Characterization of Patients With Borderline Pulmonary Arterial Pressure

[ Original Research Pulmonary Vascular Disease ] Characterization of Patients With Borderline Pulmonary Arterial Pressure Gabor Kovacs, MD ; Alexander Avian, PhD ; Maria Tscherner, MD ; Vasile Foris, MD ; Gerhard Bachmaier, PhD ; Andrea Olschewski, MD ; and Horst Olschewski, MD, FCCP BACKGROUND: Resting mean pulmonary artery pressure (mpap) values between 20 and 25 mm Hg are above normal but do not fulfill the criteria for pulmonary hypertension (PH). The clinical relevance of such borderline hemodynamics is a matter of discussion. METHODS: We focused on patients who underwent right-sided heart catheterization during rest and exercise for symptoms indicative of PH or due to underlying disease associated with an increased risk for pulmonary arterial hypertension and characterized the patients according to their resting mpap. Patients with manifest PH (mpap 25 mm Hg) were excluded. RESULTS: We included 141 patients, 32 of whom presented with borderline hemodynamics (20, mpap, 25 mm Hg). Borderline patients were older (65.8 12.5 years vs 57.3 12.5 years, P 5.001) and more often had cardiac comorbidities (53% vs 15%, P,.001) or decreased lung function (47% vs 16%, P,.001) as compared with patients with resting mpap, 21 mm Hg. After correction for age, borderline patients had significantly increased pulmonary vascular resistance (2.7 0.7 Wood units vs 1.8 0.8 Wood units, P,.001) and mpap/cardiac output (CO) and transpulmonary gradient/co slopes (both P,.001) as well as lower peak oxygen uptake (16.9 4.6 ml/min/kg vs 20.9 4.7 ml/min/kg, P 5.009) and 6-min walk distance (383 120 m vs 448 92 m, P 5.001). During follow-up (4.4 1.4 years), the mortality rate of borderline patients vs patients with resting mpap, 21 mm Hg was 19% vs 4%. CONCLUSIONS: In patients undergoing right-sided heart catheterization with exclusion of manifest PH, borderline elevation of pulmonary arterial pressure is associated with cardiac and pulmonary comorbidities, decreased exercise capacity, and a poor prognosis. CHEST 2014; 146(6): 1486-1493 Manuscript received January 22, 2014; revision accepted June 2, 2014; originally published Online First June 26, 2014. ABBREVIATIONS: CO 5 cardiac output; mpap 5 mean pulmonary artery pressure; PAP 5 pulmonary artery pressure; PAWP 5 pulmonary artery wedge pressure; PH 5 pulmonary hypertension; PVR 5 pulmonary vascular resistance; TPG 5 transpulmonary gradient AFFILIATIONS: From the Department of Internal Medicine, Division of Pulmonology (Drs Kovacs, Tscherner, Foris, and H. Olschewski), Institute for Medical Informatics, Statistics and Documentation (Drs Avian and Bachmaier), and Department of Experimental Anesthesiology (Dr A. Olschewski), Medical University of Graz; and Ludwig Boltzmann Institute for Lung Vascular Research (Drs Kovacs, Avian, Tscherner, Foris, A. Olschewski, and H. Olschewski), Graz, Austria. Part of this article have been presented in abstract form [Kovacs G, Avian A, Olschewski H. The predictive value of resting pulmonary arterial pressure for exercise hemodynamics. Am J Respir Crit Care Med. 2013;187(suppl):A4704]. FUNDING/SUPPORT: The authors have reported to CHEST that no funding was received for this study. CORRESPONDENCE TO : Gabor Kovacs, MD, Ludwig Boltzmann Institute for Lung Vascular Research, Stiftingtalstrasse 24, 8010 Graz, Austria; e-mail: gabor.kovacs@klinikum-graz.at 2014 AMERICAN COLLEGE OF CHEST PHYSICIANS. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians. See online for more details. DOI: 10.1378/chest.14-0194 1486 Original Research [ 146 # 6 CHEST DECEMBER 2014 ]

