Published ahead of print on July 10, 2008, doi:10.1164/rccm.200804-613OC
© 2008 American Thoracic Society doi: 10.1164/rccm.200804-613OC
Portopulmonary HypertensionSurvival and Prognostic Factors1 Université Paris-Sud 11, UPRES 2705, Service de Pneumologie et Réanimation, Hôpital Antoine Béclère, Assistance Publique-Hôpitaux de Paris, Clamart, France; and 2 INSERM U773, Centre de Recherche Biomédical Bichat-Beaujon, CRB3 and Service d'Hépatologie, Hôpital Beaujon, Assistance Publique-Hôpitaux de Paris, Clichy, France Correspondence and requests for reprints should be addressed to Dr. Jérôme Le Pavec, M.D., Service de Pneumologie, Hôpital Antoine Béclère, 157 rue de la Porte de Trivaux, 92140 Clamart, France. E-mail: lepavec{at}gmail.com
Rationale: Portopulmonary hypertension (PoPH) can be defined as elevation of pulmonary arterial pressure and pulmonary vascular resistance in the setting of portal hypertension. Survival results in PoPH are contrasting, and prognostic factors need to be identified. Objectives: To analyze long-term survival in a large cohort of patients with PoPH with the aim of determining the independent variables affecting survival. Methods: We retrospectively analyzed charts of all patients referred to the French Referral Center for pulmonary arterial hypertension with the diagnosis of PoPH between 1984 and 2004. Measurements and Main Results: The study population comprised 154 patients; 57% male. Mean age at diagnosis was 49 ± 11 years, 60% of patients were in New York Heart Association functional class III-IV, and mean 6-minute walk distance was 326 ± 116 m. Hemodynamic measurements showed a mean pulmonary arterial pressure of 53 ± 13 mm Hg, cardiac index of 2.9 ± 0.9 L/min/m2, and pulmonary vascular resistance of 752 ± 377 dyn/s/cm5. Portal hypertension was related to cirrhosis in 136 patients, with a severity assessed as follows: Child-Pugh class A 51%, Child-Pugh class B 38%, Child-Pugh class C 11%. Overall survival rates at 1, 3, and 5 yr were 88, 75, and 68%, respectively. Multivariate regression analysis individualized the presence and severity of cirrhosis and cardiac index as major independent prognostic factors. Conclusions: Prognosis in PoPH is mainly related to the presence and severity of cirrhosis and to cardiac function. The place of pulmonary arterial hypertension–specific therapies remains to be determined in the setting of PoPH.
Key Words: pulmonary hypertension • portal hypertension • portopulmonary hypertension • cirrhosis • survival
Pulmonary arterial hypertension (PAH) is a life-threatening disease characterized by marked and sustained elevation of pulmonary vascular resistance, leading to an increase in pulmonary artery pressure, right ventricular failure, and death (1). PAH refers to a spectrum of pulmonary microvasculature lesions, including smooth muscle cell proliferation, intimal thickening, in situ thrombosis, and vasoconstriction, that increase pulmonary vascular resistance, thus impairing pulmonary blood flow (2). Chronic liver disease is a known cause of pulmonary vascular diseases, including hepatopulmonary syndrome, characterized by intrapulmonary vascular dilatations with hypoxemia and portopulmonary hypertension (PoPH), defined by the presence of PAH in the setting of portal hypertension. Portal hypertension, rather than the presence of underlying liver disease, seems to be the main determining risk factor for the development of PAH (3). Prospective hemodynamic studies show that 2 to 6% of patients with portal hypertension demonstrate significant pulmonary hypertension (4–6). This subgroup of patients has been shown to represent up to 10% of the PAH population in the French Registry of PAH (7). The presence of PAH in patients with liver disease contributes significantly to morbidity at the time of liver transplantation (8), and survival in the PoPH population is still a matter of debate (9). Some retrospective studies involving a small number of patients suggest that these patients have a poorer outcome when compared with patients with idiopathic PAH (10, 11), whereas other studies have reported better survival rates in patients with PoPH (12, 13). Moreover, prognostic factors in PoPH have never been determined. This study analyzed long-term survival in a large cohort of patients with PoPH followed at the French Referral Center for PAH with the aim of determining prognostic factors in PoPH. Some of the results of this study have been previously reported in the form of abstracts (14, 15).
