2.1.

Subjects

Eleven healthy male subjects participated in this study. All subjects were nonsmokers and had no history of head injury or cardiorespiratory or neuromuscular diseases. Their average ($\pm \mathrm{SD}$) age, weight, and BMI were $29\pm 6$ years, $71\pm 6\mathrm{kg}$, and $22\pm 2\mathrm{kg}{\mathrm{m}}^{-2}$, respectively. The study was approved by the local ethics committee and performed according to the Declaration of Helsinki. All subjects gave their written informed consent to participate in the study and were instructed to drink no alcoholic or caffeinated beverages, and to avoid any physical activity for at least 12 h before each testing day.

2.2.

Experimental Design

The study protocol included two sessions (hypoxic and normoxic sessions) performed at the same time of the day $\pm 1\mathrm{h}$ and at least 1 week apart. Subjects seated quietly in a comfortable chair with legs down on the floor, while breathing a gas mixture delivered via a face-mask connected to a mixing gas chamber (Altitrainer 200®, SMTEC, Nyon, Switzerland). After $>15\min$ stabilization of the cardiorespiratory variables, baseline data were collected in normoxic condition [inspiratory ${\mathrm{O}}_2$ fraction (${\mathrm{FiO}}_2$) of the gas mixture: 0.21] during a 3-min period. Subjects were then exposed rapidly (within 2 min) to 4 h of normobaric hypoxia ($\mathrm{Fi}{\mathrm{O}}_2=0.12$, hypoxic session) or normoxia ($\mathrm{Fi}{\mathrm{O}}_2=0.21$, normoxic session) and then again to normoxia for 15 min. This time frame may be critical regarding the genesis of hypoxia-induced cerebral perturbations (e.g., AMS symptoms), and the reduced ${\mathrm{FiO}}_2$ used in the present study (equivalent to an altitude of $\sim 4000\mathrm{m}$) is known to induce values of $\mathrm{Sp}{\mathrm{O}}_2$ as frequently observed in clinical practice ($\sim 85$ to 90%). Subjects were blinded to the treatment (hypoxic or normoxic session and changes in ${\mathrm{FiO}}_2$ during each session). Subjects were asked to breathe normally and to refrain from sleeping, talking, or moving to avoid any interference with the normal course of physiological adaptations to the gas inhalation. Temperature and humidity inside the experimental room were maintained at $23\pm 1°\mathrm{C}$ and $43\pm 9\%$, respectively.

2.3.

Measurement of Systemic Oxygen Saturation/End-Tidal ${\mathrm{CO}}_2$

Systemic $\mathrm{Sp}{\mathrm{O}}_2$ was measured continuously using a pulse oximeter (Pulsox 300, Konica Minolta, Osaka, Japan) placed on the subject’s right index finger, and end-tidal $\mathrm{C}{\mathrm{O}}_2$ partial pressure ($\mathrm{EtC}{\mathrm{O}}_2$) was measured using a respiratory gas monitor (Ohmeda RGM, GE Healthcare, Little Chalfont, UK) connected to the face-mask.

2.4.

Measurement of Tissue Oxygenation

2.5.

Measurement of Heart Rate Variability and Mean Arterial Pressure

Full mean arterial blood pressure (MABP) and heart rate variability (HRV) evaluations (i.e., all stages for both sessions) were measured in 7 and 8 out of the 11 subjects, respectively.

2.6.

Dynamics of Tissue Oxygenation and Cardiorespiratory Responses

$\mathrm{Sp}{\mathrm{O}}_2$, $\mathrm{EtC}{\mathrm{O}}_2$, HR, MABP, and NIRS parameters were averaged over 3-min periods at the end of the initial normoxic period (BL), at $+10$, 20, 30, 40 and 50 min to study HE-induced adaptations in the first hour, and at BL, $+20$, 80, 140, 200, and 240 min and 15 min after return to normoxia ($\mathrm{N}+15$) to study HE-induced adaptation over the 4-h exposition period. Steady-state MABP and HRV parameters were collected over 10-min periods at BL, $+20$, 80, 140, 200, and 240 min in the 4-h exposition period and at $\mathrm{N}+15$. The same averaging was carried out in normoxic session to evaluate the possible effects of time independent of HE.

2.7.

