Portopulmonary hypertension

INTRODUCTION — Portopulmonary hypertension (PPHTN) refers to pulmonary arterial hypertension that is associated with portal hypertension; it is a well-recognized complication of portal hypertension due to chronic liver disease or extrahepatic causes [1-3].

In this topic review, the epidemiology, pathogenesis, diagnostic evaluation, treatment, and prognosis of PPHTN are reviewed. Portal hypertension is discussed separately. (See "Portal hypertension in adults" and "Noncirrhotic portal hypertension".)

DEFINITION — Pulmonary hypertension is classified into five groups [4] (groups 1 through 5) (table 1). Group 1 (also known as pulmonary arterial hypertension [PAH]) has several etiologies of which PPHTN is one. PPHTN is considered present when PAH exists in a patient who has coexisting portal hypertension, and no alternative cause of the PAH exists (eg, heritable PAH, collagen vascular disease, congenital heart disease, human immunodeficiency virus, or drugs) [1]. (See "Portal hypertension in adults" and "Noncirrhotic portal hypertension" and "Clinical features and diagnosis of pulmonary hypertension of unclear etiology in adults", section on 'Postdiagnostic testing and classification'.)

EPIDEMIOLOGY — PPHTN occurs in about 2 to 16 percent of patients with portal hypertension. Reported prevalence rates depend upon the patient population studied and the hemodynamic definitions used to define PPHTN. The prevalence does not appear to be influenced by the severity of portal hypertension or of liver disease [3,5,6]. Although higher rates are reported among patients with end-stage liver disease undergoing liver transplantation, this may reflect the greater risk for the development of portal hypertension in this population [7,8].

●In patients with liver cirrhosis, the prevalence rates range from 2 percent in those with chronic liver disease not listed for liver transplantation to 16 percent in those with end-stage liver disease listed for liver transplantation [3,7,9-14]. However, some of these studies did not use the criterion of elevated pulmonary vascular resistance (PVR), which appears to be important in distinguishing patients with chronic liver disease who have increased pulmonary flow from high cardiac output (characterized by low PVR) from those with true pulmonary arterial hypertension (PAH) characterized by high PVR >3 Wood units (240 dynes/sec/cm^-5).

●Among those with PAH, rates also vary. In the United States, registries report that 5 percent of patients with PAH have associated portal hypertension while in France, the proportion is about 15 percent [15,16]. However, hemodynamic definitions of PPHTN also vary among these registries, which may explain the wide range.

●Among patients with portal hypertension (eg, cirrhosis, portal vein thrombosis, hepatic vein sclerosis, congenital portal circulation abnormalities, and periportal fibrosis without cirrhosis), no etiology of portal hypertension or chronic liver disease is consistently associated with the development of PPHTN. While PPHTN was found in some studies to be more frequently associated with autoimmune cirrhosis and less frequently with hepatitis C-related cirrhosis [5,17], another study reported that alcohol and hepatitis C were the most common etiologies associated with PPHTN [6].

●In a series of 17,901 autopsied patients, pathologic changes consistent with PAH were found in 0.7 percent of patients with cirrhosis [2]. Although infrequent, it was more than five times the expected frequency.

Most studies report PPHTN equally in both men and women but some studies report that women with portal hypertension may be at greater risk of developing PPHTN [5,6].

PATHOGENESIS

Proposed mechanisms — The cause of PPHTN is unknown, although numerous theories have been proposed:

●Imbalance of vasoconstrictive and vasodilatory mediators – The most widely accepted hypothesis is that a humoral substance(s) (which would normally be metabolized by the liver) is able to reach the pulmonary circulation through portosystemic collaterals, resulting in PPHTN [18]. Candidates for this humoral substance include serotonin, interleukin-1, endothelin-1, glucagon, secretin, thromboxane B2, and vasoactive intestinal peptide [19-22]. Increased plasma concentrations of these mediators have been detected in patients with portal hypertension.

●Genetic predisposition – The observation that PPHTN is infrequent and sporadic among patients with portal hypertension is consistent with this proposal. Unlike hereditary pulmonary arterial hypertension (PAH), germline or other mutations in the bone morphogenetic protein receptor type II (BMPR2) gene have not been found in PPHTN [23-26]. However, several mutations have been proposed in pathways that involve estrogen signaling, cell growth, apoptosis, and oxidative stress [27].

●Thromboembolism from the portal venous system – Thromboembolism from the portal venous system also has been proposed as a cause of PPHTN. In this theory, thrombi from the portal circulation pass through portosystemic shunts and reach the pulmonary circulation, resulting in PAH [28,29]. Arguing against this proposal, a large autopsy study failed to show a significant number of thrombi simultaneously in the portal and pulmonary vascular beds [30], and some histopathologic features suggest that the pulmonary arterial thrombi sometimes observed in PPHTN usually arise in situ [31].

●Hyperdynamic pulmonary circulation – The hyperdynamic circulation seen in patients with chronic liver disease may also contribute to the development of PPHTN, with increased right ventricular output and blood flow through the pulmonary vascular bed causing increased shear stress on the vascular wall, possibly resulting in PAH. However, this hypothesis is not supported by data suggesting that increased blood flow is readily accommodated by the pulmonary vasculature [19].

●Inflammation – Other proposed mechanisms include increased local inflammation since many cytokines including interferon gamma are elevated in patients with cirrhosis [32], although data from humans with PPHTN are lacking.

Pathology — The pulmonary histopathology of PPHTN is indistinguishable from PAH due to other causes (picture 1 and table 1) [29,33]. These findings include medial hypertrophy, remodeling of the muscular pulmonary artery walls, and in situ thrombosis, the details of which are discussed separately. (See "The epidemiology and pathogenesis of pulmonary arterial hypertension (Group 1)", section on 'Pathology'.)

