Portopulmonary hypertension (PoPH) is defined by the European Respiratory Society and the European Association for the Study of the Liver as the presence of pulmonary arterial hypertension (PAH) associated with cirrhotic or noncirrhotic portal hypertension.1 Natural history of PoPH is pejorative with a slow increase in PAH. Mortality rate is 86% at 5 years, mostly related to the evolution of the liver disease.2,3
Up to 60% of the patients are asymptomatic at the time of diagnosis,4 and initial presentation is typically nonspecific. Therefore, transthoracic echocardiography with measures of systolic pulmonary arterial pressure (sPAP) is systematically performed as a screening test before liver transplantation (LT).5 Diagnosis is suggested by either elevation of the right ventricular systolic pressure greater than 50 mm Hg or right ventricular dysfunction on echocardiography and requires confirmation by right heart catheterization (RHC).5,6 The European Respiratory Society based the definition of PoPH on the association of 3 hemodynamic parameters measured during RHC7: (i) mean pulmonary arterial pressure (mPAP) greater than 25 mm Hg at rest, (ii) pulmonary vascular resistance (PVR) greater than 240 dynes/s per cm−5 (iii) pulmonary occlusion pressure less than 15 mm Hg.
Prevalence of PoPH ranges between 0.5% and 5%4,5,8-11 among patients with portal hypertension. The risk of PoPH is independent from the severity of associated liver disease but may be associated to female sex and autoimmune hepatitis.12 Although PoPH is characterized by mechanisms similar to those of other forms of PAH, its exact pathophysiology remains unknown.13
Perioperative mortality after LT in patients with PoPH seems to be related to the severity of the condition: mild PoPH (mPAP inferior to 35 mm Hg) does not increase mortality risk after LT, but moderate disease (mPAP between 35 and 50 mm Hg) is associated with 50% perioperative mortality. Severe disease (mPAP superior to 50 mm Hg) is universally fatal.11,14
Hence, uncontrolled PoPH remains a contraindication to LT. Small series suggest an excellent outcome after LT with low cardiac-related mortality if mPAP can be reduced to less than 35 mm Hg2,10,15,16 before transplant, acting as a « bridging therapy». However, no prognostic factors of PAH evolution and of the necessity of vasoactive drugs after LT17 have been identified.
The aim of our study was to study outcomes after LT in patients with PoPH and to identify factors associated with recovery of PAH after LT.
MATERIALS AND METHODS
Patients and Centers
All patients with PoPH who underwent LT between 2008 and 2016 in 8 of 14 French transplant centers (Besançon, Clermont-Ferrand, Lille, Lyon, Montpellier, Strasbourg, Toulouse, and Tours) were retrospectively included.
PoPH was established by RHC before or during LT according to the definition of the European Respiratory Society. Patients whose diagnosis of PoPH was performed after LT or on echocardiography only were not included.
Identification of recipients with pretransplant PoPH was retrieved from the list of the French national transplant agency (Agence de la Biomédecine) and from local data in each transplant center. Since 2008, patients with PoPH can benefit from exceptions to the French Model for End-Stage Liver Disease (MELD) score-based liver allocation system and are listed by the French national transplant agency. Medical records of selected patients were reviewed for characteristics of the underlying liver disease, cardiac hemodynamic values, vasodilator treatment, and postliver transplant outcomes.
Routine Cardiopulmonary Assessment Before LT
In France, echocardiogram and functional respiratory tests were systematic before LT. In case of PAH suspicion on echocardiogram (sPAP > 50 mm Hg or right ventricular dysfunction), an RHC was performed. Stress test, dobutamine stress echocardiogram and coronarography were performed according to the cardiologist.
Hemodynamic indicators that were used for PoPH assessment were mPAP, PVR, and pulmonary occlusion pressure as recommended by The European Respiratory Society13: (i) mPAP greater than 25 mm Hg at rest, (ii) PVR greater than 240 dynes/s per cm−5 and (iii) pulmonary occlusion pressure less than 15 mm Hg.
