T1436: EUS-Guided Gallbladder Drainage for Treatment of Acute Cholecystitis and Obstructive Jaundice

Abstracts was made a telephone interview determined in detail with management to the patients before EUS-FNA and after the results of EUS-FNA another interview was done to determine if the result of EUS-FNA had changed the management of patient. Results: In this period were consecutively recruited 302 patients with PCLs. 159 patients (53%) were asymptomatic. The identification of IPCs was made after control of genitourinary problems (26%) and evaluation of the elevated liver function tests (18%). The mean age was 55 yrs (range 27-84 yrs), 118 (74%) were female. 110 (71%) had cysts smaller than 3.0 cm. The average size was 2.3 cm (range 0.2-7.1cm). 95 (61%) were located in the head and 10 (6%) had IPCs in more than 1 segment of the pancreas. 91 had their final diagnosis obtained by EUS-FNA while 68 were operated. The diagnoses of the patients were as follows: Benign lesions were found in 93 (58%) patients, malignant potential in 36 (23%), non-invasive malignancies in 14 (9%), malignant precursor lesion (PanIN) in 10 (6%) and invasive malignancies in 6 (4 %). 114 of 159 (71.7%) patients were significantly altered management after EUS-FNA. The management change was as follow: 33 (47%) patients were referred for surgery, 70 (44%) patients were “discharged” in this cohort because they did not need to control after the result of EUS-FNA and 11 (6.9%) patients were given the followthrough imaging tests. There were 3 minor complications after EUS-FNA. Conclusion: EUS-FNA is a safe and well tolerated technique for the characterization and management of the incidental pancreatic cysts. This prospective study shows that EUS-FNA has a significant clinical impact with a change in management of over 50% of patients with IPCs.

T1433 MRCP vs. EUS-FNA: What Is the Best Way to Diagnosis and Classification (Type and Extent) of IPMT? Jose´ C. Ardengh, Giuseppe D Ipolitto, Djalma E. Coelho Neto, Jose´ Fla´vio E. Coelho, Eder R. Lima, Jose´ Luiz P. Mo´dena, Jose Sebastiao S. Dos Santos Background: Intraductal papillary mucinous tumors of the pancreas are being diagnosed with increasing frequency. CT scanning commonly serves as the primary imaging modality before surgery. The MRCP provides better characterization of IPMT type/extent, which more closely matches actual pathology. The endosonography-guided fine needle aspiration (EUS-FNA) has been increasingly used in order to make the histological diagnosis preoperatively, and to determine the type and extent of this tumor. The aim of this study is to compare the results of MRCP and EUS-FNA for the diagnosis and classification of IPMT with surgical and pathological findings. Methods: 218 patients were referred consecutively for EUS-FNA for evaluation of PCLs between Aug/2002 the set/2008 at a Tertiary Referral Centre (Hospital 9 de Julho). Of 96 patients treated with mucinous cystic tumor (MCT), 26 had both preoperative MRCP and EUS-FNA. All patients met imaging study criteria. Independent readers performed retrospective, blinded analyses using standardized criteria for IPMT type and extent. Results: Thirteen patients underwent pancreatoduodenectomy, 11 a subtotal pancreatectomy with splenectomy and 2 to laparotomy. Pathological examination revealed the presence of IPMT in 24 patients. Twentyone had a non-invasive (12 adenoma, 7 borderline and 2 carcinoma in situ) and invasive form in 3 patients. The MRCP and EUS have made the correct diagnosis, classified, located the damaged segment and identified the presence of nodules and/or vegetation at 62.5% and 71%, 80.7% and 96.1%, 80.7% and 100%, and 45.4 and 100% (p ⫽ 0326), respectively. The sensitivity, specificity, PPV, NPV and accuracy of EUS-FNA for histological diagnosis of IPMT was 80%, 100%, 80%, 100% and 93.3% respectively with kappa ⫽ 0.87 (optimum). Conclusions: The EUS and MRCP have no statistical differences for the correct diagnosis, classification and location of the affected pancreatic segment. Moreover the EUS shows absolute results better than the CPRM to identify the presence of nodules and / or vegetation. EUS-FNA obtains the diagnosis, minimizes the doubts of the MRCP, and should be employed for optimal management of patients with IPMT.

