The effects of acute normobaric hypoxia on vestibular-evoked balance responses in humans

Abstract

BACKGROUND:

Hypoxia influences standing balance and vestibular function.

OBJECTIVE:

The purpose here was to investigate the effect of hypoxia on the vestibular control of balance.

METHODS:

Twenty participants (10 males; 10 females) were tested over two days (normobaric hypoxia and normoxia). Participants stood on a force plate (head rotated leftward) and experienced random, continuous electrical vestibular stimulation (EVS) during trials of eyes open (EO) and closed (EC) at baseline (BL), after 5 (H1), 30 (H2) and 55-min (H3) of hypoxia, and 10-min into normoxic recovery (NR). Vestibular-evoked balance responses were quantified using cumulant density, coherence, and gain functions between EVS and anteroposterior forces.

RESULTS:

Oxyhemoglobin saturation, end-tidal oxygen and carbon dioxide decreased for H1-3 compared to BL; however, end-tidal carbon dioxide remained reduced at NR with EC (p≤0.003). EVS-AP force peak-to-peak amplitude was lower at H3 and NR than at BL (p≤0.01). At multiple frequencies, EVS-AP force coherence and gain estimates were lower at H3 and NR than BL for females; however, this was only observed for coherence for males.

CONCLUSIONS:

Overall, vestibular-evoked balance responses are blunted following normobaric hypoxia >30 min, which persists into NR and may contribute to the reported increases in postural sway.

Keywords

Hypoxia low oxygen balance control vestibular system electrical vestibular stimulation

1 Introduction

Hypoxia negatively influences human function and is present in certain medical conditions (e.g. cancer, chronic obstructive pulmonary disease, obstructive sleep apnea syndrome) [14, 36], occupations and leisure activities. For example, unacclimatized lowlanders who ascend to high altitude experience acute effects of hypoxia within minutes to days upon their arrival and may exhibit chronic symptoms over a number of weeks. Individuals often ascend to high altitude for travel, recreation, and various occupations (e.g., mining, road construction, or military). Therefore, from both a leisure and occupational health perspective it is important to understand how hypoxia alters human physiology and motor control (e.g., standing balance) as it can help identify hypoxia-related factors influencing safety and overall function.

Quiet standing is disrupted during hypoxia, such that the magnitude and velocity of postural sway increases when compared to normoxia [24]. Standing balance control relies on multiple sensorimotor signals, which are integrated and processed within the central nervous system [61]. Therefore, a disruption in any sensorimotor pathway may lead to an impairment in quiet standing balance when exposed to hypoxia. Early reports have described vestibular disturbances upon arrival to high altitude (∼3505 m), which were characterized by increased prevalence of spontaneous and positional nystagmus [65]. Further, reports using a rodent model indicate normobaric hypoxia (0.10 fraction of inspired oxygen - F_IO₂) and hypoxic artificial cerebrospinal fluid (0.05 O₂) lead to increased glutamate release and medial vestibular neuronal firing rates [39, 76]. The hypoxia-related increase in vestibular neuron firing rates could be a source of dysfunction within the vestibular system of humans and influence overall balance control. Presently, the influence of hypoxia on the vestibular control of balance, one of the primary functions of the vestibular system, is unclear.

The vestibular control of standing balance can be evaluated using a non-invasive, but powerful tool, known as electrical vestibular stimulation (EVS) [12 , 51], which modulates the discharge rates of all types of primary vestibular afferents [47]. A bipolar, binaural EVS signal produces a virtual signal of head motion [22 , 60], which is interpreted by the central nervous system as an isolated ex-afferent vestibular perturbation [31]. A benefit of EVS is that it affords the evaluation of how the central nervous system responds to an isolated vestibular error signal free from other sensory inputs [5, 22]. The central nervous system then produces a compensatory adjustment within postural muscles [2 , 74] that summate to produce a whole-body balance response [50 , 66]. Currently, it is unclear how vestibular-evoked balance responses are altered during hypoxia. Therefore, the purpose of this study was to determine the influence of normobaric hypoxia on whole-body vestibular-evoked balance responses. It was hypothesized vestibular-evoked balance responses would be greater during normobaric hypoxia compared with normoxia.

2 Methods

2.1 Participants

Twenty (10 males: 26.4±6.1 years; 180.1±10.0 cm; 81.0±16.2 kg; 10 females: 24.1±3.2 years; 168.2±5.8 cm; 63.9±5.8 kg) healthy, young participants were recruited from the university population. Based on data from a previous study in our lab [53] that found significant differences in vestibular-evoked balance responses for a within subject design (alpha = 0.05, partial eta squared effect size = 0.48, power = 0.8), an a priori sample size calculation determined that only 5 participants were required. However, to ensure a representation of both sexes, we included 10 males and 10 females in the current study. All participants were free from chronic diseases affecting balance and/or respiratory function and were excluded if they were pregnant, had uncontrolled hypertension (systolic blood pressure above 135 mmHg), experienced common occurrences of dizziness, or were a current or ex-smoker. All participants granted informed consent prior to data collection and the study procedures were approved by the University of British Columbia’s Clinical Research Ethics Board (H17-03178) in accordance with Canada’s Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans.

2.2 Experimental set-up

Each participant stood upright, and barefoot on a force plate (AMTI Model OR6-5-1 Biomechanics Platform, Advanced Mechanical Technology Inc, Newton, MA, USA) with medial malleoli touching, and arms relaxed at their sides. Participants rotated their head in yaw 90° leftward from anatomical (Fig. 1A). A target was placed on a wall and a laser pointer, positioned over the right ear, was used to maintain a consistent head posture during the EVS trials (Reid’s plane tilted ∼19° upward from horizontal; Fig. 1A). During trials where vision was occluded with a blindfold, experimenters verbally directed participants to maintain the required head position. Whole-body ground reaction forces were collected using a force plate. A data acquisition board (Power1401-3A, Cambridge Electronic Design Limited, Milton, CB, UK) and Spike2 software (version 8; Cambridge Electronic Design Limited, Milton, CB, UK) were used to digitize force plate data at 2041 Hz.

Fig. 1

A, Experimental setup. Participants stood with their feet together on a force plate while looking over their left shoulder, wearing a facemask, and pulse oximeter on their left index finger. A laser pointer was secured above the right ear (Reid’s plane tilted ∼19° upward from horizontal). Electrical vestibular stimulation (EVS) was delivered via carbon-rubber electrodes positioned, bilaterally, over the mastoid processes. (B) Unprocessed data of 5 s of EVS (mA) and 5 s of anteroposterior force tracings (AP Force; N). (C) Example of a derived vestibular-evoked balance response using a cumulant density function for EVS-AP Force. The dashed lines represent 95% confidence intervals. (D) The vestibular-evoked balance response was assessed at five timepoints: Baseline (BL; ∼5 min pre-hypoxia), Hypoxia 1 (H1; 5 min), Hypoxia 2 (H2; 30 min), Hypoxia 3 (H3; 55 min), and Normoxic recovery (NR; 10 min post hypoxia). On Day 1, BL and NR were evaluated in normoxia, and H1-3 were tested in normobaric hypoxia where the fraction of inspired oxygen was lowered from ∼0.21 to 0.12 for 60 min. On Day 2, all testing blocks occurred in normoxia. The boxes indicate time periods for EVS with eyes open (unfilled) or closed (filled). The order of visual conditions was randomized between participants, but consistent within participants.

All respiratory parameters were acquired at 200 Hz with an analog-to-digital converter (Powerlab/16SP ML 880; AD Instruments, Colorado Springs, CO, USA). A hypoxic generator and facemask system (HYP-123, Hypoxico, New York City, NY, USA) was used to create normobaric hypoxia (Fig. 1A). As previous work has demonstrated vestibular disturbances at ∼3505 m (spontaneous and positional nystagmus) [65], the current study induced normobaric hypoxia with a F_IO₂ of 0.12 (∼4000 m). Respired gas samples were taken at the mouth and analyzed for the partial pressure of end-tidal oxygen (P_ETO₂) and carbon dioxide (P_ETCO₂) (ML206, AD instruments, Colorado Springs, CO, USA). The participants were also instrumented with a finger pulse oximeter (7500FO, Nonin Medical Inc, Plymouth, MN, USA) on their left arm, to evaluate blood oxyhemoglobin saturation levels (SpO₂).

