Clinical value of the Integrated Pulmonary Index® during sedation for interventional upper GI-endoscopy: A randomized, prospective tri-center study.

BACKGROUND AND STUDY AIMS: The Integrated Pulmonary Index® (IPI) is a mathematically-determined factor based on parameters of capnography and pulse oximetry, which should enable sensitive detection of impaired respiratory function. Aim was to investigate whether an additional measurement of the IPI during sedation for interventional endoscopy, compared to standard monitoring alone, allows a reduction of sedation-related respiratory depression. PATIENTS AND METHODS: 170 patients with standard monitoring randomly underwent either a blinded recording of capnography (control group, n=87) or capnography, including automated IPI calculation (IPI group, n=83), during deep sedation with midazolam and propofol. The primary endpoint was the maximum decrease of oxygen saturation from the baseline level before sedation. Secondary endpoints: incidence of hypoxemia (SaO2<90%), other sedation-related complications (apnea rate, bradycardia, hypotension), patient cooperation and satisfaction (VAS). RESULTS: Mean propofol dose in the IPI group (245±61mg) was comparable to the control group (225±47mg). The average drop of the oxygen saturation in the IPI group (6.5±4.1%) was nearly identical to that of the control group (7.1±4.6%, p=0.44). Apnea episodes >15s was found in 46 patients of the control and 31 of the IPI group (p<0.05). Frequency of occurrence of a drop in pO2-saturation <90%, bradycardia <50/min or a drop of systolic pressure <90mmHg were not significantly different in both groups. Mechanical ventilation was not required in any case. Patient cooperation and satisfaction were assessed similar in both groups. CONCLUSION: A clinically appealing advantage of IPI-assessment during deep sedation with midazolam and propofol for interventional endoscopy could not be documented. However, IPI registration was more effective in reducing the incidence of apnea episodes.

Prospective evaluation of the routine use of a nasopharyngeal airway (Wendl Tube) during endoscopic propofol-based sedation.

BACKGROUND/AIMS: Insertion of a nasopharyngeal airway (NPA) during endoscopic sedation is only recommended in the event of respiratory problems. We evaluated the safety and efficacy of routine insertion of an NPA during sedation in gastrointestinal (GI) endoscopy. METHODS: Between July 2009 and April 2012, patients with colonoscopy or expected longer-lasting or therapeutic upper GI endoscopy were pseudo-randomized in a weekly alternating fashion to perform sedation (midazolam in combination with propofol) with or without NPA insertion. The primary outcome measure was respiratory depression (oxygen saturation <90%). Secondary measures included hypotension (systolic blood pressure <90 mm Hg), bradycardia (heart rate <40 beats/min) or nasopharyngeal damage after NPA insertion. RESULTS: 216 (106 females, mean age 60.7 ± 9.65 years) were enrolled. Colonoscopy was performed in 131 patients and upper endoscopy in 85 patients. In 105 patients an NPA was used (intervention group). Five (4.7%) of those patients showed minor nasopharyngeal injury. Respiratory depression (13.5 vs. 1.9%, p = 0.002) and hypotension (11 vs. 5%, p = 0.09) occurred more frequently in the control than in the intervention group. CONCLUSION: The routine placement of an NPA can reduce the frequency of hypoxemic events during endoscopic sedation with minor risks for nasopharyngeal injury.

Endoscopic approach to achalasia.

Achalasia is a primary esophageal motor disorder. The etiology is still unknown and therefore all treatment options are strictly palliative with the intention to weaken the lower esophageal sphincter (LES). Current established endoscopic therapeutic options include pneumatic dilation (PD) or botulinum toxin injection. Both treatment approaches have an excellent symptomatic short term effect, and lead to a reduction of LES pressure. However, the long term success of botulinum toxin (BT) injection is poor with symptom recurrence in more than 50% of the patients after 12 mo and in nearly 100% of the patients after 24 mo, which commonly requires repeat injections. In contrast, after a single PD 40%-60% of the patients remain asymptomatic for ≥ 10 years. Repeated on demand PD might become necessary and long term remission can be achieved with this approach in up to 90% of these patients. The main positive predictors for a symptomatic response to PD are an age > 40 years, a LES-pressure reduction to < 15 mmHg and/or an improved radiological esophageal clearance post-PD. However PD has a significant risk for esophageal perforation, which occurs in about 2%-3% of cases. In randomized, controlled studies BT injection was inferior to PD and surgical cardiomyotomy, whereas the efficacy of PD, in patients > 40 years, was nearly equivalent to surgery. A new promising technique might be peroral endoscopic myotomy, although long term results are needed and practicability as well as safety issues must be considered. Treatment with a temporary self expanding stent has been reported with favorable outcomes, but the data are all from one study group and must be confirmed by others before definite recommendations can be made. In addition to its use as a therapeutic tool, endoscopy also plays an important role in the diagnosis and surveillance of patients with achalasia.

