Research work · 2008

Portopulmonary hypertension†‡

Jennifer T. Wells, James R. Runo, Michael Ronan Lucey

Hepatology

Abstract

Portopulmonary hypertension refers to the condition of pulmonary arterial hypertension (PAH) in association with portal hypertension. Portopulmonary hypertension was first described in 1951 when Mantz and Craige published their experience with a 53-year-old female who underwent surgical exploration for complaints of hoarseness and hematemesis.1 Intraoperative findings included esophageal varices and an enlarged pulmonary artery with "pulsations simulating that of the aorta". While under anesthesia, the patient developed hypotension, hemodynamic instability, and expired. Postmortem examination revealed a small portal vein with a portocaval shunt. Pulmonary examination demonstrated abnormalities of the pulmonary arteries and arterioles such as intimal thickening in medium and large pulmonary arteries as well as thrombotic changes and endothelial proliferation of terminal arterioles. This constellation of histologic findings is now recognized to be characteristic of portopulmonary hypertension.2 At the Third World Symposium in Venice, Italy in 2003, portopulmonary hypertension became reclassified as a unique entity within the modified World Health Organization Group 1 category of PAH.3 AASLD, American Association for the Study of Liver Diseases; ERS-EASL, European Respiratory Society–European Association for the Study of the Liver; HCV, hepatitis C virus; MELD, Model for End-Stage Liver Disease; mPAP, mean pulmonary artery pressure; PAH, pulmonary arterial hypertension; PCWP, pulmonary capillary wedge pressure; PVR, vascular resistance; SLE, systemic lupus erythematosus; TTE, transthoracic echocardiography. Resting mean pulmonary artery pressure (mPAP) > 25mmHg Pulmonary capillary wedge pressure < 15 mmHg Pulmonary vascular resistance > 240 dynes · s · cm−5 Awareness and high clinical suspicion of portopulmonary hypertension is paramount due to the nonspecific nature of the symptoms and physical findings early in the disease course. Subsequent complaints of chest pain, syncope, and findings of right heart strain or failure should raise concern for advanced disease.6, 7 Prompt and accurate diagnosis is important with regular follow-up of disease progression so that medical therapy can be instituted when appropriate. Mantz and Craige uncovered an important outcome when their patient developed irreversible intraoperative hemodynamic instability.1 Portopulmonary hypertension is associated with increased mortality, with a reported perioperative mortality in up to 36% of patients.8 The presence of pulmonary hypertension of any severity increases the perioperative and long-term risks of liver transplantation.9 An attempt to stratify patients according to their mortality risk has produced the following classification: mild (mPAP 25-35 mmHg), moderate (mPAP 35-45 mmHg), and severe (mPAP > 45 mmHg). Krowka et al. retrospectively studied patients with portopulmonary hypertension and peritransplant mortality.10 They found no increased mortality in patients with mild disease; however, untreated moderate pulmonary hypertension with PVR ≥ 250 dynes · s · cm−5 was associated with 50% mortality. The outcome in those patients with severe disease was uniformly fatal. A more recent report from the same group has suggested less dire outcomes based on mPAP and PVR although severe portopulmonary hypertension remains associated with unacceptable mortality.11 Despite previous descriptions of portopulmonary hypertension from single-center retrospective reviews, little is understood about the clinical associations or mechanisms that underlie its development, and we have a poor understanding of optimal therapeutic intervention.12 In this issue of HEPATOLOGY, Kawut et al.13 report a prospective nested case-controlled study of portopulmonary hypertension conducted among seven tertiary-care liver transplant centers in the United States. The authors' goals were to identify risk factors associated with portopulmonary hypertension, including whether the severity of liver failure or type of liver disease stratified with the development of portopulmonary hypertension. The cohort comprised patients under evaluation for liver transplantation who had abnormal estimated mPAPs according to transthoracic echocardiography (TTE). Strict criteria, based on hemodynamic data from right heart catheterization, were adopted for the diagnosis of portopulmonary hypertension. The observed prevalence of portopulmonary hypertension was 6% (34 of 536 patients), which is congruent with other reports in similar populations.14 In a multivariate analysis, female sex and underlying autoimmune liver disease were significantly associated with portopulmonary hypertension. Although females are known to have an increased