Pathophysiology of portal hypertension and implications for its pharmacological control

A comparison of Doppler flowmetry with conventional assessment of acute changes in hepatic blood flow

The validity and clinical relevance of Doppler flowmetry in measuring changes in regional blood flow are uncertain. In the present study we compared changes induced ketanserin in regional splanchnic blood flow as measured by Doppler flowmetry with changes in conventionally measured systemic and in hepatic haemodynamic indices estimated pharmacokinetically using indocyanine green. Fourteen patients with alcoholic cirrhosis and portal hypertension were evaluated. On multivariate analyses, significant associations were noted for only three indices: changes in estimated hepatic blood flow were predicted jointly by changes in flow in the main and right portal veins and hepatic artery (R2 = 0.80); changes in intrahepatic shunting (indocyanine green extraction) were predicted by changes in flow in the main and right portal veins (R2 = 0.55); and changes in sinusoidal perfusion (indocyanine green clearance) were significantly predicted by changes in main portal vein flow alone (R2 = 0.76). These data support the validity of Doppler flowmetry in quantifying change in regional blood flow, but highlight the limitations in its clinical application and interpretation. The association of changes in main portal vein flow with changes in sinusoidal perfusion has clinical potential but requires confirmation using other modulating drugs.

Comparison of the effects of various vasodilators on the rat portal vein and mesenteric artery

The effects of various vasodilators on isolated helical strips of rat portal vein and mesenteric artery were examined. Dilazep, ibudilast, nifedipine, verapamil and papaverine relaxed the KCl- and norepinephrine-induced contractions of the portal vein to a greater extent than they relaxed those of mesenteric artery. Dibutyryl cyclic AMP and NKH477 did not show any significant difference in terms of the concentrations that produced 50% inhibition (IC50) of the KCl-induced contraction of the portal vein and mesenteric artery. The relaxant effects of nitroprusside and isosorbide dinitrate on the KCl- and norepinephrine-induced contractions of the portal vein were less potent than their effects on contractions of the mesenteric artery. The agents that inhibited the contractions of the portal vein more potently than those of the mesenteric artery were associated with parallel shifts to the right of the concentration-response curves for CaCl2 in both preparations. These results suggest that drugs with a Ca2+ channel blocking action may be preferable to those that cause a decrease in portal pressure for treatment of portal hypertension.

Percutaneous transhepatic measurement of the pressure gradient between the portal and hepatic veins

BACKGROUND

Knowledge of the portal pressure may be of value in the assessment of patients with chronic liver disease but its measurement is problematic.

AIMS

To evaluate the ease and safety of percutaneous transhepatic measurement of the pressure gradient between the portal and hepatic veins and to determine directly the need for an internal zero.

METHODS

Sixty-one patients undergoing liver biopsy for suspected liver disease had pressures in branches of portal and hepatic veins measured using a flexible 22G (Chiba) needle.

RESULTS

The procedure was successful in all patients, took less than ten minutes in most, and was associated with minimal discomfort. Post-procedure morbidity was similar to that of liver biopsy. Portal pressure using an external zero (either puncture site or sternal angle) was inaccurate compared with pressures obtained using the generally accepted gold standard internal zero, hepatic venous pressure, and led to incorrect classification of the presence or absence of portal hypertension in at least 10% of patients. Variations in hepatic venous pressure were not predictable on clinical grounds. The only histopathological feature predictive of portal hypertension was cirrhosis, 20 of 25 patients with and four of 36 patients without cirrhosis having portal hypertension. Of routine biochemical and haematological tests, only plasma albumin and platelet count jointly (and negatively) predicted hepatic venous pressure gradient on multiple regression analysis (R2 = 0.40).

CONCLUSIONS

The use of an internal zero is essential for accurate measurement of portal pressure and this can be achieved safely using the percutaneous, transhepatic route in patients with well compensated liver disease.

Duplex Doppler ultrasound of the ligamentum teres and portal vein: a clinically useful adjunct in the evaluation of patients with known or suspected chronic liver disease or portal hypertension

The prevalence and potential value of the detection of signs of portal hypertension by duplex Doppler ultrasound (DDU) of the ligamentum teres and portal vein in patients with known or suspected chronic liver disease and/or portal hypertension was studied in 136 consecutive patients undergoing clinical assessment including that of liver histopathology. Portal hypertension was considered to be present when any of the following DDU signs, previously demonstrated to be specific for portal hypertension, were present: an enlarged and/or patent para-umbilical vein, portal vein obstruction or hepatofugal flow in the portal vein. Of 123 patients with parenchymal liver disease, eighty-three had cirrhosis and, of these, portal hypertension was detected on DDU criteria in 86% of alcoholic cirrhotics and 67% of non-alcoholic cirrhotics. Of the 42 patients with non-cirrhotic liver disease, 1 of 7 patients with metastatic liver disease and 3 of 5 patients with alcoholic hepatitis had DDU signs of portal hypertension. Thus, in patients with parenchymal liver disease, DDU had a sensitivity of 73%, specificity of 90% and predictive values of 94 and 62% for positive and negative studies respectively for the detection of cirrhosis. In all 14 patients with portal hypertension secondary to vascular occlusive diseases, DDU examination of the ligamentum teres, portal vein and hepatic vein gave an accurate guide to the site of the occluding lesion. The high positive predictive value of DDU and its ability to aid in localizing the site of increased resistance to flow through the liver suggest that DDU of the ligamentum teres and portal vein is a potentially useful non-invasive adjunct in the assessment of patients with suspected or known liver disease or portal hypertension.