Journal Logo

CLINICAL SCIENCES: Clinical Investigations

Impaired Systemic Oxygen Extraction at Maximum Exercise in Pulmonary Hypertension


Author Information
Medicine & Science in Sports & Exercise: January 2008 - Volume 40 - Issue 1 - p 3-8
doi: 10.1249/mss.0b013e318159d1b8
  • Free


Maximum exercise tolerance is reduced in patients with both pulmonary arterial (PAH) and venous hypertension. By the Fick principle, and in the absence of a pulmonary mechanical limit, maximum oxygen uptake (V˙O2peak) is dependent on peak exercise cardiac output (Qt), arterial oxygen content (CaO2), and systemic O2 extraction. In congestive heart failure, the inability to generate a normal maximum cardiac output is thought to limit V˙O2peak (22). Ambulatory heart failure patients do not usually exhibit arterial O2 desaturation, and whereas abnormalities of skeletal muscle capillary structure and reduced mitochondrial number and function have been described (6), the systemic extraction ratio (SER) at maximum exercise seems to be normal or even increased (22).

In PAH, the Fick principle determinants of impaired V˙O2peak have been less well studied. Sun et al. (23) have shown that abnormal surrogates of exercise cardiac output relate to the severity of resting pulmonary hypertension. Abnormal widening of the alveolar-arterial O2 difference at peak exercise is a hallmark of PAH, but in mild disease it does not materially decrease arterial O2 content (30).

Very little is known about systemic O2 uptake and use in PAH. Most clinical trials in PAH treatment have demonstrated improvements in constant-load exercise capacity, with little to no improvement in central hemodynamics (18). Similarly, V˙O2peak and walking for 6 min improve in PAH with pulmonary rehabilitation, without any change in pulmonary artery systolic pressure (15). These data suggest that a treatable abnormality of oxygen uptake and use by skeletal muscle may impair exercise performance in PAH. We hypothesize that abnormal systemic oxygen extraction associated with skeletal muscle dysfunction contributes to the exercise limit in PAH.



Complete data from one hundred forty-seven consecutive, clinically indicated cardiopulmonary exercise tests performed in the Massachusetts General Hospital Cardiopulmonary Exercise Laboratory between 2002 and 2004 were collected in database format and retrospectively reviewed. All patients signed a written informed consent. The tests were performed with radial and pulmonary arterial catheters in place, as well as first-pass radionuclide ventriculographic scanning. Ninety-three patients who met criteria for an exercise limit, as will be defined below, because of PAH, left ventricular (LV) diastolic dysfunction (DD), or LV systolic dysfunction (SD), were included. Patients who did not meet these criteria, or who had a primary pulmonary mechanical limitation to exercise, were excluded. The study was approved by the Partners Human Research Council.

Cardiopulmonary exercise test.

Radial and pulmonary arterial catheters were inserted percutaneously. Systemic and pulmonary artery pressures were measured with HP1290A quartz pressure transducers (Hewlett-Packard Co., Andover, MA). Transducers were interfaced with an MT95K2 recorder (Astro-Med Inc., W. Warwick, RI), and mean end-expiratory values were obtained for right atrial, mean pulmonary arterial (mPAP), and mean systemic arterial pressure. Two-milliliter samples of systemic and pulmonary arterial blood were obtained at rest and during exercise and were analyzed for PO2, PCO2, pH (model 1620; Instrumentation Laboratories, Lexington, MA), hemoglobin concentration ([Hb]), and O2 saturation and content (model 482; Instrumentation Laboratories).

Pulmonary gas exchange and minute ventilation (VE) were measured breath-by-breath, using a commercially available metabolic cart (Model CPX/D; Medical Graphics Corporation (MGC), St. Paul, MN). The pneumotachograph was calibrated using a 3-L syringe at five different flow rates. A zirconia cell O2 analyzer and single-beam infrared CO2 analyzer were calibrated with room air and a 5% CO2/12% O2 gas. Maximum voluntary ventilation (MVV) was calculated as FEV1 × 35 (3).

Exercise protocol.

