Key words: exercise capacity; heart failure; hemodynamics; inspiratory capacity; lung function; peak oxygen uptake

Transcription

1 Resting Lung Function and Hemodynamic Parameters as Predictors of Exercise Capacity in Patients With Chronic Heart Failure* Serafim Nanas, MD; John Nanas, MD, PhD; Ourania Papazachou, MD; Christos Kassiotis, MD; Antonios Papamichalopoulos, MD; Joseph Milic-Emili, MD; and Charis Roussos, MD, MSc, PhD, MRS, FCCP Study objectives: The aim of this study was to examine the role of resting pulmonary function and hemodynamic parameters as predictors of exercise capacity in patients with chronic heart failure. Measurements and results: Fifty-one patients with chronic heart failure underwent resting pulmonary function testing, including inspiratory capacity (IC) and symptom-limited, treadmill cardiopulmonary exercise testing (CPET). Right-heart catheterization and radionuclide ventriculography were performed within 2 days of CPET. Mean ( SD) left ventricular ejection fraction was 31 12% and cardiac index was L/min/m 2. Percentage of predicted FEV 1 was 92 14%, percentage of predicted FVC was 94 15%, FEV 1 /FVC was 81 4%, and percentage of predicted IC was 84 18%. Mean peak oxygen uptake (peak V O2 ) was ml/kg/min. Analysis of variance among the three functional Weber classes showed statistically significant differences for pulmonary capillary wedge pressure (PCWP) and IC. Specifically, the more severe the exercise intolerance, the lower was IC and the higher was PCWP. In a multivariate stepwise regression analysis, using peak V O2 (liters per minute) as the dependent variable and the pulmonary function test measurements as independent variables, the only significant predictor selected was IC (r 0.71, p < ). In a final stepwise regression analysis including all the independent variables of the resting pulmonary function tests and hemodynamic measurements, the two predictors selected were IC and PCWP (r ). Conclusions: In patients with chronic heart failure, IC is inversely related to PCWP and is a strong independent predictor of functional capacity. (CHEST 2003; 123: ) Key words: exercise capacity; heart failure; hemodynamics; inspiratory capacity; lung function; peak oxygen uptake Abbreviations: ANOVA analysis of variance; AT anaerobic threshold; CI cardiac index; CPET cardiopulmonary exercise testing; FEF 25 forced expiratory flow at 25% of FVC; FEF maximal midexpiratory flow rate; FEF 50 forced expiratory flow at 50% of FVC; FEF 75 forced expiratory flow at 75% of FVC; IC inspiratory capacity; LVEF left ventricular ejection fraction; PAP pulmonary artery pressure; PCWP pulmonary capillary wedge pressure; PEF peak expiratory flow; RAP right atrial pressure; TLC total lung capacity; V co 2 carbon dioxide output; V e minute ventilation; V o 2 oxygen uptake Reduced exercise capacity, a main characteristic of chronic heart failure, correlates weakly with resting hemodynamic indexes. 1 While dyspnea and muscle fatigue represent major limiting factors, the precise pathophysiologic mechanisms leading to exercise intolerance have not been fully clarified. Respiratory abnormalities associated with chronic heart failure include restrictive and obstructive changes, 2 4 associated with decreased lung compliance, 5 decreased diffusion capacity, 6 hyperventilation, 7 ventilation-perfusion mismatch, 8 bronchial hyperresponsiveness, 9 and respiratory muscle weakness. 10 The *From the Pulmonary and Critical Care Medicine Department (Drs. S. Nanas, Papazachou, Kassiotis, Papamichalopoulos, and Roussos), National and Kapodestrian University; Clinical Therapeutics Department (Dr. J. Nanas), National and Kapodestrian University; and Meakins-Christie Laboratories (Dr. Milic-Emili), McGill University, Montreal, Canada. The study was supported by a grant from a National and Kapodestrian University of Athens Special Account for Research Grants, and by Thorax Foundation. Manuscript received January 7, 2002; revision accepted September 17, Reproduction of this article is prohibited without written permission from the American College of Chest Physicians ( permissions@chestnet.org). Correspondence to: Serafim Nanas, MD, National and Kapodestrian University, Pulmonary and Critical Care Medicine Department, Evgenidio Hospital, 20, Papadiamantopoulou str, Athens , Greece; snanas@cc.uoa.gr 1386 Clinical Investigations

