Controversy and discussion on blood supply and interventional therapy of cavernous hemangiomas of the liver

Objective: To further explore the blood supply and interventional therapy of adult cavernous hemangiomas of the liver (CHL). Methods: Recently some authors reported that a satisfactory effect resulted from transcatheter portal venous embolization was obtained in few cases of CHL with blood supply of portal vein, and raised an objection to the standpoint that CHL was commonly supplied by hepatic artery completely. In order to get a scientific and reasonable explanation for it, this paper reviewed the vascular embryology and histology of the liver, the pathologic features of CHL as well as the relative literature, and combined with the investigation results of blood dynamic changes of CHL that had been performed in 2000-2002 by us. Results: CHL was caused by the arrested development of hepatic sinusoids at the embryonic stage. Pathologically, CHL was consisted of a lot of enlarged abnormal sinusoids,which were variant in size and closely related with the hemodynamic changes of CHL ( in inverse proportion). The CHL consisted of even and tiny abnormal sinusoids (diameter less than 50 μm) pathologically usually presented a high flow. During the hepatic artery angiography or CTHA, it was rapidly filled by the arterial blood containing contrast media and frequently showed dense opacification or enhancement. Simultaneously this could result in increased pressure of abnormal sinusoids. When the sinusoidal pressure exceeded that of the connecting portal venules, the arterial blood containing contrast media filled in the abnormal sinusoids could lead to retrograde flow in the portal venules. That was arterial-portal venous shunts (APVS). These appearances described as above could also occur in some CHL with intermediate flow, in which many tiny abnormal sinusoids located in the peripheral area were identified pathologically. On the contrary, the CHL consisted of larger abnormal sinusoids (diameter more than 500 μm) pathologically usually presented a low flow. During the hepatic artery angiography or CTHA, it was filled very slowly by the arterial blood containing contrast media and was difficult to opacify or enhance. Simultaneously this could result in a low pressure of abnormal sinusoids. When the sinusoidal pressure was lower than that of the connecting portal venules, the portal venous blood containing contrast media could easily flow into the abnormal sinusoids and make it enhanced during the direct or indirect portography (or CTAP). Conclusion: CHL is really a congenital venous malformation. All the CHL with high flow and some CHL with intermediate flow are surely supplied by the hepatic artery and drained primarily by the peripheral branches of portal vein. However, in few CHL with marked lower flow, the portal vein should become a primary supply vessel, so a direct or indirect portography (or CTAP) must often be taken to identify the diagnosis. Thereby, the technique of transcatheter embolization of CHL including the aim, indication, approach, and the used sclerotic or embolic drugs, etc, should also be reconsidered in order to improve its therapeutic efficacy. (authors)