Objectives: Primary pulmonary hypertension (PPH) is a pulmonary vasculopathy resulting in exercise intolerance, usually due to dyspnea. We hypothesized that venlilation is increased during exercise in PPH relative to normal because the ventilated lung is underperfused, cardiac output increase is restricted, and arterial hypoxemia may develop. Our aim was to determine the size of the reduction in end-tidal PC[O.sub.2] (PETC[O.sub.2]) as a reflection of the abnormality in ventilatory efficiency and ventilatory drive in PPH patients.
Methods: We performed cardiopulmonary exercise testing (CPET) in 52 PPH patients. All had hemodynamic measurements to confirm the diagnosis of PPH. A subgroup of 29 patients who underwent right-heart catheterization within 50 days of CPET were studied to compare their CPET responses to resting hemodynamics. Nine healthy volunteers matched for age and gender served as CPET control subjects.
Results: In PPH patients, the percentage of predicted peak oxygen uptake (V[O.sub.2]) correlated significantly with mean pulmonary artery pressure (mPAP) [r = - 0.59, p = 0.0007, n = 29]. PETC[O.sub.2] values at rest, anaerobic threshold (AT), and peak V[O.sub.2] were proportionately reduced as percentage of predicted peak V[O.sub.2] decreased (r = 0.66 to 0.72, p < 0.0001, n = 52). PETC[O.sub.2] values at rest, AT, and peak V[O.sub.2] were also reduced as mPAP increased (r = - 0.51 to - 0.53, p < 0.005, n = 29). In contrast to normal subjects in whom PETC[O.sub.2] increased from rest to AT, PETC[O.sub.2] decreased in PPH patients, except for two patients with mild PPH in whom there was no change. Also, PETC[O.sub.2] increased rather than decreased further at the start of recovery, in contrast to normal. Although usually normal at rest, oxyhemoglobin saturation decreased during exercise in most PPH patients.
Conclusions: In patients with PPH, PETC[O.sub.2] at rest and exercise is significantly reduced in proportion to physiologic disease severity. The range of values is unusually low. Furthermore, the directional changes of PETC[O.sub.2] during exercise and early recovery are in the opposite direction of normal.
Key words: anaerobic threshold; end-tidal C[O.sub.2]; mean pulmonary artery pressure; peak oxygen uptake; primary pulmonary hypertension; ventilatory equivalent for C[O.sub.2]
Abbreviations: ANOVA = analysis of variance; AT = anaerobic threshold; CPET = cardiopulmonary exercise testing; EIS = exercise-induced right-to-left shunt; mPAP = mean pulmonary artery pressure; NYHA = New York Heart Association; [O.sub.2]sat = oxyhemoglobin saturation; PAIC[O.sub.2] = ideal alveolar PC[O.sub.2]; PETC[O.sub.2] = end-tidal carbon dioxide tension; PPH = primary pulmonary hypertension; VC[O.sub.2] = carbon dioxide output; VD/VT = physiological dead space/ tidal volume- VE = minute ventilation. VE/VC[O.sub.2] = ventilatory equivalents for carbon dioxide; V[O.sub.2] = oxygen uptake
Primary pulmonary hypertension (PPH) is a progressive disease characterized by a pulmonary vasculopathy that increases pulmonary vascular resistance, eventually leading to right-heart failure and death. (1,2) Rich et al (1) reported that the mean interval from the onset of symptoms to diagnosis was 2 years. Although the earliest symptoms are usually dyspnea and/or fatigue during exercise, the earliest physiologic variables reflecting exercise intolerance are unclear. While exercise intolerance provides the clinical clue that pulmonary hypertension might account for the symptoms, pulmonary hypertension during right-heart catheterization, without evidence of left ventricular failure or primary lung disease, is needed to make a definite diagnosis.
