Demographics and Outcomes of Patients Diagnosed With Pulmonary Hypertension With Pulmonary Capillary Wedge Pressures 16 to 18 mm Hg

Transcription

1 CHEST Original Research PULMONARY VASCULAR DISEASE Demographics and Outcomes of Patients Diagnosed With Pulmonary Hypertension With Pulmonary Capillary Wedge Pressures 16 to 18 mm Hg Insights From the REVEAL Registry Adaani E. Frost, MD, FCCP ; Harrison W. Farber, MD, FCCP ; Robyn J. Barst, MD, FCCP ; Dave P. Miller, MS ; C. Gregory Elliott, MD, FCCP ; and Michael D. McGoon, MD Background: The Registry to Evaluate Early and Long-term Pulmonary Arterial Hypertension Disease Management (REVEAL Registry) is a multicenter, US-based, observational study of patients diagnosed with group 1 pulmonary hypertension enrolled consecutively from March 2006 to December Of 3,128 patients in this analysis, inclusion criteria permitted enrollment of 268 patients with mean pulmonary capillary wedge pressure (PCWP) 16 to 18 mm Hg at diagnostic right-sided heart catheterization (RHC) (above currently accepted pulmonary arterial hypertension [PAH] diagnostic criteria). This study compared the demographics and outcomes of those 268 patients with an elevated mean PCWP to patients with a mean PCWP 15 mm Hg. Methods: Demographic characteristics and outcomes were compared for patients with mean PCWP 12, 13 to 15, and 16 to 18 mm Hg at diagnostic and follow-up RHC. Results: At diagnostic RHC, patients with PCWP 16 to18 mm Hg were older, had more severe hemodynamic impairments, and were more likely to be obese and have other comorbidities than patients with PCWP 15 mm Hg. There were no clinically relevant differences in 5-year survival rates from diagnostic RHC regardless of PCWP at diagnosis ( 15 mm Hg vs mm Hg, P 5.07). Two-year survival rates of 108 patients with PAH whose PCWP increased to 19 mm Hg (regardless of PCWP at diagnosis) were significantly lower than that of patients with PAH with PCWP 18 mm Hg at subsequent RHC. Conclusion: Patients with PCWP 16 to 18 mm Hg who were diagnosed and treated for PAH were older, heavier, and more likely to have comorbidities associated with left ventricular diastolic dysfunction at diagnosis than those with PCWP 15 mm Hg. Five-year survival rates were similarly low for all PCWP subgroups. Trial registry: ClinicalTrials.gov; No.: NCT ; URL: CHEST 2013; 143(1): Abbreviations: ERA 5 endothelin receptor antagonist; LV 5 left ventricular; mpap 5 mean pulmonary artery pressure; PAH 5 pulmonary arterial hypertension; PCWP 5 pulmonary capillary wedge pressure; PDE-5 5 phosphodiesterase-5; PH 5 pulmonary hypertension; REVEAL Registry 5 Registry to Evaluate Early and Long-term Pulmonary Arterial Hypertension Disease Management; RHC 5 right-sided heart catheterization; TPG 5 transpulmonary gradient The Registry to Evaluate Early and Long-term Pulmonary Arterial Hypertension Disease Management (REVEAL Registry) is an observational, multi center, US-based registry whose purpose is to provide information about current demographics and treatment practices for patients diagnosed with group 1 pulmonary hyper tension (PH), also called pulmonary arte rial hypertension (PAH). PAH is currently defined by a mean pulmonary artery pressure (mpap) 25 mm Hg, mean pul mo nary capillary wedge pressure (PCWP) 15 mm Hg at rest, and increased pulmonary vascular resistance. 1,2 To better reflect patterns of practice in the United States and to be more inclusive in analyses of treatments and outcomes over time, the hemodynamic enrollment criteria for the REVEAL Registry were expanded to include those journal.publications.chestnet.org CHEST / 143 / 1 / JANUARY

2 with mean PCWP 18 mm Hg at diagnosis, as long as other causes (ie, PH groups 2-5) had been excluded. 3 Patients with mean PCWP. 15 mm Hg have historically been excluded from registries and controlled clinical trials to avoid inadvertent inclusion of patients with group 2 PH (ie, those with PH due to left-sided heart disease). 4,5 The hemodynamic enrollment criteria for the REVEAL Registry were expanded to include the patients with nonconforming PCWP because: (1) having a risk fac tor for systolic or diastolic left-sided heart dysfunction may not preclude coexisting PAH 6,7 ; (2) ventricular interdependence can increase PCWP, thereby excluding patients with PAH from enrollment 8 ; and (3) the valid ity of including such patients is unknown and determining their phenotype and outcome is important. The objectives of the analyses in this study were as follows: (1) to compare outcomes and characteristics between patients with elevated mean PCWP (16-18 mm Hg; ie, nonconforming criteria) at the time of diagnostic right-sided heart catheterization (RHC) with those with traditional PAH criteria (mean PCWP 15 mm Hg) at the time of diagnostic RHC; and (2) to explore whether these patients could have been misclassified as having group 1 PAH when they should have been classified as having unrecognized left ventricular (LV) 9 dysfunction (eg, group 2 PH). One limitation is that the second objective cannot conclusively be addressed with an observational study such as the REVEAL Registry. However, in controlled clinical trials evaluating PAH-specific therapies for LV systolic dysfunction, the results have been unimpressive and the treatments can be detrimental. 