Both authors contributed equally to this work.The prevalence of hepatopulmonary syndrome (HPS) and its influence on survival before and after liver transplantation (LT) remain controversial. Additionally, the chronology of post-LT reversibility is unclear. This study prospectively analyzed 316 patients with cirrhosis who were evaluated for LT in 2002LT in -2007 177 underwent LT at a single reference hospital. HPS was defined by a partial pressure of arterial oxygen (PaO 2 ) <70 mmHg and/or an alveolar-arterial oxygen gradient (A-a PO 2 ) !20 mmHg in the supine position and positive contrast echocardiography. The prevalence of HPS was 25.6% (81/316 patients), and most patients (92.6%) had mild or moderate HPS. High Child-Pugh scores and the presence of ascites were independently associated with HPS. Patients with and without HPS did not significantly differ in LT waiting list survival (mean 34.6 months vs. 41.6 months, respectively; logrank, p ¼ 0.13) or post-LT survival (mean 45 months vs. 47.6 months, respectively; log-rank, p ¼ 0.62). HPS was reversed in all cases within 1 year after LT. One-fourth of the patients with cirrhosis who were evaluated for LT had HPS (mostly mild to moderate); the presence of HPS did not affect LT waiting list survival. HPS was always reversed after LT, and patient prognosis did not worsen. Hepatopulmonary syndrome (HPS) is characterized by blood oxygen changes that are caused by pulmonary vascular dilations in patients with liver disease (1,2). The criteria that are used to define HPS and the reported prevalence of HPS in patients with liver cirrhosis and in candidates for liver transplantation (LT) vary widely in the literature (4-32%) (3-7). The cause of HPS is unknown, and the association between liver dysfunction and portal hypertension remains controversial, as reported by different studies (4-10). Orthodeoxia is considered to be characteristic of HPS; however, the prevalence of this condition in patients with HPS is only 25% (1,11).The association between HPS and increased mortality in patients with cirrhosis who are candidates for LT has been disputed. A recent prospective multicenter study identified associations between HPS, independent of its severity, and increased mortality and poorer quality of life (7). Other studies of HPS and LT were retrospective (12) or were based on small sample sizes (6).LT is the treatment of choice for HPS. Early results of LT in patients with HPS have been disappointing and have

Obtaining blood gas measurements in the supine position and the use of modern criteria are more sensitive for the diagnosis of HPS. Blood gas analysis with the patient seated detects a greater number of severe and very severe cases. The presence of HPS was not associated with an increase in mortality regardless of blood gas criterion used.

The 99mTc-MAA is a low sensitivity test for the diagnosis of HPS that can be useful in patients who have concomitant lung disease and in severe to very severe cases of HPS. It was not related to mortality, and brain uptake normalized after LT.