Cardiopulmonary exercise testing (CPET) has been shown to be useful in assessing the severity and prognosis of PPH. (3-5) Symptoms develop during exercise because recruitment of pulmonary vascular bed needed for exercise is impaired. The symptoms and clinical course can be attributed to three easily identified pathophysiologies: (1) failure to perfuse the ventilated lung, thereby increasing the physiologic dead space and ventilatory requirement (3,6,7); (2) failure to increase cardiac output ([O.sub.2] transport) appropriately in response to exercise, causing a low work rate lactic acidosis (increased C[O.sub.2] production relative to 0,2 consumption), thereby increasing acid ventilatory drive; and (3) exercise-induced hypoxemia in most PPH patients, thereby increasing the hypoxic ventilatory drive. We hypothesized that since the key pathophysiologies in PPH increase ventilatory drive (lower the PaC[O.sub.2] set point) or reduce gas exchange efficiency (increase physiologic dead space), end-tidal carbon dioxide tension (PETC[O.sub.2]) should be decreased and the ventilatory equivalent for carbon dioxide (VE/VC[O.sub.2]) should be increased in proportion to the physiologic severity of the disease.
PETC[O.sub.2] is a measurement made at the airway during exercise without requiring any special calculations. Our goal was to determine the magnitude of the reduction in PETCO2 at rest, at anaerobic threshold (AT) [before the ventilatory compensation for exercise-induced lactic acidosis] at peak oxygen consumption (V[O.sub.2]), and early recovery in patients with a PPH diagnosis. Based on our earlier studies in which peak V[O.sub.2] was shown to correlate significantly with New York Heart Association (NYHA) symptom class in patients with PPH, (3) and its success as a prognostic indicator of survival in chronic left ventricular failure, (8) we used percentage of predicted peak V[O.sub.2] to grade physiologic severity in the patients used in this study. In addition, we correlated PETCO2 to mean pulmonary artery pressure (mPAP) as a measure of pulmonary vascular disease in a subgroup of patients in whom right-heart catheterization study was done within 50 days of the CPET study. Our studies demonstrate that PETCO2 is surprisingly low in patients with PPH, as compared to normal subjects and to previously reported patient groups. (9-11) In addition, the pattern of change in PETCO2 in response to exercise and recovery is abnormal.
MATERIALS AND METHODS
We retrospectively investigated the exercise pathophysiology in 52 patients with PPH referred for evaluation and treatment to our Pulmonary Hypertension Referral Clinic, and 9 healthy volunteers of similar distribution in age and gender (3 men and 6 women; mean age, 39.9 [+ or -] 4.8 years [mean [+ or -] SE]). The diagnosis of PPH was based on clinical findings and the diagnostic criteria described by the National Institutes of Health registry for PPH and the World Health Organization. (1) The Human Subjects Committee of our institution approved the study.
The first CPET performed on each patient was used in this trial. Therefore, most patients (39 of 52 patients) were not yet receiving pulmonary vasodilator therapy. Seven patients were receiving an IV-administered prostacyclin analog, three patients were receiving a subcutaneously administered prostacyclin analog, and three patients were receiving an endothelin receptor blocker. All patients were receiving warfarin.
Following an explanation of the exercise study and related procedures, the patient performed a physician-supervised, symptom-limited, progressively increasing exercise test on an electro-magnetically braked, upright cycle ergometer (Medical Graphics; St. Paul, MN). Although the test was unencouraged, we advised patients to do their best, but they could voluntarily stop cycling at any time that they believed that they could not continue. We monitored heart rate, BP, ECG, and gas exchange, breath-by-breath. All patients were in sinus rhythm. In the few patients in whom ventricular or atrial premature beats developed during exercise, the arrhythmia disappeared soon after the beginning of recovery. There were no complications during CPET in any patient.