Pulmonary hypertension (PH) is a progressive disease characterized by a mean pulmonary artery pressure (mpap) 25 mm Hg at rest. It may lead to right ventricular failure and eventually death. 1 According to some studies, the development of PH may be predicted by pulmonary arterial pressures (PAPs) derived from right-sided heart catheterization (excessive increase of PAP during exercise, 2-4 borderline resting mpap [21-24 mm Hg]) 5 and an increased transpulmonary gradient (TPG). 5 In addition, some studies suggested that these conditions may overlap. 6,7 Although these hemodynamic conditions may be of great clinical interest and potentially represent new targets for therapy, 8,9 there are limited long-term follow-up data, and comorbidities have not been analyzed systematically. We aimed to compare data from a real-life patient population with borderline mpap values with data from patients with normal resting mpap and historical data from healthy control subjects. A further objective was to describe the relation between resting and exercise hemodynamics in the examined patients. Materials and Methods We retrospectively analyzed all consecutive patients of our clinic between 2006 and 2011who underwent right-sided heart catheterization with hemodynamics during rest and exercise. Patients with mpap 25 mm Hg were excluded. Patients were eligible for the study when PH was suspected due to disease associated with an increased risk for pulmonary arterial hypertension (collagen vascular disease, myelodysplastic syndrome, or liver cirrhosis) or when complaints such as dyspnea on exertion could not be explained by heart or lung disease. Right-sided heart catheterization at rest and during exercise with cycle ergometry was routinely performed as described earlier. 10 The zero reference line was placed at the anterior axillary line in all supine measurements. Hemodynamic measurements were available at rest, 25 W, 50 W, and peak exercise. As part of the routine workup, 6-min walk distance and N-terminal pro-brain natriuretic peptide level were assessed; transthoracic echocardiography and pulmonary function tests were performed. Patients unable to exercise were excluded from this analysis. A relevant cardiac comorbidity was predefined as the presence of confirmed coronary heart disease or previous myocardial infarction, chronic atrial fibrillation, arterial hypertension with left ventricular hypertrophy, or impaired systolic left ventricular function (ejection fraction, 50%). A relevant respiratory impairment was predefined as FEV 1 /FVC, 70% or FEV 1, 65% or the presence of OSA treated by CPAP or noninvasive ventilation. A historical control group comprising 51 healthy subjects aged, 50 years with available mpap, cardiac output (CO), and pulmonary artery wedge pressure (PAWP) values at rest and at least at two different exercise levels was used to compare the study patients with completely asymptomatic subjects without any risk factors for PH. 11 Data are presented as mean SD for continuous variables and absolute and relative frequency for categorical data. Patient characteristics were compared with t, x 2, and Fisher exact tests. Correlations with mpap and PAWP values at various exercise levels were sought by using partial correlations controlling for age. Between-group differences were evaluated with age-adjusted analysis of variance. To compare changes in CO related to changes in mpap and pulmonary vascular resistance (PVR) between patients with normal resting mpap or borderline resting mpap and the control group, means with 95% CIs were calculated and plotted. Multiple testing P value adjustment was not performed because this was an exploratory analysis of retrospectively collected data. P,.05 was considered significant. Statistical analysis was performed with SPSS version 20.0.0. (IBM Corporation) software. The study was approved by the local Committee on Biomedical Research Ethics (NR: 25-408 ex 12/13). Results This analysis is based on resting and exercise hemodynamics data of 141 patients (107 women) from our center (age, 59.2 13.0 years; height, 166 8 cm; weight, 72 15 kg). Seventy-three patients were included due to dyspnea, 60 due to collagen vascular disease, and eight due to myelodysplastic syndrome or liver cirrhosis with or without dyspnea. Of the 141 patients, 109 had mpap, 21 mm Hg and 32 had mpap between 21 and 24 mm Hg at rest. Eighty-nine patients had no cardiac or pulmonary disease, whereas 52 had a relevant pulmonary or cardiac comorbidity; these patients were generally characterized by higher mpap and decreased exercise capacity compared with patients with no relevant pulmonary or cardiac comorbidities ( Table 1 ). Age was positively correlated with resting and exercise mpap (rest, r 5 0.32; 25 W, r 5 