Pulmonary hypertension (PH) is a chronic and progressive disease that increases pulmonary arterial pressure (PAP) and pulmonary vascular resistance (PVR), ultimately leading to right ventricular failure. Regardless of the underlying disease, PH is almost always associated with progressive exercise intolerance, dyspnea, and increased mortality [1, 2, 3].

In clinical settings, the cardiopulmonary exercise test (CPET) is very useful for pulmonologists and cardiologists in the follow-up of patients with PH. It is used to evaluate exercise tolerance, exertional dyspnea, and related underlying pathophysiological mechanisms; it can suggest the diagnosis of PH and differentiate between possible causes. Moreover, some of the exercise variables are used as prognostic factors, mainly in pulmonary arterial hypertension (PAH) and chronic thromboembolic pulmonary hypertension (CTEPH) [4, 5]. In CTEPH, CPET is useful in detecting the disease, establishing severity, and identifying causes of exercise intolerance [6, 7]. Likewise, it has been described that patients with CTEPH have greater dead space, ventilatory inefficiency, and more severe alterations in gas exchange during exercise than patients with PAH [8].

High altitude is an elevation over 2500 m (~8200 feet) [9]. Although the physiological responses to hypobaric hypoxia start at lower elevations, they are more pronounced above this altitude, and the risk of developing altitude illness also increases substantially [10]. At altitude, the barometric pressure (BP) decreases, meaning the inspired oxygen pressure (PIO${}_2$) and arterial oxygen pressure (PaO${}_2$) also decrease. In Bogotá, a city located at high altitude (2640 m, BP: 560 mmHg), the PaO${}_2$ at rest in healthy subjects is around 65 mmHg, with values lower than 60 mmHg in older individuals [11]; values are even lower in patients with cardiopulmonary disease. In patients with chronic obstructive pulmonary disease (COPD) living at high altitude, we have observed more severe gas exchange alterations during exercise than described at sea level, mainly in those with a higher degree of obstruction or with coexistent PH [12, 13].

In patients with PH who live at high altitude, in addition to the pathophysiological alterations related to pulmonary vascular compromise, changes related to the decrease in PIO${}_2$ are added, meaning gas exchange alterations in these patients are expected to be more severe than those described at sea level. Since there are no studies evaluating the impact of high altitude on exercise capacity, ventilatory and cardiovascular responses, and gas exchange in patients with PH, we designed this study to describe and compare these variables in a CPET between patients with PAH and CTEPH.

The main findings of this study, which assessed a significant number of PH patients residing at high altitude were the following: (1) In comparison to the controls, PH patients had lower exercise capacity (peak VO${}_2$ and WR) and severe gas exchange alterations. (2) There was no difference in exercise capacity between PAH and CTEPH patients. (3) There was more dyspnea, greater ventilatory inefficiency, and more severe gas exchange alterations during exercise in patients with CTEPH than with PAH. (4) Compared to what is described at sea level, due to the lower PIO${}_2$ at altitude, for both CTEPH and PAH, the PaO${}_2$ and SaO${}_2$ during exercise were lower, and due to the compensatory increase in ventilation, the VE/VCO${}_2$ ratio was greater.

In this study, in subjects living at high altitude, as expected, there was lower exercise capacity (peak VO${}_2$ and WR) and severe alterations in gas exchange in patients with PH compared to controls. PAH and CTEPH patients had lower VO${}_2$/HR and higher VE/VCO${}_2$, VD/VT, Pa-ETCO${}_2$, hypoxemia, and PA-aO${}_2$ than controls, which is related to pulmonary vascular compromise [4, 23]. These lower VO${}_2$/HR values manifest the alteration in the stroke volume that can be seen in the presence of PH. The high VE/VCO${}_2$ ratio is a hallmark abnormality in patients with pulmonary vascular disease, primarily resulting from high VD/VT. In patients with PH, an increase in the VE/VCO${}_2$ ratio has been related to different, usually coexisting mechanisms, including high VD/VT, as already mentioned, abnormalities in gas exchange, increased chemosensitivity, and an abnormal PaCO${}_2$ set point [24]. The higher Pa-ETCO${}_2$ and PA-aO${}_2$ values reflect ventilation/perfusion imbalance [4, 23, 24].

Although there were no differences between PAH and CTEPH in hemodynamics, peak VO${}_2$, and WR, similar to previous studies conducted at sea level [8, 25, 26], the ventilatory inefficiency and gas exchange alterations in CTEPH during exercise were more severe in comparison to PAH, with higher VE/VCO${}_2$, VD/VT, Pa-ETCO${}_2$, and PA-aO${}_2$, and lower PaO${}_2$ and SaO${}_2$. We highlight that, unlike controls, which significantly increased PaO${}_2$ during exercise, the increase in patients with PAH was much smaller, while there was no significant change in those with CTEPH. Similarly, in normal subjects, PaCO${}_2$ decreased significantly from exercise but did not change in those with PAH and CTEPH.

