Cardiovasc Intervent Radiol [1992) 15:342-348
CardioVascular andInterventional
Radiology
9 Springer-Vertag New York Inc. 1992
Morphologic and Clinical Results of the Transjugular Intrahepatic Portosystemic Stent-Shunt (TIPSS) G. Noeldge, l G.M. Richter,-" M. Roessle, 3 K. Haag, 3 B.T. Katzen, 4 G.J. Becker, 4 and J.C. Palmaz 5 Departments of XDiagnostic Radiology and 3Internal Medicine II, Albert-Ludwigs-University Freiburg, FRG: -'Department of Diagnostic Radiology Ruprecht-Karls-University Heidelberg, FRG: 4Miami Vascular Institute. Miami, Florida, USA; and 5Section for Special Procedures, University of Texas Health and Science Center, San Antonio, Texas, USA
Abstract. The c o n c e p t o f transjugular intrahepatic portosystemic stent-assisted shunt (TIPSS) using the Palmaz iliac stent has been successfully accomplished in 18 o f 24 patients representing a technical success rate of 75%. F o u r t e e n were male, 4 female with a mean age o f 60 years (range 34-84 years). According to classification of Child's and Turcotte, 6 were in stage A, 6 in stage B, and 6 in stage C. Five patients were treated on an emergency basis because of massive active bleeding. In I0 patients the portosystemic tract was created between the middle hepatic vein and the right main stem of the portal vein in 8, and the left main stem in 2 patients. In 8 patients, the shunt was established between the right hepatic vein and the right main branch of the portal vein. The p o r t o s y s t e m i c gradient in 18 patients was 29.9 + 6 m m Hg and dropped to an average of 16.9 + 4 mm Hg after shunt establishment. Within the early p o s t p r o c e d u r a l period of 30 days, 1 patient died of direct complications of the procedure. Because o f c a t h e t e r dislocation, embolization of the p e r c u t a n e o u s transhepatic approach to the portal vein after successful shunt " c r e a t i o n " could not be done and was followed by intraabdominal exsanguination. One patient died of an ARDS after TIPSS. A third d e v e l o p e d pulmonary infection. In 13 patients, because o f h e m a t o m a s at the puncture s~te o~ ~'he transhepat~c approach, only the transjugular approach was elected for establishing TIPSS. The mean p o r t o s y s t e m i c gradient in 18 patients prior to TIPSS was 29 -+ 6 m m Hg (range 19-41 mm Hg), dropped to an average o f 16.9 _+ 4 mm Hg (range 7-21 mm Hg), and s h o w e d no significant change 6 Gerd Noeldge, M.D., PR.D., Radiologische Universitfitsklinik, Abt. Rontgendiagnostik, Hugstetterstrasse 55, D-7800 Frmburg im Breisgau, FRG
Address reprint requests to:
months after T I P S S with a pressure of 16 -+ 1.8 mm Hg. The 1-year survival rate was 75% (8/12); the 2-year rate was 50% (3/6).
Key words: L i v e r cirrhosis, h y p e r t e n s i o n - - P o r t a l bleedings, v a r i c e a l - - T r a n s j u g u l a r p o r t o s y s t e m i c s h u n t - - T I P S S - - S t e n t , vascular
In 1985 Pamaz described experimentally a nonsurgical method for creating a portocaval anastomosis with the aid o f balloon e x p a n d a b l e vascular stents in a normotensive canine model [1]. This model proved to be valid in a n o t h e r canine model with portal h y p e r t e n s i o n [2]. F o r clinical application in patients with severe portal hypertension, we followed this principle of stent scaffolding an artificial tract b e t w e e n the main stem o f one of the hepatic veins and either o f the main portal vein branches. We ,were first successful in 1988 [3]. Since then, we have been applying this transjugular intrahepatic portosystemic stent-shunt (TIPSS) c o n c e p t in a pilot study on 24 patients. Eligibility for the study required failed e n d o s c o p i c t r e a t m e n t and ineligibility for shunt surgery.
Materials and Methods All patients included in this study had to meet the following criteria: high prospective surgical risk (age >70 years), recurrent variceal bleeding without further benefit from sclerotherapy or gastric varices ineligible for ~cler,~lhe ra p y or failed surgical shunt, mtcroscopicall} p r o : e n il~ c r .2irr[io,,i,,. portosystemm gradient ->20 mm Hg, lack of inflammatory disease, and the patient's informed consent after knowing the character of the study.
G. Noeldge et al,: Morphologic and Clinical Results of TIPSS
Preprocedural Diagnostic Steps The general and disease-specific medical history was recorded along with weight, height, abdominal girth, and skin and mucosal changes specific for liver disease, Hepatic encephalopathy stage I and II was tested using the number connection test INCT) according to Corm [4] in the age-corrected version by Schomerus eta[. [5]. Hepatic encephalopathy stage III and 1V were diagnosed clinically by stupor (stage IlI) and hepatic coma (IVJ.
Noninvasive Diagnostic Procedures Before TIPSS, the following laboratory tests were performed: LDH, GPT, GOT, GLDH. CHE, AP, bitirubim serum protein level including protein electrophoresls, renal excretion function, coagulation status, RBC, WBC, thrombocytes, electrolytes, ammonium, specific liver tumor markers (c~-fetoprotein),and hepatitis serology. Furthermore, we performed abdominal sonograpby (B- and M-mode) to qualify and quantify ascitic fluid and portal flow pattern, chest X-ray film and ECG
343 Several parameters were carefully monitored, such as gastrointestinal bleeding, change ofascitic fluid volume, pulmonary complications, renal function, imbalance of electrolytes, liver function, and hepatic encephalopathy. Further hospital stay depended on the clinical status of the patient. Antlcoagulation therapy was not given systematically and was mostly performed in Child's A and B patients with good liver function and no coagulation deficits, In such cases, heparin was administered intravenously in a dosage that increased the partial thrombin time by a factor of 2 for a maximum of 7 days.
Follow-up Studies These included similar laboratory tests as performed before TIPSS for liver and renal function, coagulation, and blood status, The tests were performed 1 and 3 months after the procedure and from then on at 3-month intervals. Imaging procedures included B-mode and duplex ultrasound and endoscopy parallel to the laboratory tests. Transvenous portography, either transfemoral or transjugular, was done in 6-month intervals. In the first week after TIPSS, a chest X-ray film was taken daily.
Staging of Portal Hypertension This was based on the Child's and Turcotte classification [6] in the modified version of Corm [7] applying the scaling of five clinical and biochemical parameters cnutritional status, bdirubm, ascites, serum albumin level, and hepatic encephalopathy) up to a maximum score of 15 points: a score of 5-8 points corresponded to stage A, 9-12 points to stage B, and 13-15 points to stage C.