Carbon-rubber electrodes (3×4 cm) were coated with conducting gel (Spectra 360, Parker Laboratories, Fairfield, NJ, USA) and positioned, bilaterally, over the mastoid processes for the vestibular stimulation (Fig. 1A). The electrodes were secured with Durapore tape (3M innovations, St. Paul, MN, USA) and an elastic headband. The stimulus was created using MATLAB software (R2020a, MathWorks Inc., Natick, MA, USA) with freely available code [32]. The EVS signal was a low-pass filtered (25 Hz; 6th order Butterworth) white noise that was scaled to a peak-to-peak amplitude of±4 mA with a bandwidth of 0–25 Hz (root-mean-square amplitude: 0.9 mA). Using Spike2 software (version 8; Cambridge Electronic Design Limited, Milton, CB, UK) and a data acquisition interface (Power1401-3A, Cambridge Electronic Design Limited, Milton, CB, UK), the EVS signal was delivered via an isolated bipolar current stimulator (DS5, Digitimer LTD, Welwyn Garden City, HD, UK) and sampled at 2041 Hz.

2.3 Experimental protocol

The study was completed over 2 testing sessions (a hypoxia day and normoxia day) separated by >24 hours. The hypoxia day was intended to determine the effects of normobaric hypoxia on vestibular-evoked balance responses; whereas the normoxia day was used to determine if the experimental protocol induced habituation of the vestibular-evoked balance response. During both testing sessions, participants were exposed to five timepoints of EVS. Each timepoint began with 1 min of quiet standing before completing two 90-s trials of stimulation (Trial 1 = eyes open [EO]; Trial 2 = eyes closed [EC]) with each trial separated by 1 min (∼5 min timepoints; Fig. 1D). A blindfold was used in the EC condition to occlude vision. The starting visual condition and testing day session was pseudo-randomized and counter-balanced between participants. The first timepoint for all experimental sessions occurred in normoxia following a 5-min period of quiet standing where the participant became accustomed to breathing through the facemask system, followed by a 60-min exposure to normobaric hypoxia (F_IO₂ = 0.12). During the hypoxia exposure, 3 stimulation timepoints occurred at 5, 30, and 55 min (Fig. 1D). Then, the participants were returned to normoxia by removing the connection to the hypoxic generator. The facemask remained in place for all conditions. Ten min following the termination of hypoxia, the final timepoint of vestibular stimulation occurred (Fig. 1D). Between each timepoint, participants were seated. During the normoxia testing session, the same order of timepoints occurred while the participants breathed normoxic gas while wearing the facemask for the duration of the protocol (Fig. 1D).

2.4 Data analysis

Commercially available software was used to analyse the ventilatory and cardiovascular variables (LabChart V7.1, AD Instruments, Bella Vista, NSW, AU). The last 60 s of recorded data (P_ETCO₂, P_ETO₂, and SpO₂) during each EVS trial were averaged and used for statistical analysis.

Center of pressure (CoP) data were analyzed in MATLAB (R2020a, Mathworks, Natick, MA, USA). Two outcome variables were analyzed to assess postural control during each timepoint for EO and EC: mean CoP displacement standard deviation (CoP-SD) and full-wave rectified CoP velocity (over 90 s per trial) in the AP and mediolateral (ML) directions. The CoP-SD and CoP velocity at each timepoint were normalized to values at BL and compared statistically.

The relationship between the input (EVS) and output (AP force) signal were used to characterize the effect of hypoxia on vestibular-evoked balance responses. Unprocessed data traces of the EVS and AP force signals are provided in Fig. 1B. The sampled AP forces were time-locked to the onset of the EVS signal, and each participant contributed 44 segments of data for each condition (segment length: ∼2.007 s; resolution: ∼0.498 Hz). Vestibular-evoked balance responses were evaluated within the time domain (i.e., cumulant density), where the data of each participant was examined individually, and within the frequency domain (coherence and gain), where the data were pooled to estimate relationships between the EVS signal (input) and motor output (AP force). Cumulant density, coherence, and gain estimates were derived using an archive of MATLAB (R2020a, Mathworks, Natick, MA, USA) functions based on multivariate Fourier analysis (Neurospec 2.0, http://www.neurospec.org) [37, 62].

The cumulant density function is akin to a cross-covariance and was used to represent the time domain relationships between the vestibular stimuli and motor output signal (AP force). Estimated values derived from the cumulant density function are represented by a correlation coefficient bound between 1 and –1 as they were normalized by the vector norms of the input (i.e., EVS) and motor output (i.e., forces; [19]). Within the time domain, the vestibular-evoked balance response can be characterized by a biphasic response (Fig. 1C) and for the purposes of this study, vestibular evoked-balance responses were quantified as the peak-to-peak amplitude. Before values were assessed for statistical analysis, it was determined per participant that cumulant density functions were significantly different from zero by ensuring that values exceeded the 95% confidence limits that were calculated from the total number of data segments [37, 62].

To describe the frequency bandwidth of the vestibular-evoked whole-body balance response, coherence estimates were derived. Coherence represents a measure of linear relationship between the input (i.e., EVS signal) and an output (i.e., AP force) across a range of frequencies. Coherence estimates were derived from the participant pooled data for sex (males vs females), visual condition (EO vs EC), and timepoint (BL vs H1-NR). For every frequency value, coherence varies from 0 (no linear relationship) to 1 (perfect linear relationship containing no noise; [37, 62]). The EVS-AP force coherence was considered significantly different than 0 when the values exceeded the 95% confidence limit that was constructed from the total segments per participant [37].

Differences in coherence between BL and all other timepoints (H1-NR) for the hypoxia day were assessed with the “Difference of Coherence” subroutine in NeuroSpec 2.0 [4, 62]. This analysis tests the assumption that the coherence estimates are equal within a normally distributed variance by comparing the standardized differences between coherence of the two trials and 95% confidence limits. To determine if there was a difference between EO and EC in the frequency domain on the normoxia day at BL, male and female data were pooled to generate 890 segments (segment length: 2.0069 s; resolution: 0.4983 Hz) per visual condition. Similarly, to determine if there was a sex difference within each visual condition in the frequency domain on the normoxia day at BL, males and females were pooled separately per visual condition to generate 445 segments (segment length: 2.0069 s; resolution: 0.49829 Hz). On the hypoxia day, male and female data were pooled at each timepoint per visual condition to generate 445 segments (segment length: 2.0069 s; resolution: 0.49829 Hz) for each condition per timepoint.

To describe the magnitude of the vestibular-evoked balance response across significant EVS-AP force coherence frequencies, gain functions were evaluated. The gain estimates were normalized to the gain at the lowest frequency data point (∼0.5 Hz). To compare normoxia (BL) to hypoxia (H1-3) and subsequent normoxic recovery (NR), point wise 95% confidence limits were constructed for each timepoint and the frequencies for which gain confidence limits did not overlap were considered statistically different.

2.5 Statistical analysis

Statistical analyses were conducted using SPSS (version 26, IBM, Armonk, NY, USA). To determine if habituation influenced the vestibular-evoked balance response, a three-way repeated measures analysis of variance (ANOVA) for timepoint (BL-NR), vision (EC vs EO), and sex (males vs females) on the EVS–AP force peak-peak amplitude was conducted on the normoxia day. For the hypoxia day, a three-way repeated measures ANOVA was conducted to test for differences in sex, vision, and timepoint on the EVS–AP force peak-to-peak amplitude as well as the CoP-SD and velocity in both the ML and AP directions. To determine if hypoxia influenced the participants cardiorespiratory function, a three-way repeated measures ANOVA was conducted for timepoint (BL-NR), vision (EC and EO), and sex (males vs females) on SpO₂, P_ETO₂, and P_ETCO₂. The Greenhouse-Geiser Epsilon factor was used to adjust degrees of freedom for violations of sphericity when necessary. Statistical significance was set at p≤0.05. Post-hoc tests were performed using paired t-tests with a Holm-Bonferroni correction where appropriate. Effect sizes were estimated using partial eta squared or Cohen’s d where appropriate. The results of all three-way repeated measures ANOVAs can be found in Tables 1 and 2 for the normoxia and hypoxia testing sessions, respectively. Data are presented as means±standard deviations (SD).