0.48; 50 W, r 5 0.43; maximal exercise, r 5 0.34; all P,.001) and with exercise PAWP (25 W, r 5 0.31 [ P,.001]; 50 W, r 5 0.29 [ P 5.001]; maximal exercise, r 5 0.20 [ P 5.02]) but not with resting PAWP ( P 5.41). All the following correlations were corrected for age. Patients With Borderline vs Normal Resting mpap Patients with borderline resting mpap (21-24 mm Hg) compared with patients with normal resting mpap ( 20 mm Hg) were older (65.8 12.5 years vs 57.3 12.5 years, P 5.001) and more often presented with a cardiac comorbidity (53% vs 15%, P,.001) or a respiratory limitation (47% vs 16%, P,.001). Patients with borderline resting mpap had elevated resting PVR (2.7 0.7 Wood units vs 1.8 0.8 Wood units, P,.001), resting PAWP (9.6 3.2 mm Hg vs 6.8 2.5 mm Hg, P,.001), and TPG (12.5 3.3 mm Hg vs 8.4 2.8 mm Hg, P,.001) but similar CO compared with patients with normal resting mpap ( Table 2 ). During exercise, the mpap/co slope in the borderline resting mpap group was steeper than in the normal journal.publications.chestnet.org 1487

TABLE 1 ] Major Clinical Characteristics in Patients Without vs With Cardiac and Pulmonary Limitations Characteristic Cardiac Limitations No (n 5 108) Yes (n 5 33) Pulmonary Limitations Pulmonary Limitations No (n 5 89) Yes (n 5 19) No (n 5 20) Yes (n 5 13) mpap, mm Hg 15.3 3.4 18.0 4.2 19 3 20.9 2.3 CO, L/min 5.1 1.2 5.2 1.4 4.4 1.5 5.0 1.5 Cardiac index, L/min/m 2 2.9 0.6 2.8 0.7 2.4 0,7 2.7 0.7 PAWP, mm Hg 7.0 2.7 7.2 3.1 8.7 3.6 8.5 3.1 PVR, WU 1.7 0.7 2.2 0.9 2.6 0.8 2.7 0.8 Heart rate, beats/min 72 11 75 13 73 17 70 15 SBP, mm Hg 130 17 123 16 127 19 132 27 FEV 1 /FVC, % 81 6 72 14 80 5 65 15 D LCO cv A, % predicted 86 17 81 25 90 23 85 26 FEV 1, % predicted 93 15 64 20 89 18 62 16 Peak O 2, ml/kg min 21.0 4.9 20.1 4.1 16.7 3.6 15.5 2.4 NT-proBNP, pg/ml 156 129 125 138 831 988 898 1,272 6MWD, m 457 85 427 117 384 112 328 113 Data are presented as mean SD. 6MWD 5 6-min walk distance; CO 5 cardiac output; D LCO cv A 5 diffusing capacity of lung for carbon monoxide for alveolar volume corrected for hemoglobin; mpap 5 mean pulmonary artery pressure; NT-proBNP 5 N-terminal pro-brain natriuretic peptide; PAWP 5 pulmonary artery wedge pressure; PVR 5 pulmonary vascular resistance; SBP 5 systolic BP; O 2 5 oxygen uptake; WU 5 Wood unit. resting mpap group ( Fig 1 ). The median mpap/co slope (changes in mpap divided by the changes in CO from rest to 50 W) was 5.2 mm Hg/L/min (3.0-7.2 mm Hg/L/min for 25th-75th percentile) in the borderline group and 3.2 mm Hg/L/min (2.2-4.4 for 25th-75th percentile) in the normal group ( P,.001). Notably, for the whole patient population, the mpap/co slope during exercise correlated with resting mpap ( r 5 0.37, P,.001) (e-fig 1). Similarly, the TPG/CO slope was steeper in patients with borderline resting mpap than in patients with normal resting mpap ( P 5.001) ( Fig 2 ), and the TPG/CO slope during exercise correlated with resting mpap ( r 5 0.39, P,.001) (e-fig 2). The median TPG/CO slope was 2.5 mm Hg/L/min (1.7-3.5 mm Hg/L/min for 25th-75th percentile) for patients with borderline resting mpap and 1.3 mm Hg/L/min (0.9-1.9 mm Hg/L/min for 25th-75th percentile) for patients with normal resting mpap. PVR showed a minimal, statistically nonsignificant decrease during exercise in patients with borderline resting mpap ( 21.8 23.3% from rest to 50 W, P 5.54) and a moderate, but significant decrease in patients with normal resting mpap ( 27.5 23.1% from rest to 50 W, P 5.05) ( Fig 3 ). Patients with borderline compared with those with normal resting mpap had lower peak oxygen uptake (16.9 4.6 ml/min/kg vs 20.9 4.7 ml/min/kg, P 5.009) and shorter 6-min walk distances (383 120 m vs 448 92 m, P 5.001) ( Table 2 ). Median follow-up was 4.2 years (range, 1.5-7.5 years) for all patients, 4.4 years (range 1.9-7.5 years) for patients with borderline resting mpap, and 3.6 years (range 1.5-7.5 years) for patients with normal resting mpap. During this period, eight of the 32 patients with borderline resting mpap and 29 of the 109 patients with normal resting mpap underwent right-sided heart catheterization. During the observation period, manifest PH (mpap 25 mm Hg) developed in four patients with borderline