- Email: [email protected]
GU Interventions
MATERIALS AND METHODS: We retrospectively reviewed twenty-two patients with pelvic abscesses refractory to antibiotic therapy who underwent a single step transvaginal pelvic aspiration ar drainage between 1995 and 2000.
Scientific Posters Chest I Gil GU Interventions
F Soria, Minimally Invasive Surgery Centre, Caceres, Extremadura, Spain • F Sanchez • V. Crisostomo • J. Ezquerra • E. Duran • J. Usón
RESULTS: Transvaginal aspiration ordrainage was successful in 19 of the 22 patients for an 86% success rate. Of the three patients that failed aspiration or drainage, all ultimately went on to have surgery despite undergoing repeat drainage procedures. Drainage catheters were placed in 15 of the 22 patients (68%) and left in place an average of 3.7 days. Aspiration alone resulted in a 100% success rate, while drainage with catheter placement resulted in an 80% success rate. No complications including bleeding, bowel perforation, or death were reported in any of the procedures.
PURPOSE: To assess the diagnostic value of intraluminal ultrasonography (ILUS) for ureteral stenosis and its usefulness as a follow-up method after endourological therapy.
CONCLUSION: Transvaginal ultrasound-guided aspiration or catheter p1acement with the trocar technique is a safe and effective treatment for refractory pelvic abscesses.
Poster No. 244 Endourologic Therapy for Ureteral Stenosis: Intraluminal Ultrasonographic Assessment. Experimental Study.
MATERIALS AND METHODS: 20 female pigs were used for this study. The study was divided in 3 phases. During Phase I, animals were subjected to an ILUS study using a 20MHz probe to evaluate ureterallayers in the studied segment, (5cm distally to the UPJ). An experimental model ofureteral stenosis was then performed.Phase II. Retrograde ureteropielography (RUP) and ILUS were carried out 4 weeks after model creation, in order to evaluate the presence of stenosis. Stenoses were then treated by retrograde endoureterotomy by endoballoon rupture, and subsequent deployment of a double pigtail stent, which was left in place for 6 weeks. Phase III: 4 weeks after removal of the stents, a further ILUS examination was performed. RESULTS: Phase I. Anatomical structure was carrectly depicted by ILUS in al! animals. A characteristic inner mucosa with hyperechoic lamina propia, sUITounded by a hypoechoic area representing the muscular layer, and the outer, hyperechoic serosa and periureteral fat. Phase II. RUP and ILUS demonstrated the existence of a ureteral stenosis in all animals. ILUS depicted a greatly diminished ureteral lumen with hyperechogenicity of al! ureterallayers, more evident at the muscle and serosa, indicating the existence of ureteral and periureteral fibrotic reaction and precluding discrimination of the ureteral layers. Phase III. An increase in the luminal diameter compared to the measurements made during the previous phase was seen in al! animals but 4. Ureterallayers could be clearly identified, as the fibrosis in the ureteral wall had d.isappeared. There was a rernnant of periureteral fibrosis around th.e lesion area. The 4 pigs with restenosis showed a ultrasonographic appearance similar to that seen in phase II.
Poster No. 246 Percutaneous Transgastric Pancreatic Pseudocyst Drainage. Pc. Shetty, Hurley Medical Center, Flint, MI, USA. U.B. Chintalapudi • D. Reddy • R. Sharma • A. Muniyappa PURPOSE: To assess the efficacy and safety ofPercutaneous Transgastric Pancreatic Pseudocyst Drainage under floroscopic and ultrasound guidance. MATERIALS AND METHODS: Fourteen pancreatic pseudocysts were drained using the percutaneous transgastric approach in 12 patients. Patients' age ranged from 13 to 72 years with a mean of 56 years. AJ] pseudocysts were diagnosed by CT. Anterior stornach wall was anchored to the anterior abdominal wall using two T-anchors. Stornach was punctured between the two T-anchors and pseudocyst was punctured through the posterior wall of the stornach. The tissues between the pseudocyst and the posterior stornach wall were dilated using 5mm diameter balioon catheter. Cyst was drained into the stornach using double pig tail stent in one case, extemal drainage pigtail catheter in another case and proximal cope loop catheter in the remaining 10 cases. Catheter was left in place for a mean of 22 days. A CT was performed when the catheter stopped draining, to check for any residual fluid collection. RESULTS: Thirteen cysts in 11 patients were successfully treated with this method. One patient had excessive bleeding through the drainage catheter. This patient was successfully treated surgically without any complication.
CONCLUSION: ILUS can be useful in depicting changes in the ureteral wal! and periureteral tissues secondary to stenosis. The technique can be used to monitor endourologic interventions, as it shows changes associated to ureteral fibrosis, specially at the muscular and periureteral, the most common causes of restenosis.
CONCLUSION: Percutaneous transgastric pancreatic pseudocyst drainage is an effective and safe procedure. In most patients this can avoid surgery and decrease the morbidity.
Poster No. 245
Poster No. 247
Single Step Transvaginal Aspiration and Drainage for Refractory Pelvic Abscesses. B. C. Lee, University oj Califomia, Davis Medical Center, Sacramento, CA, USA. J.P McGahan • B. Bijan
A ScJerotherapy of Symptomatic Hepatic or Rena! Cyst by Ethanolamine Oleate. K. Yamamoto, Nara Medical University, Kashihara, Japan • H. Sakaguchi. H. Anai • T. Tanaka • K. Kichikawa • H.
PURPOSE: For treatment of pelvic abscesses, use of the trocar technique avoids many of the technical challenges of the Seldinger method. The purpose of this study is to evaluate the effectiveness and safety ofultrasound-guided transvaginal aspiration or drainage with the trocar technique in pelvic abscesses that are refractory to antibiotic treatment.
Uchida, et al. PURPOSE: Large simple cyst that cause flank, back pain or deteriorating liver ar renal function require interventional therapy. Percutaneous puncture and aspiration or sclerotherapy are well performed and less invasive management options compared to surgical treatment. However, by using
S85
absolute alcohol as sclerosing agent, adverse effect which is a severe pain or feeling of intoxication is often appeared. The purpose of this study is to clarify the efficacy of ethanolarnine oleate(EO) as a sclerosing agent for symptomatic hepatic or renal cyst. MATERIALS AND METHODS: Seven patients with symptomatic liver or renal cyst were treated by sclerotherapy with EO. Four patients had renal cyst and 3 patients had hepatic cyst. The cyst size in great diameter was ranged from 6 to 12 cm and the average of its size was 8 cm in diameter. The cyst was punctured with 199auge needle under US guidance and 6.5F catheter was placed. After all of the cyst content was aspirated, an iodized contrast agent of a 10% volume of the cyst content was injected to check the absence of communication between the cyst and biliary tree or vessels and to test for the absence of leak into the peritoneum. And then, the solution of ethanolamine oleate-iopamidol mixture(EOl) was injected via catheter. After EOl injection, the patient was kept in supine, right lateral and left lateral position for every 10 minutes. After 30 minutes, the injected EOl was aspirated completely before catheter removal. A follow-up CT was performed at 3 month after treatment. The volume of cyst and its reduction rate was calculated. In addition, symptom and complication were assessed. RESULTS: The volume of cyst was ranged from 79 mI to 628 ml (mean;276ml) before treatment. After sclerotherpy, its volume was reduced at a month. And three months later after treatment, it was ranged from 2 ml to 61 mI (mean;24mI) and the reduction rate of cyst volume was more than 90% in average. Symptom caused by cyst was disappeared in all cases (717; 100%), and no major complication was encountered. Although 3 patients had a low grade fever after sclerotherapy, it was easily controlled. CONCLUSION: It is suggested that the sclerotherapy with EO might be asafe, effective and well-tolerated treatment for symptomatic hepatic or renal cyst.