incidence of idiopathic PAH,15 this is the first study to demonstrate that females are at increased risk for pulmonary hypertension in the setting of portal hypertension.16 Similarly, primary biliary cirrhosis, also an autoimmune condition predominantly in females, showed a trend toward significance (P = 0.07). Furthermore, age was not an increased risk factor. The average age in this study, 53 years, is almost two decades older than the average age of presentation in patients with idiopathic PAH (36 years).15 How can we account for the apparent link between autoimmunity and portopulmonary hypertension? One explanation is that the association is factitious, rather than causal, because the case number is low. In response, the authors point out that the predictors had a large effect size and were highly significant. Another explanation rests on the interaction between the host and his/her immune system. Further investigation of autoimmunity in patients with or at risk for pulmonary hypertension would appear to be a worthwhile avenue for study. When investigating a de novo diagnosis of portopulmonary hypertension, a thorough investigation for concomitant autoimmune conditions known to cause pulmonary hypertension such as scleroderma, rheumatoid arthritis, and systemic lupus erythematosus would be appropriate.17 In the study by Kawut et al., a diagnosis of hepatitis C viral (HCV) infection appeared to be protective. Data regarding the sex breakdown of the patients with HCV, traditionally more prominent in men, could help explain this finding. Male sex, in fact, may be more protective than the diagnosis of HCV infection. In addition, no correlation was found between mPAP and Model for End-Stage Liver Disease (MELD) score. This confirms previous reports18 that portopulmonary hypertension is not associated with the severity of liver dysfunction. In addition, several clinical and biochemical markers of liver function had either no correlation (total bilirubin and international normalized ratio) with mPAP or were negatively associated with portopulmonary hypertension (ascites and albumin). Treatment of portopulmonary hypertension remains problematic. There are no randomized controlled trials of therapy for portopulmonary hypertension. Treatment is modeled on the response of patients with idiopathic PAH in clinical trials that often exclude patients with portopulmonary hypertension. Available treatment options include vasodilators, such as prostaglandins, phosphodiesterase-5 inhibitors, and a dual endothelin receptor agonist.9 The most studied agent has been epoprostenol, a prostaglandin. Other agents have been reported in case series or small trials and include iloprost, sildenafil, and bosentan. At this time, there are no treatment guidelines and therapy is directed by clinician experiences. The role of liver transplantation remains controversial. At the moment, severe pulmonary hypertension, such as mPAP ≥ 45 mmHg, that does not respond to medical management is a relative contraindication to liver transplantation.9 Transplantation has been successful in patients who respond to therapy and many, though not all, are able to wean off or reduce vasodilator therapy over the first 6 months. Consequently, the American Association for the Study of Liver Diseases (AASLD) practice guidelines regarding liver transplantation state "Patients with severe pulmonary hypertension should be considered for liver transplantation only if the condition can be effectively controlled with medical therapy."19 Who should be screened for portopulmonary hypertension? The current practice guidelines of the AASLD suggest that all patients undergoing evaluation for liver transplantation should be screened for pulmonary hypertension with TTE.19 It is less certain whether patients whose liver disease is not sufficient to require liver transplantation should undergo screening. The observation in the study by Kawut et al. that significant portopulmonary hypertension can coexist with nonurgent cirrhosis certainly raises this possibility. Proof that such screening would be beneficial requires demonstrating that early identification, treatment, and possibly liver transplantation improves morbidity and mortality. This needs further clarification, making it difficult at this time to recommend such screening. Similarly, we lack data to guide medical therapy in the circumstances of portopulmonary hypertension in a patient with well-compensated cirrhosis. In summary, the article by Kawut et al. provides important findings suggesting that portopulmonary hypertension may be affected by sex and autoimmunity. These findings are hypothesis-generating and may ultimately lead to further clinical and scientific investigations that better elucidate pathogenesis and treatment of portopulmonary hypertension.

Download PDF

DOI: 10.1002/hep.22378