568 Gut 1999;44:568–574 Managing Budd-Chiari syndrome: a retrospective review of percutaneous hepatic vein angioplasty Gut: first published as 10.1136/gut.44.4.568 on 1 April 1999. Downloaded from and surgical shunting

N C Fisher, I McCaVerty, M Dolapci, M Wali, J A C Buckels, S P OlliV, E Elias

Abstract erable diYculties for the clinician, as a range of Background—The role of percutaneous therapeutic options have been advocated. Most hepatic vein angioplasty in the manage- published series have advocated surgical porto- ment of Budd-Chiari syndrome has not systemic shunting as the most appropriate been well defined. Over a 10 year period at option for patients with stable liver function, our unit, we have often used this technique with liver transplantation reserved for those in cases of short length hepatic vein steno- presenting with fulminant or end sis or occlusion, reserving surgical meso- stage chronic .1–6 However, percu- caval shunting for cases of diVuse hepatic taneous hepatic vein angioplasty (PHVA) may vein occlusion or failed angioplasty. also achieve symptomatic relief in patients with Aims—To review the outcome of angio- BCS caused by short length hepatic vein plasty and surgical shunting to define stenosis or occlusion. Since first described in 78 their respective roles. the early 1980s, we and others have de- —All patients treated by angio- scribed a favourable outcome for patients Patients 9–14 plasty or surgical shunting for non- treated with this technique. Successful malignant hepatic vein obstruction over a PHVA restores physiological hepatic venous ten year period from 1987 to 1996. drainage and may thus be theoretically prefer- able to surgical shunting, with the attendant Methods—A case note review of pretreat- ment features and clinical outcome. operative risks and diversion of blood from the liver resulting from the latter procedure. How- Results—Angioplasty was attempted in 21 patients with patent hepatic vein branches ever, angioplasty is sometimes a challenging and was succesful in 18; in three patients procedure for the radiologist and even when successful may be complicated by recurrent treatment was unsuccessful and these venous stenosis. http://gut.bmj.com/ patients had surgical shunts. Fifteen pa- As many of the patients treated with PHVA tients were treated by surgical shunting at our centre may otherwise have been consid- only. Mortality according to definitive ered suitable candidates for surgical shunting, treatment was 3/18 following angioplasty we have carried out a retrospective comparison and 8/18 following surgery; in most cases of patients treated by each of these procedures this reflected high risk status prior to in order to clarify their respective roles. treatment. Venous or shunt reocclusion

rates were similar for both groups and on September 24, 2021 by guest. Protected copyright. were associated with subtherapeutic war- Patients and methods farin in half of these cases. Most surviving PATIENTS patients in both groups are asymptomatic Between 1984 and 1996, 70 patients with although one surgical patient has chronic clinically and radiologically validated BCS . were referred to our centre for management. The primary treatment for these patients was Conclusion—With appropriate case selec- Liver Unit, Queen surgical shunting (n=15),PHVAor other angio- Elizabeth Hospital, tion, many patients with Budd-Chiari syn- drome caused by short length hepatic vein plasty (n=25), liver transplantation (n=10), Birmingham, UK transjugular intrahepatic portosystemic shunt- N C Fisher stenosis or occlusion may be managed suc- ing (TIPSS) (n=4), or medical treatment M Dolapci cessfully by angioplasty alone. Medium (n=9) (fig 1). In the remaining cases, definitive M Wali term outcome is good following this proce- J A C Buckels treatment was not given either because of early dure provided that anticoagulation is E Elias death (n=5) or the patient refused treatment maintained. Further follow up is required (n=2). Most of the patients undergoing liver to assess for definitive benefits but we sug- Department of transplantation were treated relatively early in Radiology, Queen gest that this should be included as a valid our series; poor results from this procedure led Elizabeth Hospital, initial approach in the algorithm for man- us to choose surgical shunting or radiological Birmingham, UK agement of Budd-Chiari syndrome. I McCaVerty angioplasty in preference to transplantation in (Gut 1999;44:568–574) S P OlliV patients without end stage liver disease. We have also recently adopted the TIPSS proce- Correspondence to: Keywords: Budd-Chiari syndrome; short length Dr N C Fisher, Department hepatic vein stenosis; hepatic vein occlusion; of , New percutaneous hepatic vein angioplasty; mesocaval Abbreviations used in this paper: BCS, Cross Hospital, shunt Wolverhampton WV10 0QP, Budd-Chiari syndrome; HV, hepatic vein; IVC, inferior UK. vena cava; PHVA, percutaneous hepatic vein angioplasty; SLHVS, short length hepatic vein stenosis; Accepted for publication The management of symptomatic Budd-Chiari TIPSS, transjugular intrahepatic portosystemic 7 October 1998 syndrome (BCS) continues to present consid- shunting. Managing Budd-Chiari syndrome 569

