- Email: [email protected]
Vascular Interventions: Portal Hypertension
embolization of hepatic artery. All patients showed neither end-organ damage nor organ ischemia related to the procedure. Recurrent bleeding did not occur in any patients during the follow-up periods of 6 to 36 months. CONCLUSION: NBCA embolization is a feasible and effective technique for ruptured pseudoaneurysms, when selective catheterization is difficult, or when rapid and complete hemostasis is not achieved with the conventional isolation technique using microcoils. Abstract No. 316 EE How To Manage Iatrogenic Vascular Injury: Imaging and Interventions. H.J. Jae, Seoul National University College of Medicine, Seoul, Republic of Korea 䡠 J.W. Chung 䡠 Y.H. So 䡠 W. Lee 䡠 J.H. Park PURPOSE: Iatrogenic vascular injury may result from any invasive diagnostic or therapeutic procedures and the incidence is increasing as the catheter-based techniques are more frequently employed. Most iatrogenic vascular injuries can be managed successfully with interventional procedures such as transcatheter embolization and deployment of stent or stent-graft. In this exhibit, we would like to present the cross-sectional and 3-dimensional imagings of various iatrogenic vascular injuries and illustrate how these injuries have been successfully managed with intervention. MATERIALS AND METHODS: During the past 7 years, we have experienced iatrogenic vascular injury in 65 patients managed by interventions. The injuries included: Post biopsy bleeding in liver and kidney, bleeding after percutaneous biliary or abscess drainage, bleeding after paracentesis or pleural tapping, gastrointestinal bleeding after endoscopic procedures, intraoperative vascular injury, iliac artery dissection after femoral catheterization, iliac arteriovenous fistula after spine surgery, et cetera. TEACHING POINTS: 1. To understand the imaging appearance of various iatrogenic vascular injuries with crosssectional and 3-dimensional reformatted CT images and angiography. 2. To understand how to manage successfully those iatrogenic vascular injuries with intervention. Abstract No. 317
PURPOSE: The purpose of this study is to evaluate the rate of recurrence and effectiveness of treatment of various vascular malformations as measured by the need for follow-up treatment. MATERIALS AND METHODS: Between 1981 and 2003, 860 direct stick and transcatheter embolizations were performed at a single center. 583 vascular malformation cases with at least five year follow-up data were evaluated. There were 476 arteriovenous malformations and 80 venous malformations. The upper or lower extremities were involved in 209 cases. The remainder included 104 pelvic AVM, 62 renal AVM, and 35 pulmonary AVM. Need for follow-up procedures was determined based on clinical signs and symptoms. Calculations were performed to assess the percentage of AVM by location requiring, 1) at least one
RESULTS: A total of 121 cases required at least one additional follow-up procedure. Thirty-five percent (26 repeat procedures in 75 patients) and 46 percent (18 in 39 patients) of leg and foot AVM, respectively, required at least one follow-up procedure. Eleven percent (13 in 115 patients) of AVM in visceral organs, including lungs, GI tract, liver, bladder, spleen, and kidneys, needed at least one follow-up treatment. Twenty percent (8 in 39 patients) of lower extremity AVM, followed by 14 percent (15 in 104 patients) of pelvic AVM, required at least two or more follow-up procedures; less than one percent of pulmonary, GI, liver, bladder, splenic, or renal AVM required similar follow-up. AVM located in the lower extremity necessitated the greatest number of total follow-up procedures; 80 repeat embolizations were needed to resolve 114 AVM in the foot and leg. CONCLUSION: Vascular malformations have high rates of recurrence due to collateralization of vessels. AVM in visceral organs have the lowest rate of recurrence as measured by need for a minimum of one follow-up procedure. Lower extremity followed by pelvic AVM had the highest rates of recurrence. Complex AVM, as measured by cases requiring extensive treatment, tend to reside in the lower extremity and pelvis; in contrast, most visceral AVM resolved with one treatment. Abstract No. 318 Ear Arteriovenous Malformation Management. W.F. Yakes, Vascular Malformation Center, Englewood, CO, USA PURPOSE: To determine the efficacy of Ethanol Endovascular Repair of Ear Arteriovenous Malformation (AVMs). MATERIALS AND METHODS: Six patients (5 female, 1 male; age range 6-39 years; mean age: 22 years) with ear AVMs presented for therapy. Two patients had failed prior embolizations (PVA/coils/nBCA/steroids) and 2 patients had other therapies (laser/excisions/grafting). All presented with grossly enlarged painful ear with intermittent bleeding. All patients underwent transcatheter and direct puncture ethanol Rx (77 procedures). RESULTS: All 6 patients were cured of the AVM at longterm follow-up (mean follow-up: 39 months). One patient had transient partial VII nerve palsy. Two patients had minor blisters and ear injuries.
POSTER SESSIONS
Long-Term Follow-up of Rates of Recurrence with Arteriovenous Malformation Treatment. S. Rao, Lenox Hill Hospital, New York City, NY, USA 䡠 R. Rosen
follow-up procedure, 2) at least two or more follow-up procedures, and 3) highest total number of procedures.
CONCLUSION: Ethanol endovascular repair of Ear AVMs can effect cures in this vexing lesion that previously was treated with resection of the ear and high recurrence rates.
Vascular Interventions: Portal Hypertension Abstract No. 319 Balloon-Occluded Retrograde Transvenous Obliteration in 42 Patients with Gastric Varices or Hepatic Encephalopathy. M. Honda, Showa University Hospital, Tokyo, Tokyo-to, Japan 䡠 T. Hashimoto 䡠 T. Baba 䡠 K. Sekiyama 䡠 T. Kitanosono 䡠 T. Gokan
S115
PURPOSE: To evaluate the clinical efficacy of balloonoccluded retrograde transvenous obliteration (B-RTO) for the treatment of gastric varices or hepatic encephalopathy, and to describe technical problems and complications experienced during this procedure.
examination, Doppler ultrasonography and if necessary even portal venography. In all 4 groups the primary patency after the specific intervention was calculated using KaplanMeier analysis. These results of primary shunt patency were mutually compared using Cox’s F-test and log-rank test.
MATERIALS AND METHODS: From November 1994 through August 2006, we performed B-RTO in 42 patients (25 men and 17 women; range, 35-83 years; mean age, 63.2 years) with gastric varices (31 patients) or hepatic encephalopathy (11 patients). Gastric varices or hepatic encephalopathy were associated with portal hypertension as a result of viral liver cirrhosis in 25 patients, and alcoholic liver cirrhosis in 9 patients. Child-Pugh class was A in 11 patients, B in 25 patients, and C in 6 patients. 9 patients had hepatocellular carcinoma. Gastrorenal shunts were seen in 37 patients. Shunt tracts draining into the gonadal vein were seen in 5 patients. 47 procedures were performed for these patients. Transfemoral approach was used in 28 procedures, transjugular approach in 16 procedures, and both approaches were used in 3 procedures. After balloon-occluded retrograde venography, the sclerosant (ethanolamine oleate: EO) was injected into the gastric varices or shunt tracts during balloon occlusion. The maximum dose of EO at 5% was 20 mL/day. 3 patients underwent dual balloon-occluded embolotherapy using both methods of transjugular intrahepatic portosystemic shunt (TIPS) and B-RTO. After BRTO, CT and endoscopy were used to assess whether the varices or shunt tracts were thrombosed.
RESULTS: The intervention was successful in 120 cases (the overall technical success of all interventions was 99.2%). Primary shunt patency after the intervention in the group treated with plain angioplasty was 49.7% after 12 months and 25.3% after 24 months, in the group with deployment of bare stents it was 74.9% after 12 months and 64.9% after 24 months. In the non-dedicated ePTFE-covered stent-grafts group it was 75.2% after 12 months and 64.5% after 24 months and after deploying dedicated ePTFE-covered stent-grafts 88.1% were still patent at 12 months and 80.8% after 24 months after intervention. A statistically significant improvement in shunt patency was achieved in the group with implanted dedicated ePTFEcovered stent-grafts and in the group with bare stents in comparison with the group with plain angioplasty (p ⱕ 0.01).
RESULTS: Technical success was achieved in all patients. Follow-up CT after the procedure revealed thrombosed gastric varices or shunt tracts. All gastric varices were markedly decreased in size. During the follow-up period, one patient showed regrowth of gastric varices and the second B-RTO was performed. Hepatic encephalopathy was improved in 11 patients. Although extravasation of contrast media was seen in 2 patients, no significant sequela was observed in these patients. We have experienced no lifethreatening complications.
