International Journal of Cardiology 171 (2014) 73–77
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Lung function and gas exchange in Eisenmenger syndrome and their impact on exercise capacity and survival☆ Craig S. Broberg a,⁎,1, Ryan C. Van Woerkom a,1, Elizabeth Swallow b,1, Kostas Dimopoulos d,1, Gerhard-Paul Diller d,1, Gopal Allada c,1, Michael A. Gatzoulis d,1 a
Adult Congenital Heart Disease Program, Division of Cardiovascular Medicine, Oregon Health and Science University, Portland, OR, United States Respiratory Muscle Laboratory, Royal Brompton Hospital and Harefield NHS Foundation Trust, National Heart and Lung Institute, Imperial College, London, UK Division of Pulmonology, Oregon Health and Science University, Portland, OR, United States d NIHR Cardiovascular Biomedical Research Unit, Royal Brompton and Harefield NHS Foundation Trust, National Heart and Lung Institute, Imperial College London, UK b c
a r t i c l e
i n f o
Article history: Received 2 May 2013 Received in revised form 13 November 2013 Accepted 18 November 2013 Available online 23 November 2013 Keywords: Eisenmenger Cyanosis Pulmonary arterial hypertension Lung function Blood gas Hypocapnia
a b s t r a c t Background: Eisenmenger physiology may contribute to abnormal pulmonary mechanics and gas exchange and thus impaired functional capacity. We explored the relationship between lung function and gas exchange parameters with exercise capacity and survival. Methods: Stable adult patients with Eisenmenger syndrome (N = 32) were prospectively studied using spirometry, lung volumes, diffusion capacity, and blood gas analysis, as well as same day measurement of 6-minute walk distance and cardiopulmonary maximal treadmill exercise. Patients were followed prospectively to determine survival (7.4 ± 0.5 years). Abnormalities were identified and appropriate comparisons were made between affected and unaffected individuals between respiratory mechanics, exercise function, and survival. Results: Obstruction (FEV1/FVC ratio b0.70) was found in 13 patients (41%), who were older but not otherwise different. Restriction was uncommon. Diffusion transfer coefficient, which was b 80% in half the patients, correlated with exercise duration (r = 0.542, P = 0.005), and was worse in non-survivors (N = 6). Nearly all patients had a compensated respiratory alkalosis (PaCO2 32 ± 4.4 mm Hg). PaCO2 was less reduced in older patients (r = 0.438, P = 0.022), and correlated independently with exercise duration (R = − 0.463, P = 0.03), yet PaO2, not PaCO2, was associated with survival. Conclusions: Eisenmenger patients show evidence of obstructive lung disease, diffusion abnormalities, and hypocapnia; likely from hyperventilation. Understanding expected lung mechanics and gas exchange may facilitate more appropriate clinical management. © 2013 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Eisenmenger syndrome (ES) develops from an unrestricted communication between the systemic and pulmonary circulation (such as a large intracardiac shunt or patent ductus arteriosus), which if unrepaired progresses to elevated pulmonary artery pressures, pulmonary vascular disease, and reversal of shunt leading to hypoxemia, cyanosis, and erythrocytosis [1]. There are several potential mechanisms ☆ Support: This work was supported by a grant from the Waring Trust. Dr. Broberg is supported by a Clinical Research Grant (K23HL093024) from the National Heart, Lung and Blood Institute, United States. The Royal Brompton Hospital Adult Congenital Heart Disease Center and Centre for Pulmonary Hypertension have received support from the British Heart Foundation. ⁎ Corresponding author at: UHN 62, 3181 SW Sam Jackson Pk Rd., Portland, OR 97239, United States. Tel.: +1 503 494 8750; fax: +1 503 494 8550. E-mail address: [email protected] (C.S. Broberg). 1 Each author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation. 0167-5273/$ – see front matter © 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijcard.2013.11.047
for abnormal lung mechanics, in addition to the pulmonary vascular changes themselves. These include scoliosis, present in roughly 10% of patients with congenital heart disease, or significant enlargement of the heart and/or central pulmonary arteries that may have spaceoccupying effects [2]. Patients with ES have very impaired exertional capacity, the lowest of any group of patients with congenital heart disease [3]. It is unknown to what extent, if any, lung function contributes to this physical limitation. Our institution recently published a large series evaluating restrictive lung disease in congenital heart patients including ES [4]. However, patients with obstructive lung disease were excluded, and the study did not focus on the unique aspects of ES. While there are few studies describing pulmonary function parameters such as obstructive lung disease in other groups with pulmonary arterial hypertension (PAH) [5–7], including a small number of patients with Eisenmenger physiology [8], none, to our knowledge, have specifically explored lung mechanics specifically in ES, nor their potential impact on exercise capacity or survival. Therefore, in this prospective study we aim to
74
C.S. Broberg et al. / International Journal of Cardiology 171 (2014) 73–77
describe all aspects of lung mechanics and gas exchange specifically in a stable cohort of ES patients, and explore their prognostic value. 2. Methods Consecutive adults with ES seen at our tertiary referral center between 2003 and 2005 were prospectively invited to participate and all patients gave informed consent. The study received approval from the Ethics Committee at the Royal Brompton Hospital. Inclusion criteria were known ES syndrome patients with a resting oxygen saturation b92%, seen as stable outpatients. Inpatients or those with acute illness such as bronchitis/pneumonia or recent large-volume hemoptysis were initially excluded, but invited to participate after resolution of the issue and return to outpatient status. Patients with difficulty complying with test instructions (such as those with developmental delay) were excluded from tests needing cooperation (i.e. spirometry or cardiopulmonary exercise) but eligible for other portions of the study and subsequent analysis. Each patient underwent spirometry, lung volumes, diffusion capacity quantification, and earlobe capillary blood gas measured at rest. All testing was done in a single visit and in accordance with accepted clinical standards [9]. Spirometry was performed three times and results averaged. Lung volumes were measured in an enclosed gas cabin using Helium dilution. Diffusion of carbon monoxide (CO) was quantified and corrected for hemoglobin measured on the test day. Patients also underwent a 6-minute walk test and a maximal cardiopulmonary exercise testing on the test day. Oxygen saturation and heart rate at baseline and at the end of the 6 minute walk test were recorded in addition to distance walked. Maximal cardiopulmonary treadmill exercise testing was performed using a modified Bruce protocol with continuous measurement of oxygen consumption (VO2) and carbon dioxide production (VCO2) [10,11]. All patients were followed prospectively at the Royal Brompton Hospital. Any deaths were recorded together with date and mode of death, if known. For survivors, date of last clinic visit was used for survival analysis. All patients were accounted for within six months of the time of final analysis. 2.1. Statistical analysis Spirometry, lung volume, and diffusion results were reported as percent of predicted (%pred) values for historic normal individuals matched for age, gender, and height. This allowed categorization into accepted severity categories (mild/moderate/severe), defined below. Statistical comparisons between categories, when described, were made using appropriate parametric testing (Students t-test or chi-square test). Strength of association between continuous variables was measured using Pearson's correlation coefficient. A single forward step-wise multivariate analysis was done to determine the potential interaction of PaCO2 and age as predictors of exercise duration. Analysis was performed using SPSS for Windows version 11.0 (SPSS Inc., Chicago, Illinois). No correction was made for multiple tests as this was a descriptive study. Because there were only 6 deaths, survival analysis was limited to simple comparisons between survivors and non-survivors. Results are expressed as mean ± SD unless stated otherwise. P b 0.05 was considered statistically significant.
