Advertisement
The Journal of Heart and Lung Transplantation
International Society for Heart and Lung Transplantation.

Prognostic implications of serial risk score assessments in patients with pulmonary arterial hypertension: A Registry to Evaluate Early and Long-Term Pulmonary Arterial Hypertension Disease Management (REVEAL) analysis

Open AccessPublished:September 25, 2014DOI:https://doi.org/10.1016/j.healun.2014.09.016

      Background

      Data from the Registry to Evaluate Early and Long-Term Pulmonary Arterial Hypertension Disease Management (REVEAL) were used previously to develop a risk score calculator to predict 1-year survival. We evaluated prognostic implications of changes in the risk score and individual risk-score parameters over 12 months.

      Methods

      Patients were grouped by decreased, unchanged, or increased risk score from enrollment to 12 months. Kaplan-Meier estimates of subsequent 1-year survival were made based on change in the risk score during the initial 12 months of follow-up. Cox regression was used for multivariable analysis.

      Results

      Of 2,529 patients in the analysis cohort, the risk score was decreased in 800, unchanged in 959, and increased in 770 at 12 months post-enrollment. Six parameters (functional class, systolic blood pressure, heart rate, 6-minute walk distance, brain natriuretic peptide levels, and pericardial effusion) each changed sufficiently over time to improve or worsen risk scores in ≥5% of patients. One-year survival estimates in the subsequent year were 93.7%, 90.3%, and 84.6% in patients with a decreased, unchanged, and increased risk score at 12 months, respectively. Change in risk score significantly predicted future survival, adjusting for risk at enrollment. Considering follow-up risk concurrently with risk at enrollment, follow-up risk was a much stronger predictor, although risk at enrollment maintained a significant effect on future survival.

      Conclusions

      Changes in REVEAL risk scores occur in most patients with pulmonary arterial hypertension over a 12-month period and are predictive of survival. Thus, serial risk score assessments can identify changes in disease trajectory that may warrant treatment modifications.