All patients completed a single bout of incremental cycle ergometer (CPE 2000; MGC) exercise. Two minutes of rest were followed by 2 min of unloaded cycling. Work rate was then continuously increased by 6.25-25 W·min−1 on the basis of history of exertional tolerance. Mean arterial pressure, right atrial pressure, and mPAP were measured continuously; pulmonary capillary wedge pressure (PCWP) was obtained at rest and at each minute of exercise. Two-milliliter blood samples were simultaneously drawn from the radial and pulmonary arterial catheters during rest and for the last 15 s of each minute of exercise. At the cessation of exercise, patients were asked which of the following symptoms caused them to stop: shortness of breath, leg fatigue or pain, chest pain, or a combination of the above. No exercise test was stopped by the supervising technician or physician; all tests were stopped by the patients themselves.

Data analysis.

Resting ventilatory and gas-exchange data were averaged from the final 30 s of the 2-min rest period. Exercise ventilatory and gas-exchange data were averaged for contiguous 30-s intervals. Alveolar-arterial O2 difference was calculated from standard formulae.

Values for V˙O2peak, predicted using age, gender, and height, were those of Hansen and colleagues (8). Qt was calculated from the Fick principle Qt = V˙O2/(Ca-v¯O2). Predicted maximal Qt was calculated from predicted V˙O2peak and an assumed maximal arterial-venous O2 content difference = ([Hb] × 10) (27). Oxygen delivery (DO2) was determined by Qt × CaO2. Systemic oxygen extraction ratio (SER) was determined by (Ca-v¯O2)/CaO2.

At maximum exercise, PAH was defined as mPAP ≥ 30 mm Hg, PCWP < 20 mm Hg, and PVR ≥ 80 dyn·s·cm−5. LV DD was defined as PCWP ≥ 20 mm Hg, LV ejection fraction (LVEF) ≥ 0.55, and no evidence of mitral or aortic valve regurgitation by increased LV/RV stroke count ratio or increased radionuclide/Fick Qt. LV SD was defined as PCWP ≥ 20 mm Hg and LVEF < 0.55. Where indicated, lumped DD and SD data are referred to as pulmonary venous hypertension. Peak heart rate ≥ 80% predicted, or peak respiratory exchange ratio (RER = VCO2/V˙O2) ≥ 1.00, were used as indicators of maximum effort. Pulmonary mechanical limitation was defined as a VE/MVV ≥ 0.7 at peak exercise.

Central tendencies and variabilities were expressed as means ± standard deviations and compared either by ANOVA with Neumann-Keuls finishing test for continuous variables, or by Fisher's exact test for binomial variables. At peak exercise, SER was analyzed by linear regression as a function of cardiac output for each group and compared by ANCOVA with a Tukey's finishing test. Statistical analysis was performed using XLstat Pro (Addinsoft) and GraphPad Prism 5 (GraphPad Software). A P value < 0.05 was considered significant.



All tests were clinically indicated and were performed either for chronic (> 3 months) dyspnea of unclear origin or for preoperative evaluation for heart or lung transplantation. Medication history was collected; no medications were held as part of the exercise testing protocol. Patient demographics are summarized in Table 1.

Patient demographics.

Maximum exercise capacity and gas exchange.

All patients completed a symptom-limited bout of exercise, and 91 of 93 patients demonstrated adequate maximum effort, as defined by heart rate ≥ 80% predicted or RER ≥ 1.00. All patients stopped either because of shortness of breath, leg fatigue, or both, with four patients (one PAH, three DD, zero SD) additionally experiencing chest pain. No patients stopped for chest pain alone.

During rest, mPAP was 21.6 ± 9.2, 18.1 ± 3.8, and 27.0 ± 10.1 mm Hg, PCWP was 7.2 ± 2.8, 8.3 ± 2.4, and 14.8 ± 7.4 mm Hg, and PVR was 283 ± 212, 184 ± 58, and 278 ± 145 dyn·s·cm−5 in PAH, DD, and SD, respectively (P < 0.05 among groups for each). At peak exercise, V˙O2 percentage predicted, Qt percentage predicted, and DO2 for SD were reduced in comparison with PAH and DD (Table 2). PAH patients had a lower maximum exercise SaO2 than pulmonary venous hypertension (Table 2). SER at peak exercise was most impaired in PAH, highest in SD, and intermediate in DD (Fig. 1).