Miniprobe endoscopic ultrasound accurately stages esophageal cancer and guides therapeutic decisions in the era of neoadjuvant therapy: results of a multicenter cohort analysis.

BACKGROUND: Despite recent advances in imaging techniques, adequate classification of esophageal lesions is still challenging. Accurate staging of tumors of the esophagus is a precondition for targeted therapy. In this retrospective, multicenter study, we report the role of high-frequency endoscopic ultrasound (EUS) catheter probes in pretherapeutic staging of esophageal neoplasms and thus guiding treatment decisions. METHODS: A total of 143 patients (mean age of 63.8 ± 10.7 years) with esophageal carcinoma were recruited from five German centers (Münster, Oldenburg, Hannover, Wiesbaden, and Lüneburg). Tumor type was adenocarcinoma in 112 (78 %) cases and squamous cell carcinoma in 31 (22 %). Tumor localization was as follows: proximal 3, mid esophagus 7, and distal third 133. Histological correlation either through EMR or surgery was available. In all patients, pretherapeutic uT and uN classifications were compared to pT/pN classification obtained from surgically (esophagectomy, n = 93) or endoscopically (EMR, n = 50) resected tissue. RESULTS: Overall, accuracy of uT classification was 60 % and of uN classification was 74 %. Sensitivity, specificity, and accuracy rates for local tumor extension were as follows (%): T1: 68/97/83; T2: 39/84/75; T3: 72/81/79; T4: 13/97/93; T1/2: 73/81/75; T3/4: 78/82/81. Relating to positive lymph node detection, sensitivity and specificity were 76 and 71 %, respectively. CONCLUSIONS: Miniprobe EUS is an established method for the staging of esophageal tumors. Our large multicenter cohort shows a solid accuracy of miniprobe EUS with respect to differentiating locally advanced from limited cancer and assisting to determine the treatment regimen in the era of neoadjuvant therapy; consequently, 78 % of patients would have been assigned to the adequate therapeutic regimen, whereas 11 % of patients would have been overtreated and 11 % undertreated.

Capnographic monitoring reduces the incidence of arterial oxygen desaturation and hypoxemia during propofol sedation for colonoscopy: a randomized, controlled study (ColoCap Study).

OBJECTIVES: The aim of this randomized study was to determine whether intervention based on additional capnographic monitoring reduces the incidence of arterial oxygen desaturation during propofol sedation for colonoscopy. METHODS: Patients (American Society of Anesthesiologists classification (ASA) 1-3) scheduled for colonoscopy under propofol sedation were randomly assigned to either a control arm with standard monitoring (standard arm) or an interventional arm in which additional capnographic monitoring (capnography arm) was available. In both study arms, detection of apnea or altered respiration induced withholding propofol administration, stimulation of the patient, chin lift maneuver, or further measures. The primary study end point was the incidence of arterial oxygen desaturation (defined as a fall in oxygen saturation (SaO(2)) of ≥5% or <90%); secondary end points included the occurrences of hypoxemia (SaO(2) <90%), severe hypoxemia (SaO(2) ≤85%), bradycardia, hypotension, and the quality of sedation (patient cooperation and patient satisfaction). RESULTS: A total of 760 patients were enrolled at three German endoscopy centers. The intention-to-treat analysis revealed a significant reduction of the incidence of oxygen desaturation in the capnography arm in comparison with the standard arm (38.9% vs. 53.2%; P<0.001). The numbers of patients with a fall in SaO(2) <90% and ≤85% were also significantly different (12.5% vs. 19.8%; P=0.008 and 3.7 vs. 7.8%; P=0.018). There were no differences regarding the rates of bradycardia and hypotension. Quality of sedation was similar in both groups. Results of statistical analyses were maintained for the per-protocol population. CONCLUSIONS: Additional capnographic monitoring of ventilatory activity reduces the incidence of oxygen desaturation and hypoxemia during propofol sedation for colonoscopy.

How best to approach endoscopic sedation?