Abstract No. 321 EE
CONCLUSION: B-RTO is a safe and effective method of treatment for patients with gastric varices or hepatic encephalopathy due to portal hypertension. Abstract No. 320 Influence of the Secondary Deployment of ePTFE-Covered Stent-Grafts on the Maintenance of TIPS Patency. V. Jirkovsky, Teaching Hospital of Charles University in Hradec Kralove, Hradec Hralove, East Bohemia, Czech Republic 䡠 T. Fejfar 䡠 P. Hulek 䡠 A. Krajina 䡠 V. Safka 䡠 V. Chovanec, et al. PURPOSE: Evaluate the effects of secondary deployment of ePTFE-covered stent-grafts in the treatment of dysfunctional TIPS in comparison with other ordinary approaches (plain angioplasty or implantation of bare stents). MATERIALS AND METHODS: In our center, in the period from 2000 to 2004, a total of 121 cases of dysfunctional TIPS were treated with shunts that had initially been established using bare stents. The group was divided into 4 sub-groups based on the type of intervention: plain angioplasty (52 cases, 43%), deployment of a bare stent (35 cases, 28.9%), deployment of non-dedicated ePTFE-covered stentgraft (15 cases, 12.4%) and deployment of dedicated ePTFE-covered stent-graft (19 cases, 15.7%). Following the intervention all patients were monitored at regular intervals to observe the patency of the shunt with the aid of clinical S116
CONCLUSION: Amongst all widely used interventional methods for the treatment of dysfunctional TIPS, the best consequent patency was obtained by deploying dedicated ePTFE-covered stent-grafts.
Local Transplantation of Endothelial Progenitor Cells to Reduce Atherosclerotic Plaque Formation in a Rabbit Model. M. ZhanLong, Zhong-Da Hospital, Southeast University, Nanjing, Jiangsu, China 䡠 T. GaoJun 䡠 M. XiaoLi PURPOSE: To investigate whether catheter-based, local transplantation of homogeneous endothelial progenitor cells (EPCs) could prevent atherosclerotic plaque formation in a New Zealand rabbit model of atherosclerosis. MATERIALS AND METHODS: EPCs of New Zealand rabbits were isolated, confirmed, expanded and then incubated with home synthesized Fe2O3-PLL for 24 hours. Prussian blue stain was performed to identify intracellular iron. The carotid artery injured model in rabbits was induced by dilation with a 2.5-F balloon. Magnetically labeled-EPCs and fluorescently labeled-EPCs were transplanted into the injured endothelium in group A (n ⫽ 11) and group B (n ⫽ 5), respectively, while an equal volume of saline without EPCs was injected into the injured endothelium in group C (n ⫽ 12). MR imaging and, histopathology were performed at 7 days after transplantation for 2 rabbits randomly chosen from each group. All the remaining rabbits were kept to be fed with a high lipid diet until 15 weeks when MR examination was done. Histological examinations including HE staining were done immediately after the MR imaging. TEACHING POINTS: Iron-labeling efficiency within EPCs reached more than 95% identified by Prussian blue stain. At 7 days after EPCs transplantation, MR showed signal intensity loss of the injured carotid artery on T2*WI MRI only in group A, which was correlated with the areas where the most Prussian blue staining-positive cells were in histopathology, while histopathological slides showed the fluorescence-positive cells in the injured endothelium in group B. At 15 weeks after EPCs transplantation, less plaque or no plaque were exhibited in the rabbits of group A and B
receiving EPCs transplantation compared to that of the group C (P ⬍ 0.05). The catheter-based, local transplantation of ex vivo expanded homogeneous endothelial progenitor cells contribute to prevent atherosclerotic plaque formation in New Zealand rabbit model of atherosclerosis. Abstract No. 322 Foam Sclerotherapy for Gastric Varices and Splenorenal Shunt under CT or FACT. J. Koizumi, Tokai University, Isehara, Kanagawa Prefecture, Japan 䡠 Y. Kawawa 䡠 T. Hashimoto 䡠 K. Myojin 䡠 S. Kawada 䡠 Y. Imai PURPOSE: For gastroduodenal varices, percutaneous transhepatic obliteration (PTO), balloon occluded retrograde transvenous obliteration (BRTO) or dual balloon occlusion embolotherapy (dBOE) can be used. As one of the embolic materials, ethanolamine oleate (EO) is mainly used for BRTO. However, the overdosage of EO may cause severe complications such as hemolysis, allergy, ARDS, etc. Thus foam sclerotherapy was introduced to reduce the amount of EO to obliterate large varices under combined angiographyCT or flat detector angiographic CT (FACT) system. MATERIALS AND METHODS: For ten patients, seven BRTO, two PTO, and one dBOE were performed. Under balloon occlusion, a mixture of 20 mL of 5% EO with contrast media and 20 mL of air using a pumping method was injected into the target varices until full occupation was obtained. In the last three patients with gastric varices, 1 mL of polidocanol (Aethoxysklerol, Kaigen) mixed with 4 mL of air was used with or without EO. During the balloon occlusion CT or FACT was performed to confirm the filling of air mixed with the sclerosant into the target vessels. After thrombosis was achieved the catheter was retrieved and followed by CT and endoscopy. RESULTS: In all patients, air mixed with the sclerosant was observed in the target vessels during the procedures and full thrombosis was confirmed on postcontrast CT one week after the procedures. No procedure-related complications occurred. When performing BRTO, the light foam sclerosants tended to ascend directly into the target gastric varices which are located more ventral than the gastro-renal shunt. On the contrary, the foam sclerosants induced splenic venous thrombosis possibly because the splenic vein is more ventrally located than the target gastric varices during PTO.
Abstract No. 323 Direct Intrahepatic Portosystemic Shunts in Patients with Anatomic Contraindications to Traditional TIPS. J.C. Smith, Loma Linda University Medical Center, Loma Linda, CA, USA 䡠 G.E. Watkins 䡠 E.W. Palmer 䡠 M.H. Mendler 䡠 B. Petersen PURPOSE: An intravascular ultrasound (IVUS)-guided direct intrahepatic portosystemic shunt (DIPS) procedure has recently been described in patients with normal anatomy. We present a small series of 13 patients with various disease processes, in whom traditional transjugular intrahepatic por-
MATERIALS AND METHODS: From October 21, 2003, to September 8, 2006, 13 direct inferior vena cava-to-portal vein shunts were created in 13 patients using real-time guidance with a variable 5-10 MHz sagittal IVUS imaging system. Indications for DIPS included refractory ascites (n ⫽ 8), variceal bleeding (n ⫽ 4), and hepatohydrothorax (n ⫽ 1). Entities precluding routine TIPS placement included portal vein thrombus (n ⫽ 6), Budd Chiari syndrome (n ⫽ 1), polycystic liver disease (n ⫽ 1), hepatohydrothorax (n ⫽ 1), and previous failed TIPS attempts (n ⫽ 4). Shunts were created using 10-mm-diameter Viatorr stent grafts, but they were generally dilated to only 8 mm. Follow-up was performed clinically and using primarily transabdominal ultrasound imaging. RESULTS: All shunts were deployed successfully. One intended extrahepatic portal vein puncture led to clinically significant hemorrhage, which responded to blood transfusion. One patient developed acute liver failure leading to death at 2 days. No other immediate procedural related complication was identified. The mean portosystemic gradient was reduced from 26 mm Hg before to 8 mm Hg after DIPS placements. CONCLUSION: IVUS-guided DIPS placement can safely and effectively provide portal decompression in patients with venous occlusive disease or other anatomic abnormalities that may make traditional TIPS technically challenging or impossible to perform. Abstract No. 324 Endovascular Treatment of Portal Vein Abnormalities in Patients Undergoing Orthotopic Liver Transplantation. M. Marini, C. H. U. Juan Canalejo, A Corun˜a, Galicia, Spain 䡠 I. Cao 䡠 M. Gomez-Gutierrez 䡠 D. Gulias 䡠 J. Quintela 䡠 J. Aguirrezabalaga, et al. PURPOSE: To evaluate the efficacy of endovascular procedures for the treatment of acquired portal vein (PV) abnormalities from the time of transplantation to hepatic graft loss. MATERIALS AND METHODS: For the last 10 years, 25 patients with orthotopic liver transplants (OLT) and PV abnormalities have undergone endovascular procedures. Intraoperative period (n ⫽ 11): 2 patients with a re-graft had PV rupture during hepatectomy phase and underwent stent implantation in the interposition vein graft, and 9 patients with PV thrombosis (PVT) were treated by splanchnic vessel recanalization, primary stent and closure of competitive shunts during post-reperfusion phase. Perioperative period (n ⫽ 3): one PV rethrombosis and one PV torsion was treated by a stent, and a patient with anastomotic PV stenosis and stenosis in an interposition vein graft was treated by venoplasty. Early postoperative period (n ⫽ 4): 3 patients with anastomotic PV stenosis were treated by stent and one with stenosis in a jump graft was treated by venoplasty. Late postoperative period (n ⫽ 7): 4 patients with bleeding, PVT and chronic rejection were treated by PV recanalization, stent and a TIPS; one patient with chronic PVT and bleeding was treated by PV recanalization and stent; in the last 2 patients, with hypersplenism and ascites due to chronic rejection and PVT, one was treated by PV recanalization, stent and a TIPS. In the other, PV recanali-
POSTER SESSIONS
CONCLUSION: Foam sclerotherapy allows the reduction of the sclerosant theoretically to less than half dose. BRTO seems to be a better indication for foam sclerotherapy than PTO. Combined angiography- CT or FACT during the procedures was useful in confirming air mixed with the sclerosant in the target varicose veins protruding into the gastric lumen.
tosystemic shunt (TIPS) creation was considered difficult or impossible, who subsequently underwent successful DIPS placements.