Poster No. 248 Tangential Radial Sampling or Hepatic Lesions Via Double Coaxial (TRASH-DOC) Technique. B. Bijan, University oj California Davis Medical Center, Sacramento, CA, USA. R. Kazemaini • J.P. McGahan PURPOSE: Obtaining detailed information regarding the in vivo behavioral traits of a lesion is crucial in diagnosis and therapy planning: Our technique emphasizes the utilization of cross-sectional imaging modalities to obtain more informative sampIes. The information related to the lesional differential zones can be obtained by utilizing our sampling technique more reliably and equally safely. MATERIALS AND METHODS: Atotal of 23 consecutive lesions were biopsied using TRASH-DOC technique. US/CT were used for guidance. Single access port through Glisson was created via Double Coaxial Technique. 5 radial sampIes were obtained starting tangentially from the outermost normal liver parenchyma, continuing centrally, in attempt to obtain at least 2 out of 5 tumorai zones in each sample. A control CTIUS was used to rule out early complication prior to withdrawal of the needle. The blood patch, prepared from patient blood, was routinely applied. In coagulopathic cases, it was mixed with FFP or alternatively Gel foam slurry. Iodinated contrast was mixed to render radio-opacity, detectable by CT.
586
RESULTS: In our series, tangential approach provided a larger sample from each zone, with less contamination from the nearby zones. Providing this type of sampIes helped the
pathologist to study the tumor-liver interface in more details, which is vital for evaluation of tumoral behavioral traits, differentiation of locally invasive versus non-invasive lesions, and response to therapy. The procedure time and cost were increased slightly. However, considering the cost of resampling, using TRASH-DOC technique appears to be both cost and time efficient. No major complication leading to any additional hospital stay or additional intervention was noted. The sampIes were judged by our pathology department to be adequate in all cases. The reduced risk of complications, including track seeding, seen in this technique were analyzed. CONCLUSION: Using double coaxial technique with multiple radial sampling obtained tangentially from different zones with subsequent blood patching is a safe, accurate and effective mean of obtaining tissue for histological sampling of focal hepatic lesions.
Poster No. 249 Predictors or Infected Bile among Patients Undergoing Percutaneous Cholecystostomy. S. Beardsley, University ojPennsylvania, Philadelphia, PA, USA. T. WI. Clark PURPOSE:Patients may not achieve a clinical benefit after undergoing percutaneous cholecystostomy (PC) due to the inherent difficulty in identifying patients who truly have infected gallbladders. We sought to identify imaging and biochemical parameters which would help to predict which patients have infected gallbladders. MATERIALS AND METHODS: AlI patients who underwent PC over a 48-month period were retrospectively identified. Of these 78 patients, bile culture results from sampIes drawn at the time of PC. were available in 52 patients. We used logistic regression to assess whether the following variabIes were predictive of infected bile: white blood cell count, alkaline phosphatase, aspartate transferase, alanine transferase, bilirubin, gallstones (CTIUS), sludge (CTIUS), gallbladder (GB) wall :e:3 mm (CTIUS), pericholecystic fluid (CTIUS) and absence of GB visualization on radionucIide hepatobiliary scanning. RESULTS: Of the 52 patients, 25 (48%) had infected bile. Organisms cultured included Entercoccus, Enterobacter, Klebsiella, Pseudomonas, E. Coli, Citrobacter, and Candida. No biochemical parameters were significantly predictive of infected bile; WBC > 15,000 was weakly associated with greater odds of infected bile (OR=2, P=NS). The presence of gallstones, sludge, GB wall thickening and pericholecystic fluid by US or CT were not predictive of infected bile (P=N.S.), alone or in combination. RadionucIide scans were performed in 31 % of patients; all were positive and 66% of these patients had infected bile. Since no patients who underwent radionuclide scans had negative studies, this variable could not be entered into the model due to collinearity. CONCLUSION: No single CT or US imaging variable was predictive of infected bile, and onIy a weak association of WBC to infected bile was seen. No other biochemical parameters had any association with infected bile. The ability of radionucIide scanning to predict infected bile was higher than US or CT. This study illustrates the continued challenge to identify cholecystitis among patients referred for percutaneous cholecystostomy.
Poster No. 250 CHnical Suspicion of Cholangitis Is Only Predictor of Bacteremia after Drainage of Malignant Biliary Obstruction. P.M. Cote Robson, Menwrial Sloan-Kellering Cancer Center, New York, NY, USA. L. Brody. K. Brown. A. Covey. G. Getrajdman
PURPOSE: Bacteremia after percutaneous biliary drainage (PBD) is a potentially life-threatening event. Tbis study was performed to identify clinical factors predicting post-PBD bacterem.ia within 48 hours in a complex population of cancer patients. MATERIALS AND METHODS: Prospective cohort study of individuals undergoing PBD at Memorial Sloan-Kettering between 10/98 and 12/00. RESULTS: 278 PBD procedures were performed in 221 patients (:21 men, 100 women; mean age 62.8 years). Major underlying diagnoses were biliary tract malignancy (74; 33.5%), pancreatic/ampullary cancer (45; 20.4%), and metastatic disease from colorectal cancer (39; 17.6%), or other primary (48; 21.7%). Most common indications for PBD were suspected cholangitis (68; 35.1 %), pruritis (106; 38.1 %), or jaundice ?reventing chemotherapy treatrnent (78; 28.1 %). Multiple indications were reported for 56 (20.1 %) procedures. Blood cultures were drawn within 48 hours after 72/278 (25.9%)procedures to evaluate suspected post-PBD bacterem.ia. Post-procedure blood cultures were positive after 21 procedures (7.6% of total). No correlation was found between post-procedure bacterem.ia and age at procedure, underlying diagnosis, level of obstruction, type of procedure (R drainage vs L drainage, Ext vs IntlExt), prior PBD, presence of an indwelling stent or catheter, pre-procedure WBC, total bilirubin, alkaJine phosphatase, or AST/ALT. Prior instrumentation (120 procedures, 43.2%) was associated with positive bile cultures (P
PURPOSE: A specially designed self-expandable covered metallic stent incorporated with beta-em.itting radioisotope, Holm.ium-166 (Ho-166), was developed at our institute for the purpose of intraluminal brachytherapy as well as for internal bile drainage in malignant biliary stricture. This study was aimed to evaluate the safety of the radioactive metallic stent on the normal cartine common bile duct (CBD) prior to the clinical application. MATERIALS AND METHODS: Self-expandable stent (diameter; 6rnm, length; 20rnm) was made of nitinol wire. Ho166 was incorporated within polyurethane (50J..Lm thickness)
which was covered over the outer surface of m.id l cm of the stent and, both ends of stent were bare portion. The stent with radioactivity of 21-135 J..LCi ( mean; 77.9 J..LCi) were placed in the CBD of 20 heal thy beagle dogs. The dogs were sacrificed 3-6 months after stent insertion and histopathologic examination of CBD was performed.
