Portopulmonary Hypertension: A Review

Portopulmonary hypertension (POPH) occurs when pulmonary arterial hypertension (PAH) develops in the setting of portal hypertension.¹  The hemodynamic definition of PAH was modified at the 6th World Symposium for Pulmonary Hypertension in 2018 in Nice, now defined as a mean pulmonary arterial pressure (mPAP) > 20 mm Hg (previously > 25 mm Hg), normal left heart filling pressure, and a pulmonary vascular resistance (PVR) of > 3 Wood units.²  The subclassifications of Group I PAH all fall under this new definition. However, specific data regarding the implications of the changed diagnostic criteria and response to treatment for many of these subgroups are lacking (Table 1).³  POPH is included in the Group 1 classification of PAH, which has important implications for treatment (see the section on “Management,” below).³  Since many patients with cirrhosis also have volume overload and the wedge pressure can be > 15 mm Hg while still having concomitant small vessel arteriopathy, some experts suggest using a transpulmonary gradient (mPAP − wedge pressure) > 12 mm Hg in the diagnosis of POPH to adjust for the confounder of intravascular volume expansion.⁴ 

In the United States and other developed countries, POPH occurs most often in the setting of chronic liver disease leading to cirrhosis.⁵  Interestingly, the development of POPH may not correlate with the severity of the patient’s liver disease. It is estimated that 4% to 8% of patients with liver disease have POPH, but this number may be an underestimate.^6–8  In some studies, up to 15% of patients with portal hypertension have pulmonary hypertension (PH).⁹  Data regarding the incidence of POPH are taken chiefly from liver transplant centers that employ routine screening via echocardiography.¹⁰ 

In the developing world, noncirrhotic POPH is more common and is often caused by infection with Schistosoma mansomi leading to extrahepatic portal hypertension. This infection is suspected to be the most common form of POPH in the world. Schistosoma first enters the skin, then travels to the lungs, where it produces an immune complex hypersensitivity reaction. Portal hypertension results from the mechanical obstruction of the presinusoidal vessels by ova; when this occurs, the ova can bypass the liver and travel to the lungs, causing a granulomatous inflammatory reaction. Pulmonary vascular remodeling is suspected to result from both mechanisms: direct inflammation and portal hypertension.¹¹ 

The specific pathology of POPH is unknown. Patients with POPH have been found to have higher levels of endothelin-1. The liver is a major site of the synthesis, clearance, and action of endothelin-1, and one theory is that in portal hypertension, more endothelin-1reaches the pulmonary circulation, resulting in vasoconstriction by binding with the endothelin A receptor.¹²  There may also be an autoimmune component to the disease since the female sex and the presence of autoimmune hepatitis seem to be particular risk factors.¹³  Recent research has suggested that estrogen signaling may play a role as well. A case-control study showed that a single-nucleotide polymorphism in aromatase rs7175922 was associated with higher levels of estradiol and increased odds of developing POPH (odds ratio: 2.38).¹⁴ 

The initial presentation of POPH is similar to other etiologies of PAH with fatigue and dyspnea. This eventually progresses to the more apparent signs of right ventricular failure with edema (both lower extremity and abdominal), elevated jugular venous pressure, lightheadedness, and syncope.¹⁵  While the insidious development of PAH is difficult to diagnose in all comers, it is particularly difficult in patients with liver disease as many of the early non-specific, and even late symptoms can be attributed to their underlying liver disease. Screening tests for POPH mirror screening in other patient populations and includes a transthoracic echocardiogram.¹⁶  Signs of right heart dysfunction or elevated estimated PA systolic pressures > 38 mm in one registry were used as positive screening tools, followed by a right heart catheterization (RHC). In other institutions, a right ventricular systolic pressure of 50 mm Hg is used due to the other confounding hemodynamic abnormalities common in the cirrhotic patient.¹⁷  The hemodynamic derangements in cirrhotic patients are complex and may present with a few different patterns, emphasizing the need for a careful hemodynamic evaluation. In cirrhotic patients without POPH, the most common hemodynamic pattern is a high cardiac output, high flow state with low systemic vascular resistance, low PVR, and low pulmonary capillary wedge pressure. This occurs in 30% to 50% of patients with cirrhosis. In patients with more advanced liver disease and volume overload, the hemodynamics change but are still characterized by high cardiac output, high circulating pulmonary pressures, and a high pulmonary capillary wedge pressure. However, the calculated PVR is still in the range of normal due to the increased cardiac output.¹⁸ 