      Keywords

      Pulmonary arterial hypertension (PAH) is a progressive syndrome in which fibrotic and proliferative changes in pulmonary arteries lead to increased pulmonary vascular resistance, right heart failure, and eventual death.
      • McGoon M.D.
      • Kane G.C.
      Pulmonary hypertension: diagnosis and management.
      The prognosis of patients with PAH has improved with modern therapies, but patient outcomes overall remain poor. Evidence-based clinical practice guidelines recommend the individualization of therapy based on disease severity to improve patient outcomes.
      • McLaughlin V.V.
      • Archer S.L.
      • Badesch D.B.
      • et al.
      ACCF/AHA 2009 expert consensus document on pulmonary hypertension a report of the American College of Cardiology 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 Association.
      The Registry to Evaluate Early and Long-Term Pulmonary Arterial Hypertension Disease Management (REVEAL) is a longitudinal registry that enrolled patients treated at pulmonary hypertension centers in the United States.
      • McGoon M.D.
      • Krichman A.
      • Farber H.W.
      • et al.
      Design of the REVEAL registry for US patients with pulmonary arterial hypertension.
      The data from REVEAL were used to develop a prognostic equation
      • Benza R.L.
      • Miller D.P.
      • Gomberg-Maitland M.
      • et al.
      Predicting survival in pulmonary arterial hypertension: insights from the Registry to Evaluate Early and Long-Term Pulmonary Arterial Hypertension Disease Management (REVEAL).
      and a simplified risk score calculator
      • Benza R.L.
      • Gomberg-Maitland M.
      • Miller D.P.
      • et al.
      The REVEAL Registry risk score calculator in patients newly diagnosed with pulmonary arterial hypertension.
      to predict 1-year survival in patients with PAH. The REVEAL risk model has been validated statistically in a number of studies.
      • Kane G.C.
      • Maradit-Kremers H.
      • Slusser J.P.
      • Scott C.G.
      • Frantz R.P.
      • McGoon M.D.
      Integration of clinical and hemodynamic parameters in the prediction of long-term survival in patients with pulmonary arterial hypertension.
      • Cogswell R.
      • Kobashigawa E.
      • McGlothlin D.
      • Shaw R.
      • De Marco T.
      Validation of the Registry to Evaluate Early and Long-Term Pulmonary Arterial Hypertension Disease Management (REVEAL) pulmonary hypertension prediction model in a unique population and utility in the prediction of long-term survival.
      • Cogswell R.
      • McGlothlin D.
      • Kobashigawa E.
      • Shaw R.
      • De Marco T.
      Performance of the REVEAL model in WHO Group 2 to 5 pulmonary hypertension: application beyond pulmonary arterial hypertension.
      • Escribano-Subias P.
      • Blanco I.
      • Lopez-Meseguer M.
      • et al.
      Survival in pulmonary hypertension in Spain: insights from the Spanish registry.
      The risk score calculator provides a prediction of prognosis based on the contribution of multiple variables, rather than a single variable, and predicts with good accuracy the survival of patients with PAH at 1 year. Although the calculator is designed to allow point-in-time assessment, none of the inputs to the calculator need to be strictly concurrent. Outside of clinical trials, having a full battery of assessments at the same time is very unusual. The variables entered into the calculator are based on the most recent assessment for each value, and the calculator can be updated any time any single variable is reassessed. This is an intended design feature of the calculator, although it must be assumed that the calculator allows for larger disparities between a patient’s true underlying health state and the measured health state. This makes validation of the calculator for repeat assessments particularly important.
      Whether serial recalculation of the REVEAL risk score is predictive of changes in survival over time remains unknown but warrants further exploration. Of the 19 variables included in the REVEAL prognostic equation, a few (e.g., gender or PAH sub-type) are not modifiable over time, and some have contended that these might mitigate the versatility of the risk score or that the effect of these immutable variables could negate the positive or negative effects of changes in other, modifiable variables such as the 6-minute walk distance (6MWD), brain natriuretic peptide (BNP), pericardial effusion status, renal functional status, or hemodynamic parameters. If that were the case, then recalculation of the risk score over time would not inform clinicians about the effects of interventions, which would suggest that efforts to improve the risk score and, hence, survival would be futile. In this analysis, we used data from patients with PAH enrolled in REVEAL to evaluate the prognostic implications of changes in the risk score, as well as the contributions of associated risk score variables to these changes, during a 12-month period.

      Methods

      A description of the design of REVEAL, including the inclusion and exclusion criteria, has been published previously.
      • McGoon M.D.
      • Krichman A.
      • Farber H.W.
      • et al.
      Design of the REVEAL registry for US patients with pulmonary arterial hypertension.
      In brief, the observational, prospective REVEAL registry involved 55 university-affiliated and community hospitals in the United States.
      • McGoon M.D.
      • Krichman A.
      • Farber H.W.
      • et al.
      Design of the REVEAL registry for US patients with pulmonary arterial hypertension.
      Patients who met the modified definition for World Health Organization Group 1 pulmonary hypertension
      • Simonneau G.
      • Galie N.
      • Rubin L.J.
      • et al.
      Clinical classification of pulmonary hypertension.
      were enrolled consecutively beginning in March 2006 through December 2009. The Institutional Review Board at each participating center approved the protocol.
      Data at enrollment and at reassessment used for the current analysis were based on the final study data from February 4, 2013. The analysis cohort included newly and previously diagnosed patients with Group 1 PAH aged ≥19 years at diagnosis with pulmonary capillary wedge pressure ≤15 mm Hg at diagnosis and ≥12 months of follow-up.
      The REVEAL risk score at enrollment was calculated from the most recent assessments before the time of enrollment and recalculated from the most recent assessments before 12 months after enrollment. The analysis did not exclude patients with no new assessments performed in the first 12 months of follow-up. In these cases, the most recent value at enrollment remained the most recent value at follow-up. The absence of follow-up data for a patient who survived through 12 months and continued in the study was not a reason for exclusion from the analysis.
      Among patients who survived to at least 365 days transplant free, Kaplan-Meier estimates were used to determine the subsequent 1-year survival for patients with an increased, decreased, or unchanged risk score during the initial 12 months of follow-up. Multivariable Cox regression was used to determine the effect on subsequent 1-year survival of the change in risk score over time and the recalculated score, controlling for risk score at enrollment (i.e., change and risk at enrollment were both included in the model). A single model was not fitted to include the 3 variables of risk score at enrollment, risk score at reassessment, and change in risk score, because if any 2 quantities are known, the third would be redundant. The log-rank test for comparison of Kaplan-Meier curves was used for calculating p values and the Wald test for assessing the significance of hazard ratios (HR). A p-value of < 0.05 was considered significant.

