Search

Background and study aims Endoscopic ultrasound (EUS)-guided fine-needle aspiration (FNA) performs poorly in the histological diagnosis of type 1 autoimmune pancreatitis (AIP). The aim of this study was to assess the performance of fine-needle biopsy (FNB) comparing reverse bevel (RB) and fork-tip (FT) needles. Patients and methods A retrospective study of prospectively maintained databases was performed. Patients with a final diagnosis of type 1 AIP who underwent EUS-FNB during diagnostic workup were included. Pathology reports were reviewed and classified as per international consensus diagnostic criteria (ICDC). The Primary outcome was EUS-FNB sensitivity in diagnosing type 1 AIP. Results Between March 2011 and December 2018, 24 patients with a final diagnosis of type 1 AIP underwent FNB. Six patients underwent biopsy with the RB needle and 18 with the FT needle. Mean age (± SD) 62.2 (± 11.4), 17 (70.8 %) male. No RB samples were diagnostic compared to 14 (78 %) FT; P = 0.001; of which 13 (72 %) were level 1. In eight (44 %) of FT cases a diagnosis was not possible without histology. Initial biopsy was diagnostic in five (62.5 %) of these cases. Including repeat biopsy, seven (87 %) had a diagnosis made by FT needle. Obliterative phlebitis (44 %) was the least frequently identified pathological feature and immunoglobulin (IgG)4 + plasma cells > 10 per high power field (78 %) the most common. Conclusion The FT needle demonstrated good performance for diagnosing type 1 AIP. The results support the preferential use of this core biopsy needle for EUS pancreatic tissue sampling.

Background A novel fork-tip fine-needle biopsy (FNB) needle has recently been introduced for endoscopic ultrasound (EUS)-guided sampling. The aim of this study was to compare the performance of fork-tip FNB histology and standard fine-needle aspiration (FNA) cytology in the diagnosis of solid pancreatic masses. Methods A randomized crossover study was performed in patients referred for EUS-guided sampling. Three passes were taken with each needle in a randomized order. Only samples reported as diagnostic of malignancy were considered positive. The primary end point was the sensitivity of diagnosis of malignancy. Secondary end points included the amount of sample obtained, ease of diagnosis, duration of tissue sampling, pathologist viewing time, and cost. Results 108 patients were recruited. Median age was 69 years (range 30 – 87) and 57 were male; 85.2 % had a final diagnosis of malignancy. There were statistically significant differences in sensitivity (82 % [95 % confidence interval (CI) 72 % to 89 %] vs. 71 % [95 %CI 60 % to 80 %]), accuracy (84 % [95 %CI 76 % to 91 %] vs. 75 % [95 %CI 66 % to 83 %]), proportion graded as a straightforward diagnosis (69 % [95 %CI 60 % to 78 %] vs. 51 % [95 %CI 41 % to 61 %]), and median pathology viewing time (188 vs. 332 seconds) (P Conclusion The diagnostic performance of the fork-tip FNB needle was significantly better than that of FNA; it was associated with ease of diagnosis, shorter pathological viewing times, and was cost neutral.

Introduction IgG4 related disease (IgG4-RD) is a rare immune mediated fibroinflammatory condition that can affect nearly any organ. Pancreaticobiliary (PB) manifestations include autoimmune pancreatitis (AIP) and cholangiopathy. [18F]-fluorodeoxyglucose positron emission tomography/computed tomography (FDG PET/CT) is the only technique that allows imaging of metabolic activity by detecting FDG accumulation in cells and correlation with anatomical structures. Increased tracer uptake is typically seen in inflammatory or neoplastic tissue thereby potentially aiding diagnosis, and assessment of disease extent and activity. There is limited data currently available on its utility in IgG4-RD and whether this varies according to presentation. The aim of this study is to determine the utility of FDG PET/CT in diagnosis, monitoring disease activity and identifying multi system involvement. Methods We performed a retrospective study of a prospectively maintained multi-disciplinary IgG4-RD database to identify patients who underwent FDG PET/CT over a 3-year period. Additional organ involvement and change in management consequent on FDG PET/CT was recorded. Fisher’s exact test was used for the comparison of proportions. Results 25 patients with a diagnosis or suspicion of IgG4-RD underwent FDG PET/CT between November 2016 and October 2019. The median age [IQR] at presentation was 59 [48.5–65.5], 18 (72%) were male. 15 (72.5%) suspected or proven PB disease, 6 (24%) head and neck (HN), 1 (4%) each of retroperitoneal, both PB and HN, pulmonary and renal. In 22 (88%) cases (15/15 PB, 7/10 non PB) FDG PET/CT findings had a direct impact on management. The difference in utility between PB (100%) and non-PB (70%) was not quite statistically significant (p=0.059). In 1 patient it enabled exclusion of PB IgG4-RD. In 15 (60%) it led to a decision to escalate therapy this included 3 AIP cases (21.4% of definite PB cases) in which new organ involvement was identified. In 6 cases (5 PB and 1 renal IgG4-RD) with concern of active disease because of persistently elevated or rising IgG4 levels it excluded FDG avid inflammation. Conclusion In this retrospective study FDG PET/CT had a clinically important impact on management of IgG4-RD. Identifying other organ involvement as well as influencing therapeutic decision making particularly in PB disease. Further studies are required to fully delineate its role in IgG4-RD.

Background Disconnected pancreatic duct syndrome (DPDS) is a complication of acute necrotizing pancreatitis in the neck and body of the pancreas often manifesting as persistent pancreatic fluid collection (PFC) or external pancreatic fistula (EPF). This systematic review and pairwise meta-analysis aimed to review the definitions, clinical presentation, intervention, and outcomes for DPDS. Methods The PubMed, EMBASE, MEDLINE, and SCOPUS databases were systematically searched until February 2020 using the PRISMA framework. A meta-analysis was performed to assess the success rates of endoscopic and surgical interventions for the treatment of DPDS. Success of DPDS treatment was defined as long-term resolution of symptoms without recurrence of PFC, EPF, or pancreatic ascites. Results Thirty studies were included in the quantitative analysis comprising 1355 patients. Acute pancreatitis was the most common etiology (95.3%, 936/982), followed by chronic pancreatitis (3.1%, 30/982). DPDS commonly presented with PFC (83.2%, 948/1140) and EPF (13.4%, 153/1140). There was significant heterogeneity in the definition of DPDS in the literature. Weighted success rate of endoscopic transmural drainage (90.6%, 95%-CI 81.0–95.6%) was significantly higher than transpapillary drainage (58.5%, 95%-CI 36.7–77.4). Pairwise meta-analysis showed comparable success rates between endoscopic and surgical intervention, which were 82% (weighted 95%-CI 68.6–90.5) and 87.4% (95%-CI 81.2–91.8), respectively (P = 0.389). Conclusions Endoscopic transmural drainage was superior to transpapillary drainage for the management of DPDS. Endoscopic and surgical interventions had comparable success rates. The significant variability in the definitions and treatment strategies for DPDS warrant standardisation for further research.

