8:55AM Abstract No. 77

Treatment of Unresectable Hepatoma Using Y-90 TheraSphere: Imaging, Laboratory & Clinical Response. R. Salem, Northwestern University, Chicago, IL, USA 'RJ Lewandowski·C. Wong' VL. Gates

PURPOSE: Recently, Yttrium-90 has become available for the treatment of hepatocellular carcinoma. TheraSphere® consists of millions of small glass microspheres 20-30 microns in diameter incorporating Yttrium-90 as a constiment of the glass matrix. It is a beta emitter with an average penetration of 2.5 mm. The half-life is 64.2 hours. Since it is administered via the hepatic arterial route, it can be viewed as " internal" beam radiation.

MATERIALS AND METHODS: 47 patients underwent TheraSphere treatment for unresectable hepatoma. The average lobar volume was 1020 cc; the average dose of Y90 was 2.2 GBq, average absorbed dose of 100-150 Gray. 28 patients were Okuda stage I, 19 were Okuda stage 2. Patients had baseline liver function tests and tumor markers on the day of treatment. Clinical follow-up and laboratory values were obtained at 30 day interval. Follow-up CT scans of the liver were obtained at 30 days, and subsequently at 90 day intervals.

RESULTS: 46 of 47 patients received treatment on an outpatient basis and were discharged 6 hours after catheterization. 30 and 90 day clinical, laboratory and CT follow-up was available in 42 of 4 7 patients . 81 % of patients complained of fatigue for 7-10 days. On CT imaging, an average tumor size decrease of20% at 3 months was noted. Average AFP drop in the 47 patients was 51 % at an average of 90 days. Median survival for Okuda I and 2 patients was 623 days (21 months) and 337 days (11 months) respectively.

CONCLUSION: TheraSphere® hepatic unilobar infusion for hepatocellular carcinoma appears to represent an efficacious therapy with minimal toxicity in a patient population whose disease is refractory to other forms of treatment. It can be safely performed on an outpatient basis. Survival data in hepatocellular carcinoma are extremely promising.

Scientific Session 15 Biopsy/Drainage

Saturday, March 27, 2004 8:00 AM - 9:30 AM Moderator(s): Stephen T. Kee, MD

Eric VanSonnenberg, MD

8:00AM

FEcHt'RED ABSTRACT

Commrntator: (;rrhard R. \\ itlich

Abstract No. 78

'Effect of Self-Hypnotic Relaxation OJ) Distress during and after Large Core Breast Biopsy. Preliminary Results. IE. V. Lang, BIDMC!Han'ard Medical chool, Bo lon, MA, USA '0. Hatsiopoulou '1V. Hats ~Y' V. Stinsliofl'C. chupp·K. Berbaum

PURPOSE: To assess whether a imple nonphannacologic intervention during large core breast biopsy decrea e womens' distress during biopsy and carries.over into recovery.

MATERIAI'sAlVD METHODS: In this ongoing study with a .target of240 patients, 154 women between the ages of 19-86 years (mean 48 years) have been enrolled to date and have been randomized to receive standard care treatment (n=46) empathic attention control (n=55) or self-hypnotic relaxatio (11=53) while being on the biopsy table. Women rated pain and anxiety on .0" I 0 seIf~reporting scales .before, during,

immedia ely after t e iopsy and for tbe 4 subsequent days in ~iarie . Biopsie were done with local ane thelic, an 11-14 g biop y-gun and under stereotactic (11=97) or ul tra nographic guidance (11=56 .

RR ULTS: Repeated measures analysis showed that anxiety d.ecreased linearly over tbe procedure time in the hypno i group, butnol in the empathy and tandard groups. Pain ros linearly throughout the procedure but at different rates: least in the hypnosis group and mo t in the empathy group. Immediately after biopsy, one-way A OVA with sub equent Tukcy tests showed significantly greater anxiety after biopsy UJlder landard or care (score 1.57) than hypnosis (score 0.67), .and grea~er anxiety under empathy ( 'core 2.00) than hypnosis (p<O . .ol ). There was a si gnificant interacti on between treatment groups before vs after biop y with hypn o i patients 'faring better (p<O.05). For alL'C iety on days 2-5 there was a statistically ignificantinteracLion orthe day foIlowin biopsy and trealment group (p=O.005) with an apparen ~ increase of anxiety in the standard group around day 3 and 4 when biopsy results were made available. Proced ure time were 51 min for standard care, 49 min for empathy, and 45 min for hypnosis treatment.

CONCLUSiON: Providing patient at tbe critical ti me orlarge tore breast biopsy with an appropriate. time-sensitive coping mechani m can decrea e distress during Qiop y and in the day thereafter.