2 association between resting lung function indexes and exercise capacity has not been studied in patients with chronic heart failure without concurrent obstructive lung disease. The aim of the present study was to examine the relationship of resting respiratory and hemodynamic variables to exercise capacity in patients with moderate-to-severe chronic heart failure. Patient Population Materials and Methods Anthropometric and clinical characteristics of the 51 subjects with chronic heart failure enrolled in this study are listed in Table 1. All patients were thought to be in clinically stable condition by their physicians at the time of study. Myocardial infarction within 2 months, FEV 1 /FVC 75%, and other noncardiac, exercise-limiting disorders were criteria of exclusion from the study. Other exclusion criteria were angina pectoris, lightheadedness, serious arrhythmias, or ECG changes consistent with ischemia induced by exercise. The diagnosis of chronic heart failure was based on a thorough clinical evaluation and laboratory testing, including blood chemistries, echocardiography, right-heart catheterization, radionuclide ventriculography, and coronary angiography. A myocardial biopsy was also performed when clinically indicated. The study was reviewed and approved by the Human Study Committee of our institution, and informed consent was formally obtained from each participant. Patients were classified into three groups according to the Weber classification. 11 Group A included 14 patients whose peak oxygen uptake (V o 2 ) was 20 ml/kg/min, group B included 19 patients whose peak V o 2 was 16 ml/kg/min and 20 ml/kg/ min, and group C included 18 patients whose peak V o 2 was 10 ml/kg/min and 16 ml/kg/min. Another group of 12 patients (9 men and 3 women; mean SD age, 51 9 years) was also studied. From those patients who underwent cardiopulmonary exercise testing (CPET) and pulmonary function testing in our laboratory on more than one occasion, we selected patients who showed a significant difference in pulmonary capillary wedge pressure (PCWP) [high PCWP low PCWP 50% of high PCWP]. Table 1 Baseline Characteristics of the 51 Study Patients* Characteristics Data Age, yr Male/female gender 41/10 Height, cm Weight, kg Body mass index New York Heart Association functional class I 2 II 12 III 30 IV 7 Underlying heart disease Ischemic cardiomyopathy 17 Dilated cardiomyopathy 29 Valvular heart disease 2 Other 3 *Data are presented as mean SD or No. Pulmonary Function Tests Measurements of FVC and FEV 1 were obtained in the sitting position with a closed-circuit spirometer (V max model 229; SensorMedics; Yorba Linda, CA), as recommended by the American Thoracic Society, 12 after familiarization of the study participants with the laboratory environment. Simultaneously, peak expiratory flow (PEF), forced expiratory flow at 25% of FVC (FEF 25 ), forced expiratory flow at 50% of FVC (FEF 50 ), forced expiratory flow at 75% of FVC (FEF 75 ), and maximal midexpiratory flow rate (FEF ) were calculated. Inspiratory Capacity Measurement: The procedure of inspiratory capacity (IC) measurement was explained in detail to every subject. In the sitting position, the patients were asked to breathe normally through a mouthpiece connected to a calibrated pneumotachograph. When they achieved a steady tidal volume and end-expiratory lung volume, they were instructed to fully inspire from end-expiratory lung volume to total lung capacity (TLC) and then breathe normally again. This was repeated four times. In all instances, they performed at least three satisfactory maneuvers, two of which did not differ by 5%. The best IC was selected. The approach has been shown previously to be reliable and reproducible. 