BACKGROUND: Pulmonary dysfunction including the hepatopulmonary syndrome (HPS) is an important complication to cirrhosis and portal hypertension. However, the precise relation to liver dysfunction and the prevalence of HPS are unclear. AIMS: We therefore aimed to assess (i) the prevalence of HPS in consecutive alcoholic cirrhotic patients, (ii) the degree of pulmonary dysfunction in relation to liver function and (iii) the response of a 100% oxygen test on cardiopulmonary and peripheral oxygenation. METHODS: Fifty patients with cirrhosis and 12 matched healthy controls were entered in this study. All underwent haemodynamic and pulmonary investigations [lung diffusing capacity for carbon monoxide (DLCO), contrast-enhanced echocardiography and detection of extrapulmonary shunt fraction]. A 100% oxygen test was performed with the assessment of arterial oxygen tension (PaO(2)), the alveolar-arterial oxygen gradient (AaPO(2)) and peripheral transcutaneous oxygen tension (tcPO(2)). RESULTS: The prevalence of HPS was 10%. PaO(2) and DLCO were reduced in 32 and 72% and AaPO(2), was increased in 60% of the patients respectively. DLCO correlated with indicators of liver dysfunction (galactose elimination capacity, P<0.01, indocyanine green clearance, P<0.001), portal hypertension (post-sinusoidal resistance, P<0.01) and central hypovolaemia (central and arterial blood volume, P<0.01). After 100% oxygen inhalation, the changes in PaO(2), AaPO(2), tcPO(2) and heart rate were abnormal in the patients compared with controls (P<0.02). CONCLUSIONS: Pulmonary dysfunction in alcoholic cirrhosis is common and relates to different aspects of liver dysfunction, whereas the prevalence of HPS is low. The haemodynamic response to oxygen inhalation is clearly impaired and HPS and pulmonary dysfunction seem to be caused by different pathophysiological mechanisms.
BACKGROUND:Pulmonary dysfunction including the hepatopulmonary syndrome (HPS) is an important complication to cirrhosis and portal hypertension. However, the precise relation to liver dysfunction and the prevalence of HPS are unclear. AIMS: We therefore aimed to assess (i) the prevalence of HPS in consecutive alcoholic cirrhoticpatients, (ii) the degree of pulmonary dysfunction in relation to liver function and (iii) the response of a 100% oxygen test on cardiopulmonary and peripheral oxygenation. METHODS: Fifty patients with cirrhosis and 12 matched healthy controls were entered in this study. All underwent haemodynamic and pulmonary investigations [lung diffusing capacity for carbon monoxide (DLCO), contrast-enhanced echocardiography and detection of extrapulmonary shunt fraction]. A 100% oxygen test was performed with the assessment of arterial oxygen tension (PaO(2)), the alveolar-arterial oxygen gradient (AaPO(2)) and peripheral transcutaneous oxygen tension (tcPO(2)). RESULTS: The prevalence of HPS was 10%. PaO(2) and DLCO were reduced in 32 and 72% and AaPO(2), was increased in 60% of the patients respectively. DLCO correlated with indicators of liver dysfunction (galactose elimination capacity, P<0.01, indocyanine green clearance, P<0.001), portal hypertension (post-sinusoidal resistance, P<0.01) and central hypovolaemia (central and arterial blood volume, P<0.01). After 100% oxygen inhalation, the changes in PaO(2), AaPO(2), tcPO(2) and heart rate were abnormal in the patients compared with controls (P<0.02). CONCLUSIONS:Pulmonary dysfunction in alcoholic cirrhosis is common and relates to different aspects of liver dysfunction, whereas the prevalence of HPS is low. The haemodynamic response to oxygen inhalation is clearly impaired and HPS and pulmonary dysfunction seem to be caused by different pathophysiological mechanisms.