Scientific Session 1 Hemodialysis Interventions

Scientific Session Objectives As a ,"esult ofattending the sCientific sessions, participants will be able to:

1. Evaluate currenl research in inlervenlional radiology. 2. Identify current and future technical and clinical developments. 3. Modify their academic and clinical practices. 4. Identify new research methods and areas in need of investigation.

Scientific Session 1 Hemodialysis Interventions

12:41 pm

Abstract No.2

Microbubble Potentiated Ultrasound as a Method of Decloning Thrombosed Dialysis Grafts. W.C CUIp, U ofNebraska Med Centel; Omaha, NE, USA • TR. Porter. F. Xie • TC Goertzen • TC McCowan • B.N. Vonk, et at. PURPOSE: Intravenous perfluorocarbon containing microbubbles (PESDA) have been shown to recanalize acutely thrombosed small arteries in the presence of low frequency ultrasound (LFUS). We sought to develop a microbubble-potentiated ultrasound method for declOlting full-size arteriovenous dialysis grafts without creating a systemic lytic effect.

Moderator: Richard Gray, MD

Sunday, March 4,2001 12:30 pm-2:00 pm 12:30 pm

CONCLUSION Thrombolysis of clotted hemodialysis grafts using 1 unit reteplase and 4000 units heparin is safe and effective.

Abstract No. 1

Thrombolysis of Cloned Hemodialysis Grafts with Reteplase. A. Falk, Mt. Sinai Medical Center, New York, NY, USA • j. Guller • S. Nowakowski • V. Teodorescu • j. Uribam • j. Vassalotti, et al. PURPOSE: To prospectively evaluate the efficacy of

reteplase in treating thrombosed hemodialysis arteriovenous PTFE grafts. MATERIALS AND MElHODS: 33 thrombosed PTFE dialysis grafts in 24 patients (7 male, 17 female, ages 20-83, means 55.0 years) presented for declotting. The first 6 grafts were treated with 0.5 units reteplase and 3000 units heparin. The next 27 grafts were treated with 1 unit reteplase and 4000 units heparin. This was injected into the graft through a 4 French coaxial dilatOr in the angiography suite after access imo the graft was confirmed with passage of an 018 wire. Prospective data collection included demographiC information, technical details of each procedure, the elapsed time from injection of reteplase until routine PTA and embolectomy, lOtal roOm procedure times, immediate outcomes, complications and patency rates. RESULTS: Technical success, defined as complete graft

recanalization with a palpable or auscultated thrill, plus one successful hemodialysis, was acheived in 29/33 grafts. Four procedure related complications occured: (l)bleeding from old puncture sites successfully treated with manual compression, (2) embolization of thrombus into the brachial and radial arteries treated with embolectomy and overnight heparinization of the patient, and (3) rupture of the brachial outflow vein just above the elbow in 2 forearm loop grafts. Mean lysis time was 8 minutes and mean room procedure time (including lysis time) was 56 minutes. The 30 day primary patency rate was 55%. The follow up period was 13-191 days. No patients were lost to follow up. Two patients died during the study period.

MATERIALS AND METHODS: Six femoral dialysis grafts in three dogs were cannulated for test fluid injection and angiography. After thrombosis by ligation three LFUS (I megahertz, 0.6 W/cm 2) declotting techniques were randomized: (1) direct injection of PESDA; (2) direct injection of saline (contro!); and (3) intravenous PESDA. Each test fluid was given in 0.5 to 1.0 ml doses for a total oJ' 4.5 ml in 30 minutes. Patency was assessed by co or ultrasound and cine angiography. Declotting was .raded in each third of the graft on a scale of 0 to 4 (maximum lOtal score =12) with success defined as good flow and > 70% clearing of thrombus in 30 minutes. RESULTS: 25 primary declotting procedures showed

mean patency scores of 11.1 for direct PESDA, 8.3 for IV PESDA, and 4.8 for direct saline. All are statistically different with direct PESDA vsIV PESDA P = 0.0009, direct PESDA vs saline p = 0.0001, and IV PESDA vs saline p = 0.0002. All 8 direct PESDA injections achieved lysis and good flow. None of 8 direct saline injections succeeded (p = <0.01). Intravenous PESDA succeeded in 3 of 9 (p = <0.01 vs direct PESDA, p = 0.21 vs saline). No systemic or local complications related to the PESDA or ultrasound were encountered. CONCLUSION Direct injection of PESDA with transcutaneous LF S is a successful method of lysing large clots and recanalizing thrombosed fisntlas without systemic effects.

