Whole‐body vibration preconditioning reduces the formation and delays the manifestation of high‐altitude‐induced venous gas emboli

What is the central question of this study? Is performing a 30‐min whole‐body vibration (WBV) prior to a continuous 90‐min exposure at 24,000 ft sufficient to prevent venous gas emboli (VGE) formation? What is the main finding and its importance? WBV preconditioning significantly reduces the formation and delays the manifestation of high‐altitude‐induced VGE. This study suggests that WBV is an effective strategy in lowering decompression stress.


INTRODUCTION
Decompression sickness (DCS) is a risk associated with high-altitude aviation, space exploration, compressed gas diving and tunnelling. A rapid pressure reduction leads to gas supersaturation and bubble formation in various body tissues (MacMillan, 2006;Stepanek, 2002), including in venous blood (venous gas emboli; VGE). VGE are common and often asymptomatic (silent bubbles), with the severity of DCS being largely influenced by the size, number and the location of bubbles (Eatock, 1984;Gardette, 1979;Nishi et al., 2003). In an aviation setting, mild DCS symptoms (e.g., joint pain) could impair a pilots' performance and, resultantly, necessitate that a mission is altered or even be aborted, whereas, in some instances, severe DCS may cause neurological symptoms which can lead to an immediate incapacitation and thus, constitute a serious aviation safety hazard (MacMillan, 2006;Neubauer et al., 1988;Odland et al., 1959).
Reducing the level of nitrogen dissolved in body fluids/tissues by breathing 100% oxygen prior to decompression, a process referred to as denitrogenation, decreases the potential for supersaturation, VGE formation and the likelihood of developing DCS symptoms.
The diffusion gradient for nitrogen is reduced with progression of denitrogenation, hence, each hour of oxygenation is less effective than the one preceding it . However, while the benefits of pre-oxygenation are well documented across the literature (Waligora et al., 1987;Webb & Pilmanis, 1993, this preconditioning strategy is not currently employed in fighteraircraft operations, as it is time consuming (1-6 h) and therefore not operationally feasible. Accordingly, identifying a simple and timeefficient preconditioning strategy that will effectively lower and/or delay the development of high-altitude-induced VGE will be of particular interest to military services.
It is generally accepted that VGE emanate from pre-existing gas micronuclei attached to the vessel endothelium (Christman et al., 1986; New Findings • What is the central question of this study? Is performing a 30-min whole-body vibration (WBV) prior to a continuous 90-min exposure at 24,000 ft sufficient to prevent venous gas emboli (VGE) formation?
• What is the main finding and its importance?
WBV preconditioning significantly reduces the formation and delays the manifestation of high-altitude-induced VGE. This study suggests that WBV is an effective strategy in lowering decompression stress. Lee et al., 1993;Vann et al., 1980;Yount et al., 1979). Changes in buoyancy and flow-induced shear forces can destabilise these precursors and dislodge them from their nucleation site, subsequently being eliminated by the combined effects of surface tension and the oxygen window (Blatteau et al., 2006;Masurel, 1989). Since in the absence of gas micronuclei, homogeneous bubble nucleation requires very high levels of supersaturation (>∼100 ATA) (see review by Jones et al., 1999), reducing these precursors prior to decompression would, theoretically, lower VGE production during high-altitude flying. Indeed, performing an exercise bout 1-2 h prior to decompression has been evinced to effectively lower the prevalence of VGE during hypobaria (Webb et al., 1996). However, this preconditioning strategy is difficult to implement across personnel [i.e., exercising at specific intensities (70-75% of an individual's peak oxygen uptake)] in addition to necessitating access to specialised equipment. More recently, whole-body vibration (WBV) has been investigated as another possible approach for reducing VGE formation (Germonpré et al., 2009). Interestingly, a 30-min WBV (35)(36)(37)(38)(39)(40) session completed an hour before diving led to a significant reduction in post-dive VGE count compared with control (i.e., pre-dive supine rest) (Germonpré et al., 2009

Ethical approval
Ethics approval for this human study was granted by the regional Human Ethics Committee in Stockholm, Sweden (approval no: 2019-06395), and all experimental procedures conformed to the Declaration of Helsinki, except for registration in a database. All subjects provided written informed consent before the study.