Self-Reported AMS Questionnaires

To evaluate AMS symptoms, subjects completed self-reported questionnaires before each testing session and in the last 30 min of the 4-h gas exposition period. These evaluations comprised four sections of the Lake Louise AMS questionnaire (LLS, i.e., “headache,” “gastrointestinal distress,” “fatigue/weakness,” “dizzy/light-headedness”),³⁷ the neurologic part of the Environmental Symptom Questionnaire (ESQ-III AMS-C, 11 items graded from 0 to 5 covering “lightheadedness,” “faintness,” “weakness,” “nausea,” “I feel hung-over,” and so on),³⁸ and a visual analog scale for headache (VAS, score from 0 to 100 mm). AMS is defined as an LLS score $>3$ or an AMS-C score $\ge 0.7$.

2.8.

Statistical Analysis

Normality of distribution and homogeneity of variances of the main variables were confirmed using a Skewness-Kurtosis normality test and the Levene’s test, respectively. For each analysis (over the first hour and over the entire gas exposition), two-way ($\text{session}\times \text{time}$) analysis of variance (ANOVA) with repeated measures were performed for each dependent variable. Post hoc Fisher’s LSD-tests were applied to determine a difference between two mean values if the ANOVA revealed a significant main effect or interaction effect. Relationships among changes in tissue oxygenation parameters, cardiorespiratory function parameters, $\mathrm{Sp}{\mathrm{O}}_2$, and symptomatic data were determined by Pearson product correlations. For all statistical analyses, an alpha level of 0.05 was used as the cut-off for significance. Data are presented as $\text{mean}\pm \mathrm{SD}$ in text and tables and as $\text{mean}\pm \mathrm{SE}$ in figures.

3.1.

Adaptations Over the First Hour of HE

3.1.1.

Arterial oxygenation profile

Over the first hour, $\mathrm{Sp}{\mathrm{O}}_2$ was significantly reduced during the hypoxic session compared to BL with a plateau from $+20\min$, while no significant changes occurred in the normoxic session (interaction effect, ${\mathrm{F}}_{(5,50)}=81.2$, $P<0.001$) [Fig. 1(a)].

Fig. 1

Mean changes ($\pm \mathrm{SE}$) in arterial oxygen saturation ($\mathrm{Sp}{\mathrm{O}}_2$) and end-tidal $\mathrm{C}{\mathrm{O}}_2$ partial pressure ($\mathrm{EtC}{\mathrm{O}}_2$) during exposure to 12% ${\mathrm{O}}_2$ hypoxia (solid lines). Dashed lines identify values for the control normoxic session. Data are shown over the first hour [panels (a) and (b)] and over 4 h of hypoxic exposure [panels (c) and (d)]. BL, baseline; $+15\mathrm{N}$, 15 min after normoxic washout. *$P<0.05$ versus BL; ‡$P<0.05$ versus BL and $+20\min$; $$P<0.05$ versus normoxic session.

3.1.2.

Tissue oxygenation profiles

At the muscle site, the hypoxic session resulted in lower $\mathrm{\Delta }[{\mathrm{O}}_2\mathrm{Hb}]$ (interaction effect, ${\mathrm{F}}_{(5,50)}=5.3$, $P<0.001$) and higher $\mathrm{\Delta }[\mathrm{HHb}]$ (interaction effect, ${\mathrm{F}}_{(5,50)}=4.6$, $P=0.001$) from $+10\min$ compared with BL, whereas $\mathrm{\Delta }[\mathrm{tHb}]$ did not change over the first hour of gas exposition in both hypoxic and normoxic sessions [Figs. 2(a) to 2(c)]. The hypoxic session resulted in reduced PFC $\mathrm{\Delta }[{\mathrm{O}}_2\mathrm{Hb}]$ (interaction effect, ${\mathrm{F}}_{(5,50)}=15.5$, $P<0.001$) and increased $\mathrm{\Delta }[\mathrm{HHb}]$ (interaction effect, ${\mathrm{F}}_{(5,50)}=46.5$, $P<0.001$) from $+10\min$ compared with BL, while $\mathrm{\Delta }[\mathrm{tHb}]$ increased from $+20\min$ in both hypoxic and normoxic sessions (main effect of time, ${\mathrm{F}}_{(5,50)}=5.0$, $P<0.001$) [Figs. 2(d) to 2(f)]. The hypoxic session also resulted in lower MC $\mathrm{\Delta }[{\mathrm{O}}_2\mathrm{Hb}]$ (interaction effect, ${\mathrm{F}}_{(5,50)}=21.4$, $P<0.001$) and higher $\mathrm{\Delta }[\mathrm{HHb}]$ (interaction effect, ${\mathrm{F}}_{(5,50)}=31.4$, $P<0.001$) from $+10\min$ compared with BL, whereas $\mathrm{\Delta }[\mathrm{tHb}]$ slightly decreased at $+50\min$ in both sessions (main effect of time, ${\mathrm{F}}_{(5,50)}=3.8$, $P<0.01$) [Figs. 2(g) to 2(i)]. As shown in Fig. 2 [especially, Figs. 2(e) and 2(g)], 30 to 40 min was required to stabilize the time course of oxygenation changes for PFC and MC at their highest magnitude.

