The reported prevalence of HPS varies depending on the population studied and the criteria used to define arterial deoxygenation and IPVD. The severity of the syndrome is not directly correlated with the severity or cause of the underlying liver disease, it can be seen also in patients with relatively mild liver diseases ⁴⁹. The morbidity of HPS between men and women has no difference, and HPS can also occur in children but no statistical data ⁵⁰. According to different literatures, the frequency of HPS ranges between 5% and 47% ^51-54, most of these data are from western countries. A recent study from India shown the morbidity of HPS in patients with cirrhosis was 6.7% ⁵⁵. The reported HPS morbidity in children with polysplenia syndrome, biliary atresia, Budd-Chiari syndrome and chronic liver diseases, is 0.5, 20, 28 and 8-13%, respectively,and 17-19% in the recipients of LT ^56-61. Whitworth et al ⁶² recently reported that advanced liver disease increases the frequency of HPS in children. Anand et al ⁶³ reported HPS is present in 17.5% patients with cirrhosis, 13.3% with non-cirrhotic portal fbrosis and 10% with extrahepatic portal vein obstruction ⁶³. According to a retrospective study, HPS occurred more frequently in cirrhosis (40%) than extrahepatic portal venous obstruction (13%) when the age is between 1 to 18 years of age (total 135 children). Schenk et al ⁵⁴ have reported that 27 (24%) patients had HPS among 111 patients with cirrhosis. The median survival among patients with HPS was significantly shorter (10.6 months) than patients without HPS (40.8 months). If adjusting the severity of underlying liver diseases and excluding patients who underwent liver transplantation during follow-up, the mortality in those with HPS is higher than those without HPS. In the aspect of clinical manifestations, the incidence of dyspnea, platypnea, clubbing and spider nevi is associated with patients with HPS.

Although HPS typically occurs in the setting of cirrhosis or portal hypertension, it may also develop in patients with acute and chronic hepatitis in which the portal hypertension is not established, and in the presence of vascular abnormalities in which the hepatic venous outfow to the lung is limited, such as Abernathy malformation, cavopulmonary shunts, Budd-Chiari syndrome. There are no significant risk factors for the development of HPS, and this maybe because patients detected with HPS is not enough. However, greatly higher morbidity of HPS in cirrhotic patients indicates hepatic dysfunction and portal hypertension probably involved in this abnormality. HPS particularly occurs in cirrhotic portal hypertensive patients with severe hepatic dysfunction. For example, chronic viral hepatitis, biliary atresia, primary biliary cirrhosis,alcoholic liver cirrhosis, alpha 1 antitrypsin deficiency, cryptogenic liver cirrhosis, wilson's disease, tyrosinemia. HPS also reported occurred in patients with cystic fbrosis ⁶⁴, schistosomal liver disease ⁶⁵, Gaucher disease ⁶⁶, autoimmune liver cirrhosis ⁶⁷. It has also been detected in acute hepatic conditions, such as ischemic hepatitis and viral hepatitis, and in non-cirrhotic portal hypertension, such as inferior vena cava obstruction ^68-70. Regev et al and Fuhrmann et al demonstrated that HPS could be transient in the cases of acute hepatitis ^{71, 72}.

The discovery of HPS requires a high degree of clinical suspicion and its diagnosis depends on evidence that demonstrate the presence of arterial gas exchange abnormalities, intrapulmonary vascular dilatation and the setting of liver dysfunction ⁷³. It is appropriate and cost effective to screen HPS in these patients with clubbing and (or) dyspnoea and without potential intrinsic cardiopulmonary diseases as well as in those being considered for liver transplantation ⁷⁴. It is particularly important to diagnose and differentiate HPS and PoPH, considering that the treatment and candidacy and priority for liver transplantation are different in these disorders. In most cases,once the intrinsic cardiorespiratory diseases are excluded, arterial blood gases analysis and a study to detect intrapulmonary shunt are suffcient to determine HPS.

Pulse oximetry is a simple, low cost, noninvasive, quick, accurate, quite reproducible and widely available technique that forecasts the presence and severity of hypoxemia in patients with HPS ^{75, 76}. Using pulse oximetry screening routinely can improve the detection and management of HPS in cirrhotic patients. In consideration of the utility of pulse oximetry for detecting hypoxemia in a wide range of disorders, this technique is also benefcial for screening all populations with cirrhosis ⁷³. However, arterial blood gas analysis in the sitting position,breathing room air, remains the confrmatory strategy to assess the details and severity of arterial hypoxemia associated with liver diseases, because of a lowered oxygen saturation (SpO ₂) alone is not suffcient to diagnose HPS due to other abnormalities, such as hydrothorax and ascites, may coexist in HPS patients and contribute to hypoxemia, especially when the SpO ₂ values are 97%or less ⁷⁷. Gas exchange abnormalities detected by arterial blood gas analysis were quantifed by assessing for partial pressure of oxygen (PaO ₂) and calculating alveolar-arterial oxygen gradient(AaPO ₂). Recommended cutoff values for the diagnosis of HPS are PaO ₂＜80 mmHg or AaPO ₂＞15 mmHg. Based on the findings that mortality is increased in this subset of patients with these abnormalities (increased AaPO ₂, PaO ₂＞80 mmHg) when compared with cirrhotic patients without HPS, so these abnormalities also included in the diagnostic criteria for HPS ⁷³. As AaPO ₂ increased with age, cutoff values that PaO ₂＜70 mmHg or AaPO ₂＞20 mmHg are suggested in patients older than 64 years ⁷⁷. PaO ₂ can stage the severity of HPS according to the classifcation system proposed by ERS Task Force. That severe HPS is PaO ₂＜50 mmHg, moderate HPS is PaO ₂ in between 50 and 60 mmHg, and mild HPS is PaO ₂ in between 60 and 80 mmHg. It is important to stage the degree of HPS since this degree is a means of predicting survival and determining the timing and risks of liver transplantation.

