Evan L. Fogel

Risk factors for post-ERCP pancreatitis: a prospective multicenter study.

OBJECTIVES:Pancreatitis is the most common and serious complication of diagnostic and therapeutic ERCP. The aim of this study is to examine the potential patient- and procedure-related risk factors for post-ERCP pancreatitis in a prospective multicenter study.METHODS:A 160-variable database was prospectively collected by a defined protocol on patients undergoing diagnostic or therapeutic ERCP at 15 centers in the Midwest Pancreaticobiliary Group and participating in a randomized controlled study evaluating whether prophylactic corticosteroids will reduce the incidence of post-ERCP pancreatitis. Data were collected prior to the procedure, at the time of procedure, and 24–72 h after discharge. Post-ERCP pancreatitis was diagnosed and its severity graded according to consensus criteria.RESULTS:Of the 1,115 patients enrolled, diagnostic ERCP with or without sphincter of Oddi manometry (SOM) was performed in 536 (48.1%) and therapeutic ERCP in 579 (51.9%). Suspected sphincter of Oddi dysfunction (SOD) was the indication for the ERCP in 378 patients (33.9%). Pancreatitis developed in 168 patients (15.1%) and was graded mild in 112 (10%), moderate in 45 (4%), and severe in 11(1%). There was no difference in the incidence of pancreatitis or the frequency of investigated potential pancreatitis risk factors between the corticosteroid and placebo groups. By univariate analysis, the incidence of post-ERCP pancreatitis was significantly higher in 19 of 30 investigated variables. In the multivariate risk model, significant risk factors with adjusted odds ratios (OR) were: minor papilla sphincterotomy (OR: 3.8), suspected SOD (OR: 2.6), history of post-ERCP pancreatitis (OR: 2.0), age <60 yr (OR: 1.6), ≥2 contrast injections into the pancreatic duct (OR: 1.5), and trainee involvement (OR: 1.5). Female gender, history of recurrent idiopathic pancreatitis, pancreas divisum, SOM, difficult cannulation, and major papilla sphincterotomy (either biliary or pancreatic) were not multivariate risk factors for post-ERCP pancreatitis.CONCLUSION:This study emphasizes the role of patient factors (age, SOD, prior history of post-ERCP pancreatitis) and technical factors (number of PD injections, minor papilla sphincterotomy, and operator experience) as the determining high-risk predictors for post-ERCP pancreatitis.

A Randomized Trial of Rectal Indomethacin to Prevent Post-ERCP Pancreatitis

BACKGROUND Preliminary research suggests that rectally administered nonsteroidal antiinflammatory drugs may reduce the incidence of pancreatitis after endoscopic retrograde cholangiopancreatography (ERCP). METHODS In this multicenter, randomized, placebo-controlled, double-blind clinical trial, we assigned patients at elevated risk for post-ERCP pancreatitis to receive a single dose of rectal indomethacin or placebo immediately after ERCP. Patients were determined to be at high risk on the basis of validated patient- and procedure-related risk factors. The primary outcome was post-ERCP pancreatitis, which was defined as new upper abdominal pain, an elevation in pancreatic enzymes to at least three times the upper limit of the normal range 24 hours after the procedure, and hospitalization for at least 2 nights. RESULTS A total of 602 patients were enrolled and completed follow-up. The majority of patients (82%) had a clinical suspicion of sphincter of Oddi dysfunction. Post-ERCP pancreatitis developed in 27 of 295 patients (9.2%) in the indomethacin group and in 52 of 307 patients (16.9%) in the placebo group (P=0.005). Moderate-to-severe pancreatitis developed in 13 patients (4.4%) in the indomethacin group and in 27 patients (8.8%) in the placebo group (P=0.03). CONCLUSIONS Among patients at high risk for post-ERCP pancreatitis, rectal indomethacin significantly reduced the incidence of the condition. (Funded by the National Institutes of Health; ClinicalTrials.gov number, NCT00820612.).

Triple-tissue sampling at ERCP in malignant biliary obstruction

BACKGROUND Procurement of cytologic samples by brushing is common practice at endoscopic retrograde cholangiopancreatography (ERCP) but has low sensitivity for cancer detection. Limited data are available on other techniques, including endoluminal fine-needle aspiration and forceps biopsy. This series reviews the yield of these three stricture sampling methods. METHODS In this prospective study, patients with biliary obstruction with a clinical suspicion of malignancy underwent triple-tissue sampling at one ERCP session. Final cancer diagnosis was based on all sampling methods plus surgery, autopsy, and clinical follow-up. Tissue specimens were reported as normal, atypia, or malignant. RESULTS A total of 133 patients were evaluated: 104 had cancer and 29 had benign strictures. Tissue sampling sensitivity varied according to the type of cancer; the highest yield was seen in ampullary cancers (62% to 85%). The cumulative sensitivity of triple-tissue sampling in the cancer patients was as follows: sensitivity was 52% if atypia was considered benign and 77% if it was considered malignant. The addition of a second or third technique increased sensitivity rates in most instances. No serious complications occurred from the tissue sampling methods. CONCLUSIONS Tissue sampling sensitivity varied according to the type of cancer. Combining a second or third method increased sensitivity; general use of at least two sampling methods is therefore recommended.