Subjects
Eight, healthy, male subjects volunteered to participate in this study (mean ± standard deviation (SD); age, 46 ± 11 years; body mass, 85 ± 14 kg; height, 1.83 ± 0.08 m; body mass index, 25.27 ± 4.20 kg/m 2 ). Prior to the onset of the experimental sessions, potential subjects underwent a physical examination and an electrocardiogram by a physician, with only individuals who satisfied the inclusion criteria with a clean health record (i.e., no history of cardiorespiratory disorders nor any other health conditions such as epilepsy and diabetes) being included in the study. Additionally, subjects were briefed in detail about the purpose of the study, the experimental procedures and the potential risks and benefits, prior to giving their written consent. It was also pointed-out to them that their participation was voluntary, and they were entitled to withdraw from the study at any time without providing any reason.
The right to include or exclude their data was also made clear to them.

Familiarisation session
Approximately a week prior to the experimental procedures, subjects underwent a familiarisation session which introduced them to the trial conditions, requirements, testing environment and equipment.

Experimental protocol
All experimental procedures were conducted at the Division of Environmental Physiology of the Royal Institute of Technology. During each testing day, subjects were instructed to report to the laboratory following abstinence from caffeine-and alcohol-containing beverages.
In addition, subjects were asked to refrain from strenuous physical activity for 24 h prior to and during each testing day.

Preliminary measurements and preconditioning strategies
Following arrival at the laboratory, the subjects' body mass and height were evaluated (Vetek, Väddö, Sweden). Thereon, the sub-jects completed a 5-min seated rest period followed by measurement of their resting capillary oxyhaemoglobin saturation levels (SpO 2 ) (Radical-7, Masimo, Irvine, CA, USA), heart rate and arterial blood pressure (M3, Omron, Kyoto, Japan). At completion of the resting period, each subject performed, in a Latin-square fashion, and on separate days (i.e., separated by ≥48 h), one of the following protocols: (i) 40-min seated rest (control protocol), (ii) 30-min seated rest followed by 150 deep knee-squats (exercise protocol) being completed over a 10-min period (Dervay et al., 2002), and (iii) 30-min of WBV (40 Hz, 2 mm amplitude), using a commercially available vibration platform (inSPORTline, VibroGym, Badhoevedorp, Netherlands), followed by 10-min of seated rest. During the WBV protocol, subjects stood flat-footed in an upright position and were instructed to stand as natural as possible with their feet shoulder-width apart and to look forward with their hands placed on the rigid lever arms.

2.4.2
Hypobaric exposure The inside experimenter was breathing 100% O 2 via a full-face diving mask and a demand valve during, and for 1 h preceding each experiment.
Subjects were then exposed to a reduced ambient pressure corresponding to an altitude of 24,000 ft (7315 m M e d i a n 2 3 2 3 0 1 connection with this, the subjects were asked and checked for any signs and/or symptoms of DCS. Prevalence of VGE was estimated from the cardiac ultrasound images using the Eftedal-Brubakk 5-degree scale (0 = no visible bubbles, 1 = occasional bubbles, 2 = at least one bubble every fourth heartbeat, 3 = at least one bubble every heartbeat, 4 = at least one bubble/cm 2 ) (Nishi et al., 2003). End-point criteria for the altitude exposures were a consistent VGE score of 4 and/or symptoms/signs of DCS (i.e., mild joint pain).
The Kisman integrated severity score (KISS) was calculated according to the following formula: where t i is time of observation in minutes after reaching altitude (for time points 1 to n), d i ultrasound score (grades 0-4) observed at time t i and α = 3 (the parameter α takes into account that the bubble grade is not a linear measure of bubble quantity) (Kisman et al., 1978;Nishi et al., 2003;Pontier & Lambrechts, 2014).

