SciELO - Scientific Electronic Library Online

vol.39 número2Papel da ultra-sonografia na avaliação da efusão pleuralEnvolvimento hepático na sarcoidose: relato de caso índice de autoresíndice de assuntospesquisa de artigos
Home Pagelista alfabética de periódicos  

Serviços Personalizados



  • texto em Português
  • Português (pdf)
  • Artigo em XML
  • Como citar este artigo
  • SciELO Analytics
  • Curriculum ScienTI
  • Tradução automática


Links relacionados


Radiologia Brasileira

versão On-line ISSN 1678-7099

Radiol Bras v.39 n.2 São Paulo mar./abr. 2006 



Ultrasonographic features of portal vein thrombosis*



Márcio Martins MachadoI; Ana Cláudia Ferreira RosaII; Orlando Milhomem da MotaIII; Daniela Medeiros Milhomem CardosoIV; Patrícia Medeiros MilhomemV; Leonardo Medeiros MilhomemV; Rodrigo Alvarenga NunesVI; Thaís Bittencourt Gonçalves TelesVII; Flávio Bittencourt Gonçalves TelesVII; Letícia Martins AzeredoVIII; Nestor de BarrosIX; Giovanni Guido CerriX

IGuest Professor at Department of Radiology of Faculty of Medicine, Universidade Federal de Goiás
IIMD Radiologist at Hospital of Clinics, Faculty of Medicine, Universidade Federal de Goiás
IIIHead at Department of Digestive Tract Diseases of Hospital Araújo Jorge (Hospital do Câncer) of the Associação de Combate ao Câncer em Goiás
IVMD Resident at Surgery Department at Faculty of Medicine, Universidade Federal de Goiás
VMedicine Academician at Faculty of Medicine, Universidade de Ribeirão Preto
VIMedicine Academician at Faculty of Medical Sciences, Universidade do Vale do Sapucaí
VIIMedicine Academicians at Faculty of Medicine, Universidade Federal de Goiás
VIIIMD Ultrasonographist at Department of Radiology of Faculty of Medicine, Universidade Federal de Minas Gerais
IXDoctor Professor at Department of Radiology of Faculty of Medicine, Universidade de São Paulo
XTitular Professor at Department of Radiology of Faculty of Medicine, Universidade de São Paulo

Mailing address




Portal vein thrombosis may be associated with many alterations, such as the presence of tumors, (for example, hepatocellular carcinoma, metastatic liver disease and carcinoma of the pancreas) as well as pancreatitis, hepatitis, septicemia, trauma, splenectomy, portacaval shunts, hypercoagulable conditions (for example, pregnancy) in neonates (for example, omphalitis and umbilical vein catheterization) and acute dehydration. The authors discuss herein the sonographic features of portal vein thrombosis as well as some aspects of clinical relevance.

Keywords: Portal vein; Portal vein thrombosis; Ultrasound.




The portal vein thrombosis (Figure 1) is the most common cause for pre-hepatic portal hypertension. The arteriovenous fistulae involving the liver or the splanchnic circulation, are the second most common cause. Among the causes for portal vein thrombosis, we have dehydration, shock conditions, malignant neoplasms like hepatocarcinomas or hepatic metastases, pancreas carcinoma, primary leiomyosarcoma of the portal vein, chronic pancreatitis, hepatitis, splenectomy, hypercoagulable conditions like pregnancy, pylephlebitis (infectious process involving the portal vein, resulting, for example, from acute appendicitis or abdominal sepsis), other coagulopathies, portal invasion by extrahepatic tumors, portal vein compression (by acute pancreatitis, lymph nodes), trauma, catheterization, increase in the resistance to flow (cases of cirrhosis, Budd-Chiari syndrome or hepatosplenic schistosomiasis), neonate omphalitis and even after esophageal varices sclerotherapy(1). The bypass surgeries (mesenteric-caval, portacaval, distal splenic-renal) and the disconnection surgeries (azygoportal disconnection and splenectomy) may result in portal vein thrombosis as well as other spleno-mesenteric-portal venous stem related vessels thrombosis, as complications.



In this paper, the authors discuss the presentation aspects of portal vein thrombosis, also evaluating relevant clinical aspects in this context.



Although the complete portal vein thrombosis may course associated with portal hypertension, as previously mentioned, partial isolated thrombosis (Figure 2) will not course with portal hypertension. These cases would be recognized in routine tests, or in patients presenting associated risk factors. In these cases, the presence of echogenic material attached to the vessel wall is identified, causing partial obstruction of the lumen. The Doppler may even demonstrate portal flow with reasonably preserved characteristics. In the total thrombosis, the ultrasonography usually demonstrates the portal vein with echogenic material fulfilling its lumen. Collateral portal vessels, increase in portal vein caliber and cavernomatous transformation (Figure 3) also can be found in cases of thrombosis (especially total thrombosis).