Tissue sampling at ERCP in suspected malignant biliary strictures (Part 2)

Part I of this review in the previous issue of Gastrointestinal Endoscopy outlined the key points of tissue sampling at ERCP and considered intraductal bile aspiration cytology, cytologic/histologic analysis of retrieved plastic biliary stents and fine needle aspiration cytology. Specimen adequacy, slide preparation, and accuracy of slide interpretation, which is often influenced by the interpretation “philosophy” of individual cytopathologists, are fundamental to the effort to optimize cancer detection by using tissue sampling techniques at ERCP. Intraductal bile aspiration is simple and inexpensive but adds little to the other methods, which have higher rates of cancer detection. Therefore, bile aspiration is recommended only when other sampling techniques cannot be used. Cytologic evaluation of material from retrieved plastic biliary stents is relatively insensitive (32%) and impractical as a firstline approach to the diagnosis of malignant biliary obstruction because diagnosis is delayed until the stent is removed. However, it can be considered in patients undergoing stent exchange when other methods of tissue sampling fail to confirm the suspected diagnosis of malignancy. Endoscopic fineneedle aspiration (FNA) cytology has a lower rate of cancer detection (33%) than initially reported and is technically difficult. It is usually used to supplement other simpler methods. Part II of this review considers the remaining tissue sampling methods for use at ERCP: brush cytology, endobiliary forceps biopsy, and the multimodal tissue sampling. Methods for improving diagnostic yield are discussed. BRUSH CYTOLOGY

Large-diameter biliary orifice balloon dilation to aid in endoscopic bile duct stone removal: a multicenter series

BACKGROUND The utility and safety of endoscopic biliary orifice balloon dilation (EBD) for bile duct stone removal (with use of large-diameter balloons) after biliary endoscopic sphincterotomy (BES) is currently not well established. OBJECTIVE Our purpose was to evaluate the efficacy and complications of BES followed by > or = 12 mm diameter EBD for bile duct stone removal. DESIGN Retrospective, multicenter series. SETTING Five ERCP referral centers in the United States. PATIENTS AND INTERVENTIONS Patients who underwent attempted removal of bile duct stones by BES followed by EBD with > or = 12 mm diameter dilating balloons were identified by searching the prospectively recorded endoscopic databases from 1999 to 2007. Clinical parameters, endoscopic data, and outcomes were collected and analyzed. RESULTS One hundred three patients, mean age 70 +/- 17 years (range 23-98 years), with 56 (54%) women, underwent 107 procedures. Eleven patients (11%) had a prior history of acute pancreatitis. Pancreatogram was performed in 15 (14%) patients. Median stone size and median balloon diameter used was 13 mm. Complete stone removal in the first session of EBD was accomplished in 102 (95%) procedures, and mechanical lithotripsy was required in 29 (27%). Six patients (5.4%) had documented procedure-related complications including one patient with severe bleeding and one with severe cystic duct perforation. No acute pancreatitis occurred. CONCLUSION EBD with a large-diameter balloon in conjunction with BES for bile duct stone removal is effective and relatively safe. This technique appears to be a reasonable alternative option when standard BES and basket or balloon sweep are inadequate to remove bile duct stones.

Classification and management of perforations complicating endoscopic sphincterotomy.

BACKGROUND The management of perforations after endoscopic sphincterotomy (ES) is controversial. The purpose of this study was to analyze the treatments and outcome of patients with ES perforations. METHODS Between January 1994 and July 1998, in a series of 6040 endoscopic retrograde cholangiopancreatographies, 2874 (48%) ESs were performed: 40 patients (0.6%) with perforation were identified and retrospectively reviewed. RESULTS All patients (n = 14) with guidewire perforation (group I) were recognized early, managed medically, and discharged after a mean hospital stay of 3.5 days. Twenty of 22 patients with periampullary perforation (group II) were identified early; 18 patients (90%) had aggressive endoscopic drainage, and none required operation. Of the 2 patients identified late, 1 patient required operation and subsequently died. Mean hospital stay for this group was 8.5 days. Only 1 of 4 patients with duodenal perforations (group III) was identified early; all required operation; 1 patient died, and the mean hospital stay was 19.5 days. CONCLUSIONS ES perforation has 3 distinct types: guidewire, periampullary, and duodenal. Guidewire perforations are recognized early and resolve with medical treatment. Periampullary perforations diagnosed early respond to aggressive endoscopic drainage and medical treatment. Postsphincterotomy perforations diagnosed late (particularly duodenal) require surgical drainage, which carries a high morbidity and mortality rate.