Invasive Procedures Before TIPSS These included endoscopy eventually accompamed by sclerotherapy and visceral angiography to demonstrate arterial and portal flow patterns.
Patient Preparation This included correction of coagulation deficit, hematocrit loss and low albumin level to improve the performance status with the aim of creating an elective procedure character. This was accompanied by intense medical control and eventual drainage of ascitic fluid and bowel sterihzation and irrigation to prevent inadvertent protein resorption upon and after the procedure,
TIPSS This technique is described in detad m previous reports [8-10]. In the last 7 patients, the original method has been essentially modified. The most significant change includes the avoidance of a transhepatic approach to the portal vein while the portal bifurcation is visualized by angiography (electronically stored from the previous indirect spleno-portography) and ultrasound instead [11]. Another change was the use of newly designed and better suitable puncture devices (TIPSS-Set, Angiomed Inc., Karlsruhe, Germany or Transjugular Cholangiography Set, Cook, Inc., Bloomington, IN, USA).
Results T I P S S has b e e n s u c c e s s f u l l y e s t a b l i s h e d in 18 o f the 24 p a t i e n t s , r e p r e s e n t i n g a t e c h n i c a l s u c c e s s rate o f 75%. T h e m e a n age o f the p a t i e n t s w a s 60 y e a r s with a range of 34-84 years; 4 patients were female and 14 male. In 4 p a t i e n t s , l i v e r c i r r h o s i s had d e v e l o p e d after h e p a t i t i s B i n f e c t i o n ; the o t h e r s h a d h a d p o s t a l c o h o l i c c i r r h o s i s . T h i r t e e n p a t i e n t s had o t h e r s e v e r e u n r e l a t e d d i s e a s e s , s u c h as d i a b e t e s ( N = 5), malign a n t t u m o r s ( h e p a t o m a , s q u a m o u s cell c a r c i n o m a o f the h y p o p h a r y n x ) (2), s i g n i f i c a n t c o r o n a r y a r t e r y d i s e a s e (2), r e n a l i n s u f f i c i e n c y (2), a b d o m i n a l a n e u r y s m (1), a n d p u l m o n a r y e m p h y s e m a (I). Six patients w e r e classified as stage C h i l d ' s A, 6 as stage B, a n d 6 as stage C. T h e m e a n d u r a t i o n o f s y m p t o m a t i c liver d i s e a s e w a s 10 y e a r s , with a r a n g e o f 3 - 1 7 years. F i v e p a t i e n t s w e r e a c t i v e l y b l e e d i n g u p o n i n t e r v e n t i o n , a n d the p r o c e d u r e was p e r f o r m e d o n an e m e r g e n c y basis. Subclinical hepatic encephalopa t h y w a s p r e s e n t in 10 p a t i e n t s (stage I o f the N T C test), a n d c l i n i c a l h e p a t i c e n c e p h a t o p a t h y in 4 (stup o r in 3 a n d c o m a i n 1) p a t i e n t s b e f o r e T I P S S . I n the g r o u p o f s u c c e s s f u l c a s e s , 5.7 s e v e r e v a r i c e a l bleeding episodes had been recorded and an average o f 6.2 ( r a n g e : 0 - 3 3 ) s c l e r o s i n g p r o c e d u r e s w e r e p e r f o r m e d . S c l e r o t h e r a p y w a s n o t d o n e in 2 p a t i e n t s because e n d o s c o p y revealed gastric varices stage IV and excluded major esophageal varices.
Immediate Results Postprocedural Patient Care Following successful intervention, the patients remained in the intensive care unit until clinically stable conditmns were ensured.
Morphologic Aspects.
The technical success was 75%. I n 5 p a t i e n t s the p u n c t u r e p r o c e d u r e w a s n o t s u c c e s s f u l w i t h i n a r e a s o n a b l e time s p a n ( > 2 h). A s
344
G. Noeldge et al.. Morpholog)c and Clinical Results of TIPSS
,,~ r'
~
~
~
....
A Fig. 1. A 70-year-old patient with liver ctrrhosm. Child's class C, undergoing stentmg of the intrahepatlc transparenchymal tract. A Percutaneous transhepatic portography demonstrating the large caliber of the coronary gastric vein and the vance,,,. B Plato film AP vmw: after placement of a tandem stent.
a result of severe periportal liver fibrosis, we failed to bring a catheter into the portal system in another patient despite successful puncture and positioning of a coronary guidewire in the mesenteric veto. The average procedure time in the successful cases was 5 h (range of 2.5-9 h). In 10 patients the artificial shunt tract was created by bridging the middle hepatic vein with the right main stem of the portal vein in 8, and with the left main stem in 2 patients. In 8 patients the shunt connected the right hepatic vein with the right main stem of the portal vein. In 4 patients one stent, in 12 patients two stents, and in 2 patients three stents were used for scaffolding the intraparenchymal tract (Fig. 1). In 8 patients the diameter of the shunt was 8 ram, in 6 patients it was 9 mm, and in 4 patients it was 10 mm.
mm Hg 40,
30'
20
10'
before TIPSS
Hemodynamic Aspects, Before successful stent shunting, the mean portosystemic gradient of the 18 patients was 29.9 -+ 6.0 mm Hg, range 19-41 mm Hg. After TIPSS, the gradient dropped to an average of 16.9 + 4 mm Hg, range 7-21 mm Hg (Fig. 2). This equals an average reduction of 43% which is highly significant (p < 0.0001).
Complications. During the early postprocedural period of 30 days, t patient died of a direct complication of the procedure. Because of catheter dislocation, we were unable to embolize the lateral percutaneous transhepatic access to the portal system after successful shunt creation. Six hours later, the patient developed substantial intraabdominal bleeding. Emergency packing failed to stop the
after TIPSS
Fig. 2. Immedmte change of the average portosystemic gradient after transjugular intrahepatic porto systemic stent shunt (TIPSS).
bleeding. A second liver packing was successful 3 days later. However, the patient already developed an ARDS, to which he succumbed 12 days later. Another patient developed pulmonary infection 10 days after the procedure and died. Hematomas were present in 5 patients, in 4 at the puncture site for the transhepatic approach and in 1 at the transjugular puncture site. As a consequence, from patient 13 on, we avoided the additional percutaneous approach to the portal system. Another patient, in whom we had
G. Noeldge etal.: Morpholog~c and Climcal Results of TIPSS
345
been unable to establish the stent shunt, also died of bleeding through the transhepatic tract to the portal system.