Fig. 1. Box plots showing mean, interquartile range, maximum and minimum values for major lung function variables, expressed as a percent of predicted values for age/gender matched populations. Outliers are shown as open circles. FEV1 = force expiratory volume in 1 s. FVC = forced vital capacity. TLC = total lung capacity. RV = residual volume. VA = alveolar volume. IVC = inspiratory vital capacity, KCO = carbon monoxide diffusion coefficient, KCOC = carbon monoxide diffusion coefficient corrected for measured hemoglobin, PEF = peak expiratory flow, PIF = peak inspiratory flow, VO2 = peak oxygen consumption.
Fig. 2. Association between diffusion coefficient (corrected for hemoglobin) and exercise duration during cardiopulmonary exercise.
3. Results Thirty-two patients were enrolled in the study, of which 23 (71%) were female. Mean age was 41 ± 14 years, resting oxygen saturation 81 ± 7%, hemoglobin 20 ± 3 g/dl, all consistent with confirmed Eisenmenger physiology. Three patients had atrial septal defects, 20 had ventricular septal defects, and 9 had a patent ductus arteriosus as their main source of shunting. Not all patients could or wished to complete all aspects of the study. Six patients had some form of developmental delay. Boxplots for major lung function variables expressed as a percentage of predicted values are shown (Fig. 1), including percent predicted peak VO2 during exercise. Additionally, peak VO2 (N = 27) was 11.1 ± 3.4 ml/kg/min, Ve/VCO2 slope was 82 ± 68, anaerobic threshold was 8.76 ± 2.5 ml/kg/min, and respiratory exchange ratio was 0.98 ± 0.1, all consistent with significantly reduced exercise capacity as previously documented [12,13]. 3.1. Obstructive or restrictive lung disease Reliable spirometry data were obtained in 31 patients. Forced Expiratory Volume (FEV1) was 3.86 ± 1.05 L, or 75 ± 16%pred, range 39– 111%. Forced vital capacity (FVC) was 4.62 ± 1.65 L, or 89 ± 18%pred, range 44–116%. There were 13 patients (42%) with evidence of obstruction, defined as FEV1/FVC ratio b0.70. Based on the %pred FEV1, there was 1 patient with grade 1 (FEV1 N 80%pred), 11 with grade 2 (FEV1 50– 80%pred), and 1 with grade 3 (FEV1 b 50%pred) according to the 2011
Fig. 3. Scatterplot of PaCO2 and pH demonstrating acid/base status of the patient cohort. The majority of patients had a respiratory alkalosis with varying degrees of metabolic compensation.
C.S. Broberg et al. / International Journal of Cardiology 171 (2014) 73–77
75
b80%pred. In these, 6 minute walk distance tended to be lower (352 ± 80 vs. 423 ± 122 m, P = 0.071) and exercise duration was significantly lower (4.5 ± 1.3 vs.7.0 ± 2.1 min, P b 0.001). There was a positive correlation between KCOc and exercise duration (r = 0.542, P = 0.005) (Fig. 2). 3.3. Other pulmonary function testing Peak inspiratory and expiratory flows (N = 32) were slightly reduced at 80 ± 16 and 77 ± 23%pred values, respectively. Peak expiratory flow correlated with 6 minute walk distance (r = 0.360, P = 0.046), and negatively with age (r = −0.477, P = 0.007), as did peak inspiratory flow (r = −0.392, P = 0.029 respectively). However, neither variable correlated with any measure of maximal exercise on cardiopulmonary exercise testing. 3.4. Acid–base status
Fig. 4. Relationship of PaO2 (A) and PaCO2 (B) and exercise duration on maximum treadmill exercise testing. Although PaO2 was not associated with exercise duration, PaCO2 had a negative association with exercise duration.
GOLD Criteria [14]. Patients with any obstruction were older (52 ± 13 vs. 34 ± 9 years, P b 0.001). There was no difference in smokers (N = 2, one had obstructive physiology and the other did not), or patients on beta blockers (N = 4). No difference in exercise capacity was found between patients with or without obstruction. Total lung capacity (TLC), obtained in 30 patients, was 4.96 ± 1.02 L, or 99 ± 13%pred (range 77–130%). For the group, mean residual volume (RV) was 2.03 ± 0.57 L, or 126 ± 27%pred, range 84– 199%pred. Vital capacity (VC), calculated as TLC–RV, was 2.96 ± 0.84 L. RV as a percentage of TLC, an indicator of the degree of air trapping, was 41 ± 10%, but did not correlate with any measure of exercise capacity. Only two people met criteria for mild restrictive lung disease (TLC 78% and 77%pred), one of whom had grade 2 obstructive lung disease. Neither had used amiodarone. TLC was not different in patients with obstruction. Of patients with a TLC ≥ 110%pred, alveolar volume was higher (103 ± 13 vs. 86 ± 13%pred, P = 0.011), and FEV1 was lower (87 ± 10 vs 71 ± 14%pred, P = 0.041), although the FEV1/FVC ratio was not different. TLC did not correlate with exertional metrics. No patient had significant scoliosis.
