The hepatic venous pressure gradient (HVPG) is becoming increasingly used clinically. It is useful in the differential diagnosis of portal hypertension and provides a prognostic index in cirrhotic patients. Performance of serial measurements has been shown to be useful in guiding pharmacological therapy of portal hypertension and variceal hemorrhage. The technique is safe to perform; however, many patients are anxious and reluctant to undergo serial measurements. The effects of sedatives on portal pressure measurements have not yet been defined. The objective of this study was to evaluate the effects of midazolam on the HVPG. Twenty patients with compensated cirrhosis were included in this prospective, double-blind study. The HVPG was determined by subtracting the free hepatic venous pressure (FHVP) from the wedged hepatic venous pressure (WHVP). Patients were randomized to receive either placebo, 0.02 mg/kg midazolam, or 0.03 mg/kg midazolam, administered intravenously over 3 minutes. Immediately after drug administration and every 3 minutes thereafter, for a total of 30 or 40 minutes, measurements were repeated. Three hours later, patients were asked to state whether the sedative affected their state of comfort/relaxation. The effects of both doses of midazolam on HVPG did not differ significantly from those of placebo. Furthermore, neither dose of midazolam induced significant changes in HVPG as compared with baseline values. However, higher-dose midazolam (0.03 mg/kg) was associated with significant reductions in FHVP from baseline and a tendency for a reduction in WHVP. Both doses significantly increased patient comfort and relaxation during the test. Midazolam, used at a dose of 0.02 mg/kg, is effective in increasing patient comfort and relaxation during hepatic venous pressure measurements, without significantly affecting pressures (HVPG, WHVP, or FHVP). It is therefore an acceptable option for patients undergoing serial hepatic venous pressure measurements