Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online before print June 13, 2005, 10.1148/radiol.2361040530
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2361040530v1
236/1/360    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Tesdal, I. K.
Right arrow Articles by Jaschke, W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Tesdal, I. K.
Right arrow Articles by Jaschke, W.
(Radiology 2005;236:360-367.)
© RSNA, 2005

Vascular and Interventional Radiology

Transjugular Intrahepatic Portosystemic Shunts: Adjunctive Embolotherapy of Gastroesophageal Collateral Vessels in the Prevention of Variceal Rebleeding1

I. Kaare Tesdal, MD, Thomas Filser, MD, Christel Weiss, Dr Sc Hum, Eggert Holm, MD, Christoph Dueber, MD and Werner Jaschke, MD2

1 From the Departments of Clinical Radiology (I.K.T., C.D., W.J.) and Statistics (C.W.) and Medical Clinic (T.F., E.H.), Universitätsklinikum Mannheim, Germany. Received March 19, 2004; revision requested May 28; revision received August 20; accepted October 1. Address correspondence to I.K.T., Institut für Radiologie und Nuklearmedizin Städtisches Krankenhaus Friedrichshafen, Röntgenstr 2, 88048 Friedrichshafen, Germany (e-mail: k.tesdal{at}krankenhaus-fn.de).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
PURPOSE: To prospectively compare rebleeding rates in patients treated with transjugular intrahepatic portosystemic shunt (TIPS) creation alone and those treated with TIPS creation combined with adjunctive embolotherapy of gastroesophageal collateral vessels.

MATERIALS AND METHODS: Informed consent and ethics committee approval were obtained. Between November 1991 and November 2002, the authors prospectively followed up 95 consecutive patients (61 men, 34 women; age range, 30–81 years) with variceal bleeding due to cirrhosis and portal hypertension. The patients were divided into two groups on the basis of splenoportographic findings after TIPS creation. The patients were treated with TIPS alone (group 1, 42 patients) or in combination with variceal embolotherapy (group 2, 53 patients). Embolotherapy with sclerosing agents in combination with coils was performed when varices continued to fill and the pressure gradient was more than 12 mm Hg. Rebleeding was defined as any hemorrhage necessitating a transfusion of 2 or more units of blood. Estimates for the cumulative patency, survival, and rebleeding rates were calculated by using life-table analysis; the log-rank test was used to compare the two treatment modalities. The prognostic relevance of treatment and selected variables with respect to rebleeding and survival were analyzed with multiple logistic regression.

RESULTS: Mean follow-up time (± standard deviation) was 48.7 months ± 37.8 (range, 1–127 months). The proportion of patients (Kaplan-Meier estimation) with TIPS who remained free of bleeding was 61% after 2 years and 53% after 4 years. Patients who underwent both the TIPS procedure and embolotherapy remained free of bleeding in 84% of cases after 2 years and in 81% of cases after 4 years. With respect to the rebleeding rate, the difference between the groups was statistically significant (log-rank test, P = .02). Results of multiple logistic regression analysis showed that variceal embolotherapy significantly reduced the risk of rebleeding (Wald test, P < .001).

CONCLUSION: The results suggest that TIPS and adjunctive embolotherapy of gastroesophageal collateral vessels significantly lower the rebleeding rate in comparison to TIPS alone.

© RSNA, 2005


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
To achieve portal decompression and to prevent rebleeding in patients with portal hypertension, transjugular intrahepatic portosystemic shunt (TIPS) creation is now an established technique, and results of both randomized and nonrandomized studies have strengthened the evidence that TIPS is more effective than endoscopic therapy in the prevention of variceal bleeding (1,2). The main problems with TIPS (2–6) are the low primary patency rate (8%–48% after 2 years), the relatively high rate of rebleeding (23%–40% after 2 years), and an increased risk of encephalopathy of about 29% (7%–55%). At the time of our writing this report, the preliminary experiences with polytetrafluoroethylene-covered stent grafts are very promising with regard to patency but have shown a high incidence of encephalopathy (7,8).

The effectiveness of variceal embolotherapy itself has been studied in the era before TIPS, and percutaneous transhepatic embolotherapy of gastroesophageal varices was described by Lunderquist and Vang (9). Since portal hypertension could not be improved by means of percutaneous transhepatic embolotherapy, recurrent bleeding occurred in 35%–65% of the patients within a few months after embolotherapy (10). Thus, the purpose of our study was to prospectively compare the rebleeding rates between patients treated with TIPS alone and those treated with TIPS combined with adjunctive embolotherapy of gastroesophageal collateral vessels.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 References
 
Patients
Between November 1991 and January 1998, 102 consecutive patients with cirrhosis and variceal bleeding underwent TIPS procedures. In four patients the intervention failed technically, and three other patients were lost to follow-up. Thus, the final study population consisted of 95 patients.

All patients had severe bleeding that had failed to respond to endoscopic sclerotherapy of esophageal varices or had gastric varices not amenable to sclerotherapy. Twelve patients were treated on an emergency basis within 48 hours of admission. All the other patients were prepared for treatment by performing a variety of clinical and laboratory tests, including color Doppler ultrasonography (US) and computed tomography (CT). CT was used to define the anatomy (especially the relationship between portal vein and hepatic vein), to determine vascular patency, and to rule out the presence of hepatocellular carcinoma. All patients were evaluated for possible surgical shunt creation, and a final decision was made to perform TIPS creation after an interdisciplinary conference between the gastroenterologist, the surgeon, and the radiologist. All patients were informed about the procedure in detail, and they gave written informed consent according to the guidelines of the local ethics committee that approved our study.