Sedation is the drug-induced reduction of a patient's consciousness. The aim of sedation in endoscopic procedures is to increase the patient's comfort and to improve endoscopic performance, especially in therapeutic procedures. The most commonly used sedation regimen for conscious sedation in gastrointestinal endoscopy is still the combination of benzodiazepines with opioids. However, the use of propofol has increased enormously in the past decade and several studies show advantages of propofol over the traditional regimes in terms of faster recovery time. It is important to be aware that the complication rate of endoscopies increases when sedation is used; therefore, a thorough risk evaluation before the procedure and monitoring during the procedure must be performed. In addition, properly trained staff and emergency equipment should be available. The best approach to sedation in endoscopy is to choose a sedation regimen for the individual patient, tailored according to the clinical risk assessment and the anxiety level of the patient, as well as to the type of planned endoscopic procedure.

Extended monitoring of the sedated patient: bispectral index, Narcotrend and automated responsiveness monitor.

It has been demonstrated that a processed EEG derivative may reflect the patient's level of consciousness during endoscopic sedation. However, to date a clinical benefit of EEG monitoring with respect to relevant safety parameters in routine gastrointestinal endoscopy has not been demonstrated. During long-lasting procedures (e.g. ERCP, ESD) a propofol-sparing effect has been shown, but a benefit regarding patient safety could not be documented. The use of an automated responsiveness monitor may help to achieve and to maintain the level of conscious sedation with propofol, but its precise role in endoscopic sedation is undefined.

Propofol sedation in gastrointestinal endoscopy: a gastroenterologist's perspective.

The present article describes the recommendations regarding the use of propofol by non-anesthesiologists from published guidelines. Furthermore, safety and efficacy data regarding the use of propofol in the hands of gastroenterologists are also reviewed. Although there are no studies comparing the safety and efficacy of propofol administration by anesthesiologists versus non-anesthesiologists for sedation during endoscopy, there is strong evidence that propofol administration by non-anesthesiologists is safe and efficacious for the majority of patients undergoing routine endoscopic procedures.

Dual access endoscopic necrosectomy of infected pancreatic necrosis: a case report.

A 73-year-old male developed fever and jaundice 6 months after an episode of acute necrotizing pancreatitis. During endoscopic retrograde cholangiography, a distal bile duct compression was documented and stent insertion led to resolution of jaundice, however, the febrile condition persisted. A pancreatic necrosis measuring 11x7 cm was shown by computed tomography (CT) and the patient was referred for necrosectomy. During the first endoscopic session, spontaneous drainage of pus was observed in the duodenal bulb. Therefore, the pancreatic necrosis was first punctured under endoscopic ultrasound-guidance transduodenally. The pancreatic necrosis was then additionally punctured transgastrically and the necrotic cavity was entered with a standard upper gastrointestinal scope. Despite extensive irrigation and necrosectomy we felt the transgastric approach was not sufficient enough to treat the large necrotic cavity and decided to perform the further treatment by using both accesses. Endoscopic debridement was repeated daily through the transgastric as well as the transduodenal approach over 5 days. The clinical condition of the patient dramatically improved and he became afebrile. Two months after the initial endoscopic necrosectomy, a CT scan showed nearly complete resolution of the pancreatic necrosis and the bile duct stenosis resolved. Six months later, CT scans showed no residual necrosis and an atrophic but otherwise normal pancreas.

Propofol sedation for upper gastrointestinal endoscopy in patients with liver cirrhosis as an alternative to midazolam to avoid acute deterioration of minimal encephalopathy: a randomized, controlled study.