The protocol consisted of 3 min of rest, 3 min of unloaded (freewheeling) cycling at 60 revolutions per minute, followed by a progressively increasing work rate of 5 to 15 W/min (44 of 52 patients were 10 W/min) to the maximum tolerance, and 2 min of recovery. Twelve-lead ECGs were continuously monitored. ECGs and arterial BP, measured with an automatic cuff manometer, were recorded every 2 min. Oxyhemoglobin saturation (O2sat) determined by pulse oximetry was continuously recorded. Gas exchange measurements were computer calculated breath-by-breath and averaged over 15-s intervals (11) using a diagnostic system (Cardiorespiratory Diagnostic System; Medical Graphics). The volumes of the flowmeter and mouthpiece (50 mL x breathing frequency) were subtracted from minute ventilation (VE) for the VE/VC[O.sub.2] calculations. AT was determined by the V-slope method. (11)
Patients were classified into four groups of physiologic disease severity based on the reduction in percentage of predicted peak V[O.sub.2] as previously reported (3): mild (65 to 79% predicted), moderate (50 to 64% predicted), severe (35 to 49% predicted), and very severe (< 35% predicted). The division of physiologic grading was based on the correlation of percentage of predicted peak V[O.sub.2] with the NYHA symptom class. (3) PETC[O.sub.2] was related to percentage of predicted peak V[O.sub.2] in all 52 patients and the control subjects, and to mPAP in the 29 patients who underwent right-heart catheterization diagnostic studies within 50 days of their first CPET.
The predicted peak V[O.sub.2] and AT were calculated as previously described. (11) Exercise PETC[O.sub.2] values were compared using repeated-measures analysis of variance (ANOVA). Data among the groups were compared using one-way ANOVA and the Scheffe multiple comparison test. Unpaired t tests were used to compare two groups. Values of p < 0.05 were considered significant. Pearson correlation coefficients were determined using the least-square method. Averaged data were expressed as mean [+ or -] SE.
Patient and control subject characteristics are shown in Table 1. There were 7 men and 45 women (age, 43.5 [+ or -] 1.8 years) enrolled in the patient group and 3 men and 6 women enrolled in the control group. The patient group was further divided into four groups according to the physiologic grade of severity defined by percentage of predicted peak V[O.sub.2]. (12) The percentage of predicted AT and VE/VC[O.sub.2] at AT were progressively more abnormal as physiologic severity increased (Table 1). The average respiratory exchange ratio (RER) at peak exercise ranged between 1.15 and 1.18 and was not significantly different as related to physiologic severity (Table 1).
Change in PETC[O.sub.2] During Exercise Testing
The kinetics of changes in PETC[O.sub.2] for progressively increasing work rate are shown in Figure 1 for a representative normal subject and for six PPH patients with disease of differing severity and exercise-induced right-to-left shunt (EIS) effect. PETC[O.sub.2] typically increases during exercise from rest to AT in normal subjects (Fig 1, dashed line) and then decreases when ventilatory compensation for the exercise lactic acidosis begins. It then decreases further in recovery. In contrast, PETC[O.sub.2] is reduced at rest and is decreased further during exercise in PPH patients (Fig 1, Moderate, Severe, and Very Severe panels), except for the mildly impaired patient (Fig 1, Mild panel; peak V[O.sub.2], 74% of predicted). The decrease in PETC[O.sub.2] was slow during exercise if the patient had no right-to-left shunt (thin continuous lines, Fig 1) and abrupt if the patient had a right-to-left shunt (heavy continuous lines, Fig 1) as previously described. (12) In the immediate recovery period, PETC[O.sub.2] increased in the PPH patients, in contrast to control (normal) subjects, in whom PETC[O.sub.2] decreases in early recovery.
[FIGURE 1 OMITTED]
The mean values for PETC[O.sub.2] at rest, AT, peak, and 2 min of recovery are shown for each level of physiologic impairment in the PPH subjects and for the control subjects in Figure 2. The values get progressively lower as related to increasing severity. The mean values for each PPH severity class paralleled the changes shown for the representative subjects for each severity class in Figure 1. Differing from the control group, PETC[O.sub.2] does not increase at AT and decrease during early recovery in the patient groups (Fig 2). The changes are opposite of normal in the patient group.