Background and study aims Radiofrequency ablation (RFA) is used to treat blocked biliary stents in patients with pancreaticobiliary (PB) tumors with varying results. We report our experience with a novel temperature-controlled probe for treatment of blocked metal stents. Patients and methods Patients with histologically proven PB cancers and a blocked biliary stents were treated using ELRATM electrode (Taewoong Medical) under fluoroscopic guidance. Demographics, clinical outcome, stricture diameter improvements, complications and mortality at 30 days were prospectively recorded. Results Nine procedures were performed on seven patients (4 male, 3 female); mean age 65.33 (range 56 – 82 years). Mean stricture diameter prior to RFA was 1.13 mm (SD ± 0.54) and 4.42 mm (SD ± 1.54) following RFA (P Conclusion These are the first reported data on use of a temperature-controlled RFA catheter in humans to treat blocked metal biliary stents. The device is safe but further randomized trials are required to establish the efficacy and survival benefits of this probe.

Background and Aims A new core biopsy needle with a novel tip, opposing bevel, and sheath design has recently been introduced for EUS-guided fine-needle biopsy (FNB). The diagnostic utility of this needle for differentiating solid pancreatic masses is currently unknown. The aim of this study was to compare the diagnostic performance and yield for tissue acquisition from solid pancreatic lesions of the opposing bevel needle with those of a reverse bevel EUS-FNB needle. Methods Consecutive patients with solid pancreatic masses undergoing EUS-FNB using the opposing bevel (n = 101) and the reverse bevel (n = 100) core biopsy needles were included in the study. Final diagnosis was based on positive histology or at least 12 months of follow-up in cases with a negative biopsy. The primary outcome was the diagnostic performance of the 2 needles for malignant pancreatic masses. A secondary outcome was the diagnostic yield. Results Compared with the reverse bevel needle, using strict criteria the opposing bevel needle provided significantly higher sensitivity (71.1% vs 90.1%; P = .0006) and overall accuracy (74% vs 92%; I = 0.0006) for discriminating malignant from benign solid pancreatic masses. The proportion of samples classified as adequate for histologic analysis was 87% for the reverse bevel needle versus 99% for the opposing bevel needle ( p = 0.002) Multivariate analysis controlling the needle gauge and site did not show any significant difference in accuracy and sensitivity between the 2 groups. There were no adverse events in either group. Conclusions In this first, large, single-center preliminary cohort study, an EUS core biopsy needle with a novel tip, opposing bevel, and sheath design afforded substantially superior tissue yield and diagnostic performance compared with a reverse-bevel needle. If replicated by randomized controlled trials, our findings suggest that similarly designed needles could become the standard of care for EUS-guided tissue acquisition from solid pancreatic masses.

Endoscopic ultrasound (EUS) is now firmly established as one of the essential tools for diagnosis in most gastrointestinal MDTs across the UK. However, the ability to provide therapy with EUS has resulted in a significant impact on the management of the patients. These include drainage of peripancreatic collections, EUS-guided endoscopic retrograde cholangiopancreatogram, EUS-guided coeliac plexus blocks, etc. The rapid development of this area in endoscopy is a combination of newer tools and increasing expertise by endosonographers to push the boundaries of intervention with EUS. However, the indications are limited and we are at the start of the learning curve for these high-risk procedures. These therapies should, therefore, be confined to centres with a robust multidisciplinary team, including interventional endoscopists, radiologists and surgeons.

Introduction Endoscopic ultrasound (EUS) fine needle aspiration (FNA) has poor performance in diagnosing autoimmune pancreatitis (AIP) due to small sample size and lack of preserved tissue architecture. In mass forming disease non-diagnostic EUS-FNA may result in a presumptive diagnosis of malignancy and unnecessary surgery. A core biopsy needle with a novel fork-tip design (SharkCore™) has been introduced with the aim of improving yield and maintaining tissue architecture. Literature on its performance in the diagnosis of AIP is limited to a few case reports. Aim To assess the diagnostic performance of EUS tissue sampling with a fork-tip needle in the diagnosis of type 1 AIP. Methods Retrospective review of a prospectively maintained AIP database in a tertiary center to identify patients with a final diagnosis of type 1 AIP who underwent EUS-TS during diagnostic workup. Pathology reports were reviewed and classified as per International Consensus Diagnostic Criteria (ICDC); Level 1 (highly suggestive) requires the presence of 3 or all 4 histological features and level 2 (probable) requires 2 features. Results Between March 2006 and November 2018, 35 procedures were performed on 28 individuals (29 lesions) with a final diagnosis of Type 1 AIP. Mean age ( ± SD) 63 (±11.4), 21 male. 29 procedures were for a mass lesion and or biliary obstruction. 2 patients underwent surgical resection and 2 laparoscopic biopsy. There were 8 procedures with an FNA needle, 7 reverse bevel and 20 fork-tip. There were 6 inadequate samples (3 FNA, 2 reverse bevel and 1 fork-tip). Of the 29 adequate samples, 15 (51.7%) met ICDC criteria for diagnosis of AIP all at level 1. 0/13 FNA or reverse bevel samples were diagnostic compared to 15/19 (79%) fork-tip samples (p=0.0001). Obliterative phlebitis was identified in 10/19 (52.6%), storiform fibrosis in 14/19 (70%), dense lymphoplasmocytic infiltrate in 14/19 (70%) and IgG4 positive plasma cell count > 10 per high power field in 14/19 (70%) of adequate fork-tip samples. Of the 17 individuals who underwent 20 procedures (2 repeat and 1 sampling of 2 separate lesions) with a fork-tip needle, 14 (83.2%) had a final ICDC level 1 diagnostic fork-tip biopsy. Conclusion In this the largest study to date of the fork-tip core biopsy needle in the diagnosis of AIP, the needle demonstrated very good overall diagnostic performance. This study supports the preferential use of this needle in suspected type 1 AIP.