OBJECTIVE: Midazolam sedation for upper gastrointestinal (GI) endoscopy exacerbates minimal hepatic encephalopathy (HE) in patients with liver cirrhosis, therefore an alternative drug regimen for these patients is warranted. The aim of this randomized, controlled study was to assess whether the use of the short-acting propofol as a sedative for GI endoscopy could prevent the exacerbation of minimal HE in patients with liver cirrhosis. MATERIAL AND METHODS: The study comprised patients with liver cirrhosis without clinical HE who had undergone upper GI endoscopy for therapeutic purposes (intended variceal band ligation). Sixty patients were randomly assigned into two groups to receive propofol (n=40) or midazolam (n=20) for upper GI endoscopy. The study groups were matched for age, gender and Child-Pugh score. All patients completed number connecting tests (NCTs), as well as a porto-systemic encephalopathy (PSE) syndrome test before and at 2 h after completion of the endoscopic procedure. Time needed to fulfill the tests was documented. Baseline results of the psychomotor test batteries were compared with the post-interventional evaluations. Data were also compared with the results of a healthy control group (n=20) that did not undergo endoscopic sedation. Recovery time and quality (score system) were evaluated. RESULTS: The differences in the NCT times before and after sedation (median delta NCT, midazolam group, 11 s (95% CI, -1.2 to 16.1 s) versus the propofol group, -9.5 s (95% CI, -15.7 to -4.6 s), p=0.002) and in the PSE scores (median delta PSE, midazolam group, -1 (95% CI, -1.5 to 0.2) versus the propofol group, 1 (95% CI, 0.5 to 1.5), p=0.0009) differed significantly between the two groups. In addition, the recovery time and quality in patients receiving propofol were significantly improved compared with in the midazolam group (7.8+/-2.9 min versus 18.4+/-6.7 min, 6.1+/-1.1 versus 8.2+/-1.3, both p<0.001). CONCLUSIONS: The study demonstrates that propofol sedation for upper GI endoscopy does not cause acute deterioration of minimal hepatic encephalopathy and is associated with improved recovery in patients with liver cirrhosis. Propofol should be recommended for these patients as an alternative to midazolam.

Endoscopist-directed administration of propofol: a worldwide safety experience.

BACKGROUND & AIMS: Endoscopist-directed propofol sedation (EDP) remains controversial. We sought to update the safety experience of EDP and estimate the cost of using anesthesia specialists for endoscopic sedation. METHODS: We reviewed all published work using EDP. We contacted all endoscopists performing EDP for endoscopy that we were aware of to obtain their safety experience. These complications were available in all patients: endotracheal intubations, permanent neurologic injuries, and death. RESULTS: A total of 646,080 (223,656 published and 422,424 unpublished) EDP cases were identified. Endotracheal intubations, permanent neurologic injuries, and deaths were 11, 0, and 4, respectively. Deaths occurred in 2 patients with pancreatic cancer, a severely handicapped patient with mental retardation, and a patient with severe cardiomyopathy. The overall number of cases requiring mask ventilation was 489 (0.1%) of 569,220 cases with data available. For sites specifying mask ventilation risk by procedure type, 185 (0.1%) of 185,245 patients and 20 (0.01%) of 142,863 patients required mask ventilation during their esophagogastroduodenoscopy or colonoscopy, respectively (P < .001). The estimated cost per life-year saved to substitute anesthesia specialists in these cases, assuming they would have prevented all deaths, was $5.3 million. CONCLUSIONS: EDP thus far has a lower mortality rate than that in published data on endoscopist-delivered benzodiazepines and opioids and a comparable rate to that in published data on general anesthesia by anesthesiologists. In the cases described here, use of anesthesia specialists to deliver propofol would have had high costs relative to any potential benefit.

Sedation with propofol for interventional endoscopic procedures: a risk factor analysis.

OBJECTIVE: Propofol sedation for mainly diagnostic endoscopic procedures has proved safe in recent trials, with no need for endotracheal intubation. However, there is evidence that cardiorespiratory side effects occur more frequently and that assisted ventilation may be necessary if propofol sedation is performed for interventional endoscopic procedures. MATERIAL AND METHODS: Over a 6-year period, all adverse events (defined as premature termination of the procedure due to sedation-related events or either the need for assisted ventilation or admission to ICU) occurring during 9547 endoscopic interventions (UGI, n = 5.374, ERCP, n = 3.937, EUS, n=236) under propofol sedation were assessed. RESULTS: A total of 135 adverse events (1.4%) were documented. Assisted ventilation was necessary in 40 patients (0.4%); 9 patients required endotracheal intubation (0.09%); 28 needed further monitoring on the ICU (0.3%); and 4 patients died, 3 potentially due to sedation-related side effects (mortality, 0.03%). Independent risk factors for sedation-related side effects were emergency endoscopic examinations and a total propofol dose >100 mg. CONCLUSIONS: Interventional endoscopy under propofol sedation is not risk-free. Increased attention must be focused on close monitoring of vital parameters, particularly when undertaking long-lasting interventions and emergency procedures.

Routine colonoscopy with a standard gastroscope. A randomized comparative trial in a western population.