13 Hemodynamic Measurements Right-heart catheterization was performed within 48 h of the CPET to measure PCWP, right atrial pressure (RAP) and pulmonary artery pressure (PAP). Left ventricular ejection fraction (LVEF) was measured by radionuclide ventriculography. The cardiac index (CI) was measured at rest by the noninvasive single-breath acetylene (C 2 H 2 ) technique 14 just before the onset of CPET. CPET Each patient underwent a symptom-limited, incremental CPET on a treadmill (Marquette Electronics 2000; Marquette Electronics; Milwaukee, WI) on the same day as the pulmonary function tests. A history of exercise tolerance was obtained before CPET. The exercise protocol (modified Bruce or modified Naughton) was chosen according to the New York Heart Association class to target test duration between 5 min and 15 min. All parameters were recorded for 2 min at rest, throughout exercise, and for the first 5 min of recovery. Peripheral oxygen hemoglobin saturation was monitored by pulse oximetry. Heart rate and rhythm were monitored by 12-lead ECG (MAX 1 system; Marquette Electronics), and systemic BP was measured every 2 min with a standard mercury sphygmomanometer. The patients were encouraged to exercise to exhaustion, intolerable leg fatigue, or dyspnea. Gas exchange was studied with the patient breathing through a low resistance valve, with the nose clamped. V o 2, carbon dioxide output (V co 2 ), and minute ventilation (V e) were measured on a breath-by-breath basis with a V max 229 monitor for pulmonary and metabolic studies (SensorMedics). Respiratory rate was recorded throughout the exercise test and recovery period. The ratio of V e to maximal voluntary ventilation was used to assess respiratory reserve. The system was calibrated with a gas mixture of known concentration before each test. These measurements were obtained in the upright position before and during exercise, and during the first 5 min of recovery with the subject sitting in a chair. Baseline V o 2 was calculated by averaging the measurements made for 2 min before the onset of exercise. The values of peak V o 2,V co 2, and V e were calculated as the average of the measurements made during the 20 s period before the end of exercise. The anaerobic threshold (AT) was deter- CHEST / 123 / 5/ MAY,

3 mined using the V-slope technique, 15 and the result was confirmed by a graph on which respiratory equivalent for oxygen (V e/v o 2 ) and carbon dioxide (V e/v co 2 ) were plotted simultaneously against time. The ventilatory response was calculated as the slope of the relation between V e and V co 2 from the beginning of exercise to AT. This was obtained by the method of the least-squares linear regression analysis. In order to evaluate the oxygen consumption kinetics during recovery, the first degree slope of V o 2 for the first minute of recovery period (V o 2 /t slope) 16 was calculated by linear regression using an appropriate computerized statistical program. The time required for a 50% fall from peak V o 2 was also calculated. Statistical Analysis Results are presented as means SD unless otherwise stated. Correlations were tested by Pearson s correlation coefficient. Analysis of variance (ANOVA) and Bonferroni post hoc test of significance were used for the statistical evaluation of the differences among Weber groups. A multivariate linear regression analysis was used to test the independent association of lung function and hemodynamic indexes with oxygen kinetics, followed by a stepwise regression analysis. Equations were calculated by the same method. A p value 0.05 was considered statistically significant. The significance of differences between means in the subgroup of 12 patients was examined by paired Student t test. Results Resting hemodynamic measurements of the whole population and of the three Weber class groups are presented in Table 2. Among all hemodynamic variables measured, only PCWP differed significantly between group A and group C (p 0.01), although the differences in PAP nearly reached statistical significance (p 0.057). Table 3 presents the measurements of resting pulmonary function in the