Fig. 2

Mean changes ($\pm \mathrm{SE}$) in oxy-([${\mathrm{O}}_2\mathrm{Hb}$]), deoxy-([HHb]), and total-hemoglobin ([tHb]), during the first hour of exposure to 12% ${\mathrm{O}}_2$ hypoxia (solid lines). Dashed lines identify values for the control normoxic session. Data are shown for the vastus lateralis muscle [panels (a) to (c)], for the prefrontal cortex [PFC, panels (d) to (f)], and for the motor cortex [MC, panels (g) to (i)]. BL, baseline. *$P<0.05$ versus BL; ‡$P<0.05$ versus BL and $+20\min$; $$P<0.05$ versus normoxic session.

In the hypoxic session, changes in muscle [HHb], [${\mathrm{O}}_2\mathrm{Hb}$], and [tHb] did not correlate with $\mathrm{Sp}{\mathrm{O}}_2$, whereas PFC and MC $\mathrm{\Delta }[\mathrm{HHb}]$ exhibited high inverse correlation levels with $\mathrm{Sp}{\mathrm{O}}_2$ (at $+50\min$: $R^2=-0.40$, $P<0.05$ and $R^2=-0.62$, $P<0.01$, respectively). Also, muscle [Hb] changes did not correlate with cerebral [Hb] changes over the first hour of HE. $\mathrm{\Delta }[\mathrm{HHb}]$ and $\mathrm{\Delta }[{\mathrm{O}}_2\mathrm{Hb}]$ were tightly correlated between PFC and MC ($R^2=0.55$, $P<0.01$ and $R^2=0.56$, $P<0.01$, respectively), but there was no significant correlation between cerebral sites for $\mathrm{\Delta }[\mathrm{THb}]$ ($P>0.05$).

3.2.

Adaptations Over 4-h HE

3.2.2.

Tissue oxygenation profiles

At the muscle site, the hypoxic session resulted in decreased $\mathrm{\Delta }[{\mathrm{O}}_2\mathrm{Hb}]$ from $+140\min$ compared to BL [Fig. 3(a), interaction effect, ${\mathrm{F}}_{(6,60)}=2.6$, $P<0.05$]. Increased $\mathrm{\Delta }[\mathrm{HHb}]$ compared to BL was observed until $+140\min$ during the hypoxic session, while a significant decrease in $\mathrm{\Delta }[\mathrm{HHb}]$ occurred in the normoxic session from $+80\min$ [Fig. 3(b), interaction effect, ${\mathrm{F}}_{(6,60)}=4.9$, $P<0.001$]. Significant reduction in muscle $\mathrm{\Delta }[\mathrm{THb}]$ was observed at $+200\min$ in both sessions [Fig. 3(c), main effect of time, ${\mathrm{F}}_{(6,60)}=6.0$, $P<0.001$]. During the hypoxic session, PFC (interaction effect, ${\mathrm{F}}_{(6,60)}=9.7$, $P<0.001$) and MC (interaction effect, ${\mathrm{F}}_{(6,60)}=9.3$, $P<0.001$) $\mathrm{\Delta }[{\mathrm{O}}_2\mathrm{Hb}]$ was lower compared to BL, while it increased significantly in the normoxic session from $+80\min$ in PFC and $+240\min$ in MC [Figs. 3(d) and 3(g)]. The hypoxic session induced a marked increased in $\mathrm{\Delta }[\mathrm{HHb}]$ compared to BL in both PFC (interaction effect, ${\mathrm{F}}_{(6,60)}=30.3$, $P<0.001$) and MC (interaction effect, ${\mathrm{F}}_{(6,60)}=27.4$, $P<0.001$) [Figs. 3(e) and 3(h)]. Both hypoxic and normoxic sessions resulted in significant cerebral $\mathrm{\Delta }[\mathrm{THb}]$ increases [Figs. 3(f) and 3(i)] from $+20\min$ in PFC (main effect of time, ${\mathrm{F}}_{(6,60)}=23.3$, $P<0.001$) but only from $+240\min$ in MC (main effect of time, ${\mathrm{F}}_{(6,60)}=5.5$, $P<0.001$).