Data were obtained in 28 patients. All were hypoxemic (PaO2 61 ± 35 mm Hg) as expected (Fig. 1). The majority had a respiratory alkalosis (Fig. 3) with varying amounts of metabolic compensation (pH 7.42 ± 0.03, PaCO2 32 ± 4.4 mm Hg, HCO3 20.6 ± 3.2 mmol/L, base excess − 2.8 ± 3.0). Only one patient had a PaCO2 N40 mm Hg. PaCO2 correlated with age (r = 0.440, P = 0.022), such that older patients had a PaCO2 that was closer to normal. PaCO2 and PaO2 did not differ in patients with evidence of obstructive lung disease. PaCO2 also correlated negatively with exercise duration (r = − 0.462, PaO2 did not correlate with exercise duration (Fig. 4A), peak VO2, or walk distance. PaCO2 correlated negatively with exercise duration (Fig. 4B, r = –0.462, P = 0.03), whereas age had a weaker, non-significant association. In a stepwise, multiple regression model with PaCO2 and age as predictors of exercise duration, PaCO2 remained predictive, whereas age did not. PaCO2 was not associated with other exercise parameters, however, including walk distance and Ve/VCO2 slope. 3.5. Survival After a mean follow up of 7.4 ± 0.5 years, outcome of all patients was accounted for. There were 6 deaths. Mean follow up time prior to death was 3.4 ± 3.1 years (range 4 months to 8 years). Two died of congestive heart failure, two were found dead at home (one had been complaining of acute abdominal pain), and one died soon after heart/ lung transplantation. The cause of death in the 6th patient was unknown. Compared to survivors, non-survivors had lower oxygen saturation at rest, PaO2, diffusion capacity, and higher Ve/VCO2 slope (Table 1). Age, exercise duration, and VO2 at peak exercise did not differ. 4. Discussion Our study describes expected respiratory mechanics in adult patients specifically with ES. Generally, obstructive lung disease was fairly common, with physiologic evidence of air trapping. We found reduced diffusion capacity as in other etiologies of PAH [8,15]. While some Table 1 Comparisons of baseline values in patients who died during follow up vs. survivors.
3.2. Diffusion CO transfer factor (DlCOc), obtained in 30 subjects, was 62 ± 16%pred (range 34–108%pred). Specifically, 13 patients had mild reduction (60–80%), 10 moderate (40–60%), and 1 severe (b 40%). Mean alveolar ventilation was 89 ± 15%pred. Patients with moderate-severe DlCOc had lower alveolar ventilation (79 ± 15% vs. 95 ± 13%, P b 0.05). The CO transfer coefficient (KCOc) was also abnormal (mean predicted 75 ± 15%, range 45–111). Fifteen patients (51%) had
Age (years) Resting heart rate (bpm) O2 saturation (%) Diffusion coefficient (% predicted) PaCO2 (mm Hg) PaO2 (mm Hg) Exercise duration (min) VO2 peak (ml/kg/min) Ve/VCO2 slope
Alive (N = 26)
Dead (N = 6)
P
40.3 79 85 78.2 32 50 5.7 11.1 70.2
45.8 89 79 63.5 34 42 5.2 10.8 146.8
0.373 0.167 0.010 0.026 0.424 0.004 0.64 0.82 0.034
± ± ± ± ± ± ± ± ±
12.6 14 5 14.4 4 5 1.8 3.3 19.1
± ± ± ± ± ± ± ± ±
17.6 13 6 13.0 6 4 3.7 4.4 170.3
76
C.S. Broberg et al. / International Journal of Cardiology 171 (2014) 73–77
studies of pulmonary mechanics in patients with PAH of other etiologies showed them to be largely normal [5,6], subsequent work found an association between obstructive pulmonary disease [7,16] and peripheral airway obstruction [17]. Obstructive physiology has been found in PAH, including in some with congenital heart disease [8], yet this study demonstrates the finding specifically in ES, even in the absence of usual etiologies for obstructive lung disease. Mild restriction has been reported in patients with other forms of congenital heart disease [18], including from our group [4], but was surprisingly rare in this study. None of our patients had significant scoliosis, which can be contributory to restriction in other congenital heart patients [4]. More importantly, the majority of Eisenmenger patients have not had prior cardiothoracic surgery. Therefore the relative absence of restriction argues that thoractomy or sternotomy may potentially contribute to chest wall changes and restriction as proposed elsewhere [19]. Mixed restrictive/obstructive lung disease was not reported among congenital heart disease patients with PAH [7]. Another novel finding in our study was that the majority of ES patients had a compensated respiratory alkalosis. Dissolved CO2 in venous blood that shunts right to left should in theory raise measured PaCO2 levels. But our data suggest that this amount is negligible at rest given that most have low PaCO2 levels. The respiratory alkalosis found indicates hyperventilation secondary to hypoxemia. One implication of this finding is that since PaCO2 rises towards normal in older patients, hyperventilation may diminish over time. Yet this is not a longitudinal study and cannot conclude this definitively. Other plausible explanations for less respiratory alkalosis in older individuals are that muscular mechanics decrease over time and may be responsible for the more normal PaCO2 in older patients, or alternatively, alveolar dead space may increase, as can occur in normal individuals with age. Peak expiratory flow correlated with age in our study, in support of age-related changes in ventilatory efficiency that may account for the slow rise in PaCO2. Furthermore, pulmonary vascular changes progress over time, which in turn may lead to increased CO2 shunting. Intriguingly, hypocapnia was associated with better exercise capacity. This raises several questions regarding the physiology of exercise in this unique group. We have previously shown that Eisenmenger patients have the highest Ve/VCO2 slope during exercise of any other congenital heart group [13], either due to right to left shunting of CO2 (i.e. lower VCO2), or from increased ventilation per work performed (i.e. higher Ve), which may indicate that patients reach their respiratory capacity more quickly. A high Ve/VCO2 is also a feature in other PAH patients [13], and likely due to worsening dead space as patients increase ventilation but not pulmonary perfusion [20]. While PaCO2 was associated with exercise duration (though not other parameters of exercise capacity), PaO2 was not. One important difference between PaO2 and PaCO2 is the fact that O2 delivery is compensated for by secondary erythrocytosis, whereas increased hemoglobin has no impact on PaCO2. Therefore, the combined findings suggest that exercise capacity in ES patients may have less to do with tissue oxygen delivery than CO2 clearance. Further, our data show that results of submaximal and maximal exercise tests are not necessarily comparable; there was a poor correlation between 6 minute walk test distance and Peak VO2. This enforces the interpretation that other factors such as ventilatory capacity and CO2 delivery are more rate-limiting than tissue oxygen delivery. Many patients did not achieve a respiratory exchange ratio N 1.1, despite maximal effort, which may also be consistent with CO2 clearance limiting exercise. The hypocapnia finding also highlights the necessity of avoiding potential causes of hypoventilation. Relevant causes to consider are central causes such as sedation or narcotic use, excessive oxygen supplementation, mechanical effects such as pneumonia or pulmonary edema that decrease alveolar ventilation, diaphragmatic palsy, or neuromuscular weakness and deconditioning. Over a mean follow up of more than 7 years, mortality was 18%, which is in harmony with previous data, and better than reported