and three patients with normal resting mpap. Of the four patients with PH and originally borderline resting mpap, collagen vascular disease was present in one, a relevant cardiac comorbidity (coronary heart disease, atrial fibrillation, and hypertensive heart disease) in three, and a relevant pulmonary comorbidity (lung fibrosis and COPD) in two. Of the three patients with PH and originally normal resting mpap, collagen vascular disease was present in two, and relevant cardiac or pulmonary comorbidities were not present. The proportion of patients dying within 4 years was higher in patients with borderline resting mpap than in those with normal resting mpap. Six patients (19%) with borderline resting mpap died within 4 years, whereas 1488 Original Research [ 146 # 6 CHEST DECEMBER 2014 ]

TABLE 2 ] Major Clinical Characteristics: Patients With Borderline Elevated vs Normal Pulmonary Artery Pressure and Historical Control Subjects P Values Characteristic All Patients a (N 5 141) Resting mpap, 21 mm Hg (n 5 109) Resting mpap 21-24 mm Hg (n 5 32) Historical Control Subjects (n 5 51) Overall, 21 vs 21-24 mm Hg 21-24 mm Hg vs Historical Control Subjects, 21 mm Hg vs Historical Control Subjects Age, y 59.2 13.0 57.3 12.5 65.8 12.5 23.0 4.1,.001.001,.0001,.001 BMI, kg/m 2 26.0 4.9 25.6 4.5 27.5 5.9 21.3 2.1,.001.024,.001,.001 Cardiac disease 33(23) 16 (15) 17 (53)......,.001...... Pulmonary disease 32(23) 17 (16) 15 (47)......,.001...... Collagen vascular disease 60(43) 55 (51) 5 (16)......,.001...... mpap, mm Hg 16.8 4.0 15.2 3.0 22.1 1.1 13.8 2.5,.001,.001,.001.405 CO, L/min 5.0 1.3 5.1 1.3 4.8 1.3 7.3 1.8.044.889.582.020 Cardiac index, L/min/m 2 2.8 0.7 2.8 0.7 2.7 0.7.......708...... PAWP, mm Hg 7.4 2.9 6.8 2.5 9.6 3.2 8.6 2.6,.001,.001.910.373 TPG, mm Hg 9.3 3.4 8.4 2.8 12.5 3.3 5.3 1.5,.001,.001.980,.001 PVR, WU 2.0 0.8 1.8 0.8 2.7 0.7 0.8 0.3,.001,.001,.001.809 Heart rate, /min 73 13 72 12 74 15 77 11.379.254.170.589 SBP, mm Hg 129 18 130 18 123 18.......005...... FEV 1, % predicted 86 20 88 19 77 22......,.001...... D LCO cv A, % predicted 86 19 86 17 86 28.......919...... FEV 1 /FVC, % 78 10 79 8 74 14.......096...... Peak O 2, ml/min/kg 20.3 4.9 20.9 4.7 16.9 4.6.......009...... NT-proBNP, pg/ml 282 547 188 234 653 1,062.......058...... 6MWD, m 434 102 448 92 383 120.......001...... Data are presented as mean SD or No. (%). See Table 1 legend for expansion of abbreviations. a Without historical control subjects. journal.publications.chestnet.org 1489

Figure 1 mpap (mm Hg) vs cardiac output (CO) (L/min) at rest and during exercise in patients with borderline elevated pulmonary artery pressure ( ), patients with normal resting pulmonary artery pressure ( L ), and historical control subjects ( ). mpap 5 mean pulmonary artery pressure. four (4%) died in the normal resting mpap group ( Fig 4 ). Causes of death in the borderline resting mpap group were cardiac decompensation in two patients, lung cancer in one, and traumatic subarachnoid bleeding in one. The remaining two patients died at home, and the true cause of death is not known. Causes of death in the normal resting mpap group were stroke in one patient, cardiac decompensation and concomitant renal insufficiency in one, COPD exacerbation in one, and severe asthma in one. Comparing resting and exercise hemodynamics of the study patients with the historical control group ( Figs 1-3, Table 2 ), the resting mpap and PVR values were lower in the control group than in the borderline Figure 3 PVR (Wood units) vs CO (L/min) at rest and during exercise in patients with borderline elevated pulmonary artery pressure ( ), patients with normal resting pulmonary artery pressure ( L ), and historical control subjects ( ). PVR 5 pulmonary vascular resistance. See Figure 1 legend for expansion of other abbreviation. resting mpap group but (after correction for age) not significantly different from the normal resting mpap group. In the control group, the median mpap/co slope was 0.8 mm Hg/L/min (0.5-1.0 mm Hg/L/min for 25th-75th percentile) ( Fig 1 ), and the median TPG/CO slope was 0.5 mm Hg/L/min (0.3-0.7 mm Hg/L/min for 25% and 75%) ( Fig 2 ). The resistance-compliance relationship was not different among the