Table 1
Results of the three-way repeated measures ANOVA (Vision × Timepoint × Sex) on the normoxia day for cardiorespiratory variable (SpO₂ - oxyhemoglobin saturation, P_ETO₂ - end-tidal oxygen, and P_ETCO₂ - end-tidal carbon dioxide), and anteroposterior (AP) force peak-to-peak amplitude

Normoxia testing session

Variable Factors F p η _p ²

SpO₂ (%) Sex 8.65 0.01* 0.32

Vision 0.73 0.40 0.03

Vision×Sex 0.40 0.53 0.02

Timepoint 1.46 0.22 0.07

Timepoint×Sex 0.42 0.78 0.02

Vision×Timepoint 0.70 0.59 0.03

Vision×Timepoint×Sex 1.70 0.15 0.08

P_ETO₂ (mmHg) Sex 3.92 0.06 0.17

Vision 7.92 0.01* 0.30

Vision×Sex 1.54 0.23 0.07

Timepoint 0.19 0.94 0.01

Timepoint×Sex 1.34 0.96 0.007

Vision×Timepoint 0.94 0.44 0.05

Vision×Timepoint×Sex 1.50 0.20 0.07

P_ETCO₂ (mmHg) Sex 8.30 0.01* 0.31

Vision 32.64 <0.001* 0.64

Vision×Sex 0.81 0.37 0.04

Timepoint 0.98 0.41 0.05

Timepoint×Sex 0.15 0.95 0.009

Vision×Timepoint 0.85 0.49 0.04

Vision×Timepoint×Sex 1.72 0.15 0.08

EVS-AP force peak-to-peak amplitude Sex 0.95 0.34 0.05

Vision 4.42 0.05* 0.19

Vision×Sex 0.03 0.95 <0.001

Timepoint 0.95 0.43 0.05

Timepoint×Sex 0.56 0.60 0.03

Vision×Timepoint 0.66 0.61 0.03

Vision×Timepoint×Sex 1.51 0.20 0.07

Normoxia testing session
SpO₂ (%)	Sex	8.65	0.01*	0.32
	Vision	0.73	0.40	0.03
	Vision×Sex	0.40	0.53	0.02
	Timepoint	1.46	0.22	0.07
	Timepoint×Sex	0.42	0.78	0.02
	Vision×Timepoint	0.70	0.59	0.03
	Vision×Timepoint×Sex	1.70	0.15	0.08
P_ETO₂ (mmHg)	Sex	3.92	0.06	0.17
	Vision	7.92	0.01*	0.30
	Vision×Sex	1.54	0.23	0.07
	Timepoint	0.19	0.94	0.01
	Timepoint×Sex	1.34	0.96	0.007
	Vision×Timepoint	0.94	0.44	0.05
	Vision×Timepoint×Sex	1.50	0.20	0.07
P_ETCO₂ (mmHg)	Sex	8.30	0.01*	0.31
	Vision	32.64	<0.001*	0.64
	Vision×Sex	0.81	0.37	0.04
	Timepoint	0.98	0.41	0.05
	Timepoint×Sex	0.15	0.95	0.009
	Vision×Timepoint	0.85	0.49	0.04
	Vision×Timepoint×Sex	1.72	0.15	0.08
EVS-AP force peak-to-peak amplitude	Sex	0.95	0.34	0.05
	Vision	4.42	0.05*	0.19
	Vision×Sex	0.03	0.95	<0.001
	Timepoint	0.95	0.43	0.05
	Timepoint×Sex	0.56	0.60	0.03
	Vision×Timepoint	0.66	0.61	0.03
	Vision×Timepoint×Sex	1.51	0.20	0.07

*Indicates p≤0.05. η_p² (partial eta squared) is a measure of effect size.

Table 2

Statistical results of the three-way repeated measures ANOVA (Vision × Timepoint × Sex) on the hypoxia day for cardiorespiratory (SpO₂ - oxyhemoglobin saturation, P_ETO₂ - end-tidal oxygen, and P_ETCO₂ - end-tidal carbon dioxide), and center of pressure (CoP) standard deviation (SD) and velocity variables, and anteroposterior (AP) force peak-to-peak amplitude

Hypoxia testing session
Variable	Factors	F	p	η _p ²
SpO₂ (%)	Sex	0.30	0.58	0.17
	Vision	5.98	0.02*	0.24
	Vision×Sex	0.10	0.75	0.006
	Timepoint	240.17	<0.001*	0.93
	Timepoint×Sex	1.61	0.17	0.08
	Vision×Timepoint	1.31	0.27	0.06
	Vision×Timepoint×Sex	1.56	0.19	0.08
P_ETO₂ (mmHg)	Sex	0.56	0.46	0.03
	Vision	15.18	0.001*	0.45
	Vision×Sex	0.09	0.75	0.005
	Timepoint	1043.84	<0.001*	0.98
	Timepoint×Sex	0.16	0.95	0.009
	Vision×Timepoint	1.28	0.28	0.06
	Vision×Timepoint×Sex	1.96	0.11	0.09
P_ETCO₂ (mmHg)	Sex	4.20	0.05*	0.18
	Vision	6.13	0.02*	0.25
	Vision×Sex	0.00	0.98	<0.001
	Timepoint	29.59	0.00*	0.62
	Timepoint×Sex	0.38	0.81	0.02
	Vision×Timepoint	2.63	0.04*	0.12
	Vision×Timepoint×Sex	0.49	0.73	0.02
CoP AP Velocity (%)	Sex	3.61	0.07	0.19
	Vision	3.85	0.06	0.20
	Vision×Sex	0.04	0.84	0.003
	Timepoint	0.76	0.55	0.04
	Timepoint×Sex	1.55	0.19	0.09
	Vision×Timepoint	1.84	0.13	0.10
	Vision×Timepoint×Sex	0.54	0.70	0.04
CoP ML Velocity (%)	Sex	1.68	0.22	0.10
	Vision	2.99	0.10	0.16
	Vision×Sex	0.03	0.86	0.002
	Timepoint	0.61	0.65	0.03
	Timepoint×Sex	1.97	0.11	0.11
	Vision×Timepoint	1.60	0.18	0.09
	Vision×Timepoint×Sex	0.19	0.94	0.01
CoP AP SD (%)	Sex	0.91	0.35	0.05
	Vision	0.40	0.53	0.02
	Vision×Sex	0.58	0.45	0.03
	Timepoint	2.74	0.03*	0.15
	Timepoint×Sex	1.71	0.15	0.10
	Vision×Timepoint	1.01	0.40	0.06
	Vision×Timepoint×Sex	0.83	0.50	0.05
CoP ML SD (%)	Sex	2.98	0.10	0.16
	Vision	1.43	0.24	0.08
	Vision×Sex	0.00	0.94	<0.001
	Timepoint	1.30	0.27	0.08
	Timepoint×Sex	1.20	0.31	0.07
	Vision×Timepoint	1.18	0.32	0.07
	Vision×Timepoint×Sex	1.04	0.39	0.65
EVS-AP force peak-to-peak amplitude	Sex	0.07	0.93	<0.001
	Vision	1.26	0.27	0.07
	Vision×Sex	0.06	0.80	0.003
	Timepoint	5.23	<0.001*	0.23
	Timepoint×Sex	0.29	0.89	0.02
	Vision×Timepoint	0.79	0.53	0.04
	Vision×Timepoint×Sex	1.16	0.34	0.06

*Indicates p≤0.05. η_p² (partial eta squared) is a measure of effect size.

3 Results

3.1 Technical limitations

There were 3 participants whose CoP data were outliers (>3 SDs from the mean of all participants). Thus, for CoP SD and CoP velocity, 17 participants were included in the final data set (9 males: 27.0±6.2 years; 179.7±10.5 cm; 81.2±17.2 kg; 8 females: 24.0±3.4 years; 168.6±6.1 cm; 63.5±6.1 kg). Further, owing to technical difficulties, the SpO₂ values for one participant were not sampled during EO BL on the normoxia day. Therefore, to calculate this missing value, the “Impute Missing Data Values” function on SPSS (version 26) was used. This function generated five possible imputations based off the SpO₂ values from the remaining participants in the study on the normoxia day (n = 19, mean = 96.1 %, SD = 1.54 %, minimum value = 93.7 %, and maximum value = 99.6 %). The imputation generated 96.7, 96.2, 97.0, 97.2, 96.2 % as possible values at BL. These five values were then averaged to calculate an SpO₂ value of 96.7 % for the participant, which was then used for statistical analysis.

3.2 Normoxia testing session

3.2.1 Cardiorespiratory data

There was no detectable timepoint effect for SpO₂, P_ETO₂, and P_ETCO₂ (Table 1). However, there was a vision effect (Tables 1 and 3) whereby P_ETO₂ was larger in EO compared to EC (p = 0.01), and P_ETCO₂ had larger values in EC than EO (p≤0.001). There was a main effect of sex on SpO₂ and P_ETCO₂ (Tables 1 and 3) where females had larger SpO₂ (p = 0.009), but lower P_ETCO₂ values than males (p = 0.01). There was no main effect of sex for P_ETO₂ (Table 1).