Fig. 3

Mean changes ($\pm \mathrm{SE}$) in oxy-([${\mathrm{O}}_2\mathrm{Hb}$]), deoxy-([HHb]), and total-hemoglobin ([tHb]), over a 4-h exposure to 12% ${\mathrm{O}}_2$ hypoxia (plenty lines). Dashed lines identify values for the control normoxic session. Data are shown for the vastus lateralis muscle [panels (a) to (c)], for the PFC [panels (d) to (f)], and for the MC [panels (g) to (i)]. BL, baseline; $+15\mathrm{N}$, 15 min after normoxic washout. *$P<0.05$ versus BL; ‡$P<0.05$ versus BL and $+20\min$; $$P<0.05$ versus normoxic session.

Cerebral hemodynamics assessed by NIRS during the hypoxic session were strongly correlated between PFC and MC (at $+240\min$: $R^2=0.56$, $P<0.001$; $R^2=0.39$, $P<0.05$; $R^2=0.50$, $P<0.05$, for $\mathrm{\Delta }[\mathrm{HHb}]$; $\mathrm{\Delta }[{\mathrm{O}}_2\mathrm{Hb}]$; $\mathrm{\Delta }[\mathrm{THb}]$, respectively), with no significant correlations with muscle hemodynamics. At $+240\min$ of HE, PFC $\mathrm{\Delta }[\mathrm{HHb}]$ but not MC $\mathrm{\Delta }[\mathrm{HHb}]$ was negatively correlated with $\mathrm{Sp}{\mathrm{O}}_2$ ($R^2=-0.54$, $P=0.01$). As observed during the first hour, muscle hemodynamics did not correlate with $\mathrm{Sp}{\mathrm{O}}_2$ over the 4 h of HE.

4.1.

Decoupling between ${\mathrm{SpO}}_2$ and Muscle/Cerebral Oxygenation Over a 4-h Sustained HE

Concomitant cerebral oxygenation and $\mathrm{Sp}{\mathrm{O}}_2$ reduction have been observed in unacclimatized trekkers at moderate altitudes or during progressive ascent to high altitude.^39,40 While the magnitude of reduction in the systemic and local tissue oxygenation status has been described during hypoxic exercise loading,^41,42 their time courses under prolonged hypoxic resting conditions is poorly understood. In the present study, we observed a significant deoxygenation in muscle and cerebral tissues (PFC and MC) after 10 min of resting HE. While the reduction in $\mathrm{Sp}{\mathrm{O}}_2$ ($\sim 13\%$) plateaued after 20 min of HE, our findings indicate a time to plateau of 30 to 40 min for PFC and MC and of 10 min for the skeletal muscle.