Of the 95 patients, 42 were treated with TIPS alone (group 1) and 53 were treated with TIPS and embolotherapy (group 2). Embolotherapy was performed on the basis of splenoportographic findings after TIPS creation. The clinical profile of the two groups (Table 1) was the same except for the high percentage of patients treated on an emergency basis in group 2. The majority of the patients in both groups had Child-Pugh class B cirrhosis (11).


View this table:
[in this window]
[in a new window]

  		TABLE 1. Clinical Characteristics of 95 Patients with Variceal Bleeding Due to Cirrhosis and Portal Hypertension

 
Angiographic and Procedural Protocol
Treatment was performed with patients under general anesthesia in the angiography suite. The technique of placing a transjugular intrahepatic stent to establish a portosystemic shunt has been described previously (3,6,12,13). The portosystemic gradient between the portal vein and right atrium was measured in centimeters of water, which we then converted to millimeters of mercury.

Our angioarchitectural and radiographic classification of thoracoabdominal collateral vessels was based on direct portography performed before treatment (Fig 1). Grading was as follows: grade 0, cephalad collateral vessels not visible; grade I, visible abdominal varices from one (1) cephalad collateral vessel; grade II, visible abdominal varices from two or more (1–3) cephalad collateral vessels; grade III, visible thoracoabdominal varices from one (1) cephalad collateral vessel; and grade IV, visible thoracoabdominal varices from two or more (1,2) cephalad collateral vessels. Our classification was a modification of the radiographic classification at celiac and superior mesenteric arteriography published in 1995 (14). If possible, our angiographic grading of varices was retrospectively compared with findings at endoscopy.



View larger version (61K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. Angioarchitectural and radiographic classification of portosystemic collateral vessels based on flow patterns demonstrated with injection of contrast medium into the proximal splenic vein (direct portography) before treatment (grades: I, visible abdominal varices from one cephalic collateral vessel; II, visible abdominal varices from two or more cephalic collateral vessels; III, visible thoracoabdominal varices from one cephalic collateral vessel; IV, visible thoracoabdominal varices from two or more cephalic collateral vessels). The diaphragmatic surface (dashed line) and varices (*) are indicated (gastric varices below dashed line, esophageal varices above). 1 = left gastric vein, 2 = posterior gastric veins, 3 = short gastric veins.

 
After dilation of the intrahepatic tract, the patency of the TIPS was facilitated by deployment of balloon-expandable stents (Palmaz P308 M, Johnson & Johnson, Norderstedt, Germany; Strecker tantalum 9 mm x 6 cm, Boston Scientific Europe, Ratingen, Germany) or self-expanding stents (Wallstent, Schneider, Bülach, Switzerland; Memotherm, Angiomed, Karlsruhe, Germany). The two types of self-expanding stents were both 10 mm in diameter but were different lengths. Except for the Strecker stent (6 cm in length, 9 mm in diameter), all stents were dilated to 8 mm. Whenever the pressure gradient was higher than 12 mm Hg or the decrease in pressure gradient was less than 40% (6), an additional dilation was performed to increase the stent diameter (if possible a 9-, 10-, or 12-mm balloon was used). Forty-two patients (44%) underwent the TIPS procedure alone (group 1).

Fifty-three patients (56%) underwent a TIPS procedure and variceal embolotherapy (group 2) with use of sclerosing agents (bezylalcohol, Varigloban 4%, Kreussler Pharma, aden-Biebrich, Germany; or natriumamidotrizoat, Ethibloc, Ethicon GmbH, Norderstedt, Germany) and coils. Embolotherapy was performed immediately after the TIPS procedure (at the same setting, via the transjugular vein). The rationale of performing embolotherapy was to use long-acting occluding agents for embolization of proximal and peripheral collateral veins by combining liquid and mechanical materials to prevent formation of new collateral veins and reperfusion. The indication to perform embolization of varices during the TIPS procedure was an elevated pressure gradient (>12 mm Hg) with persistent filling of varices, or, in patients undergoing TIPS procedures on an emergency basis, visible varices after shunt creation, regardless of the pressure gradient.

Procedural details, including length and diameter of balloons, technical and clinical complications, duration of fluoroscopy, and overall procedure time, were documented. The procedure was ended after an angiographic study (direct portography). Antibiotics were administered periprocedurally (4 g of amoxicillin) and after treatment (250 mg of amoxicillin four times a day for 3 days), and patients with maintained coagulation capacity (patients with Child-Pugh class A or B disease) were administered heparin (350 U per kilogram of body weight per 24 hours) to prolong the partial thromboplastin time to 40–60 seconds for the first 3 days. All interventional procedures were performed by one of two interventional radiologists (I.K.T. or W.J., with 6 and 7 years of experience, respectively, in vascular and nonvascular interventional radiology). These two authors had 5 years of experience with percutaneous transhepatic embolotherapy of gastroesophageal varices at the start of our study.

Follow-up
All patients were informed of and asked to enroll in the follow-up protocol, which included routine portal angiography and direct pressure measurements (performed by I.K.T., W.J., and C.D.) within 6 months after the TIPS procedure. The patients were free to unenroll themselves or to withdraw from the protocol.