overall population and in each Weber class group. A significant difference (p 0.05) in IC (percentage predicted) was found between group A and group B, and group A and group C. A nonsignificant trend toward a decrease in FVC (percentage predicted) was also noted (p 0.066). A marked decrease in FEF 75 (percentage predicted) was observed in the total population, but there was no significant difference between the Weber groups. The CPET indexes are presented in Table 4. All parameters differed significantly among the three groups. In all groups, peak V o 2 was 80% predicted. The significant correlations of peak V o 2 to resting lung function and hemodynamic variables are listed in Table 5. In a multivariate stepwise regression analysis, we used peak V o 2 (liters per minute) as the dependent variable and the resting pulmonary function data that were significantly correlated with peak V o 2 (IC, FEV 1, FVC, PEF, FEF 25, FEF 50, FEF 75, FEF ) as the independent variables. It was found that the only independent predictor was IC (liters). The corresponding equation was as follows: peak V o IC 0.2 (SE), where R was 0.71 and R 2 was 0.5 (F statistic 49.5). Figure 1 shows the correlation of peak V o 2 to IC in our 51 patients. In a further analysis, all resting pulmonary function and hemodynamic data that are listed in Table 5 were used as independent variables. In this final stepwise regression analysis, the only significant predictors found were IC (liters) and PCWP (millimeters of mercury). The corresponding equation was as follows: peak V o IC PCWP 0.23 (SE), where the correlation (R) and determination (R 2 ) coefficients were 0.76 and 0.58, respectively (F statistic 28.8). The significant correlations of IC to resting lung function and hemodynamic variables are listed in Table 6. In a multivariate stepwise regression analysis, using resting IC (liters) as the dependent variable Table 2 Resting Hemodynamic Parameters Stratified by Weber Class in 51 Patients With Chronic Heart Failure* Parameters All patients (n 51) Weber Class A (n 14) Weber Class B (n 19) Weber Class C (n 18) LVEF, % CI, L/min/m RAP, mm Hg PCWP, mm Hg PAP, mm Hg Heart rate, beats/min BP, mm Hg Systolic Diastolic *Data are presented as mean SD. Significantly different between group A and group C (p 0.01). ANOVA, p Clinical Investigations

4 Table 3 Resting Pulmonary Function Data by Weber Class* Variables All Patients (n 51) Weber Class A (n 14) Weber Class B (n 19) Weber Class C (n 18) FEV 1, % predicted FVC, % predicted FEV 1 /FVC, % IC, % predicted PEF, % predicted FEF 25, % predicted FEF 50, % predicted FEF 75, % predicted FEF 25 75, % predicted *Data are presented as mean SD. Significantly different between group A to group B or group C (p 0.05). ANOVA, p and the hemodynamic data as the independent variables, PCWP was the only significant predictor selected. The correlation of IC to PCWP is depicted in Figure 2. There was no statistically significant difference in age between the 51 patients of the basic group and the 12 patients of the additional group. In that group, when PCWP was low ( mm Hg), IC was L or 89 13% predicted; when PCWP was high ( mm Hg), IC was relatively decreased ( L or 80 15% predicted) [Fig 3]. The difference in PCWP and IC was statistically significant on the two occasions as tested by paired t test (t 6.6, p 0.001, and t 2.9, p 0.015, respectively). The change in PCWP and IC was mm Hg and L, respectively, and there was a significant correlation between these two parameters (r 0.6, p 0.04). Also, there was a statistically significant correlation between IC and V o 2 peak (r 0.52, p 0.018), similar to the group of 52 patients. Discussion The main finding of this study is that in patients with chronic heart failure, IC and PCWP are the only significant predictors of V o 2 peak among resting lung function and hemodynamic parameters according to multivariate stepwise regression analysis. The decrease of IC observed in our study was related to the severity of the syndrome. Also, changes in IC are related to changes in PCWP. The confounding factor