Endoscopic snare papillectomy for tumors of the duodenal papillae

BACKGROUND Tumors of the major and the minor duodenal papillae can be malignant or premalignant, and traditionally are treated by surgical excision. This study evaluated the safety and the outcome of endoscopic snare resection of such tumors. METHODS All patients with tumors of the major or the minor papilla treated by endoscopic snare resection over a 10-year period (1994-2003) were identified from an ERCP database. Patients with tumors that had endoscopic features of malignancy and those proven to be cancerous by biopsy were excluded. Papillectomy was performed by electrosurgical snare resection. A pancreatic stent usually was placed before or after excision. Residual tumor was eradicated by repeated procedures. Endoscopic surveillance was at the discretion of the endoscopist. RESULTS Seventy snare resections were performed in 55 patients (mean age 59 years). Histopathologic diagnoses were the following: adenoma (45 patients; 7 with focal high-grade dysplasia, 6 with intraductal extension), adenocarcinoma (5), carcinoid tumor (2), gastric heterotopia (1), and normal histology (2). Fourteen patients had familial adenomatous polyposis. Of the 39 patients with isolated extraductal adenoma per cholangiogram, two underwent surgical resection because of persistent high-grade dysplasia, and 37 were successfully treated by endoscopic papillectomy alone. During follow-up (mean 30 months), 18 of 37 patients (49%) had no recurrence, 7 had recurrent adenoma (mean time interval to recurrence 27 months), two died of unrelated illnesses, and 10 are awaiting follow-up. Of the 6 patients with intraductal adenoma per cholangiogram, two underwent surgical resection, two had intraductal photodynamic therapy, and two had endoscopic snare resection. Intraductal tumor in the 4 latter patients was eliminated, although it recurred in one of the patients who had photodynamic therapy. Of the 7 patients with adenocarcinoma or carcinoid tumor, pancreaticoduodenectomy was performed in 3 and palliative papillectomy was performed in 4 unsuitable for surgery. One patient with carcinoid tumor of the minor papilla is alive, without recurrence, at 5 years after papillectomy. There were 10 procedure-related complications (14.5%), including pancreatitis (5), bleeding (4), and mild perforation (1). There was no procedure-related death. CONCLUSIONS Most adenomas of the duodenal papillae without intraductal extension can be fully resected by snare papillectomy. However, adenoma recurs in about a third of patients. Endoscopic therapy appears to be a reasonable alternative to surgery for management of papillary tumors. Longer follow-up is needed to determine the true recurrence rate and if endoscopic re-treatments are effective.

Frequency of abnormal pancreatic and biliary sphincter manometry compared with clinical suspicion of sphincter of Oddi dysfunction

BACKGROUND Sphincter of Oddi manometry as performed at ERCP is the most accepted method to evaluate for sphincter of Oddi dysfunction. To fully assess for sphincter of Oddi dysfunction, both the pancreatic and the bile ducts must be evaluated. We assessed the frequency of pancreatic and biliary sphincter abnormalities in a large series of patients. METHODS A total of 593 patients underwent manometry of the biliary and pancreatic ducts at one endoscopic retrograde cholangiopancreatography session. Basal sphincter pressure greater than or equal to 40 mm Hg was considered abnormal. Phasic waves were not evaluated. Manometric abnormalities were correlated with the clinical presentation as categorized using a modified Geenen/Hogan classification. RESULTS Of 360 patients with intact sphincters, 18.9% had abnormal pancreatic sphincter basal pressure alone, 11.4% had abnormal biliary basal sphincter pressure alone, and in 31.4% the basal pressure was abnormal for both sphincters; thus, 60.1% of the patients had sphincter dysfunction. The frequency of sphincter of Oddi dysfunction did not differ whether typed by biliary or pancreatic criteria: approximately 65% type II and 59% type III. Of patients without pancreatitis, 55.9% had an abnormal basal sphincter pressure, whereas sphincter dysfunction was present in 72.3% of those with idiopathic pancreatitis and 53.9% of patients with chronic pancreatitis. Of patients with an ablated biliary sphincter, 45.9% had abnormal basal pancreatic sphincter pressure and only 0.6% had an abnormal biliary sphincter pressure alone. Abnormal pressure in both sphincters was found in 9.3%. CONCLUSION If both portions of the sphincter of Oddi are studied simultaneously, abnormalities are found very commonly (55% to 72%). Assessment of both sides of the sphincter is necessary. Classifying patients according to both pancreatic and biliary sphincter segments is cumbersome, and may be replaced by an overall type. Using this modified classification, the frequency of sphincter of Oddi dysfunction is similar in both type II and type III patients (59% to 67%).