Patient Population We retrospectively analyzed charts of all consecutive patients with the diagnosis of PoPH referred to the French National Center for PAH between 1984 and 2004. According to French legislation, the agreement of an ethics committee and informed consent are not required for retrospective collection of data corresponding to current practice. However, the database was anonymous and complied with the restrictive requirements of the Commission Nationale Informatique et Liberté, the organization dedicated to privacy, information technology, and civil rights in France. This study was approved by our Institutional Review Board. All patients with PAH and portal hypertension with or without cirrhosis were included. Exclusion criteria were chronic thromboembolic pulmonary hypertension; pulmonary hypertension secondary to left heart diseases; pulmonary hypertension associated with significant lung diseases; and PAH associated with other conditions, including HIV infection, congenital left-to-right shunts, connective tissue diseases, and exposure to anorexigens. The diagnosis of portal hypertension was based on hemodynamic measurement of a hepatic venous pressure gradient of more than 5 mm Hg or the presence of esophageal varices at endoscopy (16). Cirrhosis was documented by liver biopsy findings or typical clinical signs.
Functional and Hemodynamic Evaluations
Treatment of Pulmonary Hypertension
Statistical Analysis For the survival analysis, we used the first hemodynamic evaluation as the start point to determine the survival duration. The cutoff date was January 1, 2006. Patients lost to follow-up were censored as of the date of the last visit. The Kaplan-Meier method was used to estimate the survival proportion at each time point. The log-rank test was used for curve comparison. Univariate analysis based on the proportional-hazards model was used to examine the relation between survival and selected baseline demographic, functional, and hemodynamic variables. Results are expressed as hazard ratios with 95% confidence intervals. Multivariate analysis based on the Cox proportional hazards model was used to examine the independent effect on survival of each variable, controlling for possible confounders. Variables with a P value of less than or equal to 0.10 in the univariate analysis without colinearity were included in the multivariate model.
Patient Population The study population comprised 154 patients. Baseline demographic and clinical data, NYHA functional class, six-minute-walk distance, and hemodynamic parameters are shown in Table 1. Fifty-three patients (32%) had clinical signs of right heart failure, 24 (15%) had history of syncope, and three (2%) reported Raynaud's phenomenon. Patients with PoPH without cirrhosis were younger and had more severe pulmonary hypertension with a higher level of PVR.
Etiologies of Portal Hypertension Table 1 summarizes the etiologies of liver diseases. Child-Pugh scores were available for 119 out of the 136 patients with cirrhosis. Of these, 61 patients (51%) were classified into Child-Pugh class A, 45 (38%) into class B, and 13 (11%) into class C.
Baseline Exercise Capacity and Pulmonary Hemodynamics
Baseline Characteristics According to Child Pugh Score Portal hypertension was related to cirrhosis in 136 patients, with severity assessed as follows: Child-Pugh class A, 51% (n = 61); Child-Pugh class B, 38% (n = 45); Child-Pugh class C, 11% (n = 13). There were no significant differences in sex, NYHA functional class, six-minute-walk distance, and other hemodynamic parameters according to Child-Pugh score. Only the level of right atrial pressure was significantly lower in patients with Child-Pugh A cirrhosis, compared with patients with Child-Pugh B and C cirrhosis (8 ± 5 vs. 11 ± 5 mm Hg and 12 ± 8 mm Hg, respectively; P < 0.03).
Survival Analysis
The results of univariate analysis including variables measured at baseline in the overall population are shown in Table 3. The presence of cirrhosis, cirrhosis with B or C Child Pugh scores, high right atrial pressure, low cardiac index, and a decrease in mixed venous oxygen saturation were significantly associated with a poor outcome.