RESULTS: There was no stent migration or stent ocelusion in any cases. Grossly, stricture or perforation was not seen in all cases. Varying degrees of papalliary mucosal hyperplasia was observed at uncovered area. Fibrosis was noted in Ho-166 coated area, instead of mucosal hyperplasia. lnf1arnmation was demonstrated, but dyplasia or ulceration was not in all cases. The membrane of Ho-166 coat was intact in all cases. CONCLUSJON: These studies suggest that radioactive covered self-expandable metal lic stent delivering therapeutic radiation dose can be used and it can be an alternative therapeutic modality for palliative treatment of the biliary tract malignancy. Poster No. 252 Factors Associated with Mortality in Patients Undergoing Percutaneous Cholecystostomy. B.D. Davison, Brigham and Womens Hospital, Boston, MA, USA. l. F. Polak
PURPOSE: While cholecystectomy has been shown to be an effective treatrnent for acute and chronic cholecystitis it appears to be largely reserved for the most critically ill patients. In this population it is associated with a high mortality rate, and is often requested without elear clinical criteria. We looked at our own experience in a tertiary care hospital to evaluate the predictors of short and long-term mortalityof percutaneous cholecystostomy in our patient population. MATERIALS AND METHODS: The medical records of 81 consecutive patients in whom cholecystostomy was performed between July 1995 and July 2001 were retrospectively reviewed. Biochernical variabIes and chnical outcome were determ.ined by chart review. Analyses were conducted with Cox proportional hazrds models with mortality as the primary outcome. RESULTS: Percutaneous cholecystostomy was technically successful in 81 (100%). Overall mortality at 30 days was 2% (first mortality recorded at 18 days and not related to the procedure). By l year, mortality was 57% and reached 70% at 2 years. Factors associated with mortality (after controlling for age and gender, were elevated tota! bilirubin (hazard ration 1.18 for every l mg/dl increase) and elevated AST (HR of 1.01 for every l lU increase). Lawer risk of mortality was seen when the direct bilirubin was also elevated. CONCLUSION: Percutaneous cholecystostomy in the treatment of acute cholecystitis in high-surgical-risk patients is a relatively safe procedure. Otber factors related to patient co-morbidities account for a bigh mortality in the days and months following the procedure, at least in tertiary care institutions.
S87
Poster No. 253 The Cecostomy Butlon- An Alternative to Chronic Cholecystostomy Drains. M.K Varma, Tripler Army Medical Center, Honolulu, HI, USA • A. W Allen. M. Meyermann PURPOSE: Percutaneous cholecystostomy is a safe altemative in critically ill patients with cholecystitis and serves as a conduit to cholecystectomy in stabilized patients with calculous cholecystitis. However. there is a subset of patients in which medical co-morbidities make cholecystectomy prohibitive. These patients must then łive with an external drain, which often have a prominent extemal profile and high patient dissatisfaction. The aim of this study was to determine if the Chait Trapdoor Cecostomy Button could be used as an alternative to the standard pigtail cholecystostomy tube. MATERJALS AND METHODS: At our institution, 4 patients in the last 24 months were diagnosed with calculous cholecystitis who were not operative candidates. Intially. 10 FR pigtail, ultrasound-guided percutan,eous cholecystostomy tubes were placed (transhepatic route-3 patients; transperitoneal route- I patient). On standard follow-up in 812 weeks, a 10.2 FR Chait Trapdoor Cecostomy Button catheter was płaced in all 4 patients. We then folIowed the patients with standard tube changes every 8-12 weeks with cholangiograms and patient interview (length of follow-up: range 8-26 mos, mean 14.7 mos).
Magnetic fiducials were embedded in the stylet of an 18gauge needle or 8French catheter. Positional data were sampled at a rate of 10 Hz. Axial CT scans of the phantom were obtained after płacement offour external fiducials·and a single internal fiducial (sensor-containing catheteror needle) within the foam model. Following registration, needle trajectories and suxface puncture sites were identified on MPR images. The goal was successful simultaneous puncture of two vessels. Punctures were performed by a single experienced and a single inexperienced practitioner. The needle placement time, tipto-target distance (magnetic transformation), and tip-to-vessel wall distance (from biplane radiographic images) were measured. RESULTS: Successful punctures of 5mm vessels by an experienced operator occurred in 16/26 total attempts. Fewer successful punctures occurred at a higher respiratory rate. Successful punctlJres were achieved by the experienced and inexperienced operators in 7114 and 2113 sequentiał attempts respectively. Depth errors occurred in 13118 missed punctures. Punctures of single superficial3mm and 5mm vessels resulted in success in 15/20 and 27127 attempts respectively. CONCLUSION: Factors which appear to contribute to errant magnetic guided punctures include smaller vessel diameter, depth of the target vessel, higher respiratory rate, and operator experience. In most cases the errant passes lie within one vessel diameter of the target. The system continues to show promise in an in vitro testbed. ORBH.::I
RESULTS: Teaching Points: 1. In patients with chronic cholecystostomy tubes with calcułous cholecystitis who are not candidates for cholecystectomy, a Chait, Trapdoor Cecostomy Button is better tołerated because of its lower external profile and an equally effective alternative to standard pigtail catheter drainage. We recommend placing this catheter after a mature tract has formed with an initial pigtail cholecystostomy tube placement. 2. We have found that the cecostomy catheter placement was successful in trans-hepatic but not· in trans-peritoneal placement. The trans-peritoneal route was associated with excessive leakage around the catheter. With a shorter route to the gallbladder transperitoneally and a more mature tract, a cecostomy catheter may be feasible. 3. The standard tract length for insertion of the cecostomy catheter is recommended by the manufacturer to be less than 6 cm. Because of the helicał extensile coils distally on the catheter, a tract length of up to 9 cm was used without dislodgement transhepatically.
Poster No. 254 Respiratory-Gated Magnetic Tracking-Guided Płacement System: Work in Progress.
Needłe
E.B. Levy, Georgetown University Hospitał, Washington, DC, USA • J. Tang • D. Lindisch • K Cleary PURPOSE: Magnetic track.ing of needles and catheters holds promise as a method to accurately target interventions and therapies without the use of ionizing radiation. We report our progress with the Aurora magnetic tracking guidance system with respiratory gating in an vitro model.
S88
MATERIALS AND METHODS: A previously tested abdominal torso containing a foam hepatic model mounted upon a servomotor platform was used. Respiratory motion was simulated by to-fro movement of the servomotor platform at 12-20 cycles per minute with an excursion distance of 12cm in the y-axis. The Aurora magnetic tracking system includes a field generator and up to two receiving channels.
Poster No. 255 Safety, Diagnostic Accuracy and Limitation oC CT-Guided Coaxial Cutting Needle Biopsies oC Lung Lesions. KM. Yeow, Chang Gung Memoriał Hospital, Kwei Shan, Tao Yuan, Taiwan, R.O.C • KP. Tsay • KW Lui • Cheung • K T Pan. A.S.B. Chou
Y.e.
