Richard C.K. Wong

Risk factors for complications after performance of ERCP

BACKGROUND ERCP has become widely available for the diagnosis and treatment of benign and malignant pancreaticobiliary diseases. In this prospective study, the overall complication rate and risk factors for diagnostic and therapeutic ERCP were identified. METHODS Data were collected prospectively on patient characteristics and endoscopic techniques from 1223 ERCPs performed at a single referral center and entered into a database. Univariate and multivariate analyses were used to identify risk factors for ERCP-associated complications. RESULTS Of 1223 ERCPs performed, 554 (45.3%) were diagnostic and 667 (54.7%) therapeutic. The overall complication rate was 11.2%. Post-ERCP pancreatitis was the most common (7.2%) and in 93% of cases was self-limiting, requiring only conservative treatment. Bleeding occurred in 10 patients (0.8%) and was related to a therapeutic procedure in all cases. Nine patients had cholangitis develop, most cases being secondary to incomplete drainage. There was one perforation (0.08%). All other complications totaled 1.5%. Variables derived from cannulation technique associated with an increased risk for post-ERCP pancreatitis were precut access papillotomy (20%), multiple cannulation attempts (14.9%), sphincterotome use to achieve cannulation (13.1%), pancreatic duct manipulation (13%), multiple pancreatic injections (12.3%), guidewire use to achieve cannulation (10.2%), and the extent of pancreatic duct opacification (10%). Patient characteristics associated with an increased risk of pancreatitis were sphincter of Oddi dysfunction (21.7%) documented by manometry, previous ERCP-related pancreatitis (19%), and recurrent pancreatitis (16.2%). Pain during the procedure was an important indicator of an increased risk of post-ERCP pancreatitis (27%). Independent risk factors for post-ERCP pancreatitis were identified as a history of recurrent pancreatitis, previous ERCP-related pancreatitis, multiple cannulation attempts, pancreatic brush cytology, and pain during the procedure. CONCLUSIONS The most frequent ERCP-related complication was pancreatitis, which was mild in the majority of patients. The frequency of post-ERCP pancreatitis was similar for both diagnostic and therapeutic procedures. Bleeding was rare and mostly associated with sphincterotomy. Other complications such as cholangitis and perforation were rare. Specific patient- and technique-related characteristics that can increase the risk of post-ERCP complications were identified.

High-resolution endoscopic imaging of the GI tract using optical coherence tomography.

BACKGROUND Optical coherence tomography (OCT) has demonstrated the microscopic structure of the gastrointestinal (GI) tract mucosa and submucosa in vitro. We evaluated a prototype OCT system and assessed the feasibility of OCT in the human GI tract. METHODS The 2.4 mm diameter prototype OCT probe, inserted through an endoscope, provides a 360-degree radial scan. Images (6.7 frames/sec) are displayed on a television monitor. Tissue contact is not required. In patients undergoing elective endoscopy, OCT images were obtained of normal mucosa (confirmed by biopsy). RESULTS Seventy-two sites were imaged (38 patients): esophagus (21), stomach (12), duodenum (11), terminal ileum (4), colon (15), and rectum (9). Varying the distance between the probe and the mucosal surface produced images of the GI wall of varying depth. When held about 1 mm above the mucosal surface, the images consisted of mucosal structures such as colonic crypts, gastric pits, and duodenal villi. With the probe held against the wall, the OCT image comprised several layers interpreted as mucosa, muscularis mucosae, and submucosa. Structures including blood vessels were evident within the submucosa. A probe with a 0.5 mm working distance to the focal point provided the best images. Reducing the frame rate to 4.0 per second facilitated image interpretation. CONCLUSIONS OCT is feasible in the human GI tract and provides interpretable high-resolution images of mucosa and submucosa.

Real-time in vivo imaging of human gastrointestinal ultrastructure by use of endoscopic optical coherence tomography with a novel efficient interferometer design

We report on the design and initial clinical experience with a real-time endoscopic optical coherence tomography (EOCT) imaging system. The EOCT unit includes a high-speed optical coherence tomography interferometer, endoscope-compatible catheter probes, and real-time data capture and display hardware and software. Several technological innovations are introduced that improve EOCT efficiency and performance. In initial clinical studies using the EOCT system, the esophagus, stomach, duodenum, ileum, colon, and rectum of patients with normal endoscopic findings were examined. In these initial investigations, EOCT imaging clearly delineated the substructure of the mucosa and submucosa in several gastrointestinal organs; microscopic structures such as glands, blood vessels, pits, villi, and crypts were also observed.