BACKGROUND AND AIMS: Colorectal cancer screening increased the workload of colonoscopic procedures in endoscopic units. Recent developments advocated the use of smaller and more flexible scopes to achieve the goal of a complete examination to the cecum. Therefore, the use of an upper GI scope for colonoscopy can be considered. MATERIALS AND METHODS: Six-hundred and fifty consecutive patients (age, 64 +/- 20 years, 395 women) undergoing routine colonoscopy were examined with either a standard colonoscope or an upper GI scope in randomized order. In case of an incomplete examination, colonoscopy was repeated with the alternative instrument in the same session. All patients underwent bowel preparation with 4 1 of a polyethylene glycol solution, and the examinations were performed under conscious sedation (midazolam and pethidine i.v.) by a single investigator. RESULTS: Because of insufficient colonic preparation or refusal to participate, 28 patients had to be excluded. Both groups (colonoscope group, n = 315, upper GI scope group, n = 307) were well comparable with respect to their demographic data, previous abdominal surgery, the presence of diverticulosis, and the number of former colonoscopic examinations, respectively. Successful cecum intubation was achieved in 96% of the cases in the colonoscope group and in 93% of the patients from the upper GI scope group (p = 0.82). However, the time until the cecum was reached was prolonged to 8.7 +/- 2.4 min when using the upper GI scope compared with 8.2 +/- 2.3 min in the colonoscope group (p = 0.006). In the colonoscope group, a switch to the upper GI scope enables a complete colonoscopy in all but three cases (11/14, 79%) whereas this aim was only achieved vice versa in the upper GI scope group (using a colonoscope) in 7 of 21 patients (33%, p = 0.04). CONCLUSION: Routine colonoscopy can be performed effectively with standard upper GI scopes in a western population if no colonoscope is available.

Endoscopic pancreatic duct stenting for relief of pancreatic cancer pain.

INTRODUCTION: Little is known about the effects of endoscopic pancreatic duct (PD) decompression in patients with 'obstructive type' pain from pancreatic carcinoma. METHODS: Twenty patients with unresectable carcinoma of the pancreas, PD obstruction and postprandial epigastric pain were enrolled. The pain intensity, opioid dose and quality-of-life index were documented pre-treatment and at 4-weekly intervals. PD stenting was attempted in all patients. PD stent change was performed if biliary stents had to be changed or intense pain relapsed. RESULTS: Endoscopic PD drainage was successful in 19/20 patients by placement of a 7 F (n = 9) or 10 F (n = 10) plastic stent. Median follow-up was 16 weeks. Two patients were alive at the end of follow-up. Eleven patients received gemcitabine chemotherapy. The pre-interventional pain score was 6.7 +/- 0.9 points and decreased to 3.1 +/-1.4 points at 4 weeks (P < 0.001). It remained lowered significantly at 8/12 weeks and at the final visit (4.2 +/- 1.5 points, n = 13). The mean pre-stenting fentanyl dose was 85.5 +/- 34.7 microg/h, and it was decreased to 57.9 +/- 39.1 microg/h after 4 weeks (P < 0.01), 60.5 +/- 38.9 microg/h after 8 weeks and 64.1+/-39.8 microg/h (P < 0.01 versus pre-treatment) after 12 weeks but increased to 82.7 +/- 41.3 microg/h (NS) at the final visit. The quality-of-life index improved 4 weeks after stenting (from 4.8+/-1 to 6.2+/-1.5 points, P < 0.01) but was lowered to 5.5 +/- 2.3 points at 12 weeks (NS). The response was independent of stent diameter and chemotherapy. CONCLUSION: PD stenting achieved significant pain relief and short-term improvement of the quality of life in the majority of patients with PD obstruction due to pancreatic carcinoma.

Endoscopic debridement of paraesophageal, mediastinal abscesses: a prospective case series.

BACKGROUND: Mediastinal abscesses after esophageal perforation or postoperative leakage nearly always require surgical intervention. METHODS: Patients with paraesophageal abscesses were treated with EUS-guided or endoscopic mediastinal puncture if the abscess was >2 cm and sepsis was present. Abscess cavities were entered with a 9.5-mm endoscope after balloon dilation to allow irrigation and drainage. Debris was removed with a Dormia basket. Concomitant pleural effusions were treated with transthoracic drains. Patients received intravenous antibiotics and enteral/parenteral nutrition. RESULTS: Twenty patients fulfilled the entry criteria. Simple drainage was sufficient in 4 cases, and puncture was impossible in one case. Of the 15 treated patients (age 39-76 years, 5 women) the etiology of perforation was Boerhaave's syndrome (n = 8), anastomotic leak (n = 3), and iatrogenic perforation (n = 4). Debridement was successful in all cases and required a median of 5 daily sessions (range 3-10). All patients became apyrexial, with a C-reactive protein < 5 mg/L within a median of 4 days (range 2-8 days). Esophageal defects were closed with endoclips (n = 7), fibrin glue (n = 4), metal stents (n = 1), or spontaneously healed (n = 3). One patient died from a massive pulmonary embolism one day after successful debridement (mortality 7%). No other complications were seen. Median follow-up was 12 months (range 3-40 months). CONCLUSIONS: Nonoperative endoscopic transesophageal debridement of mediastinal abscesses appears safe and effective.