Long-term Results Clinical Success. The early (procedural) mortality was 1t%, as described above. Within the observation period of 5-28 months, the l-year survival rate was 75% (8/12), and the 2-year survival rate 50% (3/6). Presently, 11 patients are living with an average survival time of 13 months. A total of 5 patients died with an average survival time of 7 months. Death was unrelated to the procedure in 4 cases 12 pneumonias, 1 hepatoma, I left heart insufficiency) and in 1 patient, late shunt occlusion occured 18 months after TIPSS, resulting in a lethal variceal hemorrhage. Noninvasive Imaging. Sonographically, the shunt was easily identified in all patients, because of mild acoustic reverberations from the metallic stent struts. Doppler signals not only penetrated the stent but were also strong enough to allow flow measurements. In these duplex-based flow measurements, diameter error was avoided as the luminal dimension was well known for each individual patient. Therefore, the flow volume was easily determined and varied from 340 to 1900 cc/min with an average of 1210 cc/min. In 15 of 16 patients the individual flow values did not show a significant variation (+-20% over the observation period), In one patient duplex ultrasound showed a 40%, decrease of shunt flow. Shunt angiography from a transjugular approach demonstrated luminal narrowing of the liver vein proximal to the upper end of the shunt tract. Another stent was deployed overlapping the stent shunt. However, 6 months later the patient rebled. Duplex ultrasound confirmed late shunt occlusion (18 months after TIPSS). Invasive Studies. During the observation period, the shunt was angiographically followed by either transfemoral or transjugular approach in 13 patients 6 months after TIPSS, in 5 patients 12 months after TIPSS, and in 3 patients 24 months after TIPSS (Figs. 3, 4). The results are summarized in Table 1, which also includes the measurements of shunt flow and pressure gradients. Six months after TtPSS, a thin neointimal layer was demonstrated angiographically, ranging between 0. I and 0.5 ram. With the exception of our fifth patient (see above), there was no difference at the 12 month and 24 month followup studies in all other patients. The average portosystemic gradient 6 months
J Fig, 3. Four-month follow-up (same patient as m Fig. 11.
after TIPSS was 16.0 + 1.8 mm Hg: this was not statistically different from the immediate results. However, in 1 patient (P 2, Table 1) the gradient measured 23 mm Hg (endoscopy had shown mild erosive gastritis). As a value of >20 mm Hg had been considered as the threshold for sufficient portal decompression, her shunt diameter was increased to 10 mm by transjugular balloon dilatation, resulting in a decrease of her portosystemic gradient to 14 mm Hg. This procedure could be performed on an outpatient basis. Twelve months after TIPSS, the average portosystemic gradient of all patients at risk was 18.2 + 4.5 mm Hg (Table I). This increase resulted from the venous outflow stenosis of 1 patient already mentioned above. After correction by insertion of another stent, his gradient dropped from 26 mm Hg to 12 mm Hg. As a restllt the average gradient of the cohort group dropped also to 15.4 mm Hg (Table 1).
Discussion The concept of TIPSS corresponds to the hemodynamic situation of a small caliber surgical H-interposition shunt leading to partial portal decompression. With TIPSS, the interposition is established by a stented parenchymal tract bridging either of the main stems of the portal vein with the right or middle hepatic vein. From a hemodynamic point of view this concept offers a major advantage compared with
346
G Noeldge et al." Morpholog~c and Clinical Result~ of TIPSS
Table 1, Transjugular intrahepat~c portosystemic stent-shunt: angiograph~c and hemodynam~c results o f patients sur,,qving more than 6 m o n t h s
Patient
6 mo
12 mo
24 m o
Shunt
PS-grad.
Shunt
PS-grad.
Shunt
lumen (mm~
(ram Hgl
lumen Imm)
( m m Hg}
lumen ~mm)
P2 P3 P4 P5 P6 P8 Pl0 Pll P12 PI3 PI4 PI5 P16
7.7/10" 7.8 Not performed 9.5 8.9 8.9 8.5 8.8 7.7 6.0 8.6 8 7 9.9
23/14" 16
9,7 7,8 7.7 5.6/10 ~' 8.6 D~ed
18 15 16 26/12" 16
Dmd 7.8 7.7 D~ed 8.9
M e a n -+ SD
8 6 -+ I I
16 -* 1.8
14 18 18 17 17 15 17 16 15 t2
To To To To To To To 7_9 = 1.5 8.7 = t.1
be be be be be be be
PS-grad. (ram Hg)
t5 15 14
performed performed performed performed pert;ormed performed performed
18.2 -*- 4.5 t5.4 m 2,2
PS-grad = p o r t o s y s t e m i c gradmn{ " Redilatation o f the shunt ~gradmnt > 2 0 m m Hg) by a 10-mm balloon ~' B e c a u s e of s t e n o m s in the Liver vein proximal to the stent shunt, implantation of another stent 110 ram)
bu
9
9
"
',- g /',3!i
Fig. 4. Two-year follow-up (same patient as Fig. 3). A Plain film A P view shows kinking o f the t a n d e m stent, B indirect splenoport o g r a m s h o w i n g a patent shunt despite kinking of the t a n d e m stent: no varices.
