were reported by participants whose complaints may have developed from the preceding session that required suspension and discontinuity.

3.3 Cardiovascular change

3.3.1 Systolic blood pressure, diastolic blood pressure, rate-pressure product changes

Figure 7.

Effect of FBE on rating of perceived exertion using the Borg scale.

The amplitude $\times$ time interaction effect was statistically significant for SBP (F2, 66 $=$ 3.69, $P\leqslant$ 0.001) and DBP (F2, 66 $=$ 10.51, $P\leqslant$ 0.05). Overall main effect of amplitude and time was also found significant (Table 1). Regarding the main effect of amplitude, contrast analysis revealed that both the small-amplitude and large-amplitude FBE protocols induced significantly higher SBP than zero-amplitude, by an average of 4.85 mmHg (95% CI: 2.40, 7.30; $P\leqslant$ 0.001) and 7.82 mmHg (95% CI: 5.68, 9.98; $P\leqslant$ 0.001) respectively (Fig. 4A). However, DBP was decreased significantly by an addition of small-amplitude with 2.82 mmHg (95% CI: 0.82, 4.83; $P=$ 0.007) and large-amplitude with 6.71 mmHg (95% CI: 4.81, 8.60; $P\leqslant$ 0.001) (Fig. 4B).

The RPP (F2, 666 $=$ 72.96, $P\leqslant$ 0.001) immediately after the exercise session were significantly higher than the values at baseline (Fig. 4C). The amplitude $\times$ time interaction effect (F2, 66 $=$ 14.26, $P\leqslant$ 0.001) was also significant. Differences in RPP was not significant among main effect (F2, 66 $=$ 1.73, $P=$ 0.185) of amplitude (Table 1).

Table 2

The effect of FLEXI-BAR exercises on outcome measurements

Variable	$F$	$P$ -value	$F$	$P$ -value	$F$	$P$ -value	Mean difference (95% CI)	$P$ -value	Mean difference (95% CI)	$P$ -value	Mean difference (95% CI)	$P$ -value
	Amplitude*posture interaction effect		Main effect of amplitude		Main effect of posture		Post-hoc contrast analysis (main effect of amplitude)
							Zero- Vs small-amplitude		Zero- Vs large-amplitude		Small- Vs large-amplitude
HR	1.47	0.232	93.71	$\leqslant$ 0.001	8.78	$\leqslant$ 0.001	$-$ 14.16 ( $-$ 19.58, $-$ 8.74)	$\leqslant$ 0.001	$-$ 30.75 ( $-$ 37.22, $-$ 24.26)	$\leqslant$ 0.001	$-$ 16.59 ( $-$ 21.64, $-$ 11.55)	$\leqslant$ 0.001
HRV-HF	0.11	0.978	31.47	$\leqslant$ 0.001	2.96	0.059	12.79 (5.60, 19.98)	$\leqslant$ 0.001	22.46 (15.47, 29.45)	$\leqslant$ 0.001	9.67 (2.36, 16.97)	0.006
HRV-LF	0.11	0.979	31.30	$\leqslant$ 0.001	3.02	0.056	$-$ 12.82 ( $-$ 20.07, $-$ 5.58)	$\leqslant$ 0.001	$-$ 22.57 ( $-$ 29.64, $-$ 15.51)	$\leqslant$ 0.001	$-$ 9.75 ( $-$ 17.09, $-$ 2.41)	0.006

Table 3

The results of amplitude*posture interaction effect

3.3.2 High frequency for heart rate variability

A significant main effect of amplitude (F2, 66 $=$ 31.47, $P\leqslant$ 0.001) was found (Fig. 5A), while posture showed no significant main effect (F2, 66 $=$ 2.96, $P=$ 0.059) (Table 3). The amplitude $\times$ posture interaction effect, however, was not significant (F4, 132 $=$ 0.11, $P=$ 0.978). Contrast analysis revealed that overall, both the small-amplitude and the large-amplitude FBE protocols produced significantly lower HF of HRV than the zero-amplitude, by an average of 12.79 (95% CI: 5.60, 19.98; $P\leqslant$ 0.001) and 22.46 (95% CI: 15.47, 29.45; $P\leqslant$ 0.001) respectively. The large-amplitude induced significantly 9.67 (95% CI: 2.36, 16.97; $P=$ 0.004) lower HF than the small-amplitude (Table 3).

