Survival and Prognostic Factors in Systemic Sclerosis Associated Pulmonary Hypertension

Transcription

1 ARTHRITIS & RHEUMATISM Vol. 65, No. 9, September 2013, pp DOI /art , American College of Rheumatology Survival and Prognostic Factors in Systemic Sclerosis Associated Pulmonary Hypertension A Systematic Review and Meta-Analysis Guillaume Lefèvre, 1 Luc Dauchet, 2 Eric Hachulla, 1 David Montani, 3 Vincent Sobanski, 1 Marc Lambert, 4 Pierre-Yves Hatron, 4 Marc Humbert, 3 and David Launay 1 Objective. Pulmonary hypertension (PH) is a frequent and life-limiting complication of systemic sclerosis (SSc). However, data on survival rates and their evolution over time, as well as prognostic factors in SSc complicated by PH, are still conflicting. The aim of this study was to conduct a systematic review and meta-analysis of cohort studies to assess pooled survival and prognostic factors for survival in patients with SSc-associated PH. Methods. For this systematic review and metaanalysis, we searched the Medline and EMBase databases (January 1960 to January 2012). All cohort studies in which survival and/or prognostic factors for SSc-associated PH were reported were included in the analysis. We calculated the pooled survival rates and 1 Guillaume Lefèvre, MD, Eric Hachulla, MD, PhD, Vincent Sobanski, MD, David Launay, MD, PhD: Université Lille Nord de France, IMPRT IFR 114, Hôpital Claude-Huriez, and Centre Hospitalier Régional Universitaire de Lille, Lille, France; 2 Luc Dauchet, MD, PhD: Université Lille Nord de France, Hôpital Calmette, Centre Hospitalier Régional Universitaire de Lille, Institut Pasteur de Lille, and INSERM 744, Lille, France; 3 David Montani, MD, PhD, Marc Humbert, MD, PhD: Université Paris-Sud, AP-HP, and Hôpital Bicêtre, Le Kremlin-Bicêtre, France, and INSERM U999 and Centre Chirurgical Marie-Lannelongue, Le Plessis-Robinson, France; 4 Marc Lambert, MD, PhD, Pierre-Yves Hatron, MD: Université Lille Nord de France, Hôpital Claude-Huriez, and Centre Hospitalier Régional Universitaire de Lille, Lille, France. Drs. Montani and Humbert have received consulting fees, speaking fees, and/or honoraria from Actelion, Bayer, GlaxoSmithKline, Novartis, Pfizer, and Lilly (less than $10,000 each). Dr. Launay has received consulting fees, speaking fees, and/or honoraria from Actelion, GlaxoSmithKline, and Pfizer (less than $10,000 each). Address correspondence to David Launay, MD, PhD, Service de Médecine Interne, Hôpital Claude-Huriez, Rue Michel Polonovski, CHRU Lille, F Lille Cedex, France. david.launay@ chru-lille.fr. Submitted for publication December 2, 2012; accepted in revised form May 16, analyzed their evolution over time and identified prognostic factors for survival. Results. Twenty-two studies were included, representing a total of 2,244 patients with SSc-associated PH. The pooled 1-, 2-, and 3-year survival rates were 81% (95% confidence interval [95% CI] 79 84%), 64% (95% CI 59 69%), and 52% (95% CI 47 58%), respectively. Meta-regression did not reveal a significant change in survival over time, while baseline hemodynamic measures of PH severity were significantly correlated with survival. In patients with SSc complicated by pulmonary arterial hypertension (PAH), age, male sex, diffusing capacity for carbon monoxide (DLCO), pericardial effusion, and the parameters classically associated with the severity of idiopathic PAH, including the 6-minute walk distance, mean pulmonary artery pressure, cardiac index, and right atrial pressure, were significant prognostic factors. DLCO and pericardial effusion were the only prognostic factors in patients with interstitial lung disease related PH. Conclusion. Our meta-analysis revealed a poor pooled 3-year survival rate of 52% in patients with SSc-associated PH. Baseline hemodynamic measures of PAH severity, but not the period of time during which patients were included in the studies, correlated significantly with survival in patients with SSc-associated PAH. All of the prognostic factors typically observed in idiopathic PAH, including the 6-minute walk distance and right atrial pressure, were also prognostic factors in SSc-associated PAH. Pulmonary hypertension (PH) is a frequent complication of systemic sclerosis (SSc) and is one of the leading causes of morbidity and mortality in patients with this disease (1,2). In SSc, PH can be ascribed to 2412

2 SURVIVAL AND PROGNOSIS IN SSc-ASSOCIATED PULMONARY HYPERTENSION 2413 precapillary and isolated pulmonary arterial hypertension (PAH; group 1 of the updated clinical classification of PH), interstitial lung disease (ILD) related PH (included in group 3), or postcapillary PH (included in group 2) (3,4). Many studies have addressed survival and prognostic factors for survival in SSc-associated PAH (5 25), and it has usually been asserted that survival among patients with SSc-associated PH has improved over the years. In 1986, Stupi et al reported a 2-year cumulative survival rate of 40% (26); in 2003, Kawut et al reported 1-year survival of 55% for patients who were seen between 1997 and 2001 (21), and more recently, 3-year survival of 50% has been reported (14,24,25). However, real improvement in the prognosis over time has not yet been clearly demonstrated (27). Aside from the availability of new treatments and their possible impact, other factors explaining the observed variation in survival over time should be addressed, such as the baseline characteristics and disease severity of patients, which may differ between studies (27). Prognostic factors for survival in SSc-associated PAH and SSc-associated PH complicated by ILD have also been assessed in several studies (5 25,28) and were recently reviewed (29,30). It is important to identify such factors in order to ensure optimal care and facilitate appropriate monitoring (29 31). However, there are marked discrepancies between studies focused on SSc, and there is a lack of consensus on factors that usually predict mortality in idiopathic PAH (29,30). For example, a few studies have suggested that the 6-minute walk distance (14,19) and right atrial pressure (14) are significant prognostic factors in SSc-associated PAH, whereas the majority of other studies did not confirm these results (8,9,11,24,25). Therefore, right atrial pressure and the 6-minute walk distance usually are not considered to be