DIAGNOSTIC EVALUATION

Suspecting PAH

Clinical presentation — Since manifestations of portal hypertension typically precede those of pulmonary arterial hypertension (PAH), PPHTN should be suspected in patients with portal hypertension who present with symptoms and signs suggestive of PAH (eg, dyspnea on exertion, atypical chest pain, elevated jugular venous pressure, leg edema) [3,34]. The pulmonary manifestations of PPHTN are identical to other types of PAH [34], which are discussed in detail separately. (See "Clinical features and diagnosis of pulmonary hypertension of unclear etiology in adults", section on 'Clinical manifestations'.)

The interval between the first manifestation of portal hypertension and the documentation of PAH ranges from 2 to 15 years [18,34]. However, the threshold to suspect PPHTN should be low since patients with chronic liver disease often have dyspnea for many different reasons; in addition, some studies have reported high rates of asymptomatic patients (up to 60 percent) [3].

Portal hypertension is rarely discovered during the routine evaluation of PAH for underlying etiologies.

Early detection in chronic liver disease — We and others recommend that symptomatic patients with chronic liver disease and/or patients who are liver transplant candidates be evaluated for pulmonary hypertension (PH) with transthoracic echocardiography (TTE) [35]. In liver transplantation candidates, particularly those with hypoxemia, some experts use contrast echocardiography (also known as a "bubble" study) to distinguish patients with hypoxemia due to PH from those with hepatopulmonary syndrome (HPS). In HPS, contrast echocardiography would typically demonstrate an intrapulmonary shunt while in most cases of PPHTN, a shunt is not typically present. (See "Hepatopulmonary syndrome in adults: Prevalence, causes, clinical manifestations, and diagnosis".)

Although there is agreement on who should be evaluated with TTE, there is no consensus on what the criteria should be to prompt a right heart catheterization (RHC) for a definitive diagnosis of PPHTN [1,7,13]. Guidelines suggest that those with intermediate to high probability PH (ie, a peak tricuspid regurgitant velocity [TRV] >2.8 meters/second or those with a peak TRV ≤2.8 meters/second and other signs of PH) should undergo RHC. However, many experts perform RHC based upon several factors including symptoms, echocardiographic appearance, and liver transplantation candidacy with high threshold to perform RHC in those who do not need liver transplantation and a low threshold to perform RHC in those who are liver transplantation candidates. The role of echocardiography and RHC during the diagnostic evaluation of suspected PH is discussed in further detail separately. (See "Clinical features and diagnosis of pulmonary hypertension of unclear etiology in adults", section on 'Right heart catheterization' and "Clinical features and diagnosis of pulmonary hypertension of unclear etiology in adults", section on 'Echocardiography'.)

Differential diagnosis — The differential diagnosis of PPHTN is wide since dyspnea is the most common presenting symptom of PAH. Major etiologies of dyspnea in patients with chronic liver disease that should also be considered include the following:

●HPS – Patients with HPS may additionally have associated platypnea, orthodeoxia, and evidence of an intrapulmonary shunt on echocardiography that may help to distinguish it from PAH. (See "Hepatopulmonary syndrome in adults: Prevalence, causes, clinical manifestations, and diagnosis".)

●Heart failure with preserved or reduced ejection fraction – Patients with heart failure may have risk factors for cardiovascular disease together with manifestations of left-sided heart failure including bibasilar crackles, a gallop, and evidence of left ventricular dysfunction on TTE as well as evidence of an elevated pulmonary capillary wedge pressure on RHC. (See "Clinical manifestations and diagnosis of advanced heart failure".)

●Diaphragmatic compromise from ascites – Patients with severe chronic liver disease in whom no other cause is apparent may have dyspnea due to mechanical displacement of the diaphragm into the thoracic cavity from massive ascites. In such cases, dyspnea is often relieved by large volume paracentesis. (See "Diagnostic and therapeutic abdominal paracentesis" and "Evaluation of adults with ascites".)

●Deconditioning – Many patients with chronic liver disease are deconditioned. In such cases, deconditioning is often inferred after an extensive evaluation including chest computed tomography, pulmonary function testing, and TTE that have been unrevealing. Some experts additionally subject patients to cardiopulmonary exercise testing to facilitate the diagnosis of deconditioning. (See "Cardiopulmonary exercise testing in cardiovascular disease".)

●Other causes of dyspnea – Several comorbid conditions that cause chronic dyspnea (asthma, chronic obstructive pulmonary disease, interstitial lung disease) should be considered, the differential diagnosis of which is provided separately. (See "Approach to the patient with dyspnea".)

Diagnosis — The diagnosis of PPHTN requires confirmation of PAH by the exclusion of alternative causes of PAH and PH (table 1) as well as confirmation of portal hypertension.

Confirming PAH and excluding alternative causes of PAH and PH — The evaluation of suspected PAH in patients with portal hypertension is similar to that in all patients with suspected PAH, during which alternative causes of group 1 PAH and alternative types of PH (groups 2 through 5) are excluded (algorithm 1). Similar to patients with idiopathic PAH (IPAH), PAH is confirmed when RHC demonstrates:

●Elevated mean pulmonary artery pressure (mPAP) >20 mmHg at rest

●Normal or low pulmonary capillary wedge pressure ≤15 mmHg at rest

●An elevated pulmonary vascular resistance (PVR; ≥3 Wood units [240 dynes/sec/cm^-5])

An elevated PVR is important since it distinguishes patients with precapillary disease from those who have a passive elevation in the mPAP from hyperdynamic circulatory changes associated with chronic liver disease. Notably, a value of 3 Wood units (240 dynes/sec/cm^-5) rather than 2 Wood units (160 dynes/sec/cm^-5) is indicative of a population at high risk for liver transplantation. Patients with a PVR between 2 and 3 Wood units have a high risk of progression of their PVR and should be monitored closely [36].

Unlike patients with IPAH, if chronic liver disease is present, systemic vascular resistance (SVR) may be reduced and cardiac index elevated [37,38].

The diagnostic evaluation and criteria for patients with suspected PAH are discussed in detail separately. (See "Clinical features and diagnosis of pulmonary hypertension of unclear etiology in adults", section on 'Diagnosis'.)