TIPS patency and flow velocity were examined by means of noninvasive color duplex US at discharge and at 3, 6, and 12 months after the procedure and then every 6 months thereafter (all US studies were performed by T.F.). The US work-up included the following criteria: intrahepatic flow direction of portal branches, extrahepatic portal vein flow velocity, in-stent flow velocity, and flow velocity in the draining hepatic vein (2,5). TIPS insufficiency was diagnosed when the reduction of the maximum portal vein flow velocity was below 20 cm/sec or when reduction of the early increment of the portal vein flow velocity was more then 50%. Follow-up visits were scheduled prospectively.

We analyzed the incidence and severity of TIPS dysfunction (I.K.T., W.J., C.D., T.F.), the incidence and severity of rebleeding (I.K.T., W.J., T.F., E.H.), and the rate of complications and survival (I.K.T, W.J., T.F., E.H.). Rebleeding was defined as any gastrointestinal hemorrhage that necessitated a transfusion of 2 or more units of blood, and if possible the source of bleeding was endoscopically verified (endoscopy was performed by experienced endoscopists). All suspected TIPS abnormalities were evaluated by means of direct portography and pressure measurement after TIPS catheterization (I.K.T., W.J., C.D.). At angiography, TIPS stenosis was defined as at least 50% luminal narrowing of the stent and/or the adjacent hepatic vein in combination with an elevated portosystemic pressure gradient.

During follow-up (after hospitalization), 79 (83%) of 95 patients were examined at initial shunt venography after a mean of 6.9 months (range, 3–24 months). The primary end points of the study were the first rebleeding, liver transplantation, surgical shunt, or death.

Statistical Analysis
Estimates for the cumulative patency, survival, and rebleeding rates were calculated by using life-table analysis with the Kaplan-Meier method; the log-rank test was used to compare the two different treatment modalities. The prognostic relevance of treatment and selected variables with respect to rebleeding and survival were analyzed by using multiple logistic regression. Results of this analysis are presented as odds ratio estimates with corresponding 95% confidence intervals and P values from the Wald test (15). To determine statistically significant differences between two groups, the t test, the Mann-Whitney U test, or the {chi}2 test was used. The Wilcoxon signed rank test or the paired t test was used to determine statistically significant differences before and after treatment. Before using a t test, we used the Shapiro-Wilk test to check if the normality assumption was met. For ordinal data (ie, Child-Pugh score), a nonparametric test was used. All tests were two sided. A statistically significant difference was assessed for any of the analyses with results of P < .05. All computations were performed with SAS software (version 8.02; SAS Institute, Cary, NC). The statistical analyses were performed by C.W. and I.K.T.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 References
 
Procedural and Angiographic Results
The overall duration of fluoroscopy had a mean value of 38 minutes ± 8.2 (range, 14–94 minutes); the duration was 34 minutes ± 7.1 in group 1 and 41 minutes ± 9.4 in group 2 (t test, P < .001). This corresponds to a mean procedure time of approximately 2 hours. By using the technique described in references 6, 12, and 13, we did not observe any extrahepatic puncture of the portal vein. Technical complications did not lead to any clinical complications.

The mean portosystemic pressure gradient was 17.2 mm Hg ± 4.3 before TIPS creation and 9.7 mm Hg ± 3.7 after the procedure (paired t test, P < .001), with a mean balloon diameter of 9.2 mm (range, 8–12 mm). The angiographic results before and after treatment, which are listed in Table 2, are based on our angioarchitectural and radiographic classification of thoracoabdominal collateral vessels (Fig 1). The number of visible portosystemic collateral vessels after treatment was significantly reduced (Wilcoxon signed rank, P < .001), with no visualization of collateral vessels in 66 patients after treatment versus no visualization of collateral vessels in four patients before treatment (in these four patients, direct portograms did not show any retrograde filling of thoracoabdominal collateral vessels), and there was no significant difference between groups 1 and 2. The angiographic grading was retrospectively compared with endoscopic findings in 76 patients (80%), and we found a good correlation in terms of grading of varices before and after treatment in 72 of these patients (95%). In the four patients without variceal filling, there was proved cirrhosis and portal hypertension (median pressure gradient, 17.3 mm Hg; range, 16.9–17.6 mm Hg), and the endoscopic classification of the esophageal varices was grade 3–4 in combination with recurrent variceal bleeding.


View this table:
[in this window]
[in a new window]

  		TABLE 2. Angiographic Results before and after Treatment in 95 Patients

 
Follow-up Data
The mean follow-up duration was 48.7 months ± 37.8 (range, 1–127 months).

Rebleeding and patient survival.—Rebleeding rate for all patients and rebleeding rate for patients in groups 1 and 2 are shown in Figure 2 and Figure 3, respectively. Altogether, 23 patients experienced rebleeding. Of these patients, 21 (91%) underwent endoscopy, and variceal bleeding or stigmas of recent bleeding was verified in 18 of these 21 patients. Endoscopically, the other three patients had grade 2 varices without stigmas of bleeding at the time of endoscopy. Since shunt venography and reintervention were performed 5–14 days later, bleeding may have occurred from varices. The rebleeding rate at 30 days was 3% (three of 95 patients). Two of these patients underwent TIPS procedures and variceal embolotherapy on an emergency basis, and after inadequate reduction in the portosystemic pressure gradient (16 mm Hg in one patient, 14 mm Hg in the other) and still-patent collateral veins, we observed rebleeding. The variceal bleeding ceased after stent dilation and extensive re-embolization of collateral veins. Shunt venograms in the third patient (group 1) demonstrated stent occlusion (two Palmaz stents had been used), and the patient underwent recanalization by means of lysis (urokinase) and repeated stent placement.