Improved stent characteristics for prophylaxis of post-ERCP pancreatitis

BACKGROUND & AIMS Pancreatic stenting is an effective method to prevent post-endoscopic retrograde cholangiopancreatography (ERCP) pancreatitis in high-risk patients. This retrospective study evaluated the impact of modified stent characteristics on the rate of post-ERCP pancreatitis, spontaneous stent dislodgment, and stent-related sequelae. METHODS A total of 2283 patients underwent 2447 ERCPs over a 6-year period with placement of 3-4F diameter, unflanged pancreatic stents. The indication for stenting was pancreatitis prophylaxis predominantly in suspected sphincter of Oddi dysfunction (SOD), pancreas divisum therapy, and precut sphincterotomy. An abdominal radiograph was obtained 10-14 days later to assess spontaneous stent passage. Post-ERCP pancreatitis was defined according to established criteria. A total of 479 patients underwent repeat ERCPs after an initial ERCP with pancreatic stent placement. The prestenting pancreatogram was then compared with follow-up studies. RESULTS The pancreatitis rate with 3F, 4F, 5F, and 6F stents was 7.5%, 10.6%, 9.8%, and 14.6%, respectively (3F vs. 4F, 5F, 6F: P = 0.047). Spontaneous stent dislodgment was 86%, 73%, 67%, and 65%, respectively (3F vs. 4F, 5F, 6F: P < 0.0001). The frequency of ductal changes was 24% in patients with 3-4F stents compared with 80% with 5-6F stents. Ductal perforation from the stents occurred in 3 patients (0.1%). CONCLUSIONS Small diameter (3-4F), unflanged pancreatic stents are more effective than the traditionally used stents (5-6F) in preventing post-ERCP pancreatitis. Stent-induced ductal changes and the need for endoscopic removal are also significantly less with 3-4F stents. The 3F stent appears to be superior in all aspects studied. Additional studies are needed to define the ideal method to eliminate post-ERCP pancreatitis.

Role of EUS for preoperative evaluation of cholangiocarcinoma: a large single-center experience.

BACKGROUND Accurate preoperative diagnosis and staging of cholangiocarcinoma (CCA) remain difficult. OBJECTIVE To evaluate the utility of EUS in the diagnosis and preoperative evaluation of CCA. DESIGN Observational study of prospectively collected data. SETTING Single tertiary referral hospital in Indianapolis, Indiana. PATIENTS Consecutive patients with CCA from January 2003 through October 2009. INTERVENTIONS EUS and EUS-guided FNA (EUS-FNA). MAIN OUTCOME MEASUREMENTS Sensitivity of EUS for the detection of a tumor and prediction of unresectability compared with CT and magnetic resonance imaging (MRI); sensitivity of EUS-FNA to provide tissue diagnosis, by using surgical pathology as a reference standard. RESULTS A total of 228 patients with biliary strictures undergoing EUS were identified. Of these, 81 (mean age 70 years, 45 men) had CCA. Fifty-one patients (63%) had distal and 30 (37%) had proximal CCA. For those with available imaging, tumor detection was superior with EUS compared with triphasic CT (76 of 81 [94%] vs 23 of 75 [30%], respectively; P < .001). MRI identified the tumor in 11 of 26 patients (42%; P = .07 vs EUS). EUS identified CCA in all 51 (100%) distal and 25 (83%) of 30 proximal tumors (P < .01). EUS-FNA (median, 5 passes; range, 1-12 passes) was performed in 74 patients (91%). The overall sensitivity of EUS-FNA for the diagnosis of CCA was 73% (95% confidence interval, 62%-82%) and was significantly higher in distal compared with proximal CCA (81% vs 59%, respectively; P = .04). Fifteen tumors were definitely unresectable. EUS correctly identified unresectability in 8 of 15 and correctly identified the 38 of 39 patients with resectable tumors (53% sensitivity and 97% specificity for unresectability). CT and/or MRI failed to detect unresectability in 6 of these 8 patients. LIMITATION Single-center study. CONCLUSION EUS and EUS-FNA are sensitive for the diagnosis of CCA and very specific in predicting unresectability. The sensitivity of EUS-FNA is significantly higher in distal than in proximal CCA.