mortality in idiopathic PAH [21]. PaCO2 was associated with exercise duration yet not mortality, while the inverse was true for PaO2. Lower PaO2 may be a surrogate of chronic tissue ischemia, including kidneys or the myocardium with their associated adverse impact on outcome including survival. In support of this hypothesis, we recently demonstrated elevated BNP tends to be associated with increased mortality in ES [22]. 4.1. Study limitations This is an exploratory, descriptive study with inherent limitations. Although the study represents an unselected group of patients, the relatively small number and the heterogeneity of this cohort limit applicability to larger populations. Serial measurement of gas exchange would provide more insight into changes with age. Eisenmenger patients have the lowest exercise capacity of any group of congenital heart defects [23] and abnormal pulmonary function is only one such factor among many that likely affect exertional capacity. None of our findings should be interpreted as the primary cause of exertional limitation, but were included here to discover their potential significance. Not every patient was able to complete each portion of the protocol, namely those with developmental delay, yet we felt it important to include these individuals in the analysis. We performed earlobe capillary blood gas, which has been shown to have excellent correlation with radial arterial blood gas at rest and during exercise, and only trivial differences during hypoxia [24]. 5. Conclusions Obstructive lung disease and diffusion abnormalities are common findings in ES, though not universal. Patients exhibit a compensated respiratory alkalosis with PaCO2 levels that seem to rise with age. Patients with lower PaCO2 have a better exercise capacity. The data pose several questions about the nature of CO2 clearance and its relation to functional capacity in ES. References [1] Vongpatanasin W, Brickner ME, Hillis LD, Lange RA. The Eisenmenger syndrome in adults. Ann Intern Med 1998;128:745–55. [2] Broberg C, Ujita M, Babu-Narayan S, et al. Massive pulmonary artery thrombosis with haemoptysis in adults with Eisenmenger's syndrome: a clinical dilemma. Heart 2004;90:e63. [3] Inuzuka R, Diller GP, Borgia F, et al. Comprehensive use of cardiopulmonary exercise testing identifies adults with congenital heart disease at increased mortality risk in the medium term. Circulation 2012;125:250–9. [4] Alonso-Gonzalez R, Borgia F, Diller GP, et al. Abnormal lung function in adults with congenital heart disease: prevalence, relation to cardiac anatomy, and association with survival. Circulation 2013;127:882–90. [5] Williams Jr MH, Adler JJ, Colp C. Pulmonary function studies as an aid in the differential diagnosis of pulmonary hypertension. Am J Med 1969;47:378–83. [6] Wessel HU, Kezdi P, Cugell DW. Respiratory and cardiovascular function in patients with severe pulmonary hypertension. Circulation 1964;29:825–32. [7] Jing ZC, Xu XQ, Badesch DB, et al. Pulmonary function testing in patients with pulmonary arterial hypertension. Respir Med 2009;103:1136–42. [8] Romano AM, Tomaselli S, Gualtieri G, et al. Respiratory function in precapillary pulmonary hypertension. Monaldi Arch Chest Dis 1993;48:201–4. [9] Crapo RO, Jensen RL. Standards and interpretive issues in lung function testing. Respir Care 2003;48:764–72. [10] Broberg CS, Bax BE, Okonko DO, et al. Blood viscosity and its relation to iron deficiency, symptoms, and exercise capacity in adults with cyanotic congenital heart disease. J Am Coll Cardiol 2006;48:356–65. [11] Broberg CS, Ujita M, Prasad S, et al. Pulmonary arterial thrombosis in Eisenmenger syndrome is associated with biventricular dysfunction and decreased pulmonary flow velocity. J Am Coll Cardiol 2007;50:634–42. [12] Diller GP, Dimopoulos K, Okonko D, et al. Exercise intolerance in adult congenital heart disease: comparative severity, correlates, and prognostic implication. Circulation 2005;112:828–35. [13] Dimopoulos K, Okonko DO, Diller GP, et al. Abnormal ventilatory response to exercise in adults with congenital heart disease relates to cyanosis and predicts survival. Circulation 2006;113:2796–802. [14] http://www.goldcopd.org/guidelines-global-strategy-for-diagnosis-management. html.
C.S. Broberg et al. / International Journal of Cardiology 171 (2014) 73–77 [15] Kiakouama L, Cottin V, Glerant JC, Bayle JY, Mornex JF, Cordier JF. Conditions associated with severe carbon monoxide diffusion coefficient reduction. Respir Med 2011;105:1248–56. [16] Lupi-Herrera E, Sandoval J, Seoane M, Bialostozky D. Behavior of the pulmonary circulation in chronic obstructive pulmonary disease. Pathogenesis of pulmonary arterial hypertension at an attitude of 2,240 meters. Am Rev Respir Dis 1982;126:509–14. [17] Fernandez-Bonetti P, Lupi-Herrera E, Martinez-Guerra ML, Barrios R, Seoane M, Sandoval J. Peripheral airways obstruction in idiopathic pulmonary artery hypertension (primary). Chest 1983;83:732–8. [18] Rowe SA, Zahka KG, Manolio TA, Horneffer PJ, Kidd L. Lung function and pulmonary regurgitation limit exercise capacity in postoperative tetralogy of fallot. J Am Coll Cardiol 1991;17:461–6. [19] Zaqout M, De Baets F, Schelstraete P, et al. Pulmonary function in children after surgical and percutaneous closure of atrial septal defect. Pediatr Cardiol 2010;31:1171–5.
77
[20] Reybrouck T, Mertens L, Schulze-Neick I, et al. Ventilatory inefficiency for carbon dioxide during exercise in patients with pulmonary hypertension. Clin Physiol 1998;18:337–44. [21] Hopkins WE, Ochoa LL, Richardson GW, Trulock EP. Comparison of the hemodynamics and survival of adults with severe primary pulmonary hypertension or Eisenmenger syndrome. J Heart Lung Transplant 1996;15:100–5. [22] Diller GP, Alonso-Gonzalez R, Kempny A, et al. B-type natriuretic peptide concentrations in contemporary Eisenmenger syndrome patients: predictive value and response to disease targeting therapy. Heart 2012 May;98(9):736–42. [23] Diller GP, Gatzoulis MA. Pulmonary vascular disease in adults with congenital heart disease. Circulation 2007;115:1039–50. [24] Mollard P, Bourdillon N, Letournel M, et al. Validity of arterialized earlobe blood gases at rest and exercise in normoxia and hypoxia. Respir Physiol Neurobiol 2010;172:179–83.
