Investigation Of Several Specific Issues Regarding Portal Vein Thrombosis

A significant proportion of portal vein thrombosis is insidious with no or unspecific symptom. Accordingly, this kind of thrombosis has been considered as a rare disease. So far, little is known about portal vein thrombosis, leading to inconsistent ideas on the treatment. In the present study, we investigated several neglected or controversial issues using the data accumulated in our department for many years.Part1:The reasonable classification and its clinical significance for symptomatic portal vein thrombosisPurpose:The classification of portal vein thrombosis is very important for the option of treatment. In general, portal vein thrombosis has been divided into acute and chronic types according to the duration of thrombus existence or the presence of cavernous transformation of portal vein. However, this classification is ambiguous because it is uncertain about the time of thrombus formation, and also the portal vein cavernoma is not always associated with the duration of thrombus existence. In this part of the study, we attempted to clarify the rational of acute and chronic classification for portal vein thrombosis by reviewing the data obtained from60patients with symptomatic portal vein thrombosis.Materials and Methods:Sixty patients,40men and20women, were recruited for the study. The average age was48.0±13.8years old. The upper abdomen was scanned using contrast-enhanced computed tomography (CT). The CT image was measured using Image J program for assessing the degree of occlusion of portal vein and its intrahepatic branches.Results:There was thrombus in portal vein in all60patients, and44of them were associated with thrombosis in intrahepatic branches. Thirty one patients had portal vein cavernoma. The ability of portal vein occlusion to discriminate between portal veins with and without cavernous transformation was examined by construction of receiver operating characteristic (ROC) curves. The cut-off point was>82.9%for portal vein. Logestic analysis showed when the portal vein was occluded>82.9%the likelihood of cavernous transformation was32.5times higher than that was occluded <82.9%(p <0.001). In29patients without porta cavernoma, only4(13.8%) had portal vein occluded>82.9%. In contrast,26of31patients (83.9%) with portal cavernoma had portal vein occluded>82.9%. Five patients with symptoms<2days as well as portal cavernoma were all associated with a>82.9%occlusion of portal vein. However, in8patients with symptoms>20days and no portal cavernoma, no one had portal vein occluded>82.9%. These results suggest that cavernous transformation of portal vein is associated with the degree of portal vein occlusion rather than the symptom duration. The complication of portal hypertension had no significant relationship to portal vein occlusion and cavernous transformation. In addition,24of60patients (40%) could not be classified into acute or chronic thrombosis according to the current classification criteria.Conclusion:There is a close relationship between portal vein occlusion and cavernoma. In patients with a>83%portal vein occlusion, the risk of cavernous transformation of portal vein increased dozens times. In addition, our results showed that about40%of patients could not fall into the category of acute and chronic portal vein thrombosis, implying that this is a defected classification. We suggest that the portal vein thrombosis is better to be categorized into complete and partial types, and each type is subclassified into with and without cavernoma. The different types of thrombosis may convert from one to the other with the development of disease and progress in treatment.Part2:Factors contributing to the failure of intrahepatic portalsystemic shunt creation in the interventional therapy of portal vein thrombosisPurpose:The interventional therapy has become a common means in the management of portal vein thrombosis. However, the highly occluded portal vein may lead to failure of traditional TIPS procedure. Therefore, an intrahepatic portalsystemic shunt (IPS) must be created before TIPS procedure. IPS is able to recanalize the occluded portal vein, assisting in the success of TIPS. In the clinic, IPS can be difficultly created in some patients with thrombosis in portal vein and its intrahepatic branches, which has been paid little attention thus far. In this part of the study, we focused to the factors contributing to the failure of IPS creation.Materials and Methods:Fifty one patients,36men and15women, were included in the study. The average age was47.3±13.9years old. Twenty patients underwent percutenous transhepatic angioplasty (PTA) and31underwent transhepatic assistant TIPS (THA-TIPS). All patients had