Based on the ERS Task Force, PoPH severity was defined by mPAP value: mild disease if mPAP is less than 35 mm Hg, moderate disease if 35 ≤ mPAP is less than 50 mm Hg and severe disease when mPAP ≥ 50 mm Hg.18
Clinical indicators were dyspnea with New York Heart Association (NYHA) heart failure classification, varying from class I (no dyspnea) to class IV (dyspnea while at rest) score19 and 6-minute walk test (6MWT) with the distance in meters travelled by the patient in a 6-minute period.20
The study was approved by the regional ethics committee (number 2016-059).
Several parameters at the time of diagnosis were tested for PAH recovery after LT: (i) sex and age (50 years, cutoff) (ii) hemodynamic values: mPAP (50 mm Hg, cutoff), PVR (500 dynes/s per cm−5, cutoff), cardiac index (3.8 mL/min per m2, cutoff), (iii) NYHA score (NYHA I-II/III-IV), (iv) MELD score (17, cutoff), (v) combination therapy or monotherapy.
Combination therapy was defined by the association of at least 2 therapeutic classes among the 3 usually used in PoPH: endothelin receptor antagonists, phosphodiesterase type-5 inhibitors, and prostanoids.
Because RHC was not available after treatment onset and before LT for all the patients, the endpoint of the study was the normalization of pulmonary artery pressure (PAP) after LT.
Quantitative values were described as median with extremes ranges into brackets. Qualitative values were described as absolute numbers and percentages. Quantitative variables (hemodynamic parameters and pulmonary symptoms) were compared between different timepoints using the paired t test or the Wilcoxon matched-pairs signed rank test according to the number of patients. The Fisher test was used to compare qualitative variables between nonpaired groups. Kaplan-Meier curves were used for time to PAH normalization and patient survival after LT. Univariate analysis of prognostic factors for PAH recovery was performed with the Wilcoxon test; odds ratios (OR) are displayed with 95% confidence interval (CI). In regard to the low number of event, a multivariate analysis could not be performed. All P values were 2-sided and considered significant when less than 0.05.
Statistical analysis was performed using XLstat Version 2014.06.01, with GraphPad PRISM Version 5.0. for graphs.
During the study period, 3315 LT procedures were performed in the 8 centers. Twenty-eight (0.8%) patients presented PoPH. Among them, 5 patients were not included in the analysis for the following reasons: diagnosis of PAH was only confirmed after the LT procedure (n = 3), diagnosis of PoPH was performed on echocardiography without RHC (n = 1), LT was cancelled because of severe PoPH diagnosis in the operating room (n = 1).
Median follow-up of the 23 patients included in the analysis was 47 (0-94) months.
Characteristics of the patients are displayed in Table 1. Almost half of the patients were female (n = 10, 43.5%) and median age at diagnosis was 54 (42-64) years old. Half of the patients had hepatocellular carcinoma (n = 12, 52.2%). Main cause of cirrhosis was alcohol consumption (n = 16, 69.6%) and none of them had noncirrhotic portal hypertension. Liver disease was severe with 47.8% (n = 11) of the patients being Child C. Indication for LT was based on the severity of liver disease and/or presence of hepatocellular carcinoma. The presence of PoPH alone did not represent an indication for transplantation in these patients, even if 2 patients (8.7%) benefited from an exception to the French MELD score based liver allocation system. Median time between PoPH diagnosis and LT was 11 (0-88) months.
Median time on the waiting list was 8 (1-18) months (data available for 10 patients). For the 2 patients with an exception to the French MELD score based liver allocation system for PoPH, the waiting time was 4 and 10 months.
At diagnosis, 20 (87.0%) patients presented dyspnea on exertion (NYHA functional class II or III), one (4.3%) patient had dyspnea at rest and the 2 last one (8.7%) were asymptomatic (NYHA I). Seven (30.4%) patients were under beta-adrenergic blocking agents that were stopped shortly after diagnosis of PoPH. Median mPAP was 46 (28-55) mm Hg and median PVR was 459 (288-1620) dynes/s per cm−5 (see Table 2). Seven (29.1%) patients presented with severe disease (mPAP ≥ 50 mm Hg), 13 (54.2%) patients with moderate disease (35 ≤ mPAP < 50 mm Hg) and 4 (16.7%) patients with mild disease (mPAP < 35 mm Hg).