8:15AM Abstract No. 79

Sterile Fluid Collections in Acute Pancreatitis­Catheter Drainage Versus Simple Aspiration. E.M. Walser, UTMB, Galveston, TX, USA·S. W Champion'W Nealon'S. Raza

PURPOSE: Do clinical benefits justify catheter drainage of non-infected acute pancreatic fluid collections?

MATERIALS AND METHODS: We reviewed the clinical and imaging data of patients with acute pancreatic fluid collections from 1998-2003. Referral for fluid sampling was based on elevated white blood cell count and fevers. Those patients with culture negative drainages or needle aspirations were included in the study. Eleven patients had aspiration only (A) and 21 patients had aspiration and then catheter placement for chronic evacuation of fluid (C) . We excluded patients with grossly purulent collections and chronic pseudocysts. In this group, we also recorded the number of sinograms and catheter changes and duration of catheter drainage. The CT severity index and Ransom scores were calculated for all patients at presentation. The total length of hospital stay (LOS), length of stay after the percutaneous procedure (LOS-P), and conversions to surgical drainage/debridement were recorded as well as the 30 day mortality and long-term survival. Statistical analysis was dDne using Kaplan-Meier survival curves and Kruskal-Wallis one-way analysis of variance to compare the two groups.

RESULTS: The CT severity index and acute Ransom scores were not different between the two groups (p=0.19 and p= 0.23, respectively). Duration of hospitalization did not differ significantly between the two groups with a mean LOS and LOS-P of 31.8 days and 22 days in group A and 33.8 days and 22.8 days in goup C, respectively (p > 0.80). The 30 day mortality was 1111 (9%) in group A and none in group C. Kaplan-Meier survival curves for both groups were not significantly different (p=0.18). Surgical conversions occurred in 211 I (18%) of patients in group A and 8/21 (38%) of patients in group C. Patients undergoing catheter drainage required an average of 5.3 sinograms/tube changes and kept catheters in for an average of 62.8 days. S171

SI72

CONCLUSION: There is no apparent clinical benefit for catheter drainage of sterile fluid collections arising in acute pancreatitis. Suspicious fluid collections should be sampled or aspirated initially and catheter drainage should be deferred until bacteriologic analysis and culture results are known.

8:26AM Abstract No. 80

CT-Guided Transvenous, Transcaval, Needle Aspiration Biopsy of Pancreatic and Peripancreatic Masses. C T Sofocleous, Memorial Sloan K ettering Cancer Center, New York, NY, USA·K. T Brown'J Schubert·L. Brody ·A.M Covey·G. Getrajdman

PURPOSE: To evaluate the safety and efficacy of Needle Aspiration Biopsy (NAB) of pancreatic and peripancreatic masses using CT guidance and an approach that transgressed the inferior vena cava (lVC) or the renal vein.

MATERIALS AND METHODS: From January 2000 to July 2003, 55 patients underwent 59 biopsies of masses located within the pancreas (28) or in a peri pancreatic location (21). An 18 - 22 gauge needle was used to obtain specimen through the IVC or the renal vein . Accurate needle position ' was confirmed by CT imaging prior to obtaining the specimen. All specimens were evaluated by cytology. Immediate post-biopsy CT imaging was repeated through the biopsy site, to evaluate for bleeding. Cytology reports and medical records of all patients were retrospectively reviewed to evaluate diagnostic accuracy, and complication rates.

RESULTS: Masses were accessed with a direct (non-coaxial) pathway which transgressed the IVC (55) or/and the renal vein (7). In 7/59 (12%) cases the sample was deemed non diagnostic. Ofthose 4 were diagnosed later by endoscopy (I), surgical biopsy (2) and repeat NAB (I). Cytological examination was positive for malignancy in 39/59 biopsies. Benign lesions were demonstrated in 13/59 biopsies. Faise­negative result was noted in I case which was eventually diagnosed by repeat NAB. CT evidence of peri Ie siona I blood was seen in 8/55 patients (8/59 -14%-biopsies) which remained asymptomatic. There were no clinically significant complications.

CONCLUSION: CT-guided NAB using a direct approach via the IVC or renal vein is safe and effective in obtaining diagnostic specimens from pancreatic and peri pancreatic masses.

8:37AM Abstract No. 81

Evaluating the Utility of Biopsy Forceps in Transjugular Liver Biopsies. J C Tham, Beth Israel Deaconess Medical Center, Boston, MA, USA 'MF Mastomalleo ·A . Rao

PURPOSE: To evaluate the efficacy and safety of biopsy forceps in percutaneous transjugular liver biopsy.

MATERIALS AND METHODS: The medical records of forty­four patients with contraindications to percutaneous liver biopsy who underwent transjugular liver biopsy were studied. Forty (91 %) patients had thrombocytopenia and/or elevated prothrombin times, and seven (J 6%) patients had ascites. Boston Scientific T-Rex biopsy forceps were used in all patients.