S117
zation was unsuccessful and we performed a shunt of Leevin.
Abstract No. 326
RESULTS: Technical success was achieved in 24 of 25 patients (96%). None died during the procedure, allograft portal perfusion was adequate and portal hypertension solved. All venoplasties, stents and TIPS remained patent without further intervention, even in patients who died. One patient died within a week of TIPS of hepatic failure. At follow-up 7 patients died (19,39 mo 2-60) but it was not procedural related, 17 patients remain alive (40 mo 3-90): 2 underwent a re-graft and the others required no further intervention.
Clinical Efficacy of Embolization with NBCA for Esophageal-Gastric Fundal Varices in Portal Hypertension. K. Zhu, The Third Affiliated Hospital, Sun Yat-sen University, Guangzhou, Guangdong, China 䡠 H. Shan 䡠 Z. Li 䡠 Z. Jiang
CONCLUSION: Endovascular techniques in OLT are safe and effective and, in our opinion, the best choice in acquired PV abnormalities in OLT. Although stent primary patency in medium- and long-term follow-up was good, results must be confirmed. Abstract No. 325 Transjugular Intrahepatic Portosystemic Shunt: Evaluation of Impact of Stent Configuration on Patency Rate. P.J. Schaefer, University Hospital Schleswig-Holstein Campus Kiel, Kiel, Schleswig-Holstein, Germany 䡠 T. Jahnke 䡠 F.K.W. Schaefer 䡠 M. Heller 䡠 S. Mueller-Huelsbeck PURPOSE: To evaluate the impact of the configuration of the stent on the patency rate after transjugular intrahepatic portosystemic shunt creation with a self-expanding stent in patients with refractory ascites or variceal rebleeding caused by portal hypertension. MATERIALS AND METHODS: Totally, 80 patients (60 male, 20 female; mean-age 56 ⫾ 9.6, range 37-81) with transjugular intrahepatic portosystemic shunt were evaluated. 30 patients had refractory ascites, 32 variceal rebleeding, and 18 both. Primary technical success rate, interventional revision rate, and mean patency rate due to KaplanMeier were calculated. The angle as deviation of the bloodflow at the portal venous inflow and central venous outflow were measured on projected angiograms. The following five angle-groups were set up: 1) portal venous inflow, 2) central venous outflow, 3) maximum and 4) minimum angle, and 5) sum of both angles in the shunt system. Within each group, Mann-Whitney test and after dichotomic partition by the median Pearson’s Chi square test and Fisher’s exact test were carried out to prove for dependency of patency on the stent⬘s configuration. RESULTS: Primary technical success rate was 93%, interventional revision rate 28%, and mean patency rate 17.5 months. Mean/ standard deviation/ median angle were for the 1) portal venous inflow 66.5°/ 19.0°/ 65°, 2) central venous outflow 43.4°/ 14.0°/ 40°, 3) maximum angle 69.0°/ 16.1°/ 65°, 4) minimum angle 40.6°/ 13.3°/ 40°, and 5) sum of both angles 109.6°/ 21.9°/ 107.5°. The 2-sided values of significance were in the Mann-Whitney test/ Chi square test/ Fisher exact test for the 1) portal venous inflow 0.124/ 0.069/ 0.091, 2) central venous outflow 0.714/ 0.673/ 0.810, 3) maximum angle 0.178/ 0.046/ 0.056, 4) minimum angle 0.578/ 0.622/ 0.632, and 5) sum of both angles 0.157/ 0.228/ 0.335. CONCLUSION: The shunt’s patency rate with use of a self-expanding stent is not dependent on the stent⬘s configuration regarding the deviation of the blood-flow at the portal venous inflow and central venous outflow, and the maximum, minimum and total deviation in the shunt. S118
PURPOSE: To evaluate the efficacy of embolization with N-butyl-2-cyanoacrylate (NBCA) for esophageal-gastric varices in portal hypertension. MATERIALS AND METHODS: 28 cirrhotic patients with esophageal-gastric varices were enrolled in this study. Before this study, 15 patients received endoscopic variceal ligation, 13 received conservative pharmacotherapy, and all the patients suffered from rebleeding. Percutaneous transhepatic or transplenic portography was performed in all 28 patients; then, esophagogastric varices were embolized with NBCA. The proportion of NBCA and lipiodol was determined according to the size and blood flow of varices, ranging from 1:4 to 1:8. For the patients who had a large gastrorenal shunt (GRS), a catheter with a balloon was introduced into the GRS via the right femoral and left renal vein before the embolization, and the balloon was inflated to block the flow of GRS. Portal venous pressure was measured before and after embolization in all patients. RESULTS: Of the 28 patients, we successfully embolized the left gastric vein, posterior gastric vein and short gastric vein in 27 cases; balloon-occluded technique was applied in 8 patients with large GRS. Portal venous pressure significantly raised from mean, 34.7 cm H2O before embolization to mean, 38.7 cm H2O (P ⬍ 0.05) after embolization. In four patients, little NBCA entered into distal pulmonary artery branches. Two of them suffered transient irritable cough; no patient developed severe pulmonary embolism. Endoscopic examination was performed in 11 patients after embolization, which demonstrated that the varices of all 11 patients were markedly alleviated. In the 14 patients who underwent CT 3 to 6 months after treatment, the embolized varices were still obstructed in all of them, but some other collateral veins were observed in three patients. In this study, the patients were followed up from 3 to 26 months (mean, 9.9 months). Rebleeding was observed in 5 patients (17.9%). CONCLUSION: Embolization with NBCA is a safe and effective method for the treatment of esophagogastric varices; but it is necessary to block the large GRS via retrograde approach to prevent pulmonary embolism. Abstract No. 327 EE Measuring the Hepatic Vein Pressure Gradient: Review of Technique with Emphasis on Accuracy and Standardization. J.M. LaBerge, University of California, San Francisco, San Francisco, CA, USA 䡠 R.K. Kerlan PURPOSE: The hepatic vein pressure gradient (HVPG) is an indirect reflection of portal pressure that has received renewed interest with the recent observation that, in patients with compensated cirrhosis, HVPG is the best predictor of the development of varices and that a reduction in HVPG in response to pharmacologic therapy correlates with bleeding risk. The technique of obtaining HVPG while not new has not been standardized. Recently, a NIH consensus panel (June 2006) has called for standardization of HVPG technique in anticipation of an expanded role of HVPG mea-
surement in the treatment of portal hypertension. The purpose of this teaching demonstration is to describe current HVPG measurement technique and identify steps that may improve accuracy. In addition, recent literature citing the value of HVPG measurement in monitoring portal hypertension will be reviewed. MATERIALS AND METHODS: We describe and illustrate the steps necessary to perform HVPG measurement with emphasis on the following: - Selection of the hepatic vein and proper location for measurement - Use of end hole versus balloon catheter - Number of repetitive measurements necessary to insure reproducibility - Calibration and use of pressure monitor equipment for accuracy - The value of waveform analysis and use of printed tracing versus digital readouts. Recent work by Grozsmann RJ, et al (NEJM 2005; 353; 2254-61) has demonstrated the relationship between HVPG and clinical endpoints in patients treated with pharmacologic agents. Results of this paper and other studies demonstrating the clinical utility of HVPG measurement will be reviewed. HVPG is currently the only method for accurately monitoring portal pressure and tailoring the delivery of pharmacologic therapy to the needs of an individual patient. TEACHING POINTS: Measurement of HVPG may play an increasingly important role in monitoring patients with portal hypertension. Standardized techniques that yield accurate and reproducible measurements are necessary for the widespread acceptance of this procedure. These include use of balloon occlusion catheter, electronic pressure transducer calibrated for venous readings, and pressure tracings with readings averaged over three stable values. Abstract No. 328 Using the Self-expanding Sinus Superflex Visual Stent System for TIPS Creation: Initial Experiences. J. Wiskirchen, University Hospital Tuebingen, Tuebingen, Baden-Wuerttemberg, Germany 䡠 A. Fischmann 䡠 G. Tepe 䡠 U. Lauer 䡠 W. Steurer 䡠 P.L. Pereira PURPOSE: Aim of this work is to analyze the experiences with a 6-F self-expanding nitinol stent system (Sinus Superflex Visual, Optimed, Germany) in creating transjugular portosystemic shunts (TIPS) in patients suffering from acute bleeding or refractory ascites.