significant prognostic factors for survival in SSc-associated PAH. Similarly, there are major discrepancies concerning the cardiac index in SSc, because some studies showed that it was a prognostic factor (11,14,24,25), whereas other studies demonstrated that it was not (8,9,12,17,19). Conversely, right atrial pressure, the 6-minute walk distance, and the cardiac index are recognized as important and indisputable prognostic factors for survival in idiopathic PAH (31 34). One explanation for this discrepancy is that increases in cardiopulmonary pressures (e.g., right atrial pressure and mean pulmonary artery pressure [mpap]) do not correlate with survival in SSc. Another explanation could be a lack of power in studies of SSc to accurately identify prognostic factors. Indeed, studies of idiopathic PAH that focused on prognostic factors have involved 264 1,262 patients, while similar studies in SSc involved only patients (10,28). In the current study, we conducted a systematic review and meta-analysis of cohort studies to assess pooled survival and prognostic factors for survival in SSc-associated PH. We also investigated the correlation between survival and the baseline hemodynamic measures of PAH severity or the period of time during which patients were included in the studies (time of inclusion). PATIENTS AND METHODS This meta-analysis was performed according to the Preferred Reporting Items for Systematic Reviews and Meta- Analyses (PRISMA) Statement protocol (35). Search strategy. We searched Medline and EMBase databases (January 1960 to January 2012) using 2 strategies that were adapted to the specificities of each database. The first strategy included the following terms: scleroderma, systemic OR systemic sclerosis AND pulmonary arterial hypertension OR pulmonary hypertension. The second strategy included the following terms: scleroderma, systemic OR systemic sclerosis AND death OR mortality. In addition, reference lists of the articles initially detected were searched by hand to identify additional relevant reports. The eligibility of references retrieved by the search was assessed independently by 2 of the authors (DL and GL), and disagreements were resolved at each step by consensus. The full text of the remaining articles, including the references, was examined to determine whether the articles contained relevant information. Selection criteria. Studies were included if they met the following criteria: the study group comprised adult patients with a diagnosis of SSc according to the American College of Rheumatology (36) and/or the criteria for the classification of early SSc described by LeRoy and Medsger (37); the participants had precapillary PH as defined by right heart catheterization; and 1-, 2-, and 3-year survival rates for patients with PH were reported. Studies were excluded if the patients with SSc were included in a group of patients with connective tissue disease or if only echocardiography was used to diagnose PH. Quality assessment of individual studies was performed independently by 2 of the authors (DL and GL), using the Newcastle Ottawa Scale for cohort studies. The scale allocates stars (maximum of 9) for quality of selection, comparability, exposure, and outcome of study participants (38). Any discrepancies were addressed by joint reevaluation of the original article. Data extraction. Data were extracted from the selected studies independently by 2 of the authors (DL and GL), using a predefined standardized form. When the available information was incomplete, attempts were made to contact the corresponding authors of the studies for additional information. Classification of PH as either PAH or ILD-associated PH was performed according to the authors criteria. According to the different studies, ILD-associated PH was defined as the

3 2414 LEFÈVRE ET AL association between precapillary PH and either ILD on highresolution computed tomography (HRCT) associated with a restrictive pattern defined by a total lung capacity (TLC) of 70% predicted or forced vital capacity (FVC) of 70% predicted (8); total extent of ILD on HRCT of 30% or FVC of 70% predicted (9); FVC of 60% predicted or moderateto-severe fibrosis (involvement of more than one-third of lung fields) on HRCT (14); TLC of 60% predicted or TLC of 60 70% predicted and moderate-to-severe fibrosis (grade 3 4) evident on HRCT (17); or extensive change on HRCT and/or TLC of 70% predicted (22). Some studies focused on patients with either PAH or ILD-associated PH. Other studies combined patients with PAH and ILD-associated PH in the same group but presented the survival and prognostic factors for these 2 subgroups separately. For the studies in which results for the subgroups were reported separately, we chose to present the results separately, as 2 different studies (8,14,17,22). Outcome assessment. The primary analysis focused on assessing pooled 1-, 2-, and 3-year survival and prognostic factors for survival in SSc-associated PH. Statistical analysis. We calculated weighted pooled summary estimates of baseline characteristics (mean and proportion), mortality rate, and hazard ratio (HR) for mortality. For each meta-analysis, we used the DerSimonian and Laird method. Accordingly, studies were considered to be a random sample from a population of studies. Heterogeneity was quantified using a chi-square heterogeneity statistic and by means of an I 2 statistic for each analysis. A random-effects model was used to combine data. The overall effect was estimated using a weighted average of individual effects, with weights inversely proportional to variance in observed effects. For baseline characteristics, Freeman-Tukey transformation for proportions was performed. To assess the pooled mortality rate, proportions were Arcsin transformed. When the mortality rate and/or 95% confidence interval (95% CI) was not available, these values were estimated according to data obtained from the survival curve, using an actuarial method. The pooled mortality rate and HRs were estimated, with 95% CIs. Metaregression was used to assess the impact of mid cohort year and baseline pulmonary vascular resistance (PVR) on the survival rate. Separate analyses were performed for PAH and ILD-associated PAH. All analyses were performed using R software and the metafor package (39). P values less than 0.05 