Budd-Chiari syndrome (BCS) is a rare disorder defined as a hepatic venous outflow obstruction at any level between the hepatic veins and the right atrium [1] but generally implies thrombosis of the hepatic veins and/or the intrahepatic or suprahepatic inferior vena cava (IVC). Up to 50% of all cases of BCS are due to chronic myeloproliferative disorders like polycythemia vera (PV) [2] or coagulopathies like factor V (Leiden) gene mutation [3–5]. The clinical presentation is highly variable but may be categorized as acute and perhaps fulminant hepatic failure, as subacute without evidence of cirrhosis or as chronic with evidence of portal hypertension and cirrhosis.

Treatment depends on the underlying cause, the anatomic location, the extent of the thrombotic process and the severity of liver disease. Treatment options can be divided into medical treatment including anticoagulation and thrombolysis [6–8], radiological procedures such as angioplasty [9] and transjugular intrahepatic porto-systemic shunt (TIPS) [10–14] and surgical procedures including porto-systemic shunting (PSS) [15–17] and orthotopic liver transplantation (OLT) [18, 19]. Anticoagulation alone is unlikely to lead to sufficient recanalization of occluded vessels, or development of adequate collateral circulation. However, satisfactory long-term survival with only medical therapy has been reported [6, 20]. Recent data failed to show a favorable impact of PSS on survival [17, 21], while TIPS has shown encouraging results [11–13, 22–25].

This may give rise to redefine the role of OLT which may now be preserved for patients failing TIPS. The present treatment recommendations of BCS [26, 27] are not based on randomized studies but on a small number of retrospective studies [8, 25, 27] and one prospective study [20].

The aim of this single centre retrospective study was to further enlarge the body of patients evaluated with the intention to allow steady optimization of present treatment strategies.

In this study, we assessed the clinical course and long term outcome of patients with BCS referred to our university hospital analysing the data of all patients admitted with a primary diagnosis of BCS between 1988 and 2008.

Sequential use of increasingly invasive procedures based upon the clinical response has been recommended for the treatment of BCS [8, 25, 27]. In one study in 14 patients, treated with diuretics and anticoagulation alone, a mortality rate of 86% during a period of 6 months was reported [36], however acceptable long term survival rates with only medical therapy has been reported in other trials [37]. Surprisingly, in the recently published first prospective study [20] nearly 50% of the patients were managed conservatively, too. But there may be limitations of this study; on the one hand the study was not restricted to patients with severe disease and on the other hand the median follow-up was only 17 months. In our present trial only 2 patients have not had an invasive therapeutic procedure and are still without clinical symptoms under anticoagulation therapy. Both patients have incomplete hepatic venous outflow obstruction and presented with chronic or subacute BCS. In addition, in one of these two patients the follow-up period of half a year is relatively short. On the basis of this very limited experience, we may confirm the value of medical treatment alone in patients with limited disease. In contrast, in the patients with severe disease or symptomatic cirrhosis, medical treatment was not effective and interventional treatment (TIPS) or transplantation had to be applied.

Different from the last Baveno consensus [26] treatment was only partly applied in a stepwise manner. On the basis of radiological imaging and the severity of clinical presentation a decision concerning adequate treatment was reached. In our collective 13 patients underwent TIPS as initial therapy. In the follow-up, one patient died in the course of the underlying hematologic disease so that an overall survival of 92% was achieved. In all patients transplantation could be avoided. Similar 5 year transplant free survival rates between 77 and 100% after TIPS were observed in other studies [11–13, 22–25]. These results seem to justify primary TIPS treatment in patients with severe disease without awaiting failure of a previous treatment step. Both the high survival and the prevention of OLT in the patients initially treated with TIPS demonstrate an advantage over a stepwise strategy recommended by Baveno and applied in the study by Plessier et al. [8]. Although, the latter found high survival rates using therapeutic procedures by order of increasing invasiveness, the need for transplantation after TIPS was high (38%) compared to our study. This may be due to the fact that TIPS is more efficient when inserted early. Therefore, our results suggest that TIPS should no longer be considered as a bridge to OLT but as a definitive treatment option in BCS, especially when prompt intervention and a high grade of interventional experience is available. With regard to the current situation of liver donor shortage these results are important.

Although, our protocol involved resumption of anticoagulation immediately after TIPS placement, 85% of our patients had at least one reintervention within a mean follow-up of 6 years. In all these patients TIPS revision was technically successful and effective. However, TIPS dysfunction seems to be a common problem [38, 39]. Since, covered stents have a considerable advantage over bare stents with a lower dysfunction rate for the treatment of BCS patients [40], covered stents should be preferred.

Four patients underwent OLT as initial therapy. Two were transplanted before TIPS had been introduced as a treatment option of BCS and two had cirrhosis with signs of chronic liver failure. Survival following OLT depends upon the underlying cause of BCS and the patients condition at the time of transplantation [41]. A large series with 510 patients [42] found a 3 year patient survival of 85% after introduction of the Model for End-Stage Liver Disease (MELD) score versus a 3 year patient survival of 73% in the pre-MELD era. In two other studies [41, 43] 10 year survival rates were reported between 69 and 68%, respectively. It has been suggested, that the outcome of OLT in BCS patients does not differ from that of other etiologies of liver failure when adequate longterm anticoagulation is administered [44]. Even in patients with myeloproliferative diseases survival rates are similar to those in patients with other underlying etiologies [41]. Although, the number of patients transplanted in the present trial does not allow any conclusions on survival, the overall survival rate of 75% during a mean follow-up period of 11 years is comparable to the reported results.

Each patient who underwent OLT as initial therapy, had to undergo re-OLT, which is one of the most important results of our study. All patients had graft failure due to thrombotic or vascular complications. These complications have been described in other series and severe thrombotic complications occurred despite routine early posttransplantation anticoagulation [8, 41]. In accordance to other reports, these findings support the concept that aggressive anticoagulation as early as possible may reduce the risk of thrombembolic complications as well as the late recurrence of BCS after OLT [41, 43].

However, it seems likely, that in some cases thrombembolic complications will not be preventable, despite of aggressive anticoagulation as observed in one patient in our series.

In conclusion, therapy for BCS usually requires TIPS or OLT. The treatment modality is dependent on duration of illness, extent of thrombosis and degree of liver dysfunction. TIPS is successful as initial therapy as it promotes clinical improvement in the long run even though shunt revisions are generally needed. In the present study the necessity of subsequent OLT was remarkably low and thus TIPS could be regarded as definitive treatment option in BCS. OLT in BCS is associated with an increased risk of thrombembolic complications and early graft failure in spite of consequent anticoagulation therapy.