12:52 pm

Abstract No.3

Complications Related to Percutaneous Thrombectomy of Hemodialysis Grafts. T Vesely, Mallinckrodtlnstitute of Radiology, St. Louis, MO, USA PURPOSE: To retrospectively review the complications reported during percutaneous thrombectomy of hemodialysis grafts.

81

MATERiALS AND METHODS: A computer search of our

interventional radiology database revealed that 803 percutaneous thrombectomy procedures were performed on hemodialysis grafts between 1/1/93 and 9/1/00. The type and number of procedures include: Arrow PTD (412), pulse-spray with urokinase (240), Amplatz Thrombectomy Device (79), AngioJet (17), Oasis (15), HydrolyzeI' (0), Endovac (7), Lyse and Wait (7), Thrombex (6), Cragg brush (6), Castaneda brush (4). Complications were reported to have occurred in 30 patients. The radiology reports and medical records of these patients were reviewed. RESULTS: The overall complication rate was 3.7% (30/803). The type and number of complications included: rupture of a vein during angioplasty (13), severe respiratory arrest (4), arterial emboli (3), rigors related to urokinase (3), access site hematoma (2), moderate hypoxia (2), assorted other complications (3). There was one death resulting from a fall from the angiography table immediately following the procedure. There were 12 minor complications, requiring minimal treatment, and 18 major complications that altered the course of the procedure or treatment of the patient. Ten of these major complications were due to angioplastyinduced rupture which led to abandonment of the procedure. Fourteen complications were directly related to the thrombectomy procedure; 5 minor complications related to urokinase, 4 patients had acute respiratory arrest immediately following dislodgement of the arterial J' plug, 3 patients had arterial emboli (1 major, 2 minor), and 2 patients had chest pain and hypoxia (1 major, 1 minor) during the thrombectomy procedure. CONCLUSION: The most common complication was angioplasty-induced rupture of the vein or graft. The most severe complications occurred immediately follOWing dislodgement of the arterial plug and were likely due to acute pulmonary embolization.

1:03 pm

Abstract No.4

The Xpeedior Thrombectomy System: Early Results in 36 Cases. 1. Schur, St. Lukes - Roosevelt Hospital Center; New Yor"k,

NY, USA • E. Koh • CM. Tuite • M.O. Gudleski • M. T Grauer • ].F Santoro

PURPOSE: To evaluate our initial experience with a new

6 French over-the-wire (0.035) thrombectomy device in the treatment of thrombosed hemodialysis grafts. MATERIALS AND METHODS: Between May and Septem-

S2

ber 2000, 36 thrombosed PTFE grafts (19 loop and 7 straight) in 26 patients (16 females, 10 males, mean age 60.2 years) were percutaneously treated with the Xpeedior rheolytic catheter. FollOWing throinbectomy, PTA of the venous anastomosis and balloon mobilization of the arterial plug were performed in all cases. Three patients (8.3%) had coexistent acute thrombus in the outflow vein for 6-12cm for which the Xpeedior was also utilized. Four self-expandable stents were required:

2 at the venous anastomosis for failed PTA and 2 for a chronic occlusion of the subclavian and brachiocephalic vein. Five cases had occlusion or severe immediate outflow venous stenosis for ~8cm which were treated with long segment PTA. 5 patients had preexisting self-expandable stents, 3 from prior chronic central occlusions, and 2 from failed PTA at the venous anastomosis. Technical success, complication, and 30-day patency rates were tabulated. RESULTS: Complete thrombus removal was achieved in 35/36 cases (97.2%). Technical success, defined by restoration of a thrill with at least one adequate hemodialysis session within 24 hours, was 94.4%. In the three patients where acute thrombus also involved a long segment of the outflow vein, 1000!o clot clearance was observed with the device alone. Procedure times ranged between 38-90 minutes. Three complications(8,3%), not attributed to the device, were encountered and were all successfully treated in the angio suite. These included rapid extravasation from a prior dialysis puncture site; a small dissection of the brachial artery from passage of an adherent clot catheter and a large PTA-related perforation at the venous anastomosis. Primary and secondary patencies at 30 days were 67.6% and 87.5%, respectively. CONCLUSION: The Xpeedior thrombectomy device is very safe and efficient and has a high success rate with acceptable short-term primary and secondary patency rates.

1:14 pm

Abstract No.5

Pharmacomechanical Thrombolysis of Dialysis Grafts Utilizing the Wall Contact Thrombolytic Brush Catheter and the Newer Throm.bolytic

Agent Reteplase. F Castaneda, OSF Saint Francis Medical Centel;

Peoria, IL, USA·].L. Swischuk • TM. Brady • B. Smouse· ].D. Vrabel • K. Young PURPOSE: To determine the efficacy and safety of recanalization of thrombosed dialysis grafts with the MTI brush catheter and Reteplase. METHODOLOGY: After IRB approval prospective data

collection has been obtained in 74 patients with thrombosed PTFE dialysis grafts. Once patent and suitable venous outflow was confirmed, the MTJ-Castaneda brush was used, while infusing a mixture of 50/50 of contrastiNS to which a lU/lcc aliquot of Reteplase was added. Thrombolysis time, Reteplase dose, immediate patency rates, successful next dialysis rates, and complications were all recorded. RESULTS: The irrunediate and neX1 successful dialysis rates were 89% and 84% respectively. The mean thrombolysis time was 3.11min. The mean Reteplase dose infused was 1.28U/graft. Associated interventions included PTA is of venous anastamosis 85%; intragraft

77%, arterial anastamoses 40010, central vein 19% and stents were used in 13%. CONCLUSION: The combination of the MTI-Castaneda thrombolytic bnJsh catheter and Reteplase is a fast, effective and safe method of recanalizing thrombosed PTFE dialysis grafts.

1:25 pm

Abstract No.6

Hemodynamic Assessment of Dialysis Grafts: A Simple Technique for Arterial Pressure Measurement. R. Duszak, The Reading Hospital and Medical Center, Reading, PA, USA • D. Sacks PURPOSE: To evaluate a simple technique for measurement of dialysis graft hemodynamics during graft angiography. Contralateral blood pressure cuff measurements are often used to normalize direct venous limb graft pressures when evaluating stenoses, but arterial pressures can also be easily obtained by measuring direct graft pressures during graft outflow occlusion. The two methods are compared. MATERlALS AND MEmODS: Hemodynamic assessment of dialysis grafts was performed in 88 cases (49 patients) referred for graft angiography. Evaluation consisted of 1) pre-procedural measurement of bilateral arm cuff pressures and 2) simultaneous intraprocedural mea urement of contralateral cuff pressure, direct intragraft arterial pressure, and direct venous limb graft pressure. The direct arterial pressure measurements were obtained through the same catheter used for venous limb pressure evaluation, using firm manual graft compression to occlude outflow.

RESULTS: Systolic cuff pressures in each arm were on average similar (p = 0.70) but differed in one direction or anodler by >15 mm Hg in 18 patients (37%), suggesting a large minority widl arterial inflow disease. Direct intra-graft evaluation of arterial pressures yielded systolic measurements 20 ± 16 mm Hg higher than cuff measurements (p < 0.(01) that were frequendy not explained by cuff arm pressure differentials. Using direct intra-graft pressures instead of cuff pressures, venous limb ratios (as described by Sullivan, el al) are reduced by a mean 0.04 (p < 0.00l). CONCLUSION Blood pressure cuff measurements frequently differ substantially between arms in hemodialysis graft patients. Accordingly, conu'alateral cuff measurements used to normalize venous limb pressures often may not reflect true inflow pressures. Ipsilateral arterial pressures can be easily obtained within a graft using outflow compression. This technique, however, frequently results in higher systolic measurements, which should be accounted for when comparing ratios with published parameters for hemodynamic significance.