Statistical analysis
All data were statistically analysed using IBM SPSS Statistics software version 21 (IBM Corp., Armonk, NY, USA). The Shapiro-Wilk test was used to assess whether data were normally distributed (P < 0.05).
Sphericity was assessed using Mauchly's test of sphericity; where the assumption of sphericity was violated, the Greenhouse-Geisser correction was applied. Repeated measures ANOVA with post hoc Bonferroni contrast comparisons were used to assess for differences between resting baseline measurements and other collection time points for heart rate, stroke volume, cardiac output and SpO 2 . Ordinal data (i.e., VGE scores) were analysed using a statistical procedure proposed by Baguley (2012, pp. 354-358). VGE data were rank transformed (e.g., the 24 maximum VGE scores during supine rest; Table 1) and assigned rank numbers from the lowest to the highest. The ranks for the different protocols (e.g., WBV, Control and Exercise) were compared using a one-way repeated measures ANOVA with post hoc Bonferroni contrast comparisons. Furthermore, the VGE scores were also converted according to the KISS equation (Nishi et al., 2003), and differences between protocols were assessed using a oneway repeated measures ANOVA with post hoc Bonferroni contrast comparisons. Paired sample t-tests were used to assess the difference between the KISS scores collated during supine rest and after kneebend provocations during each protocol. Unless otherwise stated, data are reported as means ± SD and significance was accepted at P < 0.05, and P = 0.000 was reported as P < 0.001. GraphPad Prism version 7.0c (GraphPad Software Inc., La Jolla, CA, USA) was used to construct figures.

RESULTS
All experimental sessions were completed successfully, and none of the simulated altitude exposures and protocols resulted in DCS symptoms.

Venous gas emboli
VGE were observed across all subjects (n = 8) during the control and exercise protocols, whereas during the WBV protocol VGE were only detected in four out of eight subjects. Based on the statistical approach proposed by Baguley (2012), a significant difference in maximum VGE scores (rest and knee-bends) was discerned between the three protocols (P < 0.001), with a lower VGE score being reported in the WBV protocol (median (range), 1 (0-3)) compared with control (2 (1-3), P = 0.002) and exercise (3 (2-4), P < 0.001), while no differences were observed between control and exercise (P = 0.598) (Table 1).
Contrary to rest, where mean KISS score just failed to reach significance (P = 0.070), after knee-bend provocations a significant difference was found between protocols (P = 0.002) (Figures 1 and 2).
Moreover, when comparing the KISS scores collated during supine rest and after knee-bend provocations, a significant difference was observed in the control (P = 0.003) and exercise (P = 0.007) protocols but not in the WBV trial (P = 0.163) (Figure 2).
Additionally, we observed a significant difference in the time taken for the first VGE to be detected in the right ventricle (P = 0.014), with an earlier VGE manifestation being recorded during the control and exercise protocols (15 ± 14 min, P = 0.024, and 17 ± 24 min, P = 0.032, respectively) compared with WBV (54 ± 38 min).

Cardiovascular indices
During the time course of each hypobaric exposure, there were gradual reductions in heart rate (P ≤ 0.001), stroke volume (P ≤ 0.005) and cardiac output (P < 0.001) from baseline. More specifically, heart rate decreased from: 81 ± 9 bpm to 62 ± 9 bpm during the control protocol, 73 ± 10 bpm to 65 ± 7 bpm during the WBV protocol, and 73 ± 7 bpm to 60 ± 8 bpm during the exercise protocol.