[FIGURE 2 OMITTED]
The relationships between PETC[O.sub.2] and percentage of predicted peak V[O.sub.2] for the PPH patients and control subjects are shown in Figure 3. For the PPH patients, there was a significant positive correlation between percentage of predicted peak V[O.sub.2]and PETC[O.sub.2] at rest, AT, and at peak V[O.sub.2] (r = 0.72, P < 0.0001; r = 0.70, p < 0.0001; and r = 0.66, p < 0.0001, respectively). The control group showed a higher PETC[O.sub.2] in each state, as compared to the patient group. In the patient group, PETC[O.sub.2] in any given state was abnormally low, and decreased in proportion to the decrease in percentage of predicted peak V[O.sub.2] (ie, the lower the peak V[O.sub.2], the lower the PETC[O.sub.2]).
[FIGURE 3 OMITTED]
Change in PETC[O.sub.2] From Rest to AT
In normal subjects, PETC[O.sub.2] increases when going from rest to AT (Fig 4). In contrast, PETC[O.sub.2] decreased in patients with PPH from rest to AT, except for the mild subgroup, whose reduced resting PETC[O.sub.2] remained unchanged at the AT (Fig 4). In the other subgroups, the magnitude of the decrease in PETC[O.sub.2] between rest and AT was greater as the percentage of predicted AT decreased.
[FIGURE 4 OMITTED]
[O.sub.2]sat in patients with PPH decreased significantly from rest to peak exercise, except in patients with mild disease (Fig 5). There was no significant difference in [O.sub.2]sat as related to disease severity and the control group at rest. However, [O.sub.2]sat decreased to a greater degree, the more severe the disease, during exercise (Fig 5).
[FIGURE 5 OMITTED]
Exercise Gas Exchange as Related to mPAP
The percentage of predicted peak V[O.sub.2] and VE/ VC[O.sub.2] at AT was plotted against resting mPAP for the 29 patients who underwent a diagnostic cardiac catheterization study within 50 clays of their first CPET (Fig 6). The reduction in percentage of predicted peak V[O.sub.2] correlated significantly with the increase in mPAP (r = - 0.59, p < 0.001). Also, VE/VC[O.sub.2] at AT increased as mPAP increased (r = 0.45, p < 0.05) to values that are clearly abnormally elevated.
[FIGURE 6 OMITTED]
Figure 7 shows the correlation between PETC[O.sub.2] and mPAP at rest, the AT, and peak V[O.sub.2]. PETC[O.sub.2] was further reduced as mPAP increased in all three physical states, with a correlation coefficient of -0.51 to -0.53 (p < 0.005).
[FIGURE 7 OMITTED]
Normal subjects dwelling near sea level have progressively increased PETC[O.sub.2] by approximately 5 mm Hg above resting values at the AT (Fig 2). In contrast, PETC[O.sub.2] decreases rather than increases from rest to AT in PPH, except for the two subjects with mildest disease, in whom PETC[O.sub.2] did not change. This directional change in PETC[O.sub.2] from rest to AT is greater for a greater reduction in percentage of predicted AT (Fig 4).
The reduction in PETC[O.sub.2] seen in PPH may be accounted for by mechanisms that increase physiological dead space/tidal volume (VD/VT) and/or reduce ideal alveolar PC[O.sub.2] (PAIC[O.sub.2]). PAIC[O.sub.2] is the alveolar PC[O.sub.2] for an ideal lung, ie, a lung that has totally uniform ventilation-perfusion relationships and therefore the same PC[O.sub.2] in all lung units. PAIC[O.sub.2] is usually equated to PaC[O.sub.2] in the absence of a right-to-left shunt. The mathematical relationships are shown in equations 1 and 2:
VE/VC[O.sub.2] = k/PAIC[O.sub.2] (1 - VD/VT) [equation 1]
VC[O.sub.2]/VE = PAIC[O.sub.2] (1 - VD/VT)/k [equation 2]
where k is a constant that includes barometric pressure and conversion factors to express carbon dioxide output (VC[O.sub.2]) as standard temperature and pressure, dry, and VE as body temperature and pressure, saturated. From the above equations, we can discern the four factors in PPH pathophysiology that cause VE to increase relative to VC[O.sub.2].
