Exercise in Pulmonary Hypertension
In healthy subjects, exercise causes a complex cardiovascular and ventilatory response that allows for a 20-fold increase of oxygen consumption as compared to rest. In pulmonary hypertension (PH) this response is very much limited by a restriction of the maximum cardiac output (CO). This restriction is caused by a massively increased afterload of the right ventricle (RV) despite an adaptation that doubles right ventricular contractility as compared to normal controls.1
Citation: Advances in Pulmonary Hypertension 10, 4; 10.21693/1933-088X-10.4.237
If the exercise intensity or duration goes beyond these restrictions, the patient will faint because the CO is not high enough to keep the systemic arterial pressure sufficiently high. The systemic pressure is challenged because during exercise the muscle blood flow increases manifold, resulting in a massive drop in systemic vascular resistance. This clinical event, syncope, is well known to PH patients. It is one of the events that PH patients fear most once they have experienced it. In fact, repeated strenuous physical activity may contribute to the damage of the RV and the pulmonary arteries because exercise is accompanied by a massive increase in the pulmonary arterial pressure (PAP), which could cause injury and remodelling. This has long been known from the experience with mitral stenosis patients and was recently experimentally demonstrated in the monocrotaline rat.2
On the other hand, training may be good for PH patients because a higher level of fitness is associated with better quality of life and can mitigate depression and anxiety. It may also improve capillary density and thereby the utilization of peripheral blood flow.3 A controlled training program in German PH patients demonstrated a remarkable improvement in exercise capacity as measured by the 6-minute walk test.4 Those patients who participated in the training program and continued this training at home had a favorable prognosis.5 However, it is important to note that this study was performed by a highly specialized small team of physicians and physiotherapists, and the exercise levels were perfectly adapted to the individual patients by means of applying Doppler echocardiography during exercise. Although further investigation is clearly warranted, it appears that many PH patients are now being referred for a structured exercise program.
WHAT HAPPENS IN THE PULMONARY CIRCULATION DURING EXERCISE?
In healthy young subjects, CO increases 4-fold during exercise while PAP increases just 2-fold. This led to the speculation that a major vasodilatation caused by increased PAP (passive vasodilatation), increased blood flow (active vasodilatation), and recruitment of previously unperfused lung areas contributed to a massive decrease in pulmonary vascular resistance (PVR). However, under physiologic conditions there is no evidence of unperfused lung areas at rest (West Zone I) that could be recruited during exercise, and all the attempts to demonstrate an active vasodilatation in the pulmonary arteries have failed. There is no doubt that passive vasodilatation exists, but until recently it was an open question how much this decreases the resting PVR.
In healthy humans, PAP increases more or less linearly with the increase in CO. This is the conclusion from a literature review of all published right heart catheter investigations in healthy volunteers who have been studied at rest and exercise.6 The important detail is that the regression line crosses the y axis at about 7 mm Hg, which is consistent with a pulmonary arterial occlusion pressure in this range. However, this analysis did not answer the question of how much the PVR decreases during exercise. A recent analysis from our institution reviewed all published data that allowed calculation of individual PVR values from healthy subjects at different ages at rest and during exercise.7 The result was surprising. In the supine position there was a minor decrease in PVR from rest to exercise by about 12%. This was largely independent of age and gender. In the upright position there were fewer data available, but these suggested that during exercise PVR was very similar to the supine position while resting values were nearly doubled. This suggests a vasoconstrictive mechanism in the pulmonary arteries in the upright position that is active at rest and released during exercise.
WHY IS THE INCREASE IN PAP SO MUCH LOWER THAN THE INCREASE IN CO?
There are 2 factors explaining the moderate PAP increase during exercise in healthy subjects (Figure 1): 1) The regression line of PAP vs CO crosses the y axis considerably above zero; 2) the total pulmonary resistance (TPR) decreases by 25% during exercise. Both factors decrease the steepness of the PAP/CO relation. To really understand this, it is necessary to recalculate TPR.
Citation: Advances in Pulmonary Hypertension 10, 4; 10.21693/1933-088X-10.4.237
Total pulmonary resistance is typically calculated as TPR=PAP/CO, while the PVR is calculated as PVR=(PAP-PAWP)/CO=PAP/CO-PAWP/CO; correspondingly PVR=TPR-PAWP/CO and TPR= PVR+PAWP/CO. The term PAWP/CO describes how much filling pressure the left heart needs for a certain CO; ie, it describes the left ventricular filling resistance (LVFR): LVFR=PAWP/CO. In fact, TPR corresponds to the sum of the PVR and the filling resistance of the left heart.
Our literature research showed that LVFR was much lower in younger vs older individuals and that it decreased quite substantially in young individuals during moderate exercise levels, while in the elderly it remained on a higher level during moderate exercise and decreased only at the highest workloads. This suggests that LVFR contributes by two thirds and PVR by one third to the TPR decrease during exercise, except if the investigation is performed in the upright position.