Fick principle and gas-exchange variables at peak exercise.
Systemic O2 extraction at maximum exercise. SER, systemic extraction ratio; PAH, pulmonary arterial hypertension; DD, diastolic dysfunction; SD, systolic dysfunction. * P < 0.05 vs PAH; † P < 0.05 vs DD.


Mean arterial pressure was higher in PAH than in SD (Table 3). There was no significant difference in systemic vascular resistance among groups. Mean PAP was higher in SD than PAH, and PCWP was higher in both DD and SD than in PAH. There was no difference in PVR between SD and PAH.

Hemodynamics at peak exercise.

Correlations between systemic O2 extraction and central hemodynamics.

At maximum exercise, SER was inversely related to Qt for both SD and PAH (P < 0.05) (Fig. 2), but not for DD (P > 0.05). By ANCOVA, SER was lower for PAH compared with SD at any given Qt (Fig. 2). There were insignificant relationships between maximum SER and right atrial pressure, mPAP, PCWP, and PVR for each group.

Systemic extraction ratio (SER) vs peak exercise cardiac output (Qt) percentage predicted for pulmonary arterial hypertension (PAH) and systolic dysfunction (SD). PAH: P < 0.05, r = 0.48; SD: P < 0.05, r = 0.49; PAH vs SD: P < 0.05 by ANCOVA.


Impaired systemic O2 extraction in PAH.

The Fick principle variables' relative contributions to decreased V˙O2peak in PAH, DD, and SD are different. The current study confirms that for SD, the rate-limiting step for O2 transport is an impaired Qtmax, whereas arterial O2 content is normal, and SER is actually increased. In a teleologic sense, increased O2 uptake and use in the face of impaired O2 delivery is an adaptive response meant to preserve aerobic adenosine triphosphate production by the exercising skeletal muscle.

In our patients with PAH, a different pattern was found. At peak exercise, Qtmax and DO2 were preserved in comparison with SD, whereas arterial O2 content was not different. Conversely, systemic O2 extraction was impaired. The inverse relationship between SER and Qt in PAH, however, suggests some ability of the periphery to respond to reductions of whole-body O2 delivery, in a similar fashion to SD.

One possible explanation for these findings is that the relative preservation of DO2 in mild PAH did not require a compensatory increase in O2 extraction. The SERmax/Qtmax relation was shifted throughout the entire range of maximum Qt for PAH versus SD, however, suggesting pathologically impaired O2 extraction.

Another possible explanation is that patients with PAH terminate exercise prematurely because of severe symptoms of shortness of breath or chest pain, and, thus, they may have been unable to elicit full and potentially normal O2 extraction. However, nearly all patients had evidence of adequate maximum effort and did not experience chest pain. These findings suggest that impaired O2 extraction is unique to PAH among the three groups, and that it is pathologic.

Three distinct, although potentially related, mechanisms could account for reduced SERmax in PAH: skeletal muscle perfusion/oxidative metabolism mismatch, impaired O2 offloading in the systemic capillary, and an intrinsic abnormality of the skeletal muscle mitochondrion.

Perfusion/oxidative metabolism mismatch.

Exercising skeletal muscle capillary beds dilate in the face of increased muscle sympathetic nerve activity as a result of local vasodilators. In the current study, systemic vascular resistance, a surrogate for muscle sympathetic nerve activity that is known to be excessive in congestive heart failure, and that has more recently been described as abnormal in PAH (25), was not different among groups, implying that a global increase in sympathetic vascular tone was not responsible for the decreased SER in PAH.

If global perfusion is not to be implicated, a potential explanation might relate to regional mismatch in perfusion and oxidative metabolism. Using modeling techniques, Walley (26) suggests that microscopic peripheral mismatches between oxygen supply and demand worsen extraction. Kalliokoski et al. (11) have linked enhanced O2 extraction to an endurance training-induced decrease in flow heterogeneity in exercising muscle.

