Effect of Fast Ascent, Acclimatization and Re-Exposure To 5050 m On Cerebral Autoregulation in Unacclimatized Lowlanders. A Prospective Cohort Study.

Cerebral autoregulation (CA) is impaired during acute high-altitude (HA) exposure and effects of acclimatization and re-exposure on CA are unknown. In 18 healthy lowlanders (11 women), we hypothesized that the cerebral autoregulation index (ARI) assessed by the percentage change in middle cerebral artery peak blood flow velocity (Δ%MCAv)/percentage change in mean arterial blood pressure (Δ%MAP) induced by a sit-to-stand maneuver, is i) reduced on Day1 at 5050m compared to 520m, ii) is improved after 6 days at 5050m, and iii) is less impaired during re-exposure to 5050m after 7 days at 520m compared to Cycle1. Participants spent 4-8h/day at 5050m and slept at 2900m similar to real-life working shifts. High/low ARI indicate impaired/intact CA, respectively. With the sit-to-stand at 520m, mean(95%CI) in ΔMAP and ΔMCAv were -26%(-41 to -10) and -13%(-19 to -7); mean±SE in ARI was 0.58±0.63 Δ%/Δ% , respectively. On Day1 at 5050m, ARI worsened compared to 520m (3.29±0.70 Δ%/Δ% ), but improved with acclimatization (1.44±0.65 Δ%/Δ% , P<0.05 for both). ARI was less affected during re-exposure to 5050m (1.22±0.70 Δ%/Δ% , P<0.05 acute altitude-induced change between sojourns). This study showed that CA i) is impaired during acute HA exposure, ii) improves with acclimatization and iii) is ameliorated during re-exposure to HA a week later. score = acute score, P ET CO 2 = partial pressure of exhaled carbon dioxide, SpO 2 = oxygen saturation measured by finger pulse oximetry, MCAv = middle cerebral artery peak velocity, MAP = mean arterial pressure, ARI = autoregulation index.


INTRODUCTION
Recent developments had led to an increase in the number of settlements and work places at altitude, especially, astronomical observation centers and resource extraction facilities that are often situated at very high altitudes (4000-5000 m). At the Atacama Large Millimeter / Submillimeter Array (ALMA), the largest ground-based telescope station worldwide, employees sleep at 2900 m and work at 5050 m for 7 days, and recover for 7 days at their permanent residence, normally near sea level. These repeated fast ascents, brief periods of acclimatization and re-exposures to very high altitudes may have adverse effects on health, performance and safety of workers. [1][2][3] Cerebral autoregulation (CA) protects the brain by a negative feedback loop mechanism, maintaining constant cerebral blood flow and oxygen delivery independent of fluctuations in systemic blood pressure. 4 Impaired CA results in systemic blood pressure associated cerebral blood flow fluctuations and may cause under-perfusion and inadequate oxygen delivery, or over-perfusion of capillaries with consequent disruption of the blood-brain barrier, capillary damage and microhemorrhages. 5 Furthermore, impaired CA has been shown to have consequences on the cerebral function and is related to cognitive impairment. 5 At altitude, the arterial partial pressures of O 2 and CO 2 , two key factors influencing CA functionality, are reduced. 6 Studies investigating CA functionality at different altitudes have reported persistent CA impairment with acute 7 and prolonged high altitude exposure of up to 2 weeks compared to values reported near sea level. 8 However, reallife working shifts as implied in ALMA require workers to sleep at 2900 m and work at 5050 m. This pattern of exposure to high altitude induces an intermittent hypobaric hypoxic exposure, which might have beneficial effects on the CA functionality as suggested for intermittent hypoxic training in patients with Alzheimer diseases. 9 Furthermore, studies investigating the effects of repeated high altitude exposure showed milder reductions in O 2 and CO 2 when compared to previous altitude sojourns. 2,10 However, whether these improvements in arterial blood gases might have a beneficial effect on the CA has not been studied. Therefore, the purpose of this study was to test the hypotheses that CA is impaired during acute high-altitude exposure and that acclimatization (with an intermittent schedule of daily ascents to very high altitude, but sleeping at moderate altitude), and a 2 nd altitude sojourn, ameliorates CA functionality compared to the first acute high altitude exposure. Furthermore, this study assessed whether altitude exposure has sustained effects on CA functionality after descending to low altitude.