      Results

      REVEAL enrolled 3,515 patients, and 2,529 patients (80% female) were included in the current analysis (Figure 1). The analysis cohort was a mean age of 52.6 years at enrollment, and idiopathic PAH (IPAH; 47.1%) was the most common PAH etiology (Table 1).
      Figure thumbnail gr1
      Figure 1Analysis of the Registry to Evaluate Early and Long-Term Pulmonary Arterial Hypertension Disease Management (REVEAL) population. PCWP, pulmonary capillary wedge pressure; PH, pulmonary hypertension.
      Table 1Patient Demographics and Disease Characteristics at Enrollment
      Serial risk assessment cohort
      Characteristic(N = 2,529)
      N = 2,529, unless otherwise noted. Information on disease characteristics may not be available for all patients at the time of enrollment.
      Age, mean ± SD years52.6 ± 14.3
      Female, No. (%)2,024 (80.0)
      Newly diagnosed, No. (%)666 (26.3)
      WHO group, No. (%)
       Idiopathic PAH1,191 (47.1)
       Familial PAH71 (2.8)
       Associated PAH
        Congenital heart disease243 (9.6)
        Connective tissue disease644 (25.5)
        Portopulmonary hypertension138 (5.5)
        Other242 (9.6)
      Renal insufficiency, No. (%)92 (3.6)
      NYHA Functional Class, No. (%)2,328 (100.0)
       I176 (7.6)
       II853 (36.6)
       III1,175 (50.5)
       IV124 (5.3)
      Systolic blood pressure, mean ± SD mm Hg(n = 2,458) 118 ± 17.4
      Heart rate, mean ± SD beats/min(n = 2,439) 82.7 ± 14.5
      6-minute walk distance, mean ± SD m(n = 2,051) 369 ± 123
      BNP, mean ± SD pg/ml(n = 1,213) 266 ± 464
       Median (interquartile range) pg/ml115 (42–299)
      N-terminal pro-BNP, mean ± SD, pg/ml(n = 261) 1,937 ± 6,753
       Median (interquartile range) pg/ml547 (148–1,620)
      Pericardial effusion, No. (%)502/1,990 (25.2)
      Dlco, mean ± SD % predicted(n = 1,540) 59.6 ± 22.8
      mRAP ≤1 year of enrollment, mean ± SD (mm Hg)(n = 1,282) 9.1 ± 5.4
      Pulmonary vascular resistance, mean ± SD Wood units(n = 2,404) 10.0 ± 6.8
      BNP, brain natriuretic peptide; Dlco, diffusion capacity of the lung for carbon monoxide; mRAP, mean right atrial pressure; NYHA, New York Heart Association; PAH, pulmonary arterial hypertension; SD, standard deviation; WHO, World Health Organization.
      a N = 2,529, unless otherwise noted. Information on disease characteristics may not be available for all patients at the time of enrollment.
      The risk score calculated at enrollment in REVEAL used the most recent data up to and including the enrollment visit. The score was recalculated at 12 months, replacing any component of the risk score that had been reevaluated within the first 12 months of REVEAL follow-up with the most recent information up to and including the 12-month quarterly update. The change in the risk score at Month 12 as a function of the risk score at enrollment for all patients in the analysis cohort is reported in Table 2.
      Table 2Change in the Risk Score at Month 12 as a Function of the Risk Score at Enrollment (N = 2,529)
      The number of patients are reported in each category.
      Risk score at enrollment
      Risk categories by risk score: 0 – 7 = low risk; 8 = average risk; 9 = moderately high risk; 10 – 11 = high risk; 12–15 = very high risk. NA indicates not applicable because scores cannot change to values of <0.
      Change in the risk score from enrollment to Month 12
      ≥ –3–2–10+1+2≥ +3
      0NANANA0000
      1NANA04330
      2NA0111576
      31483317810
      4332055271613
      54153186332718
      6153543139682626
      7213379179984017