three groups, resulting in a hyperbolic relationship ( Fig 5 ). Patients with borderline resting mpap had the highest resistance and lowest compliance values. Correlation Between Resting and Exercise Hemodynamics We also analyzed subgroups of patients according to the predefined criteria for respiratory and cardiac comorbidities ( Fig 6, e-fig 3). In patients without a relevant Figure 2 TPG (mm Hg) vs CO (L/min) at rest and during exercise in patients with borderline elevated pulmonary artery pressure ( ), patients with normal resting pulmonary artery pressure ( L ), and historical control subjects ( ). TPG 5 transpulmonary gradient. See Figure 1 legend for expansion of other abbreviation. Figure 4 Survival of patients with borderline elevated mpap (dotted line) and patients with normal resting mpap (solid line). See Figure 1 legend for expansion of abbreviation. 1490 Original Research [ 146 # 6 CHEST DECEMBER 2014 ]

Figure 5 PVC vs PVR relationship in patients with borderline elevated pulmonary artery pressure (red), patients with normal resting pulmonary artery pressure (yellow), and historical control subjects (green). PVC 5 pulmonary vascular compliance; WU 5 Wood unit. See Figure 3 legend for expansion of other abbreviation. respiratory or cardiac comorbidity and in those with just a respiratory limitation, there was a strong correlation between resting and exercise mpap and PAWP. Patients with known cardiac disease, however, showed a weak correlation between resting and exercise mpap and resting and exercise PAWP. In the whole group, resting PVR was strongly correlated with exercise PVR (50 W) ( r 5 0.79, P,.001). This correlation was also significant, although weaker, among patients with cardiac comorbidities ( r 5 0.54, P..01). There was a weak, but significant correlation between resting PVR and the mpap/co slope ( r 5 0.24, P,.01) and between resting PVR and TPG/CO slope ( r 5 0.43, P,.001). Figure 6 mpap at rest (mm Hg) vs that at 50 W in patients without known respiratory or cardiac limitations, in patients with respiratory limitations, and in patients with cardiac comorbidities ( P,.05 in patients without known respiratory or cardiac limitations and in patients with respiratory limitations; not significant in patients with cardiac limitations). See Figure 1 legend for expansion of abbreviation. Discussion In this retrospective study, we analyzed patients who underwent right-sided heart catheterization for symptoms or signs of PH or due to disease associated with pulmonary arterial hypertension in whom mpap was, 25 mm Hg. Hemodynamics at rest and exercise in 141 patients from our center and 51 patients from the literature were included in the analysis. 11 Patients from the present study cohort were stratified according to resting mpap and cardiac or pulmonary comorbidities defined a priori. We found that patients with a resting mpap between 21 and 24 mm Hg were characterized by a higher rate of cardiac or pulmonary comorbidities, a significantly increased mpap/co and TPG/CO slope during exercise, decreased exercise capacity, and decreased survival compared with patients with normal resting mpap. Thus, mpap values between 21 and 24 mm Hg may represent a distinct functional and prognostic marker. In addition, we found a strong correlation between resting mpap and exercise mpap and between resting PVR and exercise PVR. Interestingly, these correlations were weaker in patients with cardiac comorbidities. Resting mpap and resting PVR were moderately correlated with the mpap/co and TPG/CO slopes during exercise. Comorbidities Cardiac and pulmonary comorbidities were more common among patients with borderline PAP, suggesting that these comorbidities might be the cause of the PAP elevation. This explanation appears plausible because groups 2 and 3 (patients with pulmonary hypertension due to left-sided heart disease and due to lung diseases and/or hypoxia, according to the updated Nice classification of PH 12 ) represent the most common forms of PH. In case of pulmonary comorbidities, there is evidence that borderline PAP values are clinically relevant in COPD and idiopathic pulmonary fibrosis. 