Design and study setting
Data of the current study were collected within a prospective cohort trial with the purpose to investigate the effects of acute, prolonged and repeated altitude exposure on cognitive performance in healthy subjects. 2 Results of the present study have not been published elsewhere.
The ascent and assessment schedule is illustrated in Figure 1. At Santiago de Chile (520 m), participants performed a familiarization and baseline session separated by a oneday rest interval. The day after, participants traveled by plane and bus (2h each) to the ALMA Operation Support Facility (ASF; 2900 m) and spent seven consecutive nights at 2900 m. Each morning, participants were driven by car (45 min travel time) to the ALMA Operation Site (AOS; 5050 m), where they stayed 4 -8 hours without oxygen supplementation. Measurements at 5050 m were repeated on the 1 st and 6 th day.
Measurements were repeated on the day after descending to 520 m. This assessment schedule was repeated after a 7-day recovery period at 520 m.

Participants
Healthy, altitude-naïve men and women, aged between 18 to 30 years were recruited from the University of Calgary, Canada (N = 18, altitude of 1100 m). Participants were instructed to avoid any overnight stays at altitudes >1500 m within four weeks before the study. All participants provided written informed consent. The study was approved by the Conjoint Health Research Ethics Board of the University of Calgary (Ethics ID: REB 15-2709) and the Cantonal Ethics Committee of Zurich (2016-00048). The main trial was registered at ClinicalTrials.gov (NCT02731456). The study has been performed in accordance with the Declaration of Helsinki.

Protocol and Measurements
Sitting values were collected and averaged over 120 stable and artifact-free heart beats at the end of a 10-minute sitting position in a comfortable chair. To assess CA functionality, participants stood up in less than 1 second and remained in standing position for 1 minute. 11 The effect of standing up was measured by averaging the first 15 beats in standing position. This procedure was repeated and values were averaged within each subject.
Clinical assessments: Clinical assessment included height, weight, oxygen saturation by pulse oximetry (SpO 2 ) and heart rate measurements, as well as the environmental symptoms questionnaire cerebral score (AMSc) to assess acute mountain sickness (AMS). 12 The AMSc comprises 11 questions on AMS symptoms (feel sick, feel hungover, coordination off, dim vision, lightheaded, headache, dizzy, loss of appetite, feel weak, nausea, faint) rated from 0 (not at all) to 5 (extreme). A weighted AMSc score ≥0.7 were

Outcomes
The main outcome of this study was the ARI (Δ%MCAv/Δ%MAP) and the effects of acute, prolonged and re-exposure to very high altitude. Additional outcomes were changes in explanatory cerebral, respiratory and cardiac variables. No a priori sample size estimation has been performed.

Hypotheses
The three main hypotheses were that i) acute high-altitude exposure impairs ARI (Δ%MCAv/Δ%MAP), ii) 6 days of acclimatization (with an intermittent hypobaric hypoxic protocol) improves ARI (Δ%MCAv/Δ%MAP) and iii) a 2 nd compared to the 1 st ascent to 5050 m has a lower impact on the altitude-related ARI impairment. An additional hypothesis was that CA functionality is immediately restored the day after descending to 520 m.

Data analysis
The primary outcome was tested for normality by the Shapiro Wilk test. Due to the nonnormally distributed data, mixed ordered logistic regression using the interaction between altitude (520 m and 5050 m), day at altitude (1 st and 6 th day) and number of altitude sojourns (1 st and 2 nd sojourn) as fixed effects, and participants as random effects was applied. Secondary outcomes were analyzed using mixed linear regression models.
To elucidate the underlying factors influencing ARI (Δ% / Δ%) or AMS, univariate and multivariable mixed linear regressions including baseline characteristics, SpO 2 , P ET CO 2 , MAP and MCAv were performed. Statistical significance was assumed when P<0.05 and 95% confidence intervals of mean differences did not overlap zero.

