[Hepatopulmonary syndrome in liver cirrhosis]

Hepatopulmonary syndrome (HPS) is defined by liver disease, hypoxaemia, increase of alveolar-arterial gradient, when inhaling room air, and intrapulmonary vascular dilatation. Pathoanatomical substrate of intrapulmonary vascular dilatation consists of dilated precapillary network, direct arterio-venous communication and dilated pleural blood vessels, "pleural spiders" [2]. Recently, hepatopulmonary syndrome gained clinical significance. Deterioration of arterial oxygenation in patients with liver disease indicates a very poor prognosis, because of which there are suggestions to classify hepatopulmonary syndrome as a new indication of liver transplantation [4].

The aim of the study was to examine the clinical and pathogenetic significance of intrapulmonary shunts in the development of respiratory disorders, and determination of correlation between intrapulmonary shunts and clinical, biochemical parameters in patients with liver cirrhosis.

In a prospective study over the period 1996-1999, we observed 50 patients with liver cirrhosis who were examined and treated at the Department of Gastroenterology and Hepatology of the Clinical Centre of Serbia in Belgrade. Hepatic examinations were based on medical history, physical examination, laboratory tests, ultrasound, duplex Doppler ultrasound, and histopathological findings. Hepatic failure was classified into 3 stages according to Child's classification. In this study we used 2 groups of pulmonary functional tests: analysis of arterial blood gas exchange and ventilation tests (spirometry, flow-volume curve and body pletismography). Arterial blood gas analysis was performed in supine and sitting positions while inhaling room air and after 15 min of inhaling hyperoxic mixture.

Patients were classified according to the generally accepted Child's classification into 3 stages: Sixteen patients (32%) were in Child's A stage, 20 patients (40%) in Child's B stage, and 14 (28%) in stage Child's C. HPS was diagnosed in 9 (18%) patients with liver cirrhosis. The majority of patients with HPS (6) were in Child's C stage, while the other 3 patients were in Child's B stage. Biochemical parameters were considerably worse in patients with shunts in comparison to those without shunts. However, t-test shoved no significant difference. Hypoxaemia caused by intrapulmonary arterio-venous shunt in supine position was found in 7 (14%) patients. Mean value of PaO2 was 8.72 kPa (SD = 1.02). Hypoxaemia caused by arterio-venous shunt in sitting position was found in 9 (18%) patients. The mean PaO2 value was 7.41 kPa, SD = 1.81. Orthodeoxia was confirmed in all patients with intrapulmonary shunt. Hypoxaemia without shunt in supine position was found in 18 (36%) patients, while 12 (24%) patients had hypoxaemia without shunt in sitting position. Comparing groups of patients with shunt and without shunt in supine position, we found a borderline statistically significant difference in PaO2 values (p = 0.057, t-test). In sitting position the difference in PaO2 values between these groups was statistically very significant (p = 0.001, t-test).

It is supposed that approximately 50% of patients with indication for liver transplantation have some form of arterial oxygenation disorder and 13-47% of these patients may have HPS [6, 7]. In our study, HPS was diagnosed in 18% of patients. We explain this high incidence by the fact that our study included the patients with advanced liver cirrhosis (stages Child's B and C). In studies performed up to date, there was neither correlation between biochemical liver function parameters and intrapulmonary shunts, nor any strong relation between severity of hepatic failure and degree of hypoxaemia [12, 13]. We noticed no correlation between hepatic functions (synthetic, excretory, transaminases) and PaO2 and/or intrapulmonary shunts. Some authors suggested that ventilation-perfusion disorder (Va/Q) is an important cause of hypoxaemia in