Kirk P. Bernadino

Intraprocedural tissue diagnosis during ERCP employing a new cytology preparation of forceps biopsy (Smash protocol).

OBJECTIVES:Techniques of tissue sampling at endoscopic retrograde cholangiopancreatography (ERCP) have been underutilized due to technical demands, low yield, and lack of immediate intraprocedural diagnosis. The objective of this study was to describe a new inexpensive, highly efficient ERCP tissue processing, and interpretation technique to address these issues.METHODS:A retrospective, institutional review board approved, single-center study was done at a tertiary-care medical center. Between June 2004 and February 2009, 133 patients (age 38–95 years; men 53%) with suspicious biliary strictures underwent ERCP with tissue sampling using a new technique. Small forceps biopsy specimens were forcefully smashed between two dry glass slides, immediately fixed, stained with rapid Papanicolaou, and interpreted by an on-site pathologist during the procedure (Smash protocol).RESULTS:Of the 117 proven to have cancer, true-positive Smash preps included pancreatic cancer 49/66 (74%), cholangiocarcinoma 23/29 (79%), metastatic cancer 8/15 (53%), and other 4/7 (57%). The median number of Smash biopsies to diagnosis was 3 (range 1–17). Suspicious or atypical results were considered to be negative in this study. There were no false positives and no complications. Smash had an overall sensitivity of 89/117 (76%) for all cases. The true-positive yield of immediate Smash prep cytology, combined with ERCP fine needle aspirate (FNA) and forceps biopsy histology was 77/95 (81%) for primary pancreaticobiliary cancers.CONCLUSIONS:Immediate cytopathologic diagnosis at ERCP was established in 72% of patients presenting with suspected malignant biliary obstruction using a new cytological preparation of forceps biopsies. This approach to ERCP tissue sampling permits immediate diagnosis and avoids the need for subsequent procedures, adds little cost and time, and is safe to perform.

Diagnostic Yield of Video Capsule Endoscopy for Iron Deficiency Anemia Without Overt Gastrointestinal Bleeding

Diagnostic Yield of Video Capsule Endoscopy for Iron Deficiency Anemia Without Overt Gastrointestinal Bleeding Kirk Bernadino, Peter B. Anderson, Steve P. Bensen Background: Since it’s introduction to clinical use in 2001, video capsule endoscopy (VCE) has become an important adjunct in the evaluation of occult gastrointestinal hemorrhage and chronic GI bleeding of suspected small intestine etiology. Previous studies have reported diagnostic yield in obscure, overt gastrointestinal bleeding (OGIB) as high as 87%. To date, there have been no published studies evaluating the diagnostic yield in patients with iron deficiency anemia (IDA), but without OGIB, who have undergone a negative endoscopic work up. Aims: To determine the diagnostic yield VCE in patients with IDA, but no history of OGIB. Methods: Between 5/02 and 12/03 140 patients were referred for VCE, 46 of these for IDA. Of these, 36 patients met our inclusion criteria of: 1) IDA as defined by transferrin saturation of <15% or ferritin <29ng/ml in the previous 12 months, 2) no history of melena or hematochezia and 3) previous negative EGD and colonoscopy. 44% had a previous small bowel follow through or enteroclysis that was negative. 8% had previous enteroscopy that was negative. VCE was performed in all patients and interpreted by one of two gastroenterologists. Complications including retained capsule, obstruction and capsule failure were recorded. Results: 19 of 36 (53%) patients had abnormalities on VCE. Angiodysplasia was the most common abnormality, seen in 8/36 (22%). 5/36 (14%) had small bowel erosions or ulcerations. 1/36 (3%) had scalloped duodenal folds suspicious for celiac disease which was later confirmed by elevated serum TTG. 1/36 (3%) had gastric antral vascular ectasia unrecognized on previous EGD. 6/36 (17%) patients had active bleeding from an unclear source, 2 of whom also had non-bleeding angiodysplasia. No complications were reported. In three patients the exam ended prior to the capsule reaching the cecum. Conclusion: VCE found abnormalities in 53% of patients with IDA and previous negative endoscopic evaluation. Small bowel angiodysplasia is themost common abnormal finding. The yield of VCE for IDA without overt bleeding is less than the previously reported yield for overt gastrointestinal bleeding of suspected small bowel source. Further studies are needed to determine the impact of VCE on management strategy of chronic IDA. *M1779 Prospective Controlled Multicentric Trial Comparing Capsule Endoscopy with Intraoperative Enteroscopy: Long Term Results in Patients with Chronic Gastrointestinal Bleeding Harald Schmidt, Dirk Hartmann, Frank Kinzel, Dieter Schilling, Georg Bolz, Peter Reitzig, Henning E. Adamek, Hartmut Hollerbuhl, Klaus Guenther, Klaus Schoenleben, Juergen F. Riemann, Hans J. Schulz Background: Capsule Endoscopy (CE) is a very precise and efficant non-invasive diagnostic procedure in patients with chronic gastrointestinal bleeding. Long term clinical outcome data of patients undergoing CE and intraoperative enteroscopy (IOE) for investigation of obscure gastrointestinal bleeding are lacking. The aimof this porspective trial was to evaluate the diagnostic and therapeutic value of CE in long term follow up. Methods: 33 patients with obscure gastrointestinal bleeding (23 men, 10 women, mean age 61.6617.7 years) underwent CE and IOE between 02/2002-07/2003. Complete data of 29 patients were available (4 patients lost in follow up). Clinical outcome was assessed with a standartised patient questionnaire and personal communication with reffering doctors. Results: A defintive bleeding source was detected and effective traeted with argon plasma coagulation or surgical resection in 25 patients (18 angiodysplasia, 2 erosive-ulcerous lesions, 1 hemangioma, 1 Meckel’’s diverticulum, 1 ileum diverticulosis, 1 jejunal polyp, 1 lymphoma). Mean follow up was 310.3 days (range 32-553 days). Clinical signs of recurrent gastrointestinal bleeding occurred in 9 of 29 patients (1 positiv faecal occult blood test, 2 anemia, 2 red blood on stool, 2melena, 2 hematochezia). In 5 of these patients (17.2%) no further therapywas necessary, 4 patients (13.8%) needed blood transfusions (range 2-62 units). In 2 of these patients (6.9%) further endoscopic interventions (argon plasma coagulation) were necessary to controll rebleeding caused by recurrent angiodysplasias. The 4 patients without bleeding source in CE showed no signs of gastrointestinal rebleeding during follow up. Conclusion: Confirmed by our long term results Capsule Endoscopy (CE) followed by therapeutic intraoperative enteroscopy or surgical intervention is a very precise and efficiant non-invasive diagnostic procedure in patients with severe chronic gastrointestinal bleeding.