laboratory features prior to treatment and on Liver transplant follow up as of December 1997. Statistical PHVA analysis was used to compare clinical and labo-

ratory variables before and after treatment, Gut: first published as 10.1136/gut.44.4.568 on 1 April 1999. Downloaded from Surgical shunt using the Mann-Whitney U test and Fisher’s TIPSS exact test as appropriate.

1984 1988 1992 1996 PRETREATMENT CLINICAL AND INVESTIGATIVE Year FEATURES Figure 1 Definitive procedures for BCS at Queen The spectrum of underlying diseases leading to Elizabeth Hospital Liver Unit, 1984–1996. BCS was similar for each group and is summa- Table 1 Predisposing diseases leading to BCS (divided rised in table 1; comparison of other features is according to initial treatment) summarised in table 2. All patients had symp- tomatic BCS at presentation. The frequency of Surgical PHVA shunting clinical symptoms and laboratory abnormali- Disease (n=21) (n=15) ties was similar in both groups, although patients in the surgical shunting group tended Myeloproliferative disease 4 6 to have a higher incidence of hepatic encepha- Protein C deficiency 6 1 Protein S deficiency 1 0 lopathy (NS), higher serum bilirubin levels Antithrombin III deficiency 0 1 (p<0.01), and more severe ischaemic changes Paroxysmal nocturnal 01 on liver histology as judged by a semiquantita- haemoglobinuria Oral contraceptive usage 1 0 tive scoring system (p<0.05). The majority of Factor V Leiden heterozygote* 2 1 patients (17/21 in the PHVA group and 14/15 Unknown 9 6 in the surgical group) were of Western *Each of these patients found to be heterozygous for the factor European origin. Radiological imaging (Dop- V Leiden mutation also had a concurrent predisposing cause for pler sonography and contrast venography) BCS. showed short length hepatic vein stenosis dure in preference to surgical shunting; how- (SLHVS) in all 21 patients undergoing at- ever follow up for these patients has been short. tempted PHVA. Doppler ultrasound reports Our present review is therefore restricted only were available for 17 patients and suggested to those patients undergoing percutaneous patent intrahepatic vein branches in 16 (with radiological angioplasty or surgical shunting, more than one patent hepatic vein branch visu- for non-malignant hepatic vein obstruction. alised in 10). In two patients, transjugular The 25 patients treated with radiological angio- venography subsequently showed further pat- plasty included two treated with metal stent ent hepatic vein branches. All patients under-