13,14 The prognostic relevance of increased PAP in systolic and diastolic left-sided heart disease has also been described. 15 It is remarkable that in both pulmonary and cardiac disease, even moderate PAP elevations are associated with a worse prognosis. In patients with scleroderma, a borderline PAP may be a distinct risk factor for manifest PH. In a number of patients, borderline PAP may be followed by rapid development of manifest pulmonary arterial hypertension. 5 Based on these findings in different diseases, borderline PAP might represent a global prognostic indicator. mpap/co and TPG/CO Slopes and PVR An mpap/co slope. 3 mm Hg/L/min during exercise may represent an abnormal hemodynamic response. 16 journal.publications.chestnet.org 1491

In the current patients with normal resting PAP, the median slope was around this value, whereas patients with borderline PAP had an increased slope of around 5 mm Hg/L/min. This suggests that about one-half of the patients with dyspnea and those at risk for PH presenting with normal resting mpap have an elevated mpap/co slope and that patients with a resting mpap between 21 and 24 mm Hg always have an elevated slope. The same is true for the TPG/CO slopes. The steepness of pressure-flow curves describes resistance; thus, accordingly, a steeper mpap/co slope represents increased total pulmonary resistance and a steeper TPG/CO slope represents increased PVR. The TPG/CO slopes in the present study appear to be almost linear in the examined range and when extrapolated, cross the y-axis very close to 0, suggesting only minor changes in PVR during exercise and no positive opening pressure of the pulmonary arteries. In fact, PVR showed only a minimal decrease and followed a similar pattern during exercise in the examined groups, suggesting that the resting values reliably predict exercise values. In the borderline group, even this minimal PVR decrease was missing. The missing or only minimal decrease of PVR during exercise is in line with earlier studies. 11,17,18 Comparison With Historical Control Group The comparison of the normal resting mpap group with the historical control group showed similar resting hemodynamics but significantly elevated mpap/co and TPG/CO slopes. Of course, we must consider that the patients were referred due to symptoms or a risk condition for PH, whereas the control subjects had no signs and symptoms of any cardiovascular disease. The distinction between the historical control group and both study groups is clearly visible in the mpap/co and TPG/CO slopes ( Figs 1, 2 ) and in the resistance-compliance curve, where the control group had the highest compliance and lowest resistance values and the borderline group was located at the opposite end of the curve. These findings suggest that not only borderline PAP but also increased mpap/co and TPG/CO slopes, a slightly increased resting PVR, or a reduction of pulmonary arterial compliance might represent early markers of pulmonary vascular abnormalities. One limitation of the historical control group is the younger age of subjects compared with the patients. Other hemodynamics studies found that control subjects of a similar age as the current patients may be characterized by an mpap/co slope of 1.4 mm Hg/L/min, and the limits of normal of the slopes of multipoint mpap/co relationships may range from 0.5 to 2.5 mm Hg/L/min. 18,19 This agrees with the historical control group included in the present analysis, where the median mpap/co slope was 0.8 mm Hg/L/min. In comparison, the median mpap/co slopes of the current patients with normal and borderline resting mpap were above this suggested normal range (3.2 and 5.2 mm Hg/L/min, respectively). Exercise Capacity and Mortality Patients with borderline PAP showed decreased exercise capacity compared with those with normal resting mpap, indicating that this hemodynamic condition is functionally relevant. In addition, the death rate in the borderline group was strikingly increased, suggesting that borderline PH is prognostically relevant, although the small number of events precludes a reliable statistical analysis. The higher mortality in the borderline group was most likely due to the increased rate of comorbidities, so any causative role of pulmonary