Confirming portal hypertension — The diagnosis of portal hypertension is typically present and confirmed by reviewing the history, physical examination, imaging, or pathology that suggests chronic liver disease or portal hypertension. In most cases, the diagnosis is made clinically (eg, the patients has ascites and varices) but if necessary, it can also be confirmed by hepatic venous catheterization to measure the hepatic venous pressure gradient (HVPG); the HVPG is the gradient in pressure between the estimated portal vein (assessed by wedged hepatic vein measurements) and the free hepatic vein or inferior vena cava. Patients in whom HVPG should be measured include those in whom the diagnosis of portal hypertension is in doubt. For example, a patient with PAH who has positive liver serologies but little evidence of chronic liver disease may require HVPG measurement to confirm or refute the diagnosis. Details regarding the diagnosis of portal hypertension are provided separately. (See "Portal hypertension in adults", section on 'Diagnosis'.)

TREATMENT — Patients with PPHTN should be treated with general measures which target portal hypertension as well as the complications of pulmonary arterial hypertension (PAH; eg, right heart failure, thromboembolism). Patients may also be candidates for PAH-directed therapy which targets the PAH itself, based upon the World Health Organization (WHO) functional class (table 2). (See 'General measures' below and 'Pulmonary arterial hypertension-directed therapy' below.)

In patients with PPHTN, treatment is nearly identical to that for other types of PAH (algorithm 2). Most of the treatment options for PPHTN have been extrapolated from studies performed in patients with idiopathic PAH (IPAH) and limited studies in patients with PPHTN that show similar benefits. Data that are pertinent to treatment of PPHTN are discussed here while data that discuss treatment of IPAH are provided separately. (See "Treatment of pulmonary arterial hypertension (group 1) in adults: Pulmonary hypertension-specific therapy".)

Comanagement with both a pulmonary hypertension (PH) expert and a hepatologist is recommended.

General measures — General measures are PAH- and portal hypertension-related.

Pulmonary arterial hypertension-related — Similar to patients with group 1 PAH, all patients with PPHTN should exercise as tolerated, receive routine vaccinations (figure 1), be counseled against smoking (tobacco and marijuana) and pregnancy, and when indicated, be treated with supportive measures including oxygen and diuretics. (See "Standard immunizations for nonpregnant adults" and "Pneumococcal vaccination in adults" and "The benefits and risks of aerobic exercise" and "Pulmonary rehabilitation" and "Treatment and prognosis of pulmonary arterial hypertension in adults (group 1)", section on 'General measures and supportive therapy'.)

Evidence to support this approach in patients with PAH is discussed separately (see "Treatment and prognosis of pulmonary arterial hypertension in adults (group 1)", section on 'General measures and supportive therapy'). However, two types of general measures require special discussion in patients with PPHTN:

●Anticoagulation – We do not routinely anticoagulate patients with PPHTN, since there are no direct data that demonstrate benefit and indirect data in the PAH population are conflicting (see "Treatment and prognosis of pulmonary arterial hypertension in adults (group 1)", section on 'General measures and supportive therapy'). In addition, anticoagulation may increase the risk of bleeding in those with severe chronic liver disease, due to coexistent thrombocytopenia, coagulopathy, or esophageal varices at risk of bleeding [35]. However, on occasion anticoagulant is administered (eg, those with mild extrahepatic portal hypertension with low cardiac output from severe right heart failure). In such cases, there are no clear guidelines that help the clinician select an ideal anticoagulant or therapeutic target. Consequently, practice varies and while some experts choose warfarin with an international normalized ratio (INR) target ranging from 1.5 to 2.5, others choose a direct oral anticoagulant (DOAC) because monitoring the INR while on warfarin can be challenging in patients with hepatic impairment. Regardless, patients and physicians should be aware of the challenges involved in monitoring the therapeutic response and of the increased risk of bleeding or thrombosis under these circumstances. (See "Pathogenesis of variceal bleeding in patients with cirrhosis".)

●Diuretics – Fluid overload is frequently present in patients with PPHTN from right heart failure associated with PAH and from liver dysfunction, thus, diuretics are frequently required. The use of diuretics in patients with cirrhosis, including the diuretic regimen and concerns related to fluid removal and hypokalemia, is discussed in detail separately. (See "Ascites in adults with cirrhosis: Initial therapy", section on 'Diuretic therapy'.)

Portal hypertension-related — Patients with PPHTN should be treated for their portal hypertension, although there is no evidence that this is beneficial from a PAH perspective since therapy is targeted at treating complications of portal hypertension such as esophageal varices. The use of beta-blockers and/or transjugular intrahepatic portosystemic shunts (TIPS) may be harmful to patients with PPHTN [1,39]. (See "Portal hypertension in adults", section on 'Treatment' and "Overview of transjugular intrahepatic portosystemic shunts (TIPS)".)

Avoidance of beta-blockade — When feasible, beta blockade, typically used as prophylactic agents for the treatment of varices, is generally avoided in patients with PPHTN since they can worsen right heart failure due to a reduction in right ventricle cardiac output and increase in pulmonary vascular resistance (PVR) [39]. Cautious use in select circumstances such as severe cardiovascular disease should be individualized and take into consideration the risk of adverse effects on pulmonary hemodynamics. (See "Portal hypertension in adults", section on 'Treatment' and "Cirrhosis in adults: Overview of complications, general management, and prognosis", section on 'General management' and "Primary prevention of bleeding from esophageal varices in patients with cirrhosis", section on 'General measures'.)

Avoidance of transjugular intrahepatic portosystemic shunt — TIPS can increase right ventricle preload and worsen heart failure and is consequently generally avoided in those with PPHTN [40]. Cautious use in select circumstances (eg, life-threatening bleed) should be individualized and take into consideration the risk of adverse effects on pulmonary hemodynamics, particularly in those with severe PAH. (See "Prevention of recurrent bleeding from esophageal varices in patients with cirrhosis", section on 'Options if initial strategy fails'.)