insertions for palliation of hepatic vein occlu- went attempted transjugular hepatic venogra- http://gut.bmj.com/ sion from solid malignancies, and two with iso- phy; in 12 of these patent intrahepatic vein lated inferior vena caval obstruction; these branches were shown while in the remaining patients were excluded from our analysis, leav- nine patients hepatic veins were only confirmed ing 21 patients in the PHVA and 15 in the sur- by transhepatic venography. gical shunting groups. For the purposes of this In patients referred directly for surgical review we recorded clinical, radiological, and shunts, 4/15 had patent hepatic vein branches visible on Doppler sonography. In one case this Table 2 Clinical and investigative features at presentation was confirmed by transjugular venography (but on September 24, 2021 by guest. Protected copyright. PHVA Surgical shunting Significance angioplasty was not attempted) and in the (n=21) (n=15) (p value) remaining three cases hepatic veins were not accessible by the transjugular route. All of these Age1 31 (16–65) 36 (20–57) NS2 cases were relatively early in our series and Duration of history, months1 3 (0–84) 3 (0–204) NS2 Clinical features4 none of the patients underwent attempted Ascites 18 15 NS3 transhepatic venography. The remaining 11 Variceal haemorrhage 4 3 NS3 Hepatic encephalopathy5 58 NS3 patients in the shunt group had diVuse hepatic Muscle wasting6 76 NS3 venous occlusion shown by Doppler sonogra- Radiological features4 phy (four of these also underwent hepatic Short length hepatic vein stenosis 21 4 venography, with no veins shown). Three of 21 IVC occlusion 3 2 patients in the PHVA group and 2/15 in the Laboratory findings1, 7 shunt group had associated inferior vena caval Serum bilirubin (µM/l) 31 (6–179) 60 (20–255) p<0.012 Serum AST (U/l) 48 (12–2410) 69 (23–504) NS2 (IVC) stenosis with a pressure gradient of at Serum albumin (g/l) 34 (20–45) 32 (26–39) NS2 least 10 mm Hg across the stenosis; in several Serum creatinine (µM/l)8 87 (47–247) 104 (61–522) remaining patients a degree of IVC stenosis Prothrombin INR 1.4 (1.0–2.9) 1.4 (1.2–3.2) NS2 Histological features1, 9 was present but without a significant pressure Centrilobular necrosis score 1 (1–3) 2 (0–3) p<0.052 gradient. None of the patients in either group Fibrosis score 2 (0–3) 1 (0–3) NS2 had associated portal vein occlusion. 1Median values (range given in brackets). 2Mann-Whitney U test; 3Fisher’s exact test. p values >0.05 are denoted as not significant (NS). PERCUTANEOUS HEPATIC VEIN ANGIOPLASTY 4Absolute numbers. 5 The techniques used for this procedure have Clinical grade 1 or higher. 13 14 6Defined where clinically obvious. been described extensively elsewhere. 7Normal laboratory values: bilirubin, 1–22 µM/l; AST, 5–43 U/l; albumin, 34–51 g/l; creatinine, Briefly, the procedure was generally under- 60–100 µM/l. taken where Doppler ultrasound scanning or 8Two patients in shunt group treated with haemodialysis prior to surgery. 9Centrilobular necrosis and fibrosis scores: 0=none; 1=mild; 2=moderate; 3=severe. contrast venography identified a short length AST, aspartate aminotransferase. stenosis or occlusion of the hepatic vein with a 570 Fisher, McCaVerty, Dolapci, et al

dosage adjustment to maintain the pro- A D thrombin INR between 2.5 and 4.0. Further RL RA ultrasound with Doppler flow imaging was car-

ried out at regular (six monthly) intervals, and Gut: first published as 10.1136/gut.44.4.568 on 1 April 1999. Downloaded from we have adopted a policy of contrast venogra- phy at annual intervals for most patients with O repeat angioplasty if any recurrent stenosis has occurred. If PHVA was technically impossible HV or failed to relieve symptoms, the patient was referred for surgical portosytemic shunting. IVC SURGICAL SHUNTING B E Patients were generally selected for shunting if they had no identifiable hepatic veins amenable to angioplasty, or if thrombolysis and/or angioplasty had failed to relieve symptoms, provided that the portal and mesenteric veins were considered to be patent. Mesocaval shunts were used preferentially, with mesoatrial shunts reserved for patients with suprahepatic IVC occlusion. Initially patients were shunted using woven Dacron grafts (n=6); we later used polytetrafluoroethylene grafts (n=5) and have more recently used autologous jugular vein C F grafts (n=7). All shunt procedures except one were performed by the same surgeon (JACB). All patients were anticoagulated postopera- tively with heparin followed by warfarin, with dosage adjusted to maintain the prothrombin INR between 2.5 and 4.0. Patients with exten- sive portal vein or end stage liver disease were not considered for shunting.