were considered significant. RESULTS Selection of the studies. Our search of the Medline and EMBase databases for the period from January 1960 to January 2012 resulted in 1,987 citations. After the abstracts were evaluated, 90 studies were examined for full text, and the exclusion criteria were applied. Twenty-two articles were finally selected, which included 2,244 patients (Figure 1 and Table 1; additional information is available from the corresponding author). Eighty-three percent of the patients were female, 79% had limited cutaneous SSc, 18% had diffuse cutaneous SSc, 37% had anticentromere antibodies, and 13% had anti topoisomerase I antibodies. The mean age of the patients at the time of inclusion in the studies was 60.3 years, and most of the patients (83%) were white. Seventy-four percent of the patients were classified as New York Heart Association (NYHA) functional class III/IV, and 23% were classified as NYHA functional class I/II. Pericardial effusion was present in 30% of patients. The mean distance covered in a 6-minute walk was 283 meters. The mean baseline values of hemodynamic measures were as follows: mpap, 44 mm Hg; right atrial pressure, 8 mm Hg; cardiac index, 2.45 liters/minute/m 2 ; pulmonary capillary wedge pressure (PCWP), 9 mm Hg; PVR, 694 dynes seconds cm 5 ; SvO 2, 63%. The main characteristics as well as the quality assessment of each study included in the analysis are shown in Table 1. Survival of patients. The pooled 1-, 2-, and 3-year survival rates were 81% (95% CI 79 84%), 64% (95% CI 59 69%), and 52% (95% CI 47 58%), respectively (Figure 2 and results not shown; additional information is available from the corresponding author). Tests for heterogeneity revealed significant heterogeneity at 1, 2, and 3 years. Therefore, we assessed whether ILDassociated PH, mid cohort year, baseline hemodynamic measures of PH severity, and age at the time of SSc diagnosis could explain this heterogeneity. Studies focusing on ILD-associated PH were a significant factor contributing to the heterogeneity of 2- and 3-year survival but not 1-year survival. Because of these results, and because PAH and ILD-associated PH do not belong to the same group according to the updated clinical classification of PH (3), we chose to present the pooled results as well as results categorized according to the types of studies, i.e., those focusing on either ILD-associated PH or PAH. The 1-, 2-, and 3-year survival rates among patients with PAH were 82% (95% CI 79 85%), 67% (95% CI 63 72%), and 56% (95% CI 51 61%), respectively (Figure 2 and results not shown; additional information is available from the corresponding author). Conversely, 1-, 2-, and 3-year survival rates were 77% (95% CI 70 83%), 48% (95% CI 40 55%), and 35% (95% CI 24 47%), respectively, for patients with ILD-associated PH (Figure 2 and results not shown; additional information is available from the corresponding author). Mid cohort year was not a factor contributing to the heterogeneity of 1-, 2-, and 3-year survival rates. Meta-regression did not reveal a significant correlation (P 0.064) between mid cohort year and 1-year survival rates in patients with PAH (additional information

4 SURVIVAL AND PROGNOSIS IN SSc-ASSOCIATED PULMONARY HYPERTENSION 2415 Figure 1. Flow diagram showing the search strategies used to identify publications for inclusion in the study. SSc systemic sclerosis; RCT randomized controlled trial. is available from the corresponding author). However, this trend disappeared when we adjusted for PVR (P 0.362). Similarly, there was no significant correlation between mid cohort year and 2-year (P 0.292) and 3-year (P 0.313) survival rates in patients with PAH (additional information is available from the corresponding author). In SSc patients with ILD-associated PH, no significant correlation was observed between mid cohort year and 1-, 2-, and 3-year survival rates (results not shown). Concerning hemodynamic parameters, baseline PVR was a significant factor contributing to the heterogeneity of 1-, 2-, and 3-year survival rates. Moreover, residual heterogeneity was not significant after including PVR in the meta-regression. Meta-regression showed a significant correlation between PVR and 1-, 2-, and 3-year survival rates in patients with PAH (Figure 3 and results not shown). For patients with ILD-associated PH, the correlation between PVR and 1-year and 3-year survival rates was significant (results not shown). The ages of the patients at the time of SSc diagnosis were available in 7 studies focusing on PAH (5,7,12,14 16,25), and age was significantly and negatively correlated with 2-year survival among patients with SSc-associated PAH (P 0.01) but not with 1-year and 3-year survival. No correlation between age and survival was observed in the group with ILD-associated PH. Prognostic factors for survival. The prognostic factors for survival and pooled hazard ratios are summarized in Table 2. Among patients with PAH, age at the time of PAH diagnosis, male sex, NYHA functional

5 2416 LEFÈVRE ET AL Table 1. Author (ref.) Johnson et al (5) Hurdman et al (6) Hesselstrand et al (7) Launay et al (8) Launay et al (8) Le Pavec et al (9) Chung et al (10) Mathai et al (11) Launay et al (12) Badesch et al (13) Condliffe et al (14) Condliffe et al (14) Hachulla et al (15) Hachulla et al (16) Mathai et al (17) Mathai et al (17) Fisher et al (18) Williams et al (19) Girgis et al (20) Kawut et al (21) Mukerjee et al (22) Mukerjee et al (22) Kuhn et al (23) Launay et al (24) Campo et al (25) Humbert et al/ cohort 1 (28) Humbert et al/ cohort 2 (28) Baseline characteristics of patients included in the selected studies* No. of NOS stars Study period No. of Female patients sex, no. PH subtype Age at diagnosis, Limited years SSc, no. Diffuse SSc, no. NYHA functional class I/II, no. NYHA functional class III/ IV, no. 6-minute walk distance, meters FVC, % DLCO, % predicted predicted PAH PAH PAH Pericardial effusion, no PAH ILD-PH ILD-PH PAH PAH PAH PAH PAH ILD-PH PAH PAH PAH ILD-PH PAH PAH PAH PAH PAH ILD-PH PAH PAH PAH PAH PAH * NOS Newcastle Ottawa Scale; SSc systemic sclerosis; NYHA New York Heart Association; FVC forced vital capacity; DLCO diffusing capacity for carbon monoxide; PAH pulmonary arterial hypertension; ILD-PH interstitial lung disease associated pulmonary hypertension.

