Saira Aijaz Khaderi, MD

Portopulmonary hypertension (POPH) is a form of group 1 pulmonary arterial hypertension. When treating patients with POPH, baseline assessment is necessary so that response to therapy can be measured as the change from baseline. Patients should undergo echocardiography and right heart catheterization, and their exercise capacity and NYHA functional class should be determined. Patients with POPH should be considered for treatment if they are NYHA functional class II or above and/or their mean pulmonary artery pressure (MPAP) is greater than 35 mm Hg in transplant candidates. The goal in the treatment and management of POPH is to improve pulmonary hemodynamics by reducing the obstruction to pulmonary arterial flow and to preserve right ventricular function (Table). This article, the second in a 2-part review of POPH in patients with liver disease, reviews the role of medical therapy and liver transplantation in treatment. Evaluation and diagnosis of POPH are discussed in a separate article.

Medical Therapy

Prostanoids

Although prostacyclin and prostaglandin analogs entered routine clinical practice for POPH in the 1990s, reports of investigational use date back to the 1980s. Prostanoids are potent vasodilators with antiplatelet aggregation and antiproliferative properties. Prostacyclin synthase is reduced in patients with PAH, resulting in decreased concentration of prostacyclin with vasoconstriction and proliferative changes in the pulmonary vasculature.¹

Epoprostenol

Epoprostenol is also known as synthetic prostaglandin I₂ or prostacyclin. It was the first therapy approved for the treatment of PAH in 1995 by the US Food and Drug Administration (FDA) as a continuous intravenous infusion.^2,3 It also inhibits platelet aggregation and may help modulate pulmonary vascular remodeling.^4,5 Epoprostenol is derived from the metabolism of arachidonic acid and is a potent pulmonary and systemic vasodilator. One study reported an immediate 11.8% decrease in MPAP, 24% decrease in pulmonary vascular resistance (PVR) and 28% drop in systemic vascular resistance (SVR) during an epoprostenol infusion.⁶ The authors reported that epoprostenol was a more potent vasodilator than nitric oxide and may have a role in predicting the reversibility of POPH. In a case series of 33 patients with secondary pulmonary hypertension (including 7 patients with POPH) treated with continuous intravenous prostacyclin for approximately 1 year, exercise tolerance, NYHA functional class, and pulmonary hemodynamics improved in each patient compared to baseline.⁷ Krowka et al studied 14 patients with moderate to severe POPH treated with intravenous epoprostenol.⁸ No significant side effects were noted and treatment resulted in significant improvements in PVR, MPAP, and cardiac output. In 2007, Fix et al published a large retrospective cohort of patients with moderate to severe POPH.⁹ Nineteen patients treated with epoprostenol were compared to 17 patients with no treatment. After a median treatment period of 15.4 months, the epoprostenol group showed significant improvement in MPAP, PVR and cardiac output, but survival did not differ between the 2 groups.

Epoprostenol has often been considered a bridge to transplant in patients with POPH. Sussman et al described 8 consecutive patients with POPH who were treated with intravenous epoprostenol (2 to 8 ng/kg/min dose).¹⁰ Liver transplant was considered in 7 of the 8 patients when MPAP decreased to less than 35 mm Hg. Six patients were eventually listed for liver transplant, but 2 died waiting on the list. Long-term outcomes in the group of transplanted recipients were excellent. They remained alive and well at least 9 to 18 months post-transplant, and half did not require long-term vasodilator therapy post-orthotopic liver transplant. Similarly, Ashfaq et al published their data on 16 patients with moderate-to-severe POPH who were treated with vasodilator therapy.¹¹ MPAP decreased to acceptable levels in 75% of the treated patients, and 11 went on to liver transplantation. Rates of 1- and 5-year survival in the transplanted patients were 91% and 67% respectively. None of the patients who failed vasodilator therapy survived.

Epoprostenol has a short half-life (3 to 5 minutes) and requires continuous infusion through central access via an infusion pump. Aseptic technique must be maintained to avoid blood stream infections. Pump failure or loss of vascular access can result in rebound pulmonary vasoconstriction that can be life-threatening and requires immediate attention. Side effects associated with epoprostenol include flushing, headache, nausea/vomiting, bradycardia, chest pain, jaw pain, diarrhea, and musculoskeletal pain.

Patients on epoprostenol should be monitored for prostanoid overdose. In the case of patients with chronic liver disease, epoprostenol increases systemic vasodilation in patients with already low systemic vascular tone. As a result, cardiac output may increase to the point of high cardiac output failure. MPAP will remain elevated secondary to high cardiac output rather than high PVR. In these patients, right heart catheterization will show an elevated MPAP in the setting of normal to low PVR/transpulmonary gradient (TPG) values. Lowering the epoprostenol dose will successfully reduce both cardiac output and MPAP.

Treprostinil

Treprostinil is a prostacyclin analog that is available in intravenous, inhalational, and subcutaneous form, although subcutaneous dosing may be limited by pain. Sakai et al published a small case series of 3 patients with PAH and end-stage liver disease treated with intravenous treprostinil.¹² Pulmonary hemodynamics improved in all patients, and 2 patients went on to an uneventful liver transplantation. More than 10 years later, data were published on 255 patients with PAH on therapy with bosentan or sildenafil randomized to additional inhaled treprostinil.¹³ Treprostinil proved to be safe and well tolerated, with improvement in quality of life measures but no improvement in other secondary endpoints.

Iloprost

Inhaled iloprost is another prostacyclin that has a short therapeutic half-life of 20 to 30 minutes and requires frequent administration (6 to 9 times daily). In study in which patients with severe POPH were treated for up to 3 years with inhaled iloprost,¹⁴ survival rates at 1, 2, and 3 years were 77%, 62%, and 46%, respectively. A second study published in 2010 was designed to assess the acute effects of inhaled iloprost on pulmonary hemodynamics and evaluate the clinical outcome after 12 months of treatment.¹⁵ Iloprost was found to rapidly reduce pulmonary arterial pressure and PVR. In the long-term evaluation, inhaled iloprost increased the 6-minute walk distance (6MWD) and functional class, but no change was noted in the systolic pulmonary artery pressure. The authors concluded that iloprost might provide symptomatic improvement and improvement in exercise capacity.

Selexipag

Selexipag is an oral selective IP prostacyclin receptor agonist that is structurally distinct from other prostacyclins.¹⁶ In a phase 3 randomized double blind clinical trial, PAH patients treated with selexipag had lower composite of death or complication of PAH to the end of the study period.¹⁷ This effect was consistent across all dose ranges, but POPH patients were excluded from this study. Safety and efficacy of selexipag has not been evaluated in POPH patients.

Endothelin Receptor Antagonists

Endothelin receptor antagonists block the production of endothelin-1 (ET-1), a potent vasoconstrictor and smooth muscle mitogen that may contribute to the development of PAH. Three different receptors have been described: endothelin A, endothelin B, and endothelin B2. Elevated ET-1 levels have been reported in patients with chronic liver disease and may originate from hepatosplanchnic circulation.¹⁸

Bosentan

Bosentan is an oral, nonspecific, ET-1A and ET-1B receptor antagonist. Initial use of bosentan in patients with POPH was limited because of concern for hepatotoxicity. Approximately 10% of patients on bosentan were reported to have mild hepatic side effects in the form of elevated aminotransferases, but severe injury has been reported.¹⁹ One of the first clinical experiences of bosentan in patients with POPH was published in 2005. Hoeper et al followed 11 patients with Child A cirrhosis and severe POPH.²⁰ All patients included were in NYHA functional class III or IV and were treated with bosentan for over 1 year. Exercise capacity and symptoms improved in all treated patients. The medication was tolerated well and there was no evidence of drug-induced liver injury. A single case report showed the effectiveness of bosentan in a 43-year-old man with alcohol-related liver disease (Child-Pugh A) and right ventricular enlargement and dysfunction secondary to POPH.²¹ Pulmonary arterial pressure decreased, exercise capacity increased, and improvement was maintained over 2 years.

In a group of 31 patients with Child A or B cirrhosis and severe POPH, bosentan had significantly better effects than inhaled iloprost on exercise capacity, hemodynamics, and survival.¹⁴ One, 2, and 3-year survival rates in the bosentan group were 94%, 89%, and 89% (compared to 77%, 62%, and 46% in the iloprost group). Both drugs were considered safe with no reported hepatotoxicity. In 2013, Savale et al published data on 34 patients with POPH, Child-Pugh A and/or B who were treated with bosentan for a median of 43 months.²² The authors reported significant improvements in hemodynamics, NYHA functional class, and 6WMD. Event-free survival rates at 1, 2, and 3 years were 82%, 63%, and 47%, respectively.

Ambrisentan

Ambrisentan is a highly selective ET-1A receptor antagonist with once daily dosing and a lower risk of hepatotoxicity compared to bosentan. Fourteen patients with moderate to severe POPH treated with ambrisentan in 4 German hospitals were retrospectively analyzed.²³ Median follow-up was 16 months, and the study demonstrated significant improvement in exercise capacity and clinical symptoms without significant change in liver function tests. Cartin-Ceba et al published their experience of 13 patients with moderate to severe POPH treated with ambrisentan monotherapy.²⁴ Patients were followed for a median of 613 days and on treatment for a median time of 390 days. Significant improvements were shown in pulmonary arterial pressure and PVR without adverse effect on hepatic function. Over 270 patients with PAH (6% with POPH) received ambrisentan from March 2009 through June 2013 at a large United Kingdom portal hypertension referral center.²⁵ Discontinuation due to side effects was higher than previously reported. Discontinuation due to abnormal transaminases was uncommon.

Macitentan

Macitentan is a dual endothelin-receptor antagonist developed by modifying the structure of bosentan to increase efficacy and safety. The SERAPHIN trial compared oral macitentan to placebo in 250 patients with moderate to severe PAH, some of whom were also on a stable dose of oral or inhaled therapy for PAH.²⁶ Over a 2-year period, patients treated with macitentan were less likely to have progression of their disease or die on therapy (38% and 31% versus 46%), regardless of if they were receiving additional oral therapy and more likely to have improvement of their exercise capacity and WHO functional class. Nasopharyngitis and significant anemia were more common in the macitentan group, but there was no difference in the rate of liver function test abnormalities compared to placebo. Trials with macitentan are currently ongoing in patients with POPH.

Phosphodiesterase-5 Inhibitors

Cyclic guanosine monophosphate (cGMP) is the mediator of nitric oxide–induced vasodilation. Phosphodiesterase-5 (PDE-5) inhibitors prolong the vasodilatory effects of cyclic guanosine monophosphate by preventing its hydrolysis, thereby reducing the pulmonary arterial pressure.

Sildenafil

Sildenafil is the most widely accepted PDE-5 inhibitor for POPH. Fourteen patients with moderate to severe POPH were treated with sildenafil (50 mg 3 times per day) in an observational study published by Reichenberger et al in 2006.²⁷ Eight patients were newly started on sildenafil, whereas sildenafil was added to inhaled prostanoids in the remaining 6x patients. Sildenafil significantly decreased 66MWD, MPAP, PVR, and cardiac index alone or in combination with inhaled prostanoids.

Sildenafil has also been used as a bridge to transplant in liver transplant candidates with POPH. Ten patients with POPH treated with sildenafil monotherapy were followed for a 21±16 months.²⁸ Patients improved symptomatically and increased their 6MWD at 1 year by 30 meters or more. Three patients became transplant eligible and another 3 patients were stable, without progression of their liver disease or POPH. Four patients were not considered transplant candidates, 2 because of refractory POPH and 2 for other comorbidities. The authors concluded that sildenafil monotherapy could stabilize or improve pulmonary hemodynamics in patients with POPH and eventually lead to liver transplantation. Gough et al took a similar look at 9 patients with POPH treated with sildenafil.²⁹ All patients had initial and follow-up right heart catheterizations within a period of 3 years. Mean PVR improved in all patients, decreasing from 575 to 375 dynes/s/cm^–5. MPAP decreased to ≤ 35 mmHg in 4 patients, 1 of whom went on to receive a liver transplant. Overall sildenafil improved pulmonary hemodynamics in this small cohort of POPH patients.

Tadalafil

Tadalafil is another oral PDE-5 inhibitor but with a longer half-life than sildenafil. Unlike sildenafil, which requires 3 times daily dosing, tadalafil requires once daily administration. A few case reports have demonstrated tadalafil’s effectiveness for POPH in combination with other medical therapy (eg, sildenafil, ambrisentan).^30,31

Guanylate Cyclase Stimulator

Riociguat

Riociguat is a first-in-class activator of soluble form of guanylate cyclase that increases levels of cyclic GMP. Two randomized clinical trials, PATENT, a study in PAH patients, and CHEST, a study in patients with chronic thromboembolic pulmonary hypertension showed improvement in 6MWD at 12 weeks (PATENT) or 16 weeks (CHEST), with improvement in secondary endpoints such as PVR, N-terminal pro b-type natriuretic peptide and WHO functional class.^32,33 Riociguat may have potential advantages in patients with POPH given that it has a favorable liver safety profile. A subgroup analysis of patients enrolled in the PATENT study showed that 13 had POPH and 11 were randomized to receive riociguat 2.5 mg 3 times daily dose and 2 received placebo.³⁴ Riociguat was well tolerated and improved 6MWD that was maintained over 2 years in the open label extension.

Medications to Avoid

Nonselective beta-blockers are commonly recommended in patients with portal hypertension to help prevent variceal hemorrhage. However, in patients with POPH, beta-blockers have been shown to decrease exercise capacity and worsen pulmonary hemodynamics. A study of 10 patients with moderate to severe POPH who were receiving beta-blockers for variceal bleeding prophylaxis showed that 6MWD improved in almost all of the patients, cardiac output increased by 28%, and PVR decreased by 19% when beta-blockers were discontinued.³⁵ The authors concluded that the use of beta-blockers should be avoided in this patient population.

Calcium channel blockers should not be used in patients with POPH because they can cause significant hypotension due to systemic vasodilatation and decreased right ventricular filling. Patients with portal hypertension and chronic liver disease commonly have low systemic vascular resistance and are particularly susceptible to the deleterious effects of calcium channel blockers.