Table 3
Mean values for SpO₂, P_ETO₂, and P_ETCO₂ (Oxyhemoglobin saturation, and end-tidal oxygen and carbon dioxide, respectively) obtained on the normoxia day at each timepoint for both males and females, in the eyes open and closed conditions

Normoxia testing session

Variable Sex, Visual condition BL1 H1 H2 H3 NR

SpO₂ (%) Females, Eyes closed 96.8±1.5† 96.7±1.7† 97.2±1.5† 97.4±1.7† 97.4±1.6†

Females, Eyes open 96.8±1.6† 96.7±1.8† 97.1±2.0† 97.0±2.3† 97.4±1.4†

Males, Eyes closed 95.3±1.2 95.1±1.1 95.2±1.0 95.4±1.7 95.6±1.2

Males, Eyes open 95.5±1.1 95.0±1.2 95.2±1.1 95.6±1.3 95.3±1.3

P_ETO₂ (mmHg) Females, Eyes closed 98.6±5.8* 100.7±6.8* 100.0±6.7* 100.9±6.1* 99.2±4.4*

Females, Eyes open 101.1±7.7 100.4±7.4 100.6±9.1 100.1±5.7 100.6±5.8

Males, Eyes closed 95.1±3.2* 95.4±3.1* 95.4±3.5* 94.9±2.8* 95.2±2.8*

Males, Eyes open 96.6±3.9 96.6±3.2 96.6±3.5 97.1±3.6 97.1±4.0

P_ETCO₂ (mmHg) Females, Eyes closed 32.5±2.8† 31.6±3.7† 31.8±3.0† 31.3±3.6† 31.9±2.4†

Females, Eyes open 31.3±3.3† 31.3±3.8† 31.1±3.4† 31.5±3.1† 31.4±3.0†

Males, Eyes closed 35.8±2.2 35.3±2.2* 35.5±2.5* 35.5±2.6* 35.4±2.6*

Males, Eyes open 35.2±2.7 34.6±2.2 34.9±2.4 34.7±3.0 34.7±3.0

Normoxia testing session
SpO₂ (%)	Females, Eyes closed	96.8±1.5†	96.7±1.7†	97.2±1.5†	97.4±1.7†	97.4±1.6†
	Females, Eyes open	96.8±1.6†	96.7±1.8†	97.1±2.0†	97.0±2.3†	97.4±1.4†
	Males, Eyes closed	95.3±1.2	95.1±1.1	95.2±1.0	95.4±1.7	95.6±1.2
	Males, Eyes open	95.5±1.1	95.0±1.2	95.2±1.1	95.6±1.3	95.3±1.3
P_ETO₂ (mmHg)	Females, Eyes closed	98.6±5.8*	100.7±6.8*	100.0±6.7*	100.9±6.1*	99.2±4.4*
	Females, Eyes open	101.1±7.7	100.4±7.4	100.6±9.1	100.1±5.7	100.6±5.8
	Males, Eyes closed	95.1±3.2*	95.4±3.1*	95.4±3.5*	94.9±2.8*	95.2±2.8*
	Males, Eyes open	96.6±3.9	96.6±3.2	96.6±3.5	97.1±3.6	97.1±4.0
P_ETCO₂ (mmHg)	Females, Eyes closed	32.5±2.8*†	31.6±3.7*†	31.8±3.0*†	31.3±3.6*†	31.9±2.4*†
	Females, Eyes open	31.3±3.3†	31.3±3.8†	31.1±3.4†	31.5±3.1†	31.4±3.0†
	Males, Eyes closed	35.8±2.2*	35.3±2.2*	35.5±2.5*	35.5±2.6*	35.4±2.6*
	Males, Eyes open	35.2±2.7	34.6±2.2	34.9±2.4	34.7±3.0	34.7±3.0

*Indicates a significant difference from the eyes open condition and †indicates a significant difference from males (p≤0.05).

3.2.2 Vestibular-evoked balance responses –time domain

On the normoxia day, there was a vision main effect for EVS-AP force peak-peak amplitude (Table 1), where EC had ∼9% larger values than EO (0.15±0.03 vs 0.14±0.04, respectively). There were no other main effects, nor interactions detected (Table 1).

3.2.3 Vestibular-evoked balance responses –frequency domain

The EVS-AP force coherence exceeded the 95% confidence limit for all subjects. When comparing visual conditions (EC vs EO) with sex collapsed at BL on the normoxia day, the EVS-AP force coherence was larger during the EC compared to EO at multiple frequencies ≤12.5 Hz (Supplementary Figure 1). The EVS–AP force gain in the EC condition was larger than EO at <2 Hz (Supplementary Figure 1). Because there were differences in coherence and gain between EO and EC at BL, visual conditions were analyzed separately in the frequency domain for all other statistical analyses.

When comparing sexes during BL, EVS-AP force coherence was larger in males than females at lower frequencies during EC and EO (≤4 and 3 Hz respectively: Supplementary Figure 2A and B), but lower coherence values at multiple higher frequencies ≥5 Hz than females (Supplementary Figure 2A and B). In both the vision conditions, the EVS–AP force gain was larger in females than males at lower frequencies (EC: ∼1-2 and 4–5.5 Hz; EO: ∼0.5–2, and 3.5–8 Hz; Supplementary Figure 2A and B). Because there were sex-related differences in EVS-AP force coherence and gain at BL, further statistical analyses were performed separately for males and females to determine the effects of hypoxia.

3.3 Hypoxia testing session

3.3.1 Cardiorespiratory data

There was a main effect of vision, and timepoint, but no detectable sex or interaction effects for SpO₂ (Table 2). Compared to BL, the SpO₂ values were reduced at H1-H3 (p≤0.001), but not at NR (p = 1.00; Fig. 2A). There was also a main effect of vision and timepoint, but not sex for P_ETO₂ (Table 2). The P_ETO₂ values were greater in EO compared to EC (p≤0.001) and were reduced at H1-H3 (p≤0.001), but not at NR compared with BL (p = 0.22; Table 2; Fig. 2B). Similarly, there was a main effect of vision, timepoint and sex for P_ETCO₂ (Table 2). During EC, P_ETCO₂ values were greater than EO (p = 0.02), and both visual conditions exhibited a reduction for all timepoints compared to BL (p≤0.02). Further, P_ETCO₂ values were lower for females than males throughout testing (p = 0.05, Fig. 2C). There was also an interaction between vision and timepoint for P_ETCO₂ (Table 2). When sex was collapsed and visual conditions were separated, there was a main effect of timepoint in the EC condition (F = 35.05, p≤0.001, η_p² = 0.64) where P_ETCO₂ was reduced at H1-NR compared to BL (p≤0.003; Fig. 2CC), yet in the EO conditions (F = 20.46, p≤0.001, η_p² = 0.51) P_ETCO₂ was only reduced at H1-H3 compared to BL (p≤0.001; Fig. 2C).

Fig. 2

Cardiorespiratory changes (A; SpO₂ - oxyhemoglobin saturation, B; P_ETO₂ - end-tidal oxygen, and C; P_ETCO₂ - end-tidal carbon dioxide) across timepoints during the hypoxia testing session. The SpO₂ (%), P_ETO₂ (mmHg), and P_ETCO₂ (mmHg), values all decreased during normobaric hypoxia (H1-H3) compared with baseline (BL; ^*p≤0.05) while P_ETCO₂ remained lower than BL at 10 min post hypoxia for eyes closed (^*p≤0.05). Squares represent mean values±standard deviations for females (unfilled) and males (filled). Circles indicate individual values for male (filled) and female (unfilled) participants.

3.3.2 Center of pressure

There were no main effects nor interactions detected for CoP AP velocity, CoP ML SD, or CoP ML velocity (Table 2; Fig. 3A–C). While there was no effect of vision or sex on CoP AP SD, there was a main effect of timepoint (Table 2). When sex and vision were collapsed, CoP AP SD increased at H1 (p = 0.02, d = 0.42) and H2 (p = 0.01, d = 0.58) from BL (Fig. 3D).

Fig. 3

Center of pressure (CoP) displacement standard deviation (SD) and velocity in the mediolateral (ML) and anteroposterior (AP) directions across timepoints on the hypoxia day. Values are normalized to baseline (BL; CoP AP velocity = 23.2±1.6 mm/s; CoP ML velocity = 17.9±1.1 mm/s; CoP ML SD = 7.0±0.2 mm/s; CoP AP SD = 9.0±0.3 mm/s). There were no detectable effects (p > 0.05) of normobaric hypoxia on CoP ML velocity (A), CoP ML SD (B), or CoP AP velocity (C), but H1 and H2 were greater than BL for CoP AP SD (D; ^*p≤0.05). Squares are mean values±standard deviations. Circles indicate values for individual female (unfilled) and male (filled) participants.

3.3.3 Vestibular-evoked balance responses –cumulant density

There was a main effect of timepoint for the EVS-AP force peak-to-peak amplitude, but no sex nor vision effects and no interactions were detected (Table 2). When sex and vision were collapsed, EVS-AP force peak-to-peak amplitude was reduced by ∼12 % at H3 (p = 0.01, d = –0.45) and ∼17 % at NR (p = 0.007, d = –0.56) compared to BL with no other detectable differences (p≥0.21, Fig. 4).