of the presence of COPD in our study population was controlled by exclusion criteria (FEV 1 /FVC 75%). A reduction of IC reflects an increase in functional residual capacity and/or a decrease in TLC. A restrictive pattern of pulmonary impairment has been docu- Table 4 CPET Indices Stratified by Weber Class* Variables All Patients (n 51) Weber Class A (n 14) Weber Class B (n 19) Weber Class C (n 18) V o 2 peak, ml/kg/min V o 2 peak, L/min V o 2 peak, % predicted AT, ml/kg/min V e/v co 2 slope V e/v o 2 slope Peak heart rate, beats/min V o 2 /t slope, ml/min/min T 1 2 of V o 2, min *Data are presented as mean SD. T 1 2 time required for a 50% fall; V o 2 /t slope first degree slope of V o 2 for the first minute of recovery period. p 0.05, Weber class group A to group B to group C. p 0.01, Weber class group A to group B to group C. p 0.001, Weber class group A to group B to group C. p 0.01, Weber class group B to group C. p 0.001, Weber class group B to group C. CHEST / 123 / 5/ MAY,

5 Table 5 Significant Correlations of V O2 Peak to Various Resting Cardiorespiratory Parameters Parameters mented in chronic heart failure. 2 4,17,18 This can be attributed mainly to subclinical alveolar and interstitial pulmonary edema. 19 The importance of increased lung water in chronic heart failure is also supported by the finding of improved lung function and exercise capacity observed after ultrafiltration 20 and, conversely, by the deterioration after saline solution infusion. 18 The original pathophysiologic mechanism leading to water accumulation in the lung is increased pulmonary venous r V o 2 peak, L/min p Value LVEF, % PCWP, mm Hg PAP, mm Hg FVC, L FEV 1,L IC, L PEF, L/min FEF 25, L/min FEF 50, L/min FEF 75, L/min FEF pressure, as expressed by a high resting PCWP. The fact that in our study PCWP correlates negatively to IC and V o 2 peak, and the significant correlation between PCWP and IC change, in the additional group of 12 patients, support that hypothesis. Other studies have previously shown a negative correlation of PCWP to FVC and TLC. 2 Other causes for TLC reduction in chronic heart failure are cardiomegaly, 21 increased central blood volume, 19 fibrosis from chronic congestion (stiffening of lung parenchyma), and possible pleural effusion. 22 All the above-mentioned mechanisms of lung restriction lead to decreased lung compliance, 5 increased work of breathing, and respiratory muscle weakness. 10 A relative increase in functional residual capacity could be explained by dynamic pulmonary hyperinflation and gas trapping due to expiratory flow limitation 23 and small airway closure, as evidenced by the increased closing volume 24,25 that has been observed in acute heart failure and in valvular heart disease. Although patients with obstructive lung changes have been excluded from this study, we found a marked decrease in forced expiratory flow at low lung volume (FEF 75 ) in our chronic heart failure population (Table 3). This reduction in expiratory flow reserve could promote tidal flow limitation with concurrent dynamic hyperinflation. Figure 1. Scatter graph of peak V o 2 vs IC of 51 patients with chronic heart failure (r 0.71, p 0.001) Clinical Investigations

6 Table 6 Significant Correlations Between Resting IC and Various Resting Cardiorespiratory Parameters Parameters Resting Pulmonary Function and Peak V O2 Our resting lung function data (FEV 1, FVC, and other expiratory flow parameters) were slightly reduced, while FEF 75 was severely reduced. These findings are comparable to those of previous studies that included patients in similar functional status. 