the surgical technique. Scaffolding the shunt by balloon expandable Palmaz stents allows adaptation of the shunt lumen to the individual hemodynamic situation, ranging from 7 to 14 mm and depending on the size of the balloon catheter used. The shunt diameter may be increased whenever necessary by
i]
9
! ,';'{
:-
9
"
simple transjugular approach. In our series this was performed in 1 patient, In view of the applied inclusion criteria it is evident that our series involves a relatively high number of patients with end stage liver disease. Two-thirds of the successfully treated patients had a stage Child's B or C. Compared to historical groups of shunt surgery, this represents a highly negative patient selection [I1-13]. The 6 patients in stage Child's A were enrolled either because of a prospectively high surgical risk (other unrelated diseases) or because of previously failed shunt surgery. In 5
G. Noeldge et al.: Morphologic and Chnical Results of TIPSS
cases, TIPSS was performed as an emergency intervention. The technical aspects of TIPSS included a significant learning curve during which we modified the applied materials substantially. In the first 3 cases, the procedure took roughly 8 h, but lasted 2-3 h in the later cases. The combination of increased experience with specificially designed needles helped abandon the additional access to the portal system via a midaxillary transhepatic approach. This approach had caused significant complications (two deaths). The combination of a high quality indirect splenoportogram with ultrasound could be used effectively for guidance of the transjugular puncture [101. The key element for TIPSS is the balloonexpandable stent. It allows for a smooth and regular lumen of the scaffolded tract and a predictable shunt geometry. By using the Palmaz stent type, an individual tailoring of the shunt diameter to the hemodynamic needs of each patient is feasible. Technically~ it proved to be helpful to start with an 8 mm shunt diameter and, eventually, increase it by l-mm increments, depending on the measured gradients. Presently, the Palmaz stent is the only such stent available on the market for this technique. It offers the highest resistance to radial compression compared with other stent types [3]. Angiography clearly documents the change of the portal hemodynamic situation by the TIPSS procedure. All of our patients showed a substantial decrease of variceal flow associated with a reduction of peripheral portal venous filling. In 2 patients, no peripheral portal flow was found after the procedure, while their portosystemic gradient was 14 mm Hg and 12 mm Hg, respectively. In the follow-up studies, not much changed compared with the situation immediately after TIPSS. None of the patients developed a significant decrease of liver function after TIPSS. Clinically significant hepatic encephalopathy was found for 3 days in 1 patient 9 months after TIPSS. In 2 patients, gastrointestinal rebleeding episodes were recorded. One occurred in our first patient in whom we had not drained a large amount of ascites before the procedure. Rapid abdominal fluid resorption after the well-functioning shunt may have caused uptake of fibrinolytic substances from the ascites and induced widespread mucosal bleeding. This was easily controlled by fresh blood and fresh frozen plasma. In the second patient, bleeding was associated with late shunt occlusion. These overall results compare well with the surgical shunt literature, where hepatic encephalopathy is reported as high as 50% in nonselective shunts and up to 20% in selective shunts [11, 12, 14-16]. Also, rebleeding is reported in the
347
surgical literature almost exclusively after shunt occlusion and independent of the type of surgical shunt. The mortality rate of 11% in our series seems to be acceptable, taking into account the negative selection criteria applied for the study. For comparison, emergency shunt operation for patients in a stage Child's C is reported as high as 40%-100% [13]. However, the total number of our patients is too small for definite conclusions on the basis of stage-related risk stratification. Five patients died during the long-term follow-up period: 3 died of reasons related to their liver disease (1 hepatoma, 2 severe infections), 1 died of late shunt failure, and 1 of reasons unknown, Comparing the 1-year survival rate of 6/8 and the 2-year survival rate of 3/6 in our series with the literature, the clinical validity of the concept is underscored. In 6 of 7 patients who died, a histomorphologic analysis was performed; some of the results are already published in detail [3, 8]. These results may be summarized by one simple fact: a neointima develops on the stent shunt surface indistinguishable from arterial application of the stent. In summary, hemodynamically, TIPSS represents a small H-shunt with partial portal decompression. Lately, Rypins et al. [17] and Johansen [18] reported favorable results with the surgical establishment of this shunt type. Apart from its pet-cutaneous character, the TIPSS concept offers one major advantage over the surgical shunt which is adaptability of the shunt size to the individual hemodynamic needs. The interposition is exclusively intrahepatic. The fact that the extrahepatic vascular trunks remain untouched represents a significant advantage if liver transplantation is taken into consideration, it seems feasible to prevent the risk of lethal variceal bleeding in patients already on a waiting list for organ transplantation by performing a TIPSS procedure. After our learning experience, it appears that the technique and the materials used for the procedure are sufficiently defined. A wider clinical application of the method appears warranted.
References 1. Palmaz JC, Sibbitt RR, Reuter SR, Garcia F. Tlo FO 11985) Expandable intrahepatic portacaval shunt stents: Early experience in the dog. A JR 145:821-825 2. Palmaz JC, Garcta F, Sibbit SR, Tio FO, Kopp Dr', Schwesinger W. Lancaster JL, Chang P (1986) Expandable intrahepatic portacaval shunt stents in dogs with chronic portal hy. pertension. A JR 147:125 I - 1254 3. Richter GM, Palmaz JC, Noldge G. R6ssle M, Siegerstetter V, Franke M, Wenz W t1989) Der transjugul/ire intrahepatische portosystemische Stent-Shunt (TIPSS). Radiologe 29:406-411
348 4. Conn HO (1977) Trailmaking and Number-Connection Test in the Assessment of mental state in portal systemic encephalopathy. Am J Dig Dis 22:541-550 5. Schomerus H. Hamster W, Reinhard U, Mayer K, D011e W 11981) Latent portosystemic encephalopathy. Dig Dis Sci 26:622-630 6. Child CG, Turcott JG (1964) Surgery and portal hypertension. ln: Child CG (ed) The liver and portal hypertension. WB Saunders, Philadelphia 7_ Conn HO (1981) A peek at the Child-Turcotte classification. Hepatology 1:673-676 8. Richter GM, Noeldge G, Patmaz JC, Roessle M, Siegerstetter V, Franke M, Gerok W, Wenz W, Farthmann E ( 19901Transjugular intrahepatic portacaval stent shunt: Preliminary clinical results. Radiology 174:1027-1030 9. Richter GM, Noetdge G, Palmaz JC, Roessle M (1990) The transjugular intrahepatic portosystemic stent-shunt (TIPSS): Results of a pilot study Cardmvasc Intervent Radiol 13:200-207 10. Noldge G, Richter GM, Rossle M, Perarnau JM, Palmaz JC (1991) Technical improvement of the transjugutar intrahepatic portosystemic stent-ass~sted shunt (TIPSS): A quicker and safer ultrasound-assisted method (abstract). Presented at the 16th annual meeting of the SCVIR, San Francisco 1991--Program book p 172