In the cases of recent thrombosis, called acute, the thrombi can be anechoic(2) for practical purposes making its recognition quite difficult. However, the study of vessels with high resolution equipment allows the identification of material with some echogenicity (thrombus) in the majority of cases(3). Even in the cases of acute thrombi that are not echogenic, the use of spectral Color Doppler or "power Doppler" (comprehensive Doppler) is essential for the demonstration flow absence in the vessel.

In cases of chronic portal vein thrombosis (Figure 4), the ultrasonography usually demonstrates decreased caliber portal vein and fulfillment with echogenic material.



In cases of complete thrombosis (Figure 5) and in the absence of cavernomatous transformation, the Doppler does not identify the flow. Doppler may detect arterial flow in the adjacent hepatic artery, and the operator must be attentive to this fact. Another aspect to be remembered is that, in the case of complete portal vein occlusions, the periportal collaterals may develop. In these cases of chronic thrombosis (specially the complete ones), we may find portal vein cavernomatous transformation (Figure 6), which consists of the presence of tortuous vessels in the portal vein topography (at the porta hepatis or hepatic hilum), representing periportal collateral circulation (Figure 7). At the ultrasonography, it is seen as an elongated echogenic structure, with the identification of inumerable serpiginous and tortuous channels, representing the collateral vessels(4). These serpiginous vessels may represent the recanalization of the thrombus itself, or also well developed collateral peri-hilar (periportal) vessels, representing the extra-luminal collateral circulation. With Doppler, we can notice the presence of these tortuous channels presenting flow (usually with decreased velocity) and with a commonly "flattened" pattern, without the undulating pattern associated with the cardiac cycle, which is commonly observed in the normal portal vein(3).







In the study of the portal vein we must be aware to the fact that the flow may not be seen at any given moment due to technical problems and not to an actual absence of flow. One may attempt to document the flow in adjacent vessels with the same equipment and in similar depths to make sure that there is an actual absence of flow in the portal vein. On the other hand, the absence of flow by itself does not indicate the presence of occlusive thrombosis; a hepatic resistance to flow may be in progress, indicating a significant portal hypertension(3).

One of the indirect signs of portal vein thrombosis is the finding of high frequency signals of arterial flow, which represent the attempt to increase hepatic arterial circulation to compensate for the deficiency of portal venous flow. This finding may be observed both in the hepatic arterial trunk, and in its intrahepatic branches(2).

All these described aspects of portal vein thrombosis (hilar portal trunk) are also applicable to the thrombosis of its right and left branches and its smaller intra hepatic branches, with sonographic and Doppler aspects similar to those found on the portal vein trunk.

The thrombosis of the portal vein intra hepatic branches may be caused by the extension of the process started at the portal vein trunk (more commonly), or may occur isolatedly, with no association with portal thrombosis. One must also emphasize that the process of portal thrombosis may extend or originate from other vessels of the splenic-mesenteric-portal venous trunk, besides that represented by the portal vein trunk. Echogenic material partially or totally occlusing the vessel lumen is identified. At Doppler, one observes the absence of flow (in cases of occlusive thrombosis) or peripheral flow (partial thrombosis). Mainly at intra hepatic portal branches (right and left branches or more peripheral branches), the lumen may not be identified, with the visualization of echogenic bands only, representing the thrombi. In some of these cases, the pulsed Doppler may reveal the presence of some residual flow, even in the absence of an identifiable vascular lumen(3).

In portal occlusions, well developed collateral vessels may be identified, with pericholecystic collateral vessels (Figure 8) particularly evident in these cases(5). In this situation, they play the role of a collateral pathway of the splanchnic blood circulation towards the liver (collateral hepatopetal vessels). Although referred by some as almost pathognomonic of the associated presence of portal vein thrombosis(3), other authors have observed this finding in patients with cirrhotic portal hypertension or schistosomal without associated portal thrombosis(4,6).

Studies carried out with large number of patients with cirrhotic portal hypertension, indicate that the occurrence of partial portal thrombosis can be verified in 1.8% of the cases, and the total portal thrombosis in 4.4%(7). Still related to portal vein thrombosis, concerns with this evaluation are frequent in patients suffering from primary or secondary hepatic neoplasms. In this sense, although some variations exist, some authors report the presence of portal vein occlusion in approximately 25% of the patients with hepatocellular carcinoma and in 1% of the patients with metastases(4).

The development of hepatofugal flow is another alteration that may occur in portal hypertension. Angiography studies indicate the presence of this alteration, from 0% to 14.8% of the studied cases(8–11). Many studies have indicated intermediate rates of hepatofugal flow, ranging from 5.3% to 8.3%(2,12).