Transplantation

Liver transplantation is a potential cure for POPH and its role in POPH has evolved over the past 2 decades. In 1997, Ramsay et al published their review of 1205 consecutive liver transplants at Baylor University Medical Center (BUMC) in Texas.³⁶ The incidence of POPH in this group was 8.5%, with the majority of patients having mild POPH. Liver transplant outcomes were not affected by mild and moderate pulmonary hypertension. However, patients with severe POPH (n = 7, systolic pulmonary artery pressure > 60 mm Hg) had a mortality rate of 42% at 9 months post-transplantation and 71% at 36 months post-transplant. The surviving patients continued to deteriorate with progressive right heart failure and no improvement in POPH.

To understand the effect of liver transplantation on POPH, one must understand the hemodynamic changes that occur with POPH and during liver transplant. The right ventricle is able to manage the same volume as the left ventricle under normal circumstances, but is unable to pump against a significant pressure gradient.³⁷ In the setting of POPH, right ventricular hypertrophy occurs and RV output remains stable for some time. With time, pulmonary artery pressure increases secondary to pulmonary arteriolar vasoconstriction, intimal thickening, and progressive occlusion of the pulmonary vascular bed. Right ventricular failure may occur as a result. Cardiac output increases significantly at the time of reperfusion during liver transplant (up to 3-fold in 15 minutes),³⁸ and in the setting of a noncompliant vascular bed, the patient is at risk for right heart failure. This is the likely explanation to such high perioperative mortality rates in patients with uncontrolled POPH. Failure to decrease MPAP to less than 50 mm Hg is considered a complete contraindication to liver transplant at most institutions. Many transplant centers will list patients for liver transplant if MPAP can be decreased to less than 35 mm Hg and PVR < 400 dynes/s/cm^–5. These parameters are thought to represent an adequate right ventricular reserve and a compliant pulmonary vascular bed.³⁷ However, even with good pressure control, the anesthesiology and critical care teams must be prepared to deal with acute right heart failure peri-operatively. Intraoperative transesophageal echocardiography has been recommended to closely follow right ventricular function.³⁸ Inhaled or intravenous dilators are the most effective agents in the event of a pulmonary hypertensive crisis.

Review of Outcomes

A retrospective review evaluated 43 patients with untreated POPH who underwent attempted liver transplantation.³⁹ Data were collected from 18 peer-reviewed studies and 7 patients at the authors’ institution. Overall mortality was 35% (15 patients), with almost all of the deaths secondary to cardiac dysfunction. Two deaths occurred intraoperatively and 8 deaths occurred during the transplant hospitalization. The transplant could not be successfully completed in 4 of the patients. MPAP > 50 mm Hg was associated with 100% mortality, whereas patients with MPAP between 35 mm Hg and 50 mm Hg had a 50% mortality. No mortality was noted in patients with MPAP < 35 mm Hg.

Liver transplantation has been shown to be successful in patients with controlled POPH. Sussman et al published their data on 8 patients with severe POPH in 2006. In this prospective study, all patients were treated with sequential epoprostenol infusions and 7 of the 8 patients experienced a significant reduction in MPAP and PVR. Six patients were listed for liver transplant, 4 of who were transplanted successfully and alive up to 5 years later.

The Baylor University Medical Center published their data on POPH patients who received liver transplants in 2007.¹¹ POPH was confirmed by right heart catheterization in 30 patients evaluated for liver transplant. Sixteen patients were considered to be suitable candidates for transplant and MPAP was decreased to less than 35 mmHg in 12 patients with vasodilator therapy. Eleven patients eventually underwent liver transplant and 1- and 5-year survival rates were 91% and 67%.

Compared to medical therapy or liver transplant alone, patients who receive medical therapy followed by liver transplantation have the best survival. The Mayo Clinic retrospectively reviewed 74 POPH patients identified between 1994 and 2007.⁴⁰ Patients were categorized in 1 of 3 categories: no medical therapy, medical therapy alone for POPH, or medical therapy for POPH followed by liver transplantation. Patients who received no medical therapy for POPH and no liver transplant had the worst outcomes, with a dismal 5-year survival of only 14% with over 50% deceased at 1 year of diagnosis. Five-year survival was 45% in patients who received medical therapy only. Patients who received medical therapy with prostacyclin followed by liver transplantation had the best outcomes, with a 5-year survival of 67% versus 25% in those who were transplanted without prior prostacyclin therapy.

We reported the longest follow-up study for patients undergoing liver transplantation with POPH in 2014.⁴¹ Seven patients with moderate to severe POPH received a liver transplant at our institution between June 2004 and January 2011. Mean pulmonary artery pressure was reduced to < 35 mm Hg, with appropriate POPH therapy in all of the patients. Both the graft and patient survival rates were 85.7% after a median follow-up of 7.8 years. The 1 patient who did not survive died from complications related to recurrent hepatitis C and cirrhosis, not from POPH-related issues. Four of the remaining 6 patients continue to require oral vasodilator therapy post-transplant, suggesting irreversible remodeling of the pulmonary vasculature. Two patients (4.4 and 8.5 years post-transplant) have no evidence of pulmonary hypertension post-transplant and therefore do not require medical treatment for pulmonary hypertension. We concluded that POPH responsive to vasodilator therapy is an appropriate indication for liver transplant, with excellent long-term survival.

Hollatz et al published their data on 11 patients with moderate to severe POPH who were successfully treated (mostly with oral sildenafil and subcutaneous treprostinil) as a bridge to liver transplant.⁴² The mortality rate was 0, with a follow-up duration of 7 to 60 months. Interestingly, 7 of the 11 patients (64%) were off all pulmonary vasodilators post-transplant. Ashfaq et al reported similar results.¹¹ Nine of 11 patients with treated moderate to severe POPH who received liver transplants stopped vasodilator therapy at a median period of 9.2 months post-transplant. Raevens et al described a group of 3 patients with POPH who went on to liver transplant after their pulmonary pressures were decreased with combined oral vasodilator therapy: 1 required continued long-term vasodilator therapy, another was weaned off medications after transplant, and the third patient died during the liver transplant from perioperative complications that induced uncontrolled pulmonary hypertension.⁴³

Patient Selection

In 2006, the United Network for Organ Sharing (UNOS) initiated a policy whereby a higher priority for liver transplantation was granted for highly selected patients in the United States.⁴⁴ UNOS policy 3.6.4.5.6 upgraded POPH patients to a MELD score of 22, with an increase in MELD every 3 months as long as MPAP remained < 35 mm Hg and PVR remained < 400 dynes/s/cm^–5. One hundred fifty-five patients were granted MELD exception points for POPH between 2002 and 2010 and went on to receive liver transplants.⁴⁵ Goldberg et al collected data from the Organ Procurement and Transplantation Network (OPTN) and compared outcomes of patients with approved POPH MELD exception points versus waitlist candidates with no exception points.⁴⁶ One hundred fifty-five waitlisted patients received POPH MELD exception points, with only 43.1% meeting OPTN exception requirements. One-third did not fulfill hemodynamic criteria consistent with POPH or had missing data, and 80% went on to receive a liver transplant. Waitlist candidates receiving POPH MELD exception points also had increased waitlist mortality and several early post-transplant deaths. The authors felt these data highlighted the need for OPTN/UNOS to revise their policy for POPH MELD exceptions points, revise how points are rewarded, and continue research to help risk stratify these patients to minimize perioperative complications.

Conclusion

Several effective medical treatment regimens are available, including prostanoids, endothelin receptor antagonists, and PDE-5 inhibitors. Liver transplantation is a potential cure but is only recommended if MPAP can be decreased to ≤ 35 mmHg. Long-term follow-up studies have shown these patients do well several years post-transplant but may continue to require oral therapy for their POPH.

Portopulmonary hypertension (POPH) is a form of group 1 pulmonary arterial hypertension. When treating patients with POPH, baseline assessment is necessary so that response to therapy can be measured as the change from baseline. Patients should undergo echocardiography and right heart catheterization, and their exercise capacity and NYHA functional class should be determined. Patients with POPH should be considered for treatment if they are NYHA functional class II or above and/or their mean pulmonary artery pressure (MPAP) is greater than 35 mm Hg in transplant candidates. The goal in the treatment and management of POPH is to improve pulmonary hemodynamics by reducing the obstruction to pulmonary arterial flow and to preserve right ventricular function (Table). This article, the second in a 2-part review of POPH in patients with liver disease, reviews the role of medical therapy and liver transplantation in treatment. Evaluation and diagnosis of POPH are discussed in a separate article.

Medical Therapy

Prostanoids

Although prostacyclin and prostaglandin analogs entered routine clinical practice for POPH in the 1990s, reports of investigational use date back to the 1980s. Prostanoids are potent vasodilators with antiplatelet aggregation and antiproliferative properties. Prostacyclin synthase is reduced in patients with PAH, resulting in decreased concentration of prostacyclin with vasoconstriction and proliferative changes in the pulmonary vasculature.¹

Epoprostenol

Epoprostenol is also known as synthetic prostaglandin I₂ or prostacyclin. It was the first therapy approved for the treatment of PAH in 1995 by the US Food and Drug Administration (FDA) as a continuous intravenous infusion.^2,3 It also inhibits platelet aggregation and may help modulate pulmonary vascular remodeling.^4,5 Epoprostenol is derived from the metabolism of arachidonic acid and is a potent pulmonary and systemic vasodilator. One study reported an immediate 11.8% decrease in MPAP, 24% decrease in pulmonary vascular resistance (PVR) and 28% drop in systemic vascular resistance (SVR) during an epoprostenol infusion.⁶ The authors reported that epoprostenol was a more potent vasodilator than nitric oxide and may have a role in predicting the reversibility of POPH. In a case series of 33 patients with secondary pulmonary hypertension (including 7 patients with POPH) treated with continuous intravenous prostacyclin for approximately 1 year, exercise tolerance, NYHA functional class, and pulmonary hemodynamics improved in each patient compared to baseline.⁷ Krowka et al studied 14 patients with moderate to severe POPH treated with intravenous epoprostenol.⁸ No significant side effects were noted and treatment resulted in significant improvements in PVR, MPAP, and cardiac output. In 2007, Fix et al published a large retrospective cohort of patients with moderate to severe POPH.⁹ Nineteen patients treated with epoprostenol were compared to 17 patients with no treatment. After a median treatment period of 15.4 months, the epoprostenol group showed significant improvement in MPAP, PVR and cardiac output, but survival did not differ between the 2 groups.

Epoprostenol has often been considered a bridge to transplant in patients with POPH. Sussman et al described 8 consecutive patients with POPH who were treated with intravenous epoprostenol (2 to 8 ng/kg/min dose).¹⁰ Liver transplant was considered in 7 of the 8 patients when MPAP decreased to less than 35 mm Hg. Six patients were eventually listed for liver transplant, but 2 died waiting on the list. Long-term outcomes in the group of transplanted recipients were excellent. They remained alive and well at least 9 to 18 months post-transplant, and half did not require long-term vasodilator therapy post-orthotopic liver transplant. Similarly, Ashfaq et al published their data on 16 patients with moderate-to-severe POPH who were treated with vasodilator therapy.¹¹ MPAP decreased to acceptable levels in 75% of the treated patients, and 11 went on to liver transplantation. Rates of 1- and 5-year survival in the transplanted patients were 91% and 67% respectively. None of the patients who failed vasodilator therapy survived.

Epoprostenol has a short half-life (3 to 5 minutes) and requires continuous infusion through central access via an infusion pump. Aseptic technique must be maintained to avoid blood stream infections. Pump failure or loss of vascular access can result in rebound pulmonary vasoconstriction that can be life-threatening and requires immediate attention. Side effects associated with epoprostenol include flushing, headache, nausea/vomiting, bradycardia, chest pain, jaw pain, diarrhea, and musculoskeletal pain.

Patients on epoprostenol should be monitored for prostanoid overdose. In the case of patients with chronic liver disease, epoprostenol increases systemic vasodilation in patients with already low systemic vascular tone. As a result, cardiac output may increase to the point of high cardiac output failure. MPAP will remain elevated secondary to high cardiac output rather than high PVR. In these patients, right heart catheterization will show an elevated MPAP in the setting of normal to low PVR/transpulmonary gradient (TPG) values. Lowering the epoprostenol dose will successfully reduce both cardiac output and MPAP.

Treprostinil

Treprostinil is a prostacyclin analog that is available in intravenous, inhalational, and subcutaneous form, although subcutaneous dosing may be limited by pain. Sakai et al published a small case series of 3 patients with PAH and end-stage liver disease treated with intravenous treprostinil.¹² Pulmonary hemodynamics improved in all patients, and 2 patients went on to an uneventful liver transplantation. More than 10 years later, data were published on 255 patients with PAH on therapy with bosentan or sildenafil randomized to additional inhaled treprostinil.¹³ Treprostinil proved to be safe and well tolerated, with improvement in quality of life measures but no improvement in other secondary endpoints.

Iloprost

Inhaled iloprost is another prostacyclin that has a short therapeutic half-life of 20 to 30 minutes and requires frequent administration (6 to 9 times daily). In study in which patients with severe POPH were treated for up to 3 years with inhaled iloprost,¹⁴ survival rates at 1, 2, and 3 years were 77%, 62%, and 46%, respectively. A second study published in 2010 was designed to assess the acute effects of inhaled iloprost on pulmonary hemodynamics and evaluate the clinical outcome after 12 months of treatment.¹⁵ Iloprost was found to rapidly reduce pulmonary arterial pressure and PVR. In the long-term evaluation, inhaled iloprost increased the 6-minute walk distance (6MWD) and functional class, but no change was noted in the systolic pulmonary artery pressure. The authors concluded that iloprost might provide symptomatic improvement and improvement in exercise capacity.

Selexipag

Selexipag is an oral selective IP prostacyclin receptor agonist that is structurally distinct from other prostacyclins.¹⁶ In a phase 3 randomized double blind clinical trial, PAH patients treated with selexipag had lower composite of death or complication of PAH to the end of the study period.¹⁷ This effect was consistent across all dose ranges, but POPH patients were excluded from this study. Safety and efficacy of selexipag has not been evaluated in POPH patients.

