Abstract
Background and aims
Altitude-related medical literature provides very few simple clinical studies relating to those on ‘adventure holidays’. Systemic blood pressure has seldom been studied closely in relation to altitude. This study aimed to address both these issues and to assist GPs approached by patients for pre-trek advice.
Methods and results
A total of 17 hillwalkers, evenly distributed for gender and age, trekked gradually from moderate to extreme altitude on Mera Peak in the Himalaya, noting any altitude sickness symptoms. Heart rate, blood pressure, oxygen saturation, peak expiratory flow and core temperature were measured daily. Altitude was double-checked hourly and synchronised with each set of measurements. On each day, two individuals wore 24-h ambulatory blood pressure monitors for assessment of altitude effects. Two principal findings emerged. Firstly, none of our 17 developed altitude-related symptoms below 4000 m, consistent with the recognised protective effect of slow rate of ascent; at 3500–4000 m all showed a sharp fall on O2sat and above 4500 m symptoms arose unpredictably. Secondly, hourly blood pressure monitoring showed no altitude effect below 3500 m, but above 5000 m a marked yet asymptomatic rise with delayed and prolonged peak.
Conclusion
There may be a critical altitude above which extra vigilance is required; blood pressure here needs further research.
Introduction
At a symposium on altitude at The Lancet in 2003, Basnyat and Murdoch 1 proposed that future studies should ‘observe the complete time sequence of changes that occur in response to hypobaric hypoxia’. Following a personal experience of asymptomatic prolonged extreme hypertension (215/115, normal 100/60) on Aconcagua, and soon afterwards the sudden death from intracranial haemorrhage, completely without symptoms, of a fit American climber during descent from a Himalayan 8000-m peak, the writer arranged a project 2 to attempt to fulfil Basnyat and Murdoch’s proposal with particular scrutiny of systemic blood pressure (BP).
Materials and methods
The project was unfunded, the writer having retired from clinical or university medical practice. Attempts to interest some altitude medicine experts failed. Volunteers were therefore self-financing, KE Adventure Travel organising an appropriate trek with a discounted group rate. Recruitment was largely personal, among colleagues from Expedition Medicine and Mountain Leader training and from the Great Outdoors Challenge (a self-supported trek across the Scottish Highlands), with some additional responders to publication on KE website and in the new Edinburgh store of Cotswold Outdoor. A total of 15 subjects, plus two trek leaders who allowed us to include their data when appropriate, were evenly distributed for age over or under 50 (range 29–72 years) and for gender. All were sent full details of the project, and all understood the medical aims and were enthusiastic about participating. Ethical clearance was obtained, plus written informed consent from each individual.
The parameters measured were heart rate, BP, oxygen saturation (O2sat), peak expiratory flow and core temperature; FEV1 was included for nine days until the loss of a vital component of the equipment. Lightweight, compact digital instruments were all checked for accuracy, reproducibility and non-operator-dependence before departure; in the field, non-medical trekkers were trained in their use with reliable results. We also carried three ambulatory blood pressure monitors (ABPMs) kindly loaned by Spacelabs UK and two altimeters, constantly cross-checked, for hourly altitude records. Details of all medical instruments are available in online supplementary material.
Routine measurements (not ABPM) were taken on all individuals on most days (twice on one day for pre-and post-exertion results), the timing governed by terrain and facilities for obtaining accurate data; results were charted against the altitude in metres at the same time. At high camp (4850 m altitude), we began to divide into subgroups as individuals developed altitude symptoms and turned around from different levels, leaving us insufficient instruments to continue monitoring; graphs therefore concentrate on the acclimatisation trek up to 4850 m.
The ABPMs were set to hourly recording, which could also be matched closely to altitude: two were used each 24 h, all trekkers taking a turn. We were able to continue limited use of these (although one malfunctioned) at very high and extreme altitude above 5000 m.
Practical problems in managing measurements and in the expedition generally are being reported in another paper (in preparation).
Results
Acute mountain sickness
Of primary relevance is the fact that none of our 17 volunteers developed any significant altitude-related symptoms until we approached high camp at 4850 m. Our very long trek-in, with no rest-days, produced some weariness and possibly some lowering of immunity resulting in minor infections, but slow acclimatisation appeared entirely successful in preventing acute mountain sickness (AMS).
Oxygen saturation
A striking feature is that, as we ascended through altitude levels 3500 m–4000 m, all trekkers showed a sharp drop in O2sat (Figure 1). Before this, most levels were well maintained. It has been suggested3–5 that a low O2sat level recorded early in a trek may be predictive of later altitude symptoms and reduced chance of summit success; and here, the two individuals with a pronounced, transient earlier drop were indeed among the earliest to turn around. But even the eventual summit group, and a very fit, acclimatised trek leader, showed this downturn on the day of ascent through this particular altitude and stabilisation at a lower O2sat thereafter, although all remained asymptomatic and were functioning well.
Systemic BP
Up to 3500 m–4000 m altitude, all physiological parameters remained essentially stable, and analysis of all results reveals that the oft-reported and unpredictable systemic hypertensive response to altitude did not occur; this shows on daily readings of systolic, diastolic and mean arterial pressure (MAP) averaged over the group (Figure 2) and over age and gender groups, and also on every individual’s hourly ABPM readings against hourly altitude on each day of the trek up to this level (Figures 3 and 4). (Sample graphs are shown, more available online.)
