Transhepatic portal venography: a new method of portal vein visualization.

Abstract

Technics for visualization of the portal venous system in man have been available for the past ten years. They consist of direct injection into the portal vein at operation or injection into the splenic pulp (splenoportography). The first method requires direct operative exposure and the second does not permit visualization of the intrahepatic portal system in the face of a blocked splenic or main portal vein, or with high portal pressure and reversal of flow; nor can this method be applied for intrahepatic portal vein visualization in the patient who has had a splenectomy or operative shunting of the portal venous system. In 1951, Rappaport (1) published his original and extensive study of hepatic venography in the dog. He described hepatic venous anastomotic channels which communicated freely with one another and permitted diffuse hepatic vein filling from a single large injection of contrast material. We have studied a large number of injection-corrosion casts of dog livers and have failed to demonstrate the presence of such anastomoses in the normal animal. A comparison of our studies with the anatomic distribution of the channels described by Rap-paport, with direct portal venous injections in dogs, led us to believe that in his experiments there was reflux of contrast material from the hepatic vein into the portal system. The purpose of this paper is to report our preliminary investigation of the possibilities of utilizing retrograde hepatic vein injections for the visualization of the portal venous system in animals and man. Method: An open-end opaque catheter is passed from a peripheral vein, through the right atrium, and into a wedged position in an hepatic vein. Pressures are then recorded and the catheter withdrawn approximately 1 cm. from the wedge, a procedure confirmed by a drop in pressure. A forceful hand injection of 20 c.c. sodium diatri-zoate is then made and serial filming carried out (2). In man we have used a No. 10 Cournand or a No. 280 polyethylene catheter and 40 to 50 c.c. of contrast material. ResultsThe examination was successfully performed in 20 normal mongrel dogs. In each of these animals complete visualization of the intrahepatic portal system was accomplished. Subsequent direct injection of the portal vein in vivo and in vitro confirmed the fact that the structures visualized were indeed portal veins. Artificial blocks of the portal veins were made by in vivo injection of vinyl plastic material through a Hales needle guide (3) attached to the portal vein. Transhepatic portal venograms demonstrated the failure of these blocked veins to fill with contrast material. Nine patients have been studied by this method, with satisfactory filling of the entire portal venous system of the lobe injected. To this date we have been unable to achieve reflux of the contrast material past the porta hepatis in man.