Endothelin Receptor Antagonists

Endothelin receptor antagonists block the production of endothelin-1 (ET-1), a potent vasoconstrictor and smooth muscle mitogen that may contribute to the development of PAH. Three different receptors have been described: endothelin A, endothelin B, and endothelin B2. Elevated ET-1 levels have been reported in patients with chronic liver disease and may originate from hepatosplanchnic circulation.¹⁸

Bosentan

Bosentan is an oral, nonspecific, ET-1A and ET-1B receptor antagonist. Initial use of bosentan in patients with POPH was limited because of concern for hepatotoxicity. Approximately 10% of patients on bosentan were reported to have mild hepatic side effects in the form of elevated aminotransferases, but severe injury has been reported.¹⁹ One of the first clinical experiences of bosentan in patients with POPH was published in 2005. Hoeper et al followed 11 patients with Child A cirrhosis and severe POPH.²⁰ All patients included were in NYHA functional class III or IV and were treated with bosentan for over 1 year. Exercise capacity and symptoms improved in all treated patients. The medication was tolerated well and there was no evidence of drug-induced liver injury. A single case report showed the effectiveness of bosentan in a 43-year-old man with alcohol-related liver disease (Child-Pugh A) and right ventricular enlargement and dysfunction secondary to POPH.²¹ Pulmonary arterial pressure decreased, exercise capacity increased, and improvement was maintained over 2 years.

In a group of 31 patients with Child A or B cirrhosis and severe POPH, bosentan had significantly better effects than inhaled iloprost on exercise capacity, hemodynamics, and survival.¹⁴ One, 2, and 3-year survival rates in the bosentan group were 94%, 89%, and 89% (compared to 77%, 62%, and 46% in the iloprost group). Both drugs were considered safe with no reported hepatotoxicity. In 2013, Savale et al published data on 34 patients with POPH, Child-Pugh A and/or B who were treated with bosentan for a median of 43 months.²² The authors reported significant improvements in hemodynamics, NYHA functional class, and 6WMD. Event-free survival rates at 1, 2, and 3 years were 82%, 63%, and 47%, respectively.

Ambrisentan

Ambrisentan is a highly selective ET-1A receptor antagonist with once daily dosing and a lower risk of hepatotoxicity compared to bosentan. Fourteen patients with moderate to severe POPH treated with ambrisentan in 4 German hospitals were retrospectively analyzed.²³ Median follow-up was 16 months, and the study demonstrated significant improvement in exercise capacity and clinical symptoms without significant change in liver function tests. Cartin-Ceba et al published their experience of 13 patients with moderate to severe POPH treated with ambrisentan monotherapy.²⁴ Patients were followed for a median of 613 days and on treatment for a median time of 390 days. Significant improvements were shown in pulmonary arterial pressure and PVR without adverse effect on hepatic function. Over 270 patients with PAH (6% with POPH) received ambrisentan from March 2009 through June 2013 at a large United Kingdom portal hypertension referral center.²⁵ Discontinuation due to side effects was higher than previously reported. Discontinuation due to abnormal transaminases was uncommon.

Macitentan

Macitentan is a dual endothelin-receptor antagonist developed by modifying the structure of bosentan to increase efficacy and safety. The SERAPHIN trial compared oral macitentan to placebo in 250 patients with moderate to severe PAH, some of whom were also on a stable dose of oral or inhaled therapy for PAH.²⁶ Over a 2-year period, patients treated with macitentan were less likely to have progression of their disease or die on therapy (38% and 31% versus 46%), regardless of if they were receiving additional oral therapy and more likely to have improvement of their exercise capacity and WHO functional class. Nasopharyngitis and significant anemia were more common in the macitentan group, but there was no difference in the rate of liver function test abnormalities compared to placebo. Trials with macitentan are currently ongoing in patients with POPH.

Phosphodiesterase-5 Inhibitors

Cyclic guanosine monophosphate (cGMP) is the mediator of nitric oxide–induced vasodilation. Phosphodiesterase-5 (PDE-5) inhibitors prolong the vasodilatory effects of cyclic guanosine monophosphate by preventing its hydrolysis, thereby reducing the pulmonary arterial pressure.

Sildenafil

Sildenafil is the most widely accepted PDE-5 inhibitor for POPH. Fourteen patients with moderate to severe POPH were treated with sildenafil (50 mg 3 times per day) in an observational study published by Reichenberger et al in 2006.²⁷ Eight patients were newly started on sildenafil, whereas sildenafil was added to inhaled prostanoids in the remaining 6x patients. Sildenafil significantly decreased 66MWD, MPAP, PVR, and cardiac index alone or in combination with inhaled prostanoids.

Sildenafil has also been used as a bridge to transplant in liver transplant candidates with POPH. Ten patients with POPH treated with sildenafil monotherapy were followed for a 21±16 months.²⁸ Patients improved symptomatically and increased their 6MWD at 1 year by 30 meters or more. Three patients became transplant eligible and another 3 patients were stable, without progression of their liver disease or POPH. Four patients were not considered transplant candidates, 2 because of refractory POPH and 2 for other comorbidities. The authors concluded that sildenafil monotherapy could stabilize or improve pulmonary hemodynamics in patients with POPH and eventually lead to liver transplantation. Gough et al took a similar look at 9 patients with POPH treated with sildenafil.²⁹ All patients had initial and follow-up right heart catheterizations within a period of 3 years. Mean PVR improved in all patients, decreasing from 575 to 375 dynes/s/cm^–5. MPAP decreased to ≤ 35 mmHg in 4 patients, 1 of whom went on to receive a liver transplant. Overall sildenafil improved pulmonary hemodynamics in this small cohort of POPH patients.

Tadalafil

Tadalafil is another oral PDE-5 inhibitor but with a longer half-life than sildenafil. Unlike sildenafil, which requires 3 times daily dosing, tadalafil requires once daily administration. A few case reports have demonstrated tadalafil’s effectiveness for POPH in combination with other medical therapy (eg, sildenafil, ambrisentan).^30,31

Guanylate Cyclase Stimulator

Riociguat

Riociguat is a first-in-class activator of soluble form of guanylate cyclase that increases levels of cyclic GMP. Two randomized clinical trials, PATENT, a study in PAH patients, and CHEST, a study in patients with chronic thromboembolic pulmonary hypertension showed improvement in 6MWD at 12 weeks (PATENT) or 16 weeks (CHEST), with improvement in secondary endpoints such as PVR, N-terminal pro b-type natriuretic peptide and WHO functional class.^32,33 Riociguat may have potential advantages in patients with POPH given that it has a favorable liver safety profile. A subgroup analysis of patients enrolled in the PATENT study showed that 13 had POPH and 11 were randomized to receive riociguat 2.5 mg 3 times daily dose and 2 received placebo.³⁴ Riociguat was well tolerated and improved 6MWD that was maintained over 2 years in the open label extension.

Medications to Avoid

Nonselective beta-blockers are commonly recommended in patients with portal hypertension to help prevent variceal hemorrhage. However, in patients with POPH, beta-blockers have been shown to decrease exercise capacity and worsen pulmonary hemodynamics. A study of 10 patients with moderate to severe POPH who were receiving beta-blockers for variceal bleeding prophylaxis showed that 6MWD improved in almost all of the patients, cardiac output increased by 28%, and PVR decreased by 19% when beta-blockers were discontinued.³⁵ The authors concluded that the use of beta-blockers should be avoided in this patient population.

Calcium channel blockers should not be used in patients with POPH because they can cause significant hypotension due to systemic vasodilatation and decreased right ventricular filling. Patients with portal hypertension and chronic liver disease commonly have low systemic vascular resistance and are particularly susceptible to the deleterious effects of calcium channel blockers.

Transplantation

Liver transplantation is a potential cure for POPH and its role in POPH has evolved over the past 2 decades. In 1997, Ramsay et al published their review of 1205 consecutive liver transplants at Baylor University Medical Center (BUMC) in Texas.³⁶ The incidence of POPH in this group was 8.5%, with the majority of patients having mild POPH. Liver transplant outcomes were not affected by mild and moderate pulmonary hypertension. However, patients with severe POPH (n = 7, systolic pulmonary artery pressure > 60 mm Hg) had a mortality rate of 42% at 9 months post-transplantation and 71% at 36 months post-transplant. The surviving patients continued to deteriorate with progressive right heart failure and no improvement in POPH.

To understand the effect of liver transplantation on POPH, one must understand the hemodynamic changes that occur with POPH and during liver transplant. The right ventricle is able to manage the same volume as the left ventricle under normal circumstances, but is unable to pump against a significant pressure gradient.³⁷ In the setting of POPH, right ventricular hypertrophy occurs and RV output remains stable for some time. With time, pulmonary artery pressure increases secondary to pulmonary arteriolar vasoconstriction, intimal thickening, and progressive occlusion of the pulmonary vascular bed. Right ventricular failure may occur as a result. Cardiac output increases significantly at the time of reperfusion during liver transplant (up to 3-fold in 15 minutes),³⁸ and in the setting of a noncompliant vascular bed, the patient is at risk for right heart failure. This is the likely explanation to such high perioperative mortality rates in patients with uncontrolled POPH. Failure to decrease MPAP to less than 50 mm Hg is considered a complete contraindication to liver transplant at most institutions. Many transplant centers will list patients for liver transplant if MPAP can be decreased to less than 35 mm Hg and PVR < 400 dynes/s/cm^–5. These parameters are thought to represent an adequate right ventricular reserve and a compliant pulmonary vascular bed.³⁷ However, even with good pressure control, the anesthesiology and critical care teams must be prepared to deal with acute right heart failure peri-operatively. Intraoperative transesophageal echocardiography has been recommended to closely follow right ventricular function.³⁸ Inhaled or intravenous dilators are the most effective agents in the event of a pulmonary hypertensive crisis.

Review of Outcomes

A retrospective review evaluated 43 patients with untreated POPH who underwent attempted liver transplantation.³⁹ Data were collected from 18 peer-reviewed studies and 7 patients at the authors’ institution. Overall mortality was 35% (15 patients), with almost all of the deaths secondary to cardiac dysfunction. Two deaths occurred intraoperatively and 8 deaths occurred during the transplant hospitalization. The transplant could not be successfully completed in 4 of the patients. MPAP > 50 mm Hg was associated with 100% mortality, whereas patients with MPAP between 35 mm Hg and 50 mm Hg had a 50% mortality. No mortality was noted in patients with MPAP < 35 mm Hg.

Liver transplantation has been shown to be successful in patients with controlled POPH. Sussman et al published their data on 8 patients with severe POPH in 2006. In this prospective study, all patients were treated with sequential epoprostenol infusions and 7 of the 8 patients experienced a significant reduction in MPAP and PVR. Six patients were listed for liver transplant, 4 of who were transplanted successfully and alive up to 5 years later.

The Baylor University Medical Center published their data on POPH patients who received liver transplants in 2007.¹¹ POPH was confirmed by right heart catheterization in 30 patients evaluated for liver transplant. Sixteen patients were considered to be suitable candidates for transplant and MPAP was decreased to less than 35 mmHg in 12 patients with vasodilator therapy. Eleven patients eventually underwent liver transplant and 1- and 5-year survival rates were 91% and 67%.

Compared to medical therapy or liver transplant alone, patients who receive medical therapy followed by liver transplantation have the best survival. The Mayo Clinic retrospectively reviewed 74 POPH patients identified between 1994 and 2007.⁴⁰ Patients were categorized in 1 of 3 categories: no medical therapy, medical therapy alone for POPH, or medical therapy for POPH followed by liver transplantation. Patients who received no medical therapy for POPH and no liver transplant had the worst outcomes, with a dismal 5-year survival of only 14% with over 50% deceased at 1 year of diagnosis. Five-year survival was 45% in patients who received medical therapy only. Patients who received medical therapy with prostacyclin followed by liver transplantation had the best outcomes, with a 5-year survival of 67% versus 25% in those who were transplanted without prior prostacyclin therapy.

We reported the longest follow-up study for patients undergoing liver transplantation with POPH in 2014.⁴¹ Seven patients with moderate to severe POPH received a liver transplant at our institution between June 2004 and January 2011. Mean pulmonary artery pressure was reduced to < 35 mm Hg, with appropriate POPH therapy in all of the patients. Both the graft and patient survival rates were 85.7% after a median follow-up of 7.8 years. The 1 patient who did not survive died from complications related to recurrent hepatitis C and cirrhosis, not from POPH-related issues. Four of the remaining 6 patients continue to require oral vasodilator therapy post-transplant, suggesting irreversible remodeling of the pulmonary vasculature. Two patients (4.4 and 8.5 years post-transplant) have no evidence of pulmonary hypertension post-transplant and therefore do not require medical treatment for pulmonary hypertension. We concluded that POPH responsive to vasodilator therapy is an appropriate indication for liver transplant, with excellent long-term survival.

Hollatz et al published their data on 11 patients with moderate to severe POPH who were successfully treated (mostly with oral sildenafil and subcutaneous treprostinil) as a bridge to liver transplant.⁴² The mortality rate was 0, with a follow-up duration of 7 to 60 months. Interestingly, 7 of the 11 patients (64%) were off all pulmonary vasodilators post-transplant. Ashfaq et al reported similar results.¹¹ Nine of 11 patients with treated moderate to severe POPH who received liver transplants stopped vasodilator therapy at a median period of 9.2 months post-transplant. Raevens et al described a group of 3 patients with POPH who went on to liver transplant after their pulmonary pressures were decreased with combined oral vasodilator therapy: 1 required continued long-term vasodilator therapy, another was weaned off medications after transplant, and the third patient died during the liver transplant from perioperative complications that induced uncontrolled pulmonary hypertension.⁴³

Patient Selection

In 2006, the United Network for Organ Sharing (UNOS) initiated a policy whereby a higher priority for liver transplantation was granted for highly selected patients in the United States.⁴⁴ UNOS policy 3.6.4.5.6 upgraded POPH patients to a MELD score of 22, with an increase in MELD every 3 months as long as MPAP remained < 35 mm Hg and PVR remained < 400 dynes/s/cm^–5. One hundred fifty-five patients were granted MELD exception points for POPH between 2002 and 2010 and went on to receive liver transplants.⁴⁵ Goldberg et al collected data from the Organ Procurement and Transplantation Network (OPTN) and compared outcomes of patients with approved POPH MELD exception points versus waitlist candidates with no exception points.⁴⁶ One hundred fifty-five waitlisted patients received POPH MELD exception points, with only 43.1% meeting OPTN exception requirements. One-third did not fulfill hemodynamic criteria consistent with POPH or had missing data, and 80% went on to receive a liver transplant. Waitlist candidates receiving POPH MELD exception points also had increased waitlist mortality and several early post-transplant deaths. The authors felt these data highlighted the need for OPTN/UNOS to revise their policy for POPH MELD exceptions points, revise how points are rewarded, and continue research to help risk stratify these patients to minimize perioperative complications.