Chemically Linked Abnormal Increases in Ventilatory Drive in PPH Patients, All Amplified by Exercise
Increase in VD/VT: VC[O.sub.2]/VE (equation 2, the reciprocal of VE/VC[O.sub.2]) is equal to the concentration of carbon dioxide in the mixed expired gas. The extent to which it is diluted relative to PaC[O.sub.2] is determined by the VD/VT (equation 2). Because of ventilation-perfusion inequalities, (3,6,7) the VD/VT is increased, making gas exchange less efficient than normal. This causes PETC[O.sub.2] to be diluted relative to PaC[O.sub.2]. Because pulmonary vascular disease is the hallmark of patients with PPH, it is likely that the remarkably low resting PETC[O.sub.2] seen in patients with PPH is partially due to underperfusion of ventilated lung, ie, increased VD/VT.
Lactic Acidosis at Low Work Rates: Because of the increased pulmonary vascular resistance in PPH, pulmonary blood flow (cardiac output) fails to increase normally during exercise. The blunted cardiac output response to exercise results in an increase in anaerobic glycolysis, with development of a lactic acidosis at low work rates. This is reflected in the low percentage of predicted AT (Table 1). The lactic acidosis produces acid stimuli, ie, [H.sup.+] and C[O.sub.2] (the latter released during the HC[O.sub.3.sup.-] buffering of the lactic acid) that increase ventilatory drive. This constrains the increase in blood acidity that would otherwise occur. (13)
Arterial Hypoxemia: Exercise-induced hypoxemia resulting from ventilation-perfusion mismatching (14) and/or the development of a right-to-left shunt during exercise stimulates ventilation. (4,12,15) Most PPH patients in our study had normal or near-normal resting [O.sub.2] sat (Fig 5). However, hypoxemia usually developed during exercise in our PPH patients, most markedly in those patients with the greatest physiologic impairment. Hypoxemia stimulates the carotid bodies to drive ventilation and decrease PETC[O.sub.2] during exercise.
The C[O.sub.2/[H.sup.+] Stimulus From an EIS: Because of increased pulmonary vascular resistance in PPH patients, when venous return increases during exercise, right atrial pressure might exceed left atrial pressure. If this does occur, a right-to-left shunt could develop in those patients with a potentially patent foramen ovale. This shunt delivers acidic, C[O.sub.2]-rich, [O.sub.2]-poor blood directly into the systemic circulation through the left atrium. To maintain arterial [H.sup.+] and PaC[O.sub.2]homeostasis, the arterial chemoreceptors (carotid bodies) are stimulated, resulting in an increase in ventilation. (13) This increased ventilatory drive results in hyperventilation of the blood passing through the lungs, thereby decreasing PETC[O.sub.2] in compensation for the C[O.sub.2][H.sup.+] load of the shunted blood. This hyperventilation of the non-shunted blood serves to maintain arterial pH and PC[O.sub.2] homeostasis in the presence of a right-to-left shunt, as previously described for the exercise-induced increase in right-to-left shunt in Eisenmenger syndrome. (16)
VE/VC[O.sub.2] at AT vs PETC[O.sub.2] at AT
From equation 1, above, the PAIC[O.sub.2]is hyperbolically related to VE/VC[O.sub.2] It might be expected therefore that if PETC[O.sub.2] changed parallel to the alveolar PC[O.sub.2], a hyperbolic relationship between VE/VC[O.sub.2] at AT and PETC[O.sub.2] at AT would result. This was observed in our patients with PPH (Fig 8).
[FIGURE 8 OMITTED]
Chronic heart failure and some lung diseases in which VO/VT is increased might also have low PETCOz values. But in these instances, cardiac function and/or pulmonary function are abnormal, providing evidence of these diseases. When a patient has symptoms of exercise limitation with an abnormally reduced PETC[O.sub.2], without objective evidence of lung or heart disease or anemia, PPH might be suspect.