G. Noeldge et al.: Morphologic and Clinical Results of TIPSS 1 l. Millikan W J, Warren WD, Henderson JM, Smith RB, Salam AA, Galambos JT, Kutner MH, Keen JH (1985) The Emory prospective randomized trial: Selective versus non-selective shunt to control variceal bleeding. Ann Surg 201:712-722 12. Warren WD, Millikan WJ Jr, Henderson JM, Wright L, Kutner M, Smith RB 3rd, Futenwider JT, Salam AA, Gatambos JJ (1982) Ten years portal hypertensive surgery at Emory. Results and new perspectives. Ann Surg 195:530-542 13, Galambos JT (1985) Portal hypertension. Semin Liver Dis 5:277-290 14 Foster JH, Elhson LH, Donovan Th, Anderson A (1971) Quantity and quality of survival after portosystemic shunts. Am J Surg 12:490-501 15. Pagliaro L, Burroughs AK, Sorensen TIA, Lebrec D, Morabito A, D'Amico G, Tint F (1989) Therapeutic controversies and randomised controlled trials (RCTs): Prevention of bleedmg and rebleeding in cirrhosis. Gastroenterol Int, 2"71-84 16. Rossle M, Haag K, Noeldge G, Richter G, Wenz W, Farthmann E, Gerok W (1990) Hemodynamic consequences of portal decompression: Which is the optimal shunt? Z Gastroenterol 28:630-634 17. Rypins EB, Mason GR, Conroy Rbl, Sarfeh IJ (1984) Predictability and maintenance of portal flow patterns after smalldiameter portocaval H-grafts in man. Ann Surg 200:706-710 18. Johansen K ~1989J Partial portal decompression for variceal hemorrhage. Am J Surg 157:479-482
CardioVascular andInterventional
Radiology
9 Springer-Vertag New York Inc. 1992
Morphologic and Clinical Results of the Transjugular Intrahepatic Portosystemic Stent-Shunt (TIPSS) G. Noeldge, l G.M. Richter,-" M. Roessle, 3 K. Haag, 3 B.T. Katzen, 4 G.J. Becker, 4 and J.C. Palmaz 5 Departments of XDiagnostic Radiology and 3Internal Medicine II, Albert-Ludwigs-University Freiburg, FRG: -'Department of Diagnostic Radiology Ruprecht-Karls-University Heidelberg, FRG: 4Miami Vascular Institute. Miami, Florida, USA; and 5Section for Special Procedures, University of Texas Health and Science Center, San Antonio, Texas, USA
Abstract. The c o n c e p t o f transjugular intrahepatic portosystemic stent-assisted shunt (TIPSS) using the Palmaz iliac stent has been successfully accomplished in 18 o f 24 patients representing a technical success rate of 75%. F o u r t e e n were male, 4 female with a mean age o f 60 years (range 34-84 years). According to classification of Child's and Turcotte, 6 were in stage A, 6 in stage B, and 6 in stage C. Five patients were treated on an emergency basis because of massive active bleeding. In I0 patients the portosystemic tract was created between the middle hepatic vein and the right main stem of the portal vein in 8, and the left main stem in 2 patients. In 8 patients, the shunt was established between the right hepatic vein and the right main branch of the portal vein. The p o r t o s y s t e m i c gradient in 18 patients was 29.9 + 6 m m Hg and dropped to an average of 16.9 + 4 mm Hg after shunt establishment. Within the early p o s t p r o c e d u r a l period of 30 days, 1 patient died of direct complications of the procedure. Because o f c a t h e t e r dislocation, embolization of the p e r c u t a n e o u s transhepatic approach to the portal vein after successful shunt " c r e a t i o n " could not be done and was followed by intraabdominal exsanguination. One patient died of an ARDS after TIPSS. A third d e v e l o p e d pulmonary infection. In 13 patients, because o f h e m a t o m a s at the puncture s~te o~ ~'he transhepat~c approach, only the transjugular approach was elected for establishing TIPSS. The mean p o r t o s y s t e m i c gradient in 18 patients prior to TIPSS was 29 -+ 6 m m Hg (range 19-41 mm Hg), dropped to an average o f 16.9 _+ 4 mm Hg (range 7-21 mm Hg), and s h o w e d no significant change 6 Gerd Noeldge, M.D., PR.D., Radiologische Universitfitsklinik, Abt. Rontgendiagnostik, Hugstetterstrasse 55, D-7800 Frmburg im Breisgau, FRG
Address reprint requests to:
months after T I P S S with a pressure of 16 -+ 1.8 mm Hg. The 1-year survival rate was 75% (8/12); the 2-year rate was 50% (3/6).
Key words: L i v e r cirrhosis, h y p e r t e n s i o n - - P o r t a l bleedings, v a r i c e a l - - T r a n s j u g u l a r p o r t o s y s t e m i c s h u n t - - T I P S S - - S t e n t , vascular
In 1985 Pamaz described experimentally a nonsurgical method for creating a portocaval anastomosis with the aid o f balloon e x p a n d a b l e vascular stents in a normotensive canine model [1]. This model proved to be valid in a n o t h e r canine model with portal h y p e r t e n s i o n [2]. F o r clinical application in patients with severe portal hypertension, we followed this principle of stent scaffolding an artificial tract b e t w e e n the main stem o f one of the hepatic veins and either o f the main portal vein branches. We ,were first successful in 1988 [3]. Since then, we have been applying this transjugular intrahepatic portosystemic stent-shunt (TIPSS) c o n c e p t in a pilot study on 24 patients. Eligibility for the study required failed e n d o s c o p i c t r e a t m e n t and ineligibility for shunt surgery.
Materials and Methods All patients included in this study had to meet the following criteria: high prospective surgical risk (age >70 years), recurrent variceal bleeding without further benefit from sclerotherapy or gastric varices ineligible for ~cler,~lhe ra p y or failed surgical shunt, mtcroscopicall} p r o : e n il~ c r .2irr[io,,i,,. portosystemm gradient ->20 mm Hg, lack of inflammatory disease, and the patient's informed consent after knowing the character of the study.
G. Noeldge et al,: Morphologic and Clinical Results of TIPSS
Preprocedural Diagnostic Steps The general and disease-specific medical history was recorded along with weight, height, abdominal girth, and skin and mucosal changes specific for liver disease, Hepatic encephalopathy stage I and II was tested using the number connection test INCT) according to Corm [4] in the age-corrected version by Schomerus eta[. [5]. Hepatic encephalopathy stage III and 1V were diagnosed clinically by stupor (stage IlI) and hepatic coma (IVJ.
Noninvasive Diagnostic Procedures Before TIPSS, the following laboratory tests were performed: LDH, GPT, GOT, GLDH. CHE, AP, bitirubim serum protein level including protein electrophoresls, renal excretion function, coagulation status, RBC, WBC, thrombocytes, electrolytes, ammonium, specific liver tumor markers (c~-fetoprotein),and hepatitis serology. Furthermore, we performed abdominal sonograpby (B- and M-mode) to qualify and quantify ascitic fluid and portal flow pattern, chest X-ray film and ECG
343 Several parameters were carefully monitored, such as gastrointestinal bleeding, change ofascitic fluid volume, pulmonary complications, renal function, imbalance of electrolytes, liver function, and hepatic encephalopathy. Further hospital stay depended on the clinical status of the patient. Antlcoagulation therapy was not given systematically and was mostly performed in Child's A and B patients with good liver function and no coagulation deficits, In such cases, heparin was administered intravenously in a dosage that increased the partial thrombin time by a factor of 2 for a maximum of 7 days.
Follow-up Studies These included similar laboratory tests as performed before TIPSS for liver and renal function, coagulation, and blood status, The tests were performed 1 and 3 months after the procedure and from then on at 3-month intervals. Imaging procedures included B-mode and duplex ultrasound and endoscopy parallel to the laboratory tests. Transvenous portography, either transfemoral or transjugular, was done in 6-month intervals. In the first week after TIPSS, a chest X-ray film was taken daily.
Staging of Portal Hypertension This was based on the Child's and Turcotte classification [6] in the modified version of Corm [7] applying the scaling of five clinical and biochemical parameters cnutritional status, bdirubm, ascites, serum albumin level, and hepatic encephalopathy) up to a maximum score of 15 points: a score of 5-8 points corresponded to stage A, 9-12 points to stage B, and 13-15 points to stage C.