In POPH, the hemodynamics are characterized by high pulmonary pressures with a normal pulmonary capillary wedge pressure and high, normal, or even low cardiac output; the cardiac output will vary due to the degree of hemodynamic derangement due to underlying liver disease and the degree of right heart failure. This results in a high PVR (Figure 1). Notably, a low cardiac output in patients with POPH portends a very poor prognosis since cirrhosis is typically associated with a high-output state.^12,19 

Patients with cirrhosis can also have other risk factors for PAH. Some risk factors for developing cirrhosis overlap with risk factors for developing PAH. For example, intravenous drug abuse increases the risk of acquiring both hepatitis C and human immunodeficiency virus; methamphetamine use, in particular, is associated with the development of Group 1 PAH as well.²  Therefore, to firmly establish the diagnosis of POPH, there needs to be objective evidence of elevated portal pressures. Often, this is clinically evident by history, laboratory evaluation, and physical exam (ascites, presence of varices, etc.). If the diagnosis is in question, it can be confirmed with hemodynamic measurements. Portal hypertension can be diagnosed by measuring the difference between the hepatic vein wedge and free pressure; wedged hepatic venous pressure is an estimate of pressure within the portal venous system, whereas free hepatic venous pressure reflects systemic venous pressure. A difference of > 5 mm Hg is consistent with portal hypertension.²⁰  This pressure gradient can be measured during RHC and is easier to do when the femoral vein is accessed.

Compared with other forms of Group 1 PAH, patients with POPH have a worse prognosis. Decreased survival and increased hospitalization rates were seen compared to patients with idiopathic PAH in the REVEAL registry, which is the largest data set available for POPH. Patients with POPH were 3 times more likely to die despite having “better” hemodynamic parameters.⁶  In a paper out of the French registry, outcomes were reported in 154 POPH patients, both treated and untreated. There was a 68% 5-year survival. Child B and C cirrhosis and lower cardiac index were associated with worse outcomes.¹⁹ 

Similarly, in the chronic liver disease population, outcomes for patients with POPH in the moderate to severe range (mPAP > 45 mm Hg) are worse compared with matched controls with the same model for end-stage liver disease (MELD) score.⁴ 

A retrospective screening RHC analysis of 74 patients with POPH seen at the Mayo Clinic demonstrated a 14% 5-year survival in patients that received no PH-directed treatment, a 45% survival in those who received vasodilator therapy, and a 67% survival if they received pretreatment for PH followed by orthotopic liver transplantation.⁴  This has led to the paradigm that POPH survival improves with treatment and should be treated with PH-directed therapy. Also, and of equal importance, if the hemodynamic parameters become favorable enough, liver transplantation significantly increases long-term survival.⁴ 

General management of patients with both right ventricular failure and cirrhosis includes diuretics, a low-sodium diet, and lifestyle modifications.^16,21,22  It is also worth addressing nutritional status as patients with liver disease are typically in a catabolic state, and sarcopenia is common.²³ 

Beta-blockers are typically used in cirrhotic patients with a history of varices. However, in a small series of 10 patients with POPH on beta blockers, baseline RHC and walk distance was measured. After this initial assessment, their beta blockers were stopped. Nine of the 10 patients improved their walk distance, and their cardiac output improved by 28%, with a 19% increase in PVR. Since moderate to severe PH patients may be stroke-volume limited and highly dependent on heart rate for cardiac output, beta blockers are not recommended in patients with POPH, and banding of their varices may be a better option than medical therapy.²⁴ 

Additionally, although they are commonly used to control severe sequelae of portal hypertension in cirrhotic patients, it is also recommended to avoid transjugular intrahepatic portosystemic shunts in patients with POPH. This procedure will increase flow through the pulmonary circulation, increase right-sided pressures, and may precipitate worsening right ventricular failure.²⁵  Screening echocardiograms are often done before elective transjugular intrahepatic portosystemic shunt procedures to rule out subclinical PAH.

As delineated above in Table 1, POPH is classified into Group 1 PAH, and there are several US Food and Drug Administration (FDA)-approved medications for this group.¹⁶  The pivotal trials for these medications typically excluded patients with cirrhosis, so it is unclear how generalizable the results of these randomized double-blinded controlled trials are for patients with POPH. Thus, we often rely on case series, single-center studies, and expert opinions to guide therapeutic choices for POPH. The classes of medications with Group 1 PAH indications are phosphodiesterase (PDE)-5 inhibitors and cyclic guanosine monophosphate (GMP) inhibitors (both targeting the nitric oxide pathway), endothelin antagonists, and prostacyclins.