Results PROCEDURES AND OUTCOME DURING FIRST

Figure 2 Schematic illustration of combined transjugular/transhepatic balloon angioplasty ADMISSION http://gut.bmj.com/ procedure. (A) Hepatic vein occlusion. O = occlusion, HV = hepatic vein, IVC = inferior vena cava, RA = right atrium, RL = right lung. (B) 5F or 6F sheath introduced into HV PHVA under ultrasound guidance. (C) Guidewire introduced into IVC by manipulation or by In this group, 10 patients were treated by trans- puncture. (D) Transjugular snare wire used to capture guidewire from above. (E) jugular angioplasty alone (with more than one Guidewire pulled proximally out of jugular vein. (F) Balloon catheter introduced from above and angioplasty carried out. therapeutic transjugular procedure in five patients) and 11 were treated by combined patent venous lumen proximally; this lesion has transjugular and transhepatic procedures (with been defined as a short length hepatic vein more than one therapeutic transhepatic proce- stenosis.15 The initial therapeutic approach was dure in four, including one patient with associ- on September 24, 2021 by guest. Protected copyright. transjugular in all cases; in selected cases where ated caval stenosis who underwent trans- this approach failed further attempts were jugular, transfemoral, and transhepatic made via a transhepatic route with insertion of procedures). Six of 21 patients had thrombo- a percutaneous transhepatic guidewire and lytic agents infused at the site of occlusion fol- snaring of the guidewire with a wire loop lowing transvenous catheterisation, and 4/21 inserted from the transjugular route. Trans- had hepatic vein stents inserted. Intrahepatic jugular balloon angioplasty to a maximum of luminal venous pressure measurements both 14 mm diameter was then done. In selected before and after angioplasty were available for cases repeated procedures were undertaken to seven patients; median pressure before treat- confirm or improve hepatic venous outflow, ment was 34 mm Hg (range 25–39); after and in selected cases where residual treatment median pressure was 19 mm Hg was visualised the transvenous catheter was left (range 12–30), with a median reduction in in place for regional thrombolysis. If there was pressure of 15 mm Hg (range 7–24). Intralumi- early reocclusion or severe residual stenosis, a nal pressures were not generally measured in self expanding metal stent was inserted to patients undergoing transhepatic procedures. maintain venous patency. Figures 2, 3, and 4 Early mortality was 2/18 (fig 5); both deaths illustrate these techniques. Angioplasty was occurred early in our series (in 1988 and 1990 considered to have been successful if restora- respectively). The first death was in a patient tion of or clear improvement in hepatic venous with combined hepatic vein (HV) and IVC outflow was shown by contrast occlusion with hepatic encephalopathy and and if the patients’ symptoms stabilised or ; a hepatic vein-right improved. In the first three cases of PHVA, atrial channel was formed by the transjugular patients did not receive anticoagulant treat- route but the patient died within 24 hours from ment following the procedure. Thereafter we circulatory failure; permission for postmortem revised our management to include anticoagu- examination was refused. The second case was lation, with heparin followed by warfarin with a patient who had undergone transjugular Managing Budd-Chiari syndrome 571 Gut: first published as 10.1136/gut.44.4.568 on 1 April 1999. Downloaded from

Figure 3 Angiograms obtained in a 37 year old man undergoing combined transjugular/transhepatic balloon angioplasty with subsequent stent insertion. (A) Percutaneous transhepatic angiogram at the time of the first procedure (thick arrow, area of occlusion; thin arrow, transhepatic catheter). (B) Image obtained after balloon angioplasty, with some improvement in venous drainage (thick arrow, transjugular catheter; thin arrow, irregular hepatic venous lumen). (C) Image obtained after stent insertion, showing further improvement in venous drainage (thick arrow, stent; thin arrow, metal coil inserted in liver along track of percutaneous sheath).