When feasible, esophageal varices should be treated with banding. (See "Primary prevention of bleeding from esophageal varices in patients with cirrhosis", section on 'Endoscopic variceal ligation'.)

Balloon-occluded retrograde transvenous obliteration (BRTO) is a newer therapy for the treatment of acute bleeding from gastric varices that is being increasingly used. It can increase portal pressures but the effect on PAH is unknown. (See "Methods to achieve hemostasis in patients with acute variceal hemorrhage", section on 'Alternative treatments'.)

Liver transplantation — In the past, PPHTN was considered a contraindication to liver transplantation due to the high perioperative mortality associated with PAH. However, successful transplantation is feasible in patients with treated PPHTN which has led to strict criteria regarding candidacy for liver transplantation in this population [6,41-49]. While pulmonary pressures may improve or normalize following liver transplantation, liver transplantation is not a treatment for PPHTN per se and can only be pursued in patients with end-stage liver disease who meet liver transplantation criteria, provided their PPHTN is treated and responsive to PAH-specific therapy. Outcomes due to liver transplantation and PAH-specific therapy have never been compared.

●Assessment pre-liver transplant – Every patient who is assessed for liver transplantation should undergo clinical and transthoracic echocardiography (TTE) evaluation for pulmonary hypertension. (See 'Early detection in chronic liver disease' above and "Liver transplantation in adults: Patient selection and pretransplantation evaluation".)

Findings suggestive of pulmonary hypertension by TTE (right ventricular [RV] enlargement, septal flattening or paradoxical motion, tricuspid regurgitation jet estimates of pulmonary artery systolic pressure >50 mmHg) should prompt right heart catheterization (RHC) to more fully characterize the presence and severity of pulmonary vascular disease. Since cardiac output is typically high in PPHTN, even a "normal" cardiac output, particularly when associated with a high right atrial pressure, a modestly increased mean pulmonary artery pressure (mPAP), and an elevated pulmonary vascular resistance, should raise concern regarding the risks of transplantation.

In general, patients with an mPAP >50 mmHg cannot undergo liver transplantation. Guidelines recommend treatment of PPHTN pre liver-transplant with PAH-specific therapy in those with an mPAP >35 mmHg [50] with the aim of reducing pulmonary pressures to an mPAP <35 mmHg and improving PVR to ≤2 Wood units (160 dynes/sec/cm^-5), and right ventricle function. In one study, almost 50 percent of patients with PPHTN became eligible for liver transplantation with PAH-specific therapy [51].

The Model for End-stage Liver Disease (MELD) scoring system is used to predict survival among patients with chronic liver disease, with higher scores correlating with lower survival. However, waitlist mortality in patients with PPHTN is determined by the severity of both liver and pulmonary dysfunction [52]. Thus, in the United States, PPHTN in those with chronic liver disease is a MELD exception [53,54]; patients with PPHTN have their MELD score upgraded 10 percent every three months while they are on the liver transplantation waiting list; otherwise, the MELD score will underestimate their mortality risk. This exception, in effect, reduces waitlist time and mortality. (See "Model for End-stage Liver Disease (MELD)", section on 'Other standard MELD exceptions'.)

●Perioperative course – PAH in itself is a surgical risk and in patients undergoing liver transplantation the risk may be greater due to hemodynamic changes induced by specific procedures such as clamping the inferior vena cava and reperfusion of the transplanted organ. Intraoperative hemodynamic monitoring with right heart catheter is therefore generally advised, not for treatment of the pulmonary hypertension but to better monitor the large volume shifts that occur during liver transplantation that can stress right heart function. Rarely, extracorporeal membrane oxygenation may be used intraoperatively or postoperatively for patients who decompensate from an oxygenation standpoint [55].

In the immediate postoperative period and for up to six months after liver transplantation, a proportion of patients (up to 20 percent) can develop worsening of their PPHTN and severe right heart failure that may require escalation of PAH-specific therapy [49,56,57]. During this period, new cases of PAH have also been anecdotally reported.

●Outcomes ─ After six months, many patients improve, stabilize, or normalize their pulmonary pressures with or without PAH-specific therapies [47,51,57,58]. In one study of 23 patients with PPHTN who underwent liver transplantation, pulmonary arterial pressure normalized in 61 percent of patients [49]. Use of combination PAH-specific therapy was the only factor that predicted normalization of pulmonary pressures. In another study, half of patients were weaned off their PAH-specific therapy [51].

Limited data suggest improved survival in patients treated with liver transplantation. As an example, one meta-analysis of 26 mostly observational studies (1019 patients), the risk of death was higher in patients who could only be treated with PAH-specific therapy compared with patients who could be transplanted as well (odds ratio 3.5, 95% CI 1.4-8.8) [51]. However, patients who proceeded to transplant had been selected very strictly and may have contributed favorably to survival in this analysis.

Limited data also suggest that survival following liver transplantation in patients with PPHTN may be worse than those without PPHTN. In particular, severe PAH (systolic PAP >60 mmHg) is associated with high perioperative risk and poor clinical outcome, while mild to moderate PAH may not influence mortality after liver transplantation [1,9-12,46,48,57,59]. As examples:

•A retrospective study of 1205 patients who underwent liver transplantation compared those with PAH to those without PAH [9]. The three-year mortality incrementally increased with PAH severity; patients without PAH had a three-year mortality of 28 percent, compared with those who had mild PAH (systolic PAP 30 to 44 mmHg; 33 percent), moderate PAH (systolic PAP 45 to 59 mmHg; 35 percent) and severe PAH (systolic PAP >60 mmHg; 71 percent).

•In a database review, overall mortality was 36 percent in those with PPHTN, but increased to 100 percent if the mPAP was >50 mmHg [46].

•In a meta-analysis of 12 mostly retrospective studies, following liver transplantation, patients with PPHTN had an increased one-year mortality (odds ratio 1.59) compared with those who underwent liver transplantation and did not have PPHTN [59]. The rate of graft loss was similar.