The absence of cirrhosis, Child-Pugh score, right atrial pressure, cardiac index, and the use of prostacyclin therapies were included in the multivariate model. Only the absence of cirrhosis, Child-Pugh scores B and C, and cardiac index were identified as independent prognostic factors (Table 4).
Causes of Death During the follow-up period, 54 patients died. The majority of deaths were attributed to PAH (n = 19) and liver disease (n = 18). In two cases, death was related to bowel obstruction and cancer. The cause of death remained unknown in the remaining 15 patients. According to the liver disease severity in the patients with cirrhosis, the distribution of causes of death was not different from the overall population of our patients.
First-Line–specific Therapies for Pulmonary Hypertension
Liver Transplantation
This study provides a longitudinal analysis of the largest population of patients with PoPH, showing that not only the hemodynamic profile but also the severity of the liver disease are the major independent prognostic factors in this population. Until now, epidemiologic data regarding PoPH were obtained from a few studies of small numbers of patients (4, 11, 20). None of these previous studies individualized variables associated with outcome in these patients. To the best of our knowledge, our study is the first to emphasize the role of the severity of liver disease on survival in patients with PoPH. In our patients with PoPH, the presence of cirrhosis (vs. extrahepatic portal hypertension) and its severity (Child Pugh B and C) were independent markers of worse survival. Scores such as the Model for End Stage Liver Disease and Child Pugh scores have been associated with survival in patients with cirrhosis (21), and the Child Pugh score was chosen to assess the severity of liver disease in our study. In a pattern similar to PAH, preservation of right ventricular function, as evidenced by a better cardiac index, seemed to be a major protective factor in PoPH (22). NYHA functional class was not related to survival in our study. This is an unexpected result because NYHA functional class has clearly been identified as a strong prognostic factor in idiopathic PAH (22). This raises the question of the reliability of NYHA functional class in assessing dyspnea in patients with cirrhosis. In these patients, dyspnea may be related to complex gas exchange impairments and/or muscular disorders (3). Presumably reflecting the epidemiology of portal hypertension in France, a moderate male predominance was found in our study, as it was by others (4, 10). However, when compared with the sex ratio in the cirrhotic population without PAH, the proportion of female patients was relatively higher in PoPH. This female predominance is also found in other types of PAH, underlying the potential role of a hormonal influence in the pathogenesis of PAH (1, 7). In a recent study, female sex was associated with a higher risk of PoPH than male sex (23). The etiology of cirrhosis was mainly alcohol consumption and hepatitis virus infection, and the severity of the cirrhosis showed a Child-Pugh class A predominance. These results were in accordance with the general epidemiology of portal hypertension in France (24). Another interesting result was the finding of a higher proportion of autoimmune cirrhosis compared with patients with cirrhosis who did not have PAH (24). In the study by Kawut and colleagues, autoimmune hepatitis was associated with an increased risk of PoPH (23). These results may suggest that autoimmunity could contribute to the pathogenesis of PAH in the setting of portal hypertension, as has been demonstrated in idiopathic PAH (25). The role of PAH-specific therapies remains unclear in the setting of PoPH. Encouraging results have been published in open-label studies (26–28), although therapeutic approaches for PoPH have never been properly evaluated (29). In our study, the use of various PAH therapies in 45 patients did not affect on long-term survival. However, 18 out of 45 patients (40%) have been treated with oral prostacyclin analog beraprost, which has never been demonstrated efficacious in associated forms of PAH (30). To date, few patients have been treated with the oral endothelin receptor antagonist bosentan, which has been shown to be efficacious and safe (31, 32). A small uncontrolled study suggests that patients treated with oral bosentan may have a better survival than patients treated by inhaled iloprost. In addition, the use of intravenous epoprostenol, which has a positive impact on long-term survival in patients with idiopathic PAH (22), could potentially aggravate portal hypertension in the setting of severe liver diseases (33). Furthermore, no survival difference was found between Child Pugh A patients treated with intravenous epoprostenol and a matched sample of Child Pugh A patients with the same hemodynamic severity who did not receive treatment (data not shown). Finally, period of management that is related to PAH treatment did not affect survival in our cohort of patients (data not shown). This emphasizes the need for randomized controlled trials to evaluate the efficacy and safety of PAH therapies in the setting of PoPH. Few survival data are available in PoPH, and these data lead to contrasting results (10–12). In the studies published by Robalino and colleagues (10) and Kawut and colleagues (11), PoPH is associated with a poor median survival, ranging from 8 months to 2.3 years. In the latter study, survival of patients with PoPH was worse than that of patients with idiopathic PAH (11). Our study showed evidence of better outcome in PoPH than previously reported, with overall survival rates at 1 and 3 years of 88 and 75%, respectively, versus 85 and 38% in the study by Kawut and colleagues (11). This better survival could be related to less severe pulmonary hypertension with higher cardiac output and lower PVR. In our patients with PoPH, survival was similar to that of our patients with idiopathic PAH (7).