angioplasty of a short length hepatic vein however this branch became occluded with stenosis but developed early recurrence of ste- progression of symptoms and so the patient

nosis, and died of a variceal haemorrhage while underwent emergency surgery. The third http://gut.bmj.com/ further procedures were being considered. patient had transhepatic angioplasty with the Postmortem examination confirmed hepatic catheter left in situ for local thrombolysis; vein occlusion with evidence of recent throm- however the catheter became blocked and bus. There were no other complications plans for further procedures were abandoned. relating to the procedures or anticoagulation. The median length of hospitalisation for sur- Three patients were referred for surgery viving patients was eight days (range 2–41). because of failed angioplasty. In the first patient on September 24, 2021 by guest. Protected copyright. transhepatic angioplasty was attempted but the Surgical shunting hepatic vein occlusion could not be traversed. In this group, mesocaval shunts were used for The second patient had transjugular angio- 16/18 cases (including the three patients who plasty of one hepatic vein branch, and was had failed angioplasty). Mesoatrial shunts were being considered for a transhepatic procedure used in two cases where complete IVC on a second patent hepatic vein branch; occlusion was present. Postoperative mortality (less than 30 days) was 5/18, including both patients undergoing a mesoatrial shunt (fig 5). Four of five patients who died had hepatic encephalopathy and 2/5 were receiving renal support prior to surgery. Four of five of these cases had diVuse hepatic vein occlusion while

100

80 PHVA

60 Shunt 40

20 Percentage surviving 0 321 54 Years following procedure Figure 5 Actuarial survival curves following PHVA or surgical shunting for BCS. 572 Fisher, McCaVerty, Dolapci, et al

Table 3 Current clinical status in survivors (divided according to definitive treatment) in the Middle East; this patient died. The remaining five cases of shunt occlusion were PHVA Surgical shunting Significance managed by percutaneous transfemoral dilata- (n=15) (n=10) (p value)

tion (two patients, with stenting of the stenosis Gut: first published as 10.1136/gut.44.4.568 on 1 April 1999. Downloaded from Duration of follow up, months2 36 (12–103) 67 (18–115) NS1 in one case), attempted portacaval shunt, Return to employment or education 9/9 8/9 insertion of TIPSS at another centre, and Receiving treatment for myeloproliferative disease 1 3 thrombolytic treatment (one patient each). Clinical features3 The attempted portacaval shunt and the Ascites 1 2 thrombolytic treatment were both unsuccessful Oedema 1 2 Diuretic treatment 2 3 and these patients died (fig 5). There were no Hepatic encephalopathy 0 1 other deaths in this group. Laboratory findings2 Serum bilirubin (µM/l) 13 (4–39) 28 (14–42) p<0.01 Serum AST (U/l) 24 (15–44) 34 (25–56) p<0.05 CURRENT CLINICAL STATUS IN SURVIVORS Serum albumin (g/l) 45 (37–49) 41 (34–44) p<0.01 Duration of follow up for survivors has been longer for the group undergoing surgical 1Mann-Whitney U test; p values >0.05 are denoted as not significant (NS). 2Median values (range given in brackets). shunts (n=10; median 67 months, range 3Absolute numbers. 18–115) compared with the PHVA group AST, asparate aminotransferase. (n=15; median 36 months, range 12–103, the fifth had a short segment stenosis but had diVerence not significant). Table 3 summarises not had angioplasty attempted. In all cases clinical and laboratory features. There are no death was due to development of multiorgan significant diVerences in the incidence of failure, occurring a median of two days (range residual symptoms, which is low in each group, 2–16) postoperatively. The median length of and none of the patients are clinically jaun- hospitalisation for surviving patients was 15 diced although serum levels of bilirubin remain days (range 8–28). higher (p<0.01) in survivors of surgical shunt- ing. In the surgical shunting group, one patient LATE MORBIDITY AND MORTALITY is being treated for chronic hepatic encepha- PHVA lopathy and 2/10 have developed isolated Complete or near complete occlusion of previ- hepatic nodules noticed at routine follow up ously recanalised hepatic veins occurred in ultrasound scanning. Both patients have 6/16 surviving patients, leading to recurrence undergone targeted to determine of symptoms in 4/6 cases. The median delay to the nature of the nodules and histology showed reocclusion was 2 years 11 months (range 1 normal or regenerative liver in each case. The year 8 months to 5 years 3 months). Recurrent majority or all of the patients in each group occlusion was associated with subtherapeutic who were in employment or full time education