•Another single center study reported a one-year survival of 69 percent, lower than those without pulmonary hypertension who had survival rates >94 percent [48].

One study suggested that high pulmonary vascular resistance pre-liver transplant (eg, ≥1.6 Wood units) may portend a higher risk of mortality and graft failure [60].

Pulmonary arterial hypertension-directed therapy — PAH-specific therapy is directed at the PAH itself rather than the underlying cause. Agent selection is determined by many factors, particularly WHO functional class (table 2).

WHO functional class I (observation) — WHO functional class I patients (table 2) do not require pharmacologic therapy; however, they should be clinically monitored (eg, every three to six months) for disease progression to a functional level that may warrant therapy. Any coexisting conditions that worsen pulmonary hypertension should also be treated (eg, cardiac disease, interstitial lung disease).

WHO functional class II to IV (pharmacotherapy) — Similar to patients with IPAH, patients with WHO functional class II, III, or IV (table 2) should be referred to a specialized center to be evaluated for PAH-specific pharmacotherapy. Comanagement with both a PH expert and a hepatologist is recommended. The evidence supporting PAH-specific pharmacotherapy primarily comes from studies that included mostly patients with IPAH with only a small proportion of or no patients with PPHTN included in the analyses (see "Treatment of pulmonary arterial hypertension (group 1) in adults: Pulmonary hypertension-specific therapy", section on 'Definition'). Evidence to support similar benefits in patients with PPHTN is limited and is the focus of this section of the topic review.

Agent selection — There are several classes of agents including prostacyclin pathway agonists, endothelin receptor antagonists, and nitric oxide (NO)-cyclic guanosine monophosphate (GMP) enhancers (phosphodiesterase inhibitors and guanylate cyclase stimulants) (table 3).

In general, the principles of agent selection are similar to that in the IPAH population with special considerations in PPHTN patients:

●Calcium channel blockers (CCBs) should be avoided. (See 'Avoidance of calcium channel blockers' below.)

●Endothelin receptor antagonists (ERAs), particularly bosentan, can be associated with liver toxicity (transaminitis, liver failure, cirrhosis). ERAs should be avoided in those whose transaminase level is greater than three times the upper limit of normal and in those with moderate to severe liver disease. Patients with PPHTN should have liver function monitored monthly during ERA therapy. ERAs may also place these patients at risk of fluid retention necessitating an adjustment in diuretic therapy. Ambrisentan and macitentan are less hepatotoxic than bosentan but patients develop more fluid retention with ambrisentan. (See 'Endothelin receptor antagonists' below.)

●Phosphodiesterase-5 inhibitors (PDE5Is), sildenafil and tadalafil, are commonly prescribed since their metabolism is not affected by liver dysfunction. While some data suggest that PDE5Is may increase splanchnic flow to worsen portal hypertension, data are conflicting and there is little clinical evidence to support this adverse effect [61,62].

●In general, most experts start with single agent therapy rather than up front dual agent therapy and have a slower than usual approach to escalation of therapy given the potential for altered metabolism due to liver dysfunction.

A shared decision by the patient and PAH and liver experts that weighs the risks of adverse effects, drug interactions, and benefits of a specific therapy is critical. Specific drug interactions and management suggestions may be determined by using the drug interactions program included with UpToDate.

Further details regarding agent selection in patients with IPAH are discussed separately. (See "Treatment of pulmonary arterial hypertension (group 1) in adults: Pulmonary hypertension-specific therapy", section on 'WHO functional class II and III or low/intermediate risk (combination oral therapy)' and "Treatment of pulmonary arterial hypertension (group 1) in adults: Pulmonary hypertension-specific therapy", section on 'WHO functional class IV or high risk (parenteral prostanoid-containing combination regimen)' and "Treatment of pulmonary arterial hypertension (group 1) in adults: Pulmonary hypertension-specific therapy".)

Avoidance of calcium channel blockers — In contrast to patients with IPAH who have positive vasoreactivity testing, patients with PPHTN are not typically treated with CCBs. This is because a vasoactive response only occurs in a small proportion of PPHTN patients (approximately 1.7 percent) [63]. In addition, intolerable side effects including hypotension and splanchnic vasodilation can develop in this population and result in an increase in the hepatic venous pressure gradient. Patients with PPHTN are particularly susceptible to hypotension because systemic vascular resistance (SVR) is usually low in chronic liver disease such that the administration of a CCB, a potent systemic vasodilator, can further reduce SVR and result in reduced RV filling causing life-threatening RV failure. Since CCBs are not typically prescribed, patients with PPHTN do not need to undergo routine vasoreactivity testing. (See "Treatment of pulmonary arterial hypertension (group 1) in adults: Pulmonary hypertension-specific therapy", section on 'Vasoreactive patients'.)

Agent options — PAH-specific therapy refers to the administration of agents with complex mechanisms of action including vasodilation, vascular growth inhibition, and vascular remodeling [64-66]. In general, patients are candidates for PAH-specific therapy if they remain in WHO functional class II, III, or IV despite optimization of the underlying cause of their portal hypertension (table 2). In patients with PPHTN, treatment is nearly identical to that for IPAH (algorithm 2) with the exception that CCBs are avoided and caution is advised with ERAs due to liver toxicity. Agent selection for PPHTN has been mostly extrapolated from studies performed in patients with IPAH and limited studies in patients with PPHTN that show similar benefits. Data specific to patients with PPHTN are discussed here while data that discuss agent selection in IPAH are provided separately. (See "Treatment of pulmonary arterial hypertension (group 1) in adults: Pulmonary hypertension-specific therapy", section on 'Definition' and "Treatment of pulmonary arterial hypertension (group 1) in adults: Pulmonary hypertension-specific therapy", section on 'Nonvasoreactive patients'.)

Combination therapy — Combining agents from two different classes is commonly used in patients with idiopathic PAH but has not been specifically studied in patients with PPHTN, except for occasional case reports or series [67-69]. Nonetheless, the principles of agent selection are similar such that most patients with PPHTN and WHO functional class II, III, and IV are treated with combination therapy, the details of which are discussed separately.