PURPOSE: To evaluate the diagnostic accuracy, safety and limitation of er-guided coaxiał cutting needle biopsies of lung lesions. MATERIALSANDMETHODS: FromMarch 1995 to August 2001, a consecutive 631 biopsies with finał diagnoses confirmed either by surgery, imaging and clinica1follow up of at least 24 months were analyzed. The sensitivity, specificity. negative predictive value, positive predictive value, false positive rate, false negative rate and overall diagnostic accuracy were calculated. Complications presenting as pneumothorax, chest tube insertion rate and hemoptysis were obtained. The criticallesion size that caused significant drop in diagnostic accuracy and accompanied by rising complication rates was determined. P:S:;.05 was taken as statisticał significant. RESULTS: The overall diagnostic accuracy was 95 % (95 % CI 92.7-96.2) (Sensitivity =93 %; Specificity =98 %; Negative predictive value = 86 %; Positive predictive value = 99 %; False positive rate =0.7 %; False negative rate = 15 %). The pneumothorax rate was 23 % (155 of660), chest tube insertion rate was l % (9 of 660, 5 moderate pneumothoraces were successfully needle aspirated) and the hemoptysis rate was 4 % (26 of 660). N o mortality was encountered. For lesions:S:;1.5cm diagnostic accuracy dropped to 84%, bleeding rate increased by 12-times and hemoptysis rate rose from 4% to 8%. CONCLUSION: Cf-guided coaxiał cutting needle lung biopsy is highły accurate both for benign (86 %) and malignant (95 %) łung lesions and safe (23 % pneumothorax rate, l % chest tube insertion rate and 4 % hemoptysis rate). The technique is limited by lesion size:S:;1.5cm.
Poster No. 256 Complications of Thoracic Interventions: Evaluation of Contributing Factors. TB. Kinney, UCSD MedicalCenter; Dept oj Radiology, San Diego, CA, USA. H. T Chen • S. C. Rose • K. Valji • A. C. Roberts PURPOSE: Common thoracic interventions include thoracentesis, tube thoracostomy, and percutaneous lung biopsy. The rationale for using image guidance in thoracic intervention is to reduce such complications as local hemorrhage, hemothorax, and pneumothorax. aur goal was to assess the factors which contributed to complications of thoracic interventions performed at our institution. MATERIALS AND METHODS: Chart review of 416 thoracentesis, tube thoracostomy, and percutaneous lung biopsy procedures was undertaken. These procedures were performed at three different hospitals by a group of 6 interventionaJ radiologists between 1998 and 2002. The imaging modalities used included ultrasound, CT, and f1uoroscopy. RESULTS: Atotal of 61 (14.7 %) complications occurred, including 43 (l0.3 %) pneumothoraces; 7 (1.7 %) hydropneumothoraces; 8 (1.9 %) localized hematomas; 2 (0.5 %) fatal hemothoraces; and 1 (0.2 %) hemopericardium. Other incidents included l episode each of vasovagal syncope and hemoptysis. One of the fatal hemothoraces occurred secondary to diagnostic thoracentesis perforrned portably in the ICU with ultrasound using an 18 G DSA needle. The otheroccurred upon chest tube removal (8 Fr locking pigtail catheter) in a patiem with critical aonic stenosis who was on anticoagulation therapy. Of the pneumothoraces, 41 occurred in lung biopsies while 2 occurred with thoracenteses. Of the hydropneumothoraces, 6 occurred with thoracentesis while one occured on removal of a thoracostomy tube. CONCLUSION: Chest interventions are associated with a relatively smali number of complications. aur most frequent complication was pneumothorax folIowed by hydropneumothorax. Bleeding complications while rare can be fatal in certain cases. Unlike other areas of the body, the pleural space is subjected to vacuum and can accumulate significant amounts of blood. Portable intervention is associated with technical compromises compared to that used in mobile patients within an interventional suite and these may influence complications. Precautions with chest tube removaJ include stopping anticoagulation and careful technique.
Poster No. 257 CT-Guided Biopsy of the Lung on an Outpatient Basis: Safety and Efficacy. T lhaya. Tollori Red Cross Hospital, Tollori, Japan • J. Nakanishi • T Ookawa PURPOSE: To evaluate the safety and efficacy ofCT-guided percutaneous lung biopsies of pulinonary lesions on an outpatiem basis. MATERIALS AND METHODS: Atotal of 279 biopsies of the lung were performed under CT guidance on an outpatiem basis using 21-gauge fine needles or 20-gauge cutting needles. A chest radiograph was taken just after careful observation for two llOurs in the outpatient clinic and asymptomatic patients were allowed to go home with appropriate instnlctions for any complications.
RESULTS: The definitive diagnosis was malignant in 152 palients (54.5%) and benign in 122 (43.7%). In lhilteen cases (4.7%), no adequate sample was obtained to perform a cytological or histological diagnosis. Pneumothoraces were observed in 101 patients (36.2%) on CT scan just after the procedure and in 68 patients (24.4%) on chest radiograph two hours after the biopsy. None of them had any symptoms and required an insertion of chest tube. Minor hemoptyses were noted in five patients (1.8%) but it disappeared within two hours after the procedure. OnIy two patiems (0.7%) developed pneumothoraces with breathlessness more than 24 hours after the procedure and they required in-patient treatment in both cases. There was no significant difference in the complication rate between fine needles and cutting needJes. Total cost of CT-guided biopsy on an outpatient basis was just half of in-patient biopsy. CONCLUSlON: CT-guided needle biopsy is a safe and reliable procedure and it can be performed on an outpatient basis with a low complication rate and it also provides high cost effecliveness.
Oncologic Access
Interventions/Venous Poster No. 258
Evaluation of the Celi Death after Radiofrequency Ablation in the Dog Liver. H. Rikimaru, Depl oj Diagnostic Radiology, Tohoku University, Sendai City, Miyagi Prejeclure, Japan • T Matsuhashi • H. Miyachi • H. Saito • T lshibashi • S. Takahashi, et al.
PURPOSE: To clarify whether Hernatoxylin and eosin (H&E) stain can clearly distinguish the ablated area from the normal parenchyma after radiofrequency ablation. MATERIALS AND METHODS: RFA was appliedto 23 different lobes of nine adult female beagle dogs' liver. The 17gauge monopolar needle electrode with a 2cm-long tip was used. The power output was set at 30W for 6min. The dogs were killed immediately, 2hrs, 24hrs, and 3days after RFA application. H&E stai n, Terminal deoxynucleotidyl transferase UTP nickend labeling (TUNEL) stain, and Hypoxia inducible factor (HIF)-l ex. stain were perforrned, and they correlated one another. RESULTS: Twenty-three well-demarcated circumscribed ablated lesions were created. They consisted of 3 layers on H&E stain immediately after RPA, which turned to 4 layers after 2hours, and finally to 5 layers within 3 days. At the central areas (layer l and 2) of the ablated sites, celi features did not change over the time course, while some nucleus disappeared after 3days. At the peripheral region (layer 3 and 4), on the other hand, hepatocyte shrinkage, nuclear size reduction, nuclear condensation progressed within 24hrs. After 24hours, bothTUNEL and HIF-lex. stain became positive in every cel l of the peripheral region, which correlated very well with the areas identified by H&E stain (layer 3 and 4). CONCLUSlON: The central area of the ablated sites after RPA appeared to be coagulation necrosis, while the peripheral area showed the features of apoptosis. The peripheral area also proved to be ischemic. The H&E stain could clearly identify ablated areas from the nonnal parenchyma.