or absence of warfarin treatment in 3/6 cases. prior to their treatment have been able to http://gut.bmj.com/ Further angioplasty was successful in 5/6 cases; return to these activities. one of these was complicated by haemobilia which was succesfully treated by embolisation Discussion of an intrahepatic arterial pseudoaneursym. In Our review suggests that excellent medium to the remaining case reocclusion was asympto- long term recovery from BCS may be expected matic and no further attempts at recanalisation in the majority of patients in whom hepatic were made. Late death occurred in 1/16 venous drainage is restored by PHVA. The patients (fig 5). This was a patient with lower mortality in this series in comparison on September 24, 2021 by guest. Protected copyright. combined hepatic vein and IVC stenosis who with surgical portosystemic shunting is prob- had two stents inserted at the time of her initial ably largely reflective of the more severe risk procedure; the upper stent later migrated factors at presentation in patients who were towards the atrium leading to recurrent tachy- managed by shunting alone. A lower mortality cardia. Combined cardiac and abdominal sur- in patients with BCS caused by SLHVS gery to remove this stent was undertaken 18 compared with those with diVuse hepatic vein months later, but the patient died from occlusion has been recognised previously.12 15 multiorgan failure after prolonged hospitalisa- Nevertheless, some patients in our series who tion. There were no other episodes of serious were managed successfully with PHVA had morbidity in this group although most patients presented with advanced liver disease charac- are undergoing annual transjugular venograms terised by variceal and hepatic en- to check for hepatic venous patency, with cephalopathy, which suggests that the role for further dilatation if any restenosis has oc- PHVA in BCS is not limited merely to those curred. with mild disease. From this retrospective comparison we are Surgical shunting unable to draw firm conclusions regarding the Shunt occlusion occurred in 6/13 surviving relative benefits of PHVA and surgical shunting patients, leading to recurrent symptoms in all in patients wtih SLHVS. Long term morbidity cases. The median delay to reocclusion was 2 may occur from both procedures and in years 7 months (range 2 months to 7 years 10 patients undergoing PHVA we require annual months), involving four Dacron and two jugu- check venograpy to ensure patency of the lar grafts. Shunt occlusion was associated with hepatic vein. However, we believe that restora- subtherapeutic warfarin dosage in 3/6 cases (all tion of physiological venous drainage from the Dacron shunts). One of these cases was a liver is preferable to the diversion of blood from patient who was unable to obtain warfarin the liver that is consequent on portosystemic treatment after returning to his native country shunting, and may be associated with a lower Managing Budd-Chiari syndrome 573

ence and that of others suggests that, following Symptomatic Budd-Chiari angioplasty, hepatic vein webs are more likely syndrome suspected to restenose compared with caval webs.22