Conclusion

Several effective medical treatment regimens are available, including prostanoids, endothelin receptor antagonists, and PDE-5 inhibitors. Liver transplantation is a potential cure but is only recommended if MPAP can be decreased to ≤ 35 mmHg. Long-term follow-up studies have shown these patients do well several years post-transplant but may continue to require oral therapy for their POPH.

Portopulmonary hypertension (POPH) is a form of group 1 pulmonary arterial hypertension. When treating patients with POPH, baseline assessment is necessary so that response to therapy can be measured as the change from baseline. Patients should undergo echocardiography and right heart catheterization, and their exercise capacity and NYHA functional class should be determined. Patients with POPH should be considered for treatment if they are NYHA functional class II or above and/or their mean pulmonary artery pressure (MPAP) is greater than 35 mm Hg in transplant candidates. The goal in the treatment and management of POPH is to improve pulmonary hemodynamics by reducing the obstruction to pulmonary arterial flow and to preserve right ventricular function (Table). This article, the second in a 2-part review of POPH in patients with liver disease, reviews the role of medical therapy and liver transplantation in treatment. Evaluation and diagnosis of POPH are discussed in a separate article.

Medical Therapy

Prostanoids

Although prostacyclin and prostaglandin analogs entered routine clinical practice for POPH in the 1990s, reports of investigational use date back to the 1980s. Prostanoids are potent vasodilators with antiplatelet aggregation and antiproliferative properties. Prostacyclin synthase is reduced in patients with PAH, resulting in decreased concentration of prostacyclin with vasoconstriction and proliferative changes in the pulmonary vasculature.¹

Epoprostenol

Epoprostenol is also known as synthetic prostaglandin I₂ or prostacyclin. It was the first therapy approved for the treatment of PAH in 1995 by the US Food and Drug Administration (FDA) as a continuous intravenous infusion.^2,3 It also inhibits platelet aggregation and may help modulate pulmonary vascular remodeling.^4,5 Epoprostenol is derived from the metabolism of arachidonic acid and is a potent pulmonary and systemic vasodilator. One study reported an immediate 11.8% decrease in MPAP, 24% decrease in pulmonary vascular resistance (PVR) and 28% drop in systemic vascular resistance (SVR) during an epoprostenol infusion.⁶ The authors reported that epoprostenol was a more potent vasodilator than nitric oxide and may have a role in predicting the reversibility of POPH. In a case series of 33 patients with secondary pulmonary hypertension (including 7 patients with POPH) treated with continuous intravenous prostacyclin for approximately 1 year, exercise tolerance, NYHA functional class, and pulmonary hemodynamics improved in each patient compared to baseline.⁷ Krowka et al studied 14 patients with moderate to severe POPH treated with intravenous epoprostenol.⁸ No significant side effects were noted and treatment resulted in significant improvements in PVR, MPAP, and cardiac output. In 2007, Fix et al published a large retrospective cohort of patients with moderate to severe POPH.⁹ Nineteen patients treated with epoprostenol were compared to 17 patients with no treatment. After a median treatment period of 15.4 months, the epoprostenol group showed significant improvement in MPAP, PVR and cardiac output, but survival did not differ between the 2 groups.

Epoprostenol has often been considered a bridge to transplant in patients with POPH. Sussman et al described 8 consecutive patients with POPH who were treated with intravenous epoprostenol (2 to 8 ng/kg/min dose).¹⁰ Liver transplant was considered in 7 of the 8 patients when MPAP decreased to less than 35 mm Hg. Six patients were eventually listed for liver transplant, but 2 died waiting on the list. Long-term outcomes in the group of transplanted recipients were excellent. They remained alive and well at least 9 to 18 months post-transplant, and half did not require long-term vasodilator therapy post-orthotopic liver transplant. Similarly, Ashfaq et al published their data on 16 patients with moderate-to-severe POPH who were treated with vasodilator therapy.¹¹ MPAP decreased to acceptable levels in 75% of the treated patients, and 11 went on to liver transplantation. Rates of 1- and 5-year survival in the transplanted patients were 91% and 67% respectively. None of the patients who failed vasodilator therapy survived.

Epoprostenol has a short half-life (3 to 5 minutes) and requires continuous infusion through central access via an infusion pump. Aseptic technique must be maintained to avoid blood stream infections. Pump failure or loss of vascular access can result in rebound pulmonary vasoconstriction that can be life-threatening and requires immediate attention. Side effects associated with epoprostenol include flushing, headache, nausea/vomiting, bradycardia, chest pain, jaw pain, diarrhea, and musculoskeletal pain.

Patients on epoprostenol should be monitored for prostanoid overdose. In the case of patients with chronic liver disease, epoprostenol increases systemic vasodilation in patients with already low systemic vascular tone. As a result, cardiac output may increase to the point of high cardiac output failure. MPAP will remain elevated secondary to high cardiac output rather than high PVR. In these patients, right heart catheterization will show an elevated MPAP in the setting of normal to low PVR/transpulmonary gradient (TPG) values. Lowering the epoprostenol dose will successfully reduce both cardiac output and MPAP.

Treprostinil

Treprostinil is a prostacyclin analog that is available in intravenous, inhalational, and subcutaneous form, although subcutaneous dosing may be limited by pain. Sakai et al published a small case series of 3 patients with PAH and end-stage liver disease treated with intravenous treprostinil.¹² Pulmonary hemodynamics improved in all patients, and 2 patients went on to an uneventful liver transplantation. More than 10 years later, data were published on 255 patients with PAH on therapy with bosentan or sildenafil randomized to additional inhaled treprostinil.¹³ Treprostinil proved to be safe and well tolerated, with improvement in quality of life measures but no improvement in other secondary endpoints.

Iloprost

Inhaled iloprost is another prostacyclin that has a short therapeutic half-life of 20 to 30 minutes and requires frequent administration (6 to 9 times daily). In study in which patients with severe POPH were treated for up to 3 years with inhaled iloprost,¹⁴ survival rates at 1, 2, and 3 years were 77%, 62%, and 46%, respectively. A second study published in 2010 was designed to assess the acute effects of inhaled iloprost on pulmonary hemodynamics and evaluate the clinical outcome after 12 months of treatment.¹⁵ Iloprost was found to rapidly reduce pulmonary arterial pressure and PVR. In the long-term evaluation, inhaled iloprost increased the 6-minute walk distance (6MWD) and functional class, but no change was noted in the systolic pulmonary artery pressure. The authors concluded that iloprost might provide symptomatic improvement and improvement in exercise capacity.

Selexipag

Selexipag is an oral selective IP prostacyclin receptor agonist that is structurally distinct from other prostacyclins.¹⁶ In a phase 3 randomized double blind clinical trial, PAH patients treated with selexipag had lower composite of death or complication of PAH to the end of the study period.¹⁷ This effect was consistent across all dose ranges, but POPH patients were excluded from this study. Safety and efficacy of selexipag has not been evaluated in POPH patients.

Endothelin Receptor Antagonists

Endothelin receptor antagonists block the production of endothelin-1 (ET-1), a potent vasoconstrictor and smooth muscle mitogen that may contribute to the development of PAH. Three different receptors have been described: endothelin A, endothelin B, and endothelin B2. Elevated ET-1 levels have been reported in patients with chronic liver disease and may originate from hepatosplanchnic circulation.¹⁸

Bosentan

Bosentan is an oral, nonspecific, ET-1A and ET-1B receptor antagonist. Initial use of bosentan in patients with POPH was limited because of concern for hepatotoxicity. Approximately 10% of patients on bosentan were reported to have mild hepatic side effects in the form of elevated aminotransferases, but severe injury has been reported.¹⁹ One of the first clinical experiences of bosentan in patients with POPH was published in 2005. Hoeper et al followed 11 patients with Child A cirrhosis and severe POPH.²⁰ All patients included were in NYHA functional class III or IV and were treated with bosentan for over 1 year. Exercise capacity and symptoms improved in all treated patients. The medication was tolerated well and there was no evidence of drug-induced liver injury. A single case report showed the effectiveness of bosentan in a 43-year-old man with alcohol-related liver disease (Child-Pugh A) and right ventricular enlargement and dysfunction secondary to POPH.²¹ Pulmonary arterial pressure decreased, exercise capacity increased, and improvement was maintained over 2 years.

In a group of 31 patients with Child A or B cirrhosis and severe POPH, bosentan had significantly better effects than inhaled iloprost on exercise capacity, hemodynamics, and survival.¹⁴ One, 2, and 3-year survival rates in the bosentan group were 94%, 89%, and 89% (compared to 77%, 62%, and 46% in the iloprost group). Both drugs were considered safe with no reported hepatotoxicity. In 2013, Savale et al published data on 34 patients with POPH, Child-Pugh A and/or B who were treated with bosentan for a median of 43 months.²² The authors reported significant improvements in hemodynamics, NYHA functional class, and 6WMD. Event-free survival rates at 1, 2, and 3 years were 82%, 63%, and 47%, respectively.

Ambrisentan

Ambrisentan is a highly selective ET-1A receptor antagonist with once daily dosing and a lower risk of hepatotoxicity compared to bosentan. Fourteen patients with moderate to severe POPH treated with ambrisentan in 4 German hospitals were retrospectively analyzed.²³ Median follow-up was 16 months, and the study demonstrated significant improvement in exercise capacity and clinical symptoms without significant change in liver function tests. Cartin-Ceba et al published their experience of 13 patients with moderate to severe POPH treated with ambrisentan monotherapy.²⁴ Patients were followed for a median of 613 days and on treatment for a median time of 390 days. Significant improvements were shown in pulmonary arterial pressure and PVR without adverse effect on hepatic function. Over 270 patients with PAH (6% with POPH) received ambrisentan from March 2009 through June 2013 at a large United Kingdom portal hypertension referral center.²⁵ Discontinuation due to side effects was higher than previously reported. Discontinuation due to abnormal transaminases was uncommon.

Macitentan

Macitentan is a dual endothelin-receptor antagonist developed by modifying the structure of bosentan to increase efficacy and safety. The SERAPHIN trial compared oral macitentan to placebo in 250 patients with moderate to severe PAH, some of whom were also on a stable dose of oral or inhaled therapy for PAH.²⁶ Over a 2-year period, patients treated with macitentan were less likely to have progression of their disease or die on therapy (38% and 31% versus 46%), regardless of if they were receiving additional oral therapy and more likely to have improvement of their exercise capacity and WHO functional class. Nasopharyngitis and significant anemia were more common in the macitentan group, but there was no difference in the rate of liver function test abnormalities compared to placebo. Trials with macitentan are currently ongoing in patients with POPH.

Phosphodiesterase-5 Inhibitors

Cyclic guanosine monophosphate (cGMP) is the mediator of nitric oxide–induced vasodilation. Phosphodiesterase-5 (PDE-5) inhibitors prolong the vasodilatory effects of cyclic guanosine monophosphate by preventing its hydrolysis, thereby reducing the pulmonary arterial pressure.

Sildenafil

Sildenafil is the most widely accepted PDE-5 inhibitor for POPH. Fourteen patients with moderate to severe POPH were treated with sildenafil (50 mg 3 times per day) in an observational study published by Reichenberger et al in 2006.²⁷ Eight patients were newly started on sildenafil, whereas sildenafil was added to inhaled prostanoids in the remaining 6x patients. Sildenafil significantly decreased 66MWD, MPAP, PVR, and cardiac index alone or in combination with inhaled prostanoids.

Sildenafil has also been used as a bridge to transplant in liver transplant candidates with POPH. Ten patients with POPH treated with sildenafil monotherapy were followed for a 21±16 months.²⁸ Patients improved symptomatically and increased their 6MWD at 1 year by 30 meters or more. Three patients became transplant eligible and another 3 patients were stable, without progression of their liver disease or POPH. Four patients were not considered transplant candidates, 2 because of refractory POPH and 2 for other comorbidities. The authors concluded that sildenafil monotherapy could stabilize or improve pulmonary hemodynamics in patients with POPH and eventually lead to liver transplantation. Gough et al took a similar look at 9 patients with POPH treated with sildenafil.²⁹ All patients had initial and follow-up right heart catheterizations within a period of 3 years. Mean PVR improved in all patients, decreasing from 575 to 375 dynes/s/cm^–5. MPAP decreased to ≤ 35 mmHg in 4 patients, 1 of whom went on to receive a liver transplant. Overall sildenafil improved pulmonary hemodynamics in this small cohort of POPH patients.

Tadalafil

Tadalafil is another oral PDE-5 inhibitor but with a longer half-life than sildenafil. Unlike sildenafil, which requires 3 times daily dosing, tadalafil requires once daily administration. A few case reports have demonstrated tadalafil’s effectiveness for POPH in combination with other medical therapy (eg, sildenafil, ambrisentan).^30,31

Guanylate Cyclase Stimulator

Riociguat

Riociguat is a first-in-class activator of soluble form of guanylate cyclase that increases levels of cyclic GMP. Two randomized clinical trials, PATENT, a study in PAH patients, and CHEST, a study in patients with chronic thromboembolic pulmonary hypertension showed improvement in 6MWD at 12 weeks (PATENT) or 16 weeks (CHEST), with improvement in secondary endpoints such as PVR, N-terminal pro b-type natriuretic peptide and WHO functional class.^32,33 Riociguat may have potential advantages in patients with POPH given that it has a favorable liver safety profile. A subgroup analysis of patients enrolled in the PATENT study showed that 13 had POPH and 11 were randomized to receive riociguat 2.5 mg 3 times daily dose and 2 received placebo.³⁴ Riociguat was well tolerated and improved 6MWD that was maintained over 2 years in the open label extension.

