Effects of High Altitude Exposure on Cerebral Hemodynamics in Normal Subjects
Background and Purpose— Acute mountain sickness (AMS) may be an early stage of high altitude cerebral edema. If so, AMS could result from an alteration of dynamic autoregulation of cerebral blood flow resulting in overperfusion of capillaries and vasogenic cerebral edema.
Methods— We measured middle cerebral artery blood flow velocity (Vmca) by transcranial Doppler and arterial blood pressure by finger plethysmography at 490 m and 20 hours after arrival at 4559 m in 35 volunteers who had been randomized to tadalafil, dexamethasone, or placebo in a study on the pharmacological prevention of high altitude pulmonary edema. A dynamic cerebral autoregulation index (ARI) was calculated from continuous recordings of Vmca and blood pressure during transiently induced hypotension.
Results— Altitude was associated with an increase in a cerebral-sensible AMS (AMS-C) score (P<0.001) and with a decrease in arterial oxygen saturation (Sao2), whereas average Vmca or ARI did not change. However, at altitude, the subjects with the lowest ARI combined with the lowest Sao2 presented with the highest AMS-C score (P<0.03). In addition, a stepwise multiple linear regression analysis on arterial Pco2, Sao2, and baseline or altitude ARI identified altitude ARI as the only significant predictor of the AMS-C score (P=0.01). The AMS-C score was lower in dexamethasone-treated subjects compared with high altitude pulmonary edema-susceptible untreated subjects. Neither tadalafil nor dexamethasone had any significant effect on Vmca or ARI.
Conclusions— High altitude hypoxia is associated with impairment in the regulation of the cerebral circulation that might play a role in AMS pathogenesis.
The pathogenesis of acute mountain sickness (AMS) remains incompletely understood. A leading hypothesis relates AMS to early stages of brain edema that may progress in a proportion of subjects to high altitude cerebral edema.1 In this context, cerebral blood flow (CBF) would be expected to be a potential aggravating or contributing factor. However, reported CBF measurements in AMS have been inconsistent, with normal as well as increased values, and no clear relationship to AMS symptomatology.1–4 Lassen suggested that AMS could result rather from altered autoregulation of CBF, allowing for an overperfusion of cerebral capillaries and vasogenic cerebral edema.5 This notion was challenged recently by a study that showed that static cerebral autoregulation was similarly impaired in newcomers at high altitude and in high altitude resident Sherpas.6
We took advantage of a study on the pharmacological prevention of high altitude pulmonary edema (HAPE) by tadalafil and dexamethasone to investigate CBF regulation during short-term high altitude exposure with variable AMS symptomatology. We measured dynamic cerebral autoregulation.7 Our hypothesis was that altitude exposure could alter dynamic cerebral autoregulation and arteriolar tone, thereby leading to capillary overperfusion and subclinical cerebral edema-related AMS symptomatology. Tadalafil is a phosphodiesterase-5 inhibitor used in the treatment of erectile dysfunction,8 which, if anything, could aggravate AMS in relation to cyclic GMP-related cerebral vasodilatation and associated headache.8 Dexamethasone is an effective treatment of vasogenic cerebral edema and, as such, has been successfully used for treatment of AMS.9
Materials and Methods
Thirty-five healthy volunteers (29 men and 6 women; 26 to 58 years of age; mean 40 years; weighing 74±6 kg [mean ± SD]) gave informed consent to the study, which was approved by the ethical committees of the University Hospital of Zürich and the University of Heidelberg. Twenty-six of them had at least 1 previous episode of HAPE. Subjects had been recruited into a study on the effects of tadalafil, a phosphodiesterase-5 inhibitor, and dexamethasone on the prevention of HAPE. Tadalafil was expected to prevent HAPE through inhibition of hypoxic pulmonary vasoconstriction and associated increase in pulmonary capillary pressure and dexamethasone by an activation of respiratory epithelial sodium transport.10 This subject population was expected to present with a high incidence of AMS because HAPE often occurs in the presence of established AMS symptomatology.10
CBF velocity was estimated by the measurement of middle cerebral artery blood flow velocity (Vmca) at a depth of 50 to 55 mm using a 2-MHz pulsed transcranial Doppler (WAKI 2TC; Atys Medical). The ultrasonic transducer was positioned on the temporal window and fixed with a headband, allowing continuous recording without modification of the insonation angle. The right or left middle cerebral artery was insonated according to the best quality Doppler signal.