10 Therefore, a poorer outcome with PAH therapies and/or marked differences in PAH-specific prescribing practices could provide evidence of major differences in the underlying pathophysiology in the traditional and nonconforming REVEAL Registry patient populations. The limitations to accurate measurement of PCWP are recognized. These limitations Manuscript received June 14, 2011; revision accepted May 1, Affiliations: From the Baylor College of Medicine (Dr Frost), Houston, TX; Boston University School of Medicine (Dr Farber), Boston, MA; Columbia University College of Physicians and Surgeons (Dr Barst), New York, NY; ICON Late Phase & Outcomes Research (Mr Miller), San Francisco, CA; Intermountain Medical Center and the University of Utah (Dr Elliott), Murray, UT; and Mayo Clinic (Dr McGoon), Rochester, MN. Funding/Support: Preparation of this manuscript was supported by Actelion Pharmaceuticals US, Inc. Funding for the REVEAL Registry is provided by Actelion Pharmaceuticals US, Inc. Correspondence to: Adaani E. Frost, MD, FCCP, Baylor College of Medicine, 6620 Main St, Ste 1225, Houston, TX 77030; frost@bcm.tmc.edu 2013 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: /chest are probably equally present at each measurement. The patient population in this analysis was defined by mean PCWP recorded at the time of PH diagnosis (ie, at the diagnostic RHC) rather than by the reproducibility of high PCWP over time. Outcomes and subsequent PCWP were analyzed in light of the initial assessment of mean PCWP. This study seeks to address whether patients diagnosed with PAH with an elevated mean PCWP at diagnostic RHC are treated similarly and have similar outcomes to patients with traditional PAH. Other analyses undertaken to determine the validity of including these nonconforming patients in a PAH registry included evaluation of PCWP in subsequent RHCs to determine the stability of PCWP over time (eg, do patients with elevated PCWP at enrollment have PCWP 15 mm Hg at follow-up, and vice versa?), evaluation of the impact of changes in PCWP on outcomes, and comparison of demographic features in traditional and nonconforming patient populations. Study Design and Population Materials and Methods The registry design and baseline characteristics of patients enrolled in the REVEAL Registry have been described previously. 3,11 The REVEAL Registry is an observational 55-center, prospective, US-based registry designed to provide current information about demographic characteristics, course, and management of 3,518 patients with PAH. The REVEAL protocol was approved by the institutional review board of each study center, and all patients provided written informed consent prior to study entry. A list of REVEAL Registry sites and institutional review boards is available in e-appendix 1. The REVEAL Registry enrollment criteria included newly diagnosed (ie, diagnosis by RHC 3 months prior to enrollment in the REVEAL Registry) or previously diagnosed patients with PAH. The diagnosis was defined by the earliest RHC confirm ing PAH diagnosis prior to enrollment. PAH was defined as an mpap. 25 mm Hg at rest or mpap. 30 mm Hg with exercise, contemporaneous with a mean PCWP or an LV end-diastolic pre ssure 18 mm Hg and pulmonary vascular resistance 3 Wood units. LV end-diastolic pressure, when taken by the enrolling physician, was given priority over PCWP. Group 1 PAH subgroups and PAH definitions (other than the nonconforming PCWP criteria) conformed to the guidelines of the third World Symposium on Pulmonary Hypertension in 2003, which was the most recent definition at the initiation of the REVEAL Registry. 12 Thus, in this analysis, patients were considered not to have left-sided heart disease as determined by their enrolling physicians and to have PAH based on all diagnostic studies, excluding the mean PCWP 16 to 18 mm Hg. Patients were aged 3 months at the time of diagnosis. For purposes of these analyses, patients with congenital heart disease-associated PAH (n 5 374) and those who met the hemodynamic criteria only during exercise (n 5 11) were excluded. Follow-up RHC was defined either as the most recent RHC prior to enrollment (in those cases where the diagnostic RHC was not the most recent) or as the first available postenrollment RHC (if there was no second RHC prior to or at the time of enrollment). 186 Original Research

3 Statistical Analyses The data lock for all analyses was November 19, The flow of patients in the analyses is shown in Figure 1. We analyzed patients with mean PCWP 12 mm Hg vs 13 mm Hg and 15 mm Hg vs 16 mm Hg. These cutoff values were selected because they are the PCWP values used both historically and in current guidelines 5 for studies and previously reported PAH registry enrollment. 4 For the same reason, mean PCWP, rather than transpulmonary gradient (TPG) (defined as the difference between the mpap and PCWP), was similarly identified as the defining focus for the anal ysis. Although many additional paired comparisons are of potential interest, we limited computation of P values to these two prespec ified comparisons to avoid adjustment for multiple comparisons and the corresponding loss in power. Subset analyses were performed for patients with serial RHC follow-up, and baseline comparisons were repeated in this cohort to guard against selection bias. Time-to-event analyses were conducted for all-cause mortality and all-cause hospitalization. To avoid immortal time bias, 13 time-to-event analyses were restricted to the time during which patients were prospectively followed. This was accomplished either by evaluating survival from time of enrollment using standard Kaplan-Meier product-limit estimates or by evaluating survival measured from time of diagnosis, accounting for the fact that the data are left-truncated. 