Portopulmonary Hypertension

Portopulmonary hypertension (PPH) is characterized by the development of pulmo‐ nary arterial hypertension (PAH) associated with portal hypertension, with or without liver disease. It is defined as a mean pulmonary artery pressure (MPAP) greater than 25 mmHg, pulmonary vascular resistance (PVR) above 240 dynes.s.cm -5 , pulmonary artery occlusion pressure (PAOP) normal when less than 15 mmHg or transpulmona‐ ry gradient (TPG) > 10 mmHg. In the pulmonary hypertension classification PPH is classified in Group I. Pulmonary arterial hypertension in association with cirrhosis and portal hypertension is underdiagnosed. Epidemiological studies estimated that about 2–6% of patients with portal hypertension develop PPH. Mortality is directly propor‐ tional to measured MPAP and PVR. Mean pulmonary artery pressure is an independ‐ ent predictor of mortality, and many centers consider that values greater than 50 mmHg is an absolute contraindication to liver transplantation (LT). The aim of the review is to explore the current aspects of PPH relative to concept, diagnosis, and treatment.


Introduction
Pulmonary hypertension (PH) is defined as a mean pulmonary artery pressure greater than or equal to 25 mmHg at rest, and above 30 mmHg during exercise, measured by right heart catheterization (RHC) [1].
At the Third World Pulmonary Hypertension Symposium in 2003 in Venice (Italy) and the Fourth World Symposium in 2008 in Dana Point (California, USA), PPH was categorized into Group I Pulmonary Hypertension [1,2,16,17].
During the Fifth World Pulmonary Hypertension Symposium held in 2013 in Nice, France, the consensus was to maintain the general disposition of previous classification, with some modifications and updates [18], as seen in Table 1.  Table 1. Classification of pulmonary hypertension-2013 Nice/France [18].
Based on diagnostic criteria, PPH can also be defined as: an increase in mean pulmonary artery pressure (MPAP) > 25 mmHg, increased pulmonary vascular resistance (PVR) > 240 dynes.s.cm -5 , and a mean pulmonary artery occlusion pressure (PAOP) normal < 15 mmHg, in patients with portal hypertension and no other causes of pulmonary hypertension. These hemodynamic criteria are consistent with the definitions and classification proposed by the Third World Pulmonary Hypertension Symposium, according to the European Respiratory Society (ERS) Task Force on Pulmonary-Hepatic Vascular Disorders (PHD). Furthermore, a transpulmonary gradient (TPG) > 10 mmHg was finally recommended by the ERS Task Force on PHD [19], as seen in Table 2. 1.
Approximately 30-50% of patients with cirrhosis have a high-flow circulatory state, owing to splanchnic vasodilation and hyperdynamic circulation, and this may cause an increase in MPAP, despite lack of pulmonary vasculature remodeling. The hyperdynamic circulation is characterized by a high cardiac output (CO), a low systemic vascular resistance (SVR), and a low PVR [6,20]. Therefore, the proposed classification for severity of PPH was based on MPAP [19,21], as described in Table 3.  Table 3. Classification of severity of portopulmonary hypertension based on MPAP (mean pulmonary artery pressure) [19].
Mild PPH appears to have no impact on outcomes following LT. However, significant increases in pulmonary artery pressures are associated with high mortality rates. MPAP > 50 mmHg is associated with 100% mortality in patients undergoing LT. Mortality is 35-40% in MPAP ranging from 35 to 50 mmHg and from zero to 17% in MPAP < 35 mmHg [22].