DISCUSSION
The aim of this study was to investigate the efficacy of WBV preconditioning on high-altitude-induced VGE. Our findings signify that performing a 30-min WBV prior to a 90-min continuous exposure at 24,000 ft (7315 m) significantly delays the manifestation of VGE and effectively reduces the formation of VGE compared with control and exercise preconditioning. Additionally, during each experimental session, heart rate, stroke volume and cardiac output gradually decreased from baseline, while no differences were recorded between protocols. Consistent with our hypothesis, WBV is effective in lowering the development of high-altitude-induced VGE.
It is generally accepted that VGE emanate from pre-existing micronuclei that reside on the surface of the endothelium (Arieli & Marmur, 2017;Christman et al., 1986;Lee et al., 1993;Vann et al., 1980). During rapid decompression, nitrogen diffuses out of tissues, infuses into micronuclei and causes these precursors to grow in size (Blatteau et al., 2006;Masurel, 1989). Changes in buoyancy and flow-induced shear forces can cause the gas filled nuclei to detach from their nucleation site and enter the systemic circulation as free-floating bubbles (i.e., VGE) (Arieli & Marmur, 2013). Accordingly, the greater VGE scores recorded following knee-bend provocations in the control and exercise protocols may relate to an enhanced blood-flow-induced shear stress instigated by these manoeuvres on the endothelium, consequently enhancing the liberation of existing, adherent gas bubbles in the systemic circulation.
Amongst the three protocols examined in the present study, WBV was the most effective in lowering the formation of VGE during decompression (Table 1), thereby resulting in a lower maximum VGE score both at rest and following knee-bend provocations (Table 1) as well as a lower KISS score after knee-bends ( Figure 1). Interestingly, contrary to control and exercise, there was no significant difference between supine rest and post-knee-bend KISS scores during the WBV protocol (Figure 2), further substantiating the efficacy of this strategy in reducing the development of VGE. Our findings are in agreement with those of Germonpré et al. (2009), who showed that, performing a short bout of WBV prior to a hyperbaric exposure resulted in less VGE being detected during decompression compared with control (i.e., pre-dive supine rest). Taken together, the efficacy of vibration in lowering decompression-induced VGE seems to be independent of whether this is performed standing upright (present study) or in a supine (Germonpré et al., 2009) position, as both methods appear to be equally efficacious in reducing VGE. Additionally, WBV appears to be able to reduce VGE irrespective of whether the pressure reductions occur after a saturation situation (present study) or following a short pressure increase (Germonpré et al., 2009). The question then arises: what is the underlying mechanism(s) that dictate these responses?
To date, several studies have shown that vibration is capable of enhancing limb blood flow as well as inducing a pulsatile flow (Jawed et al., 2020;Kerschan-Schindl et al., 2001;Lythgo et al., 2009;Zhang et al., 2003) through either modifying vascular hydrophobic properties by biochemical means (i.e., increases in nitric oxide and prostaglandin release) (Maloney-Hinds et al., 2009;Sackner et al., 2005a, b) or by activating the muscle spindle receptors (Cardinale & Wakeling, 2005;Kasai et al., 1992). These undulations in flow create shear forces (Hazell et al., 2007;Suhr et al., 2007) that could mechanically dislodge gas micronuclei from their nucleation site; liberating them into the venous circulation and, subsequently, they could be eliminated by the combined effects of surface tension and the oxygen window (Blatteau et al., 2006). Since in the absence of any pre-existing small bubbles or gas micronuclei, homogeneous bubble nucleation requires very high levels of supersaturation (i.e., >∼100 ATA) (Finkelstein & Tamir, 1985;Jones et al., 1999), reducing and/or eliminating these precursors prior to decompression would supposedly lower the development of VGE.
Results from a study using a protocol that could preserve micronuclei and/or reduce their clearance support this hypothesis (Gennser et al., 2018). Specifically, after 35 days of 6 • head-down bed rest (without any exercise), subjects exhibited significantly higher bubble grades and KISS score post ahyperbaric exposure compared with preconfinement (Gennser et al., 2018). Although, to this date, we only have indirect evidence supporting the existence of these precursors (micronuclei), if this hypothesis is correct, our findings might suggest that externally applied vibration prior to pressure reduction, effectively lowers decompression-induced VGE, possibly through a shear-induced mechanical dislodgement mechanism. However, further research is necessary to ascertain or refute this hypothesis.