3.4 Low frequency for heart rate variability

The main effect of amplitude (F2, 66 $=$ 31.30, $P\leqslant$ 0.001) was significant (Fig. 5B) and the posture was found to be non-significant (F2, 66 $=$ 3.02, $P=$ 0.056) (Table 3). The amplitude $\times$ posture interaction effect, however, was not significant (F4, 132 $=$ 0.11, $P=$ 0.979). Contrast analysis revealed that overall, both the small-amplitude and the large-amplitude FBE protocols induced significantly higher LF of HRV than the zero-amplitude, by an average of 12.82 (95% CI: 5.58, 20.07; $P\leqslant$ 0.001) and 22.57 (95% CI: 15.51, 29.64; $P\leqslant$ 0.001) respectively. The large-amplitude induced significantly 9.75 (95% CI: 2.41, 17.09; $P=$ 0.006) higher LF than the small-amplitude (Table 3).

3.4.1 Heart rate response

There was a significant main effect of amplitude (F2, 62 $=$ 93.705, $P\leqslant$ 0.001) and position (F2, 62 $=$ 8.780, $P\leqslant$ 0.001). The amplitude $\times$ posture interaction effect was not significant (F2.747, 85.142 $=$ 1.467, $P=$ 0.232) (Table 3). Contrast analysis revealed that small-amplitude (mean difference: 14.16 b $\cdot$ min ${}^{-1}$ ; 95% CI: 8.74, 19.58; $P\leqslant$ 0.001) and large-amplitude (mean difference: 30.75 b $\cdot$ min ${}^{-1}$ ; 95% CI: 24.26, 37.22; $P\leqslant$ 0.001) significantly increased HR than zero-amplitude. The difference of HR was also significant between the small- and large-amplitude (mean difference: 16.59 b $\cdot$ min ${}^{-1}$ ; 95% CI: 11.55, 21.64; $P\leqslant$ 0.001) (Fig. 6A). Regarding the main effect of posture, shoulder flexion to 180 ${}^{\circ}$ significantly (mean difference: 3.792 b $\cdot$ min ${}^{-1}$ ; 95% CI: 1.458, 6.125; $P=$ 0.001) and flexion to 90 ${}^{\circ}$ marginally significantly (mean difference: 2.083 b $\cdot$ min ${}^{-1}$ ; 95% CI: 0.007, 4.160; $P=$ 0.049) induced higher HR than flexion to 45 ${}^{\circ}$ (Fig. 6B).

3.4.2 Rating of perceived exertion changes

The pooled RPE data are shown in detail (Fig. 7). Out of the 315 FBE trials for each amplitude protocol (9 exercises $\times$ 35 participants), 2% (4 participants, 6 trials), 8% (10 participants, 26 trials), and 26% (24 participants, 83 trials) reported an RPE $>$ 15 for zero-amplitude, small-and large-amplitude FBE sessions, respectively. The large-amplitude FBE significant induced higher perceived effort than small-amplitude FBE protocols ( $P<$ 0.001) and zero-amplitude condition ( $P<$ 0.001) during 3 different exercise postures. Among the large-and small-amplitude FBE, pairwise comparison revealed that the RPE value increased significantly with shoulder position flex to 90 ${}^{\circ}$ and 180 ${}^{\circ}$ than 45 ${}^{\circ}$ .

3.4.3 Association with the baseline values

The Pearson’ s product moment correlations results showed that no baseline RPP, DBP, and SBP in different amplitude was significant correlated with the respective change scores except for RPP ( $r=$ 0.362, $P=$ 0.036) of the small-amplitude exercise protocol.

Out of 9 different FBE posture and amplitude combinations (3 amplitudes $\times$ 3 postures), the baseline LF and HF of HRV were significant negatively correlated with the change scores during the exercise except for 45 ${}^{\circ}$ ( $r=-$ 0.258, $P=$ 0.135) and 180 ${}^{\circ}$ ( $r=-$ 0.304, $P=$ 0.076) in LF, and 45 ${}^{\circ}$ ( $r=-$ 0.257, $P=$ 0.136) and 180 ${}^{\circ}$ ( $r=-$ 0.305, $P=$ 0.075) in HF of zero-amplitude exercise protocol. No significant correlations were identified with all the HR data.