View larger version (16K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2. Graph of Kaplan-Meier estimation of cumulative percentages of patients free of rebleeding among all 95 patients undergoing TIPS creation and variceal embolotherapy. Patients at risk at 6 months, n = 79; 12 months, n = 74; 24 months, n = 62; 48 months, n = 52; 90 months, n = 42; and 120 months, n = 39.

 


View larger version (18K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3. Graph of Kaplan-Meier estimation of cumulative percentages of patients free of rebleeding among patients undergoing TIPS creation alone (group 1) or in combination with variceal embolotherapy (group 2). The difference between treatment groups was statistically significant (log-rank test, P = .02). Patients at risk in group 1 at 6 months, n = 37; 12 months, n = 34; 24 months, n = 26; 48 months, n = 22; 90 months, n = 15; and 120 months, n = 13. Patients at risk in group 2 at 6 months, n = 42; 12 months, n = 40; 24 months, n = 36; 48 months, n = 30; 90 months, n = 27; and 120 months, n = 26.

 
The cumulative rates of patients free of rebleeding (Kaplan-Meier estimation) after 12, 24, 48, 90, and 120 months were 78%, 73%, 68%, 62%, and 49%, respectively. The corresponding life-table analysis of patients in groups 1 and 2 shows a statistically significant difference (log-rank test, P = .02); after a mean follow-up period of 48.7 months, 53% of the patients in group 1 were free of rebleeding versus 81% of the patients in group 2 (Fig 3). The 2- and 4-year rates of patients free of rebleeding were 61% and 53% in group 1 and 84% and 81% in group 2, respectively.

Figures 4 and 5 show the Kaplan-Meier estimation of cumulative survival of all patients and the estimation of cumulative survival of patients in group 1 and 2, respectively. The 30-day mortality rate was 7% (seven of 95 patients). The Kaplan-Meier estimation of cumulative survival for all patients after 12, 24, 48, 90, and 120 months was 77%, 65%, 57%, 48%, and 40%, respectively. The median survival time was 76 months (range, 0–100 months). The corresponding cumulative survival after 12, 24, 48, 90, and 120 months was 81%, 60%, 52%, 40%, and 30% for group 1 and 75%, 69%, 60%, 55%, and 55% for group 2, respectively. There was no statistically significant difference (log-rank test, P = .25) between groups 1 and 2 in terms of survival, but beyond 48 months after treatment we observed an increasing survival benefit for patients treated with TIPS in combination with variceal embolotherapy. Mortality due to rebleeding was 5% (five of 95 patients) after a mean follow-up of 48.7 months. Only one of these five patients was in the embolotherapy group (group 2). This patient in group 2 had a known TIPS occlusion that was not revised because of encephalopathy, and results at gastroscopy showed only grade 1 varices 3 months before death.



View larger version (14K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4. Graph of Kaplan-Meier estimation of cumulative survival of all 95 patients undergoing TIPS creation and variceal embolization. Patients at risk at 6 months, n = 79; 12 months, n = 74; 24 months, n = 62; 48 months, n = 52; 90 months, n = 42; and 120 months, n = 39.

 


View larger version (16K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5. Graph of Kaplan-Meier estimation of cumulative survival among patients undergoing TIPS creation alone (group 1) or in combination with variceal embolization (group 2). The difference between treatment groups was not statistically significant (log-rank test, P = .25), but beyond 48 months after treatment, we observed an increasing survival benefit for patients treated with TIPS creation in combination with variceal embolization. Patients at risk in group 1 at 6 months, n = 37; 12 months, n = 34; 24 months, n = 26; 48 months, n = 22; 90 months, n = 15; and 120 months, n = 13. Patients at risk in group 2 at 6 months, n = 42; 12 months, n = 40; 24 months, n = 36; 48 months, n = 30; 90 months, n = 27; and 120 months, n = 26.

 
Shunt patency and angiographic findings.—In all, we observed eight (8%) early (<30 days) occlusions. On the basis of all the clinical data, the primary patency rate after 6, 12, and 24 months was 49%, 28%, and 12%, respectively. The Kaplan-Meier estimation of the secondary patency rate after 12, 24, 48, 90, and 120 months was 94%, 92%, 85%, 85%, and 85%, respectively. The total number of percutaneous reinterventions necessary to maintain TIPS patency was 2.5 (range, 1–7) per patient. During the follow-up period, three patients underwent liver transplantation and four patients underwent surgical portosystemic shunt creation. Surgical portosystemic shunt creation was performed because of failure of percutaneous reintervention in three patients with completely occluded stents. One patient with an occluded stent refused further percutaneous intervention; thus, surgical therapy was performed.

During follow-up (after hospitalization), 79 patients (83%) were examined at initial shunt venography (Fig 6) after a mean of 6.9 months (range, 3–24 months). Table 3 shows the anatomic classification of portosystemic collateral vessels immediately before and after reintervention. By comparing the angiographic findings immediately after the first treatment (Table 2) with those obtained just before reintervention, we observed a significant (Wilcoxon signed rank test, P = .01) reperfusion of the collateral vessels (54% [43 of 79] grade I–IV collateral vessels before reintervention vs 31% [29 of 95] grade I–IV collateral vessels after TIPS procedure). We found a reperfusion of collateral vessels on venograms in patients with rebleeding. Of 23 patients with rebleeding, 21 underwent angiography; of these 21 patients, 18 had grade II or IV patent collateral vessels, and three patients had grade III patent collateral vessels.