Markowitz and Systrom (17) point out that the major difficulty in the diagnosis of pulmonary vascular limitation to exercise is the relative insensitivity of the clinical evaluation and the physiologic measurements made at rest. These investigators did CPET with radial artery and pulmonary artery catheters for hemodynamic, blood gas, and lactate measurements in 130 patients with exercise intolerance of unknown cause. Using the noninvasive exercise gas exchange measurement algorithm previously described by us for pulmonary vascular disease, (11) they found a 79% sensitivity and 75% specificity with a 76% accuracy. However, the algorithm did not use PETC[O.sub.2] measurements because the low values found in PPH was not fully appreciated at the time.
Matsumoto et al (9) reported average resting PETC[O.sub.2] values of 34.4 [+ or -] 0.6 mm Hg for NYHA class I, 32.7 [+ or -] 0.7 mm Hg for class II, and 32.2 [+ or -] 0.5 mm Hg for class III patients. Most of our PPH patients had PETC[O.sub.2] values below these mean values. Furthermore, during exercise to AT, PETC[O.sub.2] increased in chronic heart failure patients, (9,10) in contrast to the further decrease in PETC[O.sub.2] observed in PPH patients. Deboeck et al (18) compared the oxygen pulse and VE/VC[O.sub.2] at AT in chronic heart failure and pulmonary arterial hypertension for the same NYHA symptom class. They found both measurements to be more abnormal in the pulmonary arterial hypertension group. They did not describe the differences in PETC[O.sub.2],
While both PETC[O.sub.2] at AT and VE/VC[O.sub.2] at AT (Fig 8) are noninvasive indicators of ventilatory efficiency, the PETC[O.sub.2] is a more simply performed noninvasive measurement. When PETC[O.sub.2] at AT is < 30 mm Hg in a patient with exertional dyspnea of unknown cause, PPH should be considered as a possible diagnosis to account for the patient's symptoms because of the frequency with which these low values occur in PPH compared to other disorders. (9-11) Values of PETC[O.sub.2] < 520 mm Hg are so unusual in other diseases (11) that the likelihood of the diagnosis of PPH is even more suspect in patients with exercise intolerance of unknown cause. The diagnosis becomes still more likely if PETC[O.sub.2] progressively decreases from rest to the AT. Also finding that PETC[O.sub.2] increases rather than decreases at the start of recovery (Fig 1, 2) supports the diagnosis. Finally, a low resting PETC[O.sub.2] that decreases further with exercise and is accompanied by arterial oxygen desaturation is an unusual finding in chronic heart failure, (19) but usual in pulmonary vascular occlusive diseases. The observation of an unusually low PETC[O.sub.2] without evidence of acute hyperventilation (ie, normal RER) and with exercise-induced arterial hypoxemia during CPET might be used to select patients for right-heart catheterization to evaluate the possibility that the patient's symptoms are explained by the pulmonary vasculopathy of PPH.
We did not measure arterial blood gases in this study. Documenting that arterial [H.sup.+] changes little during exercise compared to the rate of C[O.sub.2][H.sup.+] production during exercise is based on a study of right-to-left shunts increasing during exercise in patients with Eisenmenger syndrome. (16) That study showed that arterial [H.sup.+] and PaC[O.sub.2] were well regulated during exercise at resting levels. In addition, it can be calculated that ventilation must track VC[O.sub.2] quite closely during exercise for pH to remain in the range that we know to be compatible with life.
We measured [O.sub.2]sat by pulse oximetry rather than by direct sampling of arterial blood. However, this appeared to give reliable measurements, evidenced by the differences found between normal subjects and patients and the differences between patient groups as related to severity of disease.
In summary, PETC[O.sub.2] at rest and exercise is reduced in patients with PPH, often to levels below those observed in other diseases such as chronic heart failure. Because PETC[O.sub.2] is easily measured during CPET, and its values are exceptionally low, capnography during CPET should be of considerable diagnostic value in patients with exercise intolerance. Our findings suggest that exceptionally low PETC[O.sub.2] values that decrease further during exercise and increase in early recovery, accompanied by exercise-induced arterial hypoxemia, should signal the need to investigate further to determine if the disorder causing the exercise symptoms is a primary pulmonary vasculopathy such as PPH.
* From the Department of Medicine, Los Angeles Biomedical Research Institute, Harbor-UCLA Medical Center, Torrance, CA. Supported in part by a grant from the American Heart Association (0160126Y).