Prospective assessment of the utility of EUS in the evaluation of gallstone pancreatitis

BACKGROUND The ability to identify common bile duct stones by noninvasive means in patients with acute biliary pancreatitis is limited. The aim of this study was to prospectively evaluate the ability of endosonography (EUS) to identify cholelithiasis and choledocholithiasis and predict disease severity in patients with nonalcoholic pancreatitis. METHODS EUS was performed immediately before endoscopic retrograde cholangiopancreatography (ERCP) by separate blinded examiners within 72 hours of admission. Gallbladder findings were compared between EUS and transabdominal ultrasonography (US). Using endoscopic extraction of a bile duct stone as the reference standard for choledocholithiasis, the diagnostic yield of EUS was compared with transabdominal US and ERCP. Features identified during endosonographic imaging of the pancreas were correlated with length of hospitalization. RESULTS Thirty-six patients were studied. EUS and transabdominal US were concordant in their interpretation of gallbladder findings in 92% of patients. The sensitivity of transabdominal US, EUS, and ERCP for identifying choledocholithiasis was 50%, 91%, and 92% and the accuracy was 83%, 97%, and 89%, respectively. Length of hospital stay was longer in patients with peripancreatic fluid (9.2 vs. 5.7 days, p < 0.1) and shorter in patients with coarse echo texture (2.6 vs. 7.2 days, p < 0.05) demonstrated on EUS. CONCLUSIONS EUS can reliably identify cholelithiasis and is more sensitive than transabdominal US in detecting choledocholithiasis in patients with biliary pancreatitis. EUS may be used early in the management of patients with acute pancreatitis to select those who would benefit from endoscopic stone extraction. The utility of EUS for predicting pancreatitis severity requires further investigation.

Safety of ERCP during pregnancy.

OBJECTIVES:There are few data in the literature regarding the indications, therapy, and safety of endoscopic management of pancreatico-biliary disorders during pregnancy. We report the largest single center experience with ERCP in pregnancy.METHODS:We reviewed 15 patients that underwent ERCP during pregnancy. In all patients, the pelvis was lead-shielded and the fetus was monitored by an obstetrician. Fluoroscopy was minimized and hard copy radiographs taken only when essential.RESULTS:The mean patient age was 28.9 yr (15–36 yr). The mean duration of gestation was 25 wk (12–33 wk); one patient was in the first, five in the second, and nine in the third trimester. The indications were gallstone pancreatitis (n = 6), choledocholithiasis on ultrasound (n = 5), elevated liver enzymes and a dilated bile duct on ultrasound (n = 2), abdominal pain and gallstones (n = 1), and chronic pancreatitis (n = 1). ERCP findings were bile duct stones (n = 6), patulous papilla (n = 1), bile duct debris (n = 1), normal bile duct and gallstones or gallbladder sludge (n = 3), dilated bile duct and gallstones (n = 1), normal bile duct and no gallstones (n = 2), and chronic pancreatitis (n = 1). Six patients underwent sphincterotomies and one a biliary stent insertion. One sphincterotomy was complicated by mild pancreatitis. All infants delivered to date have had Apgar-scores >8, and continuing pregnancies are uneventful. Mean fluorosocopy time was 3.2 min (SD ± 1.8). An estimated fetal radiation exposure was 310 mrad (SD ± 164) which is substantially below the accepted teratogenic dose.CONCLUSIONS:ERCP in pregnancy seems to be safe for both mother and fetus; however, it should be restricted to therapeutic indications with additional intraprocedure safety measures.

Prediction of outcome in acute lower-gastrointestinal haemorrhage based on an artificial neural network: internal and external validation of a predictive model

BACKGROUND Models based on artificial neural networks (ANN) are useful in predicting outcome of various disorders. There is currently no useful predictive model for risk assessment in acute lower-gastrointestinal haemorrhage. We investigated whether ANN models using information available during triage could predict clinical outcome in patients with this disorder. METHODS ANN and multiple-logistic-regression (MLR) models were constructed from non-endoscopic data of patients admitted with acute lower-gastrointestinal haemorrhage. The performance of ANN in classifying patients into high-risk and low-risk groups was compared with that of another validated scoring system (BLEED), with the outcome variables recurrent bleeding, death, and therapeutic interventions for control of haemorrhage. The ANN models were trained with data from patients admitted to the primary institution during the first 12 months (n=120) and then internally validated with data from patients admitted to the same institution during the next 6 months (n=70). The ANN models were then externally validated and direct comparison made with MLR in patients admitted to an independent institution in another US state (n=142). FINDINGS Clinical features were similar for training and validation groups. The predictive accuracy of ANN was significantly better than that of BLEED (predictive accuracy in internal validation group for death 87% vs 21%; for recurrent bleeding 89% vs 41%; and for intervention 96% vs 46%) and similar to MLR. During external validation, ANN performed well in predicting death (97%), recurrent bleeding (93%), and need for intervention (94%), and it was superior to MLR (70%, 73%, and 70%, respectively). INTERPRETATION ANN can accurately predict the outcome for patients presenting with acute lower-gastrointestinal haemorrhage and may be generally useful for the risk stratification of these patients.

