Is exercise-induced pulmonary hypertension ready for prime time in systemic sclerosis?


Two years ago the 4th World Symposium on Pulmonary Hypertension at Dana Point decided that exercise-induced pulmonary hypertension should be excluded from the diagnostic criteria for pulmonary arterial hypertension (1). Reasons for the decision included the perceived questionable reliability of the procedure, the lack of age-matched control data and the recognition that some athletes can develop exercise-induced pulmonary hypertension. However, several recent studies have shown that in scleroderma patients exercise catheterization can differentiate the haemodynamic profile in this at-risk patient population and demonstrate that cardiopulmonary disease is a real entity that may be missed by resting right heart catheterization.

Pathological studies show that pulmonary vascular disease is present in many scleroderma patients without evidence of resting PAH (2,3). An autopsy study looking at the morphometry of pulmonary arteries showed that there was significantly more intimal proliferation and medial thickening in scleroderma patients than those in controls (3). Those with PAH had the greatest amount of luminal occlusion but limited scleroderma patients who died of other causes also had striking increases in this finding. Even diffuse scleroderma had mild pulmonary vascular involvement which was more than controls. This intimal thickening correlated with disease duration in patients with limited cutaneous scleroderma.

In classic PAH associated with scleroderma, patients have a normal forced vital capacity but the diffusing capacity for carbon monoxide (DLCO) becomes severely decreased resulting in a high FVC%/DLCO% ratio (4). The DLCO is decreased in some idiopathic PAH patients but it rarely is as low as it is in scleroderma. Nevertheless, in IPAH the DLCO correlates with survival (5). Patients with interstitial fibrosis who develop pulmonary hypertension also have a low DLCO, but their forced vital capacity is also decreased. The DLCO is one of the best predictors of future pulmonary arterial hypertension in scleroderma with almost a linear decrease of the DLCO occurring over a 10–15 year period prior to the diagnosis of PAH (4). This finding is felt to correlate with the development of pulmonary vasculopathy. Studies are ongoing to determine if there are additional predictors of cardiopulmonary disease since not all patients with an isolated decrease of the DLCO develop PAH. One study has shown that 22% of high risk patients, those with a DLCO < 55% predicted, a FVC%/DLCO% ratio of > 1.6 or an estimated pulmonary pressure on echo of greater than 40 mmHg, developed PAH by 3 years (6).

Recent survival studies have shown that patients with SSc PAH have a worse prognosis than patients with idiopathic PAH (IPAH) (7), but they are often not diagnosed until they are NYHA/WHO symptom class III when potentially irreversible damage to the pulmonary vasculature and/or heart has occurred (8). However, since scleroderma is one disease that we know is an at risk population for PAH, it is critical that we identify and consider treatment before there is ‘end stage’ disease. One would not wait to treat coronary artery disease until someone has a heart attack, nor would we wait to control diabetes until there is renal failure. Scleroderma should be no different. We know that PAH is an incurable disease and the most common cause of scleroderma-related deaths (9). Survival is clearly better in patients with functional class I and II (10) so we must diagnose and treat these patients early.

Exercise-induced pulmonary hypertension on echo was shown to be increased in family members of patients with heritable pulmonary hypertension, and even though we know there is incomplete penetrance of the gene, it appears to be a risk factor in that disease process (11). There have been several recent studies showing that exercise-induced pulmonary hypertension is definitely present in scleroderma. An exercise echo study showed significant differences in the increase in estimated pulmonary artery systolic pressures with exercise in scleroderma patients compared with matched controls, but 13% of these patients had increased levels which were likely to represent exercise-induced pulmonary hypertension (12). While this study did not involve right heart catheterizations, other studies have shown a very good correlation of exercise induced pressures on echo compared with right heart catheterization. There have been very few false positives, although some patients do have pulmonary venous hypertension as the cause of their PH. An exercise right heart catheterization must be done to determine whether this increased PAP is from left heart disease (heart failure with normal systolic function, non-systolic heart failure, diastolic dysfunction), or whether it is true PAH. Steen, Saggar and Kovacs have all shown that there are significant numbers of high risk patients who have exercise-induced PAH on right heart catheterization (13–15). Those patients with exercise PAH had some increase in the pulmonary vascular resistance as you would expect in a PAH patient, suggesting that this is true pulmonary vasculopathy and not purely left heart disease (13–15). Steen showed that those with exercise-induced PAH had a very strong correlation with the other major risk factors, i.e. a very low DLCO and a high ratio of the FVC%/DLCO% suggesting that it is likely to represent early manifestations of PAH.

