Clinical recommendations for high altitude exposure of individuals with pre-existing cardiovascular conditions: A joint statement by the European Society of Cardiology, the Council on Hypertension of the European Society of Cardiology, the European Society of Hypertension, the International Society of Mountain Medicine, the Italian Society of Hypertension and the Italian Society of Mountain Medicine

Gianfranco Parati,  Piergiuseppe Agostoni,  Buddha Basnyat,  Grzegorz Bilo,  Hermann Brugger,  Antonio Coca, Luigi Festi,  Guido Giardini,  Alessandra Lironcurti,  Andrew M Luks ... Show more

European Heart Journal, Volume 39, Issue 17, 01 May 2018, Pages 1546–1554,
Published:  11 January 2018  Article history


Take home figure

Adapted from Bärtsch and Gibbs2 Physiological response to hypoxia. Life-sustaining oxygen delivery, in spite of a reduction in the partial pressure of inhaled oxygen between 25% and 60% (respectively at 2500 m and 8000 m), is ensured by an increase in pulmonary ventilation, an increase in cardiac output by increasing heart rate, changes in vascular tone, as well as an increase in haemoglobin concentration. BP, blood pressure; HR, heart rate; PaCO2, partial pressure of arterial carbon dioxide.

Issue Section:  Disease management


The travelling options currently available allow an increasingly large number of individuals, including sedentary people, the elderly and diseased patients, to reach high altitude (HA) locations, defined as locations higher than 2500 m above sea level (asl),1S i.e. the altitude above which many of the physiological responses that represent challenges for the human body start developing. Physiological acclimatization mechanisms impose an increased workload on the cardiovascular system, but the actual risk of adverse cardiovascular events associated with HA exposure is still a debated issue.1–4

The aim of this article is to review the available evidence on the effects of HA in cardiovascular patients and to address their risk of developing clinically relevant events. This was done through multiple Medline searches on the PubMed database, with the main aim of promoting a generally safe access to mountains. Searched terms included a combination of either ‘high altitude’ or ‘hypobaric hypoxia’ plus each of the following: ‘physiology’, ‘maladaption’, ‘cardiovascular response’, ‘systemic hypertension’, ‘pulmonary hypertension’, ‘ischaemic heart disease’, ‘cardiac revascularisation’, ‘heart failure’, ‘congenital heart disease’, ‘arrhythmias’, ‘implantable cardiac devices’, ‘stroke’, ‘cerebral haemorrhage’, ‘exercise’, ‘sleep apnea’. Compared with a previous review article on this topic,2S we now include the most recent data on hypoxia-induced changes in left ventricular (LV) systolic and diastolic function, lung function and ventilation control, blood coagulation, and on the effects of pharmacological interventions. We also offer an update on the clinical and pathophysiological findings related to the exposure to altitude of patients with pre-existing cardiovascular conditions (ischaemic heart disease, heart failure, and arterial and pulmonary hypertension).

Physics and cardiovascular physiology at high altitude

With increasing altitude, a progressive reduction in barometric pressure, air temperature and air humidity can be observed (Figure1).1S For the purpose of this article, we refer to Imray et al.’s1S classification of altitude ranges.

Figure 1

Altitude classification (Imray et al.1S) (left column); corresponding barometric pressure and fraction in inspired oxygen for different simulated altitudes in a laboratory setting, according to the 1976 US standard Atmosphere by NASA.5S (central two columns); relationship between altitude2S and environmental characteristics (temperature, humidity, and solar radiation) (box on the right-hand side). We used the 1976 US standard atmosphere model by NASA to estimate barometric pressure at a given altitude, because the former is a function not only of altitude but also of latitude. For similar altitudes, barometric pressure (and consequently also partial pressure of arterial oxygen) is higher the closer we are to the equator line.

Altitude classification Imray et al1S left column corresponding barometric pressure and fraction in inspired oxygen for different simulated altitudes in a laboratory setting according to the 1976 US standard Atmosphere by NASA5S central two columns relationship between altitude2S and environmental characteristics temperature humidity and solar radiation box on the right-hand side We used the 1976 US standard atmosphere model by NASA to estimate barometric pressure at a given altitude because the former is a function not only of altitude but also of latitude For similar altitudes barometric pressure and consequently also partial pressure of arterial oxygen is higher the closer we are to the equator line