View larger version (148K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6a. Images in a 53-year-old woman treated electively after recurrent severe variceal bleeding. (a) Pretreatment anteroposterior splenoportogram shows filling of gastroesophageal varices (arrow) from more than two cephalad collateral vessels (grade IV), no filling of the left portal radicle, and reduced flow in the right intrahepatic portal vein as a result of portal flow diversion via the large collateral vessels. Portosystemic pressure gradient was 16.5 mm Hg.(b) Anteroposterior portogram after a shunt has been established with implantation of two stents and the varices have been embolized with bezylalcohol in combination with natriumamidotrizoat (*) and coils (three coils [arrow], 6 and 8 mm in diameter, 15 and 30 cm in length). There is no filling of varices, and the flow in the intrahepatic portal vein became reversed. Portosystemic pressure gradient was 7.5 mm Hg. (c) Anteroposterior portogram 7 months after treatment shows high-grade stenoses of stent (double arrows) and hepatic vein (thick arrow) but no filling of varices. The intrahepatic portal vein shows hepatopetal flow. Portosystemic pressure gradient was 32 mm Hg. The patient was asymptomatic. (d) Right anterior oblique (20°) portogram after shunt dilation and repeated stent placement. Because of an insufficient angiographic result after dilation, a self-expanding stent was implanted. The intrahepatic portal flow is now hepatofugal. The reintervention reduced the portosystemic pressure gradient to 17.5 mm Hg; the reduction was about 55% and was therefore acceptable.

 


View larger version (140K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6b. Images in a 53-year-old woman treated electively after recurrent severe variceal bleeding. (a) Pretreatment anteroposterior splenoportogram shows filling of gastroesophageal varices (arrow) from more than two cephalad collateral vessels (grade IV), no filling of the left portal radicle, and reduced flow in the right intrahepatic portal vein as a result of portal flow diversion via the large collateral vessels. Portosystemic pressure gradient was 16.5 mm Hg.(b) Anteroposterior portogram after a shunt has been established with implantation of two stents and the varices have been embolized with bezylalcohol in combination with natriumamidotrizoat (*) and coils (three coils [arrow], 6 and 8 mm in diameter, 15 and 30 cm in length). There is no filling of varices, and the flow in the intrahepatic portal vein became reversed. Portosystemic pressure gradient was 7.5 mm Hg. (c) Anteroposterior portogram 7 months after treatment shows high-grade stenoses of stent (double arrows) and hepatic vein (thick arrow) but no filling of varices. The intrahepatic portal vein shows hepatopetal flow. Portosystemic pressure gradient was 32 mm Hg. The patient was asymptomatic. (d) Right anterior oblique (20°) portogram after shunt dilation and repeated stent placement. Because of an insufficient angiographic result after dilation, a self-expanding stent was implanted. The intrahepatic portal flow is now hepatofugal. The reintervention reduced the portosystemic pressure gradient to 17.5 mm Hg; the reduction was about 55% and was therefore acceptable.

 


View larger version (172K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6c. Images in a 53-year-old woman treated electively after recurrent severe variceal bleeding. (a) Pretreatment anteroposterior splenoportogram shows filling of gastroesophageal varices (arrow) from more than two cephalad collateral vessels (grade IV), no filling of the left portal radicle, and reduced flow in the right intrahepatic portal vein as a result of portal flow diversion via the large collateral vessels. Portosystemic pressure gradient was 16.5 mm Hg.(b) Anteroposterior portogram after a shunt has been established with implantation of two stents and the varices have been embolized with bezylalcohol in combination with natriumamidotrizoat (*) and coils (three coils [arrow], 6 and 8 mm in diameter, 15 and 30 cm in length). There is no filling of varices, and the flow in the intrahepatic portal vein became reversed. Portosystemic pressure gradient was 7.5 mm Hg. (c) Anteroposterior portogram 7 months after treatment shows high-grade stenoses of stent (double arrows) and hepatic vein (thick arrow) but no filling of varices. The intrahepatic portal vein shows hepatopetal flow. Portosystemic pressure gradient was 32 mm Hg. The patient was asymptomatic. (d) Right anterior oblique (20°) portogram after shunt dilation and repeated stent placement. Because of an insufficient angiographic result after dilation, a self-expanding stent was implanted. The intrahepatic portal flow is now hepatofugal. The reintervention reduced the portosystemic pressure gradient to 17.5 mm Hg; the reduction was about 55% and was therefore acceptable.

 


View larger version (162K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6d. Images in a 53-year-old woman treated electively after recurrent severe variceal bleeding. (a) Pretreatment anteroposterior splenoportogram shows filling of gastroesophageal varices (arrow) from more than two cephalad collateral vessels (grade IV), no filling of the left portal radicle, and reduced flow in the right intrahepatic portal vein as a result of portal flow diversion via the large collateral vessels. Portosystemic pressure gradient was 16.5 mm Hg.(b) Anteroposterior portogram after a shunt has been established with implantation of two stents and the varices have been embolized with bezylalcohol in combination with natriumamidotrizoat (*) and coils (three coils [arrow], 6 and 8 mm in diameter, 15 and 30 cm in length). There is no filling of varices, and the flow in the intrahepatic portal vein became reversed. Portosystemic pressure gradient was 7.5 mm Hg. (c) Anteroposterior portogram 7 months after treatment shows high-grade stenoses of stent (double arrows) and hepatic vein (thick arrow) but no filling of varices. The intrahepatic portal vein shows hepatopetal flow. Portosystemic pressure gradient was 32 mm Hg. The patient was asymptomatic. (d) Right anterior oblique (20°) portogram after shunt dilation and repeated stent placement. Because of an insufficient angiographic result after dilation, a self-expanding stent was implanted. The intrahepatic portal flow is now hepatofugal. The reintervention reduced the portosystemic pressure gradient to 17.5 mm Hg; the reduction was about 55% and was therefore acceptable.