Many of our patients who are being evaluated for pulmonary hypertension are not symptomatic at rest. Rather, it is not until they exert themselves (particularly with stairs or inclines) that they manifest their dyspnea. While it certainly seems logical that we should take into account their exercise haemodynamics, this assumption is not without its controversy. The first controversy rests with the appropriate control values for both exercise echo and catheterization. Should all patients have a maximal exercise PCWP of less than 15? It is not clear, especially in those with advanced age. In the field of echocardiography, it is accepted that there is a normal process of cardiovascular ageing that results in impaired LV relaxation (as defined by E:A ratio). It is not known, however, if this normal cardiovascular ageing results in a similar ‘normal’ degree of diastolic dysfunction with exercise as patients reach their sixth or seventh decades of life.

The second issue with regard to exercise is exactly how the stress is performed. A seemingly simple question raises many further questions. Typically there are two approaches used in echo and catheterization labs: cycle ergometry (supine or upright) and arm weight exercise. Cycle ergometry has been the typical exercise in echo and catheterization studies since the early days of cardiopulmonary exercise testing (CPET). This technique allows for fine control of load and can be ramped in a smooth manner. Moreover, many believe that the cycle is more consistent with typical exertion patterns of patients by working the large lower extremity muscle groups. A relative disadvantage of this technique is the fact that is cannot be done with a groin approach for the right and/or left heart catheterization. Furthermore, the specialised bicycle is a relatively expensive piece of equipment and is not available to many physicians who wish to perform exercise as a routine part of their echo or catheterization protocols. Therefore, the use of arm weights has become a standard method of exercise in many centres. By having the patients hold 3–5 lb hand weights or 1 l saline bags, they are instructed to perform ‘butterfly’ arm exercises in which the outstretched arms are brought to an upright position repeatedly thus increasing heart rate and cardiac output. The advantage to this technique is that anyone can do it regardless of access type (neck vs. groin) and it is inexpensive. Disadvantages include the fact that it is not conducive to maximal exercise testing. This technique does not work well with CPET and thus it has been disregarded by physiologists focused on reaching anaerobic threshold and maximal exercise. But is the criticism justified? The answer depends on the goals of the study. Certainly, if the goal is to reach anaerobic threshold and maximal exercise, weighted arm exercise is not enough. However, if the goal is to increase heart rate and reproduce symptoms, then weighted arm exercise can accomplish this goal. Perhaps weighted arm exercise could be an initial screening study for smaller centres and if there is an increased level of suspicion for exercise-induced cardiopulmonary disease, then the patient can be referred to a larger centre for a more complete assessment. Only research to address this issue will answer this question.

Condiffe’s large study of connective tissue disease patients with PAH identified a group with exercise-induced pulmonary hypertension (10). They described forty-two of 259 patients with exercise-induced PAH, nineteen percentage went on to develop resting PAH, and four of the five who died during follow up died of PAH. Seventeen percent required advanced PAH therapy, suggesting that there is a subset of scleroderma patients whose exercise PAH is a risk factor for the future development of resting PAH. This group of patients should be carefully studied to determine if treatment improves symptoms, haemodynamics, and ultimately survival. Perhaps therefore, it is time to recognise that exercise pulmonary hypertension in scleroderma is an important, high risk, early manifestation of PAH. Identifying the disease as early as possible offers the best prospect of improvement for our patients.


Dr. Steen has received consulting payments from Pfizer and Gilead Pharmaceutical companies and research funds from Pfizer, Gilead, Actelion, and United Therapeutics for studies related to pulmonary hypertension and Raynaud’s in Scleroderma.