Is skeletal muscle perfusion heterogeneity in PAH biologically plausible? Donato and coworkers (4) have recently demonstrated differing susceptibilities to systemic vasoconstrictors of resistance vessels feeding oxidative versus glycolytic skeletal muscle in rats. Slow-twitch soleus muscle vessels constricted more in response to endothelin (ET-1), whereas those perfusing fast-twitch gastrocnemius were more sensitive to norepinephrine and angiotensin-II. This may be clinically relevant, because resting humans with PAH have elevated blood concentrations of ET-1 (16), which may correlate directly with disease severity (29). Woodman et al. (28) have demonstrated that arterioles from detrained slow-twitch muscle fibers in rats had less endothelial nitric oxide synthase (eNOS) expression and a reduced hyperemic response to acetylcholine than the arterioles of controls, and fast-twitch muscle arterioles showed neither of these changes. Because slow-twitch fibers contribute more to V˙O2peak than their fast-twitch counterparts (17), ET-1 or nitric oxide (NO)-mediated compromise of oxidative skeletal muscle blood flow could reduce exercise capacity in PAH. Thus, regional impairment of slow-twitch skeletal muscle fiber perfusion could account for the principal findings of this study.

Systemic capillary O2 offloading.

Is it possible that exercise-induced arterial hypoxemia could have impaired systemic oxygen extraction in PAH? Despite a widened A-a O2 difference in PAH, there is no difference in CaO2 among groups at maximum exercise. Thus, by the Fick principle, arterial hypoxemia could not have materially reduced V˙O2max by convective means. Alternatively, Hogan et al. (10) suggest that a reduced PaO2 can decrease the diffusive transfer of O2 from the systemic capillary red blood cell to the skeletal muscle mitochondrion, compromising V˙O2max. However, this only occurs at a PaO2 < 40 mm Hg, and our PAH patients' mean PaO2 at maximum exercise was 79.0 mm Hg. Therefore, it is unlikely that the mild arterial hypoxemia at maximum exercise observed in this study accounted for impaired systemic O2 extraction in PAH.

Normal unloading of O2 in the systemic capillary is dependent on a rightward shift of the oxyhemoglobin-dissociation curve attributable to acidosis (21) and NO-facilitated conformational changes of hemoglobin (12). The mean mixed venous pH was not different at peak exercise between PAH and SD. We did not make any surrogate measures of systemic NO in the current study, but pulmonary eNOS is known to be reduced in PAH (7). We cannot rule out decreased systemic capillary NO as a reason for impaired SER in PAH; this is the subject of an ongoing study.

Mitochondrial dysfunction.

At peak exercise, the hallmark of a defined mitochondrial myopathy is impaired O2 extraction in the face of normal delivery (24). A growing body of 31P-magnetic resonance spectroscopy (1,13) and muscle biopsy literature (1,14) suggests that skeletal muscle's oxidative capacity is reduced in chronic lung disease. To our knowledge, there are no such studies in PAH.

Limitations of the study.

It might be argued that whole-body exercise, compared with that of small muscle groups, does not allow the inference that the extraction problem in PAH is at the level of the limb skeletal muscle. In fact, a "steal" phenomenon, diverting blood flow from locomotor to more oxidative respiratory muscles, has been suggested for competitive athletes (9) and patients with COPD breathing close to their maximum voluntary ventilation (20). In contrast to the latter groups, no patients in this study had evidence of a primary pulmonary mechanical limit.

One advantage of whole-body exercise is to determine whether reductions of SER in PAH produce a clinically relevant decrease in V˙O2peak. In addition, if PAH is a systemic disease, whole-body exercise and its attendant increased pulmonary blood flow, pressure, or shear stress may be necessary to evoke changes in humoral vasoconstrictors or vasodilators relevant to limb skeletal muscle perfusion. Exercising small muscle groups will miss the very pathophysiology responsible for the patient's exercise limit and, presumably, symptoms.

The presence of left-to-right intracardiac shunting in PAH could also confound these results, because it could falsely decrease SER by elevating mixed venous O2 content. PAH patients, however, had a Qtmax ratio of ventriculography/Fick calculation of 1.14, which suggests an absence of significant shunting.