RESULTS
A total of 18 healthy participants (11 women) with a mean age of 24 ± 4 years and BMI of 22.8 ± 3.1 kg/m 2 were included in the final analysis. Participant characteristics are presented in Table 1. Cardiorespiratory and cerebrovascular outcomes are described in Table 2.

Effect of acute altitude exposure (1 st day at 5050 m compared to 520 m)
With acute exposure to 5050 m, participants had lower values in SpO 2 , mean ± SE, 77.6 ± 0.8% vs. 96.5 ± 0.8%, lower values in P ET CO 2 , 24.5 ± 0.7 mmHg vs. 33.9 ± 0.6 mmHg and higher heart rates, 93 ± 3 bpm vs. 70 ± 3 bpm and more symptoms of AMS assessed by the AMSc score, 0.94 ± 0.09 vs. 0.12 ± 0.09 compared to 520 m (P<0.05 all comparisons). There was no acute altitude effect on MAP compared to 520 m, whereas sitting mean MCAv significantly increased from 57. 9 Figure 2).

Effect of prolonged altitude exposure (6 th versus 1 st day at 5050 m)
With 6 days of acclimatization at 5050 m and sleeping altitude at 2900 m, participants improved their SpO 2 (83.0 ± 0.6% vs. 77.6 ± 0.6%) without altering P ET CO 2 or breathing frequency, but with associated elevation in hemoglobin concentrations and lower AMS symptoms compared to the 1 st day at 5050 m. With acclimatization, MAP remained unchanged and MCAv tended to further increase without altering CVRi or CVCi.
Compared to the 1 st day at 5050 m, the sit-to-stand test caused a similar MCAv decrease despite a stronger drop in MAP, resulting in significantly improved ARI (Δ%/Δ%) compared to the 1 st day at 5050 m (Table 2, Figure 2).

Effect of descent from high altitude (520 m versus 6 th day at 5050 m)
On the day after descending from 5050 m to 520 m, participants showed persistent P ET CO 2 reductions and worse ARI (Δ%/Δ%) values compared to pre-ascent evaluations.
After 1-week recovery period at 520 m, SpO 2 and P ET CO 2 recovered to pre-ascent values.
When the participants descended from the 2 nd sojourn at 5050 m, they persistently showed lower P ET CO 2 , lower MAP and better ARI (Δ%/Δ%) values compared to the values obtained on the day after descending from the 1 st sojourn (Odds ratio of 0.12, 95%CI 0.02 to 0.82, P=0.030).

Predictors for AMS and ARI (Δ%/Δ%)
Predictors of AMS severity at 5050 m assessed by multivariable regression analysis were acute exposure, low sitting P ET CO 2 and high ARI (Δ%/Δ%) values (Table 3). Only higher MAP values remained as an independent predictor of CA impairment defined by ARI (Δ%/Δ%) ( Table 4).