Arterial blood pressure (BP) was measured by finger plethysmography (Finapres BP Monitor; Ohmeda 2300). The CBF velocity and the arterial BP signals were sampled at 200 Hz using an analog/digital converter (RTI 800; Analog Devices) and stored on a personal computer for off-line computations. Arterial Pco2, arterial Po2, and arterial oxygen saturation (Sao2) were measured with an automated blood gas analyzer (model ABL 5; Radiometer). Heart rate (HR) was measured on BP signals, which were continuously monitored.
A cerebral autoregulation index (ARI) was calculated in triplicate from continuously recorded Vmca and BP during an acute hypotension induced by the sudden release of bilateral thigh cuffs that had been kept inflated to a pressure of 30 mm Hg above systolic BP for 3 minutes as described previously.11 This maneuver allows for a decrease in mean BP by >15% for 20 to 30 s, during which, after an initial fall, Vmca smoothly returns back to baseline in 5 to 7 seconds. We calculated a dynamic rate of regulation, which expresses the rate of restoration of Vmca with respect to the decrease in mean BP, and derived an ARI using previously reported algorithm and software.12 The normal response of CBF to a fall in BP is a 20% correction per second, with a restoration to baseline in 5 seconds, yielding an ARI ≈5.11,12
The diagnosis and severity of AMS were established with use of a cerebral-sensitive (AMS-C) score of the Environmental Symptom Questionnaire, as proposed previously by Sampson et al.13 This self-report questionnaire consists of 67 questions, of which 11 are used to calculate the AMS-C score by multiplying the score of each question by a factorial weight. After that, the 11 products are added and the sum is multiplied by 0.1927. The influence of each question is then hidden for the examiner and the subject, making the assessment a very objective way to evaluate AMS. An AMS-C score of >0.7 establishes a diagnosis of AMS.13
Measurements were performed first at low altitude at the University Hospital of Zürich at 490 m (sea level), and then at 4559 m at the research laboratory hut Regina Margherita, on Monte Rosa (high altitude). One day before ascent, the study participants started taking either 10 mg tadalafil BID or 8 mg dexamethasone BID or placebo in a randomized double-blind protocol. They were transported by cable car to an altitude of 3200 m, and then climbed for 2 hours to a hut at the altitude of 3650 m, where they stayed overnight. The next morning, they walked up a gletcher for 4 to 6 hours to the Regina Margherita hut at 4559 m.
All measurements were taken on subjects in semirecumbent position, in a quiet room, and under stable conditions as assessed by unchanged Vmca, BP, and HR for ≥10 minutes. CBF with derived ARI, HR, BP, Paco2, Pao2, and Sao2 were measured at sea level and on average 20 hours after arrival at high altitude.
Results are presented as mean (SD). The statistical analysis consisted in paired t tests to compare sea level and high altitude measurements and linear regression calculations.14 The normality of the distributions of the variables was checked using normal probability plot and the Wilk Shapiro test. A logarithmic transformation (natural logarithm) was applied when the variable was not gaussian (AMS-C score). A multiple linear regression analysis was performed to determine the predictors of the AMS-C score. Sao2, Paco2, baseline ARI, and altitude ARI were used as predictors in a model built using a backward elimination procedure.15 A P value <0.05 was considered significant to stay into the final model. The regression coefficients with their SEs are presented with the P value of the t test against 0.
Thigh cuff release decreased mean BP by 25±11 mm Hg in normoxia and 18±9 in hypoxia (P<0.01). The resulting ARI calculation was not correlated to BP drop, but was inversely correlated to mean Vmca at sea level and at high altitude (r=−0.39, P<0.05, and r=−0.40, respectively). High altitude exposure was associated with an AMS-C score of 0.89±0.88.
The effects of high altitude on systemic and cerebral hemodynamics are presented in Table 1. Altitude increased HR, decreased Sao2, Pao2, and Paco2, but did not change BP, Vmca, and ARI.
When subgroups defined by 50% cut points for Sao2 and the AMS-C score were compared, altitude ARI in percentage of baseline was higher in the high-Sao2 group (118% versus 94%; P<0.03), whereas ARI was not different in the high- and low-AMS-C score groups. However, the AMS-C score was higher at 1.54±1.20 in the combined low Sao2–low ARI group; P<0.05 compared with an AMS-C score of 0.36±0.36 in the high-Sao2–high-ARI group. The AMS-C scores of the high-Sao2–low-ARI and the low-Sao2–high-ARI groups were intermediate, respectively, at 0.74±0.75 and 0.9±0.76, and not different from the 2 other groups.