26,27 Significant correlations of V o 2 peak 27 to FEV 1 / FVC and FEF (percentage predicted) have been reported, but they were not found in our data. r IC, L p Value LVEF, % 0.26 NS PCWP, mm Hg PAP, mm Hg FVC, L FEV 1,L PEF, L/min FEF 25, L/min FEF 50, L/min FEF 75, L/min FEF Our selection of patients with FEV 1 /FVC 75%, which narrowed the range of FEV 1, may explain the discrepancy. The role of IC as a predictor of peak V o 2 has not been investigated before in patients with chronic heart failure. In this study, IC was found to be the only independent predictor of peak V o 2 among all respiratory variables investigated with the stepwise linear regression analysis. Resting Hemodynamics and Peak V O2 The role of resting hemodynamic parameters as predictors of peak V o 2 is controversial. Several studies have found no correlation of peak V o 2 or exercise duration to resting left-heart hemodynamic measurements, which included CI, LVEF, systemic vascular resistance, and stroke volume index. 1,28 32 In our study, CI did not show statistically significant differences between Weber groups (ANOVA), although on average it was higher in group C than in groups A and B. This has been observed in previous studies ; however, the fact that PCWP was also significantly higher in group C may explain the difference in CI since group C is functioning in a different part of the Frank-Starling curve. Correlations between right-heart hemodynamic Figure 2. Scatter graph of PCWP vs IC of 51 patients with chronic heart failure (r 0.34, p 0.016). CHEST / 123 / 5/ MAY,

7 Figure 3. Inspiratory capacity (percentage predicted SE) before and after reduction of PCWP in a subgroup of 12 patients (paired t 2.9, p 0.01). data, including PCWP, PAP, RAP, or pulmonary vascular resistance and exercise capacity, have ranged from high 28,31 to moderate 29,32 or absent. 30 In our study, PCWP, PAP, and LVEF correlated moderately with peak V o 2. In our patients, there was a great variability in PCWP (range, 3 to 40 mm Hg), even though the patients were in stable clinical condition. This finding is consistent with other studies dealing with chronic heart failure. 28,31 In addition, despite being selected as an independent predictor of peak V o 2 in the stepwise linear regression analysis, PCWP was second to IC, contributing only an additional 8% to the overall variance of peak V o 2. Clinical Implications and Study Limitations The finding that a simple, safe, noninvasive maneuver such as measurement of IC predicts exercise capacity better than other resting cardiorespiratory parameters may prove useful in clinical practice, either when CPET is not available or when it is contraindicated. Since our measurements of lung function were limited to rest, the changes in IC between baseline and peak exercise were not investigated. Such measurements, however, should provide additional information regarding the role of IC on exercise intolerance in patients with chronic heart failure. A corroboration of the role of IC in the pathophysiology of chronic heart failure may be obtained by the measurement of hemodynamic parameters and IC before and after diuresis in patients with high PCWP. Conclusion In conclusion, among the various resting cardiorespiratory variables examined, IC was significantly decreased in both moderate chronic heart failure and severe chronic heart failure. The decreased IC along with the increased PCWP were the only significant predictors of exercise capacity, according to multivariate stepwise regression analysis. References 1 Franciosa JA, Park M, Levine TB. Lack of correlation between exercise capacity and indexes of resting left ventricular performance in heart failure. Am J Cardiol 1981; 47: Ries AL, Gregoratos G, Friedman PJ, et al. Pulmonary function tests in the detection of left heart failure: correlation 1392 Clinical Investigations

8 with pulmonary artery wedge pressure. Respiration 1986; 49: Guazzi M, Agostoni P, Matturri M, et al. Pulmonary function, cardiac function, and exercise capacity in a follow-up of patients with congestive heart failure treated with carvedilol. Am Heart J 1999; 138: Ravenscraft SA, Gross CR, Kubo SH, et al. Pulmonary function after successful heart transplantation. Chest 1993; 103: Evans SA, Watson L, Cowley AJ, et al. Static lung compliance in chronic heart failure: relation with dyspnoea and exercise capacity. Thorax 1995; 50: Puri S, Baker BL, Dutka DP, et al. Reduced alveolar-capillary membrane diffusing capacity in chronic heart failure: its pathophysiological relevance