acute or subacute symptoms when admission, including esophageal varices bleeding, ascites, abdominal pain, diarrhea and fever. The upper abdomen was scanned using contrast-enhanced CT before procedures, and digital subtractive angiography (DSA) was performed before and after procedures as well. The CT image was measured using Image J program for assessing the degree of occlusion of portal vein and its intrahepatic branches. In addition, the degree of portal vein occlusion was also measured on DSA image.Results:IPS creation was successful in43patients and failure in8patients. The symptom duration in patients with IPS failure was significantly longer than those with IPS success (p<0.001). Additionally, the likelihood of portal vein cavernoma was significantly increased in patients with IPS failure compared to those with IPS success (p=0.007). The IPS procedure was successful in all patients without portal vein cavernoma, whereas8patients with portal vein cavernoma were associated with IPS failure. The ability of portal vein and its right branch occlusion to discriminate between IPS success and failure was examined by construction of ROC curves. The cut-off point was>86.0%for right portal branch and>96.0%for portal vein. Logistic analysis showed that the risk of IPS failure was increased11.3times in patients with right portal branch occluded>86.0%(P=0.007) and43.2times in patients with portal vein occluded>96.0%(P=0.001). Both PTA and THA-TIPS procedures significantly increased portal vein patency (p<0.001), but there was no significant difference in portal vein patency between patients with PTA and THA-TIPS treatment.Conclusion:The success of IPS creation is largely dependent on the occlusion of portal vein and its intrahepatic branches. The introduction of the wire is likely to be failed when the portal vein is occluded over96%and/or the right portal branch occluded over86%, leading to a significant increase in the risk of IPS failure. We recommend that the occlusion of90%and80%can be considered as a safety threshold for portal vein and right portal branch, respectively. The risk of IPS failure should be weighed if the occlusion is over these thresholds. PTA and THA-TIPS showed similar results in the treatment of portal vein thrombosis.Part3:Interventional treatment of extensive portal vein thrombosisPurpose:Extensive portal vein thrombosis is indicative of the thrombus in portal vein extending into superior mesenteric vein and/or splenic vein. TIPS approach has been widely used to recanalize the embolized portal vein, but rarely reported for the treatment of extensive portal vein thrombosis. The purpose of the present study was to investigate the effect of TIPS on extensive portal vein thrombosis.Materials and Methods:Twenty one patients (men13, women8, age50.9±12.9) with extensive portal vein thrombosis were treated with traditional TIPS, transhepatic assistant TIPS and transhepatic assistant retrograde TIPS procedures. The degree of occlusion of portal and superior mesenteric veins was measured on DSA images obtained before and after treatment. The mean follow-up period was34.8months.Results:The average interval between initial symptom and the start of treatment was58days (5-540days). Twenty patients underwent successful TIPS procedures and all of them received mechanical thrombectomy. Nineteen patients were subsequently received urokinase thrombolysis. The average dose of urokinase was2.64million units (0.75-5.5million units), and the average administrative time was89.9hours (24-120hours). The degrees of portal vein and superior mesenteric vein occlusion were both decreased significantly after procedures (p<0.001). Recanalization of portal and superior mesenteric veins was started after mechanical thrombectomy and gradually improved after urokinase thrombolysis. The symptoms were obviously relieved in18patients (90%) following procedures. Four patients (20%) had the procedure-related complications, including hematoma beneath hepatic capsule, bloody pleural effusion and encephalopathy, within30days after procedures. No one died because of complications. During the follow-up period, shunt dysfunction occurred in5patients (25%). Patency of the shunt was restored in all patients after revision.Conclusion:Good therapeutic effect can be achieved by various TIPS procedures in the treatment of subacute and chronic extensive portal vein thrombosis. Mechanical thrombectomy is a crucial step for vessel recanalization, which can improve the outcome of thrombolysis by promoting the agent permeating into fragmented thrombus. Accordingly, optimal thrombolysis can be achieved by the use of lower dose of urokinase. In addition, the lower dose of urokinase can reduce the risk of agent-related complications.