Trebicka et al.1 reported an excellent and interesting study on the effects of atorvastatin on portal pressure. They used bile duct–ligated rats, which are known to be a useful model for hemodynamic studies in experimental cirrhosis. A dose of 15 mg/kg/day of atorvastatin showed a significant reduction in the portal pressure via NO-dependent and NO-independent pathways. We believe that some critical reading before the application of this high-dose atorvastatin regimen to patients with cirrhosis should be pointed out.
Atorvastatin is a cholesterol-lowering agent acting via hydroxyl-methyl-glutaryl-coenzyme A reductase. Regimens used in human clinical trials range from 10 to 80 mg/day. However, no human clinical trial has used a dose as high as 1 g of atorvastatin (15 mg/70 kg body weight). The study by Zafra et al.2 showed a significant acute reduction in hepatic sinusoidal resistance and increased hepatic blood flow in patients with cirrhosis at a dose of 40 mg of simvastatin. However, no modification of the hepatic venous pressure gradient was observed.2 Perhaps atorvastatin could exert different effects as it is not a prodrug and a higher dosage can be used (80 mg).
The bile duct–ligated model induces secondary biliary cirrhosis, which resembles iatrogenic biliary disease and shares common characteristics with primary biliary cirrhosis. One of the distinct findings of biliary cirrhosis is hypercholesterolemia, which is associated with a reduced ability to eliminate cholesterol through the bile and the appearance of abnormal lipoproteins in plasma (lipoprotein X).3 Most forms of liver cirrhosis in humans are characterized by hypocholesterolemia, particularly because of compromised hepatic protein synthesis and a subsequent reduction in the formation of lipoproteins.
Chronic cholesterol-lowering atorvastatin therapy in cirrhosis raises some concerns.
First, among Child-Pugh C patients with viral hepatitis–induced cirrhosis, a decrease in cholesterol serum predicts death at 2 years.4 Although this association is statistically significant, it might reflect a more compromised protein synthesis rather than a direct effect of hypocholesterolemia on patient survival. However, this has not been assessed and should be taken into consideration before a cholesterol-lowering therapy is initiated in these patients.
Second, hypocholesterolemia might also affect a patient's quality of life. A large study performed by Partonen et al.5 in Finland showed that low serum total cholesterol is associated with a low mood, major depression, and death from suicide. Several studies have confirmed this finding, but the mechanism is not clear.6–8
Third, cholesterol is necessary for macrophage action against Mycobacterium tuberculosis. Patients with tuberculosis often have hypocholesterolemia in comparison with the general population and household contacts, and a cholesterol-rich diet accelerates the bacteriological sterilization of sputum.9 Therefore, cholesterol-reducing regimens in patients with cirrhosis with normal or low serum cholesterol in the developing world should be carefully evaluated because most patients with cirrhosis live in those countries.10
In conclusion, just as for other interesting findings in the field of liver physiology, caution is advisable when we can extrapolate the results from animal models to humans.