Ambrisentan and tadalafil is the preferred combination for patients with WHO functional class II and III since it has been shown to result in a 50 percent reduction in the rate of clinical failure [70]. However, patients with PPHTN were not included in that trial. In addition, if liver toxicity is a concern with ambrisentan use, then a PDE5 inhibitor plus the oral prostanoid, selexipag, may be an alternative.

Prostacyclin pathway agonists — Limited data in patients with PPHTN have shown improvement in pulmonary hemodynamics, functional class, exercise capacity, and eligibility for liver transplant with several prostacyclin pathway agonists. With the exception of oral selexipag, which may be administered to patients with WHO functional class II and III, these agents (epoprostenol, treprostinil, iloprost) are often reserved for patients with WHO functional class IV and those who need a bridge to liver transplantation.

●Epoprostenol – Several retrospective studies have reported hemodynamic benefit with epoprostenol, a potent vasodilator with anti-platelet aggregating and antiproliferative properties in patients with PPHTN [41,42,71-74]. As examples:

•A case series of seven patients with PPHTN who were treated with continuous intravenous epoprostenol demonstrated that the WHO functional class improved in all patients one year after the initiation of therapy [71]. There were also significant improvements in exercise duration, mPAP, and PVR, compared with baseline.

•In an observational study, 14 patients with advanced liver disease and moderate to severe pulmonary hypertension (defined as an mPAP >35 mmHg) received epoprostenol [72]. Epoprostenol was associated with a reduction of mPAP, a reduction of PVR, and an increase in cardiac output (CO). Ten patients were maintained on a continuous intravenous infusion and six of them underwent repeat RHC that confirmed long-term reduction of PVR. Six of the 10 patients who received continuous epoprostenol died during the study, raising the possibility that the therapy could have adverse effects on hepatic function, portal hypertension, or other aspects of chronic liver disease, although it is more likely that these patients were among the sickest in the cohort.

•In a single center prospective study of 21 patients with PPHTN, early initiation of parenteral epoprostenol therapy was associated with improved five-year patient survival as compared with PPHTN patients from the Registry to Evaluate Early and Long-term Pulmonary Arterial Hypertension Disease Management (the REVEAL Registry). Over one-half of patients on epoprostenol were eligible for transplant within one year.

●Treprostinil – In a retrospective case series of 11 patients treated predominantly with subcutaneous treprostinil (in combination with oral sildenafil), the mPAP, PVR, CO, and the transpulmonary gradient all improved [67]. All 11 patients subsequently received liver transplants. Following transplantation, about two-thirds were weaned off PAH-specific medications, and only two patients required transient increased doses of prostacyclins.

●Iloprost – Case reports describe successful use of inhaled or intravenous iloprost in patients with PPHTN [75,76]. In addition, a retrospective cohort study of 13 patients who had severe PPHTN treated with inhaled iloprost for up to three years found one-, two-, and three-year survival rates of 77, 62, and 46 percent, respectively [77]. These survival rates are better than historical controls, but poorer than those seen with bosentan therapy [34,77]. (See 'Endothelin receptor antagonists' below.)

●Selexipag – This oral prostacyclin agonist agent has not been specifically tested in patients with PPHTN.

Adverse effects are similar to those with IPAH but hypersplenism has been reported in rare patients with PPHTN [78]. Some of these agents cause thrombocytopenia, and may worsen thrombocytopenia from chronic liver disease; thus, monitoring is advised. It is unclear whether the risk of adverse outcomes during prostacyclin pathway agonist therapy is greater in patients with PPHTN than IPAH. Specific drug interactions and management suggestions may be determined by using the drug interactions program included with UpToDate. The administration of prostacyclin pathway agonist therapy is discussed in detail elsewhere. (See "Treatment of pulmonary arterial hypertension (group 1) in adults: Pulmonary hypertension-specific therapy".)

Endothelin receptor antagonists — Bosentan and macitentan are oral agents that are dual endothelin-A and endothelin-B receptor antagonists, while ambrisentan is an oral selective endothelin-A receptor antagonist. ERAs are not typically used for patients with WHO functional class IV unless in combination with a parenteral prostanoid but can be used in patients with WHO functional class II and III alone or in combination with agents from a different class, typically a PDE5 inhibitor. ERAs have been shown in small series to be beneficial for patients with PPHTN [77,79-87]:

●Bosentan

•Thirty-four consecutive patients with PPHTN were treated with bosentan for a mean of 43 months [87]. Significant improvements were noted in WHO functional class, six-minute walk distance, and pulmonary hemodynamics. Plasma concentrations of bosentan were higher in these patients than in a cohort with IPAH. Four patients died during follow-up and seven had significant elevation in liver enzymes.

•A retrospective cohort study of 18 patients who had severe PPHTN treated with bosentan for up to three years found one-, two-, and three-year survival rates of 94, 89, and 89 percent, respectively [77]. The survival rates were better than those observed in patients treated with inhaled iloprost.

•An observational study of 11 patients with severe PPHTN who were treated with bosentan for one year found that the six-minute walking distance increased and the PVR decreased, compared with baseline [85].

●Ambrisentan – An observational study followed 13 patients with PPHTN who were treated with ambrisentan for a median of 613 days [86]. The mPAP decreased from a median of 58 to 41 mmHg, while the PVR decreased from a median of 445 dynes/sec/cm^-5 to 174 dynes/sec/cm^-5 (5.6 to 2.2 Wood units). Liver function testing was unchanged after 12 months of therapy.

●Macitentan – In a multicenter randomized trial of 85 patients with PPHTN, two-thirds of whom were already receiving therapy for pulmonary hypertension, at 12 weeks, macitentan improved PVR by 35 percent compared with placebo [88]. Adverse events, particularly peripheral edema, were more common in the macitentan group.