The 2 FDA-approved PDE-5 inhibitors for Group 1 PAH are sildenafil and tadalafil. The data available are from small trials but do show that the 6-minute walk distance is increased and the levels of N-terminal pro-brain natriuretic peptide (NT-proBNP) are improved when these medications are used in patients with POPH.²⁶ 

Riociguat, a soluble guanylate cyclase stimulator, was approved by the FDA based on the PATENT clinical trial, which showed improvement in walk distance, PVR, biomarkers, functional class, and time to clinical worsening in patients with World Health Organization Group I PAH. Patients with POPH were included in this trial (a rarity) but only comprised 3% of the trial participants.²⁷ 

Of note, since both PDE-5 inhibitors and cGMP stimulators increase nitric oxide, they cannot be used in combination with each other as the combination can lead to severe hypotension.²⁸ 

Bosentan, ambrisentan, and macitentan are the FDA-approved endothelin receptor antagonists for Group 1 PAH. Trials of endothelin receptor antagonists have demonstrated improvements in hemodynamics and walk distances.²⁹  The data for POPH patients are less robust.

Bosentan has some limited data in POPH patients showing improvements in exercise capacity and survival^30,31 ; still, its use has largely fallen by the wayside given its risk for hepatotoxicity due to interference with a bile salt transporter and the resultant need for liver function monitoring.³²  This side effect would be hazardous in patients with chronic liver disease, and it is not recommended for use in patients with POPH. Ambrisentan does not cause hepatotoxicity and is a much more attractive agent for patients with POPH. A large open-label trial looked at ambrisentan and POPH; it showed improvement in hemodynamics but no change in exercise capacity.³³  Macitentan, in a recent multicenter randomized controlled trial specifically in POPH patients, showed significant (35%) improvement in PVR but no significant improvement in 6-minute walk distance.³⁴  Macitentan, like ambrisentan, has no significant hepatotoxicity. The incidence of anemia is 13.2%³⁵ ; this should be monitored and is an important consideration in patients with chronic liver disease.

Prostacyclins have been the mainstay of PAH treatment for decades, and clinicians rely on them to improve the hemodynamics of patients with poor functional status and concerning hemodynamic parameters.³⁶  The intravenous formulations include epoprostenol and treprostinil. The latter is also available in subcutaneous, intravenous, and oral forms. Iloprost is another prostacyclin available in the intravenous (Europe only) and inhaled formulation.

The oral prostacyclin (treprostinil) and selexipag, an oral prostacyclin receptor agonist, have data in Group 1 PAH, but no data exist in the POPH population.

Prostacyclin infusions have been used to improve the hemodynamics in patients in anticipation of undergoing liver transplantation with overall success. In small single-center trials, mPAP, cardiac output, and PVR all improve on prostacyclin infusions in patients with POPH.^37–39  Specifically, in liver transplant candidates, studies have demonstrated improvement in hemodynamics followed by liver transplantation.^37,40,41 

Liver transplantation has a long and storied history in patients with POPH. Historically, due to either lack of diagnosis or lack of treatment of POPH in patients with cirrhosis, patients with POPH who underwent liver transplantation had significantly worse outcomes than matched controls without POPH.⁵  If the preoperative systolic pressure was estimated at > 60 mm Hg, the 9-month mortality after liver transplantation was 42%.⁴²  Given that, for many years, many centers had not offered transplantation for patients with moderate or severe POPH. However, with the advances in medical therapy in POPH, the data now show that if the hemodynamics can be improved with intravenous and/or oral PAH-directed medications, liver transplantation can be pursued with success.^37,40,41,43  Retrospective data suggest that if the mPAP is < 35 mm Hg and there is normal right ventricular function, the perioperative mortality is not increased beyond that of a patient without POPH undergoing transplantation in carefully selected patients. Conversely, if the mPAP is > 50 mm Hg, the perioperative mortality approaches 100%.^44–46  Therefore, if pursuing liver transplantation, the treatment goal is to maximize medical therapy to obtain an mPAP of < 35 mm Hg.⁴⁷ 