14 Left truncation can occur when patient information, such as time of diagnosis, is gathered retrospectively. In a study of disease mortality where the outcome is survival from the time of diagnosis, many patients may not have been enrolled in the study until months or years after diagnosis. Those patients, by virtue of having survived to the time of enrollment, could not have had an event between diagnosis and study enrollment. Therefore, for the purposes of estimating hazards the kernel of any survival estimate they must be removed from the risk set between diagnosis and enrollment. Models that address this problem are referred to as delayedentry or late-entry models to reflect that patients enter the risk set at different periods. 15 Comorbidities and concomitant medications were collected at study enrollment and quarterly thereafter. Metabolic syndrome was based on the presence of at least three of four factors at study enrollment: obesity (based on a calculated BMI 30 kg/m 2 ), hypertension (based on comorbidities listed in patient chart), diabetes mellitus (based on comorbidities listed in patient chart), or hypercholesterolemia defined as statin use (based on concomitant medications). Results Full-Analysis Cohort The characteristics of the 3,128 patients who met inclusion criteria at diagnostic RHC (the full-analysis cohort) are presented in Table 1 according to diagnostic PCWP. Compared with patients who had PCWP 15 mm Hg, patients with PCWP 16 mm Hg were older (mean, 54 years at diagnosis); more likely to be obese (45% had BMI 30 kg/m 2 ); more likely to have a history of systemic hypertension (52%); more likely to use statins (20%); and more likely to have diabetes mellitus (20%), renal insufficiency (10%), sleep apnea (39%), cardiomyopathy (3%), an ischemic cardiovascular event (13%), valvular heart disease (8%), or atrial fibrillation (8%), and greater hemodynamic impairment at diagnosis. In addition, patients with PCWP 13 mm Hg at diagnosis were more likely to be classified as functional class IV at both diagnosis and enrollment than those with PCWP 12 mm Hg at diagnosis ( Table 1 ). The presence of three out of four features of the metabolic syndrome at enrollment Figure 1. Flow of patients through the study. CHD-APAH 5 congenital heart disease-associated pulmonary arterial hypertension; PCWP 5 pulmonary capillary wedge pressure; RHC 5 right-sided heart catheterization. journal.publications.chestnet.org CHEST / 143 / 1 / JANUARY

4 Table 1 Patient Characteristics in the Full-Analysis Cohort Based on Mean PCWP at Diagnostic RHC (N 5 3,128) Characteristic PCWP 12 mm Hg PCWP mm Hg PCWP mm Hg P Valuea P Valueb All patients, No. 2, Mean age at enrollment, SD, y ,.001,.001 Mean age at diagnostic RHC, SD, y ,.001,.001 Female patients 1,800 (79) 445 (78) 197 (74) Newly diagnosed PAH 641 (28) 184 (32) 87 (33) FC IV at enrollment 132 (6) 52 (10) 20 (8) FC IV at PAH diagnosis 223 (13) 75 (18) 32 (17) On ERA at enrollment 913 (41) 199 (36) 92 (35) On PDE-5 inhibitor at enrollment 1,065 (48) 252 (46) 109 (42) On IV or SC prostacyclins at enrollment 650 (29) 164 (30) 77 (29) Race White Black Hispanic Other 1,642 (73) 310 (14) 198 (9) 102 (5) 411 (74) 86 (15) 42 (8) 20 (4) 210 (80) 38 (14) 11 (4) 4 (2).008 Mean BMI at enrollment, SD, kg/m ,.001,.001 Comorbidities at enrollment Metabolic syndrome (patients with at least 182 (8) 85 (15) 46 (17),.001,.001 three of the following four conditions): Obese (BMI 30 kg/m2 ) 667 (31) 240 (46) 113 (45).001,.001 Hypertension 887 (40) 255 (46) 135 (52).001,.001 Diabetes mellitus 244 (11) 98 (18) 52 (20).001,.001 Statin use 356 (16) 108 (20) 52 (20) Renal insufficiency 100 (5) 27 (5) 25 (10), Sleep apnea 433 (20) 155 (29) 97 (39),.001,.001 Cardiomyopathy 44 (2) 15 (3) 9 (3) Ischemic cardiovascular event 180 (8) 74 (13) 34 (13).040,.001 Valvular heart disease 77 (3) 22 (4) 22 (8), Atrial fibrillation c 30 (1) 17 (3) 20 (8),.001,.001 Hemodynamics at diagnostic RHC, mean SD mrap ,.001,.001 mpap at rest ,.001 Cardiac index TPG ,.001,.001 TPG. 12 2,289 (100) 567 (99) 268 (100) SBP ,.001,.001 DBP ,.001 Most recent echocardiography at enrollment: LV dysfunction d None 1,725 (92) 419 (94) 195 (91) Mild 97 (5) 22 (5) 12 (6) Moderate 30 (2) 3 (1) 4 (2) Moderate/severe or severe 15 (1) 1 (0) 4 (2) Data given as No. (%) unless otherwise indicated. DBP 5 diastolic BP; ERA 5 endothelin receptor antagonist; FC 5 functional class; LV 5 left ventricular; mpap 5 mean pulmonary artery pressure; mrap 5 mean right atrial pressure; PAH 5 pulmonary arterial hypertension; PCWP 5 pulmonary capillary wedge pressure; PDE-5 5 phosphodiesterase-5; RHC 5 right-sided heart catheterization; SBP 5 systolic BP; SC 5 subcutaneous; TPG 5 transpulmonary gradient. acomparing 15 mm Hg and 16 mm Hg. bcomparing 12 mm Hg and 13 mm Hg. catrial fibrillation was obtained from a checkbox at the most recent ECG at enrollment. d The denominators for the percentages exclude 68 patients for whom the LV function measurement was not done, 41 for whom it was unable to be measured, and 492 for whom no echocardiography was performed or available in the patient chart..027 also was compared across the three PCWP groups at diagnosis and occurred in 8%, 15%, and 17% in patients with PCWP 12 mm Hg, 13 to 15 mm Hg, and 16 to 18 mm Hg, respectively ( Table 1 ). The frequencies of use of IV or subcutaneous prostacyclins, endothelin receptor antagonists (ERAs), or phosphodiesterase-5 (PDE-5) inhibitors recorded at enrollment were similar among the PCWP groups ( Table 1 ). There were no differences in the demographic profiles of patients with idiopathic PAH, familial PAH, and anorexigen-related PAH compared with patients with associated PAH (data not shown). These observations were similar for the subset that had follow-up RHC (n 5 1,825) (Table 2 ). 188 Original Research