Prevalence and survival
The first autopsy studies were carried by McDonnell et al. in 1983. Those authors reported a prevalence of 0.13% in PAH non-cirrhotic patients compared to 0.73% in patients with cirrhosis and portal hypertension. In biopsies of other clinical studies, the prevalence of PAH ranged from 0.61% to 2% in cirrhotic patients [8,23,24].
Hemodynamic data from prospective studies revealed that approximately 2-6% of patients with portal hypertension develop PPH [25,26].
In a study involving 362 patients from 1985 to 1993, Castro et al. [27] used the criteria MPAP > 25 mmHg and PVR > 120 dynes.s.cm -5 for diagnosis of PPH. Those authors concluded that increased MPAP is common in patients with advanced liver disease (20%), although PPH occurred in only 4% of patients (15 patients).
Ramsay et al. [28] reviewed severe PH in patients with advanced liver disease in a study from Baylor University Medical Center. Those authors evaluated 1205 consecutive LTs, between December 1984 and October 1995. The incidence of PPH was 8.5% (102 patients with MPAP > 25 mmHg, and 6.72% in the mild form, 1.16% in the moderate form, and 0.58% in the severe form), using the same criteria. Mortality was 30% in three years in mild to moderate PPH, 42% in nine months in severe PPH, and 71% at three years post-LT.
In 1990, Robalino et al. [29] found that patients suffering from PAH associated with portal hypertension had a 15-month survival mean and a 50% mortality rate within six months of diagnosis, compared to those with primary pulmonary hypertension who survived two to three years and had a 57% survival rate within two years of diagnosis.
In a retrospective cohort study (data collection from 1997 to 2001 at the University of Pennsylvania, with a 3-year follow-up), Kawut et al. [30] compared survival and hemodynamics in patients with PPH (n=13) and PAH (n=33, pulmonary arterial hypertension was idiopathic, familial or associated with anorexics). Many of those patients were treated with epoprostenol. Those authors concluded that death risk in patients with PPH increased two fold compared to patients with PAH. Estimates of 1-year and 3-year-survival rates were 85% and 38% for patients with PPH, 82% and 72% for patients with PAH respectively. Although PPH patients had a higher cardiac index and lower PVR than PAH, patient outcome was worse, and could be attributed to complications of portal hypertension.
In a retrospective analysis of 154 PPH patients diagnosed from 1984 to 2004 and referred to the French Center for Pulmonary Arterial Hypertension, Le Pavec et al. [31] found a survival rate of 88%, 75%, and 68% at one, three, and five years, respectively. In this study, mortality was related to cirrhosis severity (higher in patients with Child-Turcotte-Pugh class B and C) and to low cardiac index. Nowadays with the advent of better patient selection for LT and appearance of new drugs, it is hoped that this limited scenary will be changed.

Pathophysiogenesis
The development of PPH is independent on the cause of portal hypertension and severity of underlying liver disease. It is weakly correlated with the Child-Turcotte-Pugh [35] classification and is associated with mortality beyond that predicted by the MELD score (Model End-Stage Liver Disease) [16,36].
The pathogenesis mechanisms of PPH remain unclear, and the knowledge on its development comes from PAH because of features similarity. Both disorders are characterized by obstruction of pulmonary arterial blood flow with increased PVR. The lesions detected are: medial hypertrophy, intimal proliferation and fibrosis of muscular pulmonary arteries, thickening of the adventitia, and in situ thrombosis. Plexiform lesions are typically found in small muscular arteries, adjacent to a larger parent vessel, and large arterial vasodilatation. Necrosis of muscular arteries cause leakage of plasma proteins into the arterial wall, resulting in necrotizing inflammatory arteritis, a probable precursor of plexiform lesions [8,19,37].
All these changes lead to increased pulmonary vascular resistance with vasoconstriction, arterial wall remodeling, and in situ microthrombosis, among other angiogenic factors investigated, such as genetic susceptibility, increased production of inflammatory mediators, and neurohormones [19].
It is believed that hyperdynamic circulation with high cardiac output can cause PPH, which are influenced by hepatic dysfunction caused by liver cirrhosis. This condition of increased pulmonary blood flow seen in patients with portal hypertension determines an increase shear stress at the level of vasculature, that may lead to endothelial injury and dysfunction with vasoconstriction and progressive vascular remodeling [25,38].
Investigators have postulated that high concentrations vasoactive substances secondary to an imbalance between vasoconstrictor and vasodilator factors could reach the pulmonary circulation due to portosystemic shunts or defective hepatic metabolism, and initiate the pulmonary vascular injury present in PPH [19,25].
ET-1 is produced by the pulmonary endothelium and liver, and binding ET-1A and ET-1B receptors on smooth muscle cells results in vasoconstriction and mitogenesis [19].
In a prospective multicenter case-control study of 175 patients with liver disease, Kawut et al. [40] identified 34 patients with PPH. Those authors demonstrated that the risk of developing PPH was higher in females and patients suffering from autoimmune hepatitis, and lower in those with hepatitis C virus.
In a recent study, Roberts et al. [41] showed that genetic variation in estrogen signaling and cell growth regulators is associated with PPH.
In another study, the same authors demonstrated that serotonin transporter polymorphism is not associated with PPH [42].
The fact that the presence of a high cardiac output, can result in a degree of pulmonary hypertension with normal or near normal pulmonary vascular resistance, which might have led to erroneous interpretation and overestimation of the incidence of PPH [43].