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Lung function and gas exchange in Eisenmenger syndrome and their impact on exercise capacity and survival☆ Craig S. Broberg a,⁎,1, Ryan C. Van Woerkom a,1, Elizabeth Swallow b,1, Kostas Dimopoulos d,1, Gerhard-Paul Diller d,1, Gopal Allada c,1, Michael A. Gatzoulis d,1 a
Adult Congenital Heart Disease Program, Division of Cardiovascular Medicine, Oregon Health and Science University, Portland, OR, United States Respiratory Muscle Laboratory, Royal Brompton Hospital and Harefield NHS Foundation Trust, National Heart and Lung Institute, Imperial College, London, UK Division of Pulmonology, Oregon Health and Science University, Portland, OR, United States d NIHR Cardiovascular Biomedical Research Unit, Royal Brompton and Harefield NHS Foundation Trust, National Heart and Lung Institute, Imperial College London, UK b c
a r t i c l e
i n f o
Article history: Received 2 May 2013 Received in revised form 13 November 2013 Accepted 18 November 2013 Available online 23 November 2013 Keywords: Eisenmenger Cyanosis Pulmonary arterial hypertension Lung function Blood gas Hypocapnia
a b s t r a c t Background: Eisenmenger physiology may contribute to abnormal pulmonary mechanics and gas exchange and thus impaired functional capacity. We explored the relationship between lung function and gas exchange parameters with exercise capacity and survival. Methods: Stable adult patients with Eisenmenger syndrome (N = 32) were prospectively studied using spirometry, lung volumes, diffusion capacity, and blood gas analysis, as well as same day measurement of 6-minute walk distance and cardiopulmonary maximal treadmill exercise. Patients were followed prospectively to determine survival (7.4 ± 0.5 years). Abnormalities were identified and appropriate comparisons were made between affected and unaffected individuals between respiratory mechanics, exercise function, and survival. Results: Obstruction (FEV1/FVC ratio b0.70) was found in 13 patients (41%), who were older but not otherwise different. Restriction was uncommon. Diffusion transfer coefficient, which was b 80% in half the patients, correlated with exercise duration (r = 0.542, P = 0.005), and was worse in non-survivors (N = 6). Nearly all patients had a compensated respiratory alkalosis (PaCO2 32 ± 4.4 mm Hg). PaCO2 was less reduced in older patients (r = 0.438, P = 0.022), and correlated independently with exercise duration (R = − 0.463, P = 0.03), yet PaO2, not PaCO2, was associated with survival. Conclusions: Eisenmenger patients show evidence of obstructive lung disease, diffusion abnormalities, and hypocapnia; likely from hyperventilation. Understanding expected lung mechanics and gas exchange may facilitate more appropriate clinical management. © 2013 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Eisenmenger syndrome (ES) develops from an unrestricted communication between the systemic and pulmonary circulation (such as a large intracardiac shunt or patent ductus arteriosus), which if unrepaired progresses to elevated pulmonary artery pressures, pulmonary vascular disease, and reversal of shunt leading to hypoxemia, cyanosis, and erythrocytosis [1]. There are several potential mechanisms ☆ Support: This work was supported by a grant from the Waring Trust. Dr. Broberg is supported by a Clinical Research Grant (K23HL093024) from the National Heart, Lung and Blood Institute, United States. The Royal Brompton Hospital Adult Congenital Heart Disease Center and Centre for Pulmonary Hypertension have received support from the British Heart Foundation. ⁎ Corresponding author at: UHN 62, 3181 SW Sam Jackson Pk Rd., Portland, OR 97239, United States. Tel.: +1 503 494 8750; fax: +1 503 494 8550. E-mail address: [email protected] (C.S. Broberg). 1 Each author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation. 0167-5273/$ – see front matter © 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijcard.2013.11.047
for abnormal lung mechanics, in addition to the pulmonary vascular changes themselves. These include scoliosis, present in roughly 10% of patients with congenital heart disease, or significant enlargement of the heart and/or central pulmonary arteries that may have spaceoccupying effects [2]. Patients with ES have very impaired exertional capacity, the lowest of any group of patients with congenital heart disease [3]. It is unknown to what extent, if any, lung function contributes to this physical limitation. Our institution recently published a large series evaluating restrictive lung disease in congenital heart patients including ES [4]. However, patients with obstructive lung disease were excluded, and the study did not focus on the unique aspects of ES. While there are few studies describing pulmonary function parameters such as obstructive lung disease in other groups with pulmonary arterial hypertension (PAH) [5–7], including a small number of patients with Eisenmenger physiology [8], none, to our knowledge, have specifically explored lung mechanics specifically in ES, nor their potential impact on exercise capacity or survival. Therefore, in this prospective study we aim to
74
C.S. Broberg et al. / International Journal of Cardiology 171 (2014) 73–77
describe all aspects of lung mechanics and gas exchange specifically in a stable cohort of ES patients, and explore their prognostic value. 2. Methods Consecutive adults with ES seen at our tertiary referral center between 2003 and 2005 were prospectively invited to participate and all patients gave informed consent. The study received approval from the Ethics Committee at the Royal Brompton Hospital. Inclusion criteria were known ES syndrome patients with a resting oxygen saturation b92%, seen as stable outpatients. Inpatients or those with acute illness such as bronchitis/pneumonia or recent large-volume hemoptysis were initially excluded, but invited to participate after resolution of the issue and return to outpatient status. Patients with difficulty complying with test instructions (such as those with developmental delay) were excluded from tests needing cooperation (i.e. spirometry or cardiopulmonary exercise) but eligible for other portions of the study and subsequent analysis. Each patient underwent spirometry, lung volumes, diffusion capacity quantification, and earlobe capillary blood gas measured at rest. All testing was done in a single visit and in accordance with accepted clinical standards [9]. Spirometry was performed three times and results averaged. Lung volumes were measured in an enclosed gas cabin using Helium dilution. Diffusion of carbon monoxide (CO) was quantified and corrected for hemoglobin measured on the test day. Patients also underwent a 6-minute walk test and a maximal cardiopulmonary exercise testing on the test day. Oxygen saturation and heart rate at baseline and at the end of the 6 minute walk test were recorded in addition to distance walked. Maximal cardiopulmonary treadmill exercise testing was performed using a modified Bruce protocol with continuous measurement of oxygen consumption (VO2) and carbon dioxide production (VCO2) [10,11]. All patients were followed prospectively at the Royal Brompton Hospital. Any deaths were recorded together with date and mode of death, if known. For survivors, date of last clinic visit was used for survival analysis. All patients were accounted for within six months of the time of final analysis. 2.1. Statistical analysis Spirometry, lung volume, and diffusion results were reported as percent of predicted (%pred) values for historic normal individuals matched for age, gender, and height. This allowed categorization into accepted severity categories (mild/moderate/severe), defined below. Statistical comparisons between categories, when described, were made using appropriate parametric testing (Students t-test or chi-square test). Strength of association between continuous variables was measured using Pearson's correlation coefficient. A single forward step-wise multivariate analysis was done to determine the potential interaction of PaCO2 and age as predictors of exercise duration. Analysis was performed using SPSS for Windows version 11.0 (SPSS Inc., Chicago, Illinois). No correction was made for multiple tests as this was a descriptive study. Because there were only 6 deaths, survival analysis was limited to simple comparisons between survivors and non-survivors. Results are expressed as mean ± SD unless stated otherwise. P b 0.05 was considered statistically significant.