6 SURVIVAL AND PROGNOSIS IN SSc-ASSOCIATED PULMONARY HYPERTENSION 2417 Figure 2. Forest plots showing 3-year survival in each study of patients with systemic sclerosis and pulmonary hypertension complicated by interstitial lung disease (PH-ILD) or patients with pulmonary arterial hypertension (PAH) and pooled data for the PH-ILD group, the PAH group, and the total study group. Each square represents an individual survival, with the size of the square being proportional to the weight given to the study. The dotted and dashed vertical lines represent combined survival for each subgroup and the whole population, respectively. pub publication; Phet P value for heterogeneity. class, 6-minute walk distance (Figure 4), diffusing capacity for carbon monoxide (DLCO), right atrial pressure, mpap, cardiac index, PVR, SvO 2, and pericardial effusion were significant prognostic factors. Conversely, SSc subtype, presence or absence of anticentromere antibodies, FVC, and PCWP were not significant prognostic factors (additional information is available from the corresponding author). In patients with ILD-associated PH, only DLCO and pericardial effusion were prognostic factors, although just one study including pericardial effusion was available. DISCUSSION The main results of our analysis are as follows: 1) the pooled 1-, 2-, and 3-year survival rates in the whole population were 81% (95% CI 79 84%), 64% (95% CI 59 69%), and 52% (95% CI 47 58%), respectively; 2) high heterogeneity between studies was mainly explained by the presence of ILD-associated PH as well as baseline hemodynamic measures of PAH severity but not year of inclusion; 3) PAH and ILD-associated PH do not have the same prognosis and prognostic

7 2418 LEFÈVRE ET AL Figure 3. Correlation between baseline pulmonary vascular resistance (PVR) and 3-year survival in patients with pulmonary arterial hypertension (PAH), as determined by meta-regression. Each square represents an individual survival, with the size of the square being proportional to the weight given to the study. The grey diamonds represent the theoretical 3-year survival predicted by meta-regression. factors; 4) there was a significant correlation between survival rates and baseline hemodynamic measures but not time of inclusion; and 5) the majority of the prognostic factors typically observed in idiopathic PAH, some of which are usually thought not to be as relevant in SSc-associated PAH (e.g., right atrial pressure, mpap, and 6-minute walk test), are also prognostic factors in SSc-associated PAH but not in ILD-associated PH. The pooled 3-year survival rate of 52% confirms that PH is a severe complication of SSc and explains why PH is one of the leading causes of death in patients with SSc (2). It is often purported that the prognoses of idiopathic PAH and SSc-associated PH have improved over time. Mortality in idiopathic PAH is often evaluated with an equation using 3 variables (mpap, mean right atrial pressure, and cardiac index), which was devised to predict a patient s probability of survival according to the baseline values of these variables (40). In SSc, no attempt has been made to correlate survival with baseline characteristics of patients. Interestingly, we observed no significant relationship between mid cohort year and survival in SSc-associated PAH and SSc complicated by ILD-related PH. Although there was only a trend toward an association between mid cohort year and 1-year survival, it disappeared when we adjusted for PVR. This provides evidence against improved survival of SSc patients over time. Conversely, we observed a significant correlation between baseline PVR and survival. This result suggests that the observed difference in survival between studies is attributable to baseline disease severity rather than time of inclusion. It is therefore possible that heightened awareness of PAH complicating SSc might lead to earlier referral of patients with less-severe disease to PH reference centers and improved survival over time. This is in accordance with our recent report that patients who underwent screening for SSc-associated PAH had less severe PAH and a much better prognosis than SSc patients in whom PAH was diagnosed during routine clinical followup (28). Variable treatment strategies and availability of therapeutic agents were described in the included studies. The influence of these treatments was not assessed here because of the heterogeneity of the reported treatments and numerous missing data. Thus, the results of our study suggest that we should be cautious when asserting that survival in SSc-associated PAH has im-

8 SURVIVAL AND PROGNOSIS IN SSc-ASSOCIATED PULMONARY HYPERTENSION 2419 Table 2. Prognostic factors for survival in patients with PAH, patients with ILD-associated PH, and the whole population* Prognostic factor PAH ILD-associated PH Whole population Age, years 1.02 ( ) 1.02 ( ) 1.02 ( ) Male sex 1.57 ( ) 0.89 ( ) 1.35 ( ) SSc subtype, limited vs. diffuse 0.91 ( ) 0.71 ( ) 0.70 ( ) Presence vs. absence of ACAs 0.99 ( ) 1.12 ( ) 1.25 ( ) NYHA functional class III/IV vs. class I/II 2.53 ( ) 2.06 ( ) 2.47 ( ) 6-minute walk distance, per 100-meter increase 0.65 ( ) 0.83 ( ) 0.68 ( ) Forced vital capacity, % predicted 1.00 ( ) 1.02 ( ) 1.00 ( ) DLCO, % predicted 0.97 ( ) 0.97 ( ) 0.97 ( ) Right atrial pressure, mm Hg 1.06 ( ) 0.95 ( ) 1.02 ( ) Mean PAP, per 10 mm Hg increase 1.27 ( ) 1.11 ( ) 1.22 ( ) PCWP, mm Hg 0.96 ( ) 0.95 ( ) 0.96 ( ) Cardiac index, liters/minute/m ( ) 0.95 ( ) 0.65 ( ) PVR, dynes seconds cm ( ) 0.99 ( ) 1.06 (1/ ) SvO 2, % predicted 0.95 ( ) 0.95 ( ) Pericardial effusion 1.74 ( ) 2.44 ( ) 1.88 ( ) * Values are the pooled hazard ratio (95% confidence interval). PAH pulmonary arterial hypertension; ILD interstitial lung disease; PH pulmonary hypertension; SSc systemic sclerosis; ACAs anticentromere antibodies; NYHA New York Heart Association; DLCO diffusing capacity for carbon monoxide; PAP pulmonary artery pressure; PCWP pulmonary capillary wedge pressure; PVR pulmonary vascular resistance. proved over time and indicate that SSc patients in whom PAH is less severe have a better prognosis. The influence of treatment on survival has not yet been assessed. One of the main differences between SScassociated PH and idiopathic PAH is that SSc-associated PAH can belong to either group 1 or group 3 (ILDassociated PH) according to the updated clinical classification of PH. Our study confirms that ILD-associated PH and PAH do not have the same prognosis, and that ILD-associated PH is one of the major factors contributing to the heterogeneity of the studies. The pooled 3-year survival rates were 56% (95% CI 51 61%) and 35% (95% CI 24 47%) in PAH and ILD-associated PH, respectively. The dismal prognosis of ILD-associated PH Figure 4. Forest plots showing the prognostic value of the baseline 6-minute walk distance, represented by the pooled hazard ratio (HR) per 100-meter increase in patients with PH-ILD, patients with PAH, and the whole population. Each square represents an individual HR, with the size of the square being proportional to the weight given to the study. The diamonds represent the combined HRs. Dashed and dotted vertical lines represent the combined HR for each subgroup and the whole population, respectively. See Figure 2 for other definitions.

