DISCUSSION
In this study, we investigated the pharmacokinetics of propranolol in patients with liver cirrhosis, and in healthy individuals and patients with chronic active hepatitis. We also identified the clinical factors that affect pharmacokinetics of propranolol and suggested a model for prediction of systemic exposure of propranolol. Generally, the bioavailability of a drug with high hepatic extraction is influenced by hepatic blood flow, protein binding, and intrinsic clearance; and the elimination is determined mainly by hepatic blood flow.[7] As propranolol is a typical drug with high hepatic extraction, the reduction of hepatic blood flow by portal hypertension can have a critical effect on the systemic exposure of propranolol by both increasing bioavailability and reducing elimination.
A few small studies have evaluated the pharmacokinetics of propranolol in normal subjects and patients with cirrhosis. A study with 6 normal subjects and 6 cirrhotic patients showed that the AUC during a dosing interval (AUCĪ„) was 3.5 times higher in cirrhotic patients after the administration of 80 mg of propranolol twice a day for 7 days.[12] In another study involving 5 normal subjects and 15 cirrhotic patients, AUCinf in cirrhotic patients was about 5 times higher than in normal subject after the single administration of 40 mg of propranolol.[13] In this study, the mean Cmax and AUClast in patients with liver cirrhosis were higher by 1.2 folds and 1.4 folds compared to normal subjects, however, there was no statistically significant difference. The increase of systemic exposure in cirrhotic patients were obviously low comparing to previous studies. This small increase can be because most patients with cirrhosis were in Child-Pugh class A. In this study, the pharmacokinetics of propranolol were also evaluated in the patients with CAH. The patients with CAH exhibited the lowest systemic exposure of propranolol among three groups. While the protein binding and intrinsic clearance were comparable to those in normal subjects, the hepatic blood flow increased due to inflammation in these patients, which presumably increased hepatic elimination.
The Cmax and AUClast of propranolol were predicted by sex, weight, total bilirubin level, platelet count (AUClast only) and H/L ratio using the regression model. Weight can naturally influence drug systemic exposure and total bilirubin level and platelet count are considered to represent the liver function being related to intrinsic hepatic clearance and plasma protein binding. Interestingly, sex was a prediction factor for both Cmax and AUClast. Several studies reported that the sex difference in pharmacokinetics of propranolol.[14-16] These studies have commonly reported that the clearance of propranolol in male was higher. The authors suggested that circulating gonadal hormones might influence propranolol metabolism. Our finding is consistent with these previous reports.
A total of 78 subjects including 36 of cirrhotic patients, 18 of CAH patients and 24 of normal subjects were completed this study. Not only the number of subjects was large compared to previous studies, but this study evaluated H/L ratio representing portal hypertension. To the best of our knowledge, it is the first study that the measurement of portal hypertension was used for the prediction of pharmacokinetics. The extent of portal hypertension was found to significantly influence the systemic exposure of propranolol. The developed prediction model is expected to be applied to the pharmacokinetic prediction of drugs which is mainly eliminated by metabolism and have high hepatic extraction.
Despite these positive aspects of the present study, there were some limitations that should be addressed. The important one is the limitation of disease status of patients with cirrhosis. Most patients with cirrhosis were in Child-Pugh class A except for 2 patients who were in class B. For this limitation, it is not certain that our proposed model can be applied to patients with advanced chronic liver disease. In addition, our prediction model can only be applied to a drug of high hepatic extraction, because the influencing factors on the pharmacokinetics of low hepatic extraction drugs are different from those of high hepatic extraction drugs.