Consistent with our results, a previous study described significantly lower PETCO${}_2$ and significantly higher end-tidal capillary carbon dioxide gradients in CTEPH versus PAH, both at rest and during exercise. It has been described that a gradient $>$7.0 mmHg would indicate CTEPH with a sensitivity of 75% at rest and 88% during exercise, which suggests the usefulness of this variable in the differential diagnosis between these two pathologies [26].

Several potential mechanisms could explain the differences in gas exchange and ventilatory efficiency during exercise between PAH and CTEPH. In CTEPH, there is anatomical compromise and heterogeneity in pulmonary blood flow. In addition to intravascular obstruction of the pulmonary arteries by unresolved organized fibrotic clots, pulmonary vascular remodeling can lead to severe pulmonary microvasculopathy, which affects the small muscular pulmonary arteries, pulmonary capillaries, and veins. Enlargement and proliferation of systemic bronchial arteries also occur, as well as anastomoses between the systemic and pulmonary circulations that promote the development of microvasculopathy [27, 28].

Although most patients with PH were non-smokers and those who had smoked had a low pack-year index, the FEV${}_1$/FVC ratio was lower in patients with PH than in controls, mainly in those with CTEPH. Even though restrictive alteration in pulmonary function tests has been described in patients with PH [29], obstructive ventilatory alteration has also been reported in both PAH [29, 30, 31] and CTEPH [8, 26, 32, 33, 34]. Similar to our data, in some studies that compared spirometric values between patients with PH, the FEV${}_1$ and the FEV${}_1$/FVC ratio were lower in patients with CTEPH than in PAH [8, 26, 32].

The decrease in FEV${}_1$ or the FEV${}_1$/FVC ratio has been attributed to different possible mechanisms, such as the involvement of the peripheral airways, effects of vasoactive or inflammatory substances, compression of the airways related to arterial dilation, and, less likely, to inspiratory muscle dysfunction [29, 30, 34, 35].

Dyspnea also was more severe in CTEPH patients. Although the exact mechanisms related to dyspnea in PH patients are not completely understood [4], the increased dyspnea in the CTEPH group was probably explained by the higher VD/VT [8, 26] and ventilatory inefficiency and more severe gas exchange alterations than in PAH patients. Although we did not perform inspiratory capacity measurements through exercise, another possible mechanism that could be related to the increased exertional dyspnea intensity in these patients is dynamic hyperinflation (DH) [4, 35].

It is estimated that over 500 million humans live at $\ge$1500 m, 81.6 million at $\ge$2500 m, and 14.4 million at $\ge$3500 m [36]. Living at altitude imposes a challenge on humans due to changes in oxygen pressure and climatic variables. The reduction in BP and the consequent decrease in PIO${}_2$ in the atmosphere condition changes the ventilatory pattern and causes a decrease in oxygenation, which is more pronounced in patients with cardiopulmonary disease. In previous studies at the same altitude as Bogotá, we have shown, compared to studies at sea level, a high prevalence of PH in patients with COPD, particularly in patients with less severe airflow obstruction [37], and more PH in patients with idiopathic pulmonary fibrosis [38]. In these patients, the PaO${}_2$ at rest or during exercise is lower than that reported at sea level [12, 38] and even lower in patients with COPD and the coexistence of PH [13]. Although there are several pathophysiological mechanisms related to the development of PH, alveolar hypoxia at high altitudes is probably a fundamental factor in promoting and developing PH [39, 40]. Although hypoxia-inducible factor (HIF) signaling is a mechanism related to disease progression in group 3 PH (associated with lung diseases and/or hypoxia), increased HIF-1$\alpha$ has also been observed in the lung tissue of patients with PAH and CTEPH [41, 42].

In control subjects, PaCO${}_2$ and PETCO${}_2$ were lower and VE/VCO${}_2$ higher compared to descriptions in normal subjects at sea level, which is explained by increased ventilation, a well-recognized compensatory mechanism for adaptation to altitude [43, 44]. Similarly, in PH patients residing at 2640 m, the VE/VCO${}_2$ ratio was lower than the values described in various studies at sea level: 47 to 68 in patients with CTEPH [8, 25, 45, 46] and 42 to 54 in those with PAH [8, 25, 46, 47, 48, 49, 50, 51]. Regarding oxygenation, studies at sea level in patients with CTEPH have described values of PaO${}_2$ in exercise between 62 and 70 mmHg [8, 46] and for arterial oxygen saturation by pulse oximetry (SpO${}_2$) between 91 and 93% [7, 8], which are higher than those found in our study. Similarly, in PAH patients, the PaO${}_2$ described during exercise at sea level was between 78 and 83 mmHg [8, 46], while SpO${}_2$ was between 89 and 92% [8, 48, 49], also higher than in these patients at high altitude. The lower PaO${}_2$ in subjects residing at high altitude can be explained, in addition to ventilation/perfusion ratio (V/Q) alterations related to the disease, by low PIO${}_2$ secondary to decreased BP [52].