Clinical presentation
Patients with PPH usually have symptoms similar to those observed in other forms of PAH [1,25].
Symptoms produced by the disease may be nonspecific. The most common symptoms are dyspnea, fatigue, and chest pain. Syncope, palpitations, and peripheral edema are less commonly observed. Symptoms arise when mean pulmonary artery pressure exceeds 40 mmHg [5,44].
Clinical symptoms of liver disease and portal hypertension may be present [25,45].
A prospective study by Hadengue et al. showed that 60% of patients with PPH were asymptomatic and 40% had exertional dyspnea [11,35].
Regarding cardiac auscultation, an increased pulmonic component of the second heart sound (P2) occurred in 82% of cases. A systolic murmur of tricuspid regurgitation was present in 69%, edema in 35%, and signs suggestive of right heart failure in 34% [22,29].
Differences between hepatopulmonary syndrome and portopulmonary hypertension are described according to Rodriguez-Roisin et al., as seen in  Table 4. Differences between hepatopulmonary syndrome (HPS) and portopulmonary hypertension (PPH).

Diagnosis
PPH is usually diagnosed after a diagnosis of portal hypertension is made. The mean interval between diagnoses of both conditions is 28 ± 38 months, according to a prospective study by Hadengue et al. [35]. Those authors reported that 40% of dyspneic patients were overlooked on clinical examination.
According to currently established and recognized diagnostic criteria, the American Association for the Study of Liver Disease (AASLD) has proposed transthoracic echocardiography screening of all LT candidates for noninvasive identification of any form of PH and patient selection for RHC [47].
Transthoracic echocardiography (TTE) provides a number of variables that correlate with right heart hemodynamics, including pulmonary artery pressure. Estimated pulmonary artery pressure (PAP) is based on maximum tricuspid regurgitant jet velocity. The simplified Bernoulli equation describes the relationship between tricuspid regurgitant jet velocity and peak tricuspid regurgitant pressure gradient is equal to 4X (tricuspid regurgitant jet velocity) 2 . This equation allows us to estimate systolic pulmonary artery pressure (SPAP), taking into account right atrial pressure (RAP): SPAP = (tricuspid regurgitant pressure gradient) + estimated RAP (which is equal to 5 or 10 mmHg), or Equation (1): x tricuspid regurgitation jet velocity meanRAP = + (1) In patients with severe tricuspid regurgitation, calculation of SPAP may be underestimated, thus the pulmonary hypertension is not precisely defined by Doppler for a threshold value of SPAP obtained [1].
Doppler TTE is a sensitive method for detection of PH, despite its low positive predictive value. Consequently, pulmonary hemodynamics should be measured by RHC in positive cases to substantiate diagnosis [1,19,46,47].
In a recent study, Raevens et al. [48][49][50] analyzed the accuracy of TTE in the detection of all forms of PPH for different cutoff values of SPAP. In SPAP values of 30 mmHg, those authors found a sensitivity of 100%, a specificity of 54%, a positive predictive value of 10%, and a negative predictive value of 100%. In SPAP values of 38 mmHg, findings were: 100% sensitivity, 82% specificity, 22% positive predictive value, and 100% negative predictive value. In SPAP values of 50 mmHg, 86% sensitivity, 95% specificity, 46% positive predictive value, and 99% negative predictive value were found.