S89
Scientific Posters Chest I Gil GU Interventions
F Soria, Minimally Invasive Surgery Centre, Caceres, Extremadura, Spain • F Sanchez • V. Crisostomo • J. Ezquerra • E. Duran • J. Usón
RESULTS: Transvaginal aspiration ordrainage was successful in 19 of the 22 patients for an 86% success rate. Of the three patients that failed aspiration or drainage, all ultimately went on to have surgery despite undergoing repeat drainage procedures. Drainage catheters were placed in 15 of the 22 patients (68%) and left in place an average of 3.7 days. Aspiration alone resulted in a 100% success rate, while drainage with catheter placement resulted in an 80% success rate. No complications including bleeding, bowel perforation, or death were reported in any of the procedures.
PURPOSE: To assess the diagnostic value of intraluminal ultrasonography (ILUS) for ureteral stenosis and its usefulness as a follow-up method after endourological therapy.
CONCLUSION: Transvaginal ultrasound-guided aspiration or catheter p1acement with the trocar technique is a safe and effective treatment for refractory pelvic abscesses.
Poster No. 244 Endourologic Therapy for Ureteral Stenosis: Intraluminal Ultrasonographic Assessment. Experimental Study.
MATERIALS AND METHODS: 20 female pigs were used for this study. The study was divided in 3 phases. During Phase I, animals were subjected to an ILUS study using a 20MHz probe to evaluate ureterallayers in the studied segment, (5cm distally to the UPJ). An experimental model ofureteral stenosis was then performed.Phase II. Retrograde ureteropielography (RUP) and ILUS were carried out 4 weeks after model creation, in order to evaluate the presence of stenosis. Stenoses were then treated by retrograde endoureterotomy by endoballoon rupture, and subsequent deployment of a double pigtail stent, which was left in place for 6 weeks. Phase III: 4 weeks after removal of the stents, a further ILUS examination was performed. RESULTS: Phase I. Anatomical structure was carrectly depicted by ILUS in al! animals. A characteristic inner mucosa with hyperechoic lamina propia, sUITounded by a hypoechoic area representing the muscular layer, and the outer, hyperechoic serosa and periureteral fat. Phase II. RUP and ILUS demonstrated the existence of a ureteral stenosis in all animals. ILUS depicted a greatly diminished ureteral lumen with hyperechogenicity of al! ureterallayers, more evident at the muscle and serosa, indicating the existence of ureteral and periureteral fibrotic reaction and precluding discrimination of the ureteral layers. Phase III. An increase in the luminal diameter compared to the measurements made during the previous phase was seen in al! animals but 4. Ureterallayers could be clearly identified, as the fibrosis in the ureteral wall had d.isappeared. There was a rernnant of periureteral fibrosis around th.e lesion area. The 4 pigs with restenosis showed a ultrasonographic appearance similar to that seen in phase II.
Poster No. 246 Percutaneous Transgastric Pancreatic Pseudocyst Drainage. Pc. Shetty, Hurley Medical Center, Flint, MI, USA. U.B. Chintalapudi • D. Reddy • R. Sharma • A. Muniyappa PURPOSE: To assess the efficacy and safety ofPercutaneous Transgastric Pancreatic Pseudocyst Drainage under floroscopic and ultrasound guidance. MATERIALS AND METHODS: Fourteen pancreatic pseudocysts were drained using the percutaneous transgastric approach in 12 patients. Patients' age ranged from 13 to 72 years with a mean of 56 years. AJ] pseudocysts were diagnosed by CT. Anterior stornach wall was anchored to the anterior abdominal wall using two T-anchors. Stornach was punctured between the two T-anchors and pseudocyst was punctured through the posterior wall of the stornach. The tissues between the pseudocyst and the posterior stornach wall were dilated using 5mm diameter balioon catheter. Cyst was drained into the stornach using double pig tail stent in one case, extemal drainage pigtail catheter in another case and proximal cope loop catheter in the remaining 10 cases. Catheter was left in place for a mean of 22 days. A CT was performed when the catheter stopped draining, to check for any residual fluid collection. RESULTS: Thirteen cysts in 11 patients were successfully treated with this method. One patient had excessive bleeding through the drainage catheter. This patient was successfully treated surgically without any complication.
CONCLUSION: ILUS can be useful in depicting changes in the ureteral wal! and periureteral tissues secondary to stenosis. The technique can be used to monitor endourologic interventions, as it shows changes associated to ureteral fibrosis, specially at the muscular and periureteral, the most common causes of restenosis.
CONCLUSION: Percutaneous transgastric pancreatic pseudocyst drainage is an effective and safe procedure. In most patients this can avoid surgery and decrease the morbidity.
Poster No. 245
Poster No. 247
Single Step Transvaginal Aspiration and Drainage for Refractory Pelvic Abscesses. B. C. Lee, University oj Califomia, Davis Medical Center, Sacramento, CA, USA. J.P McGahan • B. Bijan
A ScJerotherapy of Symptomatic Hepatic or Rena! Cyst by Ethanolamine Oleate. K. Yamamoto, Nara Medical University, Kashihara, Japan • H. Sakaguchi. H. Anai • T. Tanaka • K. Kichikawa • H.
PURPOSE: For treatment of pelvic abscesses, use of the trocar technique avoids many of the technical challenges of the Seldinger method. The purpose of this study is to evaluate the effectiveness and safety ofultrasound-guided transvaginal aspiration or drainage with the trocar technique in pelvic abscesses that are refractory to antibiotic treatment.
Uchida, et al. PURPOSE: Large simple cyst that cause flank, back pain or deteriorating liver ar renal function require interventional therapy. Percutaneous puncture and aspiration or sclerotherapy are well performed and less invasive management options compared to surgical treatment. However, by using
S85
absolute alcohol as sclerosing agent, adverse effect which is a severe pain or feeling of intoxication is often appeared. The purpose of this study is to clarify the efficacy of ethanolarnine oleate(EO) as a sclerosing agent for symptomatic hepatic or renal cyst. MATERIALS AND METHODS: Seven patients with symptomatic liver or renal cyst were treated by sclerotherapy with EO. Four patients had renal cyst and 3 patients had hepatic cyst. The cyst size in great diameter was ranged from 6 to 12 cm and the average of its size was 8 cm in diameter. The cyst was punctured with 199auge needle under US guidance and 6.5F catheter was placed. After all of the cyst content was aspirated, an iodized contrast agent of a 10% volume of the cyst content was injected to check the absence of communication between the cyst and biliary tree or vessels and to test for the absence of leak into the peritoneum. And then, the solution of ethanolamine oleate-iopamidol mixture(EOl) was injected via catheter. After EOl injection, the patient was kept in supine, right lateral and left lateral position for every 10 minutes. After 30 minutes, the injected EOl was aspirated completely before catheter removal. A follow-up CT was performed at 3 month after treatment. The volume of cyst and its reduction rate was calculated. In addition, symptom and complication were assessed. RESULTS: The volume of cyst was ranged from 79 mI to 628 ml (mean;276ml) before treatment. After sclerotherpy, its volume was reduced at a month. And three months later after treatment, it was ranged from 2 ml to 61 mI (mean;24mI) and the reduction rate of cyst volume was more than 90% in average. Symptom caused by cyst was disappeared in all cases (717; 100%), and no major complication was encountered. Although 3 patients had a low grade fever after sclerotherapy, it was easily controlled. CONCLUSION: It is suggested that the sclerotherapy with EO might be asafe, effective and well-tolerated treatment for symptomatic hepatic or renal cyst.