Fig. 1. Box plots showing mean, interquartile range, maximum and minimum values for major lung function variables, expressed as a percent of predicted values for age/gender matched populations. Outliers are shown as open circles. FEV1 = force expiratory volume in 1 s. FVC = forced vital capacity. TLC = total lung capacity. RV = residual volume. VA = alveolar volume. IVC = inspiratory vital capacity, KCO = carbon monoxide diffusion coefficient, KCOC = carbon monoxide diffusion coefficient corrected for measured hemoglobin, PEF = peak expiratory flow, PIF = peak inspiratory flow, VO2 = peak oxygen consumption.
Fig. 2. Association between diffusion coefficient (corrected for hemoglobin) and exercise duration during cardiopulmonary exercise.
3. Results Thirty-two patients were enrolled in the study, of which 23 (71%) were female. Mean age was 41 ± 14 years, resting oxygen saturation 81 ± 7%, hemoglobin 20 ± 3 g/dl, all consistent with confirmed Eisenmenger physiology. Three patients had atrial septal defects, 20 had ventricular septal defects, and 9 had a patent ductus arteriosus as their main source of shunting. Not all patients could or wished to complete all aspects of the study. Six patients had some form of developmental delay. Boxplots for major lung function variables expressed as a percentage of predicted values are shown (Fig. 1), including percent predicted peak VO2 during exercise. Additionally, peak VO2 (N = 27) was 11.1 ± 3.4 ml/kg/min, Ve/VCO2 slope was 82 ± 68, anaerobic threshold was 8.76 ± 2.5 ml/kg/min, and respiratory exchange ratio was 0.98 ± 0.1, all consistent with significantly reduced exercise capacity as previously documented [12,13]. 3.1. Obstructive or restrictive lung disease Reliable spirometry data were obtained in 31 patients. Forced Expiratory Volume (FEV1) was 3.86 ± 1.05 L, or 75 ± 16%pred, range 39– 111%. Forced vital capacity (FVC) was 4.62 ± 1.65 L, or 89 ± 18%pred, range 44–116%. There were 13 patients (42%) with evidence of obstruction, defined as FEV1/FVC ratio b0.70. Based on the %pred FEV1, there was 1 patient with grade 1 (FEV1 N 80%pred), 11 with grade 2 (FEV1 50– 80%pred), and 1 with grade 3 (FEV1 b 50%pred) according to the 2011
Fig. 3. Scatterplot of PaCO2 and pH demonstrating acid/base status of the patient cohort. The majority of patients had a respiratory alkalosis with varying degrees of metabolic compensation.
C.S. Broberg et al. / International Journal of Cardiology 171 (2014) 73–77
75
b80%pred. In these, 6 minute walk distance tended to be lower (352 ± 80 vs. 423 ± 122 m, P = 0.071) and exercise duration was significantly lower (4.5 ± 1.3 vs.7.0 ± 2.1 min, P b 0.001). There was a positive correlation between KCOc and exercise duration (r = 0.542, P = 0.005) (Fig. 2). 3.3. Other pulmonary function testing Peak inspiratory and expiratory flows (N = 32) were slightly reduced at 80 ± 16 and 77 ± 23%pred values, respectively. Peak expiratory flow correlated with 6 minute walk distance (r = 0.360, P = 0.046), and negatively with age (r = −0.477, P = 0.007), as did peak inspiratory flow (r = −0.392, P = 0.029 respectively). However, neither variable correlated with any measure of maximal exercise on cardiopulmonary exercise testing. 3.4. Acid–base status
Fig. 4. Relationship of PaO2 (A) and PaCO2 (B) and exercise duration on maximum treadmill exercise testing. Although PaO2 was not associated with exercise duration, PaCO2 had a negative association with exercise duration.
GOLD Criteria [14]. Patients with any obstruction were older (52 ± 13 vs. 34 ± 9 years, P b 0.001). There was no difference in smokers (N = 2, one had obstructive physiology and the other did not), or patients on beta blockers (N = 4). No difference in exercise capacity was found between patients with or without obstruction. Total lung capacity (TLC), obtained in 30 patients, was 4.96 ± 1.02 L, or 99 ± 13%pred (range 77–130%). For the group, mean residual volume (RV) was 2.03 ± 0.57 L, or 126 ± 27%pred, range 84– 199%pred. Vital capacity (VC), calculated as TLC–RV, was 2.96 ± 0.84 L. RV as a percentage of TLC, an indicator of the degree of air trapping, was 41 ± 10%, but did not correlate with any measure of exercise capacity. Only two people met criteria for mild restrictive lung disease (TLC 78% and 77%pred), one of whom had grade 2 obstructive lung disease. Neither had used amiodarone. TLC was not different in patients with obstruction. Of patients with a TLC ≥ 110%pred, alveolar volume was higher (103 ± 13 vs. 86 ± 13%pred, P = 0.011), and FEV1 was lower (87 ± 10 vs 71 ± 14%pred, P = 0.041), although the FEV1/FVC ratio was not different. TLC did not correlate with exertional metrics. No patient had significant scoliosis.