Introduction In an attempt to overcome the limitations of Endoscopic ultrasound (EUS) fine needle aspiration (FNA) a fine needle biopsy (FNB) needle (SharkCore™) with a novel fork-tip, has been introduced. This needle is designed to increase tissue yield and preserve tissue architecture. The aim of this study was to determine if FNB histology samples had better diagnostic performance for solid pancreatic masses than FNA cytology samples. Methods Consecutive patients referred for EUS-guided sampling of solid pancreatic lesions were recruited. Each patient had 3 passes with a standard (Beacon™) FNA needle and 3 passes with a core (SharkCore™) FNB needle performed in a randomised order. 25g needles were used for transduodenal sampling and 22g for transgastric. A single slide was made from each pass with the FNA needle and the remaining aspirate placed in CytoRich™ fluid and sent for liquid based cytological analysis. All samples from the SharkCore™ needle were placed in a single container of formaldehyde and sent for histological analysis.. Samples were reported by expert cytopathologists and histopathologists respectively who were blinded to the results of the other needle. Only samples reported as diagnostic of malignancy were considered positive. Inadequate samples were not excluded from the analysis. Sample quality and ease of diagnosis were assessed on a 3 point scale. The primary endpoint was the accuracy of the diagnosis of malignancy. Secondary endpoints were the quality of sample obtained, ease of diagnosis and the duration of tissue sampling and pathological reporting. Results 108 participants were recruited, 57 male; mean age 66.9 ± 10.9. 85.2% had a final diagnosis of malignancy. Median lesion size (IQR) was 25 mm (19–34.5). 62 (57.4%) of lesions were in the head of pancreas. Tissue results from the FNB needle were significantly more accurate than FNA (84.2% vs 75%, p=0.041) in discriminating malignant from benign masses. A greater proportion of FNB samples had abundant diagnostic material (59.2% vs 44.4%, p=0.017) and a straight forward diagnosis (68.9% vs 51.9%, p=0.03). Biopsy sampling time median (IQR) 685s (565–832) vs 752s (651–835), p=0.0006) and pathology reporting times (191s (134–258) vs 332s (260–358), p Conclusion The diagnostic performance of the SharkCore™ FNB needle was significantly better than that of a standard FNA needle in the diagnosis of solid pancreatic masses and was associated with better sample quality, ease of reporting and shorter sampling and pathological reporting times.

Introduction Pancreatic cystic lesions are being identified more frequently as cross sectional imaging has improved. Referral for endoscopic ultrasound (EUS) for further assessment and possible sampling is well recognised however there is little data on whether the characteristics of such lesions have changed over time. There is also a paucity of data concerning changes in pathological findings and surgical resection rates with respect to cystic lesions. The aim of this study was to assess trends in referral for EUS with respect to patient and cyst characteristics, pathological adequacy and surgical resection rates over a 10 year period to our centre. Patients and methods Retrospective analysis of the EUS database was performed over the period 2003 to 2012. EUS procedures for assessment of cystic lesions were identified and information concerning number per year, patient age and sex at referral, cyst size, cyst site, ability to give a cytological diagnosis, surgical resection and malignant surgical resection pathology were recorded. Kendall’s tau test (continuous data) or the Chi squared (categorical data) for trend test was used to determine significant changes over time. Results 417 patients (mean age 64.3 years, 163 males) underwent EUS in our unit for the assessment of cystic lesions over the study period. There was no significant difference in patient age or sex at referral over the study period. There was a significant increase in the number of procedures per year from 2003 (n=11) to 2012 (n=74)(tau 0.556, p=0.032) but a significant decrease in cyst size from 4.75cm in 2003 to 2.2cm in 2012 (tau -0.112, p=0.001). There was a significant change in the cyst site over time mainly due to an increase in the proportion of cysts found in the body of pancreas (p for trend Conclusions Cyst referrals for EUS have increased significantly but cysts are smaller, less likely to undergo resection, have a lower rate of malignancy and more likely to be in the body. Inability to give a cytological diagnosis rates also reduced significantly which may be due to improved laboratory techniques or a learning curve effect.

Background Stents are frequently placed in patients with biliary obstruction due to a mass in the head of the pancreas. The impact of plastic or self-expandable metal stents (SEMSs) on endoscopic ultrasound (EUS)-guided tissue sampling is unclear. This study aimed to assess, using strict pathological criteria, whether stents impair fine-needle aspiration (FNA) or fine-needle biopsy (FNB). Methods All patients with a solid mass in the head of the pancreas who underwent EUS-guided tissue sampling between 2010 and 2016 at our unit were included. Factors with possible impact on diagnostic performance were analyzed using logistic regression. Analysis was performed using both strict (malignant only) and less strict (suspicious for malignancy) cutoffs. Results Of 631 individuals undergoing 698 procedures, 535 (84.8 %) had a final diagnosis of malignancy, 141 had SEMS, 149 had plastic stents, and 341 had no stent. Using strict criteria, SEMS were associated with an increased occurrence of incorrect diagnosis of EUS tissue sampling, with an odds ratio (OR) of 1.96 (95 % confidence interval [CI] 1.24 – 3.10). Increasing tumor size (OR 0.72, 95 %CI 0.59 – 0.87), increasing number of passes (OR 0.84, 95 %CI 0.72 – 0.99), and fork-tip biopsy needle (OR 0.52, 95 %CI 0.31 – 0.86) were independently associated with a decrease in incorrect diagnosis. Repeat tissue sampling was more common with SEMSs (10.2 %) than with plastic stents (2.9 %) or no stents (4.5 %) (P Conclusion SEMS use had a negative impact on tissue diagnosis in pancreatic head masses, whereas use of a fork-tip biopsy needle and increasing number of passes were independently associated with improved accuracy.