Medications to Avoid

Nonselective beta-blockers are commonly recommended in patients with portal hypertension to help prevent variceal hemorrhage. However, in patients with POPH, beta-blockers have been shown to decrease exercise capacity and worsen pulmonary hemodynamics. A study of 10 patients with moderate to severe POPH who were receiving beta-blockers for variceal bleeding prophylaxis showed that 6MWD improved in almost all of the patients, cardiac output increased by 28%, and PVR decreased by 19% when beta-blockers were discontinued.³⁵ The authors concluded that the use of beta-blockers should be avoided in this patient population.

Calcium channel blockers should not be used in patients with POPH because they can cause significant hypotension due to systemic vasodilatation and decreased right ventricular filling. Patients with portal hypertension and chronic liver disease commonly have low systemic vascular resistance and are particularly susceptible to the deleterious effects of calcium channel blockers.

Transplantation

Liver transplantation is a potential cure for POPH and its role in POPH has evolved over the past 2 decades. In 1997, Ramsay et al published their review of 1205 consecutive liver transplants at Baylor University Medical Center (BUMC) in Texas.³⁶ The incidence of POPH in this group was 8.5%, with the majority of patients having mild POPH. Liver transplant outcomes were not affected by mild and moderate pulmonary hypertension. However, patients with severe POPH (n = 7, systolic pulmonary artery pressure > 60 mm Hg) had a mortality rate of 42% at 9 months post-transplantation and 71% at 36 months post-transplant. The surviving patients continued to deteriorate with progressive right heart failure and no improvement in POPH.

To understand the effect of liver transplantation on POPH, one must understand the hemodynamic changes that occur with POPH and during liver transplant. The right ventricle is able to manage the same volume as the left ventricle under normal circumstances, but is unable to pump against a significant pressure gradient.³⁷ In the setting of POPH, right ventricular hypertrophy occurs and RV output remains stable for some time. With time, pulmonary artery pressure increases secondary to pulmonary arteriolar vasoconstriction, intimal thickening, and progressive occlusion of the pulmonary vascular bed. Right ventricular failure may occur as a result. Cardiac output increases significantly at the time of reperfusion during liver transplant (up to 3-fold in 15 minutes),³⁸ and in the setting of a noncompliant vascular bed, the patient is at risk for right heart failure. This is the likely explanation to such high perioperative mortality rates in patients with uncontrolled POPH. Failure to decrease MPAP to less than 50 mm Hg is considered a complete contraindication to liver transplant at most institutions. Many transplant centers will list patients for liver transplant if MPAP can be decreased to less than 35 mm Hg and PVR < 400 dynes/s/cm^–5. These parameters are thought to represent an adequate right ventricular reserve and a compliant pulmonary vascular bed.³⁷ However, even with good pressure control, the anesthesiology and critical care teams must be prepared to deal with acute right heart failure peri-operatively. Intraoperative transesophageal echocardiography has been recommended to closely follow right ventricular function.³⁸ Inhaled or intravenous dilators are the most effective agents in the event of a pulmonary hypertensive crisis.

Review of Outcomes

A retrospective review evaluated 43 patients with untreated POPH who underwent attempted liver transplantation.³⁹ Data were collected from 18 peer-reviewed studies and 7 patients at the authors’ institution. Overall mortality was 35% (15 patients), with almost all of the deaths secondary to cardiac dysfunction. Two deaths occurred intraoperatively and 8 deaths occurred during the transplant hospitalization. The transplant could not be successfully completed in 4 of the patients. MPAP > 50 mm Hg was associated with 100% mortality, whereas patients with MPAP between 35 mm Hg and 50 mm Hg had a 50% mortality. No mortality was noted in patients with MPAP < 35 mm Hg.

Liver transplantation has been shown to be successful in patients with controlled POPH. Sussman et al published their data on 8 patients with severe POPH in 2006. In this prospective study, all patients were treated with sequential epoprostenol infusions and 7 of the 8 patients experienced a significant reduction in MPAP and PVR. Six patients were listed for liver transplant, 4 of who were transplanted successfully and alive up to 5 years later.

The Baylor University Medical Center published their data on POPH patients who received liver transplants in 2007.¹¹ POPH was confirmed by right heart catheterization in 30 patients evaluated for liver transplant. Sixteen patients were considered to be suitable candidates for transplant and MPAP was decreased to less than 35 mmHg in 12 patients with vasodilator therapy. Eleven patients eventually underwent liver transplant and 1- and 5-year survival rates were 91% and 67%.

Compared to medical therapy or liver transplant alone, patients who receive medical therapy followed by liver transplantation have the best survival. The Mayo Clinic retrospectively reviewed 74 POPH patients identified between 1994 and 2007.⁴⁰ Patients were categorized in 1 of 3 categories: no medical therapy, medical therapy alone for POPH, or medical therapy for POPH followed by liver transplantation. Patients who received no medical therapy for POPH and no liver transplant had the worst outcomes, with a dismal 5-year survival of only 14% with over 50% deceased at 1 year of diagnosis. Five-year survival was 45% in patients who received medical therapy only. Patients who received medical therapy with prostacyclin followed by liver transplantation had the best outcomes, with a 5-year survival of 67% versus 25% in those who were transplanted without prior prostacyclin therapy.

We reported the longest follow-up study for patients undergoing liver transplantation with POPH in 2014.⁴¹ Seven patients with moderate to severe POPH received a liver transplant at our institution between June 2004 and January 2011. Mean pulmonary artery pressure was reduced to < 35 mm Hg, with appropriate POPH therapy in all of the patients. Both the graft and patient survival rates were 85.7% after a median follow-up of 7.8 years. The 1 patient who did not survive died from complications related to recurrent hepatitis C and cirrhosis, not from POPH-related issues. Four of the remaining 6 patients continue to require oral vasodilator therapy post-transplant, suggesting irreversible remodeling of the pulmonary vasculature. Two patients (4.4 and 8.5 years post-transplant) have no evidence of pulmonary hypertension post-transplant and therefore do not require medical treatment for pulmonary hypertension. We concluded that POPH responsive to vasodilator therapy is an appropriate indication for liver transplant, with excellent long-term survival.

Hollatz et al published their data on 11 patients with moderate to severe POPH who were successfully treated (mostly with oral sildenafil and subcutaneous treprostinil) as a bridge to liver transplant.⁴² The mortality rate was 0, with a follow-up duration of 7 to 60 months. Interestingly, 7 of the 11 patients (64%) were off all pulmonary vasodilators post-transplant. Ashfaq et al reported similar results.¹¹ Nine of 11 patients with treated moderate to severe POPH who received liver transplants stopped vasodilator therapy at a median period of 9.2 months post-transplant. Raevens et al described a group of 3 patients with POPH who went on to liver transplant after their pulmonary pressures were decreased with combined oral vasodilator therapy: 1 required continued long-term vasodilator therapy, another was weaned off medications after transplant, and the third patient died during the liver transplant from perioperative complications that induced uncontrolled pulmonary hypertension.⁴³

Patient Selection

In 2006, the United Network for Organ Sharing (UNOS) initiated a policy whereby a higher priority for liver transplantation was granted for highly selected patients in the United States.⁴⁴ UNOS policy 3.6.4.5.6 upgraded POPH patients to a MELD score of 22, with an increase in MELD every 3 months as long as MPAP remained < 35 mm Hg and PVR remained < 400 dynes/s/cm^–5. One hundred fifty-five patients were granted MELD exception points for POPH between 2002 and 2010 and went on to receive liver transplants.⁴⁵ Goldberg et al collected data from the Organ Procurement and Transplantation Network (OPTN) and compared outcomes of patients with approved POPH MELD exception points versus waitlist candidates with no exception points.⁴⁶ One hundred fifty-five waitlisted patients received POPH MELD exception points, with only 43.1% meeting OPTN exception requirements. One-third did not fulfill hemodynamic criteria consistent with POPH or had missing data, and 80% went on to receive a liver transplant. Waitlist candidates receiving POPH MELD exception points also had increased waitlist mortality and several early post-transplant deaths. The authors felt these data highlighted the need for OPTN/UNOS to revise their policy for POPH MELD exceptions points, revise how points are rewarded, and continue research to help risk stratify these patients to minimize perioperative complications.

Conclusion

Several effective medical treatment regimens are available, including prostanoids, endothelin receptor antagonists, and PDE-5 inhibitors. Liver transplantation is a potential cure but is only recommended if MPAP can be decreased to ≤ 35 mmHg. Long-term follow-up studies have shown these patients do well several years post-transplant but may continue to require oral therapy for their POPH.

Pulmonary arterial hypertension (PAH) is a rare disease that is associated with high mortality and is characterized by pulmonary vascular remodeling. Portopulmonary hypertension (POPH) is a form of PAH that occurs in patients with portal hypertension where no alternative cause of PAH can be identified. POPH is documented in approximately 4.5% to 8.5% of liver transplant candidates,^1,2 but there is no relationship between the existence or severity of POPH and the severity of liver dysfunction.³ Mantz and Craig described the first case of POPH in a 53-year-old woman with enlarged pulmonary arteries that exhibited forceful pulsations more characteristic of the aorta than a low-pressure pulmonary trunk.⁴ Autopsy revealed findings of chronic liver disease including a stenotic portal vein, portocaval shunt, and esophageal varices. In both PAH and POPH, pre-capillary pulmonary arteries have characteristic lesions, such as intimal thickening, endothelial proliferation, and thrombotic changes. This 2-part article reviews the diagnosis and treatment of patients with POPH. Here, we review the epidemiology, prognosis, pathogenesis, and diagnosis of POPH; current treatment options for POPH are reviewed in a separate article.

Definition

The term POPH was first used by Yoshida et al in 1993 to describe the first successful liver transplant in a patient with POPH, a 39-year-old man with chronic hepatitis.⁵ The World Health Organization (WHO) classifies POPH as a form of Group 1 PAH.⁶ The criteria that must be met to make a diagnosis of POPH are shown in the Table 1.⁷

Moderate POPH is defined as a mean pulmonary artery pressure (MPAP) between 35 mm Hg and < 45 mm Hg, whereas severe POPH is MPAP ≥ 45 mm Hg. Moderate and severe POPH are considered contraindications to liver transplant because of high perioperative and postoperative mortality rates.⁸ In 2000, the Mayo Clinic retrospectively reviewed 43 patients with POPH who underwent attempted liver transplantation.⁹ The cardiopulmonary-related mortality rate in patients with a MPAP of 35 to < 50 mm Hg was 50% and 100% for those with MPAP > 50 mm Hg. No mortality was noted in patients with a pre-liver transplant MPAP of < 35 mm Hg and transpulmonary gradient (TPG) < 15 mm Hg.

Epidemiology

In 1983, a series of 17,901 autopsied patients showed a primary pulmonary hypertension prevalence of 0.13% and a prevalence of 0.73% in patients with cirrhosis.¹⁰ In 1987, Rich et al published data from the National Institutes of Health’s national registry of primary pulmonary hypertension.¹¹ The registry included data from 187 patients from 32 centers. Further analyses by Groves et al concluded that 8.3% of the patients likely had POPH.¹² Humbert et al published data on the French pulmonary hypertension registry experience in 2006.¹³ The French registry included 674 patients from 17 university hospitals; 10.4% of these patients had POPH. The largest prospective study was published by Hadengue et al in 1991.¹⁴ In this study, 507 patients hospitalized with portal hypertension but without known pulmonary hypertension underwent cardiac catheterization; 10 patients (2%) had pulmonary hypertension and more than half were clinically asymptomatic. Finally, the Registry to Evaluate Early And Long-term pulmonary arterial hypertension disease management (REVEAL registry) documented a 5.3% frequency of POPH (174 of 3525) in the United States.¹⁵

Prognosis

Individuals with POPH have worse outcomes compared to other forms of PAH. Median survival prior to the introduction of vasodilator therapy was a dismal 6 months and mean survival was 15 months.¹⁶ The cause of death in patients with POPH is equally distributed between right heart failure from POPH and direct complications of chronic liver disease.¹ Le Pavec et al retrospectively analyzed all patients referred to the French Referral Center with POPH between 1984 and 2004 (154 patients).¹ Approximately 50% of the patients were Child-Turcotte-Pugh class B or C, and 60% were classified as New York Health Association (NYHA) class III or IV. In these patients, 1-, 3-, and 5-year survival rates were 88%, 75%, and 68%, respectively. Major independent prognostic risk factors were presence and severity of cirrhosis and preservation of right ventricular function. Interestingly, NYHA functional class was not related to survival in this study, although it has clearly been identified as a strong prognostic factor in idiopathic PAH.

Krowka et al evaluated 174 patients with POPH enrolled in the REVEAL Registry,¹⁵ a multicenter, observational, US-based study comprised of more than 3500 patients with PAH. Despite having better hemodynamic parameters at diagnosis, patients with POPH had significantly poorer survival and all-cause hospitalization compared with patients with idiopathic PAH (IPAH) or hereditary PAH (HPAH). Two-year survival from enrollment was 67% in POPH versus 85% in those with IPAH/HPAH (P < 0.001). Five-year survival from time of diagnosis was 40% versus 64% (P < 0.001). Additionally, patients with POPH were less likely to be on PAH-specific therapy at enrollment, with only 25% on treatment at the time of entry. These findings were replicated in 2005 when Kawut et al retrospectively compared 13 patients with POPH with 33 patients with IPAH.¹⁷ Despite having a higher cardiac index and lower pulmonary vascular resistance than patients with IPAH, patients with POPH had a higher risk of death (hazard ratio, 2.8, P = 0.04), likely reflecting the combination of 2 serious diseases.