Poster No. 248 Tangential Radial Sampling or Hepatic Lesions Via Double Coaxial (TRASH-DOC) Technique. B. Bijan, University oj California Davis Medical Center, Sacramento, CA, USA. R. Kazemaini • J.P. McGahan PURPOSE: Obtaining detailed information regarding the in vivo behavioral traits of a lesion is crucial in diagnosis and therapy planning: Our technique emphasizes the utilization of cross-sectional imaging modalities to obtain more informative sampIes. The information related to the lesional differential zones can be obtained by utilizing our sampling technique more reliably and equally safely. MATERIALS AND METHODS: Atotal of 23 consecutive lesions were biopsied using TRASH-DOC technique. US/CT were used for guidance. Single access port through Glisson was created via Double Coaxial Technique. 5 radial sampIes were obtained starting tangentially from the outermost normal liver parenchyma, continuing centrally, in attempt to obtain at least 2 out of 5 tumorai zones in each sample. A control CTIUS was used to rule out early complication prior to withdrawal of the needle. The blood patch, prepared from patient blood, was routinely applied. In coagulopathic cases, it was mixed with FFP or alternatively Gel foam slurry. Iodinated contrast was mixed to render radio-opacity, detectable by CT.
586
RESULTS: In our series, tangential approach provided a larger sample from each zone, with less contamination from the nearby zones. Providing this type of sampIes helped the
pathologist to study the tumor-liver interface in more details, which is vital for evaluation of tumoral behavioral traits, differentiation of locally invasive versus non-invasive lesions, and response to therapy. The procedure time and cost were increased slightly. However, considering the cost of resampling, using TRASH-DOC technique appears to be both cost and time efficient. No major complication leading to any additional hospital stay or additional intervention was noted. The sampIes were judged by our pathology department to be adequate in all cases. The reduced risk of complications, including track seeding, seen in this technique were analyzed. CONCLUSION: Using double coaxial technique with multiple radial sampling obtained tangentially from different zones with subsequent blood patching is a safe, accurate and effective mean of obtaining tissue for histological sampling of focal hepatic lesions.
Poster No. 249 Predictors or Infected Bile among Patients Undergoing Percutaneous Cholecystostomy. S. Beardsley, University ojPennsylvania, Philadelphia, PA, USA. T. WI. Clark PURPOSE:Patients may not achieve a clinical benefit after undergoing percutaneous cholecystostomy (PC) due to the inherent difficulty in identifying patients who truly have infected gallbladders. We sought to identify imaging and biochemical parameters which would help to predict which patients have infected gallbladders. MATERIALS AND METHODS: AlI patients who underwent PC over a 48-month period were retrospectively identified. Of these 78 patients, bile culture results from sampIes drawn at the time of PC. were available in 52 patients. We used logistic regression to assess whether the following variabIes were predictive of infected bile: white blood cell count, alkaline phosphatase, aspartate transferase, alanine transferase, bilirubin, gallstones (CTIUS), sludge (CTIUS), gallbladder (GB) wall :e:3 mm (CTIUS), pericholecystic fluid (CTIUS) and absence of GB visualization on radionucIide hepatobiliary scanning. RESULTS: Of the 52 patients, 25 (48%) had infected bile. Organisms cultured included Entercoccus, Enterobacter, Klebsiella, Pseudomonas, E. Coli, Citrobacter, and Candida. No biochemical parameters were significantly predictive of infected bile; WBC > 15,000 was weakly associated with greater odds of infected bile (OR=2, P=NS). The presence of gallstones, sludge, GB wall thickening and pericholecystic fluid by US or CT were not predictive of infected bile (P=N.S.), alone or in combination. RadionucIide scans were performed in 31 % of patients; all were positive and 66% of these patients had infected bile. Since no patients who underwent radionuclide scans had negative studies, this variable could not be entered into the model due to collinearity. CONCLUSION: No single CT or US imaging variable was predictive of infected bile, and onIy a weak association of WBC to infected bile was seen. No other biochemical parameters had any association with infected bile. The ability of radionucIide scanning to predict infected bile was higher than US or CT. This study illustrates the continued challenge to identify cholecystitis among patients referred for percutaneous cholecystostomy.
Poster No. 250 CHnical Suspicion of Cholangitis Is Only Predictor of Bacteremia after Drainage of Malignant Biliary Obstruction. P.M. Cote Robson, Menwrial Sloan-Kellering Cancer Center, New York, NY, USA. L. Brody. K. Brown. A. Covey. G. Getrajdman
PURPOSE: Bacteremia after percutaneous biliary drainage (PBD) is a potentially life-threatening event. Tbis study was performed to identify clinical factors predicting post-PBD bacterem.ia within 48 hours in a complex population of cancer patients. MATERIALS AND METHODS: Prospective cohort study of individuals undergoing PBD at Memorial Sloan-Kettering between 10/98 and 12/00. RESULTS: 278 PBD procedures were performed in 221 patients (:21 men, 100 women; mean age 62.8 years). Major underlying diagnoses were biliary tract malignancy (74; 33.5%), pancreatic/ampullary cancer (45; 20.4%), and metastatic disease from colorectal cancer (39; 17.6%), or other primary (48; 21.7%). Most common indications for PBD were suspected cholangitis (68; 35.1 %), pruritis (106; 38.1 %), or jaundice ?reventing chemotherapy treatrnent (78; 28.1 %). Multiple indications were reported for 56 (20.1 %) procedures. Blood cultures were drawn within 48 hours after 72/278 (25.9%)procedures to evaluate suspected post-PBD bacterem.ia. Post-procedure blood cultures were positive after 21 procedures (7.6% of total). No correlation was found between post-procedure bacterem.ia and age at procedure, underlying diagnosis, level of obstruction, type of procedure (R drainage vs L drainage, Ext vs IntlExt), prior PBD, presence of an indwelling stent or catheter, pre-procedure WBC, total bilirubin, alkaJine phosphatase, or AST/ALT. Prior instrumentation (120 procedures, 43.2%) was associated with positive bile cultures (P
PURPOSE: A specially designed self-expandable covered metallic stent incorporated with beta-em.itting radioisotope, Holm.ium-166 (Ho-166), was developed at our institute for the purpose of intraluminal brachytherapy as well as for internal bile drainage in malignant biliary stricture. This study was aimed to evaluate the safety of the radioactive metallic stent on the normal cartine common bile duct (CBD) prior to the clinical application. MATERIALS AND METHODS: Self-expandable stent (diameter; 6rnm, length; 20rnm) was made of nitinol wire. Ho166 was incorporated within polyurethane (50J..Lm thickness)
which was covered over the outer surface of m.id l cm of the stent and, both ends of stent were bare portion. The stent with radioactivity of 21-135 J..LCi ( mean; 77.9 J..LCi) were placed in the CBD of 20 heal thy beagle dogs. The dogs were sacrificed 3-6 months after stent insertion and histopathologic examination of CBD was performed.