Introduction The most effective investigation for suspected gallstones between MRCP and EUS is unclear. A 2015 Cochrane systematic review of their performance in common bile duct (CBD) stones concluded that the tests were of comparable accuracy. Conversely, a 2017 meta-analysis found EUS to be more sensitive. Any superiority of EUS may be due to better accuracy in detecting small stones. MRCP is routinely favoured as the 2nd line test following a non-diagnostic abdominal ultrasound and EUS subsequently performed as the 3rd line test when suspicion remains after a non-diagnostic MRCP. The yield and clinical utility of EUS in this setting is unclear. The aim was to identify the yield of EUS in patients with prior non-diagnostic MRCP undergoing EUS in our tertiary service. Methods All EUS reports from 2017 were reviewed along with the electronic patient records to identify cases with prior MRCP. Indication for the procedure, symptoms, liver blood tests and interval between MRCP and EUS were recorded. Findings of sludge, microlithiasis (stones Results A total of 1058 diagnostic EUS were screened of whom 253 (24%) had prior MRCP and formed the study group. Median age was 58 (16–88) years, 179 (71%) were female and 91 (36%) had a cholecystectomy. Median interval between EUS and MRCP was 5.2 (0.1–37) months. Indications for EUS were: n=76 (30%) dilated CBD, n=65 (26%) query CBD stones, n=54 (21%) unexplained acute pancreatitis (AP), n=23 (9%) right upper quadrant pain, n=17 (6.7%) abnormal LFTs, n=16 (6.3%) double duct sign and n=2 (1%) dilated PD. There was a yield from EUS in 30 (12%) patients with no significant difference between those with (n=11) or without cholecystectomy (n=19). Stones were identified in 24 cases with median size of 4 mm (range 2–8) in: CBD (n=16), cystic duct (n=1) and GB (n=7). Three had abnormal CBD without stones (calcification CBD wall, thick walled CBD, polyp), 1 patient with possible stone on MRCP had no stone seen on EUS, 1 had a pancreatic mass, and 1 had chronic pancreatitis. All patients in whom EUS findings indicated an intervention (26/30) have been referred: ERCP in 13, cholecystectomy in 9, ERCP and cholecystectomy in 3 and chemotherapy in 1. Conclusion EUS following non-diagnostic MRCP is a sizeable workload accounting for 24% of diagnostic activity in our unit with a clinically significant yield in 12% of predominantly small stones. Further prospective studies are required to ascertain the most cost-effective way to incorporate EUS into the investigation of suspected gallstone disease.

Carcinoembryonic antigen (CEA) is suggested as the single most useful EUS/EUS-FNA derived test for the diagnosis of mucinous pancreatic cysts.To investigate the yield and diagnostic performance of EUS/EUS-FNA on an intention to diagnose basis and to determine the utility of the recommended CEA and amylase cut-off values.A retrospective study of a prospectively maintained database of 433 procedures performed in a 10 year period. Diagnostic performance of EUS-FNA was determined in 133 procedures with a definite diagnosis.CEA value was determined in significantly fewer procedures (58.6%) than EUS diagnosis was stated (83.4%; p 0.0001), cyst fluid appearance recorded (89.4%) or adequate sample for cytology obtained (76.7%; p 0.005). Median CEA was significantly higher in mucinous cysts than non-mucinous (175 ng/ml vs 3 ng/ml, p 0.0001) and in malignant cysts compared to benign (8945 ng/ml vs 93 ng/ml, p 0.001). On an intention-to-diagnose analysis, a CEA cut-off of 110 ng/ml was significantly less accurate (42.8%) than EUS diagnosis (67.7%), cytology (58.6%) or aspirate appearance (66.9%; p 0.05 for all comparisons). However, the combination of EUS diagnosis, cytology and CEA provided higher sensitivity (91%), specificity (75%) and accuracy (85.7%) than each component test alone (p 0.05 for all comparisons). Median amylase was significantly higher in benign compared to high-risk mucinous cysts ((11,429IU/L vs. 113IU/L; p 0.05.The combination of EUS, cytology and CEA performed well. Malignant cysts had a higher CEA value than benign cysts. On an intention to diagnose basis a CEA cut-off of 110 ng/ml performed poorly.

Background Secretin-enhanced magnetic resonance cholangiopancreatography (S-MRCP) facilitates better visualization of the pancreaticobiliary ductal system but its role in patients with acalculous biliary pain (ABP) is yet to be established. The aim of this study was to assess the diagnostic yield and the role of S-MRCP in the investigation of ABP patients. Methods This is a retrospective analysis of patients who had S-MRCP to investigate ABP over a 5-year period from June 2008 to May 2013. The findings and diagnosis as reported in the S-MRCP were compared with the findings on MRCP. The primary endpoint was the diagnostic yield of S-MRCP in ABP patients. Results A total of 117 patients with ABP [28 (24%) male] had S-MRCP during the study period. The most common abnormality identified was obstruction at the level of ampulla or in the proximal pancreatic duct. S-MRCP was able to identify significant pathological findings in 8 of 34 (22%) patients in whom MRCP did not detect any abnormality. Endoscopic ultrasound (EUS) was performed in 67% of patients. S-MRCP identified abnormalities in 21 of 41 (54%) patients who had a normal EUS. Conclusions We conclude that the diagnostic yield of S-MRCP for recognizing anatomical variants of the pancreatic ductal system, in particular ampullary or proximal pancreatic duct stricture, is better than MRCP and EUS. These findings reflect the dynamic nature of S-MRCP and its complementary role alongside MRCP, EUS and endoscopic retrograde cholangiopancreatography in ABP patients. Keywords MRCP, secretin, abdominal pain, visceral pain, endosonography Ann Gastroenterol 2016; 29 (3): 367-372

Pancreaticobiliary disorders are common in clinical practice and often require complex investigations before a final diagnosis and management plan is formulated. Many patients will undergo an endoscopic investigation or therapy, both of which can be challenging. Endoscopic ultrasound is increasingly used as one of the modalities for investigating and treating such patients and the indications for interventional endoscopic ultrasound appear to be increasing. In this review article, the role of endoscopic ultrasound in the investigation pathway of patients with benign and malignant pancreaticobiliary diseases will be considered. The limitations of endoscopic ultrasound and some areas of future expansion will also be discussed.

Background and study aims: This report describes the use of a novel, fully covered, self-expanding metal stent (FCSEMS) for endoscopic ultrasound (EUS)-guided drainage of walled-off pancreatic necrosis (WON). Patients and methods: Patients with WON, as defined by the revised Atlanta Criteria, were included in this open-lable, two-center, observational study. The WON was punctured using a cystotome, and the FCSEMS was inserted under fluoroscopic guidance. Necrosectomy procedures were performed as necessary. Results: A total of 19 patients were included. The median maximum collection size was 15 cm with a median of 50 % necrosis. A total of 14/19 patients underwent necrosectomy, requiring a median of 4 procedures. Resolution or reduction in the size of collection by at least 80 % was achieved in all patients. Percutaneous or surgical drainage was required in three patients. Five stents migrated or dislodged. One patient had abdominal pain post-procedure. Five patients died during follow-up (three from multi-organ failure, and two unrelated to pancreatitis). Conclusions: Use of this stent is feasible and safe for drainage of WON. However, stent displacement rates were high, and improvements to the stent design are required before it can be advocated for routine use in WON.