 

View this table:
[in this window]
[in a new window]

  		TABLE 3. Angiographic Results Immediately before and after First Reintervention in 79 Patients

 
The analysis of the shunt venograms showed a significant difference between the patients in groups 1 and 2 and between symptomatic and asymptomatic patients in terms of reperfusion of collateral vessels, as follows: Before reintervention, 23 (64%) of 36 patients in group 1 versus 10 (23%) of 43 patients in group 2 had hepatofugal perfusion of collateral vessels (Wilcoxon signed rank test, P < .05). A total of 23 patients had rebleeding; this included 17 (40%) of the 42 patients in group 1, and only six (11%) of the 53 patients in group 2 (Wilcoxon signed rank test, P < .01), although all symptomatic patients showed shunt insufficiency. Our study protocol did not prospectively compare endoscopic and angiographic findings, but according to reported endoscopic findings (partially from referring hospitals), we found a good correlation between variceal rebleeding, shunt stenosis (portal hypertension), and reopening of collateral vessels.

Complications.—Procedure-related complications with clinical sequelae occurred in 17 patients (18%) (encephalopathy, 11 patients; sepsis, three patients; right-sided heart failure, one patient; acute liver failure, two patients), and three of these patients died (sepsis, one patient; heart failure, one patient; acute liver failure, one patient). Because of uncontrolled encephalopathy, two shunts had to be embolized 3 and 47 months after TIPS creation, and after embolization the symptoms of encephalopathy disappeared in both patients. All the other patients with new or substantial progression of encephalopathy were treated successfully with lactulose and dietary restrictions (limits on protein). Five patients (12%) in group 1 developed fever (temperature, >38°C) within 24 hours after TIPS creation; three of these patients had symptoms of sepsis. In only one patient, results of the blood culture were positive (Staphylococcus aureus), and he died 1 week after the TIPS procedure (multiorgan failure). Fourteen patients (26%) in group 2 presented with symptoms of a postembolization syndrome (fever, abdominal pain, and nausea). This was likely due to the Ethibloc injection (a protein, alcohol, and contrast agent mixture), which commonly results in a inflammatory reaction.

Logistic regression analysis.—The results of the logistic regression analysis of treatment and further selected explanatory variables with respect to the end points of rebleeding and death are shown in Table 4. Compared with TIPS alone, the use of TIPS in combination with variceal embolotherapy significantly reduced the risk of rebleeding (Wald test, P < .001) but provided a comparable rate of survival. In addition, patients with hepatitis and patients treated on an emergency basis (with P = .009 and P = .027, respectively; Wald test) had a significantly higher risk of rebleeding.


View this table:
[in this window]
[in a new window]

  		TABLE 4. Multiple Logistic Regression Analysis of Selected Variables for the End Points of Rebleeding and Death

 
With respect to cumulative mortality, we observed a statistically significant interactive effect of age (Wald test, P = .002) and hepatitis (Wald test, P = .027). Both variables significantly increased the risk of reaching the end point of death during the study period. No other variables showed a significant effect on the end points assessed.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
References
 
Outside of controlled studies, the two accepted indications for TIPS are (a) acute variceal bleeding that cannot be successfully controlled with pharmacologic agents or with mechanical compression or endoscopic techniques and (b) recurrent variceal bleeding in patients whose conditions are refractory or intolerant to conventional medical management, including sclerotherapy and pharmacologic therapy (16). Reports of the use of TIPS in these situations have shown rebleeding rates of 23%–40% after 2 years and suggest a substantial improvement in survival (3,5,17–21). There is no consensus in the literature regarding the role of variceal embolotherapy during TIPS procedures (3,22), and, to our knowledge, there is no published data evaluating the combination of these two methods versus TIPS creation alone, although authors of many articles describe the use of different embolization materials during TIPS creation (3,6,17,23).

Our study was performed to compare the use of TIPS with the use of TIPS in combination with adjunctive embolotherapy of gastroesophageal collateral vessels. A justification for an additional, adjunctive, and invasive therapy to TIPS creation must be either a reduction in the rebleeding rate or a reduction in the complication rate. Sanyal et al (5) found that systemic venous pressures of more than 15 mm Hg, stent dysfunction, and continued alcoholism were risk factors for recurrent hemorrhage. Risk factors for hepatic encephalopathy and liver failure are advanced age (>60 years), bad liver function (a high Child-Pugh score), and the shunt diameter (4). All data regarding TIPS in the literature and the experience gained from surgical use of shunts confirm these findings.

In our study, we found that TIPS stenosis with portal hypertension (abnormal pressure gradient) in combination with patent nonembolized varices (thoracoabdominal collateral vessels) shown on the follow-up angiograms was a significant (P < .001) risk factor for rebleeding. The cumulative long-term rebleeding rate in our patients after 4 years was 32%. The long-term rebleeding rate in patients undergoing TIPS in combination with adjunctive embolotherapy of gastroesophageal collateral vessels was only 19% in comparison with 47% in patients undergoing TIPS creation alone, and this difference was statistically significant. Our data analysis showed an independent overall effect of variceal embolotherapy on rebleeding (P < .001), and there was an inverse correlation between the risk of mortality due to rebleeding and variceal embolotherapy. Our angiographic grading of varices turned out to be very useful for the estimation of the risk of rebleeding, and we found a good correlation to endoscopic findings. The combination of angiographic and endoscopic grading provided very accurate and detailed information about the size, appearance, and bleeding risk of the varices. Considering the clinical variables, only hepatitis and treatment on an emergency basis (P = .009 and P = .027, respectively) reached significance as negative explanatory variables for rebleeding.