Hemorrhoidal elastic band ligation with flexible videoendoscopes: a prospective, randomized comparison with the conventional technique that uses rigid proctoscopes.

BACKGROUND: Elastic band ligation by means of a rigid proctoscope is the treatment of choice for patients with symptoms caused by internal hemorrhoids of grade 2 to 3. In contrast to the flexible videoendoscope, the rigid proctoscope has limited maneuverability, has a narrower field of view, and does not allow adequate documentation. Therefore, a randomized trial was conducted to compare the safety and the efficacy of conventional elastic band ligation with videoendoscopic elastic band ligation. METHODS: A total of 100 consecutive patients (mean age 47 [12] years) with chronically bleeding grade 2 or 3 internal hemorrhoids were randomized to elastic band ligation or videoendoscopic elastic band ligation. For videoendoscopic elastic band ligation, a reusable multiband ligator was attached to the end of a therapeutic upper videoendoscope. From one to 3 bands were placed per session in both groups. Re-treatment was performed every 2 to 3 weeks in both groups until cessation of bleeding and eradication of the hemorrhoids (at least grade 2) were achieved. Thereafter, the patients were followed to assess complications and efficacy. Recurrent bleeding was considered a treatment failure. RESULTS: To achieve the desired therapeutic aims, a significantly lower number of treatment sessions was required in the videoendoscopic elastic band ligation group (1.8 [0.8] vs. 2.4 [0.9]; p < 0.01) and the total number of bands applied was significantly less (2.8 [1.1] vs. 3.7 [1.4]; p < 0.01). Pain was noted after ligation by 25% of patients in the elastic band ligation group compared with 27% of those who had videoendoscopic elastic band ligation. However, analgesic medications were required in only 7% after elastic band ligation and 9% after videoendoscopic elastic band ligation (NS). Post-ligation bleeding that had to be treated endoscopically occurred in 3.5% of the patients of the elastic band ligation group and 3.2% of those in the videoendoscopic elastic band ligation group (NS). Blood transfusion was not required. At a median follow-up of 12 months, there was no recurrence of bleeding in 40 patients (80%) in the conventional elastic band ligation group vs. 43 (86%) in the videoendoscopic elastic band ligation (NS). CONCLUSIONS: The long-term efficacy and safety of conventional elastic band ligation and videoendoscopic elastic band ligation are highly comparable. However, when videoendoscopic elastic band ligation is performed, significantly fewer treatment sessions are required.

Catheter probe extraductal EUS reliably detects distal common bile duct abnormalities.

BACKGROUND: This prospective study evaluated whether extraductal catheter probe EUS as an adjunct to endoscopic retrograde cholangiography can detect or rule out choledocholithiasis and other pathologic conditions of the distal common bile duct. METHODS: A total of 119 patients referred because of suspected choledocholithiasis or other bile flow obstruction for endoscopic retrograde cholangiography and papillotomy were included in this prospective study. Extraductal EUS of the distal common bile duct with a radial-scanning catheter probe was followed immediately by endoscopic retrograde cholangiography and papillotomy by a second examiner who was blinded to the EUS findings. Extraductal EUS and endoscopic retrograde cholangiography findings were compared. RESULTS: Extraductal EUS detected 33/34 bile duct stones and all papillary adenomas (16 patients). In 8/34 patients, stones were missed on cholangiography but were seen after papillotomy and stone extraction. Extraductal EUS missed 10 peripheral lesions, one pancreatic tumor, and two distal bile duct stenoses. Overall, the sensitivity of EDUS was 78% and specificity was 98%. CONCLUSIONS: Extraductal EUS accurately detects abnormalities involving the distal common bile duct, especially small stones. The use of catheter probe EUS imaging during interventional endoscopy can help to avoid unnecessary papillotomy and can influence therapeutic strategy.