Barometric pressure directly determines the inspired oxygen (O2) partial pressure and, in combination with alveolar ventilation, sets the alveolar O2 partial pressure. Its reduction leads to a condition known as ‘hypobaric hypoxia’. In practice, at sea level, O2 constitutes 20.94% of total gas molecules in inspired air, which with a normal rate of alveolar ventilation leads to an alveolar partial O2 pressure of roughly 100 mmHg for a barometric pressure of roughly 760 mmHg.5S When breathing at 3000 m altitude asl, the same percentage of O2 in the inspired air, combined with a lower barometric pressure and higher rate of ventilation, results in an alveolar partial O2 pressure of roughly 67 mmHg, corresponding to what would occur breathing a hypoxic air mixture (fraction of inspired O2 0.14) at sea level (Figure1).5S A series of physiological responses help to maintain adequate tissue O2 delivery and supply at HA, through a process called ‘acclimatization’. Its efficacy depends on the duration of individual’s exposure to altitude, age, sea level partial pressure of oxygen in arterial blood (PaO2) and minute ventilation.2,3,4S,5S These crucial processes include increase in ventilation, cardiac output, red cell mass and blood O2 carrying capacity, and other metabolic modifications at the microvascular and cellular levels (Take home figure). Some of these mechanisms are activated almost immediately, whereas others need hours to days to attain full expression.2,4–6 A more extensive description of the effects of HA exposure on cardiovascular physiology is provided in the seminal papers by Bärtsch et al.2

Take home figure

Adapted from Bärtsch and Gibbs2 Physiological response to hypoxia. Life-sustaining oxygen delivery, in spite of a reduction in the partial pressure of inhaled oxygen between 25% and 60% (respectively at 2500 m and 8000 m), is ensured by an increase in pulmonary ventilation, an increase in cardiac output by increasing heart rate, changes in vascular tone, as well as an increase in haemoglobin concentration. BP, blood pressure; HR, heart rate; PaCO2, partial pressure of arterial carbon dioxide.

Adapted from Brtsch and Gibbs2 Physiological response to hypoxia Life-sustaining oxygen delivery in spite of a reduction in the partial pressure of inhaled oxygen between 25 and 60 respectively at 2500m and 8000m is ensured by an increase in pulmonary ventilation an increase in cardiac output by increasing heart rate changes in vascular tone as well as an increase in haemoglobin concentration BP blood pressure HR heart rate PaCO2 partial pressure of arterial carbon dioxide

Systemic blood pressure and heart rate

Acute exposure to hypoxia produces endothelium-dependent and endothelium-independent systemic vasodilation,6S–9S which may initially induce some degree of blood pressure (BP) reduction. After a few hours, this is counter-balanced, however, by a generalized altitude-dependent increase in sympathetically mediated vasoconstriction, caused primarily by arterial hypoxaemia through afferent signalling to the cardiovascular control regions of the mid-brain via the arterial peripheral chemoreceptors located in the carotid bodies.5,6,10S As a result, a significant and persistent arterial BP increase occurs shortly after the arrival at HA, proportional to the altitude reached and more evident at night.7 This leads to a reduced degree of the physiological blood pressure fall during sleep,8 which persists at least over the first 7 days of altitude exposure.9 This is accompanied by an increase in heart rate (HR) both at rest and during exercise,10,11,11S although maximal HR achieved during exercise at HA is lower compared to sea level (see Supplementary material online, Table S1).2,11–14


With acute exposure to altitude, the decrease in PaO2 stimulates peripheral chemoreceptors in the carotid bodies leading to sympathetic activation and to an increase in minute ventilation.2,14 Moreover, mild interstitial lung fluid accumulation15 can occur in physically active persons and decreases alveolar diffusing capacity.14 Hypoxia-induced hyperventilation leads to hypocapnia and respiratory alkalosis, which blunts the initial full hypoxic ventilatory response. The combination of hyperventilation and resulting hypocapnia, with increased peripheral chemosensitivity and abnormal loop gain in chemoreflex-induced respiratory regulation7S may lead to the appearance of nocturnal periodic breathing (PB). This is an abnormal ventilatory pattern, characterized by periods of central apnoea or hypopnoea alternating with periods of hyperventilation, mainly occurring during sleep.10,16,17,13S,18

With exposure over days to weeks, the sensitivity of the peripheral chemoreceptors to hypoxia increases, resulting in a further increase in sympathetic activity and enhancement of ventilation, despite the progressive increase in arterial blood O2 content and lower partial pressure of arterial carbon dioxide (PaCO2) (‘ventilatory acclimatization’, see Supplementary material online, Table S1).19,14S