      8373678171892911
      9243475148591612
      102329498231175
      11131526321834
      121491015611
      133451000
      140002000
      150001000
      a The number of patients are reported in each category.
      b Risk categories by risk score: 0 – 7 = low risk; 8 = average risk; 9 = moderately high risk; 10 – 11 = high risk; 12–15 = very high risk. NA indicates not applicable because scores cannot change to values of <0.
      At the time the risk score was recalculated, 959 (38%) had no change in the risk score, 800 (32%) had a decrease (improvement) in the risk score of at least 1 point, and 770 (30%) had an increase (worsening) in the risk score of at least 1 point. The newly diagnosed patients were more likely to have decreased (improved) scores (271 of 666 [41%]) compared with the previously diagnosed patients (529 of 1,863 [28%]). At enrollment, 58.9% and of newly diagnosed and 92.7% of previously diagnosed patients were receiving a PAH-specific medication, such as an endothelin receptor antagonists, phosphodiesterase type 5 inhibitors, or prostanoid. During the first year of follow-up, 67% of newly diagnosed patients started a new PAH-specific medication. At the end of the first year of follow-up, 35% of newly diagnosed patients were receiving some form of combination therapy, and 25% were receiving parenteral prostanoids compared with 11% and 15% at enrollment, respectively. Among previously diagnosed patients, treatment changes in PAH-specific medications occurred for 59.9% of patients during the first year of follow-up, including new medication starts in 36.0%. At the end of the first year of follow-up, 54% of previously diagnosed patients were receiving some form of combination therapy, and 34% were receiving parenteral prostanoids compared with 44% and 31% at enrollment, respectively.
      Changes in individual parameters of the risk score that resulted in an improved, unchanged, or worsened risk score are reported in Table 3. The 6 individual parameters of New York Heart Association Functional Class, systolic blood pressure, heart rate, 6MWD, BNP levels, and pericardial effusion status improved sufficiently in 5.3% to 16.1% (Table 3) of patients to result in a decrease in the risk score at 12 months. These same 6 parameters also worsened sufficiently in 6.6% to 15.4% (Table 3) of patients to result in an increase in the risk score at 12 months.
      Table 3Changes in Individual Parameters of the Risk Score That Resulted in a Decreased (Improved), Unchanged, or Increased (Worsened) Risk Score at 12 Months
      ParameterPatients, No. (%)
      Decreased scoreUnchanged
      Includes patients in whom values for individual parameters may not have been measured at enrollment.
      Increased score
      Renal insufficiency
      The definition of renal insufficiency was based on the comorbidity being recorded in the patient’s medical record and not on a specific creatinine or estimated glomerular filtration rate cutoff.
      2,498 (98.8)31 (1.2)
      Age2,510 (99.2)19 (0.8)
      NYHA Functional Class407 (16.1)1,814 (71.7)308 (12.2)
      Systolic blood pressure313 (12.4)1,827 (72.2)389 (15.4)
      Heart rate274 (10.8)1,998 (79.0)257 (10.2)
      6-MWD248 (9.8)2,113 (83.6)168 (6.6)
      BNP275 (10.9)1,965 (77.7)289 (11.4)
      Pericardial effusion133 (5.3)2,200 (87.0)196 (7.8)
      Dlco23 (0.9)2,449 (96.8)57 (2.3)
      mRAP within 12 months40 (1.6)2,475 (97.9)14 (0.6)
      PVR7 (0.3)2,519 (99.6)3 (0.1)