We suggest that in patients with symptomatic Gut: first published as 10.1136/gut.44.4.568 on 1 April 1999. Downloaded from BCS in whom patent intrahepatic vein Doppler US or other non-invasive imaging branches can be identified by Doppler studies, (e.g. contrast CT or MR) strong consideration should be given towards therapeutic radiological procedures to restore Abnormal venous outflow Normal venous outflow hepatic venous outflow and we propose an algorithm for management of BCS incorporat- Hepatic venography with intraluminal Consider other diagnosis ing this therapeutic approach (fig 6). If BCS is pressure measurements suspected clinically, ultrasound and Doppler studies should be done and this is likely to con- Short length hepatic Diffuse venous occlusion firm or refute the diagnosis in the majority of 23 vein stenosis or occlusion cases. If ultrasound confirms BCS or if there is uncertainty then contrast hepatic venogra- phy should be undertaken and angioplasty Angioplasty Mesenteric angiography1 +/– anticoagulation attempted if SLHVS is confirmed. This may require repeated, including transhepatic, pro- cedures and/or stent insertion as described Successful Failed PV + MV patent PV + MV occluded earlier in order to obtain sustained improve- ment in hepatic venous outflow. If this Transhepatic approach Consider any porto- approach is successful then clinical recovery systemic shunt possible2 should follow although full recovery is occa- sionally slow (as illustrated in fig 4). A 2 Successful Failed Portosystemic shunt portosystemic shunt should be considered (TIPSS or surgical) where angioplasty is unsuccessful or clearly fails to resolve symptoms, or if diVuse hepatic Anticoagulate + arrange vein occlusion is present, provided that the further elective procedures portal and mesenteric veins are patent. The Figure 6 Suggested algorithm for management of BCS. 1Contrast CT or MR may obviate choice of portocaval, mesocaval, or mesoatrial need for direct angiography. 2Liver transplantation may be appropriate for fulminant or end stage disease. shunt may depend on the preference of the surgeon and the presence or absence of a caval risk of liver failure and/or hepatic encephalopa- stenosis; similar outcomes have been reported thy. One survivor of surgical shunting in our for each of these procedures,1246although the http://gut.bmj.com/ series has chronic hepatic encephalopathy, a use of the mesenteric vein rather than the por- recognised occasional sequel of portocaval tal vein for shunting will be less likely to com- shunting for BCS which has not been reported 16 promise any future liver transplant in the rare following PHVA. Furthermore, restoration of circumstance where this may be required. As hepatic venous drainage may theoretically an alternative to a surgical shunt, a TIPSS may reduce the extent of architectural disturbance be considered; there have been few reports of that normally follows BCS treated conserva- this procedure for BCS but the experience of on September 24, 2021 by guest. Protected copyright. tively or by portosytemic shunting, as nodular ourselves and others is that this may relieve regenerative hyperplasia is presumed to be a symptoms eVectively in selected cases.24 If por- consequence of disordered vasculature such as tal vein thrombosis is present at the outset, that occurring in BCS.17 Persisting fibrosis and therapeutic options are limited; in this situation nodule formation have been reported in a high proportion of long term survivors of BCS mesenteric angiography and early anticoagula- treated by conventional means, with develop- tion may help while consideration is given to any possible portosystemic shunting procedure ment of in some 25 cases.18 19 The duration of follow up for patients or liver transplantation. Following shunt sur- treated with PHVA in our series has been rela- gery or angioplasty long term anticoagulation tively short, but at present we have not should be given (after screening for any possi- encountered any cases of nodule formation in ble thrombophilic disorder) with close radio- these patients on routine follow up ultrasonog- logical follow up and further interventional raphy; this is now the subject of a prospective procedures as described earlier. Patients with evaluation by computed tomography (CT) or advanced liver failure and diVuse hepatic vein magnetic resonance (MR) imaging. occlusion should be considered for liver trans- Our experience of reocclusion of hepatic plantation. veins following angioplasty where warfarin was In summary, we have found that patients not given (in early cases) or was subtherapeutic with symptomatic BCS and moderate to severe has led us to believe that strict anticoagulation liver failure may be managed successfully by (with prothrombin INR greater than 2.5) is PHVA with appropriate case selection. Short mandatory to help prevent reocclusion, as is and medium term morbidity and mortality for the case for surgical shunts.20 Excellent long this procedure compare favourably with surgi- term results following angioplasty for caval cal shunting, although further follow up is webs have been reported without the use of required to assess for putative long term warfarin in oriental series,21 22 but our experi- benefits. 574 Fisher, McCaVerty, Dolapci, et al