1:36 pm

Abstract No.7

Patency of Stems in the Venous Outflow of Dialysis Grafts: Mid term Results. Gj. O'Sullivan, Rush-Presbyterian-St. Luke's Medical Center, Chicago, IL, USA • PM. Valen • j.A. Levine • AN Makris • P. Chopr'a PURPOSE: Stents may be used as a means of maintaining patency in the venous outflow of hemodialysis grafts. The optimum follow-up in these patients is uncertain. We present our experience with a variety of stents used mainly as salvage for failing or thrombosed aneriovenous hemodialysis grafts. MATERIALS AND METHODS: Prospective analysis of pa-

tients presenting with a failing or thrombosed arteriovenous graft (AVG). A total of 93 stents were inserted in 67 patients over a 12 month period. Indications for insertion included salvage, venous extravasation, and rupture. The primary, primary-assisted, and secondary patency rates for the venous outflow were analysed. Comparison was made between patients presenting with an "open" AVG as opposed to a thrombosed AVG. RESULTS: For non-thrombosed AVG's the primary and primary-assisted patency rates at 30, 90, 180, and 365 days were 87% (92%), 78%) (84%), 470/0 <73%) and 35% (67%). For thrombosed AVG's initially treated by pharmaca-mechanical thrombolysis. followed by steming at the same sitting, the primary and primary-assisted patency rates at 30, 90, 180 and 365 days were 84% (88%), 28% (73%), 12% (62010) and 3% (53%). The critical time period for stent thrombosis in the latter cohort appears to be around 60 days. There was no statistically significant difference between patencies of the different types of self-expanding stents. CONCLUSION Stenting is a rea onable method of prolonging AVG life in patients with failing hemodialysis grafts. There is a clinically and statistically significant difference between the 90 day patency of stems inserted in open versus thrombosed AVG's at the time of actual stent placement. This has impoltant implications in terms of the optimal follow-up for these patients, as well as issues of re-imbursement, as grafts that re-thrombose within a 90 day time period are not covered within the recently revised G0159 CPT-4 code. As a result of this study we have decreased dle time between initial stent insertion and routine follow-up for patients with thrombosed grafts who received stents. Results of dlis prospective study are ongoing.

1:47 pm

Abstract

o. 8

FEATURED ABSTRACT Commentator: Anne C. Roberts, MD Ly e and Wait Compared to Mechanical Thrombolysis of Hemodialysis Grafts. PM. Vog -/, utter fle(illth, . Cicramelllo, Ctl, USA PUl?PO 'E: To determine if the 4mg dose of Alteplase (TP)j elre live for the Jy-e and ait method of dial is acc(' thromboly is.

S3

MATERIALS AND METHODS: Forty patients with thrombosed -v grafts were randomized prospe tively to L W tI in' 4 mg of TPA or mechanical thromboly i with the row-Tr rotola dev~ c (PTO . Both r tip received 5.000 uni of heparin. Th tim interval to re tored graft flow, total procedure time, hemostasis time. procedure .tlce SS, complications, and patenci . were analyz d. Th ewer then ompared with historical reo ult in 20 patien! treated with th L&W technique using 250,000 unit of urokinase. RESUL7 : Immediate clinical ucce wa l()()Oitl in I th group of 20 TPA and PTD patient . Mean thrombolysis time with re tore<1 flow was 10 minur for TPA L&W and 19 m.i.l1ute PTD. 'l11e mean procedure time' weI' 39 minutes for L&W, 45 minutes for PTD. Mean time to hemostasLs using manual compres ion was 44 minute for TPA L&W, 23 mintlt for PTD (p=.052). The one and thrt:e month primary graft patency wa 5% and 56%, respectively, for 1]> L&W. PTO pati nts had 7fdYo one month od 63% three month primary patency. The historical group of 20 lJK L&W patients had 95% clinical ucces, mean 14 minute I. i time, mean 34 minute pI' edure tim, nd 26 minute time to hemorasis. On and three month primary patendes were 75% and 50%, respecliv Iy. TI1 re w re no complica:' lions in the ?TO group. Seven '!PA patients had bleeding (p=.OO'!'). Thre of til e had minor bleeding from prior puncrure ite dUring the procedure. Four pati nts had d layed bl ding 1-2 hours po t procedure re ulting in one hospitalization, one suture closure, and one graft thrombo-i from ompre. ion. None of these patients took oral anticoagulant . Four of th _0 UK !..<..'?,::W patients had minor puncture. ite bleeding during the procedure, but no delay I hleeding. CONCLu.. JON: The 4mg dose of TPA for 1&'1 technique i effe rive, but results in more bleeding complithan mechanical cation and longer hemo ta i tin thrombolysis. Unlike our experi nc' with urokina e, bleeding c mpH ations with TP. may be delayed.