The diagnosis of HPS requires demonstration of IPVD in all patients who have chronic liver disease and hypoxemia. Contrast echocardiography and technetium-99m labelled macroaggregated albumin (99mTcMAA) perfusion scanning are the most common tools ^{73, 77}.Pulmonary angiography and HRCT scanning are additional measures that can be used as adjunctive tests in selected individuals. However, contrast echocardiography is still considered the standard diagnostic tool for HPS ⁷⁷. In addition, echocardiography can exclude cardiac dysfunction and PoPH for its function in assessing cardiac function and estimating pulmonary arterial systolic pressure. During contrast enhanced echocardiography detection, microbubbles obtained from agitated saline (10 μm in diameter) are injected into a peripheral vein. In normal condition, microbubbles are stuck in the pulmonary capillary bed (normal range of the capillary diameter, 8-15 μm). However, in HPS, bubbles traverse the lung circulation and enter the left atrium. Timing of the arrival of microbubbles in the left atrium distinguishes intrapulmonary shunting in intracardiac levels ⁷⁷. In adults, contrast enhanced transthoracic echocardiography is much sensitive and commonly used screening technique for detecting IPVD ⁷⁸. Lung 99mTcMAA perfusion scanning is able to quantifcation of the degree of intrapulmonary shunting ⁷⁹. The role of 99mTcMAA is similar with microbubbles used in contrast echocardiography. Normally, most particles remained in the lung microvasculature, but in the setting of HPS, some particles escape through enlarged capillaries and exit lung. The fraction of particles that escape the lung and reach the brain can be calculated by imaging radioactive index in the lung and brain and quantitative analyzing the ratio between the brain and lung. In this way, a positive 99mTcMAA scan (shunting＞6%) is found in patients with HPS who have a PaO ₂＜60 mmHg instead of those with intrinsic lung disease alone ^{80, 81}. However, as a screening test in adults, 99mTcMAA scanning is less sensitive than contrast echocardiography in detecting IPVD, and cannot evaluate intracardiac shunting, cardiac function and pulmonary artery pressures.

Recently, several studies found the degree of dilatation observed on chest CT correlates with the severity of gas exchange abnormalities and this method is less invasive ^{82, 83}, suggesting that chest HRCT has the potential in assessing the presence and severity of HPS.

Although a number of compounds have been studied in experimental and human disease, no clearly effective medical therapy for HPS is available. Inhaled nitric oxide (NO), somatostatin,aspirin, norfoxacin, quercetin, mycophenolate mofetil, and paroxetine have all been tried without clear benefit ⁷³. Pentoxifylline, methylene blue, and garlic proved to have promising results in some studies, but results cannot be generalized to all patients due to the lack of randomized trials with proper study population ⁷³. The effects of endothelin receptor antagonists and angiogenesis inhibitors, which have been demonstrated beneficial in the animal models of HPS, in treat human HPS are still unclear. Although no studies have evaluated the survival benefit of supplemental oxygen therapy, many studies reported that is useful in hypoxemic patients with HPS. Some other interventions maybe benefcial in treat HPS have been described in a number of case reports, including inhaled prostacyclin derivatives ^{84, 85}, withdrawal of chronic methadone ⁸⁶ and lowering of portal pressure with transjugular intrahepatic portosystemic shunt (TIPS) on patients with HPS. However, the use of TIPS to reduce portal hypertension and improve oxygenation remains controversial in HPS due to the lack of large clinical series or randomized controlled trials ^{87, 88}.

Currently, LT is still the only effective treatment for patients with HPS. According to the literatures, the complete resolution of gas exchange abnormalities is reported in ＞80% of patients with HPS underwent LT ^{53, 89}. Both living donor and deceased donor liver transplantation have been reported to be effective ^90-92. Notably, an early prospective study found that postoperative mortality (30%) was markedly increased in patients with severe HPS (preoperative PaO ₂ of≤50 mmHg and 99mTcMAA shunt fraction ≥20%) when compared to those with a PaO ₂ greater than 50 mmHg (4%) ⁹³. One study found the response of hypoxemia to 100% oxygen breathing seem to be the most important prognostic factor of perioperative death rate ⁹⁴. During the stage of recovery after liver transplantation, frequent repositioning of patients and inhaling NO have been reported to be benefcial in improving oxygenation ^{95, 96}. A number of additional small studies and an analysis of the Scientifc Registry of Transplant Recipients data have found 1-3 year mortality after liver transplantation distributing widely from 5% to 42% in patients with HPS ⁷³. These results underline the necessity to defne the effects of HPS on liver transplantation outcomes more precisely.