The authors incorporated the presence or absence of right ventricle dilatation, concluding that TTE is a highly sensitive screening test for PPH detection. Currently, in the performance of RHC to confirm or rule out PPH, an SPAP cutoff of 30 mmHg may produce a high number of false-positive tests, resulting in low specificity, and low positive predictive values. An SPAP of 38 mmHg was associated with a lower number of false-positive tests and higher specificity, ensuring a negative predictive value of 100%, safely reducing the number of patients referred to RHC. An SPAP of 50 mmHg is associated with a decreased sensitivity of 86% and a risk of canceling LT at the time of surgery.
Right heart catheterization is the gold standard for diagnostic confirmation of pulmonary arterial hypertension, including PPH. RHC measures pressure, flow, and resistance, provides assessment of severity of hemodynamic impairment, and is useful for vasoreactivity testing of the pulmonary circulation. The following variables are measured systolic, diastolic and mean pulmonary artery pressure, RAP, PAOP, right ventricular pressure (RVP), cardiac output (CO) by thermodilution or by the Fick method, allowing calculation of pulmonary vascular resistance [19,48]. The PVR is calculed using following formula, Equation (2): In PPH, the vasoreactivity test should be performed to determine disease severity and identify which patients could benefit from vasodilator therapy. A acute vasodilator testing should be commonly performed using intravenous epoprostenol (IV) or inhaled nitric oxide (NO). The test is considered positive when MPAP decreases by ≥ 10 mmHg to an absolute value of MPAP ≤ 40 mmHg with increased or no change in CO [1,19].
MPAP may increase in different situations. First, many patients with advanced liver disease present a hyperdynamic, high-flow circulatory state, resulting from splanchnic vasodilation caused by portal hypertension, leading to a marked increase in MPAP and CO. However, PVR remains normal or decreased. Second, elevation of MPAP is due to increased central blood volume due to left ventricular (LV) abnormalities measured by PAOP, which reflects enddiastolic LV volume, resulting in varying effects on PVR. Traspulmonary gradient (TPG = MPAP -PAOP) can distinguish between excess volume (TPG < 10 mmHg) and vascular pulmonary abnormalities (TPG > 10 mmHg) [19]. Third, MPAP is elevated regardless of disease severity, due to increased PVR caused by changes in the pulmonary vascular bed with progressive obliteration to pulmonary arterial blood flow from the right ventricle (RV) to the lungs [19,36].  Table 5. Hemodynamic data obtained by right heart catheterization in advanced liver disease [19].

Frontiers in Transplantology
In all patients with pulmonary hypertension, RHC is essential for diagnostic confirmation and assessment of disease severity [1,19,36].
Diagnostic confirmation of cirrhosis by liver biopsy may strengthen the diagnosis of PPH [5,8].
Pulmonary artery catheterization obtained the following hemodynamic data [19,36], as observed in Table 5.