T1434 Endoscopic Ultrasonography (EUS) Drainage With SelfExpandable Metallic Stent (SEMS) in Pancreatic Pseudocyst (PP) and Peri-Pancreatic Abscess (PPA): A Prospective Study Carlo Fabbri, Carmelo Luigiano, Anna M. Polifemo, Francesco Ferrara, Sandro Macchia, Stefania Ghersi, Marco Bassi, Paola Billi, Vincenzo Cennamo, Lorenzo Fuccio, Michele Masetti, Elio Jovine, Nicola D’Imperio Introduction: EUS-guided PP and PPA drainage has become the standard and safer procedure in many centers but often requires repeated sessions, multiple stents placement, naso-cystic catheter placement and prolonged hospital stay. Aim: To evaluate feasibility, safety and outcomes of temporary SEMS placement for the treatment of PP and PA. Material and Methods: From June 2008 to July 2009 all patients with a CT diagnosis and with symptomatic PP and PPA were included in the study. All patients underwent EUS-guided puncture followed by cysto-gastrostomy and placement of a 10x40mm partially covered Wallflex biliary stent (Boston Scientific) or a fully covered Niti-S biliary stent (Taewoong).

AB276 GASTROINTESTINAL ENDOSCOPY

Volume 71, No. 5 : 2010

Results: Ten patients (7 M and 3 F; mean age 55.9-12.3) with 7 PP and 3 PPA were recruited. Mean lesions size was 11.6-3.9 cm (range: 8-20 cm). The procedure was technically feasible in 9 out of 10 patients (90%): in one patient the large PP secondary to acute pancreatitis presented multiple spontaneous enteric fistulization. In 6 patients a partially covered stent was placed, while in the other 3 the fully covered stent. Transgastric and transduodenal approach were conducted in 6 and 3 cases, respectively. There were no procedure-related complications, but one patients with decompensated alcoholic cirrhosis and chronic pancreatitis enrolled for a PPA died 10 hours after the procedure for acute respiratory failure. In the remaining 8 patients PP and PA resolved without additional intervention after a median of 25.7-3.1 days, documented by CT and EUS. SEMS were removed without difficulty in 7 patients, while in one patient a partially covered stent with flogistic tissue ingrowth was impossible to remove endoscopically; this stent was removed during a surgical intervention for a renal disease. No PP and PPA recurrence was observed during the follow-up period (mean 212-75.4 days). Conclusions: Endoscopic SEMS for PP and PPA is a safe and efficient endoscopic treatment. Higher cost of SEMS is probably offset by savings related to reduced number of procedures and hospital stay costs.

T1435 EUS-Guided Biliary Drainage for Malignant Biliary Ducts Obstruction: A Prospective Study Carlo Fabbri, Carmelo Luigiano, Anna Maria Polifemo, Francesco Ferrara, Sandro Macchia, Stefania Ghersi, Marco Bassi, Paola Billi, Vincenzo Cennamo, Lorenzo Fuccio, Michele Masetti, Elio Jovine, Nicola D’Imperio Background and aim: Endoscopic ultrasonography-guided biliary drainage (EUSBD) has been developed as an alternative drainage method in patients with obstructive jaundice after endoscopic retrograde cholangiopancreatography (ERCP) failure. Aim of this study is to evaluate the potential role of EUS-BD in patients with bilio-pancreatic malignancy after ERCP failure.Methods: From July 2008 to September 2009 all patients with bilio-pancreatic malignancy, candidates to alternative techniques of biliary decompression because of unsuccessful ERCP, were enrolled in the study and underwent to EUS-BD. All procedure details, technical success, clinical success, complications rate and follow-up findings were evaluated.Results: 13 patients (7 men and 6 women - mean age 79.1⫾9.1) were enrolled. EUS-BD was successfully performed in 9 out of 13 (69.3%) patients (7 biliary-duodenostomy with stent placement and 2 biliary rendez-vous with papillary stent placement) and a regression of cholestasis was obtained in all patients. No major complications and no procedure-related deaths occurred. Patients were followed-up for 122⫾60 days. During the follow-up, 5 patients died while 4 are actually alive and none of the patients required endoscopic reintervention.Conclusions: EUS-BD may be a useful procedure for treatment of obstructive jaundice when ERCP fails, but require further assessment in a larger cohort of patients, including comparative analysis versus percutaneous transhepatic biliary drainage. Dedicated devices for EUS-BD are needed for reliable procedures.