Diagnostic Yield of Standard Brush Cytology (BC) Versus EUS-FNA for Pancreatic Cancer with Associated Malignant Biliary Stricture. Is It Time to Bury the Standard Biliary Brush?

Diagnostic Yield of Standard Brush Cytology (BC) Versus EUS-FNA for Pancreatic Cancer with Associated Malignant Biliary Stricture. Is It Time to Bury the Standard Biliary Brush? Kirk Bernadino, Stuart R. Gordon, Douglas J. Robertson Background: Despite the use of modified biliary brushing techniques and triple sampling devices, EUS-FNA remains a superior modality of tissue sampling for pancreatic cancer. Given EUS-FNA is highly sensitive for obtaining a tissue diagnose in pancreas cancers, the value of adding BC of associated biliary strictures is unclear. In our institution all patients considered for palliative or neoadjuvant chemoradiotherapy undergo pretreatment EUS-FNA for tissue diagnosis. Those with obstructive jaundice also undergo ERCP. Aim: We compared the cancer detection rates of standard BC and EUS-FNA in patients with pancreas cancer with associatedmalignant biliary stricture to determine if BC increased the diagnostic yield beyond that of EUS-FNA alone. Methods: ERCP and EUS-FNA records from 3/01 to 11/03 were retrospectively reviewed. 23 patients with an ultimate tissue diagnosis of pancreas cancer underwent both standard BC of biliary stricture and EUS-FNA of pancreatic mass. BC and FNA cytology results were reported as negative, atypical, suspicious or positive for malignancy. For the purpose of this study, atypical or suspicious samples were considered as negative. Results: All BC and EUS-FNA samples had sufficient cellularity for cytologic evaluation. BC was positive 0/23 (0%). EUS-FNA was positive 20/23 (87%). All five cases of suspicious BC were positive by EUS-FNA. Three patients with pancreas cancer missed by EUS were diagnosed at surgery. Results: Sensitivity of BC and EUS-FNA are 0 and 87% respectively. Sensitivity increases to 22 and 91%, respectively when suspicious specimens are considered as positive. Conclusions: EUS is superior to standard BC for tissue diagnosis in patients with pancreatic cancer and associated biliary stricture. Adding standard BC to EUS-FNA does not increase the cancer detection rate. If EUS-FNA is to be performed on patients with a pancreatic mass our work suggests the effort and cost involved in obtaining BC of associated biliary strictures is not worthwhile. Further studies should be performed to determine if enhanced brushing techniques or multimodality sampling of malignant biliary strictures increases cancer detection rate beyond that of EUS-FNA in this population. patientswith pancreaticobiliary ductal strictures whohave a normalEUSorCPon EUS, these patients should be closely followed and explored when clinically warranted.