5 Table 2 Patient Characteristics for the Subset of Patients With Follow-up RHC (n = 1,825) With PCWP Classification Based on Mean PCWP at Diagnostic RHC Characteristic PCWP 12 mm Hg PCWP mm Hg PCWP mm Hg P Value a P Value b All patients, No. 1, Mean age at enrollment, SD, y ,.001 Mean age at diagnostic RHC, SD, y ,.001,.001 Female patients 1,097 (80) 250 (80) 105 (73) Newly diagnosed PAH 235 (17) 62 (20) 30 (21) FC IV at enrollment 64 (5) 25 (9) 5 (4) FC IV at PAH diagnosis 120 (12) 42 (19) 15 (16) On ERA at enrollment 615 (46) 127 (42) 63 (45) On PDE-5 inhibitor at enrollment 672 (50) 152 (50) 65 (47) On IV or SC prostacyclins at enrollment 472 (35) 111 (37) 50 (36) Race White Black Hispanic Other 990 (74) 186 (14) 100 (7) 71 (5) 228 (74) 43 (14) 23 (8) 13 (4) 116 (84) 17 (12) 5 (4) 1 (1).021 Mean BMI at enrollment, SD, kg/m ,.001 Obese (BMI 30 kg/m2 ) 414 (32) 134 (45) 62 (46).010,.001 Comorbidities at enrollment Hypertension 487 (36) 135 (44) 64 (46) Diabetes mellitus 133 (10) 49 (16) 30 (22),.001,.001 Renal insufficiency 48 (4) 15 (5) 15 (11), Sleep apnea 277 (21) 85 (28) 63 (47),.001,.001 Cardiomyopathy 20 (2) 9 (3) 6 (4) Ischemic cardiovascular event 92 (7) 37 (12) 16 (12) Valvular heart disease 48 (4) 11 (4) 11 (8) Atrial fibrillation c 18 (1) 12 (4) 9 (6),.001,.001 Hemodynamics at diagnostic RHC, mean SD mrap ,.001,.001 mpap at rest ,.001 Cardiac index TPG , TPG. 12 1,370 (100) 312 (100) 143 (100) SBP ,.001,.001 DBP Months from PCWP at diagnostic RHC to PCWP at second RHC, percentile 25, 50, 75 12, 25, 48 13, 25, 43 10, 21, Data given as No. (%) unless otherwise indicated. See Table 1 for expansion of abbreviations. acomparing 15 mm Hg and 16 mm Hg. bcomparing 12 mm Hg and 13 mm Hg. catrial fibrillation was obtained from a checkbox at the most recent ECG at enrollment. Percentages were calculated based on the analysis cohort in Table The 5-year survival rate from the time of diagnostic RHC in the full-analysis cohort (N 5 3,128) ranged from 49% to 58% among the PCWP subgroups ( Fig 2 ). Although there was no significant difference in survival rate based on PCWP classification at diagnostic RHC, the difference in survival rates in patients with the highest PCWP classification (16-18 mm Hg) trended toward significant ( P 5.072) compared with those with PCWP, 16 mm Hg. However, after adjustment for confounders identified at diagnosis (age, sex, and mean right atrial pressure), the difference in the 5-year survival rate for patients by PCWP classification was not significant ( P 5.71). Similarly, TPG was not associated with 5-year survival rate as either a univariate predictor ( P 5.20) or in a multivariable model ( P 5.88). Subset of Patients With Follow-up RHC The median time from diagnostic RHC to follow-up RHC was 21, 25, and 25 months for the diagnostic PCWP subgroups of 16 to 18 mm Hg, 13 to 15 mm Hg, and 12 mm Hg, respectively. Changes occurred in the PCWP measurements from diagnosis to follow-up RHC ( Fig 3 ): In patients with PCWP 12 mm Hg at diagnostic RHC (n 5 1,370), 75% had a PCWP in the same category at follow-up and 25% had a higher PCWP, while 10% of the latter group had PCWP 16 mm Hg at follow-up. Of the 312 patients who had PCWP 13 to 15 mm Hg at diagnosis, 59% (n 5 183) had PCWP 12 mm Hg at follow-up RHC, 22% (n 5 68) remained at 13 to 15 mm Hg, 12% (n 5 37) journal.publications.chestnet.org CHEST / 143 / 1 / JANUARY

6 Figure 2. Five-year survival rates from diagnostic RHC in the full-analysis cohort (N 5 3,128). The unadjusted difference in the 5-year survival rate between mean PCWP 15 mm Hg vs mm Hg trended toward significant (hazard ratio: 1.26; [95% CI, ] ; P 5.072). See Figure 1 legend for expansion of abbreviations. increased to PCWP 16 to 18 mm Hg, and 8% (n 5 24) increased to 19 mm Hg. At follow-up RHC, in the 143 patients with PCWP 16 mm Hg (but 18 mm Hg) at diagnostic RHC, 52% (n 5 74) had PCWP 12 mm Hg, 13% (n 5 18) had PCWP 13 to 15 mm Hg, 18% (n 5 26) remained at 16 to 18 mm Hg, and 17% (n 5 25) increased to 19 mm Hg. Of the 108 patients who had PCWP 19 mm Hg at follow-up RHC, 59 (55%) had PCWP 12 mm Hg, 24 (22%) had PCWP 13 to 15 mm Hg, and 25 (23%) had PCWP of 16 to 18 mm Hg at diagnostic RHC. Despite the changes in PCWP measurements at follow-up RHC, the patient characteristics according to PCWP at follow-up RHC were similar to those according to PCWP at diagnostic RHC (n 5 1,825) ( Table 3 ): Patients with PCWP 12 mm Hg at follow-up were younger (46 years at diagnosis); less likely to be obese (30%) or have a history of hypertension (36%), diabetes (10%), sleep apnea (21%), cardiomyopathy (1%), an ischemic cardiovascular event (8%), or atrial fibrillation (1%); and had less severely impaired hemodynamics, notably right atrial pressure, and lower systolic BP and diastolic BP compared with patients with PCWP 13 mm Hg at follow-up RHC ( Table 3 ). Two-year survival rates from follow-up RHC ranged from 65% in patients with PCWP 19 mm Hg at follow-up RHC to 82% in patients with PCWP 12 mm Hg at follow-up RHC ( Fig 4 ). Patients with follow-up PCWP 15 mm Hg had a higher 2-year survival rate from follow-up RHC than patients having follow-up PCWP 16 mm Hg ( P 5.001), and patients with follow-up PCWP 12 mm Hg had a higher Figure 3. Changes in mean PCWP from diagnostic RHC to follow-up RHC (first available postenrollment RHC; n 5 1,825). The three pie charts represent mean PCWP classifications at diagnostic RHC. The colors within the pie charts represent PCWP classification at follow-up RHC. a Percentages in pie chart add to 101% due to rounding. Dx 5 diagnosis. See Figure 1 legend for expansion of other abbreviations. 190 Original Research