T1436 EUS-Guided Gallbladder Drainage for Treatment of Acute Cholecystitis and Obstructive Jaundice Masayuki Kitano, Hajime Imai, Takamitsu Komaki, Ken Kamata, Hiroki Sakamoto, Masatoshi Kudo BACKGROUND AND AIM: Percutaneous transhepatic gallbladder drainage, which involves an external drainage tube, decreases the ability of the patient to carry out their normal daily activities. We performed a new internal drainage technique, EUS-guided gallbladder drainage (EUS-GBD), for decompression of acute cholecystitis and obstructive jaundice, and evaluated its feasibility and safety. PATIENTS AND METHODS: A total of 4 patients underwent EUS-GBD for treatment of acute cholecystitis caused by malignant obstruction of cystic duct (n⫽3) and obstructive jaundice with duodenal stenosis due to pancreatic carcinoma (n⫽1). Transpapillary drainage in those patients had been unsuccessful. An echoendoscope was introduced into the stomach. The EUS images clearly visualized a swollen gallbladder adjacent to the antrum. The echoendoscope was manipulated in order to identify an appropriate puncture route that has no interposing vessels. A 19-gauge needle was used to puncture the gallbladder. After the gallbladder was irrigated by saline, a 0.035-inch guide wire was introduced into the gallbladder (Fig.). When the puncture hole by biliary dilation catheters through the guide wire was successfully dilated, a double-pigtail stent was deployed between the gallbladder and the stomach. Feasibility and safety of EUS-GBD were evaluated. RESULTS: The gallbladder was successfully punctured by the needle and irrigated in all patients. However, dilation of the punctured hole was difficult in 2 of 4 patients. Consequently, those 2 patients were followed up after a single aspiration. The remaining 2 patients underwent stenting between the gallbladder and the stomach after dilation of the punctured hole. EUS-GBD relieved symptoms without any apparent complications in all patients. Recurrence of cholecystitis was observed

www.giejournal.org

Abstracts in a patient treated with a single aspiration, while no additional intervention was needed in the other 3 patients. CONCLUSION: This report of case series indicates that EUS-guided gallbladder drainage is an alternative route for decompression of the biliary system when ERCP is unsuccessful.