Data were obtained in 28 patients. All were hypoxemic (PaO2 61 ± 35 mm Hg) as expected (Fig. 1). The majority had a respiratory alkalosis (Fig. 3) with varying amounts of metabolic compensation (pH 7.42 ± 0.03, PaCO2 32 ± 4.4 mm Hg, HCO3 20.6 ± 3.2 mmol/L, base excess − 2.8 ± 3.0). Only one patient had a PaCO2 N40 mm Hg. PaCO2 correlated with age (r = 0.440, P = 0.022), such that older patients had a PaCO2 that was closer to normal. PaCO2 and PaO2 did not differ in patients with evidence of obstructive lung disease. PaCO2 also correlated negatively with exercise duration (r = − 0.462, PaO2 did not correlate with exercise duration (Fig. 4A), peak VO2, or walk distance. PaCO2 correlated negatively with exercise duration (Fig. 4B, r = –0.462, P = 0.03), whereas age had a weaker, non-significant association. In a stepwise, multiple regression model with PaCO2 and age as predictors of exercise duration, PaCO2 remained predictive, whereas age did not. PaCO2 was not associated with other exercise parameters, however, including walk distance and Ve/VCO2 slope. 3.5. Survival After a mean follow up of 7.4 ± 0.5 years, outcome of all patients was accounted for. There were 6 deaths. Mean follow up time prior to death was 3.4 ± 3.1 years (range 4 months to 8 years). Two died of congestive heart failure, two were found dead at home (one had been complaining of acute abdominal pain), and one died soon after heart/ lung transplantation. The cause of death in the 6th patient was unknown. Compared to survivors, non-survivors had lower oxygen saturation at rest, PaO2, diffusion capacity, and higher Ve/VCO2 slope (Table 1). Age, exercise duration, and VO2 at peak exercise did not differ. 4. Discussion Our study describes expected respiratory mechanics in adult patients specifically with ES. Generally, obstructive lung disease was fairly common, with physiologic evidence of air trapping. We found reduced diffusion capacity as in other etiologies of PAH [8,15]. While some Table 1 Comparisons of baseline values in patients who died during follow up vs. survivors.
3.2. Diffusion CO transfer factor (DlCOc), obtained in 30 subjects, was 62 ± 16%pred (range 34–108%pred). Specifically, 13 patients had mild reduction (60–80%), 10 moderate (40–60%), and 1 severe (b 40%). Mean alveolar ventilation was 89 ± 15%pred. Patients with moderate-severe DlCOc had lower alveolar ventilation (79 ± 15% vs. 95 ± 13%, P b 0.05). The CO transfer coefficient (KCOc) was also abnormal (mean predicted 75 ± 15%, range 45–111). Fifteen patients (51%) had
Age (years) Resting heart rate (bpm) O2 saturation (%) Diffusion coefficient (% predicted) PaCO2 (mm Hg) PaO2 (mm Hg) Exercise duration (min) VO2 peak (ml/kg/min) Ve/VCO2 slope
Alive (N = 26)
Dead (N = 6)
P
40.3 79 85 78.2 32 50 5.7 11.1 70.2
45.8 89 79 63.5 34 42 5.2 10.8 146.8
0.373 0.167 0.010 0.026 0.424 0.004 0.64 0.82 0.034
± ± ± ± ± ± ± ± ±
12.6 14 5 14.4 4 5 1.8 3.3 19.1
± ± ± ± ± ± ± ± ±
17.6 13 6 13.0 6 4 3.7 4.4 170.3
76
C.S. Broberg et al. / International Journal of Cardiology 171 (2014) 73–77
studies of pulmonary mechanics in patients with PAH of other etiologies showed them to be largely normal [5,6], subsequent work found an association between obstructive pulmonary disease [7,16] and peripheral airway obstruction [17]. Obstructive physiology has been found in PAH, including in some with congenital heart disease [8], yet this study demonstrates the finding specifically in ES, even in the absence of usual etiologies for obstructive lung disease. Mild restriction has been reported in patients with other forms of congenital heart disease [18], including from our group [4], but was surprisingly rare in this study. None of our patients had significant scoliosis, which can be contributory to restriction in other congenital heart patients [4]. More importantly, the majority of Eisenmenger patients have not had prior cardiothoracic surgery. Therefore the relative absence of restriction argues that thoractomy or sternotomy may potentially contribute to chest wall changes and restriction as proposed elsewhere [19]. Mixed restrictive/obstructive lung disease was not reported among congenital heart disease patients with PAH [7]. Another novel finding in our study was that the majority of ES patients had a compensated respiratory alkalosis. Dissolved CO2 in venous blood that shunts right to left should in theory raise measured PaCO2 levels. But our data suggest that this amount is negligible at rest given that most have low PaCO2 levels. The respiratory alkalosis found indicates hyperventilation secondary to hypoxemia. One implication of this finding is that since PaCO2 rises towards normal in older patients, hyperventilation may diminish over time. Yet this is not a longitudinal study and cannot conclude this definitively. Other plausible explanations for less respiratory alkalosis in older individuals are that muscular mechanics decrease over time and may be responsible for the more normal PaCO2 in older patients, or alternatively, alveolar dead space may increase, as can occur in normal individuals with age. Peak expiratory flow correlated with age in our study, in support of age-related changes in ventilatory efficiency that may account for the slow rise in PaCO2. Furthermore, pulmonary vascular changes progress over time, which in turn may lead to increased CO2 shunting. Intriguingly, hypocapnia was associated with better exercise capacity. This raises several questions regarding the physiology of exercise in this unique group. We have previously shown that Eisenmenger patients have the highest Ve/VCO2 slope during exercise of any other congenital heart group [13], either due to right to left shunting of CO2 (i.e. lower VCO2), or from increased ventilation per work performed (i.e. higher Ve), which may indicate that patients reach their respiratory capacity more quickly. A high Ve/VCO2 is also a feature in other PAH patients [13], and likely due to worsening dead space as patients increase ventilation but not pulmonary perfusion [20]. While PaCO2 was associated with exercise duration (though not other parameters of exercise capacity), PaO2 was not. One important difference between PaO2 and PaCO2 is the fact that O2 delivery is compensated for by secondary erythrocytosis, whereas increased hemoglobin has no impact on PaCO2. Therefore, the combined findings suggest that exercise capacity in ES patients may have less to do with tissue oxygen delivery than CO2 clearance. Further, our data show that results of submaximal and maximal exercise tests are not necessarily comparable; there was a poor correlation between 6 minute walk test distance and Peak VO2. This enforces the interpretation that other factors such as ventilatory capacity and CO2 delivery are more rate-limiting than tissue oxygen delivery. Many patients did not achieve a respiratory exchange ratio N 1.1, despite maximal effort, which may also be consistent with CO2 clearance limiting exercise. The hypocapnia finding also highlights the necessity of avoiding potential causes of hypoventilation. Relevant causes to consider are central causes such as sedation or narcotic use, excessive oxygen supplementation, mechanical effects such as pneumonia or pulmonary edema that decrease alveolar ventilation, diaphragmatic palsy, or neuromuscular weakness and deconditioning. Over a mean follow up of more than 7 years, mortality was 18%, which is in harmony with previous data, and better than reported