The theoretic background for using adjunctive embolotherapy to prevent rebleeding was based on the results of two studies, one with results published before TIPS creation had been introduced and the other with results published before TIPS creation had become an established technique (24,25). Results of the randomized study by Smith-Laing and colleagues (24) showed a reduction or delay of rebleeding after percutaneous transhepatic embolotherapy of gastroesophageal varices. Takase and colleagues (25) studied the relationship between embolization range and rebleeding rate by performing percutaneous transhepatic portography before and after sclerotherapy. They found that embolization of both esophageal varices and their feeder vessels is essential in lowering the recurrence rate after sclerotherapy.

The data from our angiographic follow-up study support the findings of Takase et al (25). The majority of our patients had shunt stenosis with recurrence of portal hypertension, but the patients who had originally undergone embolization at angiography seldom demonstrated a filling of collateral vessels, and these patients were free of rebleeding in most cases. In our experience, we have to embolize both the proximal and the peripheral level of the collateral vessels to achieve this and use long-acting occluding agents such as the combination of liquid and mechanical materials. If we had only placed some coils at the proximal level of the collateral vessels, it would hardly prevent the filling of distal feeder vessels and their varices (17,23).

With respect to risk and complication rate, the use of adjunctive embolotherapy during the TIPS procedure is two sided. It is more time consuming than performing TIPS creation alone (duration of fluoroscopy, 41 vs 34 minutes), which causes higher radiation exposure for personnel and patient, and embolotherapy is associated with specific complications. Among our patients, postembolization syndrome caused clinically relevant symptoms such as fever, pain, nausea, and, occasionally, a longer hospital stay. On the other hand, extensive adjunctive embolotherapy of varices allows us to use a smaller shunt diameter with a consecutively lower rate of encephalopathy. This study did not prove this hypothesis, but our rate of encephalopathy (12%) is lower than the rates in the majority of studies reported in the literature (2,4) and may indicate that the shunt diameter in our patients was lower than that generally reported. The results of our clinical research have caused us to change our therapy protocol in patients with severe bleeding after unsuccessful endoscopic sclerotherapy. If the splenoportogram demonstrates portosytemic collateral vessels of grade II, III, or IV, all patients will undergo variceal embolotherapy immediately before TIPS creation (both procedures performed at the same setting). We prefer to start with embolotherapy to take advantage of the better hepatofugal filling and perfusion of the collateral vessels.

A potential limitation of the present study was the nonrandomization of the patients, so the data can be affected by several potential biases. Nevertheless, taking the clinical profile of both groups into account, we did not identify a higher risk of rebleeding among patients in group 1. In fact, patients in group 2 were at a higher risk of rebleeding in comparison with patients in group 1 because of the high percentage of patients treated on an emergency basis in group 2 (5,26).

We believe that a patent TIPS with no portal hypertension is the most effective prevention for variceal rebleeding. Our data indicate that adjunctive embolotherapy of gastroesophageal collateral vessels prolongs the span of time between recurrent portal hypertension (shunt insufficiency) and rebleeding. In other words, the most important issue after combining TIPS and extensive embolotherapy of the blood supply to the varices is regular follow-up studies to detect and treat shunt insufficiency.


   FOOTNOTES
 

Abbreviations: TIPS = transjugular intrahepatic portosystemic shunt

2 Current address: Universitätsklinik für Radiodiagnostik, Innsbruck, Austria Back

Authors stated no financial relationship to disclose.

Author contributions: Guarantor of integrity of entire study, I.K.T.; study concepts and design, I.K.T., C.W.; literature research, I.K.T., T.F., C.W., E.H., C.D.; clinical studies, I.K.T., T.F., E.H.; data acquisition, I.K.T., T.F., E.H.; data analysis/interpretation, all authors; statistical analysis, I.K.T., C.W.; manuscript preparation, definition of intellectual content, and editing, I.K.T.; manuscript revision/review, I.K.T., C.W., W.J.; manuscript final version approval, I.K.T., C.D.