Figure. EUS-guided gallbladder drainage

T1437 EUS-Guided Choledochoduodenostomy Followed by Endoscopic Antegrade Biliary Stenting via the Fistula for Treatment of Obstructive Jaundice With Duodenal Stenosis Masayuki Kitano, Takamitsu Komaki, Hiroki Sakamoto, Ken Kamata, Hajime Imai, Masatoshi Kudo BACKGROUND AND AIM: EUS-guided choledochoduodenostomy (EUS-CDS) with insertion of a stent from the duodenal bulb to the extrahepatic bile duct is a possible drainage route in the treatment of patients with obstructive jaundice, when ERCP is unsuccessful. However, stagnant gastroduodenal juice sometimes causes reflux cholangitis through the inserted stent in case that outlet of the gastroduodenal juice is disturbed by stenosis of the duodenum. In those patients, subsequent biliary drainage toward the anal side of the stenosis is desired to prevent the bile duct from reflux of the stagnant fluid. We evaluated the utility of the endoscopic antegrade biliary stenting (EABS) via the fistula for refractory reflux cholangitis. PATIENS AND METHODS: Between Sep. 2006 and Nov. 2009, a total of 19 patients with obstructive jaundice underwent EUS-CDS because ERCP was unsuccessful. An echoendoscope was inserted into the duodenal bulb to puncture the extrahepatic bile duct by a 19-gauge needle. After insertion of a 0.035-inch guide wire, the puncture hole was dilated by biliary dilation catheters. Finally, a straight plastic stent (7 Fr.) was inserted through the puncture hole in order to make choledochoduodenostomy. In patients complicated with recurrent or sustained cholangitis after EUS-CDS, EABS via the fistula was performed. With a duodenoscope inserted into the duodenal bulb, the plastic stent deployed for choledochoduodenostomy was removed. A catheter with a 0.035-inch guide wire was inserted from the fistula toward the papilla. After passing the biliary stenosis and the papilla, they were introduced into the duodenal lumen. Subsequently, a non-covered metallic stent was deployed at the biliary stenosis in order that the proximal side of the stent passed the fistula into the bile duct. Feasibility and safety of EABS was evaluated. RESULTS: Deformity or stenosis of the duodenum had been observed in 16 of 19 patients treated with EUS-CDS. Among those 16 patients, 5 patients underwent EABS 17-115 days after the procedure because of recurrent or sustained cholangitis. The major symptoms of reflux cholangitis (high-grade fever, jaundice and abdominal pain) were relieved after the EABS in all patients. Alkaline phosphatase significantly decreased from 969⫾187 to 556⫾136 IU/L by 1 week after EABS. No additional intervention was needed in all patients. No major complication was observed after EABS. CONCLUSION: EUS-guided choledochoduodenostomy followed by endoscopic antegrade biliary stenting via the fistula is a possible treatment option for obstructive jaundice with duodenal stenosis.

T1438 Evaluation of a Novel Method for Measuring CEA Levels From Pancreas Cyst Aspirates Vinay K. Katukuri, Darren Andrade, David E. Loren, Thomas E. Kowalski Cyst fluid carcinoembroyonic antigen (CEA) and amylase have been shown to help to discriminate neoplastic from benign pancreatic cysts. Often this evaluation is limited by the inability to obtain adequate fluid for chemical analysis either due to high fluid viscosity or due to the limited volume of fluid in a small or septated cyst. A novel measurement method has recently become commercially available for measurement of CEA (RedPath Integrated Pathology, Inc., Pittsburgh, PA) that requires 75␮l of fluid compared to the amount previously required (1ml). The performance characteristics of this test have not been validated in the clinical setting. Aims: To examine the yield and diagnostic accuracy of CEA measurement of the commercial test compared to a standard academic institutional laboratory.Methods: We prospectively collected aspirates of pancreatic cyst of consecutive patients undergoing endoscopic ultrasound (EUS) for pancreatic cystic lesions. Fine needle aspiration (FNA) was performed with a 22g needle. If more than 1ml of fluid was available, the fluid was sent preferentially to our institutional laboratory, and when there was excess fluid, the remaining available fluid was sent to the commercial laboratory. When less

www.giejournal.org

than 1ml of fluid was available the entire specimen was sent to the commercial facility. Results: 13 patients underwent EUS FNA during the study period, 8 male and 5 female, mean age 71.5 and 63.6 years, respectively. Median and Mean fluid volumes were 2.5 and 5.3ml. CEA levels were available for all 13 patient samples sent to the commercial facility. The lowest volume from the commercial lab that yielded a CEA value was 50␮l. In 2/13 (15%) there was insufficient fluid for CEA analysis in the standard laboratory, despite receiving the requisite 1ml. The Pearson correlation between CEA measurements was 0.957, demonstrating an excellent agreement. When stratified by levels greater than 192ng/ml, 192⬎CEA⬎5 or 5⬍CEA, there was discordance in only 1/13 patients. Conclusions: The novel commercial method of cyst fluid analysis allows for accurate measurement of cyst fluid CEA even on cyst fluid aspirates of less than 1ml, and potentially less than 100␮l of fluid. This measurement tool increases the yield of EUS FNA for pancreatic cysts, particularly for those in whom cyst fluid volumes are small.