RESULTS: TIPS creation was successful in all patients. In 8 patients a single stent with a diameter of 12 mm and a length of 60 mm was used, in five patients stent length was 80 mm. All stents were deployed without any problem, the technical success rate was 100%. Mean pressure gradient before all interventions was 22.8 mm Hg (range 15 – 32), after the intervention the gradient dropped to 8.5 mm Hg (range 5 – 11). One patient who was treated because of subacute Budd Chiari syndrome suffered from an acute stent occlusion
CONCLUSION: The self-expanding Sinus Superflex Visual Stent can be used for TIPS creation successfully. Due to its superb visibility and simple release system, stent placement was easy in all cases. However, long term follow-up has to be performed to evaluate the long term performance of the stent. Abstract No. 329 Analysis of Procedure-Related Mortality of BRTO in Patients with Gastric Variceal Bleeding. Y. Kim, Keimyung University School of Medicine, Joonggu, Daegu, Korea 䡠 C. Kim 䡠 J. Choi 䡠 T. Shin PURPOSE: To analyze procedure-related mortality of balloon occluded retrograde transvenous obliteration (BRTO) in patients with gastric variceal bleeding. MATERIALS AND METHODS: Between March 2002 and June 2006, 75 patients with gastric variceal bleeding were enrolled in this study. In 19 patients, endoscopic therapy for acute variceal bleeding failed, and emergent BRTO within 24 hours after initial bleeding was performed to hemostasis. Ten of the 19 patients had massive bleeding with unstable vital signs. The remaining 9 patients had active bleeding with stable vital signs. In 56 patients, initial hemostasis was achieved with endoscopy or spontaneously, and elective BRTO was performed to prevent rebleeding. We analyzed cause of procedure-related mortality and stastistical difference in mortality between emergent and elective BRTO. RESULTS: Seven of the 75 patients (9.3%) died within 3 days after BRTO procedure. Causes of death were failed hemostasis in 5 patients, ARDS in 1 patient, and DIC in 1 patient. ARDS and DIC resulted from complication of ethanolamine oleate. Five patients with failed hemostasis had massive bleeding with unstable vital signs. In 4 of the 5 patients with failed hemostasis, BRTO was not technically feasible because contrast extravasation from varix to gastric lumen was noted during BRTO. In the remaining 1 patient, although BRTO was performed successfully, massive rebleeding occurred within several hours after BRTO. Mortality rate for emergent BRTO was 31.6%(6/19) compared with 1.8%(1/56) for elective BRTO (P ⫽ 0.001). Mortality rate for emergent BRTO in patients with unstable vital signs due to massive variceal bleeding was 60%(6/10).
POSTER SESSIONS
MATERIALS AND METHODS: 13 patients suffering from liver cirrhosis received a TIPS because of acute bleeding (n ⫽ 3) or refractory ascites (n ⫽ 10). 3/13 patients were female (23%), ten were male (77%). Mean age was 55.3 years. 8/13 patients had liver cirrhosis due to alcohol consumption, 3/13 patients from hepatitis. 2/13 patients from subacute Budd Chiari syndrome, one patient from cryptogenic liver cirrhosis. In 12/13 patients TIPS was created from the right liver vein to the right portal vein, in one patient from the left liver vein to the left portal vein. In 2 patients suffering from Budd Chiari, the right liver vein was recanalized prior to TIPS placement.
within the first 24 hours. The tract had to be recanalized and two additional stents were implanted. In the follow-up, no occlusion was observed anymore. Another patient treated for refractory ascites suffering from cirrhosis of his transplanted (!) organ developed a large liver hematoma (multiple punctures) during follow-up due to excessive anticoagulation with heparin. During follow-up the patient recovered well.
CONCLUSION: Emergent BRTO in patients with massive variceal bleeding has a high risk of mortality, especially in cases of extravasation from varix to gastric lumen during BRTO. Therefore an alternative therapeutic option is promptly needed in these cases not to lose the critical time for lifesaving. Abstract No. 330 EE Budd Chiari Syndrome (BCS) Interventional Radiologic Management. J. Tisnado, VCU/Medical College of Virginia, Richmond, VA, USA 䡠 M. Sydnor 䡠 D. Komorowski 䡠 D. Leung
S119
PURPOSE: BCS, i.e thrombotic occlusion of the hepatic veins and/or IVC, is an uncommon chronic entity of varied etiology. Its management has been, in the past, very difficult and ineffective. Medical therapy has been ineffective as well. During the last few years we have managed many patients with BCS with a combination of interventional radiologic procedures and prevented or postponed a necessary liver transplant. MATERIALS AND METHODS: We have been successful in maintaining patients under control and preventing an ultimate liver transplant (with its attendant morbidity and mortality and the scarcity of organs, live or cadaveric) in most of them. The IR procedures to manage BCS patients include: PTA of hepatic veins and /or IVC, thrombolytic therapy of thrombosed veins, placement of metallic stents (covered and uncovered) and TIPS placements and TIPS revisions as needed. TEACHING POINTS: We review the IR management ot this uncommon but serious and very difficult to manage entity. The radiologic management is preferred over the surgical management. This management prevents or postpones a major surgical intervention or a liver transplant. We present a review up to date of the problem.
Vascular Interventions: Uterine Fibroids Abstract No. 331 The Effect of Post-Procedure Pelvic MR Imaging in Women Who Have Undergone Uterine Artery Embolization. S. Rajeswaran, Northwestern University Medical School, Chicago, IL, USA 䡠 S. Dhand 䡠 H.B. Chrisman 䡠 A.A. Nemcek 䡠 R.L. Vogelzang 䡠 R.A. Omary, et al. PURPOSE: The utility of magnetic resonance imaging (MRI) in following up patients who have undergone uterine artery embolization (UAE) for leiomyomas is controversial. We aimed to determine how follow-up MRI affects interventional radiologists’ a) anticipated percentage of fibroid necrosis, b) projected treatment plans, and c) confidence in treatment plans for patients after UAE. MATERIALS AND METHODS: In a prospective study at a single institution, CAQ-certified attending interventional radiologists completed questionnaires (n ⫽ 20) before and after reviewing MRI scans on patients treated with UAE. Before knowing the results of the post-procedure MRI, these physicians were asked a) their anticipated percentage of fibroid necrosis (categorized as ⬍ 70%, 70-90%, or ⬎ 90%); b) their projected treatment plans (categorized as medical therapy, surgical therapy, repeat of embolization, six-month follow-up, or no therapy); and c) their confidence in this treatment plan (from 0% to 100%). These same questions were then answered following review of the MRI. We calculated the proportion of times that 1) the anticipated percentage of fibroid necrosis and 2) the projected treatment plans were changed by MRI. We also compared mean changes in confidence of treatment plans following MRI using the Wilcoxon test, with alpha ⫽ 0.05. RESULTS: 20 patients underwent follow-up MRI a mean of 5 months after UAE. Follow-up MRI scans changed the anticipated percentage of fibroid necrosis in 7 of 20 patients (35%). MRI also changed anticipated treatment plans in 9 of 20 patients (45%). Of these 9 patients that had their anticipated treatment plans changed by MRI, categorization S120
changed from no therapy to 6-month follow-up in 4 patients, 6-month follow-up to no therapy occurred in 4 patients, and repeat UAE to no therapy in 1 patient. MRI caused a 14% mean gain in confidence of treatment plans (p ⬍ 0.0001). CONCLUSION: Follow-up MRI significantly alters the a) anticipated percentage of fibroid necrosis, b) projected treatment plans, and c) confidence in treatment plans of interventional radiologists managing patients who have undergone UAE. Follow-up MRI should be considered for all patients after UAE. Abstract No. 332 Uterine Artery Embolization for Symptomatic Diffuse Leiomyomatosis of the Uterus. C. Scheurig, Universita¨tsmedizin Berlin, Berlin, Germany 䡠 T. Islam 䡠 E. Zimmermann 䡠 B. Hamm 䡠 T.J. Kroencke PURPOSE: To evaluate the clinical efficacy and magnetic resonance imaging (MRI) outcome after uterine artery embolization (UAE) in patients with the rare condition of diffuse uterine leiomyomatosis. MATERIALS AND METHODS: Six of 287 (2%) consecutively embolized patients presented with diffuse leiomyomatosis. Fibroid related symptoms (menorrhagia, pelvic pain, bulk related symptoms [brs]) were recorded before and up to 19 months after therapy using a standardized questionaire. Secondary outcome measurements included postembolization devascularization on contrast-enhanced MRI, uterine volume (UV) and morphologic changes as well as patient satisfaction. RESULTS: In 4/6 (67%) patients almost complete devascularization with loss of perfusion in 90-99% of the fibroid load was achieved. One/6 (17%) showed permanent complete devascularization. Another one/6 showed no relevant fibroid infarction after technically successful UAE and no clinical response. UV increased by 58% after one year in this case, which was classified as therapy failure. Menorrhagia and pelvic pain improved in all of the five remaining patients, brs improved in 3/4 (75%), 1/4 (25%) reported no change. One/5 did not have brs prior to therapy. UV decreased by a median of 33.2% (min.-max.:27.5-58.1; p ⫽ 0.043). MRI shows the thinned out myometrium before UAE recovering during follow-up time while the confluent fibroid load unmasks its nodules and shrinks. All five patients were either “very satisfied” or “satisfied” with the clinical outcome. CONCLUSION: Women suffering from the most extreme condition of uterine leiomyomatosis with almost complete replacement of myometrium by confluent nodules may be treated successfully with UAE. The myometrium impressively recovers while the fibroids decrease in volume. Abstract No. 333 Uterine Artery Embolization under Acupuncture on Uterine Fibroids. J. Pisco, Hospital Pulido Valente, Lisboa, Portugal 䡠 M. Teuchiya 䡠 A. Neves 䡠 M. Duarte 䡠 D. Santos PURPOSE: To evaluate if acupuncture maybe an alternative to the use of pharmacologic sedation / analgesia on uterine artery embolization (UAE) of uterine fibroids. MATERIALS AND METHODS: The study population included 12 white women who underwent UAE under acupuncture (36 – 48 years old). No pharmacologic sedation /