WHAT DOES THIS MEAN FOR PH PATIENTS?
In patients with pulmonary arterial hypertension (PAH), PVR is probably increased both at rest and exercise, and its decrease during exercise is even smaller than in healthy individuals, while LVFR is probably always in the normal range (Figure 1). Unfortunately there are very few data to substantiate this view, although most PH patients undergo a right heart catheterization. But very rarely do they undergo exercise. Most data are from scleroderma patients who represent a risk population for PAH. They typically show a slightly elevated PVR that does not change significantly from rest to exercise, and they have a moderately elevated LVFR that does not respond to exercise in half of the patients.8 In addition, PH patients with scleroderma have decreased RV contractility as compared to IPAH patients as suggested by Overbeek et al.9
WHAT IS THE EXERCISE-LIMITING FACTOR IN PH PATIENTS?
Wonisch et al exercised PH patients who had an implanted pressure transducer in the right ventricle, measuring RV systolic pressure (RVSP) and RV dp/dt online.10 They found that in patients with compensated RV function, RVSP and dp/dt increased steeply at the start of exercise and flattened out at higher workloads, while decompensated patients had a flattened response from the start. This suggests that the flattening indicates an inadequate CO increase. Interestingly, in that small study, the peak RVSP was very different between patients as was the maximum RV dp/dt; however, peak ePAP; ie, the pressure during pulmonary valve opening, was very similar between patients. This suggests that ePAP might be associated with a signal that stops exercise before syncope occurs. However, it is more likely that the inadequate increase in CO causes not only a flattening of the PAP but also a flattening or even a decrease in the systemic arterial pressure; this physiology likely contributes to the patient's sense of when to stop exercising (prior to syncope).
ESTABLISHED EXERCISE TESTS FOR PH PATIENTS
6-Minute Walk
The 6-minute walk has become the most commonly used test for the evaluation of PH patients since 6-minute walk distance (6MWD) was consistently associated with prognosis in idiopathic PAH (IPAH) patients,1112 and the authorities accepted the change in 6MWD as the primary endpoint for studies testing targeted PAH drugs. The test is relatively easy to perform, has a high reproducibility, and can be applied to most patients. In children and young adults it has been shown that there is a linear relationship between the results of 6MWD and cardiopulmonary exercise testing (CPET) up to a walking distance of 300 m,13 but then the relationship plateaus because of the ceiling effect in 6MWD. Such a ceiling effect was already known from adult PH patients14 and limits the utility of 6MWD in patients with a well preserved exercise capacity. The other drawbacks are that the prognostic value of the 6MWD has only been shown in PH patients with severe IPAH but not chronic thromboembolic PH (CTEPH) or scleroderma patients with PAH, and results depend on the exertion of the patient. However, the rate of exertion (assessed by the Borg Scale), although measured, was mostly ignored in the interpretation of the results.15 Altogether this might imply an advantage of CPET over the 6-minute walk test. The other advantage is that the ventilatory reserves, cardiac reserves, and gas exchange abnormalities can be continuously monitored and taken into account.
Cardiopulmonary Exercise Testing
When Hugo W. Knipping described his apparatus for the “exact evaluation of gas exchange” in 1924,16 he believed that this new method would be suitable for hospital services and general medicine. Over the years he defined terms like “vita maxima,” which is now called peak VO2, “ventilatory reserve,” and “ventilatory equivalent,” which should help explain the exercise-limiting factors of healthy subjects and patients with heart and lung disease.
The CPET in patients with PH is one of the most difficult issues because a complex pathological mechanism interacts with a complex investigation. Although the current technology is highly sophisticated and provides automatic interpretation of the data, interpretation of CPET is still difficult. Cardiopulmonary exercise testing has been called the “gold standard” for the description of the cardiopulmonary system in PH, and typical patterns of gas exchange abnormalities have been described; however, if an individual patient with unexplained dyspnea undergoes CPET, there is still no standardization or criteria by which to diagnose PH.
WHAT IS KNOWN?
Pulmonary hypertension patients have, on average, a reduced peak VO2 and oxygen pulse, a low anaerobic threshold, and an increased ventilatory equivalent at the anaerobic threshold corresponding to an increased VE/VCO2 slope. Characteristically, end-tidal pCO2 is lower than normal and tends to decrease further during exercise.17 This indicates that the ventilation is inefficient. Some PH patients have a significant decrease in oxygen saturation during exercise, but this is very variable. We showed that among PH patients, those with CTEPH have even significantly lower end-tidal pCO2 values than IPAH patients, and that a big difference between end-tidal pCO2 and systemic arterialized capillary pCO2 values points to CTEPH rather than IPAH.18 However, such differences can also be increased by accompanying lung diseases, heart diseases, or shunt blood flow. Therefore, CPET does not replace specific diagnostics for CTEPH.