In a recent study in patients with PAH and CTEPH, acute altitude exposure after ascending from 470 m to 2500 m caused a significant decrease in exercise capacity, ventilatory efficiency, and oxygenation [53]. It is striking that, despite these being patients at a slightly lower altitude than in our study, who also were in a better functional class and had lower pulmonary vascular resistance than the patients in our study, the level of hypoxemia at rest and during exercise was significantly higher. This indicates the presence of adaptive mechanisms in high-altitude resident subjects who are chronically exposed to hypoxia, which has been previously described in healthy subjects and in patients with other diseases, such as COPD [11, 12, 54].

This is the first study to evaluate exercise capacity and gas exchange alterations in patients with PH living at high altitude. Moreover, including patients with PAH, CTEPH, and control subjects allowed us to compare groups; meanwhile, measuring the ABG and ventilatory variables comprehensively assessed the limiting mechanisms of exercise in these patients with PH. Consistent with several studies at sea level, conducting this study at high altitude we also show a greater compromise in ventilatory efficiency and oxygenation in CTEPH than in PAH patients, with no differences in VO${}_2$ or peak WR between these groups [8, 25, 26]. However, unlike these studies, we observed modifications to the ventilatory pattern secondary to adaptive compensatory hyperventilation at altitude and lower oxygenation values, both in PAH and CTEPH [7, 8, 46, 48, 49], which can be explained by the lower PIO${}_2$ secondary to decreased BP [52].

We consider that the clinical utility of CPET is to evaluate exercise capacity in individual patients and identify alterations in gas exchange related to the pathophysiology of PAH and CTEPH that could explain the functional class and dyspnea of these patients. Considering that adaptive mechanisms are performed when living at different altitudes above sea level, we think these research data mainly apply to patients with PH who reside at high altitudes.

This study had several limitations, such as the retrospective design and the small sample size. Despite this, the patients in each group had a full evaluation and confirmation of the diagnosis at the institution’s pulmonary vascular disease group board using accepted diagnostic criteria. Even though DH has been linked to exertional dyspnea in some patients with PH [4, 35], we did not have inspiratory capacity and dyspnea measurements throughout the exercise to assess these dynamic changes.

Although at sea level, it has been established that there is a relationship between mortality in PH and some variables measured in CPET, such as peak VO${}_2$ and respiratory equivalents [4, 5], the results of these studies cannot be applied to patients who reside at high altitude due to the differences in the response to exercise, ventilatory efficiency and gas exchange variables related to the decrease in PB. For this reason, prospective studies in patients with PH are required to establish which physiological variables during exercise are related to mortality and which cut-off point has the best prognostic significance.

At high altitude, patients with PH present severe gas exchange alterations during exercise. Although there were no differences in hemodynamics at rest or in exercise capacity between patients with PAH and CTEPH, those with CTEPH had greater dyspnea, ventilatory inefficiency, and alterations in gas exchange during exercise. The CPET allowed the identification of these alterations related to the pathophysiology of the CTEPH that could explain the lower functional class and dyspnea in these patients.

AT, anaerobic threshold; BP, barometric pressure; CPET, cardiopulmonary exercise test; CTEPH, chronic thromboembolic pulmonary hypertension; FEV${}_1$, forced expiratory volume in 1 s; fR, respiratory frequency; FVC, forced vital capacity; HR, heart rate; MVV, maximal voluntary ventilation; PaCO${}_2$, carbon dioxide arterial pressure; Pa-ETCO${}_2$, arterial end-tidal carbon dioxide pressure gradient; PaO${}_2$, oxygen arterial pressure; PA-aO${}_2$, alveolar–arterial oxygen gradient; PAH, pulmonary arterial hypertension; PETCO${}_2$, end-tidal carbon dioxide pressure; PH, pulmonary hypertension; PIO${}_2$, inspired oxygen pressure; RER, respiratory exchange ratio; SaO${}_2$, arterial oxygen saturation; VCO${}_2$, carbon dioxide production; VO${}_2$, oxygen uptake; VO${}_2$/HR, oxygen pulse; VD/VT, dead space to tidal volume ratio; VE/VCO${}_2$, ventilatory equivalent for carbon dioxide; VE, minute ventilation; VT, tidal volume; WR, work rate.

The data set used for our analysis is available upon request from the corresponding author.

MGG, RCC, CRC, and ERA designed the research study. MGG analyzed the data. MGG, RCC, CRC, KD and ERA participated in the interpretation of data. MGG and ERA wrote the manuscript and all authors contributed to editorial changes. All authors participated sufficiently in the work and have agreed to be responsible for all aspects of it.

The Research Ethics Committee of the Fundación Neumológica Colombiana approved the study and the use of the anonymous data sets (approval number 202111-26803).

Not applicable.

This research received no external funding.

The authors declare no conflict of interest.