4. Discussion

FBE has been gaining popularity worldwide for many years, both for rehabilitation purposes as well as recreational training environments [19]. Investigating the acute cardiovascular response to FBE can provide critical evidence for physiotherapist to form the exercise intensity and ensure safety. This is the first study to examine the cardiovascular response to different FBE amplitude and posture in healthy adults. As hypothesized, the findings of the present study were that addition of large- and small-amplitude FBE increased significantly SBP, RPE, HRV, HR and induced lower DBP, but the increase and decrease was modest. Thus, it seems safe to test this protocol in patients, where after clinical recommendations can be provided.

Our study revealed that there was only mild increase in SBP ( $<$ 8 mmHg), and interestingly fair decrease in DBP ( $>$ 0 mmHg). The latter finding may be explained by a vasodilatory response to exercise. Besides, the SBP and DBP upper bound of the 95% CI were much lower than the change of BP induced by four sets of 4 repetition maximum (RM) and 20 RM leg-extension in healthy people (28 mmHg and 77 mmHg mean increase for SBP, while 26 mmHg and 53 mmHg mean increase for DBP, respectively) [20]. Consistent with findings by previous study [21, 22], we found that exhaustive vibration exercise elicits mild BP increase in healthy young adults.

Myocardial workload can be estimated by RPP which provide an indication of the amount of myocardial oxygen requirement [23]. The data suggests the RPP increase significantly after FBE, but there is not main effect of amplitude indicating that oxygen demand in different FBE amplitude should be similar. The histogram in chart indicates that the mean RPP at the end of FBE are 85, 89 and 93 for zero-, small- and large-amplitude respectively (Fig. 3C). This result is evenly much lower than the value induced by another kind of vibration exercise for frailty patients (mean value $>$ 104 for all session) [24]. The upper bound of the 95% CI for increased valuable of large-amplitude FBE did not even exceed 14.04, which means that the changes of RPP are acceptable for healthy young adults.

One subject withdrew from the study because of hypertension before the trial began. It turned out that the subject had unstable blood pressure and was diagnosed with hypertension after being referred to the doctor. The remaining participants reported no adverse symptoms and signs, and none of them requested to end even one trial, indicating that the FBE protocols performed in our study were safe and fairly tolerated in healthy adults. We also found that the baseline values of SBP, DBP and RPP was not correlated with change of same value after different amplitude FBE. Therefore, FBE did not induce proportionally higher BP and RPP stress to the participants with high baseline scores.

Surprisingly, data demonstrates that 2% (4 participants, 6 trials), 8% (10 participants, 26 trials), and 26% (24 participants, 83 trials) participants reported an RPE $>$ 15 for zero-, small-and large-amplitude FBE separately (Fig. 6). This result shows the addition of either small- or large-amplitude FBE induce higher effort perception. But the participants’ physiological stress may differ and not necessarily subjectively match with RPE in internal-training paradox [25].

HRV has been applied to evaluating the exercise load [26]. According to the study results, the LF increase and HF decrease suggesting the enhancement of sympathetic activity and the inhibition of parasympathetic activity during the FBE. This finding also corresponds to one similar study which found that HF power significant decreased and LF power increased in immediately after passive vibration [27]. Interestingly, the LF (or HF) immediately after FBE showed significantly higher (or Lower) in the group performing the higher amplitude indicating that a more intense FBE amplitude distinctively enhanced sympathetic activity. This phenomenon may result from the more muscles activation that occurs with higher FBE amplitude.

Compared with no HR changes exerted by the passive whole-body vibration [28], the active FBE impose HR increased. We found that large-amplitude and small-amplitude significantly increased higher HR than zero-amplitude FBE, meanwhile large-amplitude had higher HR than small-amplitude FBE during exercise. Furthermore, the shoulder flexion to 180 ${}^{\circ}$ induces higher HR than Flexion to 45 ${}^{\circ}$ , but all the increase are modest, and the increase exercise intensity are possibly resulted from contraction of trunk and upper extremities muscle maneuvered during FBE [6, 12].

Clinically, FBE may hold potential benefits especially for some patients with dysfunction of lower extremities as a kind of aerobic exercise training. In our study, the exercise intensities of large-amplitude FBE were mild compared with resting levels, and only occupied about 34% HR reserve. However, the intensities with addition of large-amplitude FBE were much lower than those moderate aerobic exercise training usually around 60% HR reserve [24, 29]. Besides, over 68% participants of the large-amplitude FBE have much higher RPE ( $>$ 15 points) than the usually used value from 12 to 15 points. Moreover, the lasting period of present protocol is only limited to one minute comparing aerobic exercise necessarily over 30 minutes [30]. As a result, our present data do not allow to make conclusion about FBE to be a kind of aerobic exercise in young healthy people.