Pulmonary arterial pressure

Alveolar hypoxia and arterial hypoxaemia (to a lesser degree) induce vasoconstriction in the pulmonary circulation, either directly and through sympathetic activation,14,18,19,14S,15S resulting in increased pulmonary vascular resistance and pulmonary artery pressure (hypoxic pulmonary vasoconstriction).14SHypoxic pulmonary vasoconstriction is a protective mechanism during regional alveolar hypoxia (e.g. pneumonia) to shift blood flow to better ventilated lung regions, but at HA, where the hypoxic stimulus is ubiquitous throughout the lungs, global alveolar hypoxia leads to general pulmonary hypertension with the risk of pulmonary oedema or right ventricular failure in the extreme cases (see Supplementary material online, Table S1).16S We found only one study investigating pre-capillary pulmonary hypertension.17SAlthough the sample size was small (n = 14) and the exposure to simulated HA was short (at rest and after 20 min of mild exertion), non-invasive measures of the right heart function demonstrated a predictable rise in pulmonary arterial systolic pressure, not associated with a deterioration in the right heart function.

Left ventricular function

The left ventricle undergoes significant changes when exposed to HA.11,20,21,17S In particular, a reduction in both diastolic and systolic LV volumes and geometrical alterations (increase in the sphericity index) occur, the diastolic function worsens and the LV contractility and LV apex twist increase, the last change being similar to what is observed in cases of subendocardial LV fibre dysfunction.20 Moreover, after 2 weeks of exposure to very HA, LV mass decreases disproportionately when compared with the concomitant reduction in total body weight (11% vs. 3% reduction for LV mass vs. body mass, adjusted for body surface area, P < 0.05).21 Lung impairment via cardiopulmonary interaction (with arterial hypoxaemia made worse due to mild interstitial oedema), especially during the first 2 weeks, and the increased LV inotropic stimulation by an increased sympathetic activity are considered to play a role in the development of these changes, but probably they cannot completely explain the extent of the observed findings.

Recent evidence suggests that hypoxia itself can be the basis for LV alterations. 31P magnetic resonance spectroscopy performed before and after ascent to Mt. Everest in healthy individuals revealed a decrease in the cardiac creatine phosphate/adenosine triphosphate (PCr/ATP) ratio by 18% (P < 0.01), similarly to what is observed in patients with diseases associated with chronic hypoxia.21 All these reductions returned to pre-trek levels 6 months after return to the sea level. The authors concluded that a decrease in energy reserve may be a ‘universal response to periods of sustained low O2 availability, underlying hypoxia-induced cardiac dysfunction both in the healthy human heart and in patients with cardiopulmonary diseases’.

Other effects

Mild dehydration and a hypoxic-mediated diuresis were found to lead to an acute increase in haematocrit and haemoglobin concentration in the first several days at HA,18S after which the renal production of erythropoietin stimulates new red blood cell production to increase red cell mass, with a further rise in haemoglobin concentration. In the acute exposure phase, this can be associated with increased blood coagulability.22 Nevertheless, an increased thrombotic risk at HA has never been convincingly demonstrated, and the limited available data are conflicting with respect to a hypothetical prothrombotic state.22,23

Heart failure

Heart failure (HF) is often associated with co-morbidities, such as pulmonary hypertension, chronic obstructive pulmonary disease, chronic kidney disease, cardiac ischaemia, anaemia, and thrombophilia. This condition is also characterized by an increased chemosensitivity.20S All these conditions are likely to make HF patients more vulnerable to the HA environment. In spite of this, brief simulated HA exposure was found to be safe for HF patients,24 even when performing mild physical exercise. Agostoni et al.24evaluated 38 patients with severe stable HF [New York Heart Association (NYHA) Class III–IV] undergoing cardiopulmonary exercise testing with a progressive reduction in inspired O2 from 21% (sea level) to 18%, 16%, and 14% (the last simulating 3000 m altitude). No episodes of angina, arrhythmias, or electrocardiographic (ECG) evidence of ischaemia occurred at any simulated altitude. The reduction in maximum work rate achieved at simulated altitude was progressively greater the more severe the exercise limitation at sea level. Finally, patients who showed the largest increase in the lung diffusion capacity for CO (DLCO) at sea level during moderate exercise were those who showed the lowest exercise capacity reduction at simulated altitude,20S linking HA exercise performance to gas exchange adaptability.