Background & Aims: Our study aimed to assess the sensitivity of EUS and EUS-FNA for pancreatic neuro-endocrine tumors (pNETs) and compare performance over two consecutive 4 year 2 month periods, to investigate the comparative performance between solid and cystic pNETs and determine the incremental yield of EUS±FNA in individuals with a mass not diagnosed as a pNET after cross-sectional imaging.Methods: A retrospective review of a prospectively maintained database was carried out to identify all pNET patients who underwent EUS-FNA between April 2003 and September 2011.Results: A final diagnosis of solid and cystic pNETs was made in 43 and 10 patients, respectively. Overall, the yield of combined EUS imaging and cytology was significantly higher than that of CT and/or MRI (p< 0.05) across all groups [solid (83.7% vs. 41.8%), cystic (70% vs. 10%) and combined solid-cystic (81.1% vs. 35.8%)]. The yield of combined EUS imaging and cytology was significantly better than EUS imaging alone (p

Introduction A new core biopsy needle (SharkCore) with a novel design has recently been introduced for endoscopic ultrasound guided biopsy. There is as yet no data on how its diagnostic performance compares to the existing core biopsy needle (Procore) in the diagnosis of pancreatic masses. The aim of this study was to compare the diagnostic performance and yield of the two core biopsy needles i.e. the ProCore and the Sharkcore needles in the biopsy of solid pancreatic lesions. This is the first study in literature to compare 2 core biopsy needles. Methods A retrospective comparison of a prospectively maintained database. Consecutive patients who had EUS guided fine needle biopsy with Procore (100) and Sharkcore (101) were included in the study. Demographic details, site of lesion, number of passes and histological diagnosis. Final diagnosis was based on positive histology or at least 6 months follow up in cases with a benign pathology. Primary outcome was to compare the diagnostic accuracy of the two needles using only samples graded as definitely malignant as diagnostic for malignancy. The secondary outcome was to compare the diagnostic yield for both needles. Results The sensitivity, specificity, negative predictive value and accuracy between the groups were - Procore =71%, 100%, 28% and 74% vs. Sharkcore =91%, 100%, 53% and 94%. There was a significant difference in sensitivity (p = 0.0003) and accuracy (p = 0.006). The proportion of samples classified as adequate for histological analysis was 87% for Procore and 99% for Sharkcore - this was significantly different (p = 0.0009). There were no complications in either group. Conclusion In this study the SharkCore biopsy needle demonstrated significantly better diagnostic accuracy and tissue yield. Prospective randomised studies are required. Disclosure of Interest None Declared

Introduction In cases of obstruction of the gallbladder (GB) or cystic duct where surgery is high risk, EUS guided GB drainage (EUS-GBD) is reported to have comparable efficacy to percutaneous drainage. 1 However, the technique has not been widely adopted due to lack of specific devices and concerns about leakage and stent migration. Lumen apposing metal stents (LAMS) have been developed to minimise the risk and simplify the procedure. A novel (Hot AXIOS™) device has recently become available and this consists of a stent and electrocautery-enhanced delivery system, enabling a single device to be used when previously multiple devices and steps were required. There is a single case report of its use for EUS-GBD in the literature. 2 A 68 year-old female, presented with obstructive jaundice, due to inoperable hilar cholangiocarcinoma. A Percutaneous transhepatic cholangiography (PTC) procedure was performed and an internal external drain placed. Shortly afterwards she developed clinical features of cholecystitis. A CT scan showed gross distension of the GB and a GB stone. There was extensive pericholecystic fluid with fat stranding. Following HPB MDT discussion, a decision was made to perform EUS-GBD using a Hot AXIOS™. Methods The procedure was undertaken under conscious sedation, using a therapeutic echoendoscope. The optimal site for access was identified as being in the antrum. The GB was punctured with the Hot-AXIOS device (15 mm x 10 mm (W x L) 24 mm flange diameter stent) and a cutting current was applied. X-ray screening was used; however, deployment was entirely under EUS control. The stent was dilated with a 10 mm balloon. Results The procedure was well tolerated and completed within 15 minutes. The stent was in a good position on EUS and fluoroscopy. A large amount of pus drained immediately. There was a rapid improvement in clinical condition and inflammatory markers over the next few days. Biliary drainage was internalised with placement of 2 metal stents. CT scans 8 days and 7 weeks post placement showed the stent in a good position with resolution of the cholecystitis. Conclusion Technical and clinical success was achieved in this case which to the best of our knowledge is the first EUS-GBD procedure in the UK utilising this device. Further experience of EUS-GBD with the Hot AXIOS™ and randomised controlled trials against percutaneous drainage are required to delineate its role in high risk surgical cases. References 1 Jang JW, et al. Gastroenterol . 2012 Apr; 142 (4):805–11. 2 Teoh AYB, et al. GIE 2014; 80 (6):1171. Disclosure of Interest None Declared

Background Recent data suggest that quantitative EUS elastography, a novel technique that allows real-time quantification of tissue stiffness, can accurately differentiate malignant from benign solid pancreatic masses. Objective To externally validate the diagnostic utility of this technique in an independent cohort. Design and Setting Prospective, single-center study. Patients, Interventions, and Methods A total of 104 patients with evidence of a solid pancreatic mass on cross-sectional imaging and/or endosonography underwent 111 quantitative EUS elastography procedures. Multiple elastographic measurements of the mass lesion and soft-tissue reference areas were undertaken, and the corresponding strain ratios (SRs) were calculated. The final diagnosis was based on pancreatic cytology or histology. Main Outcome Measurements The area under the receiver-operating characteristic curve, sensitivity, specificity, positive predictive value, negative predictive value, and overall accuracy of quantitative EUS elastography for discriminating malignant from benign pancreatic masses. Results The final diagnoses were primary pancreatic carcinoma (71.2%), neuroendocrine tumor (10.6%), metastatic cancer (1.9%), and pancreatitis (16.3%). Malignant masses had a higher SR ( P = .01) and lower mass elasticity ( P = .003) than inflammatory ones. The areas under the receiver-operating characteristic curve for the detection of pancreatic malignancy of both SR and mass elasticity (0.69 and 0.72, respectively) were less favorable than reported recently. At the cut points providing the highest accuracy in this cohort (4.65 for SR and 0.27% for mass elasticity), quantitative EUS elastography had a sensitivity of 100.0% and 95.7%, specificity of 16.7% and 22.2%, positive predictive value of 86.1% and 86.4%, negative predictive value of 100.0% and 50.0%, and overall accuracy of 86.5% and 83.8%, respectively. Limitations Relatively small number of patients with benign disease. Conclusion In the largest single-center study to date, the diagnostic utility of quantitative EUS elastography for discriminating pancreatic masses was modest, suggesting that it may only supplement rather than supplant the role of pancreatic tissue sampling in the future.