Since cerebral oxygenation is influenced both by $\mathrm{Sp}{\mathrm{O}}_2$ and CBF, the difference between cerebral oxygenation and $\mathrm{Sp}{\mathrm{O}}_2$ patterns may be explained by the alteration in CBF. In an experimental setting similar to the present one, Nishimura et al.⁷ observed that blood velocity in the middle cerebral artery (MCAv) was unchanged after 1 h of HE ($\mathrm{Fi}{\mathrm{O}}_2=0.15$) and did not differ compared to normoxic conditions after 2 h, while $\mathrm{EtC}{\mathrm{O}}_2$ was reduced by 2 to 3 mmHg and MABP was unchanged. However, Huang et al.⁴³ provide evidence supporting that the rise in CBF over the first hours of poikilocapnic hypoxic exposure at high altitude is delayed. Using Doppler ultrasound, they found a small but nonsignificant increase in blood velocities in the internal carotid and vertebral arteries after 2 to 4 h at higher altitude (4300 m). Similarly, Teppema et al.⁸ showed unchanged CBF during 4 h of poikilocapnic hypoxia (${\mathrm{PetO}}_2=50\mathrm{mmHg}$). By contrast, 12 to 24 h after arrival at high altitudes (3475 to 4559 m), several TCD studies found increases of 20 to 27% in MCAv.^14,44,45 These ultrasound findings are in accordance with previous results using other techniques. Severinghaus and coauthors,⁴⁶ using the Kety–Schmidt nitrous oxide washout method to analyze changes in CBF, found a 24% increase at 6 to 12 h and a 13% increase at 3 to 5 days after arrival at 3810 m. Hence, the kinetic of CBF changes, particularly in the early hours, in response to prolonged HE may explain, at least in part, the longer time needed to reach stable deoxygenation level within cerebral tissue compared to $\mathrm{Sp}{\mathrm{O}}_2$.

It has been suggested that cerebral oxygenation is strongly dependent on arterial $\mathrm{C}{\mathrm{O}}_2$ partial pressure ($\mathrm{PaC}{\mathrm{O}}_2$) due to its effect on CBF.⁴⁷ It could be postulated that cerebral oxygenation patterns cannot plateau until ${\mathrm{EtCO}}_2$ has reached a given critical or stable value. The time-course of ${\mathrm{EtCO}}_2$ changes in the present study indicates that early modest hypocapnia does not explain the specific cerebral oxygenation profiles observed in response to HE since similar ${\mathrm{EtCO}}_2$ values were observed over the first hour of normoxia and hypoxia. Consistent with previous reports,^5,7 prolonged HE induced significant hypocapnia compared to the normoxic session. However, no relationship between ${\mathrm{EtCO}}_2$ and tissue oxygenation was found over time, suggesting that ${\mathrm{EtCO}}_2$ had no specific effect on cerebral oxygenation, even after several hours of HE. The heterogeneity of individual responses to poikilocapnic hypoxia may explain part of the variability in the cerebrovascular tone response, but it is likely that ${\mathrm{PaCO}}_2$ did not strongly influence cerebral (de)oxygenation in our study.

It is important to emphasize that NIRS measures the balance between oxygen supply and utilization. NIRS-derived deoxygenation, as observed in our study, refers to a mix between the pre- and postcapillaries state, thus including a decrease in arterial ${\mathrm{O}}_2$ content and also possible mechanisms of exacerbated oxygen extraction to preserve the resting cerebral metabolic rate of oxygen ($\mathrm{CMR}{\mathrm{O}}_2$). There are few field studies of brain oxygen consumption or extraction at altitude because such studies would require arterial and venous (jugular) measurements. Moller and coauthors⁴⁸ studied $\mathrm{CMR}{\mathrm{O}}_2$ and CBF using the Fick and the Kety–Schmidt techniques in nine acclimatized subjects at rest and during exercise at sea level and at 5260 m. Despite profound changes in breathing, no changes were seen in both CBF and oxidative metabolism. In the same way, the $\mathrm{CMR}{\mathrm{O}}_2$ was unchanged compared with sea-level values in five subjects investigated within the first 24 h after arrival at 3810 m by Severinghaus and coauthors.⁴⁶ These data suggest a critical adjustment of the oxygen extraction fraction in response to hypoxemia, especially in the early hours of HE when the decrease in $\mathrm{Ca}{\mathrm{O}}_2$ is not always and/or immediately compensated by an increase in the delivery of ${\mathrm{O}}_2$ (through an increase in CBF). This remains however speculative, as human brain models specifying the physiologic control of oxygen delivery at capillary level (other than hemoglobin- and flow-dependent properties of oxygen transport) are scarce.⁴⁹

The fact that prolonged HE elicits measurable time delay in changes between $\mathrm{Sp}{\mathrm{O}}_2$ and tissue (especially cerebral) oxygenation may be of practical importance when drawing protocols aiming at investigating the effect of hypoxic gas inhalation on tissue responses. A hypoxic wash-in period of 30 min could be recommended in order to reach maximal cerebral tissue deoxygenation levels.