Another possible limitation to our findings is that the patients within the three groups were not entirely homogenous in terms of their demographics, including medication use. The patients with SD used more beta-blockers and renin-angiotensin antagonists than those with PAH and DD. Beta-blockers (2) and renin-angiotensin antagonists (5) have both been shown to modestly increase systemic oxygen extraction during exercise in patients with LV disease. However, the frequency of use of these medications was not different in PAH versus DD, suggesting that medical treatment did not account for the principal findings of this study.

Poor effort at maximum exercise could also blunt systemic oxygen extraction. Our indices of maximum effort, peak heart rate > 80% predicted or RER > 1.00, were relatively liberal. However, we have no evidence that the effort expended by the PAH group was less than those with SD and DD, based on peak heart rate and RER. Furthermore, because oxygen extraction during incremental exercise is hyperbolic (19), the effect of marginally decreased effort should have minimal impact on systemic oxygen extraction.

Finally, the PAH and SD populations that were studied are not typical of some previous studies. Our PAH patients had a mildly reduced V˙O2peak, preserved Qtmax, and only moderate elevations in mPAP. Our SD patients had the expected decrements in V˙O2peak and Qtmax, but they also had significant mean PAP elevation and elevated PVR. These findings are likely related to referral patterns, with PAH patients often being evaluated for dyspnea, and SD patients being evaluated for cardiac transplantation. Comparing mild PAH with the more severe pulmonary hypertension of LV SD, however, allows the inference that elevated central pressures, per se, are not responsible for impaired SER.


V˙O2peak is reduced more by an abnormality of O2 uptake and use than by its delivery in mild PAH. This may suggest that PAH is more of a systemic disease than has been previously recognized.

Aaron Waxman is supported by HL074859, and David Systrom is supported by HL04022-05. Special thanks to Paul Pappagianopoulos, MSc, for his technical assistance in the MGH Cardiopulmonary Exercise Laboratory.