Limitations as well as strengths exist in our study. Firstly, the HRV components were recorded continuously within 1 min, limited by the length of each training session. The Task Force statement suggests a minimal of 2-min recording to quantify the LF components of HRV while 1-min to address the HF component [31]. The duration of HRV recording may thus lead to measurements that are not completely exact. However, this was equal during all conditions, allowing at least for comparison between conditions. Furthermore, considering the individual responses, the small- and large-amplitude FBE might led to soreness and high RPE for those unaccustomed to regular exercise. This might partially explain the various outcome of RPE changes in different subjects. Further research is needed to examine the cardiovascular stress of long-lasting FLEXI-BAR training (e.g. 2-min or 5-min) in well-trained and homogeneous populations, as well as in clinical populations with cardiovascular disease. Secondly, standardization of the amplitudes does not seem robust enough in our study. It is difficult for participants to continuously and accurately maintain the amplitude without visual or auditory feedback. Thus, using a custom-made rack combined with test leader’s continuous observation and oral correction can offer the visual or auditory feedback to reduce the amplitude deviation, and this method had been standardized in the pilot study. It would be better to use a device like amplitude-monitoring sensor with good validity. Finally, the training protocol with different shoulder flexion angle was limited in sitting position, which may reduce the neuromuscular output thus minimize the exercise intensity on cardiovascular system. Since the training posture could alter the effect of FBE, to what extent different posture may modulate the cardiovascular response will require further investigation.

5. Conclusions

This study shows that addition of large-and small-amplitude FLEXI-BAR training increased significantly SBP, RPE, HRV, HR and induced lower DBP, but the increase and decrease was modest, suggesting that shoulder vibratory exercises impose no extra threats to cardiovascular stress for young healthy people and the present protocol of FBE are probably unsuitable to be an aerobic training exercise. Further study is needed to examine whether shoulder vibratory exercise protocol would be safe and provide a favorable effect for patients in musculoskeletal rehabilitation.

Footnotes

Acknowledgments

This work was financlly supported by the National Natural Science Foundation of China (81472155) and Guangzhou Science, Technology and Innovation Commission (201802010039).

Conflict of interest

None to report.

References

Pournot

Tindel

Testa

Mathevon

Lapole

. The acute effect of local vibration as a recovery modality from exercise-induced increased muscle stiffness. J Sports Sci Med. 2016; 15(1): 142-7.

Jepsen

Thomsen

Hansen

Jorgensen

Masud

Ryg

. Effect of whole-body vibration exercise in preventing falls and fractures: a systematic review and meta-analysis. BMJ Open. 2017; 7(12): e018342.

Cho

Lee

Kim

Hahn

Lee

. The effects of double oscillation exercise combined with elastic band exercise on scapular stabilizing muscle strength and thickness in healthy young individuals: a randomized controlled pilot trial. J Sports Sci Med. 2018; 17(1): 7-16.

Cochrane

. Vibration exercise: the potential benefits. Int J Sports Med. 2011; 32(2): 75-99.

Ruzene

JRS

Morcelli

Navega

. Equilibrium in women with osteoporosis submitted to balance training with and without an oscillating vibratory pole. J Bodyw Mov Ther. 2016; 20(1): 35-41.

Moreside

Vera-Garcia

McGill

. Trunk muscle activation patterns, lumbar compressive forces, and spine stability when using the bodyblade. Phys Ther. 2007; 87(2): 153-63.

Jin Luo McNamara

Moran

. The-use-of-vibration-training-to-enhance-muscle-strength-and-power.

Escamilla

Yamashiro

Dunning

Mikla

Grover

Kenniston

, et al. An electromyographic analysis of the shoulder complex musculature while performing exercises using the bodyblade(R) classic and bodyblade(R) pro. Int J Sports Phys Ther. 2016; 11(2): 175-89.

Gappmaier

Tavazoie

Jacketta

. Cardiorespiratory response to exercise on a large therapeutic roll. Cardiopulm Phys Ther J. 2013; 24(3): 5-13.

10.

Lavie

Arena

Swift

Johannsen

Sui

Lee

D-c

, et al. Exercise and the cardiovascular system. Circulation Research. 2015; 117(2): 207-19.

11.