7 Table 3 Patient Characteristics for the Subset of Patients With Follow-up RHC (n = 1,825) With PCWP Classification Based on Mean PCWP at Follow-up RHC a Characteristic PCWP 12 mm Hg PCWP mm Hg PCWP mm Hg PCWP 19 mm Hg P Valueb P Valuec All patients, No. 1, Mean age at enrollment, SD, y ,.001,.001 Mean age at diagnostic RHC, SD, y ,.001,.001 Female patients 1,024 (80) 244 (83) 101 (73) 83 (77) Newly diagnosed PAH 216 (17) 64 (22) 23 (17) 24 (22) FC IV at enrollment 63 (5) 16 (6) 10 (8) 5 (5) FC IV at PAH diagnosis 126 (14) 25 (12) 15 (15) 11 (14) On ERA at enrollment 569 (45) 131 (46) 66 (50) 39 (38) On PDE-5 inhibitor at enrollment 647 (52) 129 (45) 67 (50) 46 (44) On IV or SC prostacyclins at enrollment 462 (37) 91 (32) 48 (36) 32 (31) Race White Black Hispanic Other 932 (74) 167 (13) 97 (8) 66 (5) 217 (75) 42 (15) 17 (6) 13 (5) 109 (79) 19 (14) 5 (4) 5 (4) 76 (73) 18 (17) 9 (9) 1 (1).22 Mean BMI at enrollment, SD, kg/m ,.001,.001 Obese (BMI 30 kg/m2 ) 370 (30) 125 (46) 64 (53) 51 (50),.001,.001 Comorbidities at enrollment Hypertension 456 (36) 122 (42) 59 (44) 49 (47) Diabetes mellitus 124 (10) 46 (16) 23 (17) 19 (18).003,.001 Renal insufficiency 51 (4) 17 (6) 8 (6) 2 (2) Sleep apnea 262 (21) 78 (28) 40 (31) 45 (45),.001,.001 Cardiomyopathy 18 (1) 6 (2) 5 (4) 6 (6) Ischemic cardiovascular event 98 (8) 22 (8) 14 (11) 11 (11) Valvular heart disease 51 (4) 9 (3) 4 (3) 6 (6) Atrial fibrillation d 16 (1) 5 (2) 7 (5) 11 (10),.001,.001 Hemodynamics at second RHC, mean SD mrap ,.001,.001 mpap at rest ,.001 Cardiac index TPG ,.001,.001 TPG. 12 1,248 (97) 283 (96) 130 (94) 91 (84),.001,.001 SBP ,.001,.001 DBP Months from PCWP at diagnostic RHC to PCWP at second RHC, percentile 25, 50, 75 11, 24, 47 13, 27, 48 13, 23, 42 13, 24, Data given as No. (%) unless otherwise indicated. See Table 1 for expansion of abbreviations. a Follow-up RHC is defined as most recent at enrollment or first available postenrollment RHC. bcomparing 15 mm Hg and 16 mm Hg. ccomparing 12 mm Hg and 13 mm Hg. datrial fibrillation was obtained from a checkbox at the most recent ECG at enrollment. Percentages were calculated based on the analysis cohort in Table year survival rate from follow-up RHC than patients having follow-up PCWP 13 mm Hg ( P 5.004) ( Fig 4 ). These differences persisted after adjustment for age, sex, renal insufficiency, history of ischemic events, hypertension, and TPG. We also investigated 2-year survival rate from follow-up RHC according to PCWP classification at follow-up RHC, with breakdowns by diagnostic PCWP ( Fig 5 ). In patients with PCWP 12 mm Hg at diagnostic RHC, those who had PCWP 15 mm Hg at follow-up RHC had a statistically higher survival rate, although the difference between groups, excluding those with PCWP 19 mm Hg, was small (follow-up PCWP 15 mm Hg vs 16 mm Hg; P 5.013) ( Fig 5A ). When PCWP was 13 to 15 mm Hg at diagnostic RHC, a similar pattern to those with diagnostic PCWP 12 mm Hg was seen, with survival increasing with lower follow-up PCWP; however, a decrease in PCWP from diagnostic RHC to follow-up RHC did not appear to be associated with a significant difference in 2-year survival in those with follow-up PCWP 12 mm Hg vs PCWP 13 mm Hg ( P 5.62), or between those with follow-up PCWP 15 mm Hg vs 16 mm Hg ( P 5.11) ( Fig 5B ). Similarly, in patients journal.publications.chestnet.org CHEST / 143 / 1 / JANUARY

8 Figure 4. Two-year survival rates from follow-up RHC (n 5 1,825) by mean PCWP at follow-up RHC. There was a significant difference in survival for those patients with follow-up mean PCWP 15 mm Hg compared with those with follow-up mean PCWP 16 mm Hg, due to the decreased survival for those patients with mean PCWP 19 mm Hg ( P for mean PCWP 15 mm Hg vs 16 mm Hg; P for 12 mm Hg vs 13 mm Hg). Follow-up RHC was defined as either the most recent RHC prior to enrollment (in those cases where the diagnostic RHC was not the most recent) or as the first available postenrollment RHC (if there was no second RHC prior to or at the time of enrollment). See Figure 1 legend for expansion of abbreviations. with PCWP 16 to 18 mm Hg at diagnostic RHC, the 2-year survival rate from follow-up RHC did not differ between groups ( P 5.17 for follow-up PCWP 12 mm Hg vs 13 mm Hg and P 5.16 for follow-up PCWP 15 mm Hg vs 16 mm Hg) ( Fig 5C ). Discussion A threshold PCWP of 15 mm Hg in clinical studies and US Food and Drug Administration-approved PAH therapies has been used to avoid treating patients in whom pulmonary artery pressure is elevated because of left-sided heart disease and not PAH. The inclusion and analysis in this registry of patients diagnosed with PAH with PCWP 16 to 18 mm Hg did not appear to significantly affect patient survival. The use of ERAs, PDE-5 inhibitors, and IV, subcutaneous, or inhaled prostacyclin analogs was similar across all PCWP groups. This is an important observation as we are not aware of any study demonstrating sustained functional or survival benefit with ERAs or prostacyclin analogs in the management of left-sided heart disease Indeed, long-term administration of prostacyclin to patients with New York Heart Association class III or IV heart failure and severely reduced LV ejection fraction caused increased mortality vs placebo, 10 and bosentan administration to patients with reduced LV ejection fraction showed no benefit and was associated with more serious adverse events (such as hospitalizations for worsening heart failure) than placebo. 