RESULTS: There was no stent migration or stent ocelusion in any cases. Grossly, stricture or perforation was not seen in all cases. Varying degrees of papalliary mucosal hyperplasia was observed at uncovered area. Fibrosis was noted in Ho-166 coated area, instead of mucosal hyperplasia. lnf1arnmation was demonstrated, but dyplasia or ulceration was not in all cases. The membrane of Ho-166 coat was intact in all cases. CONCLUSJON: These studies suggest that radioactive covered self-expandable metal lic stent delivering therapeutic radiation dose can be used and it can be an alternative therapeutic modality for palliative treatment of the biliary tract malignancy. Poster No. 252 Factors Associated with Mortality in Patients Undergoing Percutaneous Cholecystostomy. B.D. Davison, Brigham and Womens Hospital, Boston, MA, USA. l. F. Polak
PURPOSE: While cholecystectomy has been shown to be an effective treatrnent for acute and chronic cholecystitis it appears to be largely reserved for the most critically ill patients. In this population it is associated with a high mortality rate, and is often requested without elear clinical criteria. We looked at our own experience in a tertiary care hospital to evaluate the predictors of short and long-term mortalityof percutaneous cholecystostomy in our patient population. MATERIALS AND METHODS: The medical records of 81 consecutive patients in whom cholecystostomy was performed between July 1995 and July 2001 were retrospectively reviewed. Biochernical variabIes and chnical outcome were determ.ined by chart review. Analyses were conducted with Cox proportional hazrds models with mortality as the primary outcome. RESULTS: Percutaneous cholecystostomy was technically successful in 81 (100%). Overall mortality at 30 days was 2% (first mortality recorded at 18 days and not related to the procedure). By l year, mortality was 57% and reached 70% at 2 years. Factors associated with mortality (after controlling for age and gender, were elevated tota! bilirubin (hazard ration 1.18 for every l mg/dl increase) and elevated AST (HR of 1.01 for every l lU increase). Lawer risk of mortality was seen when the direct bilirubin was also elevated. CONCLUSION: Percutaneous cholecystostomy in the treatment of acute cholecystitis in high-surgical-risk patients is a relatively safe procedure. Otber factors related to patient co-morbidities account for a bigh mortality in the days and months following the procedure, at least in tertiary care institutions.
S87
Poster No. 253 The Cecostomy Butlon- An Alternative to Chronic Cholecystostomy Drains. M.K Varma, Tripler Army Medical Center, Honolulu, HI, USA • A. W Allen. M. Meyermann PURPOSE: Percutaneous cholecystostomy is a safe altemative in critically ill patients with cholecystitis and serves as a conduit to cholecystectomy in stabilized patients with calculous cholecystitis. However. there is a subset of patients in which medical co-morbidities make cholecystectomy prohibitive. These patients must then łive with an external drain, which often have a prominent extemal profile and high patient dissatisfaction. The aim of this study was to determine if the Chait Trapdoor Cecostomy Button could be used as an alternative to the standard pigtail cholecystostomy tube. MATERJALS AND METHODS: At our institution, 4 patients in the last 24 months were diagnosed with calculous cholecystitis who were not operative candidates. Intially. 10 FR pigtail, ultrasound-guided percutan,eous cholecystostomy tubes were placed (transhepatic route-3 patients; transperitoneal route- I patient). On standard follow-up in 812 weeks, a 10.2 FR Chait Trapdoor Cecostomy Button catheter was płaced in all 4 patients. We then folIowed the patients with standard tube changes every 8-12 weeks with cholangiograms and patient interview (length of follow-up: range 8-26 mos, mean 14.7 mos).
Magnetic fiducials were embedded in the stylet of an 18gauge needle or 8French catheter. Positional data were sampled at a rate of 10 Hz. Axial CT scans of the phantom were obtained after płacement offour external fiducials·and a single internal fiducial (sensor-containing catheteror needle) within the foam model. Following registration, needle trajectories and suxface puncture sites were identified on MPR images. The goal was successful simultaneous puncture of two vessels. Punctures were performed by a single experienced and a single inexperienced practitioner. The needle placement time, tipto-target distance (magnetic transformation), and tip-to-vessel wall distance (from biplane radiographic images) were measured. RESULTS: Successful punctures of 5mm vessels by an experienced operator occurred in 16/26 total attempts. Fewer successful punctures occurred at a higher respiratory rate. Successful punctlJres were achieved by the experienced and inexperienced operators in 7114 and 2113 sequentiał attempts respectively. Depth errors occurred in 13118 missed punctures. Punctures of single superficial3mm and 5mm vessels resulted in success in 15/20 and 27127 attempts respectively. CONCLUSION: Factors which appear to contribute to errant magnetic guided punctures include smaller vessel diameter, depth of the target vessel, higher respiratory rate, and operator experience. In most cases the errant passes lie within one vessel diameter of the target. The system continues to show promise in an in vitro testbed. ORBH.::I
RESULTS: Teaching Points: 1. In patients with chronic cholecystostomy tubes with calcułous cholecystitis who are not candidates for cholecystectomy, a Chait, Trapdoor Cecostomy Button is better tołerated because of its lower external profile and an equally effective alternative to standard pigtail catheter drainage. We recommend placing this catheter after a mature tract has formed with an initial pigtail cholecystostomy tube placement. 2. We have found that the cecostomy catheter placement was successful in trans-hepatic but not· in trans-peritoneal placement. The trans-peritoneal route was associated with excessive leakage around the catheter. With a shorter route to the gallbladder transperitoneally and a more mature tract, a cecostomy catheter may be feasible. 3. The standard tract length for insertion of the cecostomy catheter is recommended by the manufacturer to be less than 6 cm. Because of the helicał extensile coils distally on the catheter, a tract length of up to 9 cm was used without dislodgement transhepatically.
Poster No. 254 Respiratory-Gated Magnetic Tracking-Guided Płacement System: Work in Progress.
Needłe
E.B. Levy, Georgetown University Hospitał, Washington, DC, USA • J. Tang • D. Lindisch • K Cleary PURPOSE: Magnetic track.ing of needles and catheters holds promise as a method to accurately target interventions and therapies without the use of ionizing radiation. We report our progress with the Aurora magnetic tracking guidance system with respiratory gating in an vitro model.
S88
MATERIALS AND METHODS: A previously tested abdominal torso containing a foam hepatic model mounted upon a servomotor platform was used. Respiratory motion was simulated by to-fro movement of the servomotor platform at 12-20 cycles per minute with an excursion distance of 12cm in the y-axis. The Aurora magnetic tracking system includes a field generator and up to two receiving channels.
Poster No. 255 Safety, Diagnostic Accuracy and Limitation oC CT-Guided Coaxial Cutting Needle Biopsies oC Lung Lesions. KM. Yeow, Chang Gung Memoriał Hospital, Kwei Shan, Tao Yuan, Taiwan, R.O.C • KP. Tsay • KW Lui • Cheung • K T Pan. A.S.B. Chou
Y.e.