9 2420 LEFÈVRE ET AL in SSc is also consistent with the poor prognosis associated with the presence of PH in patients with idiopathic pulmonary fibrosis (41 45). Because of the differences in prognosis between PAH and ILD-associated PH, we chose to present results for the whole population and separately for these 2 subgroups. In SSc patients with ILD-associated PH, none of the studied parameters except DLCO and pericardial effusion was a prognostic factor. In the setting of ILD-associated PH, the decrease in DLCO is likely the result of progressive pulmonary vascular remodeling or a reduction in lung volume due to ILD. However, lower DLCO may also be related to the development of pulmonary venoocclusive-like disease, which appears to occur frequently in SSc (46,47) and which could be difficult to diagnose in the context of ILD. Because the presence of pulmonary venoocclusive-like disease was not reported in the studies, we were unable to analyze its influence on survival. In patients with PAH, the subtype of SSc, the presence/absence of anticentromere antibodies, FVC, and PCWP were not prognostic factors. Conversely, not only the cardiac index and PVR but also right atrial pressure, mpap, SvO 2, age at PAH diagnosis, male sex, 6-minute walk distance, DLCO, and pericardial effusion were significant prognostic factors. In addition, the significance of NYHA functional class as a prognostic factor was confirmed in our analysis. Interestingly, almost all of these prognostic factors have also been reported to be important in idiopathic, familial, and anorexigen-associated PAH (31,32,34). In the French PAH registry, male sex, low 6-minute walk distance, and cardiac output were significantly and even independently associated with a worse prognosis in idiopathic PAH (32,48). Similarly, in the Registry to Evaluate Early and Long-Term Pulmonary Arterial Hypertension Disease Management (REVEAL) (33), these factors, along with DLCO, were associated with prognosis (31,34,49). Again, a lower DLCO not only could reflect extension of pulmonary vascular disease but also could be related to the development of a pulmonary venoocclusive-like disease, which could be overlooked in SSc-associated PAH (46,47). Age at the time of PH diagnosis appears to be an important prognostic factor in both idiopathic PAH and SSc-associated PAH, as we suggested in a previous study (24) and confirm here. This is of particular interest, because patients with SSc-associated PAH are older (mean age in our analysis, 60.3 years) than patients with idiopathic PAH (mean age in the French PAH registry, 52.5 years [32]). This could contribute to the possibly worse prognosis in SSc. We also observed that age at the time of SSc diagnosis was negatively correlated with 2-year survival in patients with PAH. However, this result should be cautiously interpreted, because age at the time of SSc diagnosis was provided in only 7 of the PAH studies in the current meta-analysis (5,7,12,14 16,25), and because this result was not significant for 1-year and 3-year survival. An important finding of our analysis is that male sex was associated with a worse prognosis in patients with SSc-associated PAH. This association was also suggested by 2 of the studies (19,24), whereas the other studies in our meta-analysis did not show this result. Of note, male sex is often reported to be associated with a poor prognosis in SSc without PAH (50). In the French PAH registry, male sex was also associated with a worse prognosis in patients with idiopathic PAH in univariate and multivariate analyses, although the explanation underlying this observation is not known (32). The 6-minute walk distance was also (quite unexpectedly) shown to be a prognostic factor in our study. This test is often suggested to be a poor tool for assessing the severity of PH or ILD in patients with SSc (51,52). Half of the individual studies do not support a clear association between the 6-minute walk distance and the prognosis of patients with SSc-associated PAH. The limitations of the 6-minute walk distance in SSc are well established (53). It is highly probable that this test reflects many parameters that are not directly associated with PAH, such as musculoskeletal impairment or depression (53,54). Taken together, these data support the notion that the 6-minute walk distance is not a proven and efficient prognostic factor in SSc-associated PAH. However, our analysis clearly shows that the 6-minute walk distance is a prognostic factor in SSc patients with PAH. Interestingly, the HR for the 6-minute walk distance (per 100-meter increase) was 0.65 (95% CI ) for patients with PAH in our analysis and 0.67 (95% CI ) for patients with idiopathic PAH, familial PAH, or anorexigen-associated PAH in the French PAH registry (32), thereby suggesting similar prognostic values of the 6-minute walk distance. Indeed, this result demonstrates that when we focus on welldefined SSc-associated PAH (and not ILD-associated PH) and extend the number of patients by pooling studies, some of the misinterpreted prognostic factors such as the 6-minute walk distance are shown to be valuable prognostic factors, as in idiopathic PAH. This highlights that caution should be used when interpreting data from smaller studies in SSc. We also observed that the cardiac index and PVR