Thompson

, Franklin Ba Fau Balady

, Balady Gj Fau Blair

, Blair Sn Fau Corrado

, Corrado D Fau Estes

NAM

, 3rd, Estes Na 3rd Fau Fulton

, et al. Exercise and acute cardiovascular events placing the risks into perspective: a scientific statement from the American Heart Association Council on Nutrition, Physical Activity, and Metabolism and the Council on Clinical Cardiology. (1524–4539 (Electronic)).

12.

Mileva

Kadr

Amin

Bowtell

. Acute effects of Flexi-bar vs. Sham-bar exercise on muscle electromyography activity and performance. J Strength Cond Res. 2010; 24(3): 737-48.

13.

Ozemek

Cochran

Strath

Byun

Kaminsky

. Estimating relative intensity using individualized accelerometer cutpoints: the importance of fitness level. BMC Med Res Methodol. 2013; 13: 53.

14.

Abdollahi

Nikkhoo

Ashouri

Asghari

Parnianpour

Khalaf

. A model for flexi-bar to evaluate intervertebral disc and muscle forces in exercises. Med Eng Phys. 2016; 38(10): 1076-82.

15.

Esco

Flatt

. Ultra-short-term heart rate variability indexes at rest and post-exercise in athletes: evaluating the agreement with accepted recommendations. J Sports Sci Med. 2014; 13(3): 535-41.

16.

Williams

Jarczok

Ellis

Hillecke

Thayer

Koenig

. Two-week test-retest reliability of the Polar((R)) RS800CX() to record heart rate variability. Clin Physiol Funct Imaging. 2017; 37(6): 776-81.

17.

Ozemek

Cochran

Strath

Byun

Kaminsky

. Estimating relative intensity using individualized accelerometer cutpoints: the importance of fitness level. BMC Med Res Methodol. 2013; 13: 53.

18.

Tholl

Luders

Bramlage

Dechend

Eckert

Mengden

, et al. The german hypertension league (deutsche hochdruckliga) quality seal protocol for blood pressure-measuring devices: 15-year experience and results from 105 devices for home blood pressure control. Blood Press Monit. 2016; 21(4): 197-205.

19.

Doroudian

Roostayi

Naimi

Rahimi

Baghban

. Effect of using the flexi-bar tool on erector spinae muscle activation under different standing weight-bearing conditions. J Back Musculoskelet Rehabil. 2019; 32(3): 505-9.

20.

Gjovaag

Hjelmeland

Oygard

Vikne

Mirtaheri

. Acute hemodynamic and cardiovascular responses following resistance exercise to voluntary exhaustion. Effects of different loadings and exercise durations. J Sports Med Phys Fitness. 2016; 56(5): 616-23.

21.

Rittweger

Beller

Felsenberg

. Acute physiological effects of exhaustive whole-body vibration exercise in man. Clin Physiol. 2000; 20(2): 134-42.

22.

Cochrane

Sartor

Winwood

Stannard

Narici

Rittweger

. A comparison of the physiologic effects of acute whole-body vibration exercise in young and older people. Arch Phys Med Rehabil. 2008; 89(5): 815-21.

23.

Whitman

Sabapathy

Jenkins

Adams

. Handrail support produces a higher rate pressure product in apparently healthy non-treadmill users during maximal exercise testing. Physiol Meas. 2019; 40(2): 02NT1.

24.

Liao

Jones

Pang

. Cardiovascular stress induced by whole-body vibration exercise in individuals with chronic stroke. Phys Ther. 2015; 95(7): 966-77.

25.

Meckel

Zach

Eliakim

Sindiani

. The interval-training paradox: physiological responses vs. subjective rate of perceived exertion. Physiol Behav. 2018; 196: 144-9.

26.

Buchheit

Gindre

. Cardiac parasympathetic regulation: respective associations with cardiorespiratory fitness and training load. Am J Physiol Heart Circ Physiol. 2006; 291(1): H451-8.

27.

Jiao

Chen

Wang

. Effect of different vibration frequencies on heart rate variability and driving fatigue in healthy drivers. Int Arch Occup Environ Health. 2004; 77(3): 205-12.

28.

Hazell

Thomas

Deguire

Lemon

. Vertical whole-body vibration does not increase cardiovascular stress to static semi-squat exercise. Eur J Appl Physiol. 2008; 104(5): 903-8.

29.

Ramirez-Velez

Tordecilla-Sanders

Tellez

Camelo-Prieto

Hernandez-Quinonez

Correa-Bautista

, et al. Effect of moderate versus high-intensity interval exercise training on heart rate variability parameters in inactive latin-american adults: a randomised clinical trial. J Strength Cond Res. 2017.

30.