Fig. 4

The electrical vestibular stimulation (EVS)-anteroposterior (AP) force peak-to-peak amplitude decreased by 55 min into normobaric hypoxia (H3) and remained lower following 10 min of normoxic recovery (NR) compared to baseline (BL; ^*p≤0.05). Squares are mean values±standard deviations collapsed across sexes and visual conditions. Circles indicate individual values for females (unfilled) and males (filled).

3.3.4 Vestibular-evoked balance responses –coherence and gain

The EVS-AP force coherence exceeded the 95% confidence limit for all subjects. During EC, the EVS-AP force coherence for females at H1 was greater than BL at multiple frequencies ≤15 Hz (Fig. 5A); however, for other timepoints females exhibited lower coherence at frequencies ≥7 and ≥5 Hz for EC (H3 and NR –Fig. 5C and D) and EO (H2-NR; Fig. 6B–D) compared BL, respectively. In males, EVS-AP force coherence during EC was lower for multiple frequencies ≤12 Hz at H1-NR than BL (Fig. 7A–D); whereas during EO, EVS-AP force coherence at H1, H2 and NR demonstrated greater values than BL at ∼4–9 Hz (Fig. 8A, B and D).

Fig. 5

Coherence (left column), difference of coherence (inset; DoC), and gain (right column) estimates were calculated from pooled data from females with eyes closed (EC) comparing Baseline (BL) to Hypoxia 1 (H1; A), Hypoxia 2 (H2; B), Hypoxia 3 (H3; C), and Normoxic recovery (NR; D) on the hypoxia testing day. The black line represents H1 while the grey line depicts timepoints H1-3 and NR in each respective panel and dashed lines represent 95% confidence intervals for coherence and DoC. For the DoC estimates between BL and H1-NR, the light grey shading represents greater coherence for BL compared to H1-NR, whereas dark grey denotes the opposite. The dark and light grey shaded areas represent point-wise 95% confidence limits for the BL and H1-NR, respectively. Timepoints were considered statistically different from each other when the confidence limits did not overlap. Gain estimates were enhanced at H1 and H2, but lower by H3 and into NR for multiple frequencies. EVS – electrical vestibular stimulation, AP – anteroposterior.

Fig. 6

Coherence (left column), difference of coherence (inset; DoC), and gain (right column) estimates were calculated from pooled data from females with eyes open (EO) comparing Baseline (BL) to Hypoxia 1 (H1; A), Hypoxia 2 (H2; B), Hypoxia 3 (H3; C), and Normoxic recovery(NR; D) on the hypoxia testing day. The black line represents H1 while the grey line depicts timepoints H1-3 and NR in each respective panel and dashed lines represent 95% confidence intervals for coherence and DoC. For the DoC estimates between BL and H1-NR, the light grey shading represents greater coherence for BL compared to H1-NR, whereas dark grey denotes the opposite. The dark and light grey shaded areas represent point-wise 95% confidence limits for the BL and H1-NR, respectively. Timepoints were considered statistically different from each other when the confidence limits did not overlap. Gain estimates were lower for multiple frequencies by H3 and following 10 min of NR. EVS –electrical vestibular stimulation, AP –anteroposterior.

Fig. 7

Coherence (left column), difference of coherence (inset; DoC), and gain (right column) estimates were calculated from pooled data for males with eyes closed (EC) comparing Baseline (BL) to Hypoxia 1 (H1; A), Hypoxia 2 (H2; B), Hypoxia 3 (H3; C), and Normoxic recovery (NR; D) on the hypoxia testing day. The black line represents H1 while the grey line depicts timepoints H1-3 and NR in each respective panel and dashed lines represent 95% confidence intervals for coherence and DoC. For the DoC estimates between BL and H1-NR, the light grey shading represents greater coherence for BL compared to H1-NR, whereas dark grey denotes the opposite. The dark and light grey shaded areas represent point-wise 95% confidence limits for the BL and H1-NR, respectively. Timepoints were considered statistically different from each other when the confidence limits did not overlap. Coherence was lower for multiple frequencies throughout hypoxia and following 10 min of NR, but there were limited detectable differences for gain. EVS – electrical vestibular stimulation, AP – anteroposterior.

Fig. 8

Coherence (left column), difference of coherence (inset; DoC), and gain (right column) estimates were calculated from pooled data from males with eyes open (EO) comparing Baseline (BL) to Hypoxia 1 (H1; A), Hypoxia 2 (H2; B), Hypoxia 3 (H3; C), and Normoxic recovery (NR; D) on the hypoxia testing day. The black line represents H1 while the grey line depicts timepoints H1-3 and NR in each respective panel and dashed lines represent 95% confidence intervals for coherence and DoC. For the DoC estimates between BL and H1-NR, the light grey shading represents greater coherence for BL compared to H1-NR, whereas dark grey denotes the opposite. The dark and light grey shaded areas represent point-wise 95% confidence limits for the BL and H1-NR, respectively. Timepoints were considered statistically different from each other when the confidence limits did not overlap. Gain estimates were smaller for multiple lower frequencies following 10 min of NR. EVS – electrical vestibular stimulation, AP – anteroposterior.

At lower frequencies, EVS–AP force gain was greater for H1 and H2 during EC (1–1.5 and 2–3.5 Hz; Fig. 5A, 1 Hz; Fig. 5B), yet lower at H3 and NR compared to BL for females (1-2 Hz; Fig. 5C and D). When females had their EO, EVS-AP force gain was lower at H1, H3, and NR compared to BL for multiple frequencies (≤7 Hz; Fig. 6A, C, and D). During EC, males exhibited lower EVS-AP force gain at H3 (1.5 Hz; Fig. 7C), but greater gain at NR compared to BL for some lower frequencies (1–1.5 Hz; Fig. 8B). For EO, males displayed lower EVS-AP force gain at NR (0.5–3 Hz; Fig. 8C), but greater values at H2 than BL (1 Hz; Fig. 8B) at lower frequencies.

4 Discussion

The present study explored the effect of normobaric hypoxia (F_IO₂ = 0.12) on the vestibular control of standing balance. Normobaric hypoxia (H1-3) led to expected reductions in SpO₂, P_ETO₂ and P_ETCO₂ during both visual conditions, yet P_ETCO₂ remained lower into NR compared to BL for EC. During normobaric hypoxia, CoP displacement variability in the AP direction increased at H1 and H2 compared to BL, but individuals seemed to adapt to the hypoxic stimulus as CoP AP SD at H3 was not different from BL and no other CoP parameters were altered. Whole-body vestibular-evoked balance responses within the time domain (i.e., EVS-AP force peak-to-peak amplitude) were reduced by 55 min of hypoxia at H3 and remained blunted during NR compared to BL. Within the frequency domain, there were multiple frequencies, where females exhibited less coherence and gain values during both visual conditions by the final hypoxia timepoint (H3) compared to BL and this reduction persisted into NR; however, a similar result for coherence was only consistently observed for males with EC. Further, EVS-AP force gain was only reduced at H3 for males in the EC condition and NR in the EO condition compared to BL. Taken together, our findings indicate that by ∼55 min of normobaric hypoxia, the central nervous system may decrease its sensitivity to or transform vestibular-driven signals for quiet standing balance, which persists up to 10 min of normoxic recovery.

4.1 Acute normobaric hypoxia reduces the vestibular control of balance

In the current study, while normobaric hypoxia had minimal effects on CoP variables, the whole-body vestibular-evoked balance response, as reflected by EVS-AP force peak-to-peak amplitude, was reduced at 55 min of acute normobaric hypoxia (Fig. 4). It is unclear if the hypoxia-related decrease in the vestibular-evoked balance response is a result of altered primary vestibular afferent function, decreased central sensitivity to vestibular-driven cues, or both. However, owing to the nature of the EVS technique [31, 47], it is unlikely that the vestibular-end organs play a role in the hypoxia-related decreases observed here. One explanation for the reduction in EVS-AP force peak-to-peak amplitude could be a consequence of habituation of the vestibulomotor system to multiple EVS exposures as reported previously [6, 7]. However, during our control condition (i.e., normoxia testing session) there were no detectable differences over time when comparing BL to a range of time points up to 70 min (Fig. 1D). Therefore, it is unlikely that habituation was responsible for the decrease in the vestibular-evoked whole-body balance response observed at 55 min of acute hypoxia.