      6MWD, 6-minute walk distance; BNP, brain natriuretic peptide; Dlco, diffusion capacity of the lung for carbon monoxide; mRAP, mean right atrial pressure; NYHA, New York Heart Association; PAH, pulmonary arterial hypertension; PVR, pulmonary vascular resistance.
      a Includes patients in whom values for individual parameters may not have been measured at enrollment.
      b The definition of renal insufficiency was based on the comorbidity being recorded in the patient’s medical record and not on a specific creatinine or estimated glomerular filtration rate cutoff.
      When patients were stratified by change in the risk score, 1-year survival estimates from 12 months post-enrollment were 93.7% ± 0.9% (p < 0.001) in patients whose risk score decreased (improved), 90.3% ± 1.0% in patients whose risk score was unchanged, and 84.6% ± 1.3% in patients whose risk score increased (worsened; Figure 2A). An absence of follow-up assessments resulted in 4% of patients being included in the no-change group, and results did not differ when these patients were excluded in a sensitivity analysis. The differences in the estimates were also similar in the sub-set of newly diagnosed patients (92.6% ± 1.6%, 85.3% ± 2.4%, and 80.4% ± 3.0% for the improved, unchanged, and worsened groups, respectively) and the sub-set of previously diagnosed patients (94.3% ± 1.0%, 91.7% ± 1.0%, and 85.8% ± 1.4% for the improved, unchanged, and worsened groups, respectively). When all patients were stratified by risk score recalculated at 12 months, 1-year survival estimates from 12 months post-enrollment were 97.3% ± 0.5% (p < 0.001) in patients with a risk score of 0 to 7 (low risk), 93.0% ± 1.3% in patients with a risk score of 8 (average risk), 85.3% ± 1.9% in patients with a risk score of 9 (moderately high risk), 72.5% ± 2.5% in patients with a risk score of 10 to 11 (high risk), and 50.2% ± 5.0% in patients with a risk score of 12 to 15 (very high risk; Figure 2B).
      Figure thumbnail gr2
      Figure 2One-year survival from 12 months post-enrollment (A) by change in the Registry to Evaluate Early and Long-Term Pulmonary Arterial Hypertension Disease Management (REVEAL) risk score and (B) by the REVEAL risk score recalculated at 12 months.
      Multivariable Cox models were developed to assess the significance of the REVEAL risk score at enrollment, change in the risk score at reassessment at 12 months, and the risk score at reassessment as predictors of subsequent 1-year survival in REVEAL (Table 4). In one model adjusting for risk at enrollment, change in risk score significantly predicted subsequent 1-year survival, with HRs of 1.67 (95% confidence interval [CI], 1.41–1.99; p < 0.001) for worsened score and 0.57 (95% CI, 0.47–0.69; p < 0.001) for improved score. When follow-up risk was considered concurrently with risk at enrollment in another model, the risk score at enrollment (HR, 1.10; 95% CI, 1.04–1.15; p < 0.001) and the risk score at reassessment (HR, 1.40; 95% CI, 1.33–1.47; p < 0.001) were each significant predictors of survival, although the follow-up risk score was a much stronger predictor.
      Table 4Multivariable Cox Models of Survival After Recalculation of the Risk Score at 12 Months
      ModelPredictor of survivalHR95% CIp-value
      ARisk score at REVEAL enrollment1.491.44–1.55<0.001
      Increased risk score at recalculation1.671.41–1.99<0.001
      Decreased risk score at recalculation0.570.47–0.69<0.001
      BRisk score at REVEAL enrollment1.101.04–1.15<0.001
      Risk score at recalculation1.401.33–1.47<0.001
      CI, confidence interval; HR, hazard ratio; REVEAL, Registry to Evaluate Early and Long-Term Pulmonary Arterial Hypertension Disease Management.