vascular abnormalities remains speculative. The presence of borderline PAP values might instead be a marker than a cause of a poor prognosis. Predictive Value of Resting mpap and PVR for Exercise PAP and PVR We found a strong correlation for both mpap and PVR between rest and exercise, although these correlations were considerably weaker in patients with cardiac comorbidities. The weaker correlation in patients with cardiac comorbidities may be explained by the individual changes in PAWP being only weakly associated with resting PAWP. In addition, both resting mpap and resting PVR were positively correlated with the mpap/co and TPG/CO slopes, suggesting that the observed hemodynamic changes at rest and during exercise may have the same origin and clinical relevance. This is supported by studies in patients with scleroderma, where such hemodynamic findings were associated with an increased risk for PH. 3,5 Limitations There was no control group comprising subjects without suspicion of PH who underwent right-sided heart catheterization in our center because for ethical reasons we could not subject healthy individuals with no clinical indication to an invasive study. The historical control subjects were younger than the study patients, but to our knowledge, this was the largest available group of healthy individuals (n 5 51) with multiple invasive hemodynamic measurements during exercise that could serve as a control group. Conclusions Based on the data representing a real-life population, it may be justified to distinguish patients with borderline 1492 Original Research [ 146 # 6 CHEST DECEMBER 2014 ]

PAP from those with normal resting PAP. After correction for age, patients with borderline PAP values have decreased exercise capacity, increased PVR, and a significantly increased mpap/co and TPG/CO slope during exercise and may have a poorer prognosis than patients with normal resting mpap. Acknowledgments Author contributions: G. K. had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. G. K. contributed to the study concept and design, data acquisition and interpretation, and drafting and submission of the final manuscript; A. A. contributed to the data acquisition and interpretation, statistical approach, critical revision of the manuscript for important intellectual content, and final approval of the manuscript; M. T., V. F., A. O., and H. O. contributed to the data acquisition and interpretation, critical revision of the manuscript for important intellectual content, and final approval of the manuscript; and G. B. contributed to the data acquisition and interpretation, structure of the database, critical revision of the manuscript for important intellectual content, and final approval of the manuscript. Financial/nonfinancial disclosures: The authors have reported to CHEST the following conflicts of interest: Dr Kovacs reports personal fees from Actelion Pharmaceuticals Ltd; personal fees and nonfinancial support from Bayer AG; personal fees from GlaxoSmithKline plc; nonfinancial support and personal fees from Pfizer, Inc; nonfinancial support from AOP Orphan Pharmaceuticals AG; personal fees and nonfinancial support from Boehringer- Ingelheim GmbH; personal fees from AstraZeneca; personal fees and nonfinancial support from Takeda Pharmaceutical Company Limited; personal fees from Novartis AG; and personal fees from Chiesi Farmaceutici SpA outside the submitted work. Dr Foris reports nonfinancial support from GlaxoSmithKline plc; Actelion Pharmaceuticals Ltd; Pfizer, Inc; Eli Lilly and Company; VitalAire Canada; and Novartis AG outside the submitted work. Dr H. Olschewski reports grants from Bayer AG; Unither Pharmaceuticals; Actelion Pharmaceuticals Ltd; and Pfizer, Inc; personal fees from Bayer AG; Unither Pharmaceuticals; Actelion Pharmaceuticals Ltd; Gilead Sciences, Inc; Encysive Pharmaceuticals Ltd; GlaxoSmithKline plc; and Nebu-Tec med Produkte Eike Kern GmbH; and personal fees and nonfinancial support from Bayer AG; Unither Pharmaceuticals; Actelion Pharmaceuticals Ltd; Pfizer, Inc; Eli Lilly and Company; and GlaxoSmithKline plc outside the submitted work. Drs Avian, Tscherner, Bachmaier, and A. Olschewski have reported that no potential conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article. Other contributions: The authors thank Eugenia Lamont for careful linguistic review. Additional information: The e-figures can be found in the Supplemental Materials section of the online article. References 1. Galiè N, Hoeper MM, Humbert M, et al ; Task Force for Diagnosis and Treatment of Pulmonary Hypertension of European Society of Cardiology (ESC); European Respiratory Society (ERS); International Society of Heart and Lung Transplantation (ISHLT). Guidelines for the diagnosis and treatment of pulmonary hypertension. Eur Respir J. 2009 ;34(6):1219-1263. 