In 2008 the Mayo Clinic published their retrospective analysis of patients with POPH to determine the natural history of POPH.¹⁸ Patients were categorized into 3 groups: (1) no medical therapy for POPH and no liver transplant; (2) medical therapy for POPH alone; (3) medical therapy for POPH followed by liver transplant. The study included 74 patients between 1994 through 2007; 19 patients who did not receive treatment for POPH or liver transplant truly represented the natural history of POPH. Their 5-year survival was only 14%, and over half were deceased 1 year after diagnosis. The largest group consisted of patients who received therapy for POPH but no liver transplant. This group did remarkably better than those who received no therapy at all, with a 5-year survival of 45%. However, the patients with the overall best survival were those who received a combination of treatment for POPH followed by liver transplant. Their 5-year survival was 67%. Survival at 5 years was only 25% for the small group of patients who received transplant without PAH therapy. Once again, mortality did not correlate with the severity of hepatic dysfunction or baseline hemodynamic data.

Pathogenesis

The pathogenesis of POPH is unclear. Multiple studies have shown that there is minimal, if any, association with pulmonary hypertension and the severity of liver disease or portal hypertension.^19,20 Portal hypertension is the result of an increase in intrahepatic resistance and an increase in blood flow into the portal circulation. Collateral vessels develop and blood from the splanchnic circulation is allowed directly into the systemic venous circulation, bypassing the liver. One of the most widely accepted theories is that a humoral substance, that would otherwise be metabolized by the liver, is able to reach the pulmonary circulation through collaterals, resulting in POPH.²¹ Pelicelli et al evaluated the possible role of endothelin-1, interleukin-6, interleukin 1β, and tumor necrosis factor in the pathogenesis of POPH.²² Plasma concentrations of these cytokines were compared between patients with POPH and patients with cirrhosis but no POPH. Patients with POPH had higher concentrations of endothelin-1 and interleukin-6, suggesting antagonists for these cytokines may have a role in the treatment of POPH. The role of endothelin-1 was further supported by Kamath et al in 2000²³ when they determined the pulmonary vascular bed is exposed to increased levels of circulating endothelin-1a in the setting of cirrhosis. Endothelin-1 is a potent vasoconstrictor and facilitator of smooth muscle proliferation.

In addition to collateral circulation allowing mediators to reach the pulmonary arterial bed in portal hypertension, high flow may trigger a vasoproliferative process in the pulmonary vascular bed. Patients with advanced liver disease have a low systemic vascular resistance, with a compensatory increase in cardiac output. An increase in cardiac output can lead to shear stress of the pulmonary vascular endothelial layer. Although the resistance of the pulmonary vasculature may decrease rapidly to help normalize pulmonary pressures, persistent circulatory overload could result in irreversible vascular changes. Autopsy and lung explant studies show that POPH is characterized by obstructive and remodeling changes in the pulmonary arterial bed.²⁴ Initially, medial hypertrophy with smooth muscle proliferation is present. As the disease advances, platelet aggregates, in situ thrombosis, and intimal fibrosis develop. Finally, web-like lesions involving the entire pulmonary wall develop with recanalization for the passage of pulmonary arterial flow. These changes are identical to the changes observed in patients with other forms of PAH.

Not all patients with portal hypertension develop POPH, suggesting that genetic predisposition may play a role in POPH development. The Pulmonary Vascular Complications of Liver Study Group published a multicenter case-control study that attempted to identify genetic risk factors for POPH in patients with advanced liver disease.²⁵ More than 1000 common single nucleotide polymorphisms (SNPs) in 93 candidate genes were genotyped in each patient. When compared to controls, multiple SNPs in the genes coding for estrogen receptor 1, aromatase, phosphodiesterase 5, angiopoietin 1, and calcium binding protein A4 were associated with an increased risk of POPH. One year earlier, the same study group concluded that female sex (adjusted odds ratio [OR], 2.90) and autoimmune hepatitis (adjusted OR, 4.02) were associated with a higher risk for POPH, whereas hepatitis C was associated with a decreased risk.²⁰

Clinical Presentation

Clinical presentation is variable in POPH. Patients referred to a pulmonologist will usually present with symptoms similar to patients with other forms of PAH. In a retrospective analysis of patients referred to the French Referral Center for Pulmonary Hypertension, 60% of the patients belonged to NYHA functional class III or IV.¹ In a series of 78 patients with POPH, the most common presenting pulmonary symptom was dyspnea on exertion (81%), followed by syncope, chest pain, and fatigue (< 33%).¹⁶ Symptoms such as syncope and chest pain are usually markers of severe POPH.³ Stigmata of portal hypertension, such as ascites, spider angiomata, and palmar erythema, may be present on exam. An accentuated pulmonary component of the second heart sound can be seen in 82% of patients and a systolic murmur caused by tricuspid regurgitation in 69% of patients.¹⁶ Patients with severe POPH may have jugular vein distention, peripheral edema, and a third heart sound.

Diagnostic Evaluation

Chest x-rays may show prominent pulmonary arteries and cardiomegaly in patients with POPH, whereas electrocardiogram can suggest right ventricular hypertrophy and right axis deviation. The best screening test for POPH in patients with portal hypertension is echocardiography. Routine screening for POPH is recommended during liver transplant evaluation in the practice guidelines from the American Association for the Study of Liver Disease.²⁶ Right-sided cardiac chamber enlargement and right ventricular pressure or volume overload can be assessed on echocardiography. Colle et al followed 165 patients evaluated for liver transplantation who underwent transthoracic Doppler echocardiography and right heart catheterization.²⁷ Seventeen patients met the criteria for POPH on echocardiography (presence of tricuspid regurgitation and calculated systolic pulmonary artery pressure over 30 mm Hg) and right heart catheterization confirmed the diagnosis in 10 patients. Right ventricular systolic pressure (RVSP) estimate of ≤ 30 mm Hg on 2-dimensional echo had a 100% sensitivity and negative predictive value. Positive predictive value was poor at 59%, reiterating the need for right heart catheterization in the diagnosis of POPH. When Kim et al used a RVSP threshold of 50 mm Hg, 72% had at least moderate pulmonary hypertension, including 30% with severe pulmonary hypertension.²⁸ Raevens et al analyzed data from 152 patients who underwent pretransplant echocardiography and catheterization.² Their data show a RVSP threshold of greater than 38 mm Hg by echocardiography had a specificity of 82% and sensitivity and negative predictive value of 100%. The European Respiratory Society recommendations state that PAH should be considered unlikely if echocardiography estimates a RVSP ≤36 mm Hg and likely if the RVSP is estimated at > 50 mm Hg.²⁹ We recommend repeating echocardiography every 6 to 12 months in patients listed for liver transplantation, as pulmonary hemodynamics may change over time.

Computed tomography (CT) may have a complementary role in the future for the noninvasive detection of POPH. In a study published in 2014, 49 patients referred for liver transplantation were retrospectively reviewed.³⁰ Measured CT signs included the main pulmonary artery/ascending aorta diameter ratio, the mean left and right main pulmonary artery diameter, and the enlargement of the pulmonary artery compared to the ascending aorta. Compared to the transthoracic echocardiography alone, an algorithm incorporating CT and echocardiography improved the detection of POPH (area under curve = 0.8, P < 0.0001).

A diagnosis of POPH can only be confirmed when PAH exists in a patient with portal hypertension, as determined by right heart catheterization, and no other cause of PAH can be identified. MPAP should be 25 mm Hg or greater, PVR of 240 dynes/s/cm^–5, wedge pressure of 15 mm Hg or less, and TPG greater than 12 mm Hg. Krowka et al showed the value of right heart catheterization in their 10-year prospective, echocardiography-catheterization algorithm study.¹⁹ Of 1235 liver transplant candidates who underwent echocardiography, 104 patients had a RVSP exceeding 50 mm Hg. Almost all of these patients had a right heart catheterization. All cause pulmonary hypertension (MPAP > 25 mm Hg) was confirmed in 90% of the patients, and 35% had a PVR < 240 dynes/s/cm^–5 and pulmonary capillary wedge pressure (PCWP) > 15 mm Hg, suggesting fluid overload. Forty-one patients had significant POPH, with a PVR > 400 dynes/s/cm^–5, and 24% also had an elevated PCWP. TPG was > 12 mm Hg in all of these patients, confirming POPH. As demonstrated by this study, right heart catheterization is required to confirm the diagnosis of POPH because high flow and fluid overload can lead to elevated pulmonary artery pressures.

Patients with POPH have a unique clinical profile with characteristics common to patients with primary pulmonary hypertension and chronic liver disease. In a retrospective review that compared 30 patients with PAH, 30 patients with chronic liver disease only, and 30 patients with catheterization-proved POPH,³¹ patients with POPH had elevated MPAP similar to those with primary PAH, but they also had reduced SVR and elevated cardiac index similar to those with chronic liver disease alone.

Besides POPH, 2 other common causes can lead to increased pulmonary arterial blood flow in patients with portal hypertension. First is a high-flow condition caused by increased cardiac output but with a normal PVR and PCWP. Fluid overload can also lead to pulmonary venous hypertension with increased PCWP, normal cardiac output, and normal PVR. Up to 25% of patients with POPH may present with marked excess volume caused by fluid retention.³ There can be an increase in both PCWP and PVR depending on the presence and the degree of fluid retention. TPG (MPAP – PCWP) > 12 mm Hg was introduced to make such patients less confusing and to help correct for increased PCWP secondary to fluid overload. Obstruction to pulmonary arterial flow is manifest by an increased TPG (Table 2).

POPH should be distinguished from hepatopulmonary syndrome (HPS), which is another pulmonary vascular consequence of liver disease. Unlike POPH, HPS is characterized by a defect in arterial oxygenation induced by pulmonary vascular dilation.³² Similar to other patients with liver disease, patients with HPS have a normal PVR and increased cardiac output secondary to a high-flow state. HPS is diagnosed by confirmation of an intrapulmonary shunt by echocardiogram. Injection of agitated saline results in saline bubbles being visualized in the left atrium 3 or more cardiac cycles after they appear in the right atrium. Currently, there is no effective medical treatment for HPS and liver transplantation is the only successful treatment.

Conclusion

POPH is an uncommon complication of chronic liver disease. It is defined as PAH in a patient with portal hypertension excluding other causes of PAH. The following criteria must be met to make a diagnosis of POPH: (1) evidence of portal hypertension; (2) MPAP ≥ 35 mm Hg; (3) PVR ≥ 240 dynes/s/cm⁵; (4) pulmonary capillary wedge pressure ≤ 15 mm Hg; and (5) TPG > 12 mm Hg. Individuals with POPH have worse outcomes compared to other forms of PAH, with a median survival of 6 months without medical therapy. The pathogenesis of POPH is unclear but may be related to a genetic predisposition since not all patients with portal hypertension develop POPH. Echocardiography is an excellent screening test for POPH, but a right heart catheterization must be performed to confirm the diagnosis.

Pulmonary arterial hypertension (PAH) is a rare disease that is associated with high mortality and is characterized by pulmonary vascular remodeling. Portopulmonary hypertension (POPH) is a form of PAH that occurs in patients with portal hypertension where no alternative cause of PAH can be identified. POPH is documented in approximately 4.5% to 8.5% of liver transplant candidates,^1,2 but there is no relationship between the existence or severity of POPH and the severity of liver dysfunction.³ Mantz and Craig described the first case of POPH in a 53-year-old woman with enlarged pulmonary arteries that exhibited forceful pulsations more characteristic of the aorta than a low-pressure pulmonary trunk.⁴ Autopsy revealed findings of chronic liver disease including a stenotic portal vein, portocaval shunt, and esophageal varices. In both PAH and POPH, pre-capillary pulmonary arteries have characteristic lesions, such as intimal thickening, endothelial proliferation, and thrombotic changes. This 2-part article reviews the diagnosis and treatment of patients with POPH. Here, we review the epidemiology, prognosis, pathogenesis, and diagnosis of POPH; current treatment options for POPH are reviewed in a separate article.

Definition

The term POPH was first used by Yoshida et al in 1993 to describe the first successful liver transplant in a patient with POPH, a 39-year-old man with chronic hepatitis.⁵ The World Health Organization (WHO) classifies POPH as a form of Group 1 PAH.⁶ The criteria that must be met to make a diagnosis of POPH are shown in the Table 1.⁷

Moderate POPH is defined as a mean pulmonary artery pressure (MPAP) between 35 mm Hg and < 45 mm Hg, whereas severe POPH is MPAP ≥ 45 mm Hg. Moderate and severe POPH are considered contraindications to liver transplant because of high perioperative and postoperative mortality rates.⁸ In 2000, the Mayo Clinic retrospectively reviewed 43 patients with POPH who underwent attempted liver transplantation.⁹ The cardiopulmonary-related mortality rate in patients with a MPAP of 35 to < 50 mm Hg was 50% and 100% for those with MPAP > 50 mm Hg. No mortality was noted in patients with a pre-liver transplant MPAP of < 35 mm Hg and transpulmonary gradient (TPG) < 15 mm Hg.

Epidemiology

In 1983, a series of 17,901 autopsied patients showed a primary pulmonary hypertension prevalence of 0.13% and a prevalence of 0.73% in patients with cirrhosis.¹⁰ In 1987, Rich et al published data from the National Institutes of Health’s national registry of primary pulmonary hypertension.¹¹ The registry included data from 187 patients from 32 centers. Further analyses by Groves et al concluded that 8.3% of the patients likely had POPH.¹² Humbert et al published data on the French pulmonary hypertension registry experience in 2006.¹³ The French registry included 674 patients from 17 university hospitals; 10.4% of these patients had POPH. The largest prospective study was published by Hadengue et al in 1991.¹⁴ In this study, 507 patients hospitalized with portal hypertension but without known pulmonary hypertension underwent cardiac catheterization; 10 patients (2%) had pulmonary hypertension and more than half were clinically asymptomatic. Finally, the Registry to Evaluate Early And Long-term pulmonary arterial hypertension disease management (REVEAL registry) documented a 5.3% frequency of POPH (174 of 3525) in the United States.¹⁵

Prognosis

Individuals with POPH have worse outcomes compared to other forms of PAH. Median survival prior to the introduction of vasodilator therapy was a dismal 6 months and mean survival was 15 months.¹⁶ The cause of death in patients with POPH is equally distributed between right heart failure from POPH and direct complications of chronic liver disease.¹ Le Pavec et al retrospectively analyzed all patients referred to the French Referral Center with POPH between 1984 and 2004 (154 patients).¹ Approximately 50% of the patients were Child-Turcotte-Pugh class B or C, and 60% were classified as New York Health Association (NYHA) class III or IV. In these patients, 1-, 3-, and 5-year survival rates were 88%, 75%, and 68%, respectively. Major independent prognostic risk factors were presence and severity of cirrhosis and preservation of right ventricular function. Interestingly, NYHA functional class was not related to survival in this study, although it has clearly been identified as a strong prognostic factor in idiopathic PAH.