      Discussion

      The current analysis shows that the REVEAL prognostic equation and risk score calculator have prognostic value not only at a set point in time (e.g., at enrollment) but also when used for serial assessments. This utility allows clinicians to regularly reassess risk to identify patients in whom PAH is improving, stable, or progressing and to individualize therapy to meet the specific needs of patients.
      • Benza R.L.
      • Miller D.P.
      • Gomberg-Maitland M.
      • et al.
      Predicting survival in pulmonary arterial hypertension: insights from the Registry to Evaluate Early and Long-Term Pulmonary Arterial Hypertension Disease Management (REVEAL).
      Another recent cohort study has shown that changes in clinical variables in response to therapy may have important potential prognostic implications in patients with IPAH.
      • Nickel N.
      • Golpon H.
      • Greer M.
      • et al.
      The prognostic impact of follow-up assessments in patients with idiopathic pulmonary arterial hypertension.
      Our findings show that changes over time occurred in most patients even though changes in individual components of the risk score were relatively less frequent. There were, however, 6 parameters that changed sufficiently over time to improve the risk score in at least 5% of patients or to worsen the risk score in at least 6% of patients. That no single parameter was the primary source of change in the risk score (either improvement or worsening) highlights the importance of ongoing assessments of risk and using a multifactorial assessment, such as the REVEAL risk score calculator, to set goals.
      Somewhat surprisingly, changes in renal insufficiency were very uncommon. This most likely reflects the use of a qualitative statement, abnormal vs normal, to define the variable instead of an absolute numeric value. Thus, a creatinine value could change from 2.5 mg/dl to 1.6 mg/dl and still be considered “abnormal” despite the improvement in the absolute number. However, we have noted in previous analyses that numeric creatinine cutoffs were equally prognostic, and it is likely that numeric depictions of creatinine or estimated glomerular filtration rate would be more sensitive to change.
      More importantly, our findings strongly suggest that combined improvement in the modifiable parameters of the REVEAL risk score can overcome the risk conferred by non-modifiable parameters and support the efforts of clinicians to lower overall patient risk. One model of survival found the recalculated risk score predicted subsequent 1-year survival better than the initial risk score, suggesting that a patient’s place on the risk continuum after 12 months is a relatively better predictor of survival in the subsequent year. This also reinforces the potential benefits of ongoing risk assessment as a means to monitor patient response to treatment.
      This analysis had several important limitations. Although risk is clearly multifactorial and modifiable, this analysis does not show which of the many risk factors associated with PAH are most important to modify or which treatments lead to such modifications. These are questions that are better answered with experimental designs rather than observational studies; however, more in-depth analyses of individual risk factors in both experimental and observational settings will also undoubtedly further our understanding. Because the risk score, as a composite instrument, has been repeatedly validated, we chose not to deconstruct the risk score into its component parts in this analysis. Although adding to the integrity of the validation, this choice also represents a limitation, because revisions could likely improve the risk score in the future.
      The REVEAL risk score calculator is a simple tool that can aid clinicians in individualizing therapy according to each patient’s risk and treatment goals. Indeed, all patients in REVEAL were actively managed, and treatment was changed in most over the course of the first year of follow-up. The use of serial risk score assessments may help identify patients with changes in disease severity and provide useful information when considering possible treatment modifications and subsequent evaluation of the relative success of those treatment changes. With additional validation, the risk score calculator used in the current analysis could be incorporated into routine clinical practice as a valuable tool to support clinicians in making treatment decisions best matched to individual patient’s levels of risk as they evolve over time.