(1) Rich S, Dantzker DR, Ayres SM, et al. Primary pulmonary hypertension: a national prospective study. Ann Intern Med 1987; 107:216-223
(2) Rubin LJ. ACCP consensus statement: primary hypertension. Chest 1993; 104:236-250
(3) Sun XG, Hansen JE, Oudiz RJ, et al. Exercise pathophysiology in patients with primary pulmonary hypertension. Circulation 2001; 104:429-435
(4) Wensel R, Optiz CF, Anker SD, et al. Assessment of survival in patients with primary pulmonary hypertension. Circulation 2002; 106:319-324
(5) Wax D, Garofano R, Barst RJ. Effects of long-term infusion of prostacyclin on exercise performance in patients with primary pulmonary hypertension. Chest 1999; 116:914-920
(6) Fukuchi K, Hayashida K, Nakanishi N, et al. Quantitative analysis of lung perfusion in patients with primary pulmonary hypertension. J Nucl Med 2002; 43:757-761
(7) Reybrouck T, Mertens L, Schulze-Neick I, et al. Ventilatory inefficiency for carbon dioxide during exercise in patients with pulmonary hypertension. Clin Physiol 1998; 18:337-344
(8) Mudge GH, Goldstein S, Addonizio LJ, et al. Task force 3: Recipient guidelines/prioritization. J Am Coll Cardiol 1993; 22:21-26
(9) Matsumoto A, Itoh H, Eto Y, et al. End-tidal C[O.sub.2] pressure decreases during exercise in cardiac patients: association with severity of heart failure and cardiac output reserve. J Am Coll Cardiol 2000; 36:242-249
(10) Tanabe Y, Hosaka Y, Ito M, et al. Significance of end-tidal PC[O.sub.2] response to exercise and its relation to functional capacity in patients with chronic heart failure. Chest 2001; 119:811-817
(11) Wasserman K, Hansen J, Sue DY, et al. Principles of exercise testing and interpretation. 3rd ed. Baltimore, MD: Lippincott, Williams and Wilkins, 1999
(12) Sun XG, Hansen JE, Oudiz R, et al. Gas exchange detection of exercise-induced right-to-left shunt in patients with primary pulmonary hypertension. Circulation 2002; 105:54-60
(13) Wasserman K, Whipp BJ, Koyal SN, et al. Effect of carotid body resection on ventilatory and acid-base control during exercise. J Appl Physiol 1975; 39:354-358
(14) Rafanan AL, Golish JA, Dinner DS, et al. Nocturnal hypoxemia is common in primary pulmonary hypertension. Chest 2001; 120:894-899
(15) Nootens MT, Berarducci LA, Kaufmann E, et al. The prevalence and significance of a patent foramen ovale in pulmonary hypertension. Chest 1993; 104:1673-1675
(16) Sietsema KE, Cooper DM, Perloff SK, et al. Control of ventilation during exercise in patients with central venous-to-systemic arterial shunts. J Appl Physiol 1988; 64:234-242
(17) Markowitz DH, Systrom DM. Diagnosis of pulmonary vascular limit to exercise by cardiopulmonary exercise testing. J Heart Lung Transplant 2004; 23:88-95
(18) Deboeck G, Niset G, Lamotte M, et al. Exercise testing in pulmonary arterial hypertension and in chronic heart failure. Eur Respir J 2004; 23:747-751
(19) Wasserman K, Zhang YY, Gitt A, et al. Lung function and exercise gas exchange in chronic heart failure. Circulation 1997; 96:2221-2227
Manuscript received February 26, 2004; revision accepted October 19, 2004.
Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (www.chestjournal. org/misc/reprints.shtml).
Correspondence to: Karlman Wasserman, MD, PhD, FCCP, Department of Medicine, Harbor-UCLA Medical Center, 1000 W Carson St, Box 405, Torrance, CA 90509-2910; e-mail: firstname.lastname@example.org
COPYRIGHT 2005 American College of Chest Physicians
COPYRIGHT 2005 Gale Group