The United Network for Organ Sharing (UNOS) allows for exception points to be given to transplant candidates with mortality risks not calculated by the MELD score. Given the increased mortality risk in patients with POPH compared to their MELD-matched controls, UNOS suggests adding exception points for patients with a confirmed diagnosis of POPH that have reached the goal mPAP of < 35 mm Hg and PVR < 5 Wood units.⁴⁸  Of note, this documentation needs to be updated every 3 months with a repeat RHC to remain current. In a retrospective cohort study of patients in the Organ Procurement Transplant Network database who were listed for liver transplantation with MELD exception points, 11.1% of patients were reported as removed from the transplant list for clinical deterioration or as a 1-year waitlist mortality. Age, initial MELD score, and initial PVR were predictors of waitlist mortality.^49,50 

Of note, most of the data on which the current guidelines and exception-points rules were based were collected before the advent of current medical therapies, or patients were not on any PH-directed therapy. Often, these data were obtained in the operating room at the time of transplant; therefore, mPAP was recorded, but not the full hemodynamic profile. Because of this, mPAP is the hemodynamic criteria emphasized in MELD exception guidelines. Recently investigators queried whether a strict cut-off of 35 mm Hg was necessary to ensure successful transplantation. The investigators noted that modern PH therapy may drive up the patient’s cardiac output, thereby increasing pulmonary pressures above this threshold while having a low PVR and good right ventricular function, likely better markers of transplant candidacy than mPAP alone. Sixteen patients with a diagnosis of POPH on PH-directed medications were included. The hemodynamics prior to transplant showed an mPAP > 35 mm Hg but with a PVR < 250 dynes s cm⁻⁵. One-year survival was 69%; this survival is similar to other cohorts of patients with POPH undergoing transplant. More than half of the survivors were able to discontinue PH medications posttransplantation.⁵¹  This suggests that a full hemodynamic profile, including right ventricular function, may be more important than the strict mPAP number cutoff in the modern era of PH therapy.

Despite the evidence that the survival of patients with POPH is improved with liver transplantation,⁴  the perioperative course can be complicated, and a team of PH experts to manage the hemodynamics in the intra- and perioperative periods is recommended. An intraoperative event known as reperfusion syndrome may be poorly tolerated in patients with compromised right ventricular function. Cardiac output increases acutely at the time of liver allograft reperfusion and rises to triple its baseline in as little as 15 minutes.⁵²  This can cause acute elevations in right ventricular pressures, bradycardia, and acute right ventricular failure. Prompt and adept use of inhaled or intravenous prostacyclins and nitric oxide has been used to mitigate this injury with mixed results.^43,53  It is also suspected that surgical factors play a role; less blood loss, less time under anesthesia, and fewer fluid shifts are likely associated with a smoother intraoperative course.

The 5-year survival posttransplantation in patients with POPH has been reported as 54% to 87%; worse than non-POPH patients posttransplant.^54,55  Most patients’ hemodynamics will improve posttransplant as the portal hypertension is improved; however, this is not entirely consistent, and it takes time for the pulmonary vasculature to reverse remodel; 3 to 6 months is typical, and patients should stay on their PH-directed medications until they are followed up as an outpatient. Forty percent to 50% of patients will eventually discontinue their medication, but a substantial portion may need some long-term pulmonary vasodilators.^37,40,41,56 

Because of inconsistent postoperative results, provider beliefs and attitudes on liver transplant in patients with POPH is far from unanimous, even among the experts in the field. A recent article by DuBrock et al⁵⁷  highlighted provider attitudes regarding POPH and liver transplantation at centers that performed > 50 transplants/year. The responses varied widely even among this select group (Figure 2). This survey highlighted the variability of management strategies and attitudes among POPH specialists. For example, 15% of providers believed that POPH rarely or never improves after transplantation, while 42% responded that the PH always or often improves. Fifty percent agreed that treated POPH should be an indication for liver transplantation. This, of note, does not line up with current guidelines. Again, nearly half thought that the MELD exception criteria should be modified. This survey highlights the need for multicenter prospective studies, updated practice guidelines, and adherence thereof to improve the standardization of care.⁵⁷ 

POPH is a rare complication of portal hypertension. It is characterized by progressive pulmonary arteriopathy and vasoconstriction similar to other forms of PAH, progressing to right ventricular failure over time. Its presence increases the mortality rate compared with both similarly matched patients with PH or cirrhosis alone. The past 2 decades have demonstrated the efficacy of PH-directed therapy in this patient population, showing improvements in hemodynamics and other clinical outcomes. The role of liver transplantation in POPH continues to evolve; however, it is evident that in a select group of patients who have controlled hemodynamics, liver transplantation can be curative of not only their liver disease but also of their POPH. Additional prospective multicenter studies are necessary to help determine posttransplant outcomes and hemodynamic responsiveness in the era of updated PH therapeutics.