Another explanation for the hypoxia-related reduction in vestibular-evoked balance responses is sensory reweighting [44 , 58]. In the present study, the reduction in the vestibular-evoked balance response at H3 (Fig. 4) may reflect an impairment in the processing of vestibular information within the central nervous system. Owing to alterations that may be occurring within vestibular-related pathways during hypoxia [8 , 73], the central nervous system may be less sensitive to, or have greater difficulty transforming vestibular-driven signals to maintain quiet standing balance compared to normoxia. Therefore, the central nervous system may shift to a greater reliance on other sensory inputs (e.g., proprioception) to maintain upright postural control.

Overall, the EVS-AP force coherence was lower by 55 min at multiple frequencies for females compared with BL; yet there were a wider range of frequencies affected for EC in males with limited changes in EO (Figs. 5–8C). It was also found that in both the EC and EO conditions, females had lower EVS-AP gain amplitudes at H3 compared to BL at multiple low frequencies (0.5–7 Hz; Figs. 5, 6C), yet the effects of hypoxia on EVS-AP gain for males were limited to the EC condition at ∼3 Hz. Thus, females may have experienced more consistent impairments than males in the vestibular control of balance later into normobaric hypoxia (∼55 min). A possible reason why females were more consistently affected than males, may relate to the larger reduction in P_ETCO₂ in females by 55 min of normobaric hypoxia (Fig. 2C), which corroborates reports of a lower P_ETCO₂ [1, 71] and alveolar PCO₂ [29 , 38] in females. Therefore, reductions in P_ETCO₂ in addition to reduced P_ETO₂ during normobaric hypoxia, may also contribute to vestibular function alterations observed here.

4.2 The vestibular control of balance remained blunted during short-term normoxic recovery

Indeed, normobaric hypoxia reduced EVS-AP force peak-to-peak amplitude after ∼55 min of exposure, yet this blunted response persisted into NR for at least 10 min (Fig. 4). Additionally, during NR the EVS-AP force coherence for both sexes was less than BL at multiple frequencies (Figs. 5–8D). The same was true for EVS-AP force gain in both visual conditions for females, and males in the EO condition (Figs. 5–8D). Sympathetic nerve activity remains elevated following both acute intermittent [18 , 68] and continuous hypoxia [75] when returning to normoxia for 5–180 min. Further, vestibular-driven signals can modulate systemic blood pressure via vestibulo-sympathetic responses in animals [77, 78] and humans [11 , 70]. Thus, the prolonged reduction in vestibular-evoked balance responses during NR (Figs. 4–8) could be driven by changes in sympathetic activity and may be linked to reduced P_ETCO₂ in NR (Fig. 2C).

In the present study, P_ETO₂ returned to BL values during NR, but P_ETCO₂ remained reduced (Fig. 2C). Therefore, extended reductions in the vestibular control of balance during NR may be mediated by reductions in the partial pressure of CO₂. Currently there is limited data on the effects of hypocapnia and vestibular function, yet insights from hypercapnia studies indicate potential connections between CO₂ and the vestibular control of balance. For example, lesions to the medial region of the vestibular nucleus in rats attenuated cardiorespiratory responses during hypercapnia (0.10 CO₂) but not hypoxia (0.10 O₂) [39]. Similarly, rats with inner ear injuries (damage to utricle and hair cells) exhibit lower respiratory responses when exposed to 0.08 CO₂ compared to controls [3]. As multiple cardiorespiratory responses are altered during hypoxia, future research should take into consideration changes in both P_ETO₂ and P_ETCO₂ to determine which variables are driving alterations in vestibular function and upright posture during hypoxia. Nevertheless, carbon dioxide seems to play a key role in modulating the vestibular control of balance.

4.3 The presence of static visual cues exacerbates the effect of acute hypoxia on the vestibular control of balance

This study demonstrated that hypoxia reduced EVS-AP gain across a wide range of frequencies in the EO condition at H3 (0.5–1.5, 2, and 3.5–7.5 Hz; Fig. 6C) for females and NR (Females: 0.5–1 and 4–6.5 Hz; Males: 0.5–2.5 Hz, Figs. 6and 8D , respectively) compared to BL. Conversely, there was a smaller band of frequencies in which EVS-AP force gain was reduced at H3 and NR compared to BL in the EC condition (Females: BL > H3 at 0.5–2 Hz and BL > NR at 1-2 Hz, Fig. 5C and D; Males: BL > H3 at 1.5 Hz, Fig. 7C). Therefore, within the frequency domain, hypoxia had a stronger effect on the vestibular control of balance when visual cues were present compared to occluded, supporting previous studies investigating quiet standing balance (For review: [24]). Hypoxia can induce blurred vision [27], impair visual acuity and contrast sensitivity [10 , 59], as well as increase visual thresholds of cones and rods [28]. Vestibular function (spontaneous and positional nystagmus) is also impaired at high altitude [65]. In addition, vision decreases the amplitude of vestibular-evoked balance responses at low frequencies [74] and increases variability in the transformation of the direction of the vestibular control of balance compared to vision being absent [52]. During hypoxia, the central nervous system may be less sensitive to vestibular-driven signals for balance control than normoxia and this effect may be exacerbated when another sensory signal is disrupted (e.g., vision). Because hypoxia is likely creating multiple noisy signals (e.g. vision, vestibular, and somatosensory) [24], the central nervous system may have difficulty integrating and transforming these less reliable signals for motor control. Further, the integration sites for standing balance may also be altered with hypoxia. For example, Purkinje neurons within the cerebellum of rat’s experience excitotoxicity in the presence of hypoxia [8]. As Purkinje neurons are responsible for relaying signals to downstream cerebellar and vestibular nuclei [69], which are important for standing balance, their impairment in hypoxia may interfere with sensorimotor integration. Therefore, as emphasized by the current findings, the vestibular control of standing balance may be blunted within a hypoxic environment to a greater extent in the presence of vision than with vision occluded.

4.4 Conclusion

The current study demonstrated ∼55 min of normobaric hypoxia (F_IO₂ = 0.12) blunted the vestibular control of standing balance as indicated by reductions in EVS-AP force peak-to-peak amplitude and coherence, which remain disrupted following ∼10 min of NR. It appears vestibular function during normobaric hypoxia may be influenced more when static visual cues are present than when vision is occluded (Figs. 6 and 8). Further, the effects of normobaric hypoxia are more consistent in females than males (Figs. 5–8C). Future research is necessary to determine if hypoxia-related impairments in standing balance are owing to alterations in other sensorimotor signals (vision, cutaneous, etc.), in addition to vestibular.

References

Aitken

M.L.

, Franklin

J.L.

, Pierson

D.J.

, Schoene

R.B.

Influence of body size and gender on control of ventilation, J Appl Physiol 60 (1986), 1894–1899.

Ali

A.S.

, Rowen

K.A.

, Iles

J.F.

Vestibular actions on back and lower limb muscles during postural tasks in man, J Physiol 546 (2003), 615–624.

Allen

, Juric-Sekhar

, Campbell

, Mussar

K.E.

, Seidel

, Tan

, Zyphur

, Villagracia

, Stephanian

, Koch

, Ramirez

J.M.

, Rubens

D.D.

Inner ear insult suppresses the respiratory response to carbon dioxide, Neuroscience 175 (2011), 262–272.

Amjad

A.M.

, Halliday

D.M.

, Rosenberg

J.R.

, Conway

B.A.

An extended difference of coherence test for comparing and combining several independent coherence estimates: Theory and application to the study of motor units and physiological tremor, J Neurosci Methods 73 (1997), 69–79.

Baldessera

, Cavallari

, Tassone

Effects of transmastoid electrical stimulation on the triceps brachii EMG in man, NeuroReport 1 (1990), 191–193.

Balter

S.G.T.

, Stokroos

R.J.

, Akkermans

, Kingma

Habituation to galvanic vestibular stimulation for analysis of postural control abilities in gymnasts, Neurosci Lett 366 (2004), 71–75.

Balter

S.G.T.

, Stokroos

R.J.

, Eterman

R.M.A.

, Paredis

S.A.B.

, Orbons

, Kingma

Habituation to galvanic vestibular stimulation, Acta Otolaryngol (Stockh) 124 (2004), 941–945.

Barenberg

, Strahlendorf

Hypoxia induces an excitotoxic-type of dark cell degeneration in cerebellar Purkinje neurons, Neurosci Res 40 (2001), 245–254.

Baumgartner

, Eichenberger

, Bärtsch

Postural ataxia at high altitude is not related to mild to moderate acute mountain sickness, Eur J Appl Physiol 86 (2002), 322–326.

10.

Benedek

, Kéri

, Grósz

, Tótka

, Tóth

, Benedek

Short-term hypobaric hypoxia enhances visual contrast sensitivity, Neuroreport 13 (2002), 1063–1066.

11.

Bent

L.R.

, Bolton

P.S.

, Macefield

V.G.

Modulation of muscle sympathetic bursts by sinusoidal galvanic vestibular stimulation in human subjects, Exp Brain Res 174 (2006), 701–711.