10 SURVIVAL AND PROGNOSIS IN SSc-ASSOCIATED PULMONARY HYPERTENSION 2421 are prognostic factors, which is consistent with the correlation between PVR and overall survival that we observed using meta-regression. More surprising was the fact that both right atrial pressure and mpap were also prognostic factors. Although right atrial pressure is clearly associated with the prognosis in idiopathic PAH (31,32,34), it often has been suggested to be less useful in predicting survival in SSc-associated PAH (25,29). Our analysis suggests that the sample sizes of these studies were probably too small and underpowered to show a significant association between these parameters and survival. As mentioned previously, our meta-analysis shows that a higher baseline mpap is correlated with a worse prognosis in SSc-associated PAH. There are marked discrepancies among studies of idiopathic PAH. The National Institutes of Health supported US registry for idiopathic PAH (40) and REVEAL (33) showed that mpap was a significant prognosis factor, and that a higher mpap was associated with a worse prognosis. Conversely, a French study of long-term intravenous administration of epoprostenol showed that a higher mpap is associated with a better prognosis (55). Moreover, neither a study of US patients diagnosed with idiopathic PAH and treated with epoprostenol (56) nor a study of patients in the French PAH registry (32) showed an association between baseline mpap and survival in idiopathic PAH. To explain these discrepancies in idiopathic PAH, it must be emphasized that mpap reflects only one aspect of the hemodynamic parameters and is determined not only by the PVR but also by right ventricular performance. Therefore, the mpap can decrease in patients with very severe PAH, and PVR should also be considered rather than mpap alone (57). Although most individual results were not statistically significant, our meta-analysis shows that a higher baseline mpap is significantly associated with a worse prognosis in SSc-associated PAH. Our study has several limitations. Some prognostic factors were assessed in only a few studies, thus weakening the role of meta-analysis. All of these single prognostic factors have been recently reviewed (29,30). Moreover, the number of studies and the number of SSc patients in the group with ILD-associated PH was limited. In this meta-analysis, we used inverse weighting proportional to variance. Therefore, larger studies, with presumably narrower confidence limits surrounding the point estimate of their effects, were given greater weight relative to smaller studies. A possible limitation could arise if large studies included a narrower spectrum of disease and cofactors compared with small studies. If that were the case, the current weighting would be biased toward greater internal validity (accurate representation of the cases studied) at the expense of external validity (accurate representation of the cases in the population). In order to address this potential limitation, we analyzed our results again by replacing inverse weighting proportional to variance with inverse weighting proportional to sample size. The results were very similar for each parameter, showing that internal and external validity were balanced. The only exception was observed for the relative risk associated with limited cutaneous SSc, which was 0.69 (95% CI ) with inverse weighting proportional to variance and 1.04 (95% CI ) (fixed effect) with inverse weighting proportional to sample size. This discrepancy is easily explained by one outlier study (24) that has the highest total number of subjects but the lowest number of patients with diffuse cutaneous SSc (reference class), which explains high variance despite a large sample size. In conclusion, this meta-analysis confirms poor pooled 3-year survival in SSc-associated PH. In patients with SSc, PAH and ILD-associated PH have different prognoses and prognostic factors. There was no relationship between survival and the time at which patients were included in the studies, while the prognosis was better for patients in whom the hemodynamic measures at baseline indicated lower disease severity. Almost all of the usual prognostic factors observed in idiopathic PAH were also prognostic factors in SSc-associated PAH, including the highly debated 6-minute walk distance, mpap, and right atrial pressure. The results of this analysis suggest that we should exercise caution when interpreting the results of studies with a limited number of patients with SSc-associated PH. AUTHOR CONTRIBUTIONS All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Launay 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. Study conception and design. Lefèvre, Montani, Sobanski, Hatron, Humbert, Launay. Acquisition of data. Lefèvre, Lambert, Launay. Analysis and interpretation of data. Lefèvre, Dauchet, Hachulla, Launay. REFERENCES 1. Le Pavec J, Humbert M, Mouthon L, Hassoun PM. Systemic sclerosis-associated pulmonary arterial hypertension. Am J Respir Crit Care Med 2010;181: Steen VD, Medsger TA. Changes in causes of death in systemic sclerosis, Ann Rheum Dis 2007;66:940 4.