Introduction EUS-guided FNA is the standard of care for tissue sampling of pancreatic lesions. The new Echo Tip ProCore biopsy needle (Cook Medical) was developed to obtain core tissue specimens for histological analysis improving the diagnostic yield and potentially obviating the need for onsite cytopathologist. Our aim was to compare the standard Echo tip ultra FNAC needle with the new Echo Tip Ultra ProCore FNAB needle in the sampling of solid pancreatic lesions. Method Randomised multicenter trial (REC Ref 12/EM/0189). All patients referred for EUS-guided tissue sampling of a solid pancreatic mass were invited to participate. Linear EUS (Aloka-Olympus/Hitachi-Pentax) was performed and patients were randomised to either FNAC or FNAB. Due to the nature of the study the endoscopist could not be blinded. A 22G needle was used when sampling from the stomach and a 25G from the duodenum. A maximum of 5 passes were allowed. Sampling time from first insertion of the needle into the lesion to the time the last tissue sample is transferred into the histology pot, number of passes and adequacy for diagnosis were recorded and included for analysis. Immediate and delayed complications were also recorded. Results 266 patients were recruited (median age 67 years, 52% men); 142 randomised to the FNAC group and 124 to the FNAB group. Two patients were excluded after randomization; no mass lesion was seen in 1 patient randomised to FNAC and the needle failed in 1 patient in the FNAB group and FNAC was performed. 264 patients were analysed, 141 in the FNAC group (22G=83 and 25G=58. Median max diameter of the lesion 34 mm, range 10–87 mm) and 123 in the FNAB group (22G=53 and 25G=71. Median max diameter 33 mm, range 8–85 mm). There was no statistical significant difference between groups in number of passes (median 4, range 1–5 in the FNAC group vs. median 3, range 1–5 in the FNAB group, p = 0.12), sampling time (median 13 min, range 1–32 min vs. median 13 min, range 2–42 min, p = 0.63) and adequacy of the specimen for diagnosis (90.8% vs. 87%, p = 0.33). Three technical failures were observed in the FNAB group and none in the FNAC cohort (1.6% vs. 0%, p = 0.99). No complications reported in either group. Conclusion FNAC and FNAB needles were comparable in terms of sampling time, number of passes, sample adequacy for diagnosis and safety. Technical failure was higher in the FNAB arm but not statistically significant. Disclosure of interest None Declared.

Introduction There are a number of factors which can influence the accuracy of EUS FNA but the presence of an in-room cytotechnician (IRC) has been shown to be one of the important factors. The aim of this study was to establish if the introduction of an IRC improved the diagnostic accuracy of EUS guided FNA of solid pancreatico biliary lesions Methods This is a prospective study and includes all patients with solid pancreatico biliary lesions who underwent EUS FNA from April 2009 to September 2010. We have been performing EUS FNA since 2003 but did not have an IRC till this period. The IRC attended 2 of the 4 EUS lists and therefore there were two groups identified: Group 1 – Cytotechnician absent; Group 2 – Cytotechnician present. Final diagnosis was based on a positive diagnosis on cytology, other forms of tissue acquisition and/or a year9s follow up for the benign/indeterminate cases. Only patients with a final diagnosis were included in the study. Results The final results are shown in table 1. We compared our data to the 6 months prior to this study (October 2008–March 2009) when there was no IRC available. The accuracy, sensitivity, specificity, PPV and NPV during this period were 86%, 82%, 100%, 100% and 57.6% respectively. The inadequate aspirate rate during this period was 14.2%. There was no statistical difference when this data was compared to either Groups 1 or 2. Conclusion This study shows that the presence of an IRC does not necessarily improve the diagnostic accuracy of EUS FNA of pancreatobiliary lesions. There are a number of factors which influence the results but experience of the cytopathologist/cytotechnician has shown to be one of the important factors. By keeping the number of inadequate aspirates low and reducing the number of highly suspicious samples the accuracy of the test can be improved. Large studies may be required to show a significant difference. Cost effectiveness and experience of the cytopathologist/cytotechnician should be taken into consideration when setting up this service.

Introduction Secretin enhanced magnetic resonance cholangio-pancreatography (S-MRCP) has been used as part of the diagnostic algorithm for the diagnosis of patients presenting with acalculous biliary type abdominal pain (ABAP); the exact role of this diagnostic modality is unclear. The aim of this study was to assess the diagnostic yield of S-MRCP in a large HPB tertiary referral centre in the investigation of patients with ABAP. Methods Retrospective analysis of a prospectively maintained database of all patients presenting with ABAP in whom a S-MRCP was requested from June 2008 to May 2013. The findings of S-MRCP were compared with those of MRCP and EUS (performed prior to S MRCP) and the diagnostic yield of S-MRCP in the work up of a patient presenting with ABAP was estimated. Results Of the 117 patients with ABAP [28 males and 89 female; mean age 48 yrs] were referred for S-MRCP, 114 (97.4%) patients successfully completed the scan. Of these 37 patients who had a normal MRCP, S-MRCP identified significant findings in 8 (22%) patients (Table 1). In the present study 78 (67%) patients had EUS. Out of the 41 patients who had a normal EUS, S-MRCP was able to identify significant pathology in 21 (54%) patients (Table 1). Conclusion This study suggests that S-MRCP has a 22 and 54% additional diagnostic yield in ABAP patients who have a normal MRCP and a normal EUS respectively. The commonest abnormality identified in these patients on S-MRCP was obstruction at ampulla or proximal PD. S-MRCP should be considered in the diagnostic algorithm of patients with ABAP. References 1 Mariani A, Arcidiacono PG, Curioni S, et al . Diagnostic yield of ERCP and secretin-enhanced MRCP and EUS in patients with acute recurrent pancreatitis of unknown aetiology. Dig Liver Dis 2009 Oct;41(10):753-8 2 Matos C, Metens T, Deviere J, et al . Pancreas divisum: evaluation with secretin-enhanced magnetic resonance cholangiopancreatography. Gastrointest Endosc 2001 Jun;53(7):728-33 Disclosure of Interest None Declared.