Krowka et al evaluated 174 patients with POPH enrolled in the REVEAL Registry,¹⁵ a multicenter, observational, US-based study comprised of more than 3500 patients with PAH. Despite having better hemodynamic parameters at diagnosis, patients with POPH had significantly poorer survival and all-cause hospitalization compared with patients with idiopathic PAH (IPAH) or hereditary PAH (HPAH). Two-year survival from enrollment was 67% in POPH versus 85% in those with IPAH/HPAH (P < 0.001). Five-year survival from time of diagnosis was 40% versus 64% (P < 0.001). Additionally, patients with POPH were less likely to be on PAH-specific therapy at enrollment, with only 25% on treatment at the time of entry. These findings were replicated in 2005 when Kawut et al retrospectively compared 13 patients with POPH with 33 patients with IPAH.¹⁷ Despite having a higher cardiac index and lower pulmonary vascular resistance than patients with IPAH, patients with POPH had a higher risk of death (hazard ratio, 2.8, P = 0.04), likely reflecting the combination of 2 serious diseases.

In 2008 the Mayo Clinic published their retrospective analysis of patients with POPH to determine the natural history of POPH.¹⁸ Patients were categorized into 3 groups: (1) no medical therapy for POPH and no liver transplant; (2) medical therapy for POPH alone; (3) medical therapy for POPH followed by liver transplant. The study included 74 patients between 1994 through 2007; 19 patients who did not receive treatment for POPH or liver transplant truly represented the natural history of POPH. Their 5-year survival was only 14%, and over half were deceased 1 year after diagnosis. The largest group consisted of patients who received therapy for POPH but no liver transplant. This group did remarkably better than those who received no therapy at all, with a 5-year survival of 45%. However, the patients with the overall best survival were those who received a combination of treatment for POPH followed by liver transplant. Their 5-year survival was 67%. Survival at 5 years was only 25% for the small group of patients who received transplant without PAH therapy. Once again, mortality did not correlate with the severity of hepatic dysfunction or baseline hemodynamic data.

Pathogenesis

The pathogenesis of POPH is unclear. Multiple studies have shown that there is minimal, if any, association with pulmonary hypertension and the severity of liver disease or portal hypertension.^19,20 Portal hypertension is the result of an increase in intrahepatic resistance and an increase in blood flow into the portal circulation. Collateral vessels develop and blood from the splanchnic circulation is allowed directly into the systemic venous circulation, bypassing the liver. One of the most widely accepted theories is that a humoral substance, that would otherwise be metabolized by the liver, is able to reach the pulmonary circulation through collaterals, resulting in POPH.²¹ Pelicelli et al evaluated the possible role of endothelin-1, interleukin-6, interleukin 1β, and tumor necrosis factor in the pathogenesis of POPH.²² Plasma concentrations of these cytokines were compared between patients with POPH and patients with cirrhosis but no POPH. Patients with POPH had higher concentrations of endothelin-1 and interleukin-6, suggesting antagonists for these cytokines may have a role in the treatment of POPH. The role of endothelin-1 was further supported by Kamath et al in 2000²³ when they determined the pulmonary vascular bed is exposed to increased levels of circulating endothelin-1a in the setting of cirrhosis. Endothelin-1 is a potent vasoconstrictor and facilitator of smooth muscle proliferation.

In addition to collateral circulation allowing mediators to reach the pulmonary arterial bed in portal hypertension, high flow may trigger a vasoproliferative process in the pulmonary vascular bed. Patients with advanced liver disease have a low systemic vascular resistance, with a compensatory increase in cardiac output. An increase in cardiac output can lead to shear stress of the pulmonary vascular endothelial layer. Although the resistance of the pulmonary vasculature may decrease rapidly to help normalize pulmonary pressures, persistent circulatory overload could result in irreversible vascular changes. Autopsy and lung explant studies show that POPH is characterized by obstructive and remodeling changes in the pulmonary arterial bed.²⁴ Initially, medial hypertrophy with smooth muscle proliferation is present. As the disease advances, platelet aggregates, in situ thrombosis, and intimal fibrosis develop. Finally, web-like lesions involving the entire pulmonary wall develop with recanalization for the passage of pulmonary arterial flow. These changes are identical to the changes observed in patients with other forms of PAH.

Not all patients with portal hypertension develop POPH, suggesting that genetic predisposition may play a role in POPH development. The Pulmonary Vascular Complications of Liver Study Group published a multicenter case-control study that attempted to identify genetic risk factors for POPH in patients with advanced liver disease.²⁵ More than 1000 common single nucleotide polymorphisms (SNPs) in 93 candidate genes were genotyped in each patient. When compared to controls, multiple SNPs in the genes coding for estrogen receptor 1, aromatase, phosphodiesterase 5, angiopoietin 1, and calcium binding protein A4 were associated with an increased risk of POPH. One year earlier, the same study group concluded that female sex (adjusted odds ratio [OR], 2.90) and autoimmune hepatitis (adjusted OR, 4.02) were associated with a higher risk for POPH, whereas hepatitis C was associated with a decreased risk.²⁰

Clinical Presentation

Clinical presentation is variable in POPH. Patients referred to a pulmonologist will usually present with symptoms similar to patients with other forms of PAH. In a retrospective analysis of patients referred to the French Referral Center for Pulmonary Hypertension, 60% of the patients belonged to NYHA functional class III or IV.¹ In a series of 78 patients with POPH, the most common presenting pulmonary symptom was dyspnea on exertion (81%), followed by syncope, chest pain, and fatigue (< 33%).¹⁶ Symptoms such as syncope and chest pain are usually markers of severe POPH.³ Stigmata of portal hypertension, such as ascites, spider angiomata, and palmar erythema, may be present on exam. An accentuated pulmonary component of the second heart sound can be seen in 82% of patients and a systolic murmur caused by tricuspid regurgitation in 69% of patients.¹⁶ Patients with severe POPH may have jugular vein distention, peripheral edema, and a third heart sound.

Diagnostic Evaluation

Chest x-rays may show prominent pulmonary arteries and cardiomegaly in patients with POPH, whereas electrocardiogram can suggest right ventricular hypertrophy and right axis deviation. The best screening test for POPH in patients with portal hypertension is echocardiography. Routine screening for POPH is recommended during liver transplant evaluation in the practice guidelines from the American Association for the Study of Liver Disease.²⁶ Right-sided cardiac chamber enlargement and right ventricular pressure or volume overload can be assessed on echocardiography. Colle et al followed 165 patients evaluated for liver transplantation who underwent transthoracic Doppler echocardiography and right heart catheterization.²⁷ Seventeen patients met the criteria for POPH on echocardiography (presence of tricuspid regurgitation and calculated systolic pulmonary artery pressure over 30 mm Hg) and right heart catheterization confirmed the diagnosis in 10 patients. Right ventricular systolic pressure (RVSP) estimate of ≤ 30 mm Hg on 2-dimensional echo had a 100% sensitivity and negative predictive value. Positive predictive value was poor at 59%, reiterating the need for right heart catheterization in the diagnosis of POPH. When Kim et al used a RVSP threshold of 50 mm Hg, 72% had at least moderate pulmonary hypertension, including 30% with severe pulmonary hypertension.²⁸ Raevens et al analyzed data from 152 patients who underwent pretransplant echocardiography and catheterization.² Their data show a RVSP threshold of greater than 38 mm Hg by echocardiography had a specificity of 82% and sensitivity and negative predictive value of 100%. The European Respiratory Society recommendations state that PAH should be considered unlikely if echocardiography estimates a RVSP ≤36 mm Hg and likely if the RVSP is estimated at > 50 mm Hg.²⁹ We recommend repeating echocardiography every 6 to 12 months in patients listed for liver transplantation, as pulmonary hemodynamics may change over time.

Computed tomography (CT) may have a complementary role in the future for the noninvasive detection of POPH. In a study published in 2014, 49 patients referred for liver transplantation were retrospectively reviewed.³⁰ Measured CT signs included the main pulmonary artery/ascending aorta diameter ratio, the mean left and right main pulmonary artery diameter, and the enlargement of the pulmonary artery compared to the ascending aorta. Compared to the transthoracic echocardiography alone, an algorithm incorporating CT and echocardiography improved the detection of POPH (area under curve = 0.8, P < 0.0001).

A diagnosis of POPH can only be confirmed when PAH exists in a patient with portal hypertension, as determined by right heart catheterization, and no other cause of PAH can be identified. MPAP should be 25 mm Hg or greater, PVR of 240 dynes/s/cm^–5, wedge pressure of 15 mm Hg or less, and TPG greater than 12 mm Hg. Krowka et al showed the value of right heart catheterization in their 10-year prospective, echocardiography-catheterization algorithm study.¹⁹ Of 1235 liver transplant candidates who underwent echocardiography, 104 patients had a RVSP exceeding 50 mm Hg. Almost all of these patients had a right heart catheterization. All cause pulmonary hypertension (MPAP > 25 mm Hg) was confirmed in 90% of the patients, and 35% had a PVR < 240 dynes/s/cm^–5 and pulmonary capillary wedge pressure (PCWP) > 15 mm Hg, suggesting fluid overload. Forty-one patients had significant POPH, with a PVR > 400 dynes/s/cm^–5, and 24% also had an elevated PCWP. TPG was > 12 mm Hg in all of these patients, confirming POPH. As demonstrated by this study, right heart catheterization is required to confirm the diagnosis of POPH because high flow and fluid overload can lead to elevated pulmonary artery pressures.

Patients with POPH have a unique clinical profile with characteristics common to patients with primary pulmonary hypertension and chronic liver disease. In a retrospective review that compared 30 patients with PAH, 30 patients with chronic liver disease only, and 30 patients with catheterization-proved POPH,³¹ patients with POPH had elevated MPAP similar to those with primary PAH, but they also had reduced SVR and elevated cardiac index similar to those with chronic liver disease alone.

Besides POPH, 2 other common causes can lead to increased pulmonary arterial blood flow in patients with portal hypertension. First is a high-flow condition caused by increased cardiac output but with a normal PVR and PCWP. Fluid overload can also lead to pulmonary venous hypertension with increased PCWP, normal cardiac output, and normal PVR. Up to 25% of patients with POPH may present with marked excess volume caused by fluid retention.³ There can be an increase in both PCWP and PVR depending on the presence and the degree of fluid retention. TPG (MPAP – PCWP) > 12 mm Hg was introduced to make such patients less confusing and to help correct for increased PCWP secondary to fluid overload. Obstruction to pulmonary arterial flow is manifest by an increased TPG (Table 2).

POPH should be distinguished from hepatopulmonary syndrome (HPS), which is another pulmonary vascular consequence of liver disease. Unlike POPH, HPS is characterized by a defect in arterial oxygenation induced by pulmonary vascular dilation.³² Similar to other patients with liver disease, patients with HPS have a normal PVR and increased cardiac output secondary to a high-flow state. HPS is diagnosed by confirmation of an intrapulmonary shunt by echocardiogram. Injection of agitated saline results in saline bubbles being visualized in the left atrium 3 or more cardiac cycles after they appear in the right atrium. Currently, there is no effective medical treatment for HPS and liver transplantation is the only successful treatment.

Conclusion

POPH is an uncommon complication of chronic liver disease. It is defined as PAH in a patient with portal hypertension excluding other causes of PAH. The following criteria must be met to make a diagnosis of POPH: (1) evidence of portal hypertension; (2) MPAP ≥ 35 mm Hg; (3) PVR ≥ 240 dynes/s/cm⁵; (4) pulmonary capillary wedge pressure ≤ 15 mm Hg; and (5) TPG > 12 mm Hg. Individuals with POPH have worse outcomes compared to other forms of PAH, with a median survival of 6 months without medical therapy. The pathogenesis of POPH is unclear but may be related to a genetic predisposition since not all patients with portal hypertension develop POPH. Echocardiography is an excellent screening test for POPH, but a right heart catheterization must be performed to confirm the diagnosis.

Pulmonary arterial hypertension (PAH) is a rare disease that is associated with high mortality and is characterized by pulmonary vascular remodeling. Portopulmonary hypertension (POPH) is a form of PAH that occurs in patients with portal hypertension where no alternative cause of PAH can be identified. POPH is documented in approximately 4.5% to 8.5% of liver transplant candidates,^1,2 but there is no relationship between the existence or severity of POPH and the severity of liver dysfunction.³ Mantz and Craig described the first case of POPH in a 53-year-old woman with enlarged pulmonary arteries that exhibited forceful pulsations more characteristic of the aorta than a low-pressure pulmonary trunk.⁴ Autopsy revealed findings of chronic liver disease including a stenotic portal vein, portocaval shunt, and esophageal varices. In both PAH and POPH, pre-capillary pulmonary arteries have characteristic lesions, such as intimal thickening, endothelial proliferation, and thrombotic changes. This 2-part article reviews the diagnosis and treatment of patients with POPH. Here, we review the epidemiology, prognosis, pathogenesis, and diagnosis of POPH; current treatment options for POPH are reviewed in a separate article.

Definition

The term POPH was first used by Yoshida et al in 1993 to describe the first successful liver transplant in a patient with POPH, a 39-year-old man with chronic hepatitis.⁵ The World Health Organization (WHO) classifies POPH as a form of Group 1 PAH.⁶ The criteria that must be met to make a diagnosis of POPH are shown in the Table 1.⁷

Moderate POPH is defined as a mean pulmonary artery pressure (MPAP) between 35 mm Hg and < 45 mm Hg, whereas severe POPH is MPAP ≥ 45 mm Hg. Moderate and severe POPH are considered contraindications to liver transplant because of high perioperative and postoperative mortality rates.⁸ In 2000, the Mayo Clinic retrospectively reviewed 43 patients with POPH who underwent attempted liver transplantation.⁹ The cardiopulmonary-related mortality rate in patients with a MPAP of 35 to < 50 mm Hg was 50% and 100% for those with MPAP > 50 mm Hg. No mortality was noted in patients with a pre-liver transplant MPAP of < 35 mm Hg and transpulmonary gradient (TPG) < 15 mm Hg.