11 2422 LEFÈVRE ET AL 3. Simonneau G, Robbins IM, Beghetti M, Channick RN, Delcroix M, Denton CP, et al. Updated clinical classification of pulmonary hypertension. J Am Coll Cardiol 2009;54:S Galie N, Hoeper MM, Humbert M, Torbicki A, Vachiery JL, Barbera JA, et al. Guidelines for the diagnosis and treatment of pulmonary hypertension: the Task Force for the Diagnosis and Treatment of Pulmonary Hypertension of the European Society of Cardiology (ESC) and the European Respiratory Society (ERS), endorsed by the International Society of Heart and Lung Transplantation (ISHLT). Eur Respir J 2009;34: Johnson SR, Granton JT, Tomlinson GA, Grosbein HA, Le T, Lee P, et al. Warfarin in systemic sclerosis-associated and idiopathic pulmonary arterial hypertension: a Bayesian approach to evaluating treatment for uncommon disease. J Rheumatol 2012; 39: Hurdman J, Condliffe R, Elliot CA, Davies C, Hill C, Wild JM, et al. ASPIRE registry: assessing the Spectrum of Pulmonary hypertension Identified at a REferral centre. Eur Respir J 2012; 39: Hesselstrand R, Wildt M, Ekmehag B, Wuttge DM, Scheja A. Survival in patients with pulmonary arterial hypertension associated with systemic sclerosis from a Swedish single centre: prognosis still poor and prediction difficult. Scand J Rheumatol 2011;40: Launay D, Humbert M, Berezne A, Cottin V, Allanore Y, Couderc LJ, et al. Clinical characteristics and survival in systemic sclerosis-related pulmonary hypertension associated with interstitial lung disease. Chest 2011;140: Le Pavec J, Girgis RE, Lechtzin N, Mathai SC, Launay D, Hummers LK, et al. Systemic sclerosis related pulmonary hypertension associated with interstitial lung disease: impact of pulmonary arterial hypertension therapies. Arthritis Rheum 2011;63: Chung L, Liu J, Parsons L, Hassoun PM, McGoon M, Badesch DB, et al. Characterization of connective tissue disease-associated pulmonary arterial hypertension from REVEAL: identifying systemic sclerosis as a unique phenotype. Chest 2010;138: Mathai SC, Bueso M, Hummers LK, Boyce D, Lechtzin N, Le Pavec J, et al. Disproportionate elevation of N-terminal probrain natriuretic peptide in scleroderma-related pulmonary hypertension. Eur Respir J 2010;35: Launay D, Sitbon O, Le Pavec J, Savale L, Tcherakian C, Yaici A, et al. Long-term outcome of systemic sclerosis-associated pulmonary arterial hypertension treated with bosentan as first-line monotherapy followed or not by the addition of prostanoids or sildenafil. Rheumatology (Oxford) 2010;49: Badesch DB, McGoon MD, Barst RJ, Tapson VF, Rubin LJ, Wigley FM, et al. Longterm survival among patients with scleroderma-associated pulmonary arterial hypertension treated with intravenous epoprostenol. J Rheumatol 2009;36: Condliffe R, Kiely DG, Peacock AJ, Corris PA, Gibbs JS, Vrapi F, et al. Connective tissue disease-associated pulmonary arterial hypertension in the modern treatment era. Am J Respir Crit Care Med 2009;179: Hachulla E, Launay D, Mouthon L, Sitbon O, Berezne A, Guillevin L, et al. Is pulmonary arterial hypertension really a late complication of systemic sclerosis? Chest 2009;136: Hachulla E, Carpentier P, Gressin V, Diot E, Allanore Y, Sibilia J, et al, and the ItinérAIR-Sclérodermie Study Investigators. Risk factors for death and the 3-year survival of patients with systemic sclerosis: the French ItinérAIR-Sclérodermie Study. Rheumatology (Oxford) 2009;48: Mathai SC, Hummers LK, Champion HC, Wigley FM, Zaiman A, Hassoun PM, et al. Survival in pulmonary hypertension associated with the scleroderma spectrum of diseases: impact of interstitial lung disease. Arthritis Rheum 2009;60: Fisher MR, Mathai SC, Champion HC, Girgis RE, Housten- Harris T, Hummers L, et al. Clinical differences between idiopathic and scleroderma-related pulmonary hypertension. Arthritis Rheum 2006;54: Williams MH, Handler CE, Akram R, Smith CJ, Das C, Smee J, et al. Role of N-terminal brain natriuretic peptide (N-TproBNP) in scleroderma-associated pulmonary arterial hypertension. Eur Heart J 2006;27: Girgis RE, Mathai SC, Krishnan JA, Wigley FM, Hassoun PM. Long-term outcome of bosentan treatment in idiopathic pulmonary arterial hypertension and pulmonary arterial hypertension associated with the scleroderma spectrum of diseases. J Heart Lung Transplant 2005;24: Kawut SM, Taichman DB, Archer-Chicko CL, Palevsky HI, Kimmel SE. Hemodynamics and survival in patients with pulmonary arterial hypertension related to systemic sclerosis. Chest 2003;123: Mukerjee D, St George D, Coleiro B, Knight C, Denton CP, Davar J, et al. Prevalence and outcome in systemic sclerosis associated pulmonary arterial hypertension: application of a registry approach. Ann Rheum Dis 2003;62: Kuhn KP, Byrne DW, Arbogast PG, Doyle TP, Loyd JE, Robbins IM. Outcome in 91 consecutive patients with pulmonary arterial hypertension receiving epoprostenol. Am J Respir Crit Care Med 2003;167: Launay D, Sitbon O, Cordier JF, Hachulla E, Mouthon L, Gressin V, et al. Survival and prognostic factors in patients with incident and newly diagnosed SSc-associated pulmonary arterial hypertension from the French Registry. Ann Rheum Dis. In press. 