Cardiopulmonary exercise test results may be prognostically relevant for PH patients. Wensel et al showed that a very low peak VO2 (<10.4 mL kg-1 min-1) and a very low peak systemic systolic pressure (<120 mm Hg) were independent factors indicating a very bad prognosis.19 However, the poorer prognosis of these patients could also be suspected by clinical history, by physical examination, or by performing a 6-minute walk test. Indeed, these are the same patients that have a poor quality of life and a high WHO functional class. A group from the Netherlands has recently tested prognostic factors derived from exercise tests in a cohort of 115 PH patients with a 4-year follow-up.20 As expected, they found that patients with peak VO2 >13.2 mL kg-1 min-1, VE/VCO2 slope <48, O2 pulse increase >3.3 mL/beat, and 6MWD >399 m had a better prognosis than the other patients. However, the only factor that significantly added to the prognostic value of the 6MWD was the exercise-induced O2 pulse increase.
In PH patients at rest the O2 pulse is not significantly different from that of normal controls, whereas the increase during exercise is markedly reduced. This can be explained by the changes in stroke volume (SV) and arteriovenous oxygen difference (AVDO2). To understand this, it is necessary to examine mathematical relations of O2 pulse with SV and AVDO2.
This formula explains why O2 pulse increases so little in severe PH during the transition from rest to exercise. Severe PH causes a reduced resting CO. Because VO2 remains normal, the reduced CO causes an increased AVDO2 at rest. This partly compensates for the reduced SV and makes O2 pulse look relatively normal. During the transition to exercise, however, two factors contribute to a reduced O2 pulse increase. AVDO2 increases less than normal because it was already increased at rest, and SV increases less than normal because the right ventricle is compromised.
HOW DOES CPET INDICATE THE CIRCULATORY LIMITATION IN PH?
Patients with PH undoubtedly have a limited peak VO2 because their peak CO is limited due to a reduced SV during exercise. Correspondingly, the maximum heart rate should indicate completely exhausted cardiac reserves. In contrast, PH patients typically reach just about 60%–80% of their predicted peak heart rate. Is it possible that ventilation becomes the limiting factor? We have learned that ventilation is very inefficient in PH patients; on average, these patients just reach 70%–90% of their predicted peak ventilation, much below their predicted maximum voluntary ventilation. In addition, they have significantly reduced arterial and end-tidal pCO2 values. Only occasionally the oxygen saturation drops to a level that can be considered an exercise-limiting factor. Indeed, the “hyperventilation response to exercise” is the most typical feature of PH patients in CPET. However, the true limiting factor is probably a decrease in systemic arterial pressure shortly before exercise cessation.
CAN WE DIAGNOSE OR EXCLUDE PH BY MEANS OF CPET?
Unfortunately, diagnosing or excluding PH by means of CPET is not possible. When I started investigating PH patients by means of CPET, I was confident that I would find a pattern that is both highly sensitive and specific for PH. But peak VO2 may be reduced for multiple reasons, including cardiac or pulmonary disease, deconditioning, depression, and anxiety. The same is true for an increased ventilatory equivalent and a minor O2 pulse increase.
If lung disease has been excluded by an appropriate lung function test with assessment of the pulmonary diffusion capacity, and heart disease has been excluded by a thorough echocardiography including E/E'; if the patient is physically active, and the body weight is normal and there is no psychological disorder, then the above mentioned pattern in the spiro-ergometry would suggest PH with a very high positive predictive value. In daily practice, however, we are often unclear if there is really no heart disease. Left ventricular diastolic dysfunction may be very difficult to detect noninvasively, particularly if the patient has atrial fibrillation. Unfortunately, quite a number of patients with unexplained dyspnea are overweight and deconditioned and have some form of depression or anxiety. Therefore, the only standard to exclude PH and to find an explanation for unexplained dyspnea is the combination of CPET with right heart catheterization. This has been recently demonstrated in a case series of scleroderma patients with unexplained dyspnea.21
If a patient has a peak VO2 in the normal range, the oxygen saturation stays in the normal range during exercise and the O2 pulse increases adequately, PH is very unlikely. However, even then early forms of PH can be overlooked as they may present with completely normal responses in the CPET, although PAP and PVR are significantly elevated. Maybe the combination of CPET with Doppler echo-cardiography could improve the diagnostic sensitivity.22–24 Unfortunately, this technique is very demanding and few centers worldwide have enough experience with it. In the near future we still have to rely on multiple indicators of PH forming a more or less typical pattern. There is an unmet need for specific biomarkers of pulmonary vasculopathy. With such a tool we might even detect patients at risk for PH, and this would be the way to start treatment before PH is manifest.
Professor of Medicine
Director of the Pulmonary Division
Graz University Hospital
Medical University of Graz
Graz, Austria
Mean pulmonary arterial pressure vs cardiac output. Line represents healthy young individual. Broken line represents early PAH patient. Both have a resting pulmonary arterial wedge pressure of 7 mm Hg. Note the severe limitation of exercise cardiac output as compared to the resting cardiac output in the PAH patient.
Contributor Notes