RESULTS: Right transjugular access to the hepatic veins was used in al144 patients. A mean of 5.5 (range 2-1 0) specimens were obtained from the right (41 /44), middle (6/44), and left (1 /44) hepatic veins. Mean specimen size was 4mm in length (range 2-14mm) Tissue specimens were deemed satisfactory

for histologic diagnosis in 41144 (93%) cases and definitive diagnosis was made in 40/44 (91 %). Diagnoses included cirrhosis 28/44 (64%), stage I-III fibrosis 8/44 (1 8%), Wilson's disease 2/44 (5%), cholangitis 2/44 (5%), and malignancy 1/ 44 (2%). There were no procedure-related complications.

CONCLUSION: The transjugular biopsy forceps system is a safe and effective method of performing transjugular liver biopsies.

8:48AM Abstract No. 82

Fibrin Glue Embolization of Percutaneous Lung Biopsy Tract to Prevent Pneumothorax in a Rabbit Model. K. Tay, Singapore General Hospital, Singapore, Singapore '!. Song'L. AW

PURPOSE: To determine if embolization of the biopsy tract using fibrin glue (Tisseel) is effective in preventing pneumothorax following percutaneous lung biopsy in a rabbit model.

MATERIALS AND METHODS: The research protocol was approved by the Hospital's Ethics as well as Animal Care and Use Committees. New Zealand white rabbits weighing between 4 to 5 kg were used. All procedures were performed under general anesthesia and all the animals were intubated and placed i

on positive pressure ventilation. Percutaneous lung biopsy was performed using the co-axial technique with a 17G co­axial needle and a 18G core biopsy needle. The biopsies were performed in the lung bases on both sides, with I side serving as control where embolization of the biopsy tract was not performed. For the opposite test side, fibrin glue (commercial preparation, Tisseel) was injected along the biopsy tract via the 17G coaxial needle as it was being withdrawn. The procedures were performed under fluoroscopic guidance in 3 animals. The presence of pneumothorax was assessed with a post procedure chest radiograph. In 3 other animals, the procedure was performed under CT fluoroscopic guidance and pneumothorax was assessed with a post biopsy CT scan. The animals were sacrificed at the end of the procedure and gross pathological as well as histological examinations of the biopsy tracts were performed.

RESULTS: All 6 animals sustained pneumothoraces on the control sides (1 00% pneumothorax) where tract embolization was not performed. Of these, 3 were tension pneumothoraces where marked contralateral mediastinal shifts were present. There was no evidence of pneumothorax on all 6 test sides (0% pneumothorax) where fibrin glue embolization was performed. The mean volume of fibrin glue used was 0.7 cc (range 0.5 - 1.0 cc),

CONCLUSION: Biopsy tract embolization using fibrin glue is effective in preventing pneumothorax following percutaneous lung biopsy in a rabbit model.

The research is supported by a grant from the Singapore Radiological Society Trust Fund.

8:59AM Abstract No. 83

A Comparison of Two Different Pathways for Core Liver Biopsy: Is Patient Admission Necessary? MM Pilat, LSUHSC-Shreveport, Shreveport, LA, USA ·JR. Rodriguez -NS Lemoine 'G.F Black 'CJ Schonholz 'H.R. D 'Agostino

PURPOSE: To report the results of two pathways for outpatient core liver biopsy: one that comprises 23-hour hospital admission and another one using a postprocedure 2-hour observation streamlined process.

MATERIALS AND METHODS: The study includes 177 consecutive patients that underwent core liver biopsy (percutaneous, 167 patients; transjugular, 10 patients). Biopsy indications were: hepatitis C, 149 patients; unspecified diffuse liver disease, 21 patients; hepatitis B, 3 patients; autoimmune disease, 2 patients; post transplantation, 2 patients (liver, 1; bone marrow, I). Patient evaluation involved a directed history, physical exam and coagulation studies. Local anesthesia with or without conscious sedation was utilized . Patients with normal coagulation underwent percutaneous US-guided biopsy while patients with coagulopathy had fluoroscopy-guided transjugular biopsy. Immediate complications were assessed while the patients were in the hospital. Late complications were evaluated by telephone interview. In the initial 19-months of the study, 103 patients had their biopsy using a 23-hour hospital admission (HA). A 2-hour post biopsy observation streamlined process (SP) was utilized in 74 patients thereafter. Parameters compared were: paperwork amount involved in each pathway, biopsy diagnostic efficacy, complications, and each process costs.