Diagnosis of PoPH was established in the operating room before abdominal incision through pulmonary artery catheterization in 5 (21.7%) of 23 patients. PoPH was moderate in 3 of these patients (mPAP, 38-40-46 mm Hg, respectively; PVR, 312-400-356 dynes/s per cm−5). In the other 2 patients, PoPH was severe (mPAP, 54 and 50 mm Hg, respectively; PVR, 1145 and 1620 dynes/s per cm−5) but the transplant procedure was not deferred. All these patients benefited from echocardiogram before LT: it was considered normal with sPAP less than 50 mm Hg (sPAP, 45-45-40-35-30 mm Hg).
Medical treatment of PoPH was introduced for all patients shortly after diagnosis. Three therapeutic classes were represented: endothelin receptor antagonists, phosphodiesterase type-5 inhibitors, and prostanoids. One patient was treated by prostanoids only (4.3%), 12 (52.3%) patients received a combination therapy (endothelin receptor antagonists + phosphodiesterase type-5 inhibitors + prostanoids), 7 (30.4%) patients were treated by phosphodiesterase type-5 inhibitors only (sildenafil or tadalafil), and 3 (13.0%) patients received endothelin receptor antagonists (bosentan or ambrisentan).
Treatments were well tolerated. One patient presented cytolysis as a side effect 1 month after the beginning of Bosentan, which was replaced by Sildenafil.
Sixteen (89%) of 18 patients who were diagnosed before LT procedure had a reassessment with RHC. Median RVP decreased from 498 (288-848)dynes/s per cm−5 to 216 (66-592) dynes/s per cm−5 (P < 0.001) and median mPAP decreased from 46 (28-54) mm Hg to 33 (16-50) mm Hg (P < 0.016). Median time between treatment introduction, which occurred shortly after diagnosis, and the first hemodynamic reassessment was 7 (1-17) months.
The 2 patients who did not have a repeated RHC underwent the LT procedure 1 and 3 months after the introduction of vasoactive treatment. They were reassessed by echocardiogram and sPAP was 39 and 43 mm Hg.
Median time between the first and the last evaluation was 17.5 (4-63) months.
Posttransplant Course: Mortality
Perioperative and postoperative mortality rate was 8.7% (2/23). Two patients specifically died from acute right heart failure after worsening of PoPH at day 3 and day 8 posttransplant; they both had moderate PoPH at diagnosis. One patient had RHC 5 months before LT and after vasoactive treatment introduction that showed satisfactory parameters (PVR, 296 dynes/s per cm−5; mPAP, 31 mm Hg; cardiac output [CO], 6.7 mL/min; cardiac index, 4 mL/min per m2). The other patient was diagnosed during LT (PVR, 356 dynes/s per cm−5; mPAP, 46 mm Hg; CO, 6.1 mL/min; cardiac index, 3.4 mL/min per m2).
Other causes of death after LT included hepatocellular carcinoma recurrence at 6 months in a patient with normalized mPAP before LT, and ischemic cholangitis at 9 months while Epoprostenol was about to be stopped after mPAP normalization.
Patients' survival curves are displayed in Figure 1. Survival rates since PoPH diagnosis were 95.6% at 3 months, 91.3% at 1 year, and 82.6% at 3 years. Survival rates after LT are 91.3% at 3 months, 82.6% at 1 and 3 years.
Posttransplant Course and Long-Term Evaluation: Evolution of PAH
Monitoring of PAH after LT was performed with RHC, except for 2 patients who had echocardiography only. Clinical parameters and hemodynamic values were compared at the time of diagnosis and after LT and are detailed in Table 2. Median time between LT and the last RHC available was 7 (0-17) months. Three hemodynamic parameters were significantly improved after LT: median mPAP decreased from 46 (28-55) mm Hg to 31.5 (20-60) mm Hg (P = 0.001), median PVR from 459 (288-1620) dynes/s per cm−5 to 232 (77-1624) dynes/s per cm−5 (P = 0.017), and sPAP improved from 65 (30-110) mm Hg to 41 (30-88) mm Hg (P = 0.0196) on echocardiogram. Twenty-one (91.3%) patients had a contributory echocardiogram before LT and sPAP could not be assessed for the 2 other patients because of the absence of cardiac tricuspid valve regurgitation. There was no significant difference before and after LT for cardiac index, CO, and capillary pressure, as well as for clinical parameters (dyspnea and 6MWT).