19 The patients with nonconforming PCWP criteria at enrollment were older, more likely to be obese, and were more likely to have comorbidities, including hypertension, diabetes, sleep apnea, and renal insufficiency, compared with those patients with traditional PAH hemodynamic enrollment criteria. Interestingly, this also was true of those patients with PCWP 13 mm Hg compared with those with PCWP 12 mm Hg (ie, there were significant differences among patients even within the traditional PCWP definition). It has been previously reported that the incidence of metabolic syndrome (defined as having at least three of the following four conditions: hypertension, obesity, diabetes, and hyperlipidemia) is much higher in patients with pulmonary venous hypertension (defined as PH with associated left-sided heart disease). 20 That study reports that, in contrast to patients with group 1 PH, in whom the metabolic synd rome was present in only 20% of the patients, the metabolic syndrome occurred in 70% of patients with group 2 PH. Our data show that there is a higher prevalence of the metabolic syndrome with higher PCWP and that the prevalence of the metabolic syndrome is highest in patients with PCWP 16 to 18 mm Hg at diagnosis (17%); a lower prevalence is observed in patients with PCWP 12 mm Hg at diagnosis (8%). The prevalence of the metabolic syndrome in our study is still dramatically less than that reported in patients with group 2 PH and is consistent with that reported in patients with PAH. 20 Although demographic analyses also suggest that higher diagnostic PCWP is associated with a risk of comorbidities associated with LV diastolic dysfunction, survival and treatment appear remarkably similar in the traditional and nonconforming enrollment criteria patients; in other words, both groups behave like patients with traditionally defined 192 Original Research

9 Figure 5. Two-year survival rates from follow-up RHC (n 5 1,825) by mean PCWP at follow-up RHC with breakdowns by mean PCWP at diagnostic RHC. A, Patients with mean PCWP 12 mm Hg at diagnostic RHC (n 5 1,370). B, Patients with mean PCWP mm Hg at diagnostic RHC (n 5 312). C, Patients with mean PCWP mm Hg at diagnostic RHC (n 5 143). Follow-up RHC was defined as either the most recent RHC prior to enrollment (in those cases where the diagnostic RHC was not the most recent) or as the first available postenrollment RHC (if there was no second RHC prior to or at the time of enrollment). See Figure 1 legend for expansion of abbreviations. PAH. Despite overlapping comorbidity frequencies among PCWP groups, it also is possible that some of these patients had unrecognized group 2 PH. Most importantly, the survival of the patients included in REVEAL who have higher PCWP is similar to that of patients with traditional PAH hemodynamics, even though they have many more associated comorbidities. Two- and 5-year survival rates of patients with heart failure with preserved ejection fraction associated with PH (by echocardiographic criteria) are variably reported, journal.publications.chestnet.org CHEST / 143 / 1 / JANUARY

10 (40% and 20%, respectively) but are substantially lower than the lowest survival rates reported here in patients with PCWP 16 to 18 mm Hg. 9,21 The variability of PCWP between diagnostic and follow-up RHC and the consequent reclassification of patients among the different PCWP groups at follow-up RHC (particularly from PCWP. 15 mm Hg to a traditional PCWP 15 mm Hg) could represent improvement of PAH with pulmonary vasoactive therapy. Alternatively, the variability of PCWP between measurements may be due to transient phenomena, such as altered volume status, diuresis, or variability of measurement technique. While the majority of the REVEAL Registry centers enrolled. 100 patients and were considered PAH specialty centers, a standard methodology for diagnostic RHC was not mandated by the REVEAL Registry. However, it is the standard of care to undertake RHC per the American College of Cardiology Foundation/American Heart Association consensus guidelines. The extent to which these patients changes are true changes as a result of their clinical management vs individual variation in the measurement is unknown. This emphasizes the need to consider the whole clinical scenario, and not just one parameter, when making a diagnosis and formulating a treatment plan for patients with PAH. Over one-half of the patients with PCWP 16 to 18 mm Hg at diagnostic RHC (64.4%) had PCWP 15 mm Hg at follow-up. However, within that same group there also were patients with PCWP 19 mm Hg at follow-up (n 5 25). Although these patients may have another form of PH, the change in PCWP may reflect hemodynamic consequences of severe PAH (such as myocardial ischemia or hypoxemia). An increase in PCWP to 16 mm Hg at follow-up RHC in patients who meet the traditional hemodynamic definition of PAH (PCWP 15 mm Hg) at diagnostic RHC is associated with slightly older age, obesity, hypertension, diabetes, and worse pulmonary hemodynamics at diagnosis. This change could be due to PAH-specific medications improving flow through the pulmonary vascular bed, thereby unmasking LV diastolic dysfunction. Survival rates and medication pro files in patients with PCWP 16 to 18 mm Hg at follow-up RHC, however, are remarkably similar to those in patients whose hemodynamic diagnostic parameters remain within traditional