Treatment
Specific treatment of PAH are use in PPH and includes different classes of vasodilators, such as prostacyclin analogs, endothelin receptor antagonists, and phosphodiesterase type 5 inhibitors [19,25,38].
The goal of therapies is to improve haemodynamics by reducing mean pulmonary artery pressure and pulmonary vascular resistance, to improve the haemodynamic right ventricle, thus creating possibility for patients to become eligible for LT [51,52].
These drugs are used only after diagnostic confirmation of the disease by RHC, and patients meet diagnostic criteria for PPH, according to the ERS Task Force on PHD [16,19].
A decrease of > 20% in MPAP and PVR indicates that patients are responsive to vasodilators [11]. Publications and reports of a recent small case series have indicated that use of these drugs before and after LT results in clinical improvement. However, further studies are needed [53][54][55].
Epoprostenol is administered by continuous intravenous infusion. It is a potent pulmonary and systemic vasodilator, it has antiproliferative effects, and potent inhibitor of platelet aggregation. The drug also reduces MPAP, and probably improves exercise tolerance and hemodynamic parameters, but common adverse effects and complications are attributable to this drug: jaw pain, headache, diarrhea, nausea, and vomiting; others effects are described as infecction in infusion line, ascites, right heart failure, splenomegaly, severe thrombocytopenia, and leukopenia [1,19,25,56].
Endothelin are endogenous vasoconstrictors with a major role in the pathogenesis of PAH.
Bosentan is an orally active dual antagonist of endothelin 1A and 1B that reduces PVR, improving exercise capacity, functional class, pulmonary and cardiac hemodynamics, and even prevents clinical deterioration. It can elevate liver enzymes despite limited experience in PPH [1,19].Bosentan use should be avoided in patients with moderate to severe liver dysfunction and elevated liver enzymes.
Ambrisentan is a selective ET-1A with minimal effect on liver function and sitaxentan was withdrawn from the market due fatal liver injury registration [38].
These drugs block cyclic GMP degradation. Cyclic GMP is a second messenger for nitric oxide, thereby prolonging vasodilator mediation of NO, producing lower MPAP and PVR [1]. These should be use cautiously because it may increase portal hypertension by splanchnic vasodilation [38].
Reichenberger et al. [16,57] used sildenafil in 14 patients with PPH for 12 months. Of these patients, six received inhaled iloprost or treprostinil. Hemodynamics improved significantly within three months and was maintained at 12 months, when diagnosed by RHC. Other small studies have shown clinical improvement after safe and effective use of this drug.
Yamashita et al. [58] reported cases of two patients with advanced liver disfunction and thrombocytopenia who were successfully treated with a combination of two oral vasodilators, ambrisentan and tadalafil. They concluded that it may be a safe and effective option for selected patients with severe and rapidly progressing PPH.
Retrospective studies involving postoperative liver transplant have stated that PPH was an absolute contraindication to transplantation because of high perioperative mortality. It is currently known that better preoperative evaluation, early initiation of drug, and improved anesthetic and surgical conditions offer new treatment possibilities.

Liver transplantation
Liver transplantation is a highly complex procedure, since the organ is responsible for multiple functions in the body. The first unsuccessful attempt at orthotopic LT in humans was carried out in the United States in 1963 by Thomas Earl Starzl and staff. Starzl was named the father of modern transplantation. The first successful case was recorded in 1967. By the end of the 1960s, 33 transplants had been described worldwide. Subsequently, other teams started performing this surgery with a low survival rate [14].
PPH patients have a high mortality rate related to right heart failure. There are few treatment options and LT has become an attractive therapy with a potential for cure. The role of LT in the treatment of PPH has evolved over the past 15 years [16]. Over time, better results will be achieved by advances in the understanding of new immunosuppressive drugs, biologic drug activity, metabolism, surgical technique, evaluation and intraoperative monitoring in anesthesiology and intensive care [14]. The anesthesiologist has an important role in managing these high risk patients [60].
Studies have proved successful in practice, with the introduction of pulmonary arterial vasodilators after PPH diagnosis, lower pulmonary artery pressure, and improving right ventricular function obtained for patient referral to LT [16,19,64].
Kwo et al. [65] reported that four patients with severe PPH showed a marked reduction in MPAP and PVR after long-term use of epoprostenol, providing better results for LT candidates.
Mair et al. [66] described a poor outcome in a case report. The patient received epoprostenol for eight months before LT. PVR was reduced from 12 units to 3 Wood units, but the patient developed right heart failure unresponsive to conventional inhaled therapy in the LT perioperative period, and died 28 days later.
LT is a special case of right ventricular stress with a sharp 5-10% increase in CO during reperfusion. However, an increase in CO is unpredictable and may reach up to 300%, precipitating right heart failure in a RV that is already under strain [61,67]. Increased CO probably results from removal of blood flow obstruction through the portal vein in the diseased liver, associated with systemic vasodilatation caused by acid rain, and other metabolites originating from the new graft. There is a significant decrease in myocardial contractility, chronotropy, and systemic vascular resistance [61,68]. Once this occurs, a patient suffering from pulmonary hypertension is at great risk [61].

Study justification
We believe that understanding the aspects and nuances of this severe disease may raise awareness about the issue and increase scientific knowledge. Following recommendations proposed by the international scientific community will certainly contribute to solidify work done by a multidisciplinary team to decrease morbidity and mortality in PPH patients undergoing liver transplantation. ET-1B