      Disclosure statement

      Assistance in writing the first draft of the manuscript was provided by a professional medical writer, Anna Lau, PhD, of Percolation Communications LLC, and paid by the sponsor.
      The REVEAL Registry is sponsored by Actelion Pharmaceuticals US Inc.
      R.L.B. is employed by the Allegheny Health Network, has been a consultant for Bayer and United Therapeutics, has received honoraria from Bayer, and has received research support from Actelion, Bayer, Geno Pharmaceuticals, Gilead, IKARIA, and United Therapeutics. D.P.M. and A.J.F. are employed by ICON Clinical Research the biostatistics CRO for the REVEAL Registry. A.E.F. has been a consultant for Actelion, Bayer, and Gilead, has received honoraria from Actelion, Bayer, Gilead, and United Therapeutics, and has received research support from Actelion, ARIES, Bayer, Gilead, IKARIA, and United Therapeutics. D.B.B. has been a consultant for, has received honoraria from, and has received research support from Actelion, Arena, Bayer, Gilead, IKARIA, and United Therapeutics/Lung LLC. W.W.B. is employed by Actelion. M.D.M. has been a consultant for Actelion, Glaxo-Smith Kline, and Lung LLC.

      References

        • McGoon M.D.
        • Kane G.C.
        Pulmonary hypertension: diagnosis and management.
        Mayo Clin Proc. 2009; 84: 191-207
        • McLaughlin V.V.
        • Archer S.L.
        • Badesch D.B.
        • et al.
        ACCF/AHA 2009 expert consensus document on pulmonary hypertension a report of the American College of Cardiology 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 Association.
        J Am Coll Cardiol. 2009; 53: 1573-1619
        • McGoon M.D.
        • Krichman A.
        • Farber H.W.
        • et al.
        Design of the REVEAL registry for US patients with pulmonary arterial hypertension.
        Mayo Clin Proc. 2008; 83: 923-931
        • Benza R.L.
        • Miller D.P.
        • Gomberg-Maitland M.
        • et al.
        Predicting survival in pulmonary arterial hypertension: insights from the Registry to Evaluate Early and Long-Term Pulmonary Arterial Hypertension Disease Management (REVEAL).
        Circulation. 2010; 122: 164-172
        • Benza R.L.
        • Gomberg-Maitland M.
        • Miller D.P.
        • et al.
        The REVEAL Registry risk score calculator in patients newly diagnosed with pulmonary arterial hypertension.
        Chest. 2012; 141: 354-362
        • Kane G.C.
        • Maradit-Kremers H.
        • Slusser J.P.
        • Scott C.G.
        • Frantz R.P.
        • McGoon M.D.
        Integration of clinical and hemodynamic parameters in the prediction of long-term survival in patients with pulmonary arterial hypertension.
        Chest. 2011; 139: 1285-1293
        • Cogswell R.
        • Kobashigawa E.
        • McGlothlin D.
        • Shaw R.
        • De Marco T.
        Validation of the Registry to Evaluate Early and Long-Term Pulmonary Arterial Hypertension Disease Management (REVEAL) pulmonary hypertension prediction model in a unique population and utility in the prediction of long-term survival.
        J Heart Lung Transplant. 2012; 31: 1165-1170
        • Cogswell R.
        • McGlothlin D.
        • Kobashigawa E.
        • Shaw R.
        • De Marco T.
        Performance of the REVEAL model in WHO Group 2 to 5 pulmonary hypertension: application beyond pulmonary arterial hypertension.
        J Heart Lung Transplant. 2013; 32: 293-298
        • Escribano-Subias P.
        • Blanco I.
        • Lopez-Meseguer M.
        • et al.
        Survival in pulmonary hypertension in Spain: insights from the Spanish registry.
        Eur Respir J. 2012; 40: 596-603
        • Simonneau G.
        • Galie N.
        • Rubin L.J.
        • et al.
        Clinical classification of pulmonary hypertension.
        J Am Coll Cardiol. 2004; 43: 5S-12S
        • Nickel N.
        • Golpon H.
        • Greer M.
        • et al.
        The prognostic impact of follow-up assessments in patients with idiopathic pulmonary arterial hypertension.
        Eur Respir J. 2012; 39: 589-596