After LT, PAP (i) increased in 2 (8.7%) patients, leading to death at day 3 and day 8; (ii) remained stable under treatment in 7 (30.4%) patients; and (iii) normalized in 14 (60.9%) patients.
The 7 patients who remained stable under treatment without PAP normalization improved their hemodynamic parameters between last the assessment before LT and last available RHC (see Table 1, SDC, http://links.lww.com/TP/B500).
In patients with PAH recovery, median time between vasoactive treatment onset and PAP normalization was 11.5 (1-66) months. Recovery from PAH occurred before LT in 9 (64.3%) patients and after LT in 5 (35.7%) patients. Long-term follow-up was available for the 8 patients with PAP normalization and vasoactive drugs cessation: 63.5 (35-76) months from LT and 32.5 (16-59) months after pulmonary artery pressure normalization and vasoactive drugs cessation. None of the patients benefited from RHC after PAH recovery. Echocardiography was performed for 6 patients every 2 to 3 years, and 2 patients had only clinical follow-up every 6 months. None of them experienced PoPH recurrence.
All patients (n = 18) who had diagnosis of PoPH before LT were still treated with vasoactive drugs at the time of LT. Treatment was stopped without disease recurrence in 8 (34.8%) patients 18.5 (5-60) months after LT.
Among the 5 patients whom PoPH was diagnosed in the operating room: (i) 1 patient died from acute right heart failure at postoperative day 3; (ii) 2 patients with severe PoPH at diagnosis (mPAP, 54 and 53 mm Hg) remained stable under treatment without PAH recovery; (iii) 2 patients normalized their PAP, allowing cessation of vasoactive treatment.
In the 21 patients available for long-term analysis, time to PAH recovery from diagnosis is shown in Figure 2. Median time to PAP normalization was 11.5 (1-66) months, and 1-year normalization rate was 34.8%.
Considering that patients with mild PoPH do act as patients without PoPH, we analysed the outcome for the subgroup of 19 patients with moderate or severe PoPH at diagnosis: 2 (10.5%) patients died in the postoperative course from acute right heart failure, 10 (52.6%) patients normalized their PAP (among whom 7 patients stopped vasoactive drug treatment), and 7 (36.8%) patients remained stable under treatment without PAH recovery.
Prognostic Factors of PAH Recovery
In univariate analysis, the following parameters were tested: sex, hemodynamic values at diagnosis (cardiac index, mPAP, PVR), dyspnea NYHA score, MELD at the time of LT, age at diagnosis, PoPH medical treatment (combination therapy or monotherapy) were tested for PAP normalization. The association of 2 different drugs for PAH treatment was the only significant factor associated with PAH recovery (P = 0.04; OR, 8.75; 95% CI, 1.24-61.68) (Table 3).
LT is feasible in patients with PoPH provided an effective treatment of PAH during pretransplant period. The indication of LT remains the severity of the liver disease, and PAH alone shall not advocate LT. Recovery of PAH can be obtained in 60.8% of the patients and is favored by a combination of vasoactive drugs.
Immediate posttransplant mortality attributed to right heart failure in patients with PoPH is 8.7% in our experience and remains almost twice as high as the overall posttransplant mortality rate on the French national territory (4.8%).21 However, it appears to be less than the rates in recently published series: 14.3% in another French series22 and 19.2% in the series from the United Kingdom.23 Interestingly, the use of combination vasoactive therapy was greater in our series (52%) as compared with Savale et al’s22 (less than 22%) and Verma et al’s23 (1 patient only). Indeed, the use of vasoactive combination therapy was the only significant factor associated with PAH recovery. Among the 2 patients who died secondary to acute right heart failure, 1 patient was diagnosed with moderate PoPH intra operatively and was not previously treated, and the other 1 received a monotherapy with by phosphodiesterase type-5 inhibitors.