and relationship to exercise performance Circulation 1995; 91: Clark AL, Volterrani M, Swan JW, et al. The increased ventilatory response to exercise in chronic heart failure: relation to pulmonary pathology. Heart 1997; 77: Clark AL, Volterrani M, Swan JW, et al. Ventilation-perfusion matching in chronic heart failure. Int J Cardiol 1995; 48: Cabanes LR, Weber SN, Matran R, et al. Bronchial hyperresponsiveness to methacholine in patients with impaired left ventricular function. N Engl J Med 1989; 320: Mancini DM, Henson D, LaManca J, et al. Respiratory muscle function and dyspnea in patients with chronic congestive heart failure. Circulation 1992; 86: Weber KT, Kinasewitz GT, Janicki JS, et al. Oxygen utilization and ventilation during exercise in patients with chronic heart failure. Circulation 1982; 65: American Thoracic Society. Standardization of spirometry, 1995 update. Am Rev Respir Dis 1995; 152: Marin JM, Carrizo SJ, Gascon M, et al. Inspiratory capacity, dynamic hyperinflation, breathlessness, and exercise performance during the 6-minute-walk test in chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2001; 163: Elkayam U, Wilson AF, Morrison J, et al. Non-invasive measurement of cardiac output by a single constant expiratory technique. Thorax 1984; 39: Beaver WL, Wasserman K, Whipp BJ. A new method for detecting anaerobic threshold by gas exchange. J Appl Physiol 1986; 60: Nanas S, Nanas J, Kassiotis C, et al. Respiratory muscles performance is related to oxygen kinetics during maximal exercise and early recovery in patients with congestive heart failure. Circulation 1999; 100: Faggiano P, Lombardi C, Sorgato A, et al. Pulmonary function tests in patients with congestive heart failure: effects of medical therapy. Cardiology 1993; 83: Puri S, Dutka DP, Baker L, et al. Acute saline infusion reduces alveolar-capillary membrane conductance and increases airflow obstruction in patients with left ventricular dysfunction. Circulation 1999; 99: Hauge A, Bo G, Waaler BA. Interrelations between pulmonary liquid volumes and lung compliance. J Appl Physiol 1975; 38: Agostoni PG, Marenzi GC, Pepi M, et al. Isolated ultrafiltration in moderate congestive heart failure. J Am Coll Cardiol 1993; 21: Agostoni PG, Cattadori G, Guazzi M, et al. Cardiomegaly as a possible cause of lung dysfunction in patients with heart failure. Am Heart J 2000; 140:E24 22 Chua TP, Coats AJS. The lungs in chronic heart failure. Eur Heart J 1995; 16: Duguet A, Tantucci C, Lozinguez O, et al. Expiratory flow limitation as a determinant of orthopnea in acute heart failure. J Am Coll Cardiol 2000; 35: Light RW, George RB. Serial pulmonary function in patients with acute heart failure. Arch Intern Med 1983; 143: Collins JV, Clark TJH, Brown DJ. Airway function in healthy subjects and patients with left heart failure. Clin Sci Mol Med 1975; 49: Kraemer MD, Kubo SH, Rector TS, et al. Pulmonary and peripheral vascular factors are important determinants of peak exercise oxygen uptake in patients with heart failure. J Am Coll Cardiol 1993; 21: Dimopoulou I, Tsintzas O, Daganou M, et al. Contribution of lung function to exercise capacity in patients with chronic heart failure. Respiration 1999; 66: Szlachcic J, Massie BM, Kramer BL, et al. Correlates and prognostic implication of exercise capacity in chronic congestive heart failure. Am J Cardiol 1985; 55: Metra M, Livio DC, Panina G, et al. Exercise hyperventilation chronic congestive heart failure and its relation to functional capacity and hemodynamics. Am J Cardiol 1992; 70: Chomsky DB, Lang CC, Rayos GH, et al. Hemodynamic exercise testing: a valuable tool in the selection of cardiac transplantation candidates. Circulation 1996; 94: Roul G, Moulichon ME, Bareiss P, et al. Prognostic factors of chronic heart failure in NYHA class II or III: value of invasive exercise haemodynamic data. Eur Heart J 1995; 16: Franciosa JA, Baker BJ, Seth L. Pulmonary versus systemic hemodynamics in determining exercise capacity of patients with chronic left ventricular failure. Am Heart J 1985; 110: CHEST / 123 / 5/ MAY,