25. Campo A, Mathai SC, Le Pavec J, Zaiman AL, Hummers LK, Boyce D, et al. Hemodynamic predictors of survival in sclerodermarelated pulmonary arterial hypertension. Am J Respir Crit Care Med 2010;182: Stupi AM, Steen VD, Owens GR, Barnes EL, Rodnan GP, Medsger TA Jr. Pulmonary hypertension in the CREST syndrome variant of systemic sclerosis. Arthritis Rheum 1986;29: Girgis RE, Mathai SC, Wigley FM, Hassoun PM. Survival in systemic sclerosis-related pulmonary arterial hypertension in the modern treatment era. Am J Respir Crit Care Med 2009;180: Humbert M, Yaici A, de Groote P, Montani D, Sitbon O, Launay D, et al. Screening for pulmonary arterial hypertension in patients with systemic sclerosis: clinical characteristics at diagnosis and long-term survival. Arthritis Rheum 2011;63: Johnson SR, Swiston JR, Granton JT. Prognostic factors for survival in scleroderma associated pulmonary arterial hypertension. J Rheumatol 2008;35: Johnson SR, Granton JT. Pulmonary hypertension in systemic sclerosis and systemic lupus erythematosus. Eur Respir Rev 2011; 20: Benza R, Gomberg-Maitland M, Frost AE, Frantz RP, Humbert M, McGoon MD. Development of prognostic tools in pulmonary arterial hypertension: lessons from modern day registries. Thromb Haemost 2012;108: Humbert M, Sitbon O, Chaouat A, Bertocchi M, Habib G, Gressin V, et al. Survival in patients with idiopathic, familial, and anorexigen-associated pulmonary arterial hypertension in the modern management era. Circulation 2010;122: Benza RL, Miller DP, Gomberg-Maitland M, Frantz RP, Foreman AJ, Coffey CS, 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: Benza RL, Gomberg-Maitland M, Miller DP, Frost A, Frantz RP, Foreman AJ, et al. The REVEAL Registry risk score calculator in patients newly diagnosed with pulmonary arterial hypertension. Chest 2012;141: Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred Reporting

12 SURVIVAL AND PROGNOSIS IN SSc-ASSOCIATED PULMONARY HYPERTENSION 2423 Items for Systematic Reviews and Meta-Analyses: the PRISMA Statement. Ann Intern Med 2009;151: Subcommittee for Scleroderma Criteria of the American Rheumatism Association Diagnostic and Therapeutic Criteria Committee. Preliminary criteria for the classification of systemic sclerosis (scleroderma). Arthritis Rheum 1980;23: LeRoy EC, Medsger TA Jr. Criteria for the classification of early systemic sclerosis. J Rheumatol 2001;28: Wells G, Shea B, O Connell D, Peterson J, Welch V, Losos M, et al. The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. URL: ohri.ca/programs/clinical_epidemiology/oxford.asp. 39. Viechtbauer W. Conducting meta-analyses in R with the metafor package. J Stat Software 2010;36:1 48. URL: org/v36/i03/paper. 40. D Alonzo GE, Barst RJ, Ayres SM, Bergofsky EH, Brundage BH, Detre KM, et al. Survival in patients with primary pulmonary hypertension: results from a national prospective registry. Ann Intern Med 1991;115: Minai OA, Chaouat A, Adnot S. Pulmonary hypertension in COPD: epidemiology, significance, and management: pulmonary vascular disease: the global perspective. Chest 2010;137:39S 51S. 42. Weitzenblum E, Chaouat A, Canuet M, Kessler R. Pulmonary hypertension in chronic obstructive pulmonary disease and interstitial lung diseases. Semin Respir Crit Care Med 2009;30: Chaouat A, Naeije R, Weitzenblum E. Pulmonary hypertension in COPD. Eur Respir J 2008;32: Lettieri CJ, Nathan SD, Barnett SD, Ahmad S, Shorr AF. Prevalence and outcomes of pulmonary arterial hypertension in advanced idiopathic pulmonary fibrosis. Chest 2006;129: Cottin V, Le Pavec J, Prevot G, Mal H, Humbert M, Simonneau G, et al. Pulmonary hypertension in patients with combined pulmonary fibrosis and emphysema syndrome. Eur Respir J 2010;35: Gunther S, Jais X, Maitre S, Berezne A, Dorfmuller P, Seferian A, et al. Computed tomography findings of pulmonary venoocclusive disease in scleroderma patients presenting with precapillary pulmonary hypertension. Arthritis Rheum 2012;64: Dorfmuller P, Humbert M, Perros F, Sanchez O, Simonneau G, Muller KM, et al. Fibrous remodeling of the pulmonary venous system in pulmonary arterial hypertension associated with connective tissue diseases. Hum Pathol 2007;38: Humbert M, Sitbon O, Yaici A, Montani D, O Callaghan D, Jais X, et al. Survival in incident and prevalent cohorts of patients with pulmonary arterial hypertension. Eur Respir J 2010;36: Miller DP, Gomberg-Maitland M, Humbert M. Survivor bias and risk assessment. Eur Respir J 2012;40: Fransen J, Popa-Diaconu D, Hesselstrand R, Carreira P, Valentini G, Beretta L, et al. Clinical prediction of 5-year survival in systemic sclerosis: validation of a simple prognostic model in EUSTAR centres. Ann Rheum Dis 2011;70: Garin MC, Highland KB, Silver RM, Strange C. Limitations to the 6-minute walk test in interstitial lung disease and pulmonary hypertension in scleroderma. J Rheumatol 2009;36: Avouac J, Kowal-Bielecka O, Pittrow D, Huscher D, Behrens F, Denton CP, et al, for the EPOSS Group. Validation of the 6 min walk test according to the OMERACT filter: a systematic literature review by the EPOSS-OMERACT group. Ann Rheum Dis 2010;69: Schoindre Y, Meune C, Dinh-Xuan AT, Avouac J, Kahan A, Allanore Y. Lack of specificity of the 6-minute walk test as an outcome measure for patients with systemic sclerosis. J Rheumatol 2009;36: Deuschle K, Weinert K, Becker MO, Backhaus M, Huscher D, Riemekasten G. Six-minute walk distance as a marker for disability and complaints in patients with systemic sclerosis. Clin Exp Rheumatol 2011;29 Suppl 65:S Sitbon O, Humbert M, Nunes H, Parent F, Garcia G, Herve P, et al. Long-term intravenous epoprostenol infusion in primary pulmonary hypertension: prognostic factors and survival. J Am Coll Cardiol 2002;40: McLaughlin VV, Shillington A, Rich S. Survival in primary pulmonary hypertension: the impact of epoprostenol therapy. Circulation 2002;106: Saggar R, Sitbon O. Hemodynamics in pulmonary arterial hypertension: current and future perspectives. Am J Cardiol 2012;110 Suppl:9S 15S.