The impact of mild hypoxia on postural control

@article{fujii2026impact,
  title = {{The impact of mild hypoxia on postural control}},
  author = {Manami Fujii and Daniel M. Merfeld},
  journal = {Communications in Kinesiology},
  year = {2026},
  month = nov,
  doi = {10.51224/cik.2026.8543},
  url = {https://doi.org/10.51224/cik.2026.8543},
  note = {CC BY},
}

TY  - JOUR
TI  - The impact of mild hypoxia on postural control
AU  - Manami Fujii
AU  - Daniel M. Merfeld
PY  - 2026
DA  - 2026/11/03
JO  - Communications in Kinesiology
DO  - 10.51224/cik.2026.8543
UR  - https://doi.org/10.51224/cik.2026.8543
AB  - Introduction: Over 200 million people live at altitudes exceeding 2500 m above sea level and commercial aircraft cabins are routinely pressurized to equivalent altitudes of 1800-2500 m (Debenham et al., 2021). It is important to understand how reduced blood oxygenation impairs standing balance control for everyday functional safety. Increased postural sway on a firm surface with eyes open was reported at 3619 m and 5140 m, indicating an impact of hypobaric hypoxia on standing balance (Clarke et al., 2018). Additionally, over 30 minutes of normobaric hypoxia (11 % O2) exposure significantly reduces anterior-posterior postural sway in response to electrical vestibular stimulation with eyes closed during quiet standing (Debenham et al., 2023), which suggests that hypoxia alters vestibular contributions to postural control. However, it is still unknown whether mild hypoxia - defined as an ~5 % reduction in peripheral oxygen saturation (SpO₂) - impacts postural control in response to platform perturbations and quiet stance with eyes closed on foam and firm surfaces. To evaluate impacts of mild hypoxia on postural control, we introduced six-dimensional (6D) platform continuous perturbations to quantify frequency response functions (FRFs) (i.e., normalized postural responses by the platform stimulus) (Peterka, 2002). We hypothesize that mild hypoxia will: (1) increase FRF magnitude during 6D platform perturbations and (2) increase postural sway on both firm and foam surfaces with eyes closed. Methods: As an exploratory test, six healthy adults (aged 26-67 years) underwent continuous pseudorandom platform perturbations that were simultaneously provided in six dimensions – anterior-posterior (AP), medial-lateral (ML), earth-vertical (EV) translations, right anterior-left posterior (RALP), left anterior-right posterior (LARP), and yaw rotations - using a 6DoF motion platform (Charles River Analytics) for two days – one under normoxia (21 % O2) and one under normobaric hypoxia (16 % O2) in a chamber (Hypoxico). A separate training session under normoxia was completed prior to testing to minimize learning effects. Order of the test conditions were blinded to the subject and counterbalanced across the subjects. Each day consisted of an exposure block during 1 hour of gas inhalation, followed 10 minutes post-exposure by a recovery block. The exposure block comprised three identical, back-to-back testing cycles (Trials 1–3). Each testing cycle consisted of a 5-minute perturbation trial followed by two 1-minute quiet stance trials (one firm, one foam surface) with eyes closed. The recovery block repeated this exact 5-minute perturbation and dual quiet stance sequence once. Kinetic responses (3D forces and 3D torques) were sampled at 500 Hz via AMTI force plates that were built into the motion platform. Force plate data were then processed using a discrete-time Fourier transform (DFT) to derive FRFs by normalizing spectral responses to the stimulus at each of the discrete perturbation frequencies and the subject body weight. Additionally, root-mean square (RMS) distance and mean velocity of center of pressure (CoP) in AP and ML directions were quantified for two quiet stance trials with foam and firm surfaces. Results: Both AP and ML force FRFs increased at the three highest frequencies (1.3-2.0 Hz) during hypoxia compared to normoxia. However, a three-way repeated-measures MANOVA on the complex-valued FRFs revealed no statistically significant main effect of mild hypoxia across any dimension (Pillai’s Trace, AP: p = 0.227, ML: p = 0.348, EV: p = 0.227, RALP: p = 0.586, LARP: p = 0.405, and yaw: p = 0.833). Similarly, despite observable trends toward increased postural sway in the second and third quiet-stance trials on foam under hypoxia, a two-way repeated-measures ANOVA on CoP metrics (AP/ML RMS distance and mean velocity) indicated no significant effect of hypoxia on either firm (AP: p = 0.154, ML: p = 0.117) or foam surfaces (AP: p = 0.256, ML: p = 0.127). Notably, effect size calculations showed a moderate impact of hypoxia during quiet stance on foam with eyes closed (Cohen’s d = 0.65). A prospective power analysis suggests a sample size of N = 27 is required to achieve 81 % power to detect a medium effect of hypoxia (d = 0.5) using a one-sided paired t-test (alpha = 0.05). Conclusions: Although the impact of mild hypoxia on perturbed and quiet balance did not reach statistical significance, the moderate effect sizes suggest a practically meaningful effect of mild hypoxia on postural control that warrants further investigation. We will expand the cohort to clarify whether mild hypoxia alters postural responses to 6D platform perturbations and increases quiet-stance postural sway with eyes closed. This work will provide critical insights into how the central nervous system compensates for mild hypoxia-induced vestibular impairments, which ultimately ensures functional safety and prevents falls in low-oxygen environments at high altitudes.
SN  - 2767-0732
ER  -