Acute pancreatitis after EUS-guided FNA of solid pancreatic masses: a pooled analysis from EUS centers in the United States

BACKGROUND The aim of this study was to determine the frequency and the severity of pancreatitis after EUS-guided FNA of solid pancreatic masses. A survey of centers that offer training in EUS in the United States was conducted. METHODS A list of centers in which training in EUS is offered was obtained from the Web site of the American Society for Gastrointestinal Endoscopy. Designated program directors were contacted via e-mail. The information requested included the number of EUS-guided FNA procedures performed for solid pancreatic masses, the number of cases of post-procedure pancreatitis, and the method for tracking complications. For each episode of pancreatitis, technical details were obtained about the procedure, including the location of the mass, the type of fine needle used, the number of needle passes, and the nature of the lesion. RESULTS Nineteen of the 27 programs contacted returned the questionnaire (70%). In total, 4909 EUS-guided FNAs of solid pancreatic masses were performed in these 19 centers over a mean of 4 years (range 11 months to 9 years). Pancreatitis occurred after 14 (0.29%): 95% CI[0.16, 0.48] procedures. At two centers in which data on complications were prospectively collected, the frequency of acute pancreatitis was 0.64%, suggesting that the frequency of pancreatitis in the retrospective cohort (0.26%) was under-reported (p=0.22). The odds that cases of pancreatitis would be reported were 2.45 greater for the prospective compared with the retrospective cohort (95% CI[0.55, 10.98]). The median duration of hospitalization for treatment of pancreatitis was 3 days (range 1-21 days). The pancreatitis was classified as mild in 10 cases, moderate in 3, and severe in one; one death (proximate cause, pulmonary embolism) occurred after the development of pancreatitis in a patient with multiple comorbid conditions. CONCLUSIONS EUS-guided FNA of solid pancreatic masses is infrequently associated with acute pancreatitis. The procedure appears to be safe when performed by experienced endosonographers. The frequency of post EUS-guided FNA pancreatitis may be underestimated by retrospective analysis.

Palliation of malignant gastric outlet obstruction using an endoscopically placed Wallstent

BACKGROUND Treatment options for malignant gastric outlet obstruction are limited. Surgical gastrojejunostomy, commonly performed, has significant morbidity and mortality. METHODS Over 2 years, we prospectively studied the safety, feasibility, and outcomes for use of a newly designed expandable metal stent (Wallstent Enteral; Schneider, Minneapolis, Minn.) to treat malignant gastric outlet obstruction. Stents 16 to 22 mm in diameter and 60 to 90 mm in length were deployed directly through the endoscope. RESULTS Twelve patients (ten women, two men; mean age 59.7 years) underwent stenting. Thereafter, six patients were able to eat a regular diet; three could eat pureed food. In three patients, the procedure was unsuccessful because of multiple obstructions not recognized before stenting (one) and stents deployed too proximally (one) or too distally (one). CONCLUSIONS Placement of a newly designed stent through the endoscope is safe and effective palliation for various types of malignant gastric outlet obstruction and significantly improves many aspects of patient quality of life.

Prediction of outcome of acute GI hemorrhage: a review of risk scores and predictive models.

Patients with acute GI hemorrhage are encountered on an almost daily basis by gastroenterologists, endoscopists, emergency room physicians, and other health care personnel. Acute GI hemorrhage encompasses a spectrum of bleeding lesions from the upper GI tract and the small bowel to the colon and the rectum. It is associated with substantial patient morbidity and mortality, as well as economic cost. 1-6 Over the last few decades, the diagnosis and the management of acute GI hemorrhage has undergone remarkable changes, particularly with the rapid dissemination and advancement of diagnostic and therapeutic endoscopy and pharmacotherapeutics. Moreover, by extensive early application of endoscopic evaluation, the natural history and the diverse etiology of acute GI hemorrhage is much better understood today than in the past. Physicians have long been aware that the clinical outcome of an episode of acute GI hemorrhage can be predicted with reasonable accuracy and that such a prediction may be useful in patient management. Although it is difficult to prove a beneficial impact on overall patient mortality, a strong argument can be made that improved medical decision making, particularly at the initial patient triage level, may lead to improvements in patient management, outcome, and resource utilization. 2,7-15 Thus, with the current focus on the economic aspects of health care delivery, it is not surprising that there is increased interest in the early risk stratification of patients with acute GI hemorrhage so that they can be triaged to appropriate, cost-effective levels of care. This review summarizes and critically appraises the major clinical scoring systems promulgated in

An immunodeficiency characterized by defective signal transduction in T lymphocytes.

We studied a nine-year-old boy with severe, recurrent infections. The patient was exposed in utero to azathioprine and prednisone. He had autoimmune hemolytic anemia, bronchiectasis, and Hodgkins disease. The patients circulating lymphocytes were normal in number and phenotype, but stimulation of the T-cell receptor by antigens, mitogens, and monoclonal antibodies failed to induce interleukin-2-receptor expression, interleukin-2 synthesis, or lymphocyte proliferation. The early biochemical events necessary to initiate lymphocyte activation--accumulation of the second messenger diacylglycerol, activation of the enzyme protein kinase C, and elevation of the free intracellular calcium concentration--failed to occur in this patients lymphocytes. The defect in the lymphocyte could be corrected in vitro by two agents that bypass the receptor-mediated signal mechanism (the diacylglycerol analogue phorbol and the calcium ionophore ionomycin). Further studies localized the defect in signal transduction to the interaction between cell-surface receptors and the guanine nucleotide-binding protein. We conclude that this patients immunodeficiency was caused by a defective coupling of surface receptors to signal-transducing proteins in his T lymphocytes, resulting in failure of lymphocyte activation.