We are grateful to our surgical, radiological, and medical 14 Cooper S, OlliV SP, Elias E. Recanalisation of hepatic veins colleagues for allowing us to report patients under their care, by a combined transhepatic and transjugular approach in 3 and to Bridget Gunson for help with statistical analysis. cases of Budd-Chiari syndrome. J Interv Radiol 1996;11:9– 13. 15 Valla D, Hadengue A, El Younsi M, . Hepatic venous 1 OrloV MJ, Girard B. Treatment of Budd-Chiari syndrome et al Gut: first published as 10.1136/gut.44.4.568 on 1 April 1999. Downloaded from by side-to-side portacaval shunt: long-term results. Hepatol- outflow block caused by short-length hepatic vein stenoses. ogy 1986;6:1181. 1997;25:814–19. 2HalVG, Todo S, Tzakis AG, et al. Liver transplantation for 16 Pezzuoli G, Spina GP, Opocher E, et al. Portocaval shunt in the Budd-Chiari syndrome. Ann Surg 1990;211:43–9. the treatment of primary Budd-Chiari syndrome. Surgery 3 Klein AS, Sitzmann JV, Coleman J, et al. Current 1985;98:319–23. management of the Budd-Chiari syndrome. Ann Surg 17 Maia de Sousa JM, Portmann B, Williams R. Nodular 1990;212:144–9. regenerative hyperplasia of the liver and the Budd-Chiari 4 Panis Y, Belghiti J, Valla D, et al. Portosystemic shunt in syndrome. J Hepatol 1991;12:28–35. Budd-Chiari syndrome: long-term survival and factors 18 Vons C, Belghiti J, Vilgrain V, et al. Occurrence of hepatic aVecting shunt patency in 25 patients in Western countries. nodules in patients with Budd-Chiari syndrome is a Surgery 1994;115:276–81. frequent event. J Hepatol 1997;26(Suppl.1):307. 5 Tilanus HW. Budd-Chiari syndrome [review]. Br J Surg 19 OrloV MJ, Johansen KH. Treatment of Budd-Chiari 1995; :1023–30. 82 syndrome by side-to-side portacaval shunt: experimental 6 Hemming AW, Langer B, Greig P, et al. Treatment of Budd- Chiari syndrome with portosystemic shunt or liver and clinical results. Ann Surg 1978;188:494–512. transplantation. Am J Surg 1996;171:176–81. 20 Bismuth H, Sherlock DJ. Portasystemic shunting versus 7 Meier WL, Waller RM, Sones PJ. Budd-Chiari web treated liver transplantation for the Budd-Chiari syndrome. Ann by percutaneous transluminal angioplasty. Am J Roentgenol Surg 1991;214:581–9. 1981;137:1257–8. 21 Yang XL, Cheng TO, Chen CR. Successful treatment by 8 Nishikawa M, Seki K, Miyoshi S. Treatment of the percutaneous balloon angioplasty of Budd-Chiari syn- Budd-Chiari syndrome in polycythaemia rubra vera by drome caused by membranous obstruction of inferior vena repeated percutaneous transluminal angioplasty of a cava: 8-year follow-up study. J Am Coll Cardiol 1996;28: hepatic vein stenosis. Postgrad Med J 1982;58:511–14. 1720–4. 9 Sparano J, Chang J, Trasi B, et al. Treatment of the 22 Kohli V, Pande GK, Dev V, et al. Management of hepatic Budd-Chiari syndrome with percutaneous transluminal venous outflow obstruction. Lancet 1993;342:718–22. angioplasty. Am J Med 1987;82:821–8. 23 Millener P, Grant EG, Rose S, et al. Color Doppler imaging 10 Vickers CR, West RJ, Hubscher SG, et al. Hepatic vein webs findings in patients with Budd-Chiari syndrome: correla- and resistant ascites. Diagnosis, management and implica- tion with venographic findings. Am J Roentgenol 1993;161: tions. J Hepatol 1989;8:287. 307–12. 11 Martin LG, Henderson JM, Millikan WJ. Angioplasty for long-term treatment of patients with Budd-Chiari syn- 24 Ochs A, Sellinger M, Haag K, et al. Transjugular drome. Am J Roentgenol 1990;154:1007. intrahepatic portosystemic stent-shunt (TIPS) in the treat- 12 Mahmoud AEA, Mendoza A, Meshikhes AN, et al. Clinical ment of Budd-Chiari syndrome. J Hepatol 1993;18:217– spectrum, investigations and treatment of Budd-Chiari 25. syndrome. QJMed1996;89:37–43. 25 Mahmoud AEA, Helmy AS, Billingham L, et al. Poor prog- 13 GriYth JF, Mahmoud AEA, Cooper S, et al. Radiological nosis and limited therapeutic options in patients with intervention in Budd-Chiari syndrome: techniques and Budd-Chiari syndrome and portal venous system thrombo- outcome in 18 patients. Clin Radiol 1996;51:775–84. sis. Eur J Gastroenterol Hepatol 1997;9:485–9. http://gut.bmj.com/ on September 24, 2021 by guest. Protected copyright.