In this patient with advanced liver disease, the likely causes of progressive dyspnea include anemia as well as cardiac or pulmonary disease. Anemia may be secondary to overt or occult gastrointestinal bleeding from portal gastropathy or varices, or due to poor nutrition, alcohol induced marrow toxicity or hypersplenism. Likely cardiac causes include alcohol induced cardiomyopathy (which results in impaired left ventricular function and heart failure), while potential pulmonary causes include pleural effusions, pulmonary compression secondary to ascites, or the hepatopulmonary syndrome . In this patient, the clinical findings are not supportive of anemia or heart failure and the lungs appear normal. However, the presence of central cyanosis indicates that significant hypoxemia is present. The digital clubbing may be secondary to cirrhosis, but when taken in conjunction with the central cyanosis, should also raise the suspicion of a shunt between the pulmonary and systemic circulations. Such a shunt may be intra-cardiac or extra-cardiac (including intra-pulmonary) in origin. The noteworthy point in his history is the worsening of dyspnea when sitting up. This is known as platypnea and is only caused by a few disorders : the hepatopulmonary syndrome, a left atrial myxoma or a left atrial thrombus. The blood gas analysis shows significant hypoxemia when upright, with relief upon lying down. This postural hypoxia is known as orthodoxiea and is associated with the clinical finding of platypnea. His chest x-ray appears normal, which makes significant pulmonary parenchymal disease unlikely. This is corroborated by the contrast CT scan of the chest, which additionally shows no evidence to suggest an arteriovenous shunt. The echocardiogram shows a structurally normal heart with no evidence of intra-cardiac shunting. Further evaluation via contrast echocardiography with agitated saline shows the passage of micro-bubbles from the right to the left sides of the heart. Given the absence of an intra-cardiac or arterio-venous shunt, this indicates the presence of an abnormally dilated pulmonary capillary bed - thus confirming the diagnosis to be the hepatopulmonary syndrome. Liver transplantation is the only definitive treatment for the hepatopulmonary syndrome. Antibiotics, Steroids and Interferon-Alpha have not been shown to be of benefit.
The hepatopulmonary syndrome is characterized by a defect in arterial oxygenation induced by pulmonary vascular dilatation in the setting of chronic liver disease. A prevalence of between 5% to 32% has been reported in patients undergoing liver transplantation. In the normal lung, the alveoli are uniformly ventilated and capillaries are uniformly perfused, resulting in a balance between ventilation and perfusion. In addition, the capillary diameter ranges between 8 to 15 micrometers, permitting the diffused oxygen to penetrate throughout the entire width of the vessel. In the hepatopulmonary syndrome, gross and uneven dilation of the pulmonary capillary vasculature takes place. This results in increased and non-uniform pulmonary blood flow, resulting in a ventilation-perfusion mismatch. In addition, oxygen diffusion into the center of the dilated capillaries is restricted. All of these factors contribute towards the resultant hypoxemia. Cirrhosis itself causes changes to the pulmonary vasculature - these include a reduction in vascular tone and a diminished vasoconstrictor response to hypoxia. Thus, pulmonary blood flow is further enhanced, exacerbating the ventilation-perfusion imbalance. In addition, loss of vascular tone results in the lung being unable to compensate for gravitational changes in blood flow - with subsequent development of orthodoxiea and platypnea. Dyspnea in the presence of spider nevi, clubbing, cyanosis and hypoxemia is suggestive of the hepatopulmonary syndrome. Platypnea may also be noted, but is often absent in the early stages. Most cases of the hepatopulmonary syndrome are associated with clinical evidence of portal hypertension (i.e. varices, splenomegaly or ascites). Note however that the hepatopulmonary syndrome has also been reported to occur in patients with acute liver failure or ischemic hepatitis. The chest x-ray is usually non-diagnostic, demonstrating a mild interstitial pattern in the lower lung fields in 46% to 100% of patients. This is probably secondary to the diffuse pulmonary vascular dilatation. Contrast enhanced transthoracic echocardiography with saline is the easiest method to detect pulmonary vascular dilatation. The saline is agitated to create microbubbles and injected via a peripheral vein. In the normal pulmonary vasculature, the capillary diameter is insufficient for microbubbles to pass through. However, the abnormally dilated capillary bed in patients with the hepatopulmonary syndrome allows passage of microbubbles, which is echocardiographically visualized as opacification of the left atrium 3 to 6 cardiac cycles after opacification of the right atrium. An alternative but more invasive diagnostic modality is injection of technetium tagged albumin into a peripheral vein. Uptake of technetium into the brain indicates that the injected technetium has progressed into the systemic circulation, bypassing the normal filtering mechanisms of the pulmonary capillary beds. Although the above diagnostic modalities may suggest the presence of a dilated capillary bed, it should be noted that intra-cardiac shunts and pulmonary arteriovenous shunts may also mimic these findings. Intra-cardiac shunts may be excluded at the time of echocardiography itself, while CT pulmonary angiography will exclude pulmonary arteriovenous shunts (but is only necessary if strong clinical suspicion of such a shunt exists). Liver transplantation is the only treatment proven to prolong survival in patients with the hepatopulmonary syndrome. Trials with antibiotics, beta-blockers, cyclooxygenase inhibitors, nitric oxide inhibitors and immunosuppresants have so far been uniformly unsuccessful. In addition, most patients are treated symptomatically with supplemental oxygen, although the mortality benefit, if any, is still a matter of debate. The 5-year survival rate of patients with the hepatopulmonary syndrome ranges from 23% to 63%. The causes of death are usually multifactorial and related primarily to the complications of hepatic disease.