2. Tolle JJ, Waxman AB, Van Horn TL, Pappagianopoulos PP, Systrom DM. Exercise-induced pulmonary arterial hypertension. Circulation. 2008 ;118(21): 2183-2189. 3. Condliffe R, Kiely DG, Peacock AJ, et al. Connective tissue disease-associated pulmonary arterial hypertension in the modern treatment era. Am J Respir Crit Care Med. 2009 ;179(2):151-157. 4. Saggar R, Khanna D, Furst DE, et al. Exercise-induced pulmonary hypertension associated with systemic sclerosis: four distinct entities. Arthritis Rheum. 2010 ;62(12):3741-3750. 5. Valerio CJ, Schreiber BE, Handler CE, Denton CP, Coghlan JG. Borderline mean pulmonary artery pressure in patients with systemic sclerosis: transpulmonary gradient predicts risk of developing pulmonary hypertension. Arthritis Rheum. 2013 ;65(4):1074-1084. 6. Kovacs G, Maier R, Aberer E, et al. Borderline pulmonary arterial pressure is associated with decreased exercise capacity in scleroderma. Am J Respir Crit Care Med. 2009 ;180(9):881-886. 7. Whyte K, Hoette S, Herve P, et al. The association between resting and mild-tomoderate exercise pulmonary artery pressure. Eur Respir J. 2012 ;39(2):313-318. 8. Kovacs G, Maier R, Aberer E, et al. Pulmonary arterial hypertension therapy may be safe and effective in patients with systemic sclerosis and borderline pulmonary artery pressure. Arthritis Rheum. 2012 ;64(4):1257-1262. 9. Saggar R, Khanna D, Shapiro S, et al. Brief report: effect of ambrisentan treatment on exercise-induced pulmonary hypertension in systemic sclerosis: a prospective single-center, open-label pilot study. Arthritis Rheum. 2012 ;64(12): 4072-4077. 10. Kovacs G, Maier R, Aberer E, et al. Assessment of pulmonary arterial pressure during exercise in collagen vascular disease: echocardiography vs right-sided heart catheterization. Chest. 2010 ;138(2): 270-278. 11. Kovacs G, Olschewski A, Berghold A, Olschewski H. Pulmonary vascular resistances during exercise in normal subjects: a systematic review. Eur Respir J. 2012 ;39(2):319-328. 12. Simonneau G, Gatzoulis MA, Adatia I, et al. Updated clinical classification of pulmonary hypertension. J Am Coll Cardiol. 2013 ;62(suppl 25 ):D34D-41. 13. Hamada K, Nagai S, Tanaka S, et al. Significance of pulmonary arterial pressure and diffusion capacity of the lung as prognosticator in patients with idiopathic pulmonary fibrosis. Chest. 2007 ;131(3): 650-656. 14. Kessler R, Faller M, Fourgaut G, Mennecier B, Weitzenblum E. Predictive factors of hospitalization for acute exacerbation in a series of 64 patients with chronic obstructive pulmonary disease. Am J Respir Crit Care Med. 1999 ;159(1): 158-164. 15. Bursi F, McNallan SM, Redfield MM, et al. Pulmonary pressures and death in heart failure: a community study. J Am Coll Cardiol. 2012 ;59(3):222-231. 16. Naeije R, Vanderpool R, Dhakal BP, et al. Exercise-induced pulmonary hypertension: physiological basis and methodological concerns. Am J Respir Crit Care Med. 2013 ;187(6):576-583. 17. Janicki JS, Weber KT, Likoff MJ, Fishman AP. The pressure-flow response of the pulmonary circulation in patients with heart failure and pulmonary vascular disease. Circulation. 1985 ; 72 ( 6 ): 1270-1278. 18. Argiento P, Vanderpool RR, Mulè M, et al. Exercise stress echocardiography of the pulmonary circulation: limits of normal and sex differences. Chest. 2012 ;142(5):1158-1165. 19. Lewis GD, Murphy RM, Shah RV, et al. Pulmonary vascular response patterns during exercise in left ventricular systolic dysfunction predict exercise capacity and outcomes. Circ Heart Fail. 2011 ;4(3):276-285. journal.publications.chestnet.org 1493