Patients with PPHTN should have liver function tested before therapy and monthly during ERA therapy, since ERAs have been associated with hepatotoxicity in unselected patients with IPAH. There are few pharmacokinetic data in patients with liver disease [87]. In general, initiation of ERAs is not advised for patients whose transaminases are greater than three times the upper limit of normal and for those with moderate to severe liver function impairment. Although dose adjustments can be made in those who demonstrate worsening liver function on ERAs, switching to another agent is preferred. Further data are provided in the drug interactions program included with UpToDate for bosentan, macitentan, and ambrisentan.

The administration and adverse effects of ERAs are discussed elsewhere. (See "Treatment of pulmonary arterial hypertension (group 1) in adults: Pulmonary hypertension-specific therapy".)

Nitric oxide-cyclic guanosine monophosphate enhancers — These agents include two families of drugs, the guanylate cyclase stimulant, riociguat, and the PDE5Is, sildenafil and tadalafil. These are not typically used for patients with WHO functional class IV unless in combination with a parenteral prostanoid but can be used in patients with WHO functional class II and III alone or in combination with agents from a different class, typically an ERA. Guanylate cyclase stimulants and PDE5Is cannot be combined since they belong to the same class of agents. PDE5Is are commonly used since they are well tolerated in patients with liver disease. Limited data in patients with PPHTN describe some benefit with these agents.

●Guanylate cyclase stimulant – Riociguat is a guanylate cyclase stimulant of proven benefit in patients with chronic thromboembolic pulmonary hypertension (CTEPH). It has not been directly studied in patients with PPHTN. However, in a post-hoc analysis of two major trials that studied riociguat in patients with CTEPH and PAH, among the 13 patients who had PPHTN, riociguat was well-tolerated and improved the six-minute walking distance and WHO functional class for up to two years [89].

Per the manufacturers labeling, dose-adjustment is not needed for those with mild to moderate liver disease but riociguat administration is not advised in those with severe liver dysfunction. Consequently, many experts do not use riociguat as a first line agent in PPHTN.

Major trials demonstrating benefit in patients with CTEPH and PAH are discussed separately. (See "Chronic thromboembolic pulmonary hypertension: Pulmonary hypertension-specific therapy", section on 'Medication selection' and "Treatment of pulmonary arterial hypertension (group 1) in adults: Pulmonary hypertension-specific therapy".)

●Phosphodiesterase-5 inhibitors

•Sildenafil – Case reports and case series describe the successful use of sildenafil to treat PPHTN [68,90,91]. One observational study followed 14 patients with moderate to severe PPHTN who were administered sildenafil (50 mg, three times per day) [68]. In eight patients, sildenafil was the first therapy, while sildenafil was added to pre-existing inhaled prostanoid therapy in the remaining six patients. Among the 12 patients who completed the study, there was a reduction of mPAP and PVR at three months, but not at one year, compared with baseline. Six-minute walking distance was improved at both three months and one year. There was no difference whether sildenafil was used as monotherapy or in combination therapy.

•Tadalafil – Tadalafil has not been studied in patients with PPHTN. In our experience, when tadalafil (usually in combination with ambrisentan) is indicated for patients with WHO class II/III PAH, monitoring for hepatotoxicity is indicated for the ambrisentan component of this combination.

•Vardenafil – Vardenafil, although approved for erectile dysfunction is not approved for pulmonary hypertension.

Hepatic toxicity does not appear to be an issue with PDE5Is.

There are conflicting data on whether PDE5 inhibitors alter splanchnic flow and worsen portal hypertension [61,62]. However, there is no evidence to support this adverse effect clinically. Specific drug interactions and management suggestions may be determined by using the drug interactions program included with UpToDate. Further data supporting the efficacy of PDE5 inhibitors in patients with IPAH are discussed separately. (See "Treatment of pulmonary arterial hypertension (group 1) in adults: Pulmonary hypertension-specific therapy".)

Refractory disease

Right to left shunts and lung transplantation — Shunts are appropriate (eg, atrial septostomy) as a bridge to lung transplantation in those with severe symptomatic PPHTN that is refractory to combination therapy. These options are discussed separately. (See "Treatment of pulmonary arterial hypertension (group 1) in adults: Pulmonary hypertension-specific therapy", section on 'Right-to-left shunt' and "Treatment of pulmonary arterial hypertension (group 1) in adults: Pulmonary hypertension-specific therapy", section on 'Lung transplantation'.)

PROGNOSIS — Patients with PPHTN may have worse outcomes than patients with other types of group 1 pulmonary arterial hypertension (PAH) (table 1) [6,15]. This was suggested by a retrospective cohort study of 174 patients that was performed using data from the Registry to Evaluate Early and Long-term Pulmonary Arterial Hypertension Disease Management (the REVEAL Registry) [15]. Compared with idiopathic and familial PAH, patients with PPHTN had a lower two-year survival rate (67 versus 85 percent) and five-year survival rate (40 versus 64 percent), despite having better hemodynamic measurements and functional class assignments at the time of diagnosis. Similarly, in the UK, one-, three-, and five-year survival rates were 85, 60, and 35 percent, respectively, lower than those reported for patients with idiopathic PAH [6].

It is possible that the poorer outcomes reflect the observation that some of these data were collected in an era where PAH-specific therapy was delayed or not administered in PPHTN patients [15]. For example, in a 2008 series of 154 patients with PPHTN, whose survival rates were 88 percent at one year, 75 percent at three years, and 68 percent at five years, only one-third of patients received PAH-specific therapy [17]. However, data since then do not support that theory. In a US cohort, 85 percent of whom received PAH-specific therapy (typically monotherapy), the overall survival was 89 percent at 6 months, 77 percent at one year, 51 percent at three years, and 38 percent at five years [92]; analysis of patients diagnosed from 1988 to 2007 compared with 2008 to 2019 did not reveal significant differences in mortality.