The flow reversion indicates the existence of high resistance to the intra-hepatic portal flow. In these situations, ample collateral vessels develop (porto-systemic collateral vessels), in order to provide alternate paths to the blood flow that is not being able to cross the hepatic bed towards systemic circulation. When that happens, the portal vein may have its caliper reduced, even in the presence of portal hypertension. The blood flow will be deviated to pathways with less resistance (systemic), through s collateral circulation, relieving the portal vein (portal hypertense system)(2). It has also been demonstrated that in the presence of blood flow reversion (hepatofugal flow) the risk of gastrointestinal hemorrhages is reduced, with the diameter of portal vein commonly reduced(2).

Another existing correlation is between the presence of hepatofugal flow in the splenic vein and the development of encephalopathy. This would occur due to the bypass of the portal flow to systemic circulation through the collateral vessels at the level of the splenic hilum, and mainly through splenorenal and spontaneous anastomosis(13).



The authors conclude that the causes for portal vein thrombosis are multifactorial, and that we shall always be attentive to this possibility, with the understanding that Doppler ultrasonography is a non-invasive method of great value in its identification.



1. Cerri GG, Mólnar LJ, Paranaguá-Vezozzo DC. Doppler. São Paulo: Sarvier, 1996;124–135.        [ Links ]

2. Bolondi L, Gaiani S, Barbara L. Liver and portal hypertension. In: Taylor KJW, Burns PN, Wells PNT, editors. Clinical applications of Doppler ultrasound. 2nd ed. Philadelphia: Lippincott-Raven, 1995;133–154.        [ Links ]

3. Weill FS. Ultrasound diagnosis of digestive diseases. 4th ed. Berlin: Springer-Verlag, 1996;121–135.        [ Links ]

4. Cosgrove DO. Ultrasound contrast agents. In: Meire H, Cosgrove D, Dewbury K, Farrant P, editors. Abdominal and general ultrasound. London: Churchill Livingstone, 2001;69–79.        [ Links ]

5. Cerri GG, Habr-Gama A, Paranaguá-Vezozzo D, Machado MCC, Magalhães A, Pinotti HW. Doppler demonstration of cystic vein dilatation secondary to portal vein thrombosis. Surg Endosc 1991;5:92–93.        [ Links ]

6. Machado MM, Rosa ACF, Cerri GG. Doenças hepáticas difusas, hipertensão portal e transplante de fígado. In: Cerri GG, Oliveira IRS, editores. Ultra-sonografia abdominal. Rio de Janeiro: Revinter, 2002;55–124.        [ Links ]

7. Gaiani S, Bolondi L, Fenyves D, Zironi G, Rigamonti A, Barbara L. Effect of propranolol on portosystemic collateral circulation in patients with cirrhosis. Hepatology 1991;14:824–829.        [ Links ]

8. Foster DN, Herlinger H, Miloszewski KJA, et al. Hepatofugal portal blood flow in hepatic cirrhosis. Ann Surg 1978;187:179–182.        [ Links ]

9. Okazaki K, Miyazaki M, Onishi S, Ito K. Effects of food intake and various extrinsic hormones on portal blood flow in patients with liver cirrhosis demonstrated by pulsed Doppler with Octoson. Scand J Gastroenterol 1986;21:1029–1038.        [ Links ]

10. Rector WG Jr, Hoefs JC, Hossack KF, Everson GT. Hepatofugal portal flow in cirrhosis: observations on hepatic hemodynamics and the nature of arterioportal communications. Hepatology 1988;8:16–20.        [ Links ]

11. Burcharth F, Aagaard J. Total hepatofugal portal blood flow in cirrhosis demonstrated by transhepatic portography. ROFO 1988;148:47–49.        [ Links ]

12. Kawasaki T, Moriyasu F, Nishida O, et al. Analysis of hepatofugal flow in portal venous system using ultrasonic Doppler duplex system. Am J Gastroenterol 1989;84:937–941.        [ Links ]

13. Ohnishi K, Saito M, Nakayama T, et al. Portal venous hemodynamics in chronic liver disease: effect of posture change and exercise. Radiology 1985;155:757–761.        [ Links ]



Mailing address:
Dr. Márcio Martins Machado
Rua Ruy Brasil Cavalcante, 496, Ed. Art-1, ap. 1001, Setor Oeste
Goiânia, GO, Brasil 74140-140

Received October 28, 2004.
Accepted after revision January 7, 2005.



* Study developed at Department of Radiology of Hospital das Clinicas da Faculdade de Medicina da Universidade de São Paulo, Diagnosis Center of Hospital Sírio Libanês, São Paulo, SP, and Department of Digestive Tract Diseases of Hospital Araújo Jorge (Hospital do Câncer) of the Associação de Combate ao Câncer em Goiás, Goiânia, GO.

Creative Commons License Todo o conteúdo deste periódico, exceto onde está identificado, está licenciado sob uma Licença Creative Commons