DISCUSSION
This study focused on the cerebral autoregulatory ability to protect the brain from rapid systemic blood pressure falls in maneuvers like the sit-to-stand test during acute, prolonged and repeated exposure to very high altitude (5050 m). The present findings suggest that cerebral autoregulation is impaired during acute exposure, but improved after a 7-day stay at high altitude (with a sleeping altitude of 2900 m). Strikingly, one week after altitude exposure the CA remained impaired. Moreover, novel findings of this study suggest that a 2 nd altitude sojourn after one week at low altitude has a milder effect on the CA functionality, indicating that workers exposed to repeated high altitude exposures might be protected from initial cerebral autoregulatory impairments seen during the first acute high-altitude exposure. Acute exposure to 5050 m, worse ARI (Δ%/Δ%) and lower P ET CO 2 values were associated with higher AMS severity.
This study used a straightforward approach to describe CA proposed by a recent metaanalysis, by calculating the percent change in MCAv divided by the percent change from baseline in MAP. 13 The meta-analysis comprises 49 studies, whereas 41 studies used transcranial Doppler measurements to assess MCAv, a surrogate for cerebral blood flow. staying at 5260 m). Furthermore, in healthy lowlanders, it has been shown that the functionality of cerebral autoregulation is more protective to blood pressure elevations (CA ability to vasoconstrict blood vessels) than blood pressure reductions (CA ability to vasodilate). 13 Since spontaneous blood pressure changes assessed by transfer function analysis includes both blood pressure elevations and reductions, this might provide different information about the CA functionality than exclusive blood pressure reductions induced by a sit-to-stand maneuver. This non-linear CA functionality might partly explain previously reported opposite findings at high altitude. 8,21 However, whether the altitude-induced CA impairment differs in the ability to protect the brain from blood pressure elevations versus blood pressure reductions has not been studied in detail.
Moreover, this study provides novel findings supporting the hypothesis that CA is less impaired during a second sojourn at very high altitude. Contrary to a previous study by Subudhi et al. 10 , which concluded that re-exposure to 5260 m after 7 days at sea level similarly impaired CA (based on transfer function analysis). However, they have not performed a second baseline measurement and might have missed the circumstance that the participants had persistently impaired CA after 1 week at low altitude, as seen in the current study (Table 2). This would result in a different baseline and therefore, smaller altitude-induced difference in CA functionality. The observed smaller altitudeinduced impact on CA functionality might be partly explained by preserved oxygen delivery towards the brain by elevated hemoglobin concentration and less altitudeinduced hypoxemia, or by less vasodilatation of the small cerebral arteries allowing the CA to induce vasodilation in response to the sit-to-stand maneuver.
The persistent impairment of CA throughout 1-week post-altitude exposure at low altitude came as a surprise and needs further investigation. Findings on the day after the second descent to 520 m suggest already improved CA after the 2 nd sojourn, indicating that CA functionality is normalized back at low altitude after repeated exposure. Nevertheless, sustained impairment of CA after repeated altitude sojourns would have an unknown impact on the safety and health of high-altitude workers and needs further investigation.
The assessment of the CA at high altitude is challenging; many influential physiological parameters change, therefore, invasive and highly sophisticated measurements would be required and large sample sizes would be needed to correct for the influences of various confounders. Altitude exposure is associated with various changes influencing the CA functionality, i.e. changes in arterial blood gases (causing vasoconstriction or vasodilation), change in plasma volume and associated elevation of hemoglobin concentration (lower blood volume with higher viscosity), change in driving pressure (difference between the MAP and the intracranial pressure) and change in cerebral blood vessel diameter. 4,19 The findings of this study support the hypothesis that acclimatization and re-exposure have beneficial effects on the CA against blood pressure falls. However, underlying mechanisms and physiological explanations remain speculative and further studies are needed.

CONCLUSIONS
First, this study confirms CA impairment during acute exposure to very high altitude.
Further, novel findings reported here are the beneficial effects of acclimatization to very high altitude by spending the work-day at 5050 m and sleeping at 2900 m. Furthermore, we found a milder altitude-induced impact on the CA functionality during a 2 nd similar cycle at high altitude but persistent CA impairment up to 1-week post-altitude exposure, which resolved after the 2 nd high altitude sojourn. Taken together, these findings suggest that high altitude shift work with 1-week breaks at low altitude between work cycles, might be at risk for over and under-perfusion of brain areas during commencement of high-altitude work. Whether the improved CA functionality during the second high-altitude cycle indicates a protection also for several repeated work-shift cycles at high altitude remains to be determined.    day at 5050 m; 2 nd sojourn, 1 st and 6 th day at 5050 m). To account for the low number of observations (a total of 51 observations due to single missing values), the 5 most significant predictors from the univariate regression were entered into the multivariable regression model. AMS-c score = acute mountain sickness-cerebral score, P ET CO 2 = partial pressure of exhaled carbon dioxide, SpO 2 = oxygen saturation measured by finger pulse oximetry, MCAv = middle cerebral artery peak velocity, MAP = mean arterial pressure, ARI = autoregulation index. day at 5050 m; 2 nd sojourn: 1 st and 6 th day at 5050 m). To account for the low number of observations (a total of 53 observations due to single missing values), the 5 most significant predictors from the univariate regression were entered into the multivariable regression model. P ET CO 2 = partial pressure of exhaled carbon dioxide, SpO 2 = oxygen saturation measured by finger pulse oximetry, MCAv = middle cerebral artery peak velocity, MAP = mean arterial pressure, ARI = autoregulation index.