Epoprostenol has been advocated by Montani et al24 as the reference treatment for PoPH.17,25 However, this drug is delivered by continuous injection and may be constraining for cirrhotic patients. The 2 major therapeutic classes used were phosphodiesterase type-5 inhibitors and endothelin receptor antagonists, whose administration is more convenient and that allowed PAH recovery in 9 (39.1%) of the cases.13,26-29 Therapeutic studies on vasoactive medication for PAH have not been performed in the specific subgroup of patients with portal hypertension and PoPH, as cirrhosis was mostly an exclusion criterion. The main difference between isolated PAH and PoPH remains the hyper circulatory state of portal hypertension,30 caused by pronounced splanchnic arterial vasodilatation, reduced systemic vascular resistance31,32 and the subsequent increase of the CO. As CO depends on the patient’s morphology therefore the cardiac index (CO/body surface area) seemed to us more accurate to compare patients. As mPAP is closely related to the cardiac index, the hyper dynamic state could overestimate the severity of PAH which is only based on mPAP. The hyper dynamic state and the cardiac index should therefore be taken into account in the future studies on PoPH and its medical treatment. However, we did not identify cardiac index as a prognostic factor for PAH recovery. In the literature, results are also discordant regarding the relation between cardiac index and PAH.2,33
This might allow to discriminate between “true” PoPH and hyperkinetic-related PAH. In our experience, the presence of a PAH is not an indication to LT, even if some patients may benefit from a MELD exception to the French liver score for PoPH. The indication for LT was based on the underlying liver disease, and associated PAH was considered to assess and prevent the cardiovascular risk of the transplant procedure. Our findings are consistent with the recently published experience in the UK and ILTS recommendations.18 Furthermore, LT does not always cure PAH, with only 17.4% of the patients being drug-free at 1-year posttransplantation.
To be able to treat PoPH before LT, a careful screening of PAH should be performed in the pretransplant period. In the recent ILTS recommendations,18 RHC should be performed when systolic PAP is above 50 mm Hg on Doppler echocardiography. Nevertheless, 5 patients in the present study had diagnosis of PoPH during LT and had a normal sPAP at preoperative screening echocardiogram. We hence advocate that RHC should be performed in case of clinical suspicion or sPAP above 30 mm Hg, and should be repeated on the waiting list. This had already been proposed by Colle et al (Hepatology 2003)34: with a 30-mm Hg threshold, sensibility was 100%, specificity 96%, positive predictive value 59% and negative predictive value 100% for the detection of PAH. Furthermore, right ventricular function evaluation appears to be necessary. On the echocardiogram, right ventricular dilation and tricuspid annular plane systolic excursion are the most useful markers of right heart failure. It was unfortunately rarely available on our echocardiograms.
In our study, 2 patients underwent LT procedure in spite of severe PAH assessed in the operating room. One patient was not known to have PAH, and the decision to carry on the LT procedure was taken by a multidisciplinary team of surgeons and physicians, considering the possibility to control PAH with Epoprostenol and Furosemide during the procedure and the need for an urgent transplant. The other patient received a vasoactive treatment before LT and has been reassessed before transplant by RHC that showed mPAP of 36 mm Hg. It has been considered that mPAP in the operating room was probably overestimated because of the presence of edema and ascites.
The main limits of this study are its retrospective nature and the rarity of the pathology, hence the large inclusion period and the necessity of 8 LT centers with the absence of PoPH treatment standardization. This study unfortunately failed to identify patients who were considered for LT but eventually contraindicated because of severe PAH despite vasoactive treatment. Prospective studies focusing on the treatment of PAH in the specific group of cirrhotic patients need to be performed, and decreasing protocols should be established after LT to really assess the impact of LT on PAH, as it has been proposed in the ILTS recommendations.18
The authors would like to thank the AFEF (French Association for the Study of Liver Disease) for its support in the design of the study and recruitment of participating centers.