The results of the simple linear regression analysis to determine the slopes of the relationships between AMS-C and, respectively, Sao2, Paco2, baseline ARI, and altitude ARI, and the final result of the backward procedure for the multiple regression analysis are shown in Table 2. It is apparent that baseline ARI and altitude Sao2 and Paco2 were not related to the AMS-C score. The backward procedure disclosed ARI as the only significant independent predictor of the AMS-C score (P=0.0107). The individual AMS-C scores and ARI measurements are shown in Figure 1. When the same procedure was applied to test the relationship between ARI and altitude Sao2, Pao2, and baseline ARI, only Sao2 emerged as a significant predictor.
Two of the 35 subjects included in the present study and 3 other visitors of the Regina Margherita hut experienced severe AMS and HAPE requiring emergency treatment with oxygen, acetazolamide, nifedipine, or dexamethasone. In these subjects, there were parallel evolutions between ARI and AMS-C score (Figure 2).
There were no significant differences in BP, Vmca, or ARI between the tadalafil-, dexamethasone-, or placebo-treated subjects. However, the AMS-C score of the subjects taking dexamethasone was lower than in placebo controls (0.28±0.27 versus 1.38±1.17; P<0.05). The AMS-C score of tadalafil-treated subjects tended to be lower than in the controls (0.92±0.85 versus 1.38±1.17; P=0.062).
The new finding of the present study is that AMS symptoms in subjects acutely exposed to high altitude appear to be related to an altered dynamic autoregulation of the cerebral circulation.
Transcranial Doppler was used for the noninvasive and beat-by-beat estimation of CBF. This approach is based on the reasonable assumption that hypoxia or acute hypotension would not alter the diameter of the middle cerebral artery.16,17 Finger plethysmography was used for the noninvasive beat-by-beat estimation of cerebral arterial pressure. This approach is validated for the measurement of instantaneous relative changes, and as such, has been used previously for cerebral autoregulation studies and ARI computing.11,12
Previous studies in normal subjects using the same methodology as in the present study reported a normal ARI of ≈5, range from 3 to 7, for an average BP drop ranging from 15 to 28 mm Hg.12 Our normal subjects presented with a baseline ARI of 4.44 for an average cuff release-induced drop in BP on cuff release of 18 to 25 mm Hg, which is in keeping with these previous data. It is of interest that ARI was not correlated to BP drop, as reported previously as well.11 However, contrary to previous reports,11 we found an inverse correlation between ARI and Vmca at sea level and at high altitude, suggestive of a relative inhibition of cerebral autoregulation by higher blood flow velocities or decreased cerebral resistance.
In the present study, the average ARI was not different between sea level and altitude. This is in contrast with previous reports of an impaired autoregulation in experimental animals in hypoxia18 and in healthy newcomers to high altitudes or high altitude residents.6,19 These discrepancies might be explained by compensatory enhancement of cerebral autoregulation by hypocapnia7,20 and by differences in methodological approach. The only available study showing similar decrease in cerebral autoregulation in high altitude newcomers and long-term residents did not consider the evolution of cerebral hemodynamics in relation to the quality of adaptation measured by an AMS score, and measured cerebral autoregulation in static conditions, with increases in BP induced by a phenylephrine infusion.6 Although static and dynamic cerebral autoregulation measurements have been shown to be well correlated,12 the agreement between both approaches in hypoxia is not known. Static cerebral autoregulation measurements require phenylephrine infusion. However, vascular reactivity to hypoxia may be affected by chronic as well as acute sympathetic influences and vice versa. Dynamic cerebral autoregulation measurements are performed without drug administration that might affect arteriolar tone, and within 5 seconds, thus preceding any possible vascular effect of decreased flow-induced sympathetic nervous system activation.21 Therefore, we believe that dynamically measured ARI more closely reflects physiological cerebral autoregulation and is accordingly more likely related to AMS symptomatology.
In the present study, the subjects with the most severe AMS symptomatology tended to present with lower ARI, indicating altered dynamic cerebral autoregulation and decreased arteriolar tone, and this was significant despite concomitant hypocapnia. These changes could be a cause of increased capillary perfusion, increased filtration pressure, and resultant cerebral edema. Alternatively, cerebral edema related to other yet unknown direct effects of hypoxia1 could be a cause of secondary decrease in cerebral arteriolar tone and altered cerebral autoregulation.22 Although AMS and high altitude cerebral edema present at the extremes of a spectrum of clinical neurological manifestations of high altitude intolerance, with efficacy of dexamethasone therapy likely related to its effects on vasogenic cerebral edema in general,9 data available until now do not prove the existence of increased cerebral extravascular water content related to headache symptomatology in AMS.1
It may be argued that the correlations between the AMS-C score and altitude ARI, although significant, were loose and accordingly not suggestive of a causal relationship. However, loose correlations could also be attributed to the variability of self-reporting questionnaire AMS manifestations13 and the inherent approximations of ARI estimates.11 On the other hand, for the same reasons, it appears unlikely that AMS and ARI would be significantly correlated by chance.