PURPOSE: The purpose of this study is to evaluate the rate of recurrence and effectiveness of treatment of various vascular malformations as measured by the need for follow-up treatment. MATERIALS AND METHODS: Between 1981 and 2003, 860 direct stick and transcatheter embolizations were performed at a single center. 583 vascular malformation cases with at least five year follow-up data were evaluated. There were 476 arteriovenous malformations and 80 venous malformations. The upper or lower extremities were involved in 209 cases. The remainder included 104 pelvic AVM, 62 renal AVM, and 35 pulmonary AVM. Need for follow-up procedures was determined based on clinical signs and symptoms. Calculations were performed to assess the percentage of AVM by location requiring, 1) at least one
RESULTS: A total of 121 cases required at least one additional follow-up procedure. Thirty-five percent (26 repeat procedures in 75 patients) and 46 percent (18 in 39 patients) of leg and foot AVM, respectively, required at least one follow-up procedure. Eleven percent (13 in 115 patients) of AVM in visceral organs, including lungs, GI tract, liver, bladder, spleen, and kidneys, needed at least one follow-up treatment. Twenty percent (8 in 39 patients) of lower extremity AVM, followed by 14 percent (15 in 104 patients) of pelvic AVM, required at least two or more follow-up procedures; less than one percent of pulmonary, GI, liver, bladder, splenic, or renal AVM required similar follow-up. AVM located in the lower extremity necessitated the greatest number of total follow-up procedures; 80 repeat embolizations were needed to resolve 114 AVM in the foot and leg. CONCLUSION: Vascular malformations have high rates of recurrence due to collateralization of vessels. AVM in visceral organs have the lowest rate of recurrence as measured by need for a minimum of one follow-up procedure. Lower extremity followed by pelvic AVM had the highest rates of recurrence. Complex AVM, as measured by cases requiring extensive treatment, tend to reside in the lower extremity and pelvis; in contrast, most visceral AVM resolved with one treatment. Abstract No. 318 Ear Arteriovenous Malformation Management. W.F. Yakes, Vascular Malformation Center, Englewood, CO, USA PURPOSE: To determine the efficacy of Ethanol Endovascular Repair of Ear Arteriovenous Malformation (AVMs). MATERIALS AND METHODS: Six patients (5 female, 1 male; age range 6-39 years; mean age: 22 years) with ear AVMs presented for therapy. Two patients had failed prior embolizations (PVA/coils/nBCA/steroids) and 2 patients had other therapies (laser/excisions/grafting). All presented with grossly enlarged painful ear with intermittent bleeding. All patients underwent transcatheter and direct puncture ethanol Rx (77 procedures). RESULTS: All 6 patients were cured of the AVM at longterm follow-up (mean follow-up: 39 months). One patient had transient partial VII nerve palsy. Two patients had minor blisters and ear injuries.
POSTER SESSIONS
Long-Term Follow-up of Rates of Recurrence with Arteriovenous Malformation Treatment. S. Rao, Lenox Hill Hospital, New York City, NY, USA 䡠 R. Rosen
follow-up procedure, 2) at least two or more follow-up procedures, and 3) highest total number of procedures.
CONCLUSION: Ethanol endovascular repair of Ear AVMs can effect cures in this vexing lesion that previously was treated with resection of the ear and high recurrence rates.
Vascular Interventions: Portal Hypertension Abstract No. 319 Balloon-Occluded Retrograde Transvenous Obliteration in 42 Patients with Gastric Varices or Hepatic Encephalopathy. M. Honda, Showa University Hospital, Tokyo, Tokyo-to, Japan 䡠 T. Hashimoto 䡠 T. Baba 䡠 K. Sekiyama 䡠 T. Kitanosono 䡠 T. Gokan
S115
PURPOSE: To evaluate the clinical efficacy of balloonoccluded retrograde transvenous obliteration (B-RTO) for the treatment of gastric varices or hepatic encephalopathy, and to describe technical problems and complications experienced during this procedure.
examination, Doppler ultrasonography and if necessary even portal venography. In all 4 groups the primary patency after the specific intervention was calculated using KaplanMeier analysis. These results of primary shunt patency were mutually compared using Cox’s F-test and log-rank test.
MATERIALS AND METHODS: From November 1994 through August 2006, we performed B-RTO in 42 patients (25 men and 17 women; range, 35-83 years; mean age, 63.2 years) with gastric varices (31 patients) or hepatic encephalopathy (11 patients). Gastric varices or hepatic encephalopathy were associated with portal hypertension as a result of viral liver cirrhosis in 25 patients, and alcoholic liver cirrhosis in 9 patients. Child-Pugh class was A in 11 patients, B in 25 patients, and C in 6 patients. 9 patients had hepatocellular carcinoma. Gastrorenal shunts were seen in 37 patients. Shunt tracts draining into the gonadal vein were seen in 5 patients. 47 procedures were performed for these patients. Transfemoral approach was used in 28 procedures, transjugular approach in 16 procedures, and both approaches were used in 3 procedures. After balloon-occluded retrograde venography, the sclerosant (ethanolamine oleate: EO) was injected into the gastric varices or shunt tracts during balloon occlusion. The maximum dose of EO at 5% was 20 mL/day. 3 patients underwent dual balloon-occluded embolotherapy using both methods of transjugular intrahepatic portosystemic shunt (TIPS) and B-RTO. After BRTO, CT and endoscopy were used to assess whether the varices or shunt tracts were thrombosed.
RESULTS: The intervention was successful in 120 cases (the overall technical success of all interventions was 99.2%). Primary shunt patency after the intervention in the group treated with plain angioplasty was 49.7% after 12 months and 25.3% after 24 months, in the group with deployment of bare stents it was 74.9% after 12 months and 64.9% after 24 months. In the non-dedicated ePTFE-covered stent-grafts group it was 75.2% after 12 months and 64.5% after 24 months and after deploying dedicated ePTFE-covered stent-grafts 88.1% were still patent at 12 months and 80.8% after 24 months after intervention. A statistically significant improvement in shunt patency was achieved in the group with implanted dedicated ePTFEcovered stent-grafts and in the group with bare stents in comparison with the group with plain angioplasty (p ⱕ 0.01).
RESULTS: Technical success was achieved in all patients. Follow-up CT after the procedure revealed thrombosed gastric varices or shunt tracts. All gastric varices were markedly decreased in size. During the follow-up period, one patient showed regrowth of gastric varices and the second B-RTO was performed. Hepatic encephalopathy was improved in 11 patients. Although extravasation of contrast media was seen in 2 patients, no significant sequela was observed in these patients. We have experienced no lifethreatening complications.
Abstract No. 321 EE
CONCLUSION: B-RTO is a safe and effective method of treatment for patients with gastric varices or hepatic encephalopathy due to portal hypertension. Abstract No. 320 Influence of the Secondary Deployment of ePTFE-Covered Stent-Grafts on the Maintenance of TIPS Patency. V. Jirkovsky, Teaching Hospital of Charles University in Hradec Kralove, Hradec Hralove, East Bohemia, Czech Republic 䡠 T. Fejfar 䡠 P. Hulek 䡠 A. Krajina 䡠 V. Safka 䡠 V. Chovanec, et al. PURPOSE: Evaluate the effects of secondary deployment of ePTFE-covered stent-grafts in the treatment of dysfunctional TIPS in comparison with other ordinary approaches (plain angioplasty or implantation of bare stents). MATERIALS AND METHODS: In our center, in the period from 2000 to 2004, a total of 121 cases of dysfunctional TIPS were treated with shunts that had initially been established using bare stents. The group was divided into 4 sub-groups based on the type of intervention: plain angioplasty (52 cases, 43%), deployment of a bare stent (35 cases, 28.9%), deployment of non-dedicated ePTFE-covered stentgraft (15 cases, 12.4%) and deployment of dedicated ePTFE-covered stent-graft (19 cases, 15.7%). Following the intervention all patients were monitored at regular intervals to observe the patency of the shunt with the aid of clinical S116
CONCLUSION: Amongst all widely used interventional methods for the treatment of dysfunctional TIPS, the best consequent patency was obtained by deploying dedicated ePTFE-covered stent-grafts.