1. American Thoracic Society and European Respiratory Society Statement. Skeletal muscle dysfunction in chronic obstructive pulmonary disease. Am J Respir Crit Care Med. 1999;159:S2-40.
2. Bruce RA, Hossack HF, Kusumi F, Clarke LJ. Acute effects of oral propranolol on hemodynamic responses to upright exercise. Am J Cardiol. 1979;44:132-40.
3. Clark TJ, Freedman S, Campbell EJ, Winn RR. The ventilatory capacity of patients with chronic airways obstruction. Clin Sci. 1969;36:307-16.
4. Donato AJ, Lesniewski LA, Delp MD. Skeletal muscle arteriolar vasoconstrictor reactivity to local, humoral, and neural agonists are differentially affected by muscle fiber type and exercise training. Physiologist. 2004;47:278.
5. Drexler H, Banhardt U, Meinertz T, Wollschlager H, Lehmann M, Just H. Contrasting peripheral short-term and long-term effects of converting enzyme inhibition in patients with congestive heart failure. A double-blind, placebo-controlled trial. Circulation. 1989;79:491-502.
6. Drexler H, Riede U, Munzel T, Konig H, Funke E, Just H. Alterations of skeletal muscle in chronic heart failure. Circulation. 1992;85:1751-9.
7. Giaid A, Saleh D. Reduced expression of endothelial nitric oxide synthase in the lungs of patients with pulmonary hypertension. N Engl J Med. 1995;333:214-21.
8. Hansen JE, Sue DY, Wasserman K. Predicted values for clinical exercise testing. Am Rev Respir Dis. 1984;129:S49-55.
9. Harms CA, Babcock MA, McClaran SR, et al. Respiratory muscle work compromises leg blood flow during maximal exercise. J Appl Physiol. 1997;82:1573-83.
10. Hogan MC, Roca J, Wagner PD, West JB. Limitation of maximal O2 uptake and performance by acute hypoxia in dog muscle in situ. J Appl Physiol. 1988;65:815-21.
11. Kalliokoski KK, Oikonen V, Takala TO, Sipila H, Knuuti J, Nuutila P. Enhanced oxygen extraction and reduced flow heterogeneity in exercising muscle in endurance-trained men. Am J Physiol Endocrinol Metab. 2001;280:E1015-21.
12. Kosaka H, Seiyama A. Increased oxygen dissociation by nitric oxide from RBC. Adv Exp Med Biol. 1997;428:349-54.
13. Kutsuzawa T, Shioya S, Kurita D, Haida M, Ohta Y, Yamabayashi H. Muscle energy metabolism and nutritional status in patients with chronic obstructive pulmonary disease. A 31P magnetic resonance study. Am J Respir Crit Care Med. 1995;152:647-52.
14. Maltais F, Simard AA, Simard C, Jobin J, Desgagnes P, LeBlanc P. Oxidative capacity of the skeletal muscle and lactic acid kinetics during exercise in normal subjects and in patients with COPD. Am J Respir Crit Care Med. 1996;153:288-93.
15. Mereles D, Ehlken N, Kreuscher S, et al. Exercise and respiratory training improve exercise capacity and quality of life in patients with severe chronic pulmonary hypertension. Circulation. 2006;114:1482-9.
16. Michel RP, Langleben D, Dupuis J. The endothelin system in pulmonary hypertension. Can J Physiol Pharmacol. 2003;81:542-54.
17. Pringle JS, Doust JH, Carter H, Tolfrey K, Campbell IT, Jones AM. Oxygen uptake kinetics during moderate, heavy and severe intensity "submaximal" exercise in humans: the influence of muscle fibre type and capillarisation. Eur J Appl Physiol. 2003;89:289-300.
18. Rich S. The current treatment of pulmonary arterial hypertension: time to redefine success. Chest. 2006;130:1198-202.
19. Richardson RS, Poole DC, Knight DR, et al. High muscle blood flow in man: is maximal O2 extraction compromised? J Appl Physiol. 1993;75:1911-6.
20. Simon M, LeBlanc P, Jobin J, Desmeules M, Sullivan MJ, Maltais F. Limitation of lower limb VO2 during cycling exercise in COPD patients. J Appl Physiol. 2001;90:1013-9.
21. Stringer W, Wasserman K, Casaburi R, Porszasz J, Maehara K, French W. Lactic acidosis as a facilitator of oxyhemoglobin dissociation during exercise. J Appl Physiol. 1994;76:1462-7.
22. Sullivan MJ, Knight JD, Higginbotham MB, Cobb FR. Relation between central and peripheral hemodynamics during exercise in patients with chronic heart failure. Muscle blood flow is reduced with maintenance of arterial perfusion pressure. Circulation. 1989;80:769-81.
23. Sun XG, Hansen JE, Oudiz RJ, Wasserman K. Exercise pathophysiology in patients with primary pulmonary hypertension. Circulation. 2001;104:429-35.
24. Taivassalo T, Jensen TD, Kennaway N, DiMauro S, Vissing J, Haller RG. The spectrum of exercise tolerance in mitochondrial myopathies: a study of 40 patients. Brain. 2003;126:413-23.
25. Velez-Roa S, Ciarka A, Najem B, Vachiery JL, Naeije R, van de Borne P. Increased sympathetic nerve activity in pulmonary artery hypertension. Circulation. 2004;110:1308-12.
26. Walley KR. Heterogeneity of oxygen delivery impairs oxygen extraction by peripheral tissues: theory. J Appl Physiol. 1996;81:885-94.
27. Wasserman K, Hansen JE, Sue DY, Stringer WW, Whipp BJ. Principles of Exercise Testing and Interpretation. 4th ed. Philadelphia (PA): Lipincott Williams & Wilkins; 2005.
28. Woodman CR, Schrage WG, Rush JW, et al. Hindlimb unweighting decreases endothelium-dependent dilation and eNOS expression in soleus not gastrocnemius. J Appl Physiol. 2001;91:1091-8.
29. Yamakami T, Taguchi O, Gabazza EC, et al. Arterial endothelin-1 level in pulmonary emphysema and interstitial lung disease. Relation with pulmonary hypertension during exercise. Eur Respir J. 1997;10:2055-60.
30. Yasunobu Y, Oudiz RJ, Sun XG, Hansen JE, Wasserman K. End-tidal PCO2 abnormality and exercise limitation in patients with primary pulmonary hypertension. Chest. 2005;127:1637-46.


© 2008 American College of Sports Medicine