Introduction Post-inflammatory peri-pancreatic fluid collections are frequent sequelae of severe acute pancreatitis. Collections are at risk of suppurative infection complicated by pancreatic necrosis. Over the last decade there has been an increasing emphasis on minimally invasive drainage procedures, including EUS-guided cyst-gastrostomy, and these approaches seem to be associated with lower morbidity and mortality. Access to the necrosis cavity has however been severely limited by having to maintain the tract with small diameter plastic stents. Recently, a novel flanged fully covered self-expanding metal stent (FCSEMS; NAGI stent, Taewoong Medical, Korea) has been developed to allow for better drainage of infected necrosis and easier endoscopic access into the cavity. Setting A non-randomised prospective multicentre phase II study to determine the safety and efficacy of FCSEMS endoscopic cyst-gastrostomy in the management of complex/infected pancreatic fluid collections. Methods Patients were included if they had evidence of a pancreatic fluid collection which was deemed to be amenable for EUS-guided drainage after discussion at a HPB multidisciplinary meeting. Patients selected for EUS-guided drainage had cross sectional imaging (MR or CT) performed within 2 weeks of the procedure and then an EUS assessment was made of the necrotic component. The collection was punctured using a cystotome and the FCSEMS inserted over a guidewire with fluoroscopic control. Repeat procedures were performed as necessary. Results A total of 11 patients (8 male, 3 female) were included in the study. Median age was 57.3 years. The aetiology of the collection was gallstones in 6 patients, idiopathic in 3, ischaemic in 1 and drug-induced in 1. Ten patients had evidence of at least 30% necrosis within the collection. Mean diameter of the collection was 15 cm and EUS-guided puncture was initially performed in all patients. The tract was dilated with a balloon in 6 patients. Stent insertion was either with a 20 mm (7 patients) or 30 mm (4 patients) length FCSEMS. Ten patients underwent endoscopic necrosectomy, with a median of 3 procedures (range 1–10). Significant reduction in the size of collection was achieved in all patients. Adverse events included stent migration in 3 (2 spontaneously and 1 during necrosectomy). Two patients died of complications of severe acute pancreatitis. Conclusion FCSEMS insertion is feasible and safe for drainage of pancreatic fluid collections. It allows repeated through the stent necrosectomy procedures and appears to be a major advance in the management of infected pancreatic necrosis. Disclosure of Interest None Declared.

Introduction The detection and diagnosis of pNETs remains challenging. EUS and EUS-FNA has a significant role in the detection, precise localisation and cytological confirmation of pNETs. Methods A retrospective review of all pNET patients undergoing EUS-FNA between April 2003 and September 2011 was carried out to determine the efficacy of EUS-FNA in confirming pNETs and compare performance over two consecutive 4 year period. Results 10 patients (3% of EUS procedures for cystic lesions) with cystic pNETs and 44 (4% of EUS procedures for solid lesions) with solid pNETs were identified. Table 1 compares the size, demographics and diagnostic performance of radiology, EUS & cytology in solid and cystic pNETs. 17 and 5 solid and cystic pNETs respectively were diagnosed between 2003 and 2007 while 27 and 5 solid and cystic pNETs were diagnosed between 2008 and 2011. EUS-FNA diagnosis of cystic and solid pNETs has improved from 20% and 59% respectively between 2003 and 2007 to 100% and 81% respectively between 2008 and 2011. Overall, sensitivity of EUS cytology has improved from 50% to 84.4% (p = 0.015) during this period. Malignant potential of solid pNETs was higher (54.5% vs 20%) compared to cystic pNETs. Curative resection was higher in patients with cystic pNETs (80% vs 68%) compared to solid pNETs. Conclusion EUS & EUS-FNA is a useful test in diagnosing pNETs. Overall, the sensitivity of combined EUS imaging and cytology was significantly better compared to CT/MRI (p Disclosure of Interest None Declared.

Introduction The accurate diagnosis of solid pancreatic masses is important in directing appropriate management of patients. The methods commonly used are percutaneous, laparoscopic and EUS guided biopsy of these lesions. Laparoscopic guided biopsy consumes theatre time and space and can be more expensive than the alternatives. In our unit; laparoscopic guided biopsy is reserved for patients who have an inconclusive result from an EUS guided biopsy or are considered for trial resection. The aim of this study was to look at the diagnostic performance of endoscopic ultrasound (EUS) guided biopsy (fine needle aspiration (FNA) or pro-core biopsy) and laparoscopic (lap) guided biopsy of solid pancreatic masses in a large HPB referral centre. Methods Retrospective review of patients undergoing EUS or laparoscopic guided biopsy for solid lesions between January 2011 and March 2012. Data was obtained from a dedicated prospectively maintained database in the histopathology department. Final diagnosis was based on positive histology/cytology of pancreatic adenocarcinoma. Benign cases were followed up for a period of at least six months. Results 464 specimens from the pancreas (histology + cytology including pancreatic resections) were received by the histopathology department during this period. Of these 275/464 (59%) patients had tissue biopsy of solid lesions. These included: EUS guided biopsy = 253 and Laparoscopic guided biopsy = 22. In the latter group; 10/22(45%) had a previous EUS of which 8/10 had an accurate diagnosis. 12/22(54%) patients went straight for laparoscopic guided biopsy. For the purposes of this study; highly suspicious and malignant samples were categorised as malignant. The accuracy, sensitivity and negative predictive value for EUS guided biopsy and laparoscopic guided biopsy were 92%/96%; 90%/94% and 74%/89% respectively. The inadequate aspirate rate was 5% and 0% respectively. There was no significant difference between the two groups. The cost of performing these procedures in our trust were: EUS guided biopsy £1094 and lap guided biopsy £ 2164. Conclusion In our unit;EUS guided biopsy of solid pancreaticobiliary lesions provides high diagnostic accuracy with a low inadequate aspirate rate. Our data supports the role of EUS guided FNA as the first modality of tissue acquisition from the pancreas. Though this data also shows the cost effectiveness of EUS guided biopsy over lap guide biopsy; in units with lower diagnostic accuracy of EUS guided biopsy the cost benefit may not be realised. Disclosure of Interest None Declared.