Scientific Session 2 Oncologic Intervention: Hepatic Chemoembolization

Sunday, March 4, 2001 12:30 pm-2:00 pm Abstract No, 9

Transcatheter Arterial Chemoembolization for Hepatocellular Carcinoma: Effects on Underlying Liver Function. j. Geschwind, 1heJohns Hopkins Hospital, Baltimore, MD, USA • K.Juluru • A. Arepally • G. Lund • M. Choti • P. 1huluvath

84

transcatheter

arterial

MATERIALS AND METHODS: l"ACE was performed in 55 patients with unresectable hepatocellular carcinoma. Mean tumor size was 8.8 em. Patients were treated with a mean of 2.7 TACE procedures, performed 5 weeks apart. TACE was performed in a selective or superselective manner within th eright or left hepatic artery and consisted of a mixture of chemotherapeutic agents (cisplatin, doxorubicin and mitomycin C) and lipiodol followed by polyvinyl alcohol particles. Liver function tests, physical examination and tumor markers were obtained at baseline, 4 weeks and 6 months after TACE. Child-Pugh score was calculated at each time point. The data were acquired both retrospectively and prospectively. Prior to TACE, there were 33 patients with Child A (mean score=5.0]), 22 patients with Child B (mean score=7.06) and none with Child C. RESULTS: Mean Child-Pugh score increased in all patients from 5.82 to 6.02 after first TACE and to 6.11 after second TACE (not significant). Scores did not significantly change after the third TACE or at 6 month follOW-Up. No patient died of liver failure. Mean scores also increased for each class of patients but were only significant (p<0.05) in Child A patients (5.01 to 5.51 and 5.91, after first and second TACE). Decrease in tumor markers and tumor viability were recorded in all patients.

CONCLUSION TACE did not cause any significant deterioration of liver function in patients with cirrhosis and hepatocellular carcinoma. 12:41 pm

Abstract No. 10

Chemoembolization of Hepatic Metastases from Neuroendocrine Tumors. M.F Giroux, Hosp. Univ Penn, Philadelphia, PA, USA • C Cope • R.A. Baum • RD. Shlansky-Goldberg • D.B. Freiman • M.C Soulen PURPOSE: To evaluate response and survival after hepatiC chemoembolization with cisplatin, doxorubicin, mitomycin-C, Ethiodol, and polyvinyl alcohol in patients with liver metastasis from neuroendocrine tumors.

Moderators: William Rilling, MD Laszlo Horvath, MD

12:30 pm

lar carcinoma caused by chemoembolization (TACE).

PURPOSE: To evaluate the potential damage to underlying liver function in cirrhotic patients with hepatocellu-

MATERIALS AND MEIHODS: 24 patients were treated between 1991 and 2000 because of hormonal symptoms (n=13) and/or progreSSive liver-dominant disease. On average, liver metastasis was diagnosed 20 months before commencing liver-directed therapy. 10/24 patients(42%) were previously treated with systemic chemotherapy and 50% received Sandostatin therapy. Chemoembolization was performed monthly in a lobar distribution until the entire tumor burden was treated. Pretreatment and post-t.reatment cross-sectional imaging and clinical follow-up were obtained at one month and every three months thereafter. Recurrent intrahepatic disease was treated with additional chembolization as needed.