12.

Britton

T.C.

, Day

B.L.

, Brown

, Rothwell

J.C.

, Thompson

P.D.

, Marsden

C.D.

Postural electromyographic responses in the arm and leg following galvanic vestibular stimulation in man, Exp Brain Res 94 (1993), 143–151.

13.

Bruyneel

A.-V.

, Humbert

, Bertrand

Comparison of balance strategies in mountain climbers during real altitude exposure between 1.500m and 3.200 m: Effects of age and expertise, Neurosci Lett 657 (2017), 16–21.

14.

Chen

P.-S.

, Chiu

W.-T.

, Hsu

P.-L.

, Lin

S.-C.

, Peng

I.-C.

, Wang

C.-Y.

, Tsai

S.-J.

Pathophysiological implications of hypoxia in human diseases, J Biomed Sci 27 (2020), 63.

15.

Choi

Glutamate neurotoxicity and diseases of the nervous system, Neuron 1 (1988), 623–634.

16.

Clarke

S.B.

, Deighton

, Newman

, Nicholson

, Gallagher

, Boos

C.J.

, Mellor

, Woods

D.R.

, O’Hara

J.P.

Changes in balance and joint position sense during a 12-day high altitude trek: The British Services Dhaulagiri medical research expedition, PLOS ONE 13 (2018), e0190919.

17.

Coats

A.C.

, Stoltz

M.S.

The recorded body-sway response to galvanic stimulation of the labyrinth: A preliminary study, The Laryngoscope 79 (1969), 85–103.

18.

Cutler

M.J.

, Swift

N.M.

, Keller

D.M.

, Wasmund

W.L.

, Smith

M.L.

Hypoxia-mediated prolonged elevation of sympathetic nerve activity after periods of intermittent hypoxic apnea, J Appl Physiol 96 (2004), 754–761.

19.

Dakin

C.J.

, Luu

B.L.

, van den Doel

, Inglis

J.T.

, Blouin

J.-S.

Frequency-specific modulation of vestibular-evoked sway responses in humans, J Neurophysiol 103 (2010), 1048–1056.

20.

Dakin

C.J.

, Son

G.M.L.

, Inglis

J.T.

, Blouin

J.-S.

Frequency response of human vestibular reflexes characterized by stochastic stimuli: Frequency response of human vestibular reflexes, J Physiol 583 (2007), 1117–1127.

21.

Dalton

B.H.

, Blouin

J.-S.

, Allen

M.D.

, Rice

C.L.

, Inglis

J.T.

The altered vestibular-evoked myogenic and whole-body postural responses in old men during standing, Exp Gerontol 60 (2014), 120–128.

22.

Day

B.L.

, Fitzpatrick

R.C.

Virtual head rotation reveals a process of route reconstruction from human vestibular signals: Perceiving self motion, J Physiol 567 (2005), 591–597.

23.

Day

B.L.

, Séverac Cauquil

, Bartolomei

, Pastor

M.A

, Lyon

I.N.

Human body-segment tilts induced by galvanic stimulation: A vestibularly driven balance protection mechanism, J Physiol 500 (1997), 661–672.

24.

Debenham

M.I.B.

, Smuin

J.N.

, Grantham

T.D.A.

, Ainslie

P.N.

, Dalton

B.H.

Hypoxia and standing balance, Eur J Appl Physiol 121 (2021), 993–1008.

25.

Degache

, Larghi

, Faiss

, Deriaz

, Millet

Hypobaric versus Normobaric Hypoxia: Same Effects on Postural Stability? , High Alt Med Biol 13 (2012), 40–45.

26.

Degache

, Serain

É.

, Roy

, Faiss

, Millet

G.P.

The fatigue-induced alteration in postural control is larger in hypobaric than in normobaric hypoxia, Sci Rep 10 (2020).

27.

DeHart

R.L.

Fundamentals of aerospace medicine, 2nd ed., Wolters Kluwer; 2 edition 1996.

28.

Ernest

, Krill

A.E.

The effect of hypoxia on visual function, Invest Ophthalmol 10 (1971), 323–328.

29.

FitzGerald

M.P.

The changes in the breathing and the blood at various high altitudes, Phil Trans Roy Soc London 203 (1913), 351–371.

30.

FitzGerald

M.P.

Further observations on the changes in the breathing and the blood at various high altitudes, Proc R Soc Med 88 (1914), 248–258.

31.

Fitzpatrick

R.C.

, Day

B.L.

Probing the human vestibular system with galvanic stimulation, J Appl Physiol 96 (2004), 2301–2316.

32.

Forbes

P.A.

, Dakin

C.J.

, Geers

A.M.

, Vlaar

M.P.

, Happee

, Siegmund

G.P.

, Schouten

A.C.

, Blouin

J.-S.

Electrical vestibular stimuli to enhance vestibulo-motor output and improve subject comfort, PLoS ONE 9 (2014), e84385.

33.

Fraser

W.D.

, Eastman

D.E.

, Paul

M.A.

, Porlier

J.A.G.

Decrement in postural control during mild hypobaric hypoxia, Aviat Space Environ Med 58 (1987), 768–772.

34.

Gekeler

, Schatz

, Fischer

M.D.

, Schommer

, Boden

, Bartz-Schmidt

K.U.

, Gekeler

, Willmann

Decreased contrast sensitivity at high altitude, Br J Ophthalmol (2019)..

35.

Grewal

, James

, Macefield

V.G.

Frequency-dependent modulation of muscle sympathetic nerve activity by sinusoidal galvanic vestibular stimulation in human subjects, Exp Brain Res 197 (2009), 379–386.

36.

Grocott

, Montgomery

, Vercueil

High-altitude physiology and pathophysiology: Implications and relevance for intensive care medicine, Crit Care 11 (2007), 203.

37.

Halliday

D.M.

, Rosenberg

J.R.

, Amjad

A.M.

, Breeze

, Conway

B.A.

, Farmer

S.F.

A framework for the analysis of mixed time series/point process data-Theory and application to the study of physiological tremor, single motor unit discharges and electromyograms, Prog Biophys Mol Biol 64 (1995), 237–278.

38.

Hannon

J.P.

Comparative altitude apaptability of young men and women, in: Environ Stress, Elsevier, 1978, pp. 335–350.

39.

Hernandez

J.P.

, Xu

, Frazier

D.T.

Medial vestibular nucleus mediates the cardiorespiratory responses to fastigial nuclear activation and hypercapnia, J Appl Physiol 97 (2004), 835–842.

40.

Hoshikawa

, Hashimoto

, Kawahara

, Ide

Postural instability at a simulated altitude of 5,000m before and after an expedition to Mt. Cho-Oyu (8,201m), Eur J Appl Physiol 110 (2010), 539–547.

41.

Inoue

, Yamanaka

, Okamoto

, Hosoi

Effect of a glutamate blocker, ipenoxazone hydrochloride on the hypoxia-induced firing in the medial vestibular nucleus, Acta Otolaryngol (Stockh) 124 (2004), 58–60.

42.

Johnson

B.G.

, Wright

A.D.

, Beazley

M.F.

, Harvey

T.C.

, Hillenbrand

, Imray

C.H.E.

The sharpened romberg test for assessing ataxia in mild acute mountainsickness✩, Wilderness Environ Med 16 (2005), 62–66.

43.

Khosravi-Hashemi

, Forbes

P.A.

, Dakin

C.J.

, Blouin

Virtual signals of head rotation induce gravity-dependent inferences of linear acceleration, J Physiol 597 (2019), 5231–5246.

44.

Kiemel

, Oie

K.S.

, Jeka

J.J.

Multisensory fusion and the stochastic structure of postural sway, Biol Cybern 87 (2002), 262–277.

45.

van der Kooij

, Jacobs

, Koopman

, van der Helm

An adaptive model of sensory integration in a dynamic environment applied to human stance control, Biol Cybern 84 (2001), 103–115.

46.

Krusche

, Jendrusch

, Platen

Short- and middle-term high-altitude exposure does not affect visual acuity and contrast sensitivity of healthy young people, J Sci Med Sport 22 (2019), S12–S1616.

47.

Kwan

, Forbes

P.A.

, Mitchell

D.E.

, Blouin

J.-S.

, Cullen

K.E.

Neural substrates, dynamics and thresholds of galvanic vestibular stimulation in the behaving primate, Nat Commun 10 (2019), 1–15.

48.

Leibowitz

H.W.

, Rodemer

C.S.

, Dichgans

The independence of dynamic spatial orientation from luminance and refractive error, Percept Psychophys 25 (1979), 75–79.

49.

Leuenberger

U.A.

, Hogeman

C.S.

, Quraishi

, Linton-Frazier

, Gray

K.S.

