It could be argued that, mechanistically, PAVMs related to HPS are much smaller than those associated with HHT. But it is also true that patients with HPS usually have a much larger number of microscopic PAVMs (which explains why patients with HPS can have severe right-to-left shunt). In addition, patients with advanced cirrhosis often have altered mental status as a consequence of their own disease (hepatic encephalopathy), perhaps masking neurologic signs; therefore, some of the CNS complications of their HPS could be missed. But as a counterargument, many times patients with hepatic encephalopathy do have brain images taken, and one would expect that if brain abscesses were common in them, more case reports or series should be available. Local mechanisms for the clearance of microemboli and bacteria in the lung vasculature will require further study to elucidate the cause of this striking difference. At the current time, it is not clear if this is an issue related only to the size of the PAVMs or whether some other mechanisms are involved.