T1439 Comparision of Linear EUS and ERCP for Patients With Clinically Suspicious Bile Duct Stones That Not Detected in the Abdominal CT Tae Hyeon Kim, Bong Joon Yang, Yong Hwan Ahn, Hyo Jeong Oh, Yong Leol Oh, Young Woo Sohn BACKGROUNDS/AIMS: Endoscopic retrograde cholangiopancreatography (ERCP) is being replaced by endoscopic ultrasonography (EUS) for diagnostic purposes in the pancreatobiliary disease, and it should be reserved for therapeutic indication. We investigated diagnostic accuracy, complications and benefits of an EUS-guided approach versus an ERCP-guided approach in patients with suspicious bile duct stones who were not diagnosed by CT scan. METHODS: We enrolled patients with suspicious bile duct stones, in whom a multi-detector abdominal CT scan is not diagnostic at Wonkwang university hospital from May 2007 to March 2009. These patients were prospectively randomized to either a linear EUS first, ERCP-second (n⫽54) or ERCP-only (n⫽52) procedure. ERCP were perfomed within 2 days if the common bile duct (CBD) stones were detected by a linear EUS. Patients without stones who underwent EUS were followed up over 6 months. RESULTS: There were no significant differences in clinical features, Barkun score and demographic characteristics. All EUS and ERCP were successful. 21 (38.8%) patients in the EUS group were found to have CBD stones that were all treated by ERCP with endoscopic sphincterotomy. In the ERCP group, 18 (34.6%) patients had CBD stones that were removed in the same method. The sensitivity and specificity of ERC in the ERCP group were 72.2% and 97.1%, respectively. The sensitivity and specificity of EUS were 95.2% and 97%, respectively. The post-ERCP pancreatitis rate was 5 of 52 (9.6%) in ERCP group and 1 of 54 (1.9%) in EUS group, but there was not significant difference. With EUS guided approach, diagnostic ERCP and its related complications could be spared in 67 (63.2%) patients. The independent factor predicting CBD stones was Barkun score(⬎ 67%) at the multivariable analysis. CONCLUSIONS: In patients with suspicious bile duct stones that did not detected in the abdominal CT, EUS-guided approach may be a safe and valuable strategy to select patients for therapeutic ERCP.

T1440 A Prospective, Randomized Trial Comparing the Diagnostic Yield of the 22-Gauge and 25-Gauge Needles During EUS-FNA of All Solid Lesions Jason Conway, Sarba Kundu, John A. Evans, Cynthia L. Cookingham, Kim Geisinger, Girish Mishra Background: Many endosonographers prefer the 22-gauge (g) needle when performing EUS-FNA. This may be due to more experience with the 22g needle or the belief that a larger needle will obtain more tissue and increase diagnostic yield. Aim: We compared the diagnostic yield of the 22g and 25g needles during EUS-FNA of all solid lesions. Methods: This interim analysis of an ongoing prospective study included patients undergoing EUS-FNA of solid lesions from January 1, 2009 through November 1, 2009. All lesions underwent 4 passes (2 with 22g needle, 2 with 25g needle) using the EchoTip Ultra needle (Cook Medical, Winston-Salem, NC). Ten cc of suction was used during each pass. The needle used first (either 22g or 25 g) was determined by block randomization. Both the on-site cytotechnician and the cytopathologist were blinded to the sampling technique. Data on gross appearance, cellularity and diagnosis were recorded. Diagnostic yield was defined as the acquisition of adequate cells for the pathologist to render a diagnosis. Results: Data from 75 patients were included in this interval analysis. The mean age was 66 years, 62% were male, and 83% were Caucasian. There was no difference in gender or ethnicity distribution in either needle group. Patients who had the 22g needle pass first were slightly older(69 yrs.) compared to 25g (62 yrs.) with p⫽0.02. Pancreatic lesions, 37 (49%) and lymph nodes, 17 (23%) comprised the majority of the lesions biopsied. Average lesion diameter was 31 mm. Common cytology diagnoses included: adenocarcinoma 31 (41%), reactive adenopathy 13 (17%),

Volume 71, No. 5 : 2010 GASTROINTESTINAL ENDOSCOPY AB277