Sirs, Bacteria and their products (DNA, endotoxin and cytokines) may contribute not only to deteriorate hyperdynamic circulation, but also to increase hepatic vascular resistance in cirrhotic patients, promoting by this way a further increase in portal pressure. Bacterial infections have been proposed as triggers for variceal bleeding and have been shown to be associated with a poor prognosis.1 As a consequence, whether antibiotics may contribute to ameliorating portal pressure in cirrhosis has been a matter of debate. Indeed, three studies had previously evaluated this issue (Table 1). Only two of them showed the effects of placebo on hepatic venous pressure gradient (HVPG) and the effect was completely equivalent to that of norfloxacin.2,3 In a third study,4 a nonsignificant trend to reduce HVPG was observed. However, this study did not evaluate the effect of placebo on HVPG and less than half of the patients had clinically significant portal hypertension (HVPG > 10 mmHg). A recent study published in Alimentary Pharmacology and Therapeutics5 provides novel data testing the possible efficacy of antibiotics in lowering portal pressure. The investigators treated for 1 month patients with decompensated alcoholic cirrhosis with rifaximin, a nonsystemically absorbed antibiotic, and observed a significant decrease in plasma endotoxin levels, which directly correlated with a significant drop in HVPG. The authors concluded that intestinal decontamination with rifaximine might represent a therapeutic approach in the prevention of complications of portal hypertension. The major drawback of this study was the absence of a control group. Indeed, similar effect on HVPG was observed in the two studies that had data on the effect of norfloxacin and placebo on HVPG, thus highlighting the importance of excluding other possible factors biasing the results like, for example, alcohol abstinence. The potential interest of these findings was further faded by the absence of follow-up data on the outcome of their patients during the study period. Therefore, these results should be considered preliminary and need to be confirmed. We must wait for additional studies with appropriate design and sample size to define clearly the possible role of antibiotics to improve the hemodynamic disturbances of cirrhotic patients. Declaration of personal and funding interests: The writing of this paper was funded in part by the Swiss National Science Foundation, the EASL Sheila Sherlock Fellowship. Ciberehd is funded by Instituto de Salud Carlos III.