4.2.

Tissue-Specific Sensitivity to Sustained HE

It has been suggested that oxygen delivery to tissues is tightly matched by immediate cardiac output changes to meet the peripheral demand in normal subjects at rest.⁵⁰ Some studies^40,51 reported long-term adaptations in regional cerebral oxygenation following trekking at high altitudes. Our results show time-dependent responses and distinct kinetics (amplitude and rate) of cerebral and muscle deoxygenation when $\mathrm{Sp}{\mathrm{O}}_2$ is reduced over hours. Following HE onset, muscle [HHb] increased to significance after a delay of 10 min and remained constant over the first hour, whereas cerebral [HHb] (PFC and MC) increased progressively to reach a plateau only after 30 to 40 min. Interestingly, cerebral and muscle tissues respond differently to the offset of gas exposure during the first 15 min of normoxic washout: [HHb] returned promptly to baseline values in cerebral tissue, but not in the muscle tissue (see Fig. 3).

The literature provides inconsistent results regarding changes in oxygenation status within the skeletal muscle during few minutes of HE, with either no change^41,52,53 or a rapid impact on tissue oxygenation state.⁵⁴ It is likely that these results are mainly explained by a time-dependent relationship, the level of hypoxia, and a high interindividual variability in muscle responses (especially compared to the brain) to acute hypoxia. Our findings over 1-h HE support a larger sensitivity of the cerebral cortex compared to the muscle, even if qualitative signs of skeletal muscle hypoxemia reached significance already after 10 min. Major differences were observed in cerebral and muscle tissue deoxygenation magnitude throughout HE, with a large increase in [HHb] in PFC and MC compared to the vastus lateralis muscle. Previous results have also revealed that the hypoxic cerebrovascular responsiveness has a relatively slow on-response rate.⁵⁵ The differences observed between the brain and the quadriceps might be due to differences in basal metabolism and in the microvascular architecture and responsiveness of both tissues.^53,56 Hence, muscle capillary recruitment may be a tissue-specific strategy to limit the impact of the decreased arterial saturation.⁵⁷

From $\sim 2$- to 4-h HE, changes in muscle [${\mathrm{O}}_2\mathrm{Hb}$], [HHb], and [tHb] (Fig. 3) may reflect a decrease in local blood flow and/or ${\mathrm{O}}_2$ quadriceps extraction, raising the possibility that blood flow was shifted to other organs and/or other muscle groups (e.g., those involved in the seated position), independent of gas exposure. This seems a plausible explanation since the decline in [HHb] and [tHb] was similar between normoxic and hypoxic sessions (see Sec. 4.4).

In the examined brain regions, increase in perfusion ([tHb]) was slightly different in PFC than MC. A marked increase in cerebral ${\mathrm{O}}_2$ blood volume in PFC, as indicated by the higher [tHb] values, was detected as early as after 20 min of HE and remained fairly constant between 2 and 4 h of HE. In contrast, the increased [tHb] occurred only at 240 min HE in MC region. This may be partly explained when considering the regions of the brain associated with integrating sensory input; i.e., the PFC can be considered as an active region even in resting conditions as it is an area of the brain responsible for attention and motivation.³³ One possible explanation for this PFC [tHb] elevation at rest might be the effect of being on the seat for such a prolonged time and consequently engaged in (introspective) thoughts. These two patterns were similar in both conditions despite distinct levels of $\mathrm{Sp}{\mathrm{O}}_2$ and are therefore independent of mechanisms associated with HE.

4.3.

Cardiorespiratory and NIRS Changes and Altitude Sickness Symptoms During Sustained HE