PURPOSE: To evaluate the diagnostic accuracy, safety and limitation of er-guided coaxiał cutting needle biopsies of lung lesions. MATERIALSANDMETHODS: FromMarch 1995 to August 2001, a consecutive 631 biopsies with finał diagnoses confirmed either by surgery, imaging and clinica1follow up of at least 24 months were analyzed. The sensitivity, specificity. negative predictive value, positive predictive value, false positive rate, false negative rate and overall diagnostic accuracy were calculated. Complications presenting as pneumothorax, chest tube insertion rate and hemoptysis were obtained. The criticallesion size that caused significant drop in diagnostic accuracy and accompanied by rising complication rates was determined. P:S:;.05 was taken as statisticał significant. RESULTS: The overall diagnostic accuracy was 95 % (95 % CI 92.7-96.2) (Sensitivity =93 %; Specificity =98 %; Negative predictive value = 86 %; Positive predictive value = 99 %; False positive rate =0.7 %; False negative rate = 15 %). The pneumothorax rate was 23 % (155 of660), chest tube insertion rate was l % (9 of 660, 5 moderate pneumothoraces were successfully needle aspirated) and the hemoptysis rate was 4 % (26 of 660). N o mortality was encountered. For lesions:S:;1.5cm diagnostic accuracy dropped to 84%, bleeding rate increased by 12-times and hemoptysis rate rose from 4% to 8%. CONCLUSION: Cf-guided coaxiał cutting needle lung biopsy is highły accurate both for benign (86 %) and malignant (95 %) łung lesions and safe (23 % pneumothorax rate, l % chest tube insertion rate and 4 % hemoptysis rate). The technique is limited by lesion size:S:;1.5cm.
Poster No. 256 Complications of Thoracic Interventions: Evaluation of Contributing Factors. TB. Kinney, UCSD MedicalCenter; Dept oj Radiology, San Diego, CA, USA. H. T Chen • S. C. Rose • K. Valji • A. C. Roberts PURPOSE: Common thoracic interventions include thoracentesis, tube thoracostomy, and percutaneous lung biopsy. The rationale for using image guidance in thoracic intervention is to reduce such complications as local hemorrhage, hemothorax, and pneumothorax. aur goal was to assess the factors which contributed to complications of thoracic interventions performed at our institution. MATERIALS AND METHODS: Chart review of 416 thoracentesis, tube thoracostomy, and percutaneous lung biopsy procedures was undertaken. These procedures were performed at three different hospitals by a group of 6 interventionaJ radiologists between 1998 and 2002. The imaging modalities used included ultrasound, CT, and f1uoroscopy. RESULTS: Atotal of 61 (14.7 %) complications occurred, including 43 (l0.3 %) pneumothoraces; 7 (1.7 %) hydropneumothoraces; 8 (1.9 %) localized hematomas; 2 (0.5 %) fatal hemothoraces; and 1 (0.2 %) hemopericardium. Other incidents included l episode each of vasovagal syncope and hemoptysis. One of the fatal hemothoraces occurred secondary to diagnostic thoracentesis perforrned portably in the ICU with ultrasound using an 18 G DSA needle. The otheroccurred upon chest tube removal (8 Fr locking pigtail catheter) in a patiem with critical aonic stenosis who was on anticoagulation therapy. Of the pneumothoraces, 41 occurred in lung biopsies while 2 occurred with thoracenteses. Of the hydropneumothoraces, 6 occurred with thoracentesis while one occured on removal of a thoracostomy tube. CONCLUSION: Chest interventions are associated with a relatively smali number of complications. aur most frequent complication was pneumothorax folIowed by hydropneumothorax. Bleeding complications while rare can be fatal in certain cases. Unlike other areas of the body, the pleural space is subjected to vacuum and can accumulate significant amounts of blood. Portable intervention is associated with technical compromises compared to that used in mobile patients within an interventional suite and these may influence complications. Precautions with chest tube removaJ include stopping anticoagulation and careful technique.
Poster No. 257 CT-Guided Biopsy of the Lung on an Outpatient Basis: Safety and Efficacy. T lhaya. Tollori Red Cross Hospital, Tollori, Japan • J. Nakanishi • T Ookawa PURPOSE: To evaluate the safety and efficacy ofCT-guided percutaneous lung biopsies of pulinonary lesions on an outpatiem basis. MATERIALS AND METHODS: Atotal of 279 biopsies of the lung were performed under CT guidance on an outpatiem basis using 21-gauge fine needles or 20-gauge cutting needles. A chest radiograph was taken just after careful observation for two llOurs in the outpatient clinic and asymptomatic patients were allowed to go home with appropriate instnlctions for any complications.
RESULTS: The definitive diagnosis was malignant in 152 palients (54.5%) and benign in 122 (43.7%). In lhilteen cases (4.7%), no adequate sample was obtained to perform a cytological or histological diagnosis. Pneumothoraces were observed in 101 patients (36.2%) on CT scan just after the procedure and in 68 patients (24.4%) on chest radiograph two hours after the biopsy. None of them had any symptoms and required an insertion of chest tube. Minor hemoptyses were noted in five patients (1.8%) but it disappeared within two hours after the procedure. OnIy two patiems (0.7%) developed pneumothoraces with breathlessness more than 24 hours after the procedure and they required in-patient treatment in both cases. There was no significant difference in the complication rate between fine needles and cutting needJes. Total cost of CT-guided biopsy on an outpatient basis was just half of in-patient biopsy. CONCLUSlON: CT-guided needle biopsy is a safe and reliable procedure and it can be performed on an outpatient basis with a low complication rate and it also provides high cost effecliveness.
Oncologic Access
Interventions/Venous Poster No. 258
Evaluation of the Celi Death after Radiofrequency Ablation in the Dog Liver. H. Rikimaru, Depl oj Diagnostic Radiology, Tohoku University, Sendai City, Miyagi Prejeclure, Japan • T Matsuhashi • H. Miyachi • H. Saito • T lshibashi • S. Takahashi, et al.
PURPOSE: To clarify whether Hernatoxylin and eosin (H&E) stain can clearly distinguish the ablated area from the normal parenchyma after radiofrequency ablation. MATERIALS AND METHODS: RFA was appliedto 23 different lobes of nine adult female beagle dogs' liver. The 17gauge monopolar needle electrode with a 2cm-long tip was used. The power output was set at 30W for 6min. The dogs were killed immediately, 2hrs, 24hrs, and 3days after RFA application. H&E stai n, Terminal deoxynucleotidyl transferase UTP nickend labeling (TUNEL) stain, and Hypoxia inducible factor (HIF)-l ex. stain were perforrned, and they correlated one another. RESULTS: Twenty-three well-demarcated circumscribed ablated lesions were created. They consisted of 3 layers on H&E stain immediately after RPA, which turned to 4 layers after 2hours, and finally to 5 layers within 3 days. At the central areas (layer l and 2) of the ablated sites, celi features did not change over the time course, while some nucleus disappeared after 3days. At the peripheral region (layer 3 and 4), on the other hand, hepatocyte shrinkage, nuclear size reduction, nuclear condensation progressed within 24hrs. After 24hours, bothTUNEL and HIF-lex. stain became positive in every cel l of the peripheral region, which correlated very well with the areas identified by H&E stain (layer 3 and 4). CONCLUSlON: The central area of the ablated sites after RPA appeared to be coagulation necrosis, while the peripheral area showed the features of apoptosis. The peripheral area also proved to be ischemic. The H&E stain could clearly identify ablated areas from the nonnal parenchyma.
S89