Exercise during decompression has been known for decades to initiate and exacerbate the problem of DCS (Henry, 1956;Krutz & Dixon, 1987;Pilmanis et al., 1999), and thus during World War II lower-extremity exercise was utilised at simulated altitude (30,000-35,000 ft) to 'screen out' flight bomber crew that were more susceptible to DCS (Henry, 1951). In contrast, the effect of exercise preconditioning on VGE is more complex, contentious and rather multifactorial, as depending on: (i) when the exercise is being carried out (i.e., 2 or 24 h prior to decompression) , (ii) the type of exercise being performed (i.e., running, cycling or squatting (low-impact vs. high-impact, concentric vs. eccentric)) (Blatteau et al., 2007;Dujic et al., 2004;Foster et al., 2013;Gennser et al., 2012;Jurd et al., 2011), and (iii) the duration/intensity of the exercise bout, it can either facilitate the creation (i.e., through muscle soreness, injury and cavitation) or the elimination of micronuclei (Dervay et al., 2002;Gennser et al., 2012;Harvey et al., 1944;Jurd et al., 2011;Provost et al., 1997;Vann and Thalmann, 1993;Wisløff et al., 2001;Wilbur et al., 2010). Earlier studies conducted in frogs and humans (Blinks et al., 1951;Dervay et al., 2002), highlighted that the incidence of VGE decreased when the rest interval from exercise to altitude depressurisation lengthened (1-2 h) and postulated that exercising prior to depressurisation may act to aggravate VGE formation.
However, neither of these studies used a control protocol, whereby no exercise was being performed prior to decompression, to evaluate the difference between physical activity and inactivity on high-altitudeinduced VGE. In the present study, when exercise preconditioning was replaced with seated rest, no differences were recorded between the timing of VGE manifestation, KISS scores (both at rest and after knee-bends) and bubble gradings during the simulated highaltitude exposure (Figures 1 and 2). Therefore, our study suggests that the present exercise modality (i.e., squats) performed just prior to decompression neither aggravates nor provides any additional safety against DCS compared with control, but further substantiates that exercise timing is an important factor that needs to be contemplated prior to high-altitude flying.
Current fighter aircraft operations are conducted without preoxygenation (i.e., time-consuming), consequently making pilots more susceptible to DCS (Ånell et al., 2020;Dervay et al., 2002;Diesel et al., 2002;Webb et al., 1998). While oxygen-rich inflight breathing gas mixtures render a better protection against DCS than nitrogenrich gases, in line with earlier studies (Ånell et al., 2020;Diesel et al., 2002;Webb & Pilmanis, 1993;Webb et al., 2000), we provide further evidence in supporting that a hyperoxic gas mixture (i.e., 100% O 2 ) is inadequate to ensure total safety from DCS at altitudes exceeding 22,500 ft; especially when decompression is preceded by either inactivity or exercise (Table 1 and Figure 2 Interestingly none of our subjects exhibited any DCS-related signs and/or symptoms, despite high VGE scores being recorded during both the control and exercise protocols (Table 1 and Figure 3). While this is not an uncommon finding (Conkin et al., 1998;Dervay et al., 2002;Diesel et al., 2002), it does, however, serve as an additional reminder that the relationship between VGE and DCS is far from causal (Bayne et al., 1985;Sawatzky, 1991). Notwithstanding, it is equally noteworthy that we did not observe any VGE crossover to the arterial side, suggesting that the high prevalence of VGE documented in the control and exercise protocols were not sufficient to overwhelm the pulmonary microcirculation's ability to filter them out (Butler & Hills, 1985).
In conclusion, the present study demonstrates that completing a 30-min WBV immediately before decompression significantly delays the generation of VGE and effectively reduces the prevalence of VGE during a continuous 90-min exposure at 24,000 ft compared with seated rest and exercise preconditioning. Moreover, similar to seated rest, exercise (squats) performed just prior to decompression neither aggravated nor provided any additional safety against the development of high-altitude-induced VGE.

ACKNOWLEDGEMENTS
We are grateful to all the subjects for their participation. We would also like to thank Björn Johannesson and Tommaso Tuci for their valuable technical assistance. The study was funded from the Swedish Armed Forces (grant no. 9220919).

COMPETING INTERESTS
The authors have no conflicts of interest to declare.

AUTHOR CONTRIBUTIONS
A

DATA AVAILABILITY STATEMENT
The datasets presented in this article are not readily available as sharing these will compromise the ethical standards and agreement with the subjects.