Schmid et al.25 evaluated stable HF patients (NYHA Class II) exercising during a short exposure to HA (3454 m, Jungfraujoch, Switzerland) and at sea level. During HA exercise, mean peak VO2 decreased by 22%, without causing arrhythmias or altering echocardiographic variables, with the exception of an increase in pulmonary artery systolic pressure (PAPs). Drug therapy of HF may also interfere with HA adaptation mechanisms. Critical HF drugs such as beta-blockers and angiotensin-converting enzyme (ACE) inhibitors act on the chemoreceptors and on haemodynamic responses through adrenergic receptors, which are also involved in the alveolar–capillary gas diffusion control (β2-receptors). Despite their importance in the care of patients with HF which is incontrovertible at sea level, ACE inhibitors and angiotensin receptor blockers (ARBs) do blunt the kidney’s ability to produce erythropoietin and could limit the compensatory rise in haematocrit and blood O2-carrying capacity that is important at HA.21S,22SHIGHCARE (HIGH altitude CArdiovascular REsearch) investigators reported that healthy subjects receiving the non-cardioselective β2 antagonist carvedilol reached lower peak exercise ventilation and lower VO2peak at HA than those receiving the selective β1 receptor blocker nebivolol.13 Carvedilol also reduced hyperventilation during exercise, possibly by reducing peripheral chemoreceptor sensitivity. This effect may be favourable in normoxia, because the fall in arterial O2 saturation is quite small at sea level, but can be much greater and unfavourable at HA.20S,23S Moreover, administration of diuretics should be based on the balanced evaluation of signs of early dehydration or fluid gain. Among the diuretics, acetazolamide, a carbonic anhydrase inhibitor with mild diuretic effect, which is frequently used for mountain sickness prophylaxis and treatment, should be specifically considered.24S It should be emphasized that the concomitant administration of acetazolamide and other diuretics may increase the risk of dehydration and electrolyte imbalances at HA and should thus be carefully evaluated.

A final issue is related to periodic breathing. It can be present in HF patients at sea level, both during sleep and exercise, but likely it will worsen at HA.12S,25S,26 Whether it should be suppressed with acetazolamide remains an issue yet to be clarified.

Very little is known about the effects of HA in patients who underwent heart transplant. Living at moderate altitude seems not to be harmful, but no data are available on the effects of acute HA exposure.26SRecommendations for HF patients going to be exposed to HA are summarized in Table 1.

Table 1

Recommendations for heart failure patients going to high altitude

The strength of these recommendations is to be weighted in the light of the limited evidence available.

HF, heart failure; NYHA, New York Heart Association.

Ischaemic heart disease

It has been suggested that living permanently at moderate altitude might be beneficial, reducing CV mortality.27S Acute exposure to HA, on the other hand, may represent a more challenging condition for the cardiovascular system. Given that total O2 demand is constant for a given workload and that myocardial O2 extraction is already very high at sea level, with acute HA exposure cardiac output must increase to maintain O2 delivery despite the reduced blood arterial O2 content.

In healthy subjects, epicardial coronary blood flow increases due to vasodilation during acute HA exposure and thus there may not be any significant impairment during exercise of coronary flow reserve, at least up to 4500 m.27 The actual risk of cardiac ischaemia associated with HA is indeed unclear. On the one hand, stress testing performed in healthy subjects above a simulated altitude of 8000 m27S did not induce ECG alterations; on the other hand, changes in arterial wall properties may reduce coronary O2 supply in diastole at HA, with possible clinical implications in subjects with silent coronary plaques. Evidence in this regard has been obtained through calculation of the subendocardial viability ratio (SEVR), defined as the ratio between diastolic pressure–time index (DPTI, an estimate of myocardial O2 supply based on both coronary artery driving pressure in diastole and diastolic time) and systolic pressure–time index (SPTI, an estimate of myocardial O2 consumption in systole). Subendocardial viability ratio, therefore, indirectly estimates the degree of myocardial perfusion and was found to be significantly reduced at 4559 m in healthy volunteers (from 1.63 ± 0.15 to 1.18 ± 0.17; P < 0.001).28 The administration of acetazolamide was associated with a smaller degree of SEVR reduction under hypobaric hypoxia exposure and with faster recovery after residing for 3 days at HA, suggesting that acetazolamide may offset the reduction in subendocardial O2 supply in these conditions.28 Similarly, at HA, also the O2 supply/demand ratio (SEVR-CaO2, i.e. SEVR corrected for the arterial O2 content) displayed significant reductions, which were more pronounced with placebo (from 29.6 ± 4.0 to 17.3 ± 3.0; P <