Local Transplantation of Endothelial Progenitor Cells to Reduce Atherosclerotic Plaque Formation in a Rabbit Model. M. ZhanLong, Zhong-Da Hospital, Southeast University, Nanjing, Jiangsu, China 䡠 T. GaoJun 䡠 M. XiaoLi PURPOSE: To investigate whether catheter-based, local transplantation of homogeneous endothelial progenitor cells (EPCs) could prevent atherosclerotic plaque formation in a New Zealand rabbit model of atherosclerosis. MATERIALS AND METHODS: EPCs of New Zealand rabbits were isolated, confirmed, expanded and then incubated with home synthesized Fe2O3-PLL for 24 hours. Prussian blue stain was performed to identify intracellular iron. The carotid artery injured model in rabbits was induced by dilation with a 2.5-F balloon. Magnetically labeled-EPCs and fluorescently labeled-EPCs were transplanted into the injured endothelium in group A (n ⫽ 11) and group B (n ⫽ 5), respectively, while an equal volume of saline without EPCs was injected into the injured endothelium in group C (n ⫽ 12). MR imaging and, histopathology were performed at 7 days after transplantation for 2 rabbits randomly chosen from each group. All the remaining rabbits were kept to be fed with a high lipid diet until 15 weeks when MR examination was done. Histological examinations including HE staining were done immediately after the MR imaging. TEACHING POINTS: Iron-labeling efficiency within EPCs reached more than 95% identified by Prussian blue stain. At 7 days after EPCs transplantation, MR showed signal intensity loss of the injured carotid artery on T2*WI MRI only in group A, which was correlated with the areas where the most Prussian blue staining-positive cells were in histopathology, while histopathological slides showed the fluorescence-positive cells in the injured endothelium in group B. At 15 weeks after EPCs transplantation, less plaque or no plaque were exhibited in the rabbits of group A and B
receiving EPCs transplantation compared to that of the group C (P ⬍ 0.05). The catheter-based, local transplantation of ex vivo expanded homogeneous endothelial progenitor cells contribute to prevent atherosclerotic plaque formation in New Zealand rabbit model of atherosclerosis. Abstract No. 322 Foam Sclerotherapy for Gastric Varices and Splenorenal Shunt under CT or FACT. J. Koizumi, Tokai University, Isehara, Kanagawa Prefecture, Japan 䡠 Y. Kawawa 䡠 T. Hashimoto 䡠 K. Myojin 䡠 S. Kawada 䡠 Y. Imai PURPOSE: For gastroduodenal varices, percutaneous transhepatic obliteration (PTO), balloon occluded retrograde transvenous obliteration (BRTO) or dual balloon occlusion embolotherapy (dBOE) can be used. As one of the embolic materials, ethanolamine oleate (EO) is mainly used for BRTO. However, the overdosage of EO may cause severe complications such as hemolysis, allergy, ARDS, etc. Thus foam sclerotherapy was introduced to reduce the amount of EO to obliterate large varices under combined angiographyCT or flat detector angiographic CT (FACT) system. MATERIALS AND METHODS: For ten patients, seven BRTO, two PTO, and one dBOE were performed. Under balloon occlusion, a mixture of 20 mL of 5% EO with contrast media and 20 mL of air using a pumping method was injected into the target varices until full occupation was obtained. In the last three patients with gastric varices, 1 mL of polidocanol (Aethoxysklerol, Kaigen) mixed with 4 mL of air was used with or without EO. During the balloon occlusion CT or FACT was performed to confirm the filling of air mixed with the sclerosant into the target vessels. After thrombosis was achieved the catheter was retrieved and followed by CT and endoscopy. RESULTS: In all patients, air mixed with the sclerosant was observed in the target vessels during the procedures and full thrombosis was confirmed on postcontrast CT one week after the procedures. No procedure-related complications occurred. When performing BRTO, the light foam sclerosants tended to ascend directly into the target gastric varices which are located more ventral than the gastro-renal shunt. On the contrary, the foam sclerosants induced splenic venous thrombosis possibly because the splenic vein is more ventrally located than the target gastric varices during PTO.
Abstract No. 323 Direct Intrahepatic Portosystemic Shunts in Patients with Anatomic Contraindications to Traditional TIPS. J.C. Smith, Loma Linda University Medical Center, Loma Linda, CA, USA 䡠 G.E. Watkins 䡠 E.W. Palmer 䡠 M.H. Mendler 䡠 B. Petersen PURPOSE: An intravascular ultrasound (IVUS)-guided direct intrahepatic portosystemic shunt (DIPS) procedure has recently been described in patients with normal anatomy. We present a small series of 13 patients with various disease processes, in whom traditional transjugular intrahepatic por-
MATERIALS AND METHODS: From October 21, 2003, to September 8, 2006, 13 direct inferior vena cava-to-portal vein shunts were created in 13 patients using real-time guidance with a variable 5-10 MHz sagittal IVUS imaging system. Indications for DIPS included refractory ascites (n ⫽ 8), variceal bleeding (n ⫽ 4), and hepatohydrothorax (n ⫽ 1). Entities precluding routine TIPS placement included portal vein thrombus (n ⫽ 6), Budd Chiari syndrome (n ⫽ 1), polycystic liver disease (n ⫽ 1), hepatohydrothorax (n ⫽ 1), and previous failed TIPS attempts (n ⫽ 4). Shunts were created using 10-mm-diameter Viatorr stent grafts, but they were generally dilated to only 8 mm. Follow-up was performed clinically and using primarily transabdominal ultrasound imaging. RESULTS: All shunts were deployed successfully. One intended extrahepatic portal vein puncture led to clinically significant hemorrhage, which responded to blood transfusion. One patient developed acute liver failure leading to death at 2 days. No other immediate procedural related complication was identified. The mean portosystemic gradient was reduced from 26 mm Hg before to 8 mm Hg after DIPS placements. CONCLUSION: IVUS-guided DIPS placement can safely and effectively provide portal decompression in patients with venous occlusive disease or other anatomic abnormalities that may make traditional TIPS technically challenging or impossible to perform. Abstract No. 324 Endovascular Treatment of Portal Vein Abnormalities in Patients Undergoing Orthotopic Liver Transplantation. M. Marini, C. H. U. Juan Canalejo, A Corun˜a, Galicia, Spain 䡠 I. Cao 䡠 M. Gomez-Gutierrez 䡠 D. Gulias 䡠 J. Quintela 䡠 J. Aguirrezabalaga, et al. PURPOSE: To evaluate the efficacy of endovascular procedures for the treatment of acquired portal vein (PV) abnormalities from the time of transplantation to hepatic graft loss. MATERIALS AND METHODS: For the last 10 years, 25 patients with orthotopic liver transplants (OLT) and PV abnormalities have undergone endovascular procedures. Intraoperative period (n ⫽ 11): 2 patients with a re-graft had PV rupture during hepatectomy phase and underwent stent implantation in the interposition vein graft, and 9 patients with PV thrombosis (PVT) were treated by splanchnic vessel recanalization, primary stent and closure of competitive shunts during post-reperfusion phase. Perioperative period (n ⫽ 3): one PV rethrombosis and one PV torsion was treated by a stent, and a patient with anastomotic PV stenosis and stenosis in an interposition vein graft was treated by venoplasty. Early postoperative period (n ⫽ 4): 3 patients with anastomotic PV stenosis were treated by stent and one with stenosis in a jump graft was treated by venoplasty. Late postoperative period (n ⫽ 7): 4 patients with bleeding, PVT and chronic rejection were treated by PV recanalization, stent and a TIPS; one patient with chronic PVT and bleeding was treated by PV recanalization and stent; in the last 2 patients, with hypersplenism and ascites due to chronic rejection and PVT, one was treated by PV recanalization, stent and a TIPS. In the other, PV recanali-
POSTER SESSIONS
CONCLUSION: Foam sclerotherapy allows the reduction of the sclerosant theoretically to less than half dose. BRTO seems to be a better indication for foam sclerotherapy than PTO. Combined angiography- CT or FACT during the procedures was useful in confirming air mixed with the sclerosant in the target varicose veins protruding into the gastric lumen.
tosystemic shunt (TIPS) creation was considered difficult or impossible, who subsequently underwent successful DIPS placements.
S117
zation was unsuccessful and we performed a shunt of Leevin.
Abstract No. 326
RESULTS: Technical success was achieved in 24 of 25 patients (96%). None died during the procedure, allograft portal perfusion was adequate and portal hypertension solved. All venoplasties, stents and TIPS remained patent without further intervention, even in patients who died. One patient died within a week of TIPS of hepatic failure. At follow-up 7 patients died (19,39 mo 2-60) but it was not procedural related, 17 patients remain alive (40 mo 3-90): 2 underwent a re-graft and the others required no further intervention.