mortality in idiopathic PAH [21]. PaCO2 was associated with exercise duration yet not mortality, while the inverse was true for PaO2. Lower PaO2 may be a surrogate of chronic tissue ischemia, including kidneys or the myocardium with their associated adverse impact on outcome including survival. In support of this hypothesis, we recently demonstrated elevated BNP tends to be associated with increased mortality in ES [22]. 4.1. Study limitations This is an exploratory, descriptive study with inherent limitations. Although the study represents an unselected group of patients, the relatively small number and the heterogeneity of this cohort limit applicability to larger populations. Serial measurement of gas exchange would provide more insight into changes with age. Eisenmenger patients have the lowest exercise capacity of any group of congenital heart defects [23] and abnormal pulmonary function is only one such factor among many that likely affect exertional capacity. None of our findings should be interpreted as the primary cause of exertional limitation, but were included here to discover their potential significance. Not every patient was able to complete each portion of the protocol, namely those with developmental delay, yet we felt it important to include these individuals in the analysis. We performed earlobe capillary blood gas, which has been shown to have excellent correlation with radial arterial blood gas at rest and during exercise, and only trivial differences during hypoxia [24]. 5. Conclusions Obstructive lung disease and diffusion abnormalities are common findings in ES, though not universal. Patients exhibit a compensated respiratory alkalosis with PaCO2 levels that seem to rise with age. Patients with lower PaCO2 have a better exercise capacity. The data pose several questions about the nature of CO2 clearance and its relation to functional capacity in ES. References [1] Vongpatanasin W, Brickner ME, Hillis LD, Lange RA. The Eisenmenger syndrome in adults. Ann Intern Med 1998;128:745–55. [2] Broberg C, Ujita M, Babu-Narayan S, et al. Massive pulmonary artery thrombosis with haemoptysis in adults with Eisenmenger's syndrome: a clinical dilemma. Heart 2004;90:e63. [3] Inuzuka R, Diller GP, Borgia F, et al. Comprehensive use of cardiopulmonary exercise testing identifies adults with congenital heart disease at increased mortality risk in the medium term. Circulation 2012;125:250–9. [4] Alonso-Gonzalez R, Borgia F, Diller GP, et al. Abnormal lung function in adults with congenital heart disease: prevalence, relation to cardiac anatomy, and association with survival. Circulation 2013;127:882–90. [5] Williams Jr MH, Adler JJ, Colp C. Pulmonary function studies as an aid in the differential diagnosis of pulmonary hypertension. Am J Med 1969;47:378–83. [6] Wessel HU, Kezdi P, Cugell DW. Respiratory and cardiovascular function in patients with severe pulmonary hypertension. Circulation 1964;29:825–32. [7] Jing ZC, Xu XQ, Badesch DB, et al. Pulmonary function testing in patients with pulmonary arterial hypertension. Respir Med 2009;103:1136–42. [8] Romano AM, Tomaselli S, Gualtieri G, et al. Respiratory function in precapillary pulmonary hypertension. Monaldi Arch Chest Dis 1993;48:201–4. [9] Crapo RO, Jensen RL. Standards and interpretive issues in lung function testing. Respir Care 2003;48:764–72. [10] Broberg CS, Bax BE, Okonko DO, et al. Blood viscosity and its relation to iron deficiency, symptoms, and exercise capacity in adults with cyanotic congenital heart disease. J Am Coll Cardiol 2006;48:356–65. [11] Broberg CS, Ujita M, Prasad S, et al. Pulmonary arterial thrombosis in Eisenmenger syndrome is associated with biventricular dysfunction and decreased pulmonary flow velocity. J Am Coll Cardiol 2007;50:634–42. [12] Diller GP, Dimopoulos K, Okonko D, et al. Exercise intolerance in adult congenital heart disease: comparative severity, correlates, and prognostic implication. Circulation 2005;112:828–35. [13] Dimopoulos K, Okonko DO, Diller GP, et al. Abnormal ventilatory response to exercise in adults with congenital heart disease relates to cyanosis and predicts survival. Circulation 2006;113:2796–802. [14] http://www.goldcopd.org/guidelines-global-strategy-for-diagnosis-management. html.
C.S. Broberg et al. / International Journal of Cardiology 171 (2014) 73–77 [15] Kiakouama L, Cottin V, Glerant JC, Bayle JY, Mornex JF, Cordier JF. Conditions associated with severe carbon monoxide diffusion coefficient reduction. Respir Med 2011;105:1248–56. [16] Lupi-Herrera E, Sandoval J, Seoane M, Bialostozky D. Behavior of the pulmonary circulation in chronic obstructive pulmonary disease. Pathogenesis of pulmonary arterial hypertension at an attitude of 2,240 meters. Am Rev Respir Dis 1982;126:509–14. [17] Fernandez-Bonetti P, Lupi-Herrera E, Martinez-Guerra ML, Barrios R, Seoane M, Sandoval J. Peripheral airways obstruction in idiopathic pulmonary artery hypertension (primary). Chest 1983;83:732–8. [18] Rowe SA, Zahka KG, Manolio TA, Horneffer PJ, Kidd L. Lung function and pulmonary regurgitation limit exercise capacity in postoperative tetralogy of fallot. J Am Coll Cardiol 1991;17:461–6. [19] Zaqout M, De Baets F, Schelstraete P, et al. Pulmonary function in children after surgical and percutaneous closure of atrial septal defect. Pediatr Cardiol 2010;31:1171–5.
77
[20] Reybrouck T, Mertens L, Schulze-Neick I, et al. Ventilatory inefficiency for carbon dioxide during exercise in patients with pulmonary hypertension. Clin Physiol 1998;18:337–44. [21] Hopkins WE, Ochoa LL, Richardson GW, Trulock EP. Comparison of the hemodynamics and survival of adults with severe primary pulmonary hypertension or Eisenmenger syndrome. J Heart Lung Transplant 1996;15:100–5. [22] Diller GP, Alonso-Gonzalez R, Kempny A, et al. B-type natriuretic peptide concentrations in contemporary Eisenmenger syndrome patients: predictive value and response to disease targeting therapy. Heart 2012 May;98(9):736–42. [23] Diller GP, Gatzoulis MA. Pulmonary vascular disease in adults with congenital heart disease. Circulation 2007;115:1039–50. [24] Mollard P, Bourdillon N, Letournel M, et al. Validity of arterialized earlobe blood gases at rest and exercise in normoxia and hypoxia. Respir Physiol Neurobiol 2010;172:179–83.
