Noninvasive measurement of portal venous blood flow in patients with cirrhosis: effects of physiological and pharmacological stimuli.

The present study aims to evaluate the usefulness of combined pulse Doppler-real-time ultrasonography as a noninvasive method for the measurement of portal blood flow in man. This measurement technique was performed on 12 healthy subjects and 20 patients with portal hypertension. Ten patients (group 1) were evaluated prior to and after ingestion of a standard meal (Ensure Plus) or placebo. In the remaining 10 patients (group 2), the effects of isosorbide dinitrate (5 mg/SL) administration or placebo were studied. In group 1, food intake caused a significant increase of portal blood flow (from 1038 +/- 539 to 1572 +/- 759 ml/min, P less than 0.02); this effect was due to a significant rise in mean blood velocity (from 18.5 +/- 3.7 to 23.9 +/- 3.9 cm/sec, P less than 0.02). In group 2, isosorbide dinitrate significantly reduced portal blood flow (from 985 +/- 491 to 625 +/- 355 ml/min, P less than 0.05); a significant decline of mean blood velocity (from 18.8 +/- 4.5 to 14.5 +/- 2.5 cm/sec, P less than 0.02) was observed. Placebo administration had no significant hemodynamic effects in either group. Our results suggest that Doppler measurements gave accurate noninvasive estimations of portal blood flow and that this technique may be used to monitor physiological and pharmological stimuli in patients with portal hypertension.