The effect of PAH-directed therapy on survival is poorly studied. One older study reported a six-month survival rate of 50 percent in patients who did not receive PAH-specific therapy [34]. Newer studies done since then are conflicting with one retrospective study reporting improved survival on PAH-specific therapy [93] while another showed no difference [94]. Another meta-analysis of 26 observational studies reported a one- and three-year survival in patients treated with PAH-specific therapy as 86 and 69 percent, respectively [51].

High pulmonary vascular resistance (PVR) is a risk factor for mortality [50]. Consequently, lowering of PVR with PAH-specific therapies is a reasonable goal, and one that improves candidacy for liver transplant. However, data to support improved survival with this strategy are not reported. The effect of liver transplantation on outcome is discussed above. (See 'Liver transplantation' above.)

Other risk factors for death include severity of liver disease (eg, MELD-sodium score >15), poor six-minute walk distance, high right atrial pressure, and high pulmonary capillary wedge pressure [92].

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Pulmonary hypertension in adults".)

SUMMARY AND RECOMMENDATIONS

●Portopulmonary hypertension (PPHTN) refers to pulmonary arterial hypertension (PAH) that is associated with portal hypertension due to chronic liver disease or portal hypertension from extrahepatic causes. PPHTN belongs to group 1 PAH (table 1). (See 'Definition' above.)

●PPHTN occurs in about 2 to 16 percent of patients with portal hypertension, although rates vary and are not affected by the severity of underlying portal hypertension or liver disease. Rates appear to be equal among men and women. (See 'Epidemiology' above.)

●The cause of PPHTN is unknown, although numerous theories have been proposed among which an imbalance of vasoconstrictive and vasodilatory mediators and a genetic predisposition to PPHTN are the most plausible. The pulmonary histopathology of PPHTN is indistinguishable from PAH due to other causes (picture 1). (See 'Pathogenesis' above.)

●Most patients with PPHTN present with clinical evidence of portal hypertension that precedes the development of pulmonary symptoms due to PAH (eg, dyspnea on exertion, atypical chest pain, raised jugular venous pressure, leg edema). We and others recommend that symptomatic patients with chronic liver disease and patients who are liver transplantation candidates be evaluated for pulmonary hypertension with transthoracic echocardiography (TTE) with a view to performing a diagnostic right heart catheterization (RHC) in those with an intermediate to high risk of PH. (See 'Suspecting PAH' above and "Clinical features and diagnosis of pulmonary hypertension of unclear etiology in adults", section on 'Intermediate or high probability of pulmonary hypertension on echocardiography (assess left heart)'.)

●Major etiologies of dyspnea in patients with chronic liver disease that should be considered during the diagnostic evaluation include hepatopulmonary syndrome, heart failure, diaphragmatic compromise from ascites, deconditioning, and other major causes of chronic dyspnea (eg, asthma, chronic obstructive pulmonary disease, interstitial lung disease), most of which can be excluded clinically or by TTE and RHC findings. The diagnosis of PPHTN requires confirmation of PAH by the exclusion of alternative causes of PAH and pulmonary hypertension (PH) (table 1) as well as confirmation of portal hypertension (algorithm 1). In most cases, the diagnosis of portal hypertension is confirmed clinically but if necessary, it can also be confirmed by hepatic venous catheterization to measure the hepatic venous pressure gradient (HVPG). (See 'Diagnostic evaluation' above.)

●In patients with PPHTN, treatment with general measures and PAH-specific therapies (prostacyclin pathway agonists, endothelin receptor antagonists, and nitric oxide-cyclic guanosine monophosphate enhancers [phosphodiesterase inhibitors and guanylate cyclase stimulants] (table 3)) is almost identical to that for other types of PAH (algorithm 2). Most of the treatment options for PPHTN have been extrapolated from studies performed in patients with idiopathic PAH (IPAH) and limited studies in patients with PPHTN that show similar benefits. (See "Treatment of pulmonary arterial hypertension (group 1) in adults: Pulmonary hypertension-specific therapy".)

●Aspects of treatment that are unique to patients with PPHTN include the following:

•We do not routinely anticoagulate patients with PPHTN, since there are no direct data in PPHTN patients that report benefit, indirect data from other PAH populations are conflicting, and many patients are at increased risk of bleeding from the sequelae of underlying chronic liver disease. (See 'Pulmonary arterial hypertension-related' above.)

•Patients with PPHTN should be treated for their portal hypertension, although there is no evidence that this is beneficial from a PAH perspective. However, beta-blockers and transjugular intrahepatic portosystemic shunts are potentially harmful and, therefore, typically avoided in patients with PPHTN. While pulmonary pressures may improve or normalize following liver transplantation, liver transplantation is not a treatment for PPHTN and can only be pursued in patients with end-stage liver disease who meet liver transplantation criteria, provided their PPHTN is treated and responsive to PAH-specific therapy. (See 'Portal hypertension-related' above.)

•Most patients with PPHTN do not need to undergo vasoreactivity testing since calcium channel blocker therapy is rarely of benefit and may be associated with an increased risk of adverse effects when compared with IPAH. (See 'Avoidance of calcium channel blockers' above.)

•Endothelin receptor antagonists (ERAs) can be associated with liver toxicity, particularly, bosentan. ERAs should be avoided in those whose transaminase level is greater than three times the upper limit of normal and in those with portal hypertension from moderate to severe liver disease. Patients with PPHTN should have liver function monitored monthly during ERA therapy. Phosphodiesterase-5 inhibitors (PDE5Is) may worsen portal hypertension but clinical evidence to support that potential adverse effect is poor. Specific drug interactions and management suggestions may be determined by using the drug interactions program included with UpToDate.

●Patients with PPHTN may have worse outcomes than patients with other types of group 1 PAH (table 1). The effect of PAH-specific therapy on survival is unclear. For those who undergo liver transplantation for end-stage liver disease, a small proportion (up to 20 percent) may experience deterioration of their PAH in the immediate postoperative setting and for six months following transplantation that necessitates escalation of PAH-specific therapy. However, after six months, up to two-thirds of patients experience improved or normalization of pulmonary pressures. (See 'Prognosis' above and 'Liver transplantation' above.)