Short-term intermittent hypoxia enhances sympathetic responses to continuous hypoxia in humans, J Appl Physiol 103 (2007), 835–842.

50.

Lund

, Broberg

Effects of different head positions on postural sway in man induced by a reproducible vestibular error signal, Acta Physiol Scand 117 (1983), 307–309.

51.

Luu

B.L.

, Inglis

J.T.

, Huryn

T.P.

, Van der Loos

H.F.M.

, Croft

E.A.

, Blouin

J.-S.

Human standing is modified by an unconscious integration of congruent sensory and motor signals: Vestibular-motor pathways in standing, J Physiol 590 (2012), 5783–5794.

52.

Mackenzie

S.W.

, Reynolds

R.F.

Differential effects of vision upon the accuracy and precision of vestibular-evoked balance responses: Accuracy and precision of vestibular-motor transformations, J Physiol 596 (2018), 2173–2184.

53.

McGeehan

M.A.

, Woollacott

M.H.

, Dalton

B.H.

Vestibular control of standing balance is enhanced with increased cognitive load, Exp Brain Res 235 (2017), 1031–1040.

54.

Mian

O.S.

, Day

B.L.

Determining the direction of vestibular-evoked balance responses using stochastic vestibular stimulation: Vestibular-evoked balance response directions, J Physiol 587 (2009), 2869–2873.

55.

Nashner

, Berthoz

Visual contribution to rapid motor responses during postural control, Brain Res 150 (1978), 403–407.

56.

Nashner

L.M.

, Wolfson

Influence of head position and proprioceptive cues on short latency postural reflexes evoked by galvanic stimulation of the human labyrinth, Brain Res 67 (1974), 255–268.

57.

Nordahl

S.H.

, Aasen

, Risberg

, Owe

J.O.

, Molvær

O.I.

Postural control and venous gas bubble formation during hypobaric exposure, Aviat Space Environ Med 73 (2002), 184–190.

58.

Oie

K.S.

, Kiemel

, Jeka

J.J.

Multisensory fusion: Simultaneous re-weighting of vision and touch for the control of human posture, Cogn Brain Res 14 (2002), 164–176.

59.

Paulus

W.M.

, Straube

, Brandt

Th.

Visual stabilization of posture: Physiological stimulus characteristics and clinical aspects, Brain 107 (1984), 1143–1163.

60.

Peters

R.M.

, Blouin

J.-S.

, Dalton

B.H.

, Inglis

J.T.

Older adults demonstrate superior vestibular perception for virtual rotations, Exp Gerontol 82 (2016), 50–57.

61.

Rasman

B.G.

, Forbes

P.A.

, Tisserand

, Blouin

J.-S.

Sensorimotor manipulations of the balance control loop–beyond imposed external perturbations, Front Neurol 9 (2018), 00899.

62.

Rosenberg

J.R.

, Amjad

A.M.

, Breeze

, Brillinger

D.R.

, Halliday

D.M.

The Fourier approach to the identification of functional coupling between neuronal spike trains, Prog Biophys Mol Biol 53 (1989), 1–31.

63.

Rothman

S.M.

, Olney

J.W.

Glutamate and the pathophysiology of hypoxic-ischemic brain damage, Ann Neurol 19 (1986), 105–111.

64.

Sarabon

, Mekjavić

I.B.

, Eiken

, Babič

The effect of bed rest and hypoxic environment on postural balance and trunk automatic (Re)actions in young healthy males, Front Physiol 9 (2018).

65.

Singh

, Kochhar

R.C.

, Kacker

S.K.

Effects of high altitude on inner ear functions, J Laryngol Otol 90 (1976), 1113–1120.

66.

Smith

C.P.

, Allsop

J.E.

, Mistry

, Reynolds

R.F.

Co-ordination of the upper and lower limbs for vestibular control of balance: Upper limb control for balance, J Physiol 595 (2017), 6771–6782.

67.

Stadelmann

, Latshang

T.D.

, Lo Cascio

C.M.

, Clark

R.A.

, Huber

, Kohler

, Achermann

, Bloch

K.E.

Impaired postural control in healthy men at moderate altitude M and M): Data from a randomized trial, PLOS ONE 10 (2015), e0116695.

68.

Stuckless

T.J.R.

, Vermeulen

T.D.

, Brown

C.V.

, Boulet

L.M.

, Shafer

B.M.

, Wakeham

D.J.

, Steinback

C.D.

, Ayas

N.T.

, Floras

J.S.

, Foster

G.E.

Acute intermittent hypercapnic hypoxia and sympathetic neurovascular transduction in men, J Physiol 598 (2020), 473–487.

69.

Tsubota

, Ohashi

, Tamura

Optogenetics in the cerebellum: Purkinje cell-specific approaches for understanding local cerebellar functions, Behav Brain Res 255 (2013), 26–34.

70.

Voustianiouk

, Kaufmann

, Diedrich

, Raphan

, Biaggioni

, MacDougall

, Ogorodnikov

, Cohen

Electrical activation of the human vestibulo-sympathetic reflex, Exp Brain Res 171 (2006), 251–261.

71.

Wadhwa

, Gradinaru

, Gates

G.J.

, Badr

M.S.

, Mateika

J.H.

Impact of intermittent hypoxia on long-term facilitation of minute ventilation and heart rate variability in men and women: Do sex differences exist? , J Appl Physiol 104 (2008), 1625–1633.

72.

Wagner

D.R.

, Saunders

, Robertson

, Davis

J.E.

Normobaric hypoxia effects on balance measured by computerized dynamic posturography, High Alt Med Biol 17 (2016), 222–227.

73.

Wagner

L.S.

, Oakley

S.R.

, Vang

, Noble

B.N.

, Cevette

M.J.

, Stepanek

J.P.

Hypoxia-induced changes in standing balance, Aviat Space Environ Med 82 (2011), 518–522.

74.

Wallace

J.W.

, Rasman

B.G.

, Dalton

B.H.

Vestibular-evoked responses indicate a functional role for intrinsic foot muscles during standing balance, Neuroscience 377 (2018), 150–160.

75.

Xie

, Skatrud

J.B.

, Puleo

D.S.

, Morgan

B.J.

Exposure to hypoxia produces long-lasting sympathetic activation in humans, J Appl Physiol 91 (2001), 1555–1562.

76.

Xie

, Zhang

, Pan

, Wu

, Chen

, Wang

, Liu

, Qian

, Liu

L.-J.

Decreased calcium-activated potassium channels by hypoxia causes abnormal firing in the spontaneous firing medial vestibular nuclei neurons, Eur Arch Otorhinolaryngol 272 (2015), 2703–2711.

77.

Yates

B.J.

Vestibular influences on the autonomic nervous system, Ann N Y Acad Sci 781 (1996), 458–473.

78.

Yates

B.J.

, Miller

A.D.

Physiological evidence that the vestibular system participates in autonomic and respiratory control, J Vestib Res 8 (1998), 17–25.

Normoxia testing session
Variable	Sex, Visual condition	BL1	H1	H2	H3	NR
SpO₂ (%)	Females, Eyes closed	96.8±1.5†	96.7±1.7†	97.2±1.5†	97.4±1.7†	97.4±1.6†
	Females, Eyes open	96.8±1.6†	96.7±1.8†	97.1±2.0†	97.0±2.3†	97.4±1.4†
	Males, Eyes closed	95.3±1.2	95.1±1.1	95.2±1.0	95.4±1.7	95.6±1.2
	Males, Eyes open	95.5±1.1	95.0±1.2	95.2±1.1	95.6±1.3	95.3±1.3
P_ETO₂ (mmHg)	Females, Eyes closed	98.6±5.8*	100.7±6.8*	100.0±6.7*	100.9±6.1*	99.2±4.4*
	Females, Eyes open	101.1±7.7	100.4±7.4	100.6±9.1	100.1±5.7	100.6±5.8
	Males, Eyes closed	95.1±3.2*	95.4±3.1*	95.4±3.5*	94.9±2.8*	95.2±2.8*
	Males, Eyes open	96.6±3.9	96.6±3.2	96.6±3.5	97.1±3.6	97.1±4.0
P_ETCO₂ (mmHg)	Females, Eyes closed	32.5±2.8*†	31.6±3.7*†	31.8±3.0*†	31.3±3.6*†	31.9±2.4*†
	Females, Eyes open	31.3±3.3†	31.3±3.8†	31.1±3.4†	31.5±3.1†	31.4±3.0†
	Males, Eyes closed	35.8±2.2*	35.3±2.2*	35.5±2.5*	35.5±2.6*	35.4±2.6*
	Males, Eyes open	35.2±2.7	34.6±2.2	34.9±2.4	34.7±3.0	34.7±3.0