limits at follow-up RHC. Therefore, the variation in PCWP measurements may simply represent limitations in the reproducibil ity of PCWP measurements. Irrespective of PCWP at diagnosis, if PCWP increases to 19 mm Hg at follow-up RHC, it appears that the outcome is correspondingly worse. Patients with this increase in PCWP at follow-up RHC, although few (n 5 108), had a worse outcome, with a 2-year sur vival rate from follow-up RHC of 65% compared with 82%, 81%, and 78% for patients with follow-up PCWP 12, 13 to 15, and 16 to 18 mm Hg, respectively ( P for PCWP 15 mm Hg vs 16 mm Hg; P for 12 mm Hg vs 13 mm Hg). In conclusion, this analysis from the REVEAL Registry cautiously suggests that PAH may exist in patients with elevated PCWP (16-18 mm Hg). We emphasize, however, that misdiagnosis and subsequent treatment of left-sided heart disease as group 1 PH is detrimental to patients. The results of the survival and hemodynamic follow-up analyses suggest that it was clinically appropriate to include in the REVEAL Registry the patients who were classified as having PAH despite a PCWP 16 to 18 mm Hg at diagnostic RHC. Again, it must be emphasized that these patients with higher PCWP were carefully evaluated at PAH specialty clinics and all aspects of the clinical scenario were scrutinized prior to including them in the REVEAL Registry. Formal inclusion of these nonconforming patients in group 1 PH remains controversial, particularly for patients who did not have subsequent conforming PCWP measures; however, these patients do not appear to fully conform to any clearly defined PH diagnostic criteria. The current study underscores the variability in a hemodynamic measurement critical to the diagnosis of PAH, specifically PCWP or LV end-diastolic pressure. Although it cannot be determined whether any variability in PCWP is due to changes in the patient s disease status, changes in therapies (such as more or less aggressive diuresis or after-load reduction), or due to other vagaries of measurement (these occur in all centers, including those that specialize in RHC for PAH), the data are clear. In patients diagnosed with PAH within the criteria of this study (PCWP 18 mm Hg at diagnosis), a PCWP measurement 19 mm Hg at any follow-up hemodynamic assessment portends a poorer prognosis and warrants further evaluation. Patients who are diagnosed as having PAH (group 1 PH), but also have an elevated PCWP, carry significant morbidity and mortality, similar to the severe mortality and morbidity seen in the traditionally restricted PAH group. This study provides, for the first time, an analysis of their phenotype and outcome, and compares them with those of patients who meet the widely accepted diagnostic hemodynamic criteria for PAH. Acknowledgments Author contributions: Dr Frost is the guarantor of the manuscript and takes responsibility for the integrity of the data and the accuracy of the data analysis. Dr Frost: contributed to the study design; collection, analysis, and interpretation of data; drafting and critical review of the manuscript; approval of the final manuscript; and served as principal author. Dr Farber: contributed to the study design; collection, analysis, and interpretation of data; drafting and critical review of the manuscript; and approval of the final manuscript. Dr Barst: contributed to the study design; collection, analysis, and interpretation of data; drafting and critical review of the manuscript; and approval of the final manuscript. 194 Original Research

11 Mr Miller: contributed to the study design; collection, analysis, and interpretation of data; drafting and critical review of the manuscript; and has seen and approved the final manuscript. Dr Elliott: contributed to the study design; collection, analysis, and interpretation of data; drafting and critical review of the manuscript; and has seen and approved the final manuscript. Dr McGoon: contributed to the study design; collection, analysis, and interpretation of data; drafting and critical review of the manuscript; and has seen and approved the final manuscript. Financial/nonfinancial disclosures: The authors have reported to CHEST the following conflicts of interest: Dr Frost serves as a consultant for Actelion Pharmaceuticals US, Inc and Gilead; has received honoraria from Actelion Pharmaceuticals US, Inc, Gilead, and Pfizer, Inc; and has provided expert testimony on diet-pill litigation. She has also received (through the Baylor College of Medicine) funds for institutional review board-approved research from Gilead; Actelion Pharmaceuticals US, Inc; United Therapeutics Corporation; Eli Lilly and Company; Pfizer, Inc; Novartis AG; and Bayer. Dr Frost has received honoraria for her service on the REVEAL Steering Committee, which is supported by Actelion Pharmaceuticals US, Inc. Dr Farber serves as a consultant and is on the speaker s bureau for Actelion Pharmaceuticals US, Inc and has received honoraria for service on the REVEAL Steering Committee, which is supported by Actelion Pharmaceuticals US, Inc. Dr Barst serves as a consultant for and has received honoraria from Actelion Pharmaceuticals US, Inc; Bayer; GlaxoSmithKline; GeneraMedix; Gilead; Eli Lilly and Company; MondoBIOTECH; the US National Institutes of Health National Heart, Lung and Blood Institute (NIH/NHLBI); Novartis AG, and Pfizer, Inc. Dr Barst has provided expert testimony on diet-pill litigation for the plaintiffs and has also received grants from Actelion Pharmaceuticals US, Inc; Gilead; Eli Lilly and Company; NIH/NHLBI; Novartis AG; Pfizer, Inc; and United Therapeutics