High-altitude headache has been proposed to be mainly caused by the acute reduction in cerebral regional oxygen saturation in unacclimatized trekkers,³⁹ and acetazolomide decreases AMS symptomatology, at least in part, by helping to maintain cerebral oxygenation at altitudes up to 5700 m.⁵⁸ Therefore, it is likely that the level of brain deoxygenation plays a key role in the acclimation process. Cerebral oxygenation has been shown to be influenced by sympathetic activity at rest.⁵⁹ In the present study, stimulation of sympathetic activity as revealed by HRV indices and increased HR (Table 1) during HE was concomitant with a decrease in tissue (cerebral and muscle) oxygenation. The well-known progressive increase in HR with sustained hypoxia occurred during the first 20 min and was followed by a steady state over the 4 h of HE. The observed unchanged MABP during the first hour may then result from a decline in systemic vascular resistance and supports the absence of strong systemic influences on cerebral oxygenation in the early phase ($<1\mathrm{h}$) of physiological adjustments to hypoxia. Also, blood pressure did not differ between conditions when HE was prolonged, even if MABP increased slightly over time. This is in agreement with previous studies reporting that acute short-lasting hypoxia ($<1$ day) influences only slightly systemic blood pressure in humans.^7,60

Our study and others^61,62 demonstrated a change in the sympathetic–parasympathetic balance in response to acute HE at rest, with a sympathetic tone increase together with a decrease in parasympathetic tone (i.e., $\mathrm{L}{\mathrm{F}}_{\mathrm{RR}}$ increase and $\mathrm{H}{\mathrm{F}}_{\mathrm{RR}}$ decrease over time, respectively; see Table 1). Previous studies on AMS debated the fact that increased sympathetic activity during high-altitude exposure may be involved in acute hypoxic adaptations^63,64 and that parameters like HRV could permit the predication of AMS susceptibility.⁶⁴ In accordance with Subudhi et al.,⁵ we found that being 4 h at $\mathrm{Fi}{\mathrm{O}}_2=0.12$ significantly increases altitude illness symptoms (i.e., headache, LLS, AMS-C score). These results support this time frame to be critical regarding the genesis of hypoxia-induced cerebral perturbations. However, no relationship among HRV indices, AMS symptoms, and NIRS changes was found in the present study. This confirms recent findings suggesting that changes in cardiac autonomic modulations may not be related to AMS development in the early phase of acute normobaric hypoxia.⁶²

4.4.

Methodological Considerations

Over the brain, the detected NIR light is affected not only by cortical hemodynamic changes but also by changes in optical properties of all other superficial tissue layers between the optode and the gray matter (e.g., scalp, skull, CSF).^65,66 However, to minimize the contribution of hemodynamic changes from the skin/scalp layer (e.g., vasomotion/flow motion induced by mechanical skin irritation and/or sympathetic hyperactivation^67,68) in the high-probability banana-shaped photon flux paths, we enlarged interoptode distances (up to 3.5 cm for the brain and 4 cm for the muscle) to reach the maximal light path providing a sufficient signal-to-noise ratio of the optical density measurements.¹⁹ Controlled ambient temperature and attention given to ensure head NIRS setup to be noncompressive (i.e., no scalp ischemia/hyperaemia) were also important to reduce the weight of the skin layer (i.e., extracranial blood flow in the skin tissue) in perturbing the observed chromophore concentration changes. We are confident that, taken together, the behavior of the present oxygenation changes most probably reflected the brain tissue oxygenation changes. NIRS penetration depth (i.e., approximately half the interoptode distance) limits its sensitivity to the upper 1 cm of the cerebral cortex,⁶⁹ so that it is important to consider that the measurements obtained are regional and strictly confined to the zone beneath the sensors. We acknowledge that the observed tissue oxygenation cannot be generalized to whole brain or even to cerebral lobes.

In order to investigate the kinetic of cerebral hemodynamic responses to 4-h HE, subjects had to be completely relaxed (without sleeping or speaking) in order to avoid inputs from other sources (e.g., anxiety, voluntary hypo- or hyperventilation, etc.), which would influence the measured response. Thanks to the normoxic control session, we identified the fact that staying awake but totally inactive in a sitting position for a long time might have induced some slight effects on parameters expected to remain constant (e.g., muscle and cerebral tHb on Fig. 2). A gradual decrease in cerebral blood velocity (and $\mathrm{EtC}{\mathrm{O}}_2$) has been previously reported to occur both in normoxic and hypoxic conditions in subjects sitting for 5 h.⁷ Accordingly, in our study, concomitant muscle hypoperfusion and cerebral hyperperfusion were likely due to the prolonged position rather than to the nature of the inhaled gas over time.

Finally, we used a normobaric hypoxic protocol, and whether our results would have been heightened or different if using hypobaric hypoxia is still an open question.