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 

  1. The North Italian Endoscopic Club for the Study and Treatment of Esophageal Varices. Prediction of the first variceal hemorrhage in patients with cirrhosis of the liver and esophageal varices: a prospective multicenter study. N Engl J Med 1988; 319:983–989.[Abstract]
  2. Burroughs AK, Vangeli M. Transjugular intrahepatic portosystemic shunt versus endoscopic therapy: randomized trials for secondary prophylaxis of variceal bleeding—an updated meta-analysis. Scand J Gastroenterol 2002; 37:249–252.[CrossRef][Medline]
  3. Rössle M, Haag K, Ochs A, et al. The transjugular intrahepatic portosystemic shunt. N Engl J Med 1994; 330:165–171.[Abstract/Free Full Text]
  4. Hauenstein KH, Haag K, Ochs A, et al. The reducing stent: treatment for transjugular intrahepatic portosystemic shunt-induced refractory hepatic encephalopathy and liver failure. Radiology 1995; 194:175–179.[Abstract/Free Full Text]
  5. Sanyal AJ, Freedman AM, Luketic VA, et al. The natural history of portal hypertension after transjugular intrahepatic portosystemic shunts. Gastroenterology 1997; 112:889–898.[CrossRef][Medline]
  6. Tesdal IK, Jaschke W, Bühler M, et al. Transjugular intrahepatic portosystemic shunting (TIPS) with balloon-expandable and self-expanding stents: technical and clinical aspects after 3 1/2 years experience. Cardiovasc Intervent Radiol 1997; 20:29–37.[CrossRef][Medline]
  7. Haskal ZJ. Improved patency of transjugular intrahepatic portosystemic shunts in humans: creation and revision with PTFE stent-grafts. Radiology 1999; 213:759–766.[Abstract/Free Full Text]
  8. Cejna M, Peck-Radosavljevic M, Thurnher S, et al. ePTFE-covered stent-grafts for revision of obstructed transjugular intrahepatic portosystemic shunt. Cardiovasc Intervent Radiol 2002; 25:365–372.[CrossRef][Medline]
  9. Lunderquist A, Vang J. Transhepatic catheterization and obliteration of the coronary vein in patients with portal hypertension and esophageal varices. N Engl J Med 1974; 291:646–649.
  10. Sos TA. Transhepatic portal venous embolization of varices: pros and cons. Radiology 1983; 148:569–570.[Abstract/Free Full Text]
  11. Pugh RN, Murray-Lyon IM, Dawson JL, Pietroni MC, Williams R. Transection of the oesophagus for bleeding oesophageal varices. Br J Surg 1973; 60:646–649.[Medline]
  12. Zemel G, Becker GJ, Bancroft JW, Benenati JF, Katzen BT. Technical advances in transjugular intrahepatic portosystemic shunts. RadioGraphics 1992; 12:615–662.[Abstract]
  13. Rösch J, Uchida BT, Barton RE, Keller FS. Coaxial catheter-needle system for transjugular portal vein entrance. J Vasc Interv Radiol 1993; 4:145–147.[Medline]
  14. Hashizume M, Kitano S, Tanoue K, et al. Sclerotherapy-resistant esophageal varices with enormously enlarged cepahaled collateral vessels predictable using portography. Hepatogastroenterology 1995; 42:551–556.[Medline]
  15. Gail M, Simon R. Testing for qualitative interactions between treatment effects and patient subsets. Biometrics 1985; 41:361–372.[CrossRef][Medline]
  16. Shiffman ML, Jeffers L, Hoofnagle JH, Tralka TS. The role of transjugular intrahepatic portosystemic shunt for treatment of portal hypertension and its complications: a conference sponsored by the National Digestive Diseases Advisory Board. Hepatology 1995; 22:1591–1597.[CrossRef][Medline]
  17. Coldwell DM, Ring EJ, Rees CR, et al. Multicenter investigation of the role of transjugular intrahepatic portosystemic shunt in management of portal hypertension. Radiology 1995; 196:335–340.[Abstract/Free Full Text]
  18. LaBerge JM, Somberg KA, Lake JR, et al. Two-year outcome following transjugular intrahepatic portosystemic shunt for variceal bleeding: results in 90 patients. Gastroenterology 1995; 108:1143–1151.[CrossRef][Medline]
  19. Sahagun G, Benner KG, Saxon R, et al. Outcome of 100 patients after transjugular intrahepatic portosystemic shunt for variceal hemorrhage. Am J Gastroenterol 1997; 92:1444–1452.[Medline]
  20. Sterling KM, Darcy MD. Stenosis of transjugular intrahepatic portosystemic shunts: presentation and management. AJR Am J Roentgenol 1997; 168:239–244.[Abstract/Free Full Text]
  21. Jalan R, John TG, Redhead DN, et al. A comparative study of the transjugular intrahepatic portosystemic stent-shunt (TIPSS) and oesophageal transection in uncontrolled variceal haemorrhage. Am J Gastroenterol 1995; 90:1932–1936.[Medline]
  22. Haskal ZJ. The complex TIPS. J Vasc Interv Radiol 1996; 7(suppl):S162–S167.
  23. Rosemurgy AS, Goode SE, Zwiebel BR, Black TJ, Brady PG. A prospective trial of transjugular intrahepatic portasystemic stent shunts versus small-diameter prosthetic H-graft portacaval shunts in the treatment of bleeding varices. Ann Surg 1996; 224:378–386.[CrossRef][Medline]
  24. Smith-Laing G, Scott J, Long RG, Dick R, Scherlock S. Role of percutaneous transhepatic obliteration of varices in the management of hemorrhage from gastroesophageal varices. Gastroenterology 1981; 80:1031–1036.[Medline]
  25. Takase Y, Shibuya S, Chikamori F, Orii K, Iwasaki Y. Recurrence factors studied by percutaneous transhepatic portography before and after endoscopic sclerotherapy for oesophageal varices. Hepatology 1990; 11:348–353.[CrossRef][Medline]
  26. Nevens F, Broecckaert L, Rutgeerts P, Van Steenbergen W, Fevery J. The long-term morbidity and mortality rate in a cohort of patients with liver cirrhosis and oesophageal varices. Hepatogastroenterology 1995; 42:979–984.[Medline]




This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2361040530v1
236/1/360    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Tesdal, I. K.
Right arrow Articles by Jaschke, W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Tesdal, I. K.
Right arrow Articles by Jaschke, W.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOLOGY RADIOGRAPHICS RSNA JOURNALS ONLINE