Corporation. Dr Barst has received honoraria for her service on the REVEAL Steering Committee, which is supported by Actelion Pharmaceuticals US, Inc. Mr Miller is employed by ICON Late Phase & Outcomes Research, a company that receives research support from Actelion Pharmaceuticals US, Inc and other pharmaceutical companies. Dr Elliott is employed by Intermountain Healthcare, which, with Dr Elliott as principal investigator, has received grant support during the last 5 years from Actelion Pharmaceuticals US, Inc; Pfizer, Inc; Encysive Pharmaceuticals; and United Therapeutics Corporation. Dr Elliott has received honoraria for service on the REVEAL Steering Committee, which is supported by Actelion Pharmaceuticals US, Inc. Dr McGoon serves as a consultant with Actelion Pharmaceuticals US, Inc, Gilead, Lung Rx, and Medtronic, Inc and has received grants from Gilead and Medtronic, Inc. Dr McGoon has received honoraria for his service on the REVEAL Steering Committee, which is supported by Actelion Pharmaceuticals US, Inc. Role of sponsors : The REVEAL Registry is sponsored by Actelion Pharmaceuticals US, Inc. Editorial and programming support was funded by Actelion Pharmaceuticals US, Inc. Other contributions : Jennifer M. Kulak, PhD; Latoya M. Mitchell, PhD; Kathryn W. Leonard, BS; and Mary Hines, BSc (Hons), of inscience Communications, Springer Healthcare, provided editorial support in the development of this manuscript. Ginny Lai, MPH, of ICON Late Phase & Outcomes Research provided SAS programming support. We thank the principal investigators and their study coordinators for their participation in the REVEAL Registry (available in e-appendix 2 ). Additional information: The e-appendixes can be found in the Supplemental Materials area of the online article. References 1. Badesch DB, Champion HC, Sanchez MA, et al. Diagnosis and assessment of pulmonary arterial hypertension. J Am Coll Cardiol ;54(1)(Suppl):S55-S McLaughlin VV, Archer SL, Badesch DB, et al ; ACCF/ AHA. ACCF/AHA 2009 expert consensus document on pulmonary hypertension: a report of the American College of Cardio logy Foundation Task Force on Expert Consensus Documents and the American Heart Association: developed in collaboration with the American College of Chest Physicians, American Thoracic Society, Inc., and the Pulmonary Hypertension Asso ciation. Circulation ; 119 ( 16 ): McGoon MD, Krichman A, Farber HW, et al. Design of the REVEAL registry for US patients with pulmonary arterial hypertension. Mayo Clin Proc ;83 (8 ): Frost AE, Badesch DB, Barst RJ, et al. The changing picture of pulmonary arterial hypertension (PAH) patients in the United States: how the REVEAL registry differs from historic and non-us contemporary registries. Chest ;139 (1 ): Simonneau G, Robbins IM, Beghetti M, et al. Updated clinical classification of pulmonary hypertension. J Am Coll Cardiol ;54 (1 )(suppl ):S43-S Bady E, Achkar A, Pascal S, Orvoen-Frija E, Laaban JP. Pulmonary arterial hypertension in patients with sleep apnoea syndrome. Thorax ;55 (11 ): Kessler R, Chaouat A, Weitzenblum E, et al. Pulmonary hypertension in the obstructive sleep apnoea syndrome: prevalence, causes and therapeutic consequences. Eur Respir J ;9 (4 ): Shapiro BP, McGoon MD, Redfield MM. Unexplained pulmonary hypertension in elderly patients. Chest ;131 (1 ): Damy T, Goode KM, Kallvikbacka-Bennett A, et al. Determinants and prognostic value of pulmonary arterial pressure in patients with chronic heart failure. Eur Heart J ; 31 (18 ): Califf RM, Adams KF, McKenna WJ, et al. A randomized controlled trial of epoprostenol therapy for severe congestive heart failure: The Flolan International Randomized Survival Trial (FIRST). Am Heart J ;134 (1 ): Badesch DB, Raskob GE, Elliott CG, et al. Pulmonary arterial hypertension: baseline characteristics from the REVEAL Registry. Chest ;137 (2 ): Simonneau G, Galiè N, Rubin LJ, et al. Clinical classification of pulmonary hypertension. J Am Coll Cardiol ; 43 (12 Suppl S ):5S-12S. 13. Suissa S. Immortal time bias in observational studies of drug effects. Pharmacoepidemiol Drug Saf ;16 (3 ): Tsai W, Jewell N, Wang M. A note on the product-limit estimator under right censoring and left truncation. Biometrika ;74 (4 ): Bull K, Spiegelhalter DJ. Survival analysis in observational studies. Stat Med ;16 (9 ): Anand I, McMurray J, Cohn JN, et al ; EARTH investigators. Long-term effects of darusentan on left-ventricular remod elling and clinical outcomes in the EndothelinA Receptor Antagonist Trial in Heart Failure (EARTH): randomised, double-blind, placebo-controlled trial. Lancet ;364 (9431 ): Packer M, McMurray J, Massie BM, et al. Clinical effects of endothelin receptor antagonism with bosentan in patients with severe chronic heart failure: results of a pilot study. J Card Fail ;11 (1 ): Teerlink JR. Recent heart failure trials of neurohormonal modulation (OVERTURE and ENABLE): approaching the asymptote of efficacy? J Card Fail ;8 (3 ): Kaluski E, Cotter G, Leitman M, et al. Clinical and hemodynamic effects of bosentan dose optimization in symptomatic heart failure patients with severe systolic dysfunction, associated with secondary pulmonary hypertension a multicenter randomized study. Cardiology ;109 (4 ): Robbins IM, Newman JH, Johnson RF, et al. Association of the metabolic syndrome with pulmonary venous hypertension. Chest ;136 (1 ): Kjaergaard J, Akkan D, Iversen KK, et al. Prognostic importance of pulmonary hypertension in patients with heart failure. Am J Cardiol ;99 (8 ): journal.publications.chestnet.org CHEST / 143 / 1 / JANUARY