RESULTS: HA pathway involved 16 forms while the SP pathway involved 7 forms. Diagnostic samples were obtained in 175 (98.9%) patients [HA, 101 (98%) patients; SP, 74 (I 00%) patients]. Six patients experienced minor complications (immediate, 3 patients; late, 3 patients). Of these 3 (2.7%) occurred in HA patients and 3 (4 .1 %) in SP patients. Telephone survey revealed self limited pain in 28.6% in HA patients' and in 29.6 % in SP patients. Costs from percutaneous liver biopsy ranged from $2749-2849 for HA patients and $2389 for SP patients; transjugular liver biopsy costs for HA patients ranged from $5377-5477 and was $5017 for SP patients.

CONCLUSION: Safe and effective core liver biopsy carried out using a streamline pathway had less paperwork and substantial cost savings when compared with a 23-hour hospital admission process.

Scientific Session 16 UAE

Saturday, March 27, 2004 12:30PM - 2:00 PM Moderator(s): Kevin Dickey, MD

Connie Hsu, MD

12:30 PM Abstract No. 84

Choice of Embolic Material for Uterine Embolization: Results of a Randomized Trial. K.c. Sterbis, Georgetown UniverSity Hospital, Washington, DC, USA ·S.D. Allison' MF McCullough·J M Bnmo ·P.A. Flick·JB. Spies

PURPOSE: To compare the short-term and mid-term outcomes following uterine artery embolization (UAE) of leiomyomas using polyvinyl alcohol particles (PVA) and tris­acryl gelatin microspheres (TAGM).

MATERIALS AND METHODS: One hundred patients with symptomatic leiomyomas were randomized to embolization with either PYA or TAGM. Each patient had a baseline pelvic MRl and completed a baseline symptom severity and quality oflife (QOL) questionnaire. Procedural data included volume of embolic material used, occurrences of micro-catheter occlusion and the presence of uterine artery spasm and ovarian artery flow to the uterus. Short-term outcome measures included VAS pain scores and analgesic use in hospital, as well analgesic use, VAS scores, symptom scores and twice-daily temperature during the first week post-procedure. Symptom

scores were also assessed at 2, 3 and 4 weeks after embolization. Mid-term outcomes were measured at 3 months after embolization. Symptom and QOL scores were compared to baseline and MRl was used to evaluate for uterine and fibroid volume change and the extent of fibroid infarction. Analysis controlled for the presence of ovarian artery supply and the presence of uterine artery spasm.

RESULTS: The two treatment groups were similar at baseline. At the time of embolization, there was a difference in mean volume of embolic material (9.4 ml ofTAGM versus 3.0 ml of PVA). The frequency of catheter occlusions was much higher using PYA (28%) than TAGM (4%). There were no other significant procedural differences. No difference was seen in any of the short-term outcome measures, including post­procedural pain, use of pain medication, temperature and symptom scores. Mid-term outcome data showed no significant difference in scores for bleeding or pain and pressure. The groups had no difference in overall symptom scores or quality of life scores. MRI at 3 months showed no difference in the mean change in uterine or fibroid volume or the presence of un infarcted fibroids using PYA versus TAGM.

CONCLUSION: We found no difference in short-term and mid-term outcome from uterine embolization using PYA or TAGM .

12:41 PM Abstract No. 85

Comparison of Two Different Angiographic End-Points for Uterine Fibroid Embolization with PYA. A Multicenter Study. J Golzarian, University o/Iowa, Iowa City, lA, USA·C. Torres·s. Sun-D. Valenti

PURPOSE: The stan'dard end-point of uterine artery embolization (UAE) for fibroids is complete stasis in uterine arteries (UA). Limited uterine artery (L-UAE) with occlusion of the peri-fibroid feeding vessels, preserving the flow inside normal branches of UA has beel) suggested as the endpoint when using spherical embolic agents. The aim of this study is to compare these two end-points using regular (non-spherical) polyvinyl alcohol (PVA) particles.

MATERIALS AND METHODS: This is a retrospective multicenter study from a prospectively acquired database. L­UAE was performed in 72 patients with symptomatic uterine fibroids (group A). Results were compared to a group of 62 patients with complete UAs embolization (group B). Embolization was performed using 500-700/lm or 355-500 ,urn non-spherical PYA particles. The mean dose of morphine in center 1 and overnight admission for pain in center 2 were noted for both groups. Clinical success, uterine or largest fibroid volume and complications were compared for each group.

RESULTS: Mean dose of morphine and the time spent to the hospital were significantly lower in group A. The mean duration for intense pain was respectively 105 and 522 min. The need for overnight hospital admission for pain in group A and B were respectively 6 and 23%. The uterine or largest fibroid volume reduction was not significantly different between groups. Symptoms were completely resolved or significantly improved in 93% of patients in both groups. Two patients in group A had assisted fibroid expulsion. One spontaneous expulsion occurred in group B. Transient amenorrhea was observed in one patient in group A and two patients in group B. Vaginal discharge was observed in 6 and 10 patients respectively in group A and B. The amount of embolic agent used in group A and B was respectively 2.3 and 4.7.

S173