Clinical Efficacy of Embolization with NBCA for Esophageal-Gastric Fundal Varices in Portal Hypertension. K. Zhu, The Third Affiliated Hospital, Sun Yat-sen University, Guangzhou, Guangdong, China 䡠 H. Shan 䡠 Z. Li 䡠 Z. Jiang
CONCLUSION: Endovascular techniques in OLT are safe and effective and, in our opinion, the best choice in acquired PV abnormalities in OLT. Although stent primary patency in medium- and long-term follow-up was good, results must be confirmed. Abstract No. 325 Transjugular Intrahepatic Portosystemic Shunt: Evaluation of Impact of Stent Configuration on Patency Rate. P.J. Schaefer, University Hospital Schleswig-Holstein Campus Kiel, Kiel, Schleswig-Holstein, Germany 䡠 T. Jahnke 䡠 F.K.W. Schaefer 䡠 M. Heller 䡠 S. Mueller-Huelsbeck PURPOSE: To evaluate the impact of the configuration of the stent on the patency rate after transjugular intrahepatic portosystemic shunt creation with a self-expanding stent in patients with refractory ascites or variceal rebleeding caused by portal hypertension. MATERIALS AND METHODS: Totally, 80 patients (60 male, 20 female; mean-age 56 ⫾ 9.6, range 37-81) with transjugular intrahepatic portosystemic shunt were evaluated. 30 patients had refractory ascites, 32 variceal rebleeding, and 18 both. Primary technical success rate, interventional revision rate, and mean patency rate due to KaplanMeier were calculated. The angle as deviation of the bloodflow at the portal venous inflow and central venous outflow were measured on projected angiograms. The following five angle-groups were set up: 1) portal venous inflow, 2) central venous outflow, 3) maximum and 4) minimum angle, and 5) sum of both angles in the shunt system. Within each group, Mann-Whitney test and after dichotomic partition by the median Pearson’s Chi square test and Fisher’s exact test were carried out to prove for dependency of patency on the stent⬘s configuration. RESULTS: Primary technical success rate was 93%, interventional revision rate 28%, and mean patency rate 17.5 months. Mean/ standard deviation/ median angle were for the 1) portal venous inflow 66.5°/ 19.0°/ 65°, 2) central venous outflow 43.4°/ 14.0°/ 40°, 3) maximum angle 69.0°/ 16.1°/ 65°, 4) minimum angle 40.6°/ 13.3°/ 40°, and 5) sum of both angles 109.6°/ 21.9°/ 107.5°. The 2-sided values of significance were in the Mann-Whitney test/ Chi square test/ Fisher exact test for the 1) portal venous inflow 0.124/ 0.069/ 0.091, 2) central venous outflow 0.714/ 0.673/ 0.810, 3) maximum angle 0.178/ 0.046/ 0.056, 4) minimum angle 0.578/ 0.622/ 0.632, and 5) sum of both angles 0.157/ 0.228/ 0.335. CONCLUSION: The shunt’s patency rate with use of a self-expanding stent is not dependent on the stent⬘s configuration regarding the deviation of the blood-flow at the portal venous inflow and central venous outflow, and the maximum, minimum and total deviation in the shunt. S118
PURPOSE: To evaluate the efficacy of embolization with N-butyl-2-cyanoacrylate (NBCA) for esophageal-gastric varices in portal hypertension. MATERIALS AND METHODS: 28 cirrhotic patients with esophageal-gastric varices were enrolled in this study. Before this study, 15 patients received endoscopic variceal ligation, 13 received conservative pharmacotherapy, and all the patients suffered from rebleeding. Percutaneous transhepatic or transplenic portography was performed in all 28 patients; then, esophagogastric varices were embolized with NBCA. The proportion of NBCA and lipiodol was determined according to the size and blood flow of varices, ranging from 1:4 to 1:8. For the patients who had a large gastrorenal shunt (GRS), a catheter with a balloon was introduced into the GRS via the right femoral and left renal vein before the embolization, and the balloon was inflated to block the flow of GRS. Portal venous pressure was measured before and after embolization in all patients. RESULTS: Of the 28 patients, we successfully embolized the left gastric vein, posterior gastric vein and short gastric vein in 27 cases; balloon-occluded technique was applied in 8 patients with large GRS. Portal venous pressure significantly raised from mean, 34.7 cm H2O before embolization to mean, 38.7 cm H2O (P ⬍ 0.05) after embolization. In four patients, little NBCA entered into distal pulmonary artery branches. Two of them suffered transient irritable cough; no patient developed severe pulmonary embolism. Endoscopic examination was performed in 11 patients after embolization, which demonstrated that the varices of all 11 patients were markedly alleviated. In the 14 patients who underwent CT 3 to 6 months after treatment, the embolized varices were still obstructed in all of them, but some other collateral veins were observed in three patients. In this study, the patients were followed up from 3 to 26 months (mean, 9.9 months). Rebleeding was observed in 5 patients (17.9%). CONCLUSION: Embolization with NBCA is a safe and effective method for the treatment of esophagogastric varices; but it is necessary to block the large GRS via retrograde approach to prevent pulmonary embolism. Abstract No. 327 EE Measuring the Hepatic Vein Pressure Gradient: Review of Technique with Emphasis on Accuracy and Standardization. J.M. LaBerge, University of California, San Francisco, San Francisco, CA, USA 䡠 R.K. Kerlan PURPOSE: The hepatic vein pressure gradient (HVPG) is an indirect reflection of portal pressure that has received renewed interest with the recent observation that, in patients with compensated cirrhosis, HVPG is the best predictor of the development of varices and that a reduction in HVPG in response to pharmacologic therapy correlates with bleeding risk. The technique of obtaining HVPG while not new has not been standardized. Recently, a NIH consensus panel (June 2006) has called for standardization of HVPG technique in anticipation of an expanded role of HVPG mea-
surement in the treatment of portal hypertension. The purpose of this teaching demonstration is to describe current HVPG measurement technique and identify steps that may improve accuracy. In addition, recent literature citing the value of HVPG measurement in monitoring portal hypertension will be reviewed. MATERIALS AND METHODS: We describe and illustrate the steps necessary to perform HVPG measurement with emphasis on the following: - Selection of the hepatic vein and proper location for measurement - Use of end hole versus balloon catheter - Number of repetitive measurements necessary to insure reproducibility - Calibration and use of pressure monitor equipment for accuracy - The value of waveform analysis and use of printed tracing versus digital readouts. Recent work by Grozsmann RJ, et al (NEJM 2005; 353; 2254-61) has demonstrated the relationship between HVPG and clinical endpoints in patients treated with pharmacologic agents. Results of this paper and other studies demonstrating the clinical utility of HVPG measurement will be reviewed. HVPG is currently the only method for accurately monitoring portal pressure and tailoring the delivery of pharmacologic therapy to the needs of an individual patient. TEACHING POINTS: Measurement of HVPG may play an increasingly important role in monitoring patients with portal hypertension. Standardized techniques that yield accurate and reproducible measurements are necessary for the widespread acceptance of this procedure. These include use of balloon occlusion catheter, electronic pressure transducer calibrated for venous readings, and pressure tracings with readings averaged over three stable values. Abstract No. 328 Using the Self-expanding Sinus Superflex Visual Stent System for TIPS Creation: Initial Experiences. J. Wiskirchen, University Hospital Tuebingen, Tuebingen, Baden-Wuerttemberg, Germany 䡠 A. Fischmann 䡠 G. Tepe 䡠 U. Lauer 䡠 W. Steurer 䡠 P.L. Pereira PURPOSE: Aim of this work is to analyze the experiences with a 6-F self-expanding nitinol stent system (Sinus Superflex Visual, Optimed, Germany) in creating transjugular portosystemic shunts (TIPS) in patients suffering from acute bleeding or refractory ascites.