Introduction The finding of dilated CBD or PD or both (double duct sign) on abdominal cross-sectional imaging (CT or MRI) in patients with normal LFTs frequently leads to further investigation by EUS if a clear cause is not demonstrated on imaging. There is limited literature on the yield of EUS in this setting. Methods A retrospective review of our prospectively maintained EUS database was carried out to identify patients who underwent EUS for dilated duct(s), normal LFTs and non-diagnostic cross-sectional imaging between January 2007 and August 2011. Our aim was to evaluate the yield of EUS in this setting. Minimum follow up was for 12 months. Results 83 patients (3% of pancreaticobiliary EUS procedures during this period) were identified. Mean age was 66.7 years (30 – 87). There was a female preponderance (73%). 40(48%) had dilated CBD only, 5(6%) had solitary PD dilatation and 38(46%) had both CBD and PD dilatation on prior imaging. EUS was concordant with prior imaging in 60(72%) patients and discordant in 23(28%) [partial agreement in 18, non-dilated ducts in 3 and different duct dilated in 2]. 16(19%) patients had a new finding on EUS [3 cbd stones, 3 cbd polyps, 4 microlithiasis, 1 ampullary adenoma, 3 chronic pancreatitis, 1 pancreatic duct adenocarcinoma (PDAC) and 1 portal vein compressing mid CBD]. Of these 11(13%) were felt to be the cause of duct dilatation. On subsequent MDT review, the PDAC was identified on the initial scan from the referring hospital. 45% (10/22) of males compared to 9.8% of (6/61) females had a new finding on EUS (p = 0.0008). Overall, more females had symptoms compared to males (p = 0.065). There was no significant association between symptoms and new finding on EUS (p = 0.24). 7(17.5%) of the isolated CBD dilatation, 3(60%) of the isolated PD dilatation and 1(2.5%) of the dilated CBD and PD had a causative diagnosis respectively. Follow up of patients with no cause of duct dilatation on EUS revealed that 18 patients (54.5%) in the dilated CBD cohort had prior cholecystectomy [one patient had peri-ampullary diverticulum and prior cholecystectomy] while 2 and 10 patients in the dilated CBD and PD cohort (n = 37) had ampullary stenosis and prior cholecystectomy respectively. Conclusion Our study shows that EUS has a significant yield in individuals with unexplained duct dilatation and normal LFT (13% had a causative diagnosis). The yield was highest in isolated PD dilatation. A new finding was significantly more likely in men then women. EUS should ideally follow review of original cross sectional imaging by a HPB radiologist. Disclosure of Interest None Declared.

Introduction Imaging morphology alone does not currently have high enough accuracy to determine the nature of pancreatic cystic lesions and therefore sampling of cyst fluid for biochemical markers via endoscopic ultrasound (EUS) has value. There is little data showing whether these markers are stable over time in an individual cyst. The aim of this study was to determine the variability of these markers. Methods Patients are prospectively added to the EUS FNA database when attending for EUS and the period April 2003–April 2010 was examined with reference to patients undergoing repeat EUS and aspiration of cystic lesions. These individuals had either declined surgery or there was clinical doubt concerning the neoplastic potential of the cyst. Results During the study period 267 EUS and cyst aspiration were performed. 28 individuals had 61 repeat procedures (24 had 2 procedures, 3 had 3 procedures and 1 had 4 procedures). Complete data was available on 13 individuals. Mean CEA at first aspiration was 74.8 compared to 121.0 at second aspiration (p=0.32) with a mean difference of −46.2. Mean amylase at first aspiration was 7142.4 compared to 7344.1 at second aspiration (p=0.98) with a mean difference of −186.2. There were no changes in CEA from normal to abnormal between aspirations. In two individuals there was a significant change in amylase level from normal to abnormal, both of whom were subsequently found to have undergone malignant change at surgery. Conclusion In this small study, biochemical markers appear relatively constant. Major changes in amylase may be associated with malignant change. Further study on a larger cohort is required.

Introduction The differentiation of mucinous from non-mucinous pancreatic cysts is important because of the malignant potential of the latter. EUS-FNA allows for high resolution imaging of pancreatic cysts as well as sampling for markers (CEA), cytology and a visual assessment of cyst content. The Cooperative cyst study found an elevated fluid CEA (>192 ng/ml) to be the single most accurate test in correctly predicting mucinous cysts with a sensitivity of 73%. The CEA value of 192ng/ml has subsequently been widely adopted as a definitive cut-off value. Methods The aims of the present study were to assess the utility of a cut-off value of 192 ng/ml in differentiating mucinous from non-mucinous pancreatic cysts and to compare this to general EUS assessment. IPMN and mucinous cyst adenoma/adenocarcinoma (MCA and MCAC) were considered separately. The study population comprised all the patients undergoing EUS-FNA at a tertiary centre for assessment of suspected pancreatic neoplastic cysts between June 2003 and April 2010. Results During this period 267 procedures were performed on 235 individuals, of whom 71 had a definitive diagnosis (60 resection histology, 5 histology, 6 malignant cytology), cystic degeneration of pancreatic adenocarcinoma (3) being excluded. 68 patients (51 females), (78 procedures) formed the study group. There were 25 mucinous cyst adenomas (11 MCA, 14 MCAC). There were 22 IPMN (2 malignant) and 21 non-mucinous cysts. For MCA/MCAC using a cut-off of 192 ng/ml the sensitivity, specificity, accuracy and NPV of detecting a mucinous lesion were 62.5%, 94.4%, 79.4%, 73.9%. Combining EUS morphology, cytology and visual assessment of aspirate (mucoid/non-mucoid) gave figures of 100%, 70.8%, 86%, 100%. The combination was significantly more sensitive p=0.007, but no significant difference in specificity. ROC area under the curve was numerically greater 0.861 versus 0.785 (not significant). For the IPMN patients the cut-off 192 ng/ml showed a sensitivity of 20% in the diagnosis of IPMN. EUS diagnosis had a sensitivity of 85%. Sensitivity of aspirate appearance: 87%, cytology: 50%; combining fluid appearance, EUS and cytology: 93%. Comparing the performance of CEA 192 versus combination in differentiating IPMN from non-mucinous cyst, ROC curve area was 0.864 versus 0.623 p =0.02. Conclusion A CEA cut-off of 192 ng/ml demonstrated good specificity and moderate sensitivity in the diagnosis of MCA, however it performed poorly in the diagnosis of IPMN. Combining EUS, aspirate appearance and cytology was significantly more sensitive in diagnosing mucinous cysts with a very high NPV.