Tadalafil would not be expected to affect AMS symptoms, except for an aggravation of headache that is a side effect of phosphodiesteras-5 inhibitors when used to treat erectile dysfunction.8 Sildenafil does not affect CBF velocity in normal subjects.23 The absence of cerebral vascular effects of tadalafil in our subjects is thus not surprising. On the other hand, dexamethasone is of established efficacy in the prevention and treatment of AMS.9,24 There is no previous report on the effects of dexamethasone on CBF autoregulation. Although dexamethasone showed no cerebral vascular effect in the present study, we interpreted this negative finding with caution because the study was not a priori designed and powered to answer that question.
In conclusion, altitude exposure is associated with altered dynamic cerebral autoregulation in proportion to AMS symptomatology. This observation offers rationale for further studies on the role of abnormal regulation of cerebral hemodynamics in the adaptation to high altitude.
This study was supported by grant 3.4613.04 from the Fonds de la Recherche Scientifique Medicale (FRSM), Belgium; the Hartmann-Müller Foundation, Zürich; the Pierluigi Crivelli Foundation, Lugano; and the Anna Feddersen-Wagner Fonds, University of Zürich. We are most grateful to all participants and the hut keepers, as well as the Sezione Varallo of the Italian Alpine Club for providing the facilities at the Capanna Regina Margherita.
- Received September 29, 2004.
- Revision received October 26, 2004.
- Accepted November 30, 2004.
Bärtsch P, Roach R. Acute mountain sickness and high-altitude cerebral edema. In: High Altitude. An Exploration of Human Adaptation. In: Lenfant C, Hornbein TF, Schoene RB, eds. Lung Biology in Health and Disease. New York, NY: Marcel Dekker; 2001: 731–776.
Reeves JT, Moore LG, McCullough RE, McCullough RG, Harrison G, Tranmer BI, Micco AJ, Tucker A, Weil JV. Headache at high altitude is not related to internal carotid arterial blood velocity. J Appl Physiol. 1985; 59: 909–915.
Otis SM, Rossman ME, Schneider PA, Rush MP, Ringelstein EB. Relationship of cerebral blood flow regulation to acute mountain sickness. J Ultrasound Med. 1989; 8: 143–148.
Baumgartner RW, Spyridopoulos I, Bärtsch P, Maggiorini M, Oelz O. Acute mountain sickness is not related to cerebral blood flow: a decompression chamber study. J Appl Physiol. 1999; 86: 1578–1582.
Jansen GFA, Krins A, Basnyat B, Bosch A, Odoom JA. Cerebral autoregulation in subjects adapted and not adapted to high altitude. Stroke. 2000; 31: 2314–2318.
Aaslid R, Lindegaard KF, Sorteberg W, Normes H. Cerebral autoregulation dynamics in humans. Stroke. 1989; 20: 45–52.
Mahoney PJ, Panerai RB, Deverson ST, Hayes PD, Evans DH. Assessment of the thigh cuff technique for measurement of dynamic cerebral autoregulation. Stroke. 2000; 31: 476–480.
Tiecks FP, Lam AM, Aaslid R, Newell DW. Comparison of static and dynamic cerebral autoregulation measurements. Stroke. 1995; 26: 1014–1019.
Winer BJ. Statistical Principles in Experimental Designs. 2nd ed. New York, NY: McGraw-Hill; 1971: 514–603.
Kleinbaum DG, Kupper LL, Muller KE. Applied Regression Analysis and Other Multivariate Methods. 2nd ed. Boston, Mass: PWS Publishing Company; 1988: 324–325.
Poulin MJ, Robbins PA. Indexes of flow and cross-sectional area of the middle cerebral artery using Doppler ultrasound during hypoxia and hypercapnia in humans. Stroke. 1996; 27: 2244–2250.
Larsen FS, Olsen KS, Hansen OB, Knudsen GN. Transcranial Doppler is valid for determination of the lower limit of cerebral blood flow autoregulation. Stroke. 1994; 25: 1985–1988.
Häggendal E, Johansson B. Effects of arterial carbon dioxide tension and oxygen saturation on cerebral blood flow autoregulation in dogs. Acta Physiol Scand. 1965; 66: 27–53.
Guyton AC, Batson H, Smith CM, Armstrong GG. Method for studying competence of the body’s blood pressure regulatory mechanism and effect of pressoreceptor denervation. Am J Physiol. 1951; 79: 1156–1162.