Epidemiology

In 1983, a series of 17,901 autopsied patients showed a primary pulmonary hypertension prevalence of 0.13% and a prevalence of 0.73% in patients with cirrhosis.¹⁰ In 1987, Rich et al published data from the National Institutes of Health’s national registry of primary pulmonary hypertension.¹¹ The registry included data from 187 patients from 32 centers. Further analyses by Groves et al concluded that 8.3% of the patients likely had POPH.¹² Humbert et al published data on the French pulmonary hypertension registry experience in 2006.¹³ The French registry included 674 patients from 17 university hospitals; 10.4% of these patients had POPH. The largest prospective study was published by Hadengue et al in 1991.¹⁴ In this study, 507 patients hospitalized with portal hypertension but without known pulmonary hypertension underwent cardiac catheterization; 10 patients (2%) had pulmonary hypertension and more than half were clinically asymptomatic. Finally, the Registry to Evaluate Early And Long-term pulmonary arterial hypertension disease management (REVEAL registry) documented a 5.3% frequency of POPH (174 of 3525) in the United States.¹⁵

Prognosis

Individuals with POPH have worse outcomes compared to other forms of PAH. Median survival prior to the introduction of vasodilator therapy was a dismal 6 months and mean survival was 15 months.¹⁶ The cause of death in patients with POPH is equally distributed between right heart failure from POPH and direct complications of chronic liver disease.¹ Le Pavec et al retrospectively analyzed all patients referred to the French Referral Center with POPH between 1984 and 2004 (154 patients).¹ Approximately 50% of the patients were Child-Turcotte-Pugh class B or C, and 60% were classified as New York Health Association (NYHA) class III or IV. In these patients, 1-, 3-, and 5-year survival rates were 88%, 75%, and 68%, respectively. Major independent prognostic risk factors were presence and severity of cirrhosis and preservation of right ventricular function. Interestingly, NYHA functional class was not related to survival in this study, although it has clearly been identified as a strong prognostic factor in idiopathic PAH.

Krowka et al evaluated 174 patients with POPH enrolled in the REVEAL Registry,¹⁵ a multicenter, observational, US-based study comprised of more than 3500 patients with PAH. Despite having better hemodynamic parameters at diagnosis, patients with POPH had significantly poorer survival and all-cause hospitalization compared with patients with idiopathic PAH (IPAH) or hereditary PAH (HPAH). Two-year survival from enrollment was 67% in POPH versus 85% in those with IPAH/HPAH (P < 0.001). Five-year survival from time of diagnosis was 40% versus 64% (P < 0.001). Additionally, patients with POPH were less likely to be on PAH-specific therapy at enrollment, with only 25% on treatment at the time of entry. These findings were replicated in 2005 when Kawut et al retrospectively compared 13 patients with POPH with 33 patients with IPAH.¹⁷ Despite having a higher cardiac index and lower pulmonary vascular resistance than patients with IPAH, patients with POPH had a higher risk of death (hazard ratio, 2.8, P = 0.04), likely reflecting the combination of 2 serious diseases.

In 2008 the Mayo Clinic published their retrospective analysis of patients with POPH to determine the natural history of POPH.¹⁸ Patients were categorized into 3 groups: (1) no medical therapy for POPH and no liver transplant; (2) medical therapy for POPH alone; (3) medical therapy for POPH followed by liver transplant. The study included 74 patients between 1994 through 2007; 19 patients who did not receive treatment for POPH or liver transplant truly represented the natural history of POPH. Their 5-year survival was only 14%, and over half were deceased 1 year after diagnosis. The largest group consisted of patients who received therapy for POPH but no liver transplant. This group did remarkably better than those who received no therapy at all, with a 5-year survival of 45%. However, the patients with the overall best survival were those who received a combination of treatment for POPH followed by liver transplant. Their 5-year survival was 67%. Survival at 5 years was only 25% for the small group of patients who received transplant without PAH therapy. Once again, mortality did not correlate with the severity of hepatic dysfunction or baseline hemodynamic data.

Pathogenesis

The pathogenesis of POPH is unclear. Multiple studies have shown that there is minimal, if any, association with pulmonary hypertension and the severity of liver disease or portal hypertension.^19,20 Portal hypertension is the result of an increase in intrahepatic resistance and an increase in blood flow into the portal circulation. Collateral vessels develop and blood from the splanchnic circulation is allowed directly into the systemic venous circulation, bypassing the liver. One of the most widely accepted theories is that a humoral substance, that would otherwise be metabolized by the liver, is able to reach the pulmonary circulation through collaterals, resulting in POPH.²¹ Pelicelli et al evaluated the possible role of endothelin-1, interleukin-6, interleukin 1β, and tumor necrosis factor in the pathogenesis of POPH.²² Plasma concentrations of these cytokines were compared between patients with POPH and patients with cirrhosis but no POPH. Patients with POPH had higher concentrations of endothelin-1 and interleukin-6, suggesting antagonists for these cytokines may have a role in the treatment of POPH. The role of endothelin-1 was further supported by Kamath et al in 2000²³ when they determined the pulmonary vascular bed is exposed to increased levels of circulating endothelin-1a in the setting of cirrhosis. Endothelin-1 is a potent vasoconstrictor and facilitator of smooth muscle proliferation.

In addition to collateral circulation allowing mediators to reach the pulmonary arterial bed in portal hypertension, high flow may trigger a vasoproliferative process in the pulmonary vascular bed. Patients with advanced liver disease have a low systemic vascular resistance, with a compensatory increase in cardiac output. An increase in cardiac output can lead to shear stress of the pulmonary vascular endothelial layer. Although the resistance of the pulmonary vasculature may decrease rapidly to help normalize pulmonary pressures, persistent circulatory overload could result in irreversible vascular changes. Autopsy and lung explant studies show that POPH is characterized by obstructive and remodeling changes in the pulmonary arterial bed.²⁴ Initially, medial hypertrophy with smooth muscle proliferation is present. As the disease advances, platelet aggregates, in situ thrombosis, and intimal fibrosis develop. Finally, web-like lesions involving the entire pulmonary wall develop with recanalization for the passage of pulmonary arterial flow. These changes are identical to the changes observed in patients with other forms of PAH.

Not all patients with portal hypertension develop POPH, suggesting that genetic predisposition may play a role in POPH development. The Pulmonary Vascular Complications of Liver Study Group published a multicenter case-control study that attempted to identify genetic risk factors for POPH in patients with advanced liver disease.²⁵ More than 1000 common single nucleotide polymorphisms (SNPs) in 93 candidate genes were genotyped in each patient. When compared to controls, multiple SNPs in the genes coding for estrogen receptor 1, aromatase, phosphodiesterase 5, angiopoietin 1, and calcium binding protein A4 were associated with an increased risk of POPH. One year earlier, the same study group concluded that female sex (adjusted odds ratio [OR], 2.90) and autoimmune hepatitis (adjusted OR, 4.02) were associated with a higher risk for POPH, whereas hepatitis C was associated with a decreased risk.²⁰

Clinical Presentation

Clinical presentation is variable in POPH. Patients referred to a pulmonologist will usually present with symptoms similar to patients with other forms of PAH. In a retrospective analysis of patients referred to the French Referral Center for Pulmonary Hypertension, 60% of the patients belonged to NYHA functional class III or IV.¹ In a series of 78 patients with POPH, the most common presenting pulmonary symptom was dyspnea on exertion (81%), followed by syncope, chest pain, and fatigue (< 33%).¹⁶ Symptoms such as syncope and chest pain are usually markers of severe POPH.³ Stigmata of portal hypertension, such as ascites, spider angiomata, and palmar erythema, may be present on exam. An accentuated pulmonary component of the second heart sound can be seen in 82% of patients and a systolic murmur caused by tricuspid regurgitation in 69% of patients.¹⁶ Patients with severe POPH may have jugular vein distention, peripheral edema, and a third heart sound.

Diagnostic Evaluation

Chest x-rays may show prominent pulmonary arteries and cardiomegaly in patients with POPH, whereas electrocardiogram can suggest right ventricular hypertrophy and right axis deviation. The best screening test for POPH in patients with portal hypertension is echocardiography. Routine screening for POPH is recommended during liver transplant evaluation in the practice guidelines from the American Association for the Study of Liver Disease.²⁶ Right-sided cardiac chamber enlargement and right ventricular pressure or volume overload can be assessed on echocardiography. Colle et al followed 165 patients evaluated for liver transplantation who underwent transthoracic Doppler echocardiography and right heart catheterization.²⁷ Seventeen patients met the criteria for POPH on echocardiography (presence of tricuspid regurgitation and calculated systolic pulmonary artery pressure over 30 mm Hg) and right heart catheterization confirmed the diagnosis in 10 patients. Right ventricular systolic pressure (RVSP) estimate of ≤ 30 mm Hg on 2-dimensional echo had a 100% sensitivity and negative predictive value. Positive predictive value was poor at 59%, reiterating the need for right heart catheterization in the diagnosis of POPH. When Kim et al used a RVSP threshold of 50 mm Hg, 72% had at least moderate pulmonary hypertension, including 30% with severe pulmonary hypertension.²⁸ Raevens et al analyzed data from 152 patients who underwent pretransplant echocardiography and catheterization.² Their data show a RVSP threshold of greater than 38 mm Hg by echocardiography had a specificity of 82% and sensitivity and negative predictive value of 100%. The European Respiratory Society recommendations state that PAH should be considered unlikely if echocardiography estimates a RVSP ≤36 mm Hg and likely if the RVSP is estimated at > 50 mm Hg.²⁹ We recommend repeating echocardiography every 6 to 12 months in patients listed for liver transplantation, as pulmonary hemodynamics may change over time.

Computed tomography (CT) may have a complementary role in the future for the noninvasive detection of POPH. In a study published in 2014, 49 patients referred for liver transplantation were retrospectively reviewed.³⁰ Measured CT signs included the main pulmonary artery/ascending aorta diameter ratio, the mean left and right main pulmonary artery diameter, and the enlargement of the pulmonary artery compared to the ascending aorta. Compared to the transthoracic echocardiography alone, an algorithm incorporating CT and echocardiography improved the detection of POPH (area under curve = 0.8, P < 0.0001).

A diagnosis of POPH can only be confirmed when PAH exists in a patient with portal hypertension, as determined by right heart catheterization, and no other cause of PAH can be identified. MPAP should be 25 mm Hg or greater, PVR of 240 dynes/s/cm^–5, wedge pressure of 15 mm Hg or less, and TPG greater than 12 mm Hg. Krowka et al showed the value of right heart catheterization in their 10-year prospective, echocardiography-catheterization algorithm study.¹⁹ Of 1235 liver transplant candidates who underwent echocardiography, 104 patients had a RVSP exceeding 50 mm Hg. Almost all of these patients had a right heart catheterization. All cause pulmonary hypertension (MPAP > 25 mm Hg) was confirmed in 90% of the patients, and 35% had a PVR < 240 dynes/s/cm^–5 and pulmonary capillary wedge pressure (PCWP) > 15 mm Hg, suggesting fluid overload. Forty-one patients had significant POPH, with a PVR > 400 dynes/s/cm^–5, and 24% also had an elevated PCWP. TPG was > 12 mm Hg in all of these patients, confirming POPH. As demonstrated by this study, right heart catheterization is required to confirm the diagnosis of POPH because high flow and fluid overload can lead to elevated pulmonary artery pressures.

Patients with POPH have a unique clinical profile with characteristics common to patients with primary pulmonary hypertension and chronic liver disease. In a retrospective review that compared 30 patients with PAH, 30 patients with chronic liver disease only, and 30 patients with catheterization-proved POPH,³¹ patients with POPH had elevated MPAP similar to those with primary PAH, but they also had reduced SVR and elevated cardiac index similar to those with chronic liver disease alone.

Besides POPH, 2 other common causes can lead to increased pulmonary arterial blood flow in patients with portal hypertension. First is a high-flow condition caused by increased cardiac output but with a normal PVR and PCWP. Fluid overload can also lead to pulmonary venous hypertension with increased PCWP, normal cardiac output, and normal PVR. Up to 25% of patients with POPH may present with marked excess volume caused by fluid retention.³ There can be an increase in both PCWP and PVR depending on the presence and the degree of fluid retention. TPG (MPAP – PCWP) > 12 mm Hg was introduced to make such patients less confusing and to help correct for increased PCWP secondary to fluid overload. Obstruction to pulmonary arterial flow is manifest by an increased TPG (Table 2).

POPH should be distinguished from hepatopulmonary syndrome (HPS), which is another pulmonary vascular consequence of liver disease. Unlike POPH, HPS is characterized by a defect in arterial oxygenation induced by pulmonary vascular dilation.³² Similar to other patients with liver disease, patients with HPS have a normal PVR and increased cardiac output secondary to a high-flow state. HPS is diagnosed by confirmation of an intrapulmonary shunt by echocardiogram. Injection of agitated saline results in saline bubbles being visualized in the left atrium 3 or more cardiac cycles after they appear in the right atrium. Currently, there is no effective medical treatment for HPS and liver transplantation is the only successful treatment.

Conclusion

POPH is an uncommon complication of chronic liver disease. It is defined as PAH in a patient with portal hypertension excluding other causes of PAH. The following criteria must be met to make a diagnosis of POPH: (1) evidence of portal hypertension; (2) MPAP ≥ 35 mm Hg; (3) PVR ≥ 240 dynes/s/cm⁵; (4) pulmonary capillary wedge pressure ≤ 15 mm Hg; and (5) TPG > 12 mm Hg. Individuals with POPH have worse outcomes compared to other forms of PAH, with a median survival of 6 months without medical therapy. The pathogenesis of POPH is unclear but may be related to a genetic predisposition since not all patients with portal hypertension develop POPH. Echocardiography is an excellent screening test for POPH, but a right heart catheterization must be performed to confirm the diagnosis.