At 3500–4000 m, on daily readings every individual showed a transient drop in BP, followed by a rise on further ascent to a level higher than hitherto in the trek (Figure 5), which reflects Bärtsch and Gibbs observation 6 that the initial reaction to hypobaric hypoxia is peripheral vasodilatation, thereafter overtaken by sympathetic response, raising BP to a higher level than before the dip; it also supports Parati’s suggestion 7 that for BP rise at altitudes over 5400 m, the causative mechanism differs from that for any rise below 3500 m. (Additional graphs online.)
The two ABPMs in use on the day we ascended through this range also reflect this, in systolic, diastolic, MAP and pulse pressure: the specific timing appears similar, and the BP settles afterwards to a higher resting level than in those same individuals early in the trek (Figures 6 and 7).
At altitudes above 5000 m, our limited data suggest an inevitable rise in BP, the peak of which may be delayed a few hours beyond the point of maximum altitude and may persist or recur for many hours afterwards (Figures 8 and 9); in our cases it was completely asymptomatic.
Other data at the critical altitude (illustrated online)
In both daily and hourly graphs, heart rate mostly runs parallel to effort and therefore to ascent, but it also shows a dip in the midst of this ascent through 3500–4000 m before rising again, until it settles to previous levels as subjects rest at the higher altitude.
We had also been monitoring peak expiratory flow, but were forced to a change of instrument just at this 3500–4000 m altitude on the trek, which must invalidate the drop which also appears in this reading for all. This appears worth repeating. Our FEV1 monitor was by now unavailable.
There is also a suggestion that core temperature reaches a low point at this altitude, 14 of our 15 volunteers showing a dip on this day.
Discussion
This study has two principal findings, the first a confirmation of previous knowledge about the importance of rate of ascent, but with a specific danger level even when ascent is gradual. None of our participants developed altitude-related symptoms until our day approaching high camp at 4850 m, although all showed a sustained drop in oxygen saturation at 3500–4000 m. Several developed symptoms at various levels above 4300 m. The second finding is that systemic BP did not rise in response to altitude increase in any individuals until above 4000 m, but above 5000 m the rise was marked, prolonged and delayed for some hours after maximum altitude in the two individuals studied.
The strength of this study lies in its use of typical people in the real environment in which dangerous medical problems may afflict healthy individuals; specific trek planning ensured maximum cooperation by participants. Its weakness is its small size, with insufficient numbers for statistical significance, due to lack of funding.
Planning in relation to previous research concentrated on sequential studies as suggested by Basnyat and Murdoch, 1 particularly those involving systemic BP.
Stokes et al. 3 measured BP and calculated MAP, but did not quote results except to state that MAP at 4700 m showed no difference between under-50 s and over-50 s.
Jackson et al. 8 used the Lake Louise consensus scoring system (LLS) 9 (form completed by subjects), Vardy et al. 10 used the same LLS system (from interviews) and Kayser 11 used a questionnaire (completed by trekkers) to gauge the incidence of AMS, but clinical measurements were not involved.
Karinen et al., 4 in a comprehensive and valuable study, showed the predictive value of O2sat measurement, relating it to self-reported altitude symptoms, which began to appear at 3500 m; Pomidori et al. 5 also emphasised the predictive value of O2sat measurement, especially after exercise at each altitude. BP changes were not part of these studies.
Beidleman et al. 12 studied respiratory changes using a pressure chamber and found no change in MAP, but comparison was restricted to rapid ascent to 4300 m with and without pre-staging at 2200 m.
Wagner 13 studied heart rate and body temperature continuously in a single individual from 5800 m to 6962 m, but not O2sat or BP.
A full review of literature studied is in a detailed account of this expedition, available from the author.
Clinicians presented with a request for advice by patients planning high-altitude holidays are faced with a vast literature on the topic of physiological response to altitude, much of which reports highly sophisticated and detailed science of the underlying mechanisms. Altitude-induced hypoxia is also being used to inform intensive care. This is immensely valuable and interesting, yet still preventable tragedies occur on mountains, suggesting the need for wider dissemination of knowledge of basic clinical responses among non-specialist doctors and among the interested public to increase awareness of risk and encourage simple measures to minimise it. This study showed that non-medical individuals, with basic instruction, can monitor themselves and each other, and our first finding suggests a particular altitude level which demands care even in gradual ascents.
Future research should clarify our second finding, of the possible significance of prolonged asymptomatic hypertension to deaths at extreme altitude from unknown cause; Firth et al. 14 reported that some 70% of deaths above 8000 m on Everest occur during descent from the summit or from a high turnaround point, and bodies are often left on the mountain at such altitudes, precluding autopsy. Although a few studies7,15–17 have involved ABPMs, they have been an underused tool in altitude research as it applies to mountaineers and those attempting Everest.
Footnotes
Acknowledgments
The authors acknowledge the support of KE Adventure staff and especially of all volunteers.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The generosity of individual friends who donated towards instrument purchase, of Spacelabs UK who loaned ABPMs, and of outdoor firms who offered equipment discounts is greatly appreciated. Altitude and mean oxygen saturation, all individuals. Altitude and mean systolic and diastolic pressure, daily, entire group. NC-S altitude and hourly BP Day 4. MS altitude and hourly BP Day 10. Altitude and mean arterial pressure (MAP), mean for all individuals: daily readings. Hourly altitude and MAP early in trek, GH and JN day 3. Hourly altitude and MAP, same individuals day 11. JT high-alt hourly MAP. HJ, hourly systolic and diastolic BP up to extreme altitude, days 15–16.
