Impact of periampullary diverticula on the outcome and fluoroscopy time in endoscopic retrograde cholangiopancreatography.

BACKGROUND: It is unclear whether the presence of periampullary diverticula (PAD) affects technical success and complication rates during endoscopic retrograde cholangiopancreatography (ERCP). Moreover, the impact of PAD on fluoroscopy duration is still unknown. The present study aimed to investigate the success rate and difficulty of common bile duct (CBD) cannulation, post-procedure complications and fluoroscopy duration in patients with and without PAD. METHODS: Patients from January 2008 to December 2010 with PAD (group A) and without PAD (group B) and similar indications for therapeutic ERCP were prospectively compared. The comparison included patient characteristics, findings of ERCP, and details of procedure and fluoroscopy time. The influence of papilla's location with respect to the diverticulum on procedure was also investigated. RESULTS: A total of 428 consecutive patients who had undergone therapeutic ERCP for similar indications were divided in two groups according to the presence (group A, 107 patients) or absence (group B, 321 patients) of PAD. The mean age and ASA score of the patients with PAD were significantly higher than those patients without PAD. The main indication was choledocholithiasis. Successful final CBD cannulation was achieved in 97.20% of the patients in group A vs 99.69% in group B (P=0.05). CBD diameter, number of stones and the largest stone size were significantly higher in group A than group B (P<0.001). Complete clearance of the CBD after the first attempt was achieved in 85.86% and 94.75% of the patients in groups A and B, respectively (P=0.03). In both groups, the time needed to complete the procedure and fluoroscopy time was significantly longer in patients with PAD (22.87 vs 18.99 minutes, P<0.001; 76.51 vs 47.42 seconds, P<0.001). There was no significant difference between the two groups in the complication rate. The type of papilla's location with respect to the diverticulum did not influence the total cannulation rate and post-procedure complications. CONCLUSION: The presence of a PAD does not affect the success rate and complications of therapeutic ERCP in expert hands; however, the fluoroscopy time is significantly longer in patients with PAD.

Prophylactic clip application before endoscopic resection of large pedunculated colorectal polyps in patients receiving anticoagulation or antiplatelet medications.

BACKGROUND: The aim of the present study was to evaluate the safety of prophylactic endoclipping before resection of large pedunculated colorectal polyps in patients with uninterrupted anticoagulation or antiplatelet therapy. PATIENTS AND METHODS: In a single tertiary referral center, patients with normal coagulation parameters and those with uninterrupted anticoagulation or antiplatelet medications, who underwent prophylactic endoclipping before resection of large pedunculated colorectal polyps (polyp's head diameter >10 mm), were evaluated retrospectively. Demographic, clinical, and laboratory parameters, polyp characteristics, number of clips used for endoclipping, histology of resected polyps, the polypectomy technique, and postpolypectomy complications were recorded and compared. RESULTS: In 64 patients with large pedunculated colorectal polyps, successful endoclipping before resection was achieved. Eleven patients (17.2%) had uninterrupted anticoagulation (n = 4) or antiplatelet (n = 7) medications. Statistical analysis showed no significant differences in age, sex, indications for colonoscopy, location of polyps, size of polyp head or stalk, coexisting small polyps, histology of resected polyps, and polypectomy techniques among patients with normal coagulation parameters (group A) and patients on uninterrupted anticoagulant or antiplatelet treatment (group B). Coronary artery disease, hypertension, and atrial fibrillation were significantly more prevalent in group B. No immediate or delayed postpolypectomy hemorrhage occurred in either group. One patient (1.9%) in group A developed postpolypectomy coagulation syndrome and was successfully treated conservatively. Follow-up evaluation demonstrated no recurrence of polyps or cancer development. CONCLUSIONS: According to our experience, uncomplicated polypectomy of large pedunculated colorectal polyps can be performed by prophylactic endoclipping in patients receiving anticoagulation or antiplatelet medications.

Comparison of three types of precut technique to achieve common bile duct cannulation: a retrospective analysis of 274 cases.

BACKGROUND: The aim of this retrospective study was to evaluate the success rate and complications of three precut techniques of sphincterotomy after failure of conventional techniques of cannulation of common bile duct (CBD). PATIENTS AND METHODS: Between January 2003 and October 2011, 2,903 consecutive ERCPs were performed in patients with naïve major papilla. In 283 patients in whom biliary cannulation was not achieved, precut technique was performed and these patients were included in the study. RESULTS: A total of 274 patients were included in the final analysis. Needle-knife papillotomy (NKP) was performed in 129 cases (47.1 %), suprapapillary fistulotomy (SPF) in 78 patients (28.5 %), and transpancreatic sphincterotomy (TPS) in 67 cases (24.5 %). No significant difference was observed in the initial and eventual