In patients with PoPH, the question of indication and safety of orthotopic liver transplantation is still debated (34). Indeed, patients with PoPH who undergo OLT may improve, stabilize, or worsen after surgery (34), and patients with Among the six patients who underwent OLT without presurgical treatment, PAH remained stable in two or worsened in four, leading to death in two patients. An important finding was that PAH did not resolved in any case. These results suggest that after OLT, neither correction of portal hypertension nor normalization of right ventricular function reversed the structural changes of the pulmonary arteries once they settled and led to PAH. In patients who have improved pulmonary hemodynamics after OLT, elevation of pulmonary arterial pressures is related to hyperdynamic state, which is associated with normal PVR in the vast majority of cases (34). In those patients, pulmonary pressures returned to normal levels in parallel with normalization of cardiac output. Finally, the relationship between presurgical hemodynamics and mortality risk in the present study could not be established because of the limited number of patients who underwent OLT. Nevertheless, patients who stabilized after OLT were those who had the less severe pulmonary hemodynamic impairment before surgery (Table 5). This result suggests that, in the presence of PoPH, OLT should be considered only in the presence of mild to moderate PAH. When PAH is too severe, the use of presurgical PAH treatments may help to reduce the mortality risk in patients who undergo OLT. Given the high mortality risk due to OLT in the setting of PoPH and given that cirrhosis is a major prognostic factor in those patients, a multidisciplinary approach involving physicians from liver and pulmonary vascular centers would be appropriate to target the optimum timing for OLT. There are several limitations to our study. First, we performed a single-center study using a retrospective analysis, with all potential associated biases. Because this study extended over 20 years and the availability and effectiveness of PAH therapies varied over the time, the extrapolation of our data concerning specific PAH treatment in PoPH may be impaired. In conclusion, this study supports that presence and severity of the liver disease play a key role in outcome of patients with portopulmonary hypertension. A multidisciplinary approach is mandatory to achieve appropriate care and follow-up of the pulmonary and the hepatic components of the disease to target the optimum timing for OLT. Even if preservation of right ventricular function is associated with a better survival, our study was not able to demonstrate a significant role of PAH-specific therapies on prognosis. The impact of PAH-specific therapies on exercise tolerance, hemodynamics, and morbidity-mortality has to be properly evaluated in prospective, randomized controlled trials.
Writing assistance was provided by Dr. Laura Price, SpR Respiratory and Intensive Care Medicine, NW Thames, United Kingdom.
Originally Published in Press as DOI: 10.1164/rccm.200804-613OC on July 10, 2008 Conflict of Interest Statement: None of the authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Received in original form April 25, 2008; accepted in final form July 3, 2008
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
AT A GLANCE COMMENTARY
) greater than 25 mm Hg at rest or greater than 30 mm Hg during exercise, with a pulmonary capillary wedge pressure less than 15 mm Hg and pulmonary vascular resistance (PVR) greater than 3 mm Hg/L/min (240 dyn · s · cm–5) (