1. Hervé P, Lebrec D, Brenot F, et al. Pulmonary vascular disorders in portal hypertension. Eur Respir J
2. Krowka MJ, Plevak DJ, Findlay JY, et al. Pulmonary hemodynamics and perioperative cardiopulmonary-related mortality in patients with portopulmonary hypertension undergoing liver transplantation. Liver Transpl
3. Swanson KL, Wiesner RH, Nyberg SL, et al. Survival in portopulmonary hypertension: Mayo Clinic experience categorized by treatment subgroups. Am J Transplant
4. Pilatis ND, Jacobs LE, Rerkpattanapipat P, et al. Clinical predictors of pulmonary hypertension in patients undergoing liver transplant evaluation. Liver Transpl
5. Krowka MJ, Swanson KL, Frantz RP, et al. Portopulmonary hypertension: Results from a 10-year screening algorithm. Hepatology
6. Thevenot T, Degano B, Briquez C, et al. L’hypertension porto-pulmonaire. Hépato-Gastro Oncol Dig
7. Montani D, Günther S, Dorfmüller P, et al. Pulmonary arterial hypertension. Orphanet J Rare Dis
8. Krowka MJ, Wiesner RH, Heimbach JK. Pulmonary contraindications, indications and MELD exceptions for liver transplantation: a contemporary view and look forward. J Hepatol
9. Khaderi S, Khan R, Safdar Z, et al. Long-term follow-up of portopulmonary hypertension patients after liver transplantation. Liver Transpl
10. Safdar Z, Bartolome S, Sussman N. Portopulmonary hypertension: an update. Liver Transpl
11. Krowka MJ, Miller DP, Barst RJ, et al. Portopulmonary hypertension: a report from the US-based REVEAL Registry. Chest
12. Kawut SM, Krowka MJ, Trotter JF, et al. Clinical risk factors for portopulmonary hypertension. Hepatology
13. Chabot F, Gomez E, Boyer L, et al. Porto-pulmonary hypertension. Rev Mal Respir
14. Raevens S, Geerts A, Van Steenkiste C, et al. Hepatopulmonary syndrome and portopulmonary hypertension: recent knowledge in pathogenesis and overview of clinical assessment. Liver Int
15. Ashfaq M, Chinnakotla S, Rogers L, et al. The impact of treatment of portopulmonary hypertension on survival following liver transplantation. Am J Transplant
16. Hollatz TJ, Musat A, Westphal S, et al. Treatment with sildenafil and treprostinil allows successful liver transplantation of patients with moderate to severe portopulmonary hypertension. Liver Transpl
17. Sussman N, Kaza V, Barshes N, et al. Successful liver transplantation following medical management of portopulmonary hypertension: a single-center series. Am J Transplant
18. Krowka MJ, Fallon MB, Kawut SM, et al. International Liver Transplant Society Practice Guidelines: diagnosis and management of hepatopulmonary syndrome and portopulmonary hypertension. Transplantation
19. Hurst JW, Morris DC, Alexander RW. The use of the New York Heart Association’s classification of cardiovascular disease as part of the patient’s complete Problem List. Clin Cardiol
20. ATS Committee on Proficiency Standards for Clinical Pulmonary Function Laboratories. ATS statement: guidelines for the six-minute walk test. Am J Respir Crit Care Med
22. Savale L, Sattler C, Coilly A, et al. Long-term outcome in liver transplantation candidates with portopulmonary hypertension. Hepatology
23. Verma S, Hand F, Armstrong MJ, et al. Portopulmonary Hypertension: still an appropriate consideration for liver transplantation? Liver Transpl
24. Montani D, Jaïs X, Sitbon O, et al. Pulmonary arterial hypertension. Rev Mal Respir
25. Krowka MJ, Frantz RP, McGoon MD, et al. Improvement in pulmonary hemodynamics during intravenous epoprostenol (prostacyclin): a study of 15 patients with moderate to severe portopulmonary hypertension. Hepatology
26. O’Connell C, Amar D, Boucly A, et al. Comparative safety and tolerability of prostacyclins in pulmonary hypertension. Drug Saf
27. Gall H, Sommer N, Milger K, et al. Survival with sildenafil and inhaled iloprost in a cohort with pulmonary hypertension: an observational study. BMC Pulm Med
28. Galiè N, Hoeper MM, Humbert M, 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 Heart J
29. Galiè N, Barberà JA, Frost AE, et al. Initial use of ambrisentan plus tadalafil in pulmonary arterial hypertension. N Engl J Med
30. Iwakiri Y, Groszmann RJ. The hyperdynamic circulation of chronic liver diseases: from the patient to the molecule. Hepatology
. 2006;43(2 Suppl 1):S121–S131.
31. Møller S, Bernardi M. Interactions of the heart and the liver. Eur Heart J
32. Porres-Aguilar M, Mukherjee D. Portopulmonary hypertension: an update. Respirology
33. Le Pavec J, Souza R, Herve P, et al. Portopulmonary hypertension: survival and prognostic factors. Am J Respir Crit Care Med
34. Colle IO, Moreau R, Godinho E, et al. Diagnosis of portopulmonary hypertension in candidates for liver transplantation: a prospective study. Hepatology