RESULTS: TIPS creation was successful in all patients. In 8 patients a single stent with a diameter of 12 mm and a length of 60 mm was used, in five patients stent length was 80 mm. All stents were deployed without any problem, the technical success rate was 100%. Mean pressure gradient before all interventions was 22.8 mm Hg (range 15 – 32), after the intervention the gradient dropped to 8.5 mm Hg (range 5 – 11). One patient who was treated because of subacute Budd Chiari syndrome suffered from an acute stent occlusion
CONCLUSION: The self-expanding Sinus Superflex Visual Stent can be used for TIPS creation successfully. Due to its superb visibility and simple release system, stent placement was easy in all cases. However, long term follow-up has to be performed to evaluate the long term performance of the stent. Abstract No. 329 Analysis of Procedure-Related Mortality of BRTO in Patients with Gastric Variceal Bleeding. Y. Kim, Keimyung University School of Medicine, Joonggu, Daegu, Korea 䡠 C. Kim 䡠 J. Choi 䡠 T. Shin PURPOSE: To analyze procedure-related mortality of balloon occluded retrograde transvenous obliteration (BRTO) in patients with gastric variceal bleeding. MATERIALS AND METHODS: Between March 2002 and June 2006, 75 patients with gastric variceal bleeding were enrolled in this study. In 19 patients, endoscopic therapy for acute variceal bleeding failed, and emergent BRTO within 24 hours after initial bleeding was performed to hemostasis. Ten of the 19 patients had massive bleeding with unstable vital signs. The remaining 9 patients had active bleeding with stable vital signs. In 56 patients, initial hemostasis was achieved with endoscopy or spontaneously, and elective BRTO was performed to prevent rebleeding. We analyzed cause of procedure-related mortality and stastistical difference in mortality between emergent and elective BRTO. RESULTS: Seven of the 75 patients (9.3%) died within 3 days after BRTO procedure. Causes of death were failed hemostasis in 5 patients, ARDS in 1 patient, and DIC in 1 patient. ARDS and DIC resulted from complication of ethanolamine oleate. Five patients with failed hemostasis had massive bleeding with unstable vital signs. In 4 of the 5 patients with failed hemostasis, BRTO was not technically feasible because contrast extravasation from varix to gastric lumen was noted during BRTO. In the remaining 1 patient, although BRTO was performed successfully, massive rebleeding occurred within several hours after BRTO. Mortality rate for emergent BRTO was 31.6%(6/19) compared with 1.8%(1/56) for elective BRTO (P ⫽ 0.001). Mortality rate for emergent BRTO in patients with unstable vital signs due to massive variceal bleeding was 60%(6/10).
POSTER SESSIONS
MATERIALS AND METHODS: 13 patients suffering from liver cirrhosis received a TIPS because of acute bleeding (n ⫽ 3) or refractory ascites (n ⫽ 10). 3/13 patients were female (23%), ten were male (77%). Mean age was 55.3 years. 8/13 patients had liver cirrhosis due to alcohol consumption, 3/13 patients from hepatitis. 2/13 patients from subacute Budd Chiari syndrome, one patient from cryptogenic liver cirrhosis. In 12/13 patients TIPS was created from the right liver vein to the right portal vein, in one patient from the left liver vein to the left portal vein. In 2 patients suffering from Budd Chiari, the right liver vein was recanalized prior to TIPS placement.
within the first 24 hours. The tract had to be recanalized and two additional stents were implanted. In the follow-up, no occlusion was observed anymore. Another patient treated for refractory ascites suffering from cirrhosis of his transplanted (!) organ developed a large liver hematoma (multiple punctures) during follow-up due to excessive anticoagulation with heparin. During follow-up the patient recovered well.
CONCLUSION: Emergent BRTO in patients with massive variceal bleeding has a high risk of mortality, especially in cases of extravasation from varix to gastric lumen during BRTO. Therefore an alternative therapeutic option is promptly needed in these cases not to lose the critical time for lifesaving. Abstract No. 330 EE Budd Chiari Syndrome (BCS) Interventional Radiologic Management. J. Tisnado, VCU/Medical College of Virginia, Richmond, VA, USA 䡠 M. Sydnor 䡠 D. Komorowski 䡠 D. Leung
S119
PURPOSE: BCS, i.e thrombotic occlusion of the hepatic veins and/or IVC, is an uncommon chronic entity of varied etiology. Its management has been, in the past, very difficult and ineffective. Medical therapy has been ineffective as well. During the last few years we have managed many patients with BCS with a combination of interventional radiologic procedures and prevented or postponed a necessary liver transplant. MATERIALS AND METHODS: We have been successful in maintaining patients under control and preventing an ultimate liver transplant (with its attendant morbidity and mortality and the scarcity of organs, live or cadaveric) in most of them. The IR procedures to manage BCS patients include: PTA of hepatic veins and /or IVC, thrombolytic therapy of thrombosed veins, placement of metallic stents (covered and uncovered) and TIPS placements and TIPS revisions as needed. TEACHING POINTS: We review the IR management ot this uncommon but serious and very difficult to manage entity. The radiologic management is preferred over the surgical management. This management prevents or postpones a major surgical intervention or a liver transplant. We present a review up to date of the problem.
Vascular Interventions: Uterine Fibroids Abstract No. 331 The Effect of Post-Procedure Pelvic MR Imaging in Women Who Have Undergone Uterine Artery Embolization. S. Rajeswaran, Northwestern University Medical School, Chicago, IL, USA 䡠 S. Dhand 䡠 H.B. Chrisman 䡠 A.A. Nemcek 䡠 R.L. Vogelzang 䡠 R.A. Omary, et al. PURPOSE: The utility of magnetic resonance imaging (MRI) in following up patients who have undergone uterine artery embolization (UAE) for leiomyomas is controversial. We aimed to determine how follow-up MRI affects interventional radiologists’ a) anticipated percentage of fibroid necrosis, b) projected treatment plans, and c) confidence in treatment plans for patients after UAE. MATERIALS AND METHODS: In a prospective study at a single institution, CAQ-certified attending interventional radiologists completed questionnaires (n ⫽ 20) before and after reviewing MRI scans on patients treated with UAE. Before knowing the results of the post-procedure MRI, these physicians were asked a) their anticipated percentage of fibroid necrosis (categorized as ⬍ 70%, 70-90%, or ⬎ 90%); b) their projected treatment plans (categorized as medical therapy, surgical therapy, repeat of embolization, six-month follow-up, or no therapy); and c) their confidence in this treatment plan (from 0% to 100%). These same questions were then answered following review of the MRI. We calculated the proportion of times that 1) the anticipated percentage of fibroid necrosis and 2) the projected treatment plans were changed by MRI. We also compared mean changes in confidence of treatment plans following MRI using the Wilcoxon test, with alpha ⫽ 0.05. RESULTS: 20 patients underwent follow-up MRI a mean of 5 months after UAE. Follow-up MRI scans changed the anticipated percentage of fibroid necrosis in 7 of 20 patients (35%). MRI also changed anticipated treatment plans in 9 of 20 patients (45%). Of these 9 patients that had their anticipated treatment plans changed by MRI, categorization S120
changed from no therapy to 6-month follow-up in 4 patients, 6-month follow-up to no therapy occurred in 4 patients, and repeat UAE to no therapy in 1 patient. MRI caused a 14% mean gain in confidence of treatment plans (p ⬍ 0.0001). CONCLUSION: Follow-up MRI significantly alters the a) anticipated percentage of fibroid necrosis, b) projected treatment plans, and c) confidence in treatment plans of interventional radiologists managing patients who have undergone UAE. Follow-up MRI should be considered for all patients after UAE. Abstract No. 332 Uterine Artery Embolization for Symptomatic Diffuse Leiomyomatosis of the Uterus. C. Scheurig, Universita¨tsmedizin Berlin, Berlin, Germany 䡠 T. Islam 䡠 E. Zimmermann 䡠 B. Hamm 䡠 T.J. Kroencke PURPOSE: To evaluate the clinical efficacy and magnetic resonance imaging (MRI) outcome after uterine artery embolization (UAE) in patients with the rare condition of diffuse uterine leiomyomatosis. MATERIALS AND METHODS: Six of 287 (2%) consecutively embolized patients presented with diffuse leiomyomatosis. Fibroid related symptoms (menorrhagia, pelvic pain, bulk related symptoms [brs]) were recorded before and up to 19 months after therapy using a standardized questionaire. Secondary outcome measurements included postembolization devascularization on contrast-enhanced MRI, uterine volume (UV) and morphologic changes as well as patient satisfaction. RESULTS: In 4/6 (67%) patients almost complete devascularization with loss of perfusion in 90-99% of the fibroid load was achieved. One/6 (17%) showed permanent complete devascularization. Another one/6 showed no relevant fibroid infarction after technically successful UAE and no clinical response. UV increased by 58% after one year in this case, which was classified as therapy failure. Menorrhagia and pelvic pain improved in all of the five remaining patients, brs improved in 3/4 (75%), 1/4 (25%) reported no change. One/5 did not have brs prior to therapy. UV decreased by a median of 33.2% (min.-max.:27.5-58.1; p ⫽ 0.043). MRI shows the thinned out myometrium before UAE recovering during follow-up time while the confluent fibroid load unmasks its nodules and shrinks. All five patients were either “very satisfied” or “satisfied” with the clinical outcome. CONCLUSION: Women suffering from the most extreme condition of uterine leiomyomatosis with almost complete replacement of myometrium by confluent nodules may be treated successfully with UAE. The myometrium impressively recovers while the fibroids decrease in volume. Abstract No. 333 Uterine Artery Embolization under Acupuncture on Uterine Fibroids. J. Pisco, Hospital Pulido Valente, Lisboa, Portugal 䡠 M. Teuchiya 䡠 A. Neves 䡠 M. Duarte 䡠 D. Santos PURPOSE: To evaluate if acupuncture maybe an alternative to the use of pharmacologic sedation / analgesia on uterine artery embolization (UAE) of uterine fibroids. MATERIALS AND METHODS: The study population included 12 white women who underwent UAE under acupuncture (36 – 48 years old). No pharmacologic sedation /