Invasive Procedures Before TIPSS These included endoscopy eventually accompamed by sclerotherapy and visceral angiography to demonstrate arterial and portal flow patterns.
Patient Preparation This included correction of coagulation deficit, hematocrit loss and low albumin level to improve the performance status with the aim of creating an elective procedure character. This was accompanied by intense medical control and eventual drainage of ascitic fluid and bowel sterihzation and irrigation to prevent inadvertent protein resorption upon and after the procedure,
TIPSS This technique is described in detad m previous reports [8-10]. In the last 7 patients, the original method has been essentially modified. The most significant change includes the avoidance of a transhepatic approach to the portal vein while the portal bifurcation is visualized by angiography (electronically stored from the previous indirect spleno-portography) and ultrasound instead [11]. Another change was the use of newly designed and better suitable puncture devices (TIPSS-Set, Angiomed Inc., Karlsruhe, Germany or Transjugular Cholangiography Set, Cook, Inc., Bloomington, IN, USA).
Results T I P S S has b e e n s u c c e s s f u l l y e s t a b l i s h e d in 18 o f the 24 p a t i e n t s , r e p r e s e n t i n g a t e c h n i c a l s u c c e s s rate o f 75%. T h e m e a n age o f the p a t i e n t s w a s 60 y e a r s with a range of 34-84 years; 4 patients were female and 14 male. In 4 p a t i e n t s , l i v e r c i r r h o s i s had d e v e l o p e d after h e p a t i t i s B i n f e c t i o n ; the o t h e r s h a d h a d p o s t a l c o h o l i c c i r r h o s i s . T h i r t e e n p a t i e n t s had o t h e r s e v e r e u n r e l a t e d d i s e a s e s , s u c h as d i a b e t e s ( N = 5), malign a n t t u m o r s ( h e p a t o m a , s q u a m o u s cell c a r c i n o m a o f the h y p o p h a r y n x ) (2), s i g n i f i c a n t c o r o n a r y a r t e r y d i s e a s e (2), r e n a l i n s u f f i c i e n c y (2), a b d o m i n a l a n e u r y s m (1), a n d p u l m o n a r y e m p h y s e m a (I). Six patients w e r e classified as stage C h i l d ' s A, 6 as stage B, a n d 6 as stage C. T h e m e a n d u r a t i o n o f s y m p t o m a t i c liver d i s e a s e w a s 10 y e a r s , with a r a n g e o f 3 - 1 7 years. F i v e p a t i e n t s w e r e a c t i v e l y b l e e d i n g u p o n i n t e r v e n t i o n , a n d the p r o c e d u r e was p e r f o r m e d o n an e m e r g e n c y basis. Subclinical hepatic encephalopa t h y w a s p r e s e n t in 10 p a t i e n t s (stage I o f the N T C test), a n d c l i n i c a l h e p a t i c e n c e p h a t o p a t h y in 4 (stup o r in 3 a n d c o m a i n 1) p a t i e n t s b e f o r e T I P S S . I n the g r o u p o f s u c c e s s f u l c a s e s , 5.7 s e v e r e v a r i c e a l bleeding episodes had been recorded and an average o f 6.2 ( r a n g e : 0 - 3 3 ) s c l e r o s i n g p r o c e d u r e s w e r e p e r f o r m e d . S c l e r o t h e r a p y w a s n o t d o n e in 2 p a t i e n t s because e n d o s c o p y revealed gastric varices stage IV and excluded major esophageal varices.
Immediate Results Postprocedural Patient Care Following successful intervention, the patients remained in the intensive care unit until clinically stable conditmns were ensured.
Morphologic Aspects.
The technical success was 75%. I n 5 p a t i e n t s the p u n c t u r e p r o c e d u r e w a s n o t s u c c e s s f u l w i t h i n a r e a s o n a b l e time s p a n ( > 2 h). A s
344
G. Noeldge et al.. Morpholog)c and Clinical Results of TIPSS
,,~ r'
~
~
~
....
A Fig. 1. A 70-year-old patient with liver ctrrhosm. Child's class C, undergoing stentmg of the intrahepatlc transparenchymal tract. A Percutaneous transhepatic portography demonstrating the large caliber of the coronary gastric vein and the vance,,,. B Plato film AP vmw: after placement of a tandem stent.
a result of severe periportal liver fibrosis, we failed to bring a catheter into the portal system in another patient despite successful puncture and positioning of a coronary guidewire in the mesenteric veto. The average procedure time in the successful cases was 5 h (range of 2.5-9 h). In 10 patients the artificial shunt tract was created by bridging the middle hepatic vein with the right main stem of the portal vein in 8, and with the left main stem in 2 patients. In 8 patients the shunt connected the right hepatic vein with the right main stem of the portal vein. In 4 patients one stent, in 12 patients two stents, and in 2 patients three stents were used for scaffolding the intraparenchymal tract (Fig. 1). In 8 patients the diameter of the shunt was 8 ram, in 6 patients it was 9 mm, and in 4 patients it was 10 mm.
mm Hg 40,
30'
20
10'
before TIPSS
Hemodynamic Aspects, Before successful stent shunting, the mean portosystemic gradient of the 18 patients was 29.9 -+ 6.0 mm Hg, range 19-41 mm Hg. After TIPSS, the gradient dropped to an average of 16.9 + 4 mm Hg, range 7-21 mm Hg (Fig. 2). This equals an average reduction of 43% which is highly significant (p < 0.0001).
Complications. During the early postprocedural period of 30 days, t patient died of a direct complication of the procedure. Because of catheter dislocation, we were unable to embolize the lateral percutaneous transhepatic access to the portal system after successful shunt creation. Six hours later, the patient developed substantial intraabdominal bleeding. Emergency packing failed to stop the
after TIPSS
Fig. 2. Immedmte change of the average portosystemic gradient after transjugular intrahepatic porto systemic stent shunt (TIPSS).
bleeding. A second liver packing was successful 3 days later. However, the patient already developed an ARDS, to which he succumbed 12 days later. Another patient developed pulmonary infection 10 days after the procedure and died. Hematomas were present in 5 patients, in 4 at the puncture site for the transhepatic approach and in 1 at the transjugular puncture site. As a consequence, from patient 13 on, we avoided the additional percutaneous approach to the portal system. Another patient, in whom we had
G. Noeldge etal.: Morpholog~c and Climcal Results of TIPSS
345
been unable to establish the stent shunt, also died of bleeding through the transhepatic tract to the portal system.
Long-term Results Clinical Success. The early (procedural) mortality was 1t%, as described above. Within the observation period of 5-28 months, the l-year survival rate was 75% (8/12), and the 2-year survival rate 50% (3/6). Presently, 11 patients are living with an average survival time of 13 months. A total of 5 patients died with an average survival time of 7 months. Death was unrelated to the procedure in 4 cases 12 pneumonias, 1 hepatoma, I left heart insufficiency) and in 1 patient, late shunt occlusion occured 18 months after TIPSS, resulting in a lethal variceal hemorrhage. Noninvasive Imaging. Sonographically, the shunt was easily identified in all patients, because of mild acoustic reverberations from the metallic stent struts. Doppler signals not only penetrated the stent but were also strong enough to allow flow measurements. In these duplex-based flow measurements, diameter error was avoided as the luminal dimension was well known for each individual patient. Therefore, the flow volume was easily determined and varied from 340 to 1900 cc/min with an average of 1210 cc/min. In 15 of 16 patients the individual flow values did not show a significant variation (+-20% over the observation period), In one patient duplex ultrasound showed a 40%, decrease of shunt flow. Shunt angiography from a transjugular approach demonstrated luminal narrowing of the liver vein proximal to the upper end of the shunt tract. Another stent was deployed overlapping the stent shunt. However, 6 months later the patient rebled. Duplex ultrasound confirmed late shunt occlusion (18 months after TIPSS). Invasive Studies. During the observation period, the shunt was angiographically followed by either transfemoral or transjugular approach in 13 patients 6 months after TIPSS, in 5 patients 12 months after TIPSS, and in 3 patients 24 months after TIPSS (Figs. 3, 4). The results are summarized in Table 1, which also includes the measurements of shunt flow and pressure gradients. Six months after TtPSS, a thin neointimal layer was demonstrated angiographically, ranging between 0. I and 0.5 ram. With the exception of our fifth patient (see above), there was no difference at the 12 month and 24 month followup studies in all other patients. The average portosystemic gradient 6 months
J Fig, 3. Four-month follow-up (same patient as m Fig. 11.
after TIPSS was 16.0 + 1.8 mm Hg: this was not statistically different from the immediate results. However, in 1 patient (P 2, Table 1) the gradient measured 23 mm Hg (endoscopy had shown mild erosive gastritis). As a value of >20 mm Hg had been considered as the threshold for sufficient portal decompression, her shunt diameter was increased to 10 mm by transjugular balloon dilatation, resulting in a decrease of her portosystemic gradient to 14 mm Hg. This procedure could be performed on an outpatient basis. Twelve months after TIPSS, the average portosystemic gradient of all patients at risk was 18.2 + 4.5 mm Hg (Table I). This increase resulted from the venous outflow stenosis of 1 patient already mentioned above. After correction by insertion of another stent, his gradient dropped from 26 mm Hg to 12 mm Hg. As a restllt the average gradient of the cohort group dropped also to 15.4 mm Hg (Table 1).
Discussion The concept of TIPSS corresponds to the hemodynamic situation of a small caliber surgical H-interposition shunt leading to partial portal decompression. With TIPSS, the interposition is established by a stented parenchymal tract bridging either of the main stems of the portal vein with the right or middle hepatic vein. From a hemodynamic point of view this concept offers a major advantage compared with
346
G Noeldge et al." Morpholog~c and Clinical Result~ of TIPSS
Table 1, Transjugular intrahepat~c portosystemic stent-shunt: angiograph~c and hemodynam~c results o f patients sur,,qving more than 6 m o n t h s
Patient
6 mo
12 mo
24 m o
Shunt
PS-grad.
Shunt
PS-grad.
Shunt
lumen (mm~
(ram Hgl
lumen Imm)
( m m Hg}
lumen ~mm)
P2 P3 P4 P5 P6 P8 Pl0 Pll P12 PI3 PI4 PI5 P16
7.7/10" 7.8 Not performed 9.5 8.9 8.9 8.5 8.8 7.7 6.0 8.6 8 7 9.9
23/14" 16
9,7 7,8 7.7 5.6/10 ~' 8.6 D~ed
18 15 16 26/12" 16
Dmd 7.8 7.7 D~ed 8.9
M e a n -+ SD
8 6 -+ I I
16 -* 1.8
14 18 18 17 17 15 17 16 15 t2
To To To To To To To 7_9 = 1.5 8.7 = t.1
be be be be be be be
PS-grad. (ram Hg)
t5 15 14
performed performed performed performed pert;ormed performed performed
18.2 -*- 4.5 t5.4 m 2,2
PS-grad = p o r t o s y s t e m i c gradmn{ " Redilatation o f the shunt ~gradmnt > 2 0 m m Hg) by a 10-mm balloon ~' B e c a u s e of s t e n o m s in the Liver vein proximal to the stent shunt, implantation of another stent 110 ram)
bu
9
9
"
',- g /',3!i
Fig. 4. Two-year follow-up (same patient as Fig. 3). A Plain film A P view shows kinking o f the t a n d e m stent, B indirect splenoport o g r a m s h o w i n g a patent shunt despite kinking of the t a n d e m stent: no varices.
the surgical technique. Scaffolding the shunt by balloon expandable Palmaz stents allows adaptation of the shunt lumen to the individual hemodynamic situation, ranging from 7 to 14 mm and depending on the size of the balloon catheter used. The shunt diameter may be increased whenever necessary by
i]
9
! ,';'{
:-
9
"
simple transjugular approach. In our series this was performed in 1 patient, In view of the applied inclusion criteria it is evident that our series involves a relatively high number of patients with end stage liver disease. Two-thirds of the successfully treated patients had a stage Child's B or C. Compared to historical groups of shunt surgery, this represents a highly negative patient selection [I1-13]. The 6 patients in stage Child's A were enrolled either because of a prospectively high surgical risk (other unrelated diseases) or because of previously failed shunt surgery. In 5
G. Noeldge et al.: Morphologic and Chnical Results of TIPSS
cases, TIPSS was performed as an emergency intervention. The technical aspects of TIPSS included a significant learning curve during which we modified the applied materials substantially. In the first 3 cases, the procedure took roughly 8 h, but lasted 2-3 h in the later cases. The combination of increased experience with specificially designed needles helped abandon the additional access to the portal system via a midaxillary transhepatic approach. This approach had caused significant complications (two deaths). The combination of a high quality indirect splenoportogram with ultrasound could be used effectively for guidance of the transjugular puncture [101. The key element for TIPSS is the balloonexpandable stent. It allows for a smooth and regular lumen of the scaffolded tract and a predictable shunt geometry. By using the Palmaz stent type, an individual tailoring of the shunt diameter to the hemodynamic needs of each patient is feasible. Technically~ it proved to be helpful to start with an 8 mm shunt diameter and, eventually, increase it by l-mm increments, depending on the measured gradients. Presently, the Palmaz stent is the only such stent available on the market for this technique. It offers the highest resistance to radial compression compared with other stent types [3]. Angiography clearly documents the change of the portal hemodynamic situation by the TIPSS procedure. All of our patients showed a substantial decrease of variceal flow associated with a reduction of peripheral portal venous filling. In 2 patients, no peripheral portal flow was found after the procedure, while their portosystemic gradient was 14 mm Hg and 12 mm Hg, respectively. In the follow-up studies, not much changed compared with the situation immediately after TIPSS. None of the patients developed a significant decrease of liver function after TIPSS. Clinically significant hepatic encephalopathy was found for 3 days in 1 patient 9 months after TIPSS. In 2 patients, gastrointestinal rebleeding episodes were recorded. One occurred in our first patient in whom we had not drained a large amount of ascites before the procedure. Rapid abdominal fluid resorption after the well-functioning shunt may have caused uptake of fibrinolytic substances from the ascites and induced widespread mucosal bleeding. This was easily controlled by fresh blood and fresh frozen plasma. In the second patient, bleeding was associated with late shunt occlusion. These overall results compare well with the surgical shunt literature, where hepatic encephalopathy is reported as high as 50% in nonselective shunts and up to 20% in selective shunts [11, 12, 14-16]. Also, rebleeding is reported in the
347
surgical literature almost exclusively after shunt occlusion and independent of the type of surgical shunt. The mortality rate of 11% in our series seems to be acceptable, taking into account the negative selection criteria applied for the study. For comparison, emergency shunt operation for patients in a stage Child's C is reported as high as 40%-100% [13]. However, the total number of our patients is too small for definite conclusions on the basis of stage-related risk stratification. Five patients died during the long-term follow-up period: 3 died of reasons related to their liver disease (1 hepatoma, 2 severe infections), 1 died of late shunt failure, and 1 of reasons unknown, Comparing the 1-year survival rate of 6/8 and the 2-year survival rate of 3/6 in our series with the literature, the clinical validity of the concept is underscored. In 6 of 7 patients who died, a histomorphologic analysis was performed; some of the results are already published in detail [3, 8]. These results may be summarized by one simple fact: a neointima develops on the stent shunt surface indistinguishable from arterial application of the stent. In summary, hemodynamically, TIPSS represents a small H-shunt with partial portal decompression. Lately, Rypins et al. [17] and Johansen [18] reported favorable results with the surgical establishment of this shunt type. Apart from its pet-cutaneous character, the TIPSS concept offers one major advantage over the surgical shunt which is adaptability of the shunt size to the individual hemodynamic needs. The interposition is exclusively intrahepatic. The fact that the extrahepatic vascular trunks remain untouched represents a significant advantage if liver transplantation is taken into consideration, it seems feasible to prevent the risk of lethal variceal bleeding in patients already on a waiting list for organ transplantation by performing a TIPSS procedure. After our learning experience, it appears that the technique and the materials used for the procedure are sufficiently defined. A wider clinical application of the method appears warranted.
References 1. Palmaz JC, Sibbitt RR, Reuter SR, Garcia F. Tlo FO 11985) Expandable intrahepatic portacaval shunt stents: Early experience in the dog. A JR 145:821-825 2. Palmaz JC, Garcta F, Sibbit SR, Tio FO, Kopp Dr', Schwesinger W. Lancaster JL, Chang P (1986) Expandable intrahepatic portacaval shunt stents in dogs with chronic portal hy. pertension. A JR 147:125 I - 1254 3. Richter GM, Palmaz JC, Noldge G. R6ssle M, Siegerstetter V, Franke M, Wenz W t1989) Der transjugul/ire intrahepatische portosystemische Stent-Shunt (TIPSS). Radiologe 29:406-411
348 4. Conn HO (1977) Trailmaking and Number-Connection Test in the Assessment of mental state in portal systemic encephalopathy. Am J Dig Dis 22:541-550 5. Schomerus H. Hamster W, Reinhard U, Mayer K, D011e W 11981) Latent portosystemic encephalopathy. Dig Dis Sci 26:622-630 6. Child CG, Turcott JG (1964) Surgery and portal hypertension. ln: Child CG (ed) The liver and portal hypertension. WB Saunders, Philadelphia 7_ Conn HO (1981) A peek at the Child-Turcotte classification. Hepatology 1:673-676 8. Richter GM, Noeldge G, Patmaz JC, Roessle M, Siegerstetter V, Franke M, Gerok W, Wenz W, Farthmann E ( 19901Transjugular intrahepatic portacaval stent shunt: Preliminary clinical results. Radiology 174:1027-1030 9. Richter GM, Noetdge G, Palmaz JC, Roessle M (1990) The transjugular intrahepatic portosystemic stent-shunt (TIPSS): Results of a pilot study Cardmvasc Intervent Radiol 13:200-207 10. Noldge G, Richter GM, Rossle M, Perarnau JM, Palmaz JC (1991) Technical improvement of the transjugutar intrahepatic portosystemic stent-ass~sted shunt (TIPSS): A quicker and safer ultrasound-assisted method (abstract). Presented at the 16th annual meeting of the SCVIR, San Francisco 1991--Program book p 172
G. Noeldge et al.: Morphologic and Clinical Results of TIPSS 1 l. Millikan W J, Warren WD, Henderson JM, Smith RB, Salam AA, Galambos JT, Kutner MH, Keen JH (1985) The Emory prospective randomized trial: Selective versus non-selective shunt to control variceal bleeding. Ann Surg 201:712-722 12. Warren WD, Millikan WJ Jr, Henderson JM, Wright L, Kutner M, Smith RB 3rd, Futenwider JT, Salam AA, Gatambos JJ (1982) Ten years portal hypertensive surgery at Emory. Results and new perspectives. Ann Surg 195:530-542 13, Galambos JT (1985) Portal hypertension. Semin Liver Dis 5:277-290 14 Foster JH, Elhson LH, Donovan Th, Anderson A (1971) Quantity and quality of survival after portosystemic shunts. Am J Surg 12:490-501 15. Pagliaro L, Burroughs AK, Sorensen TIA, Lebrec D, Morabito A, D'Amico G, Tint F (1989) Therapeutic controversies and randomised controlled trials (RCTs): Prevention of bleedmg and rebleeding in cirrhosis. Gastroenterol Int, 2"71-84 16. Rossle M, Haag K, Noeldge G, Richter G, Wenz W, Farthmann E, Gerok W (1990) Hemodynamic consequences of portal decompression: Which is the optimal shunt? Z Gastroenterol 28:630-634 17. Rypins EB, Mason GR, Conroy Rbl, Sarfeh IJ (1984) Predictability and maintenance of portal flow patterns after smalldiameter portocaval H-grafts in man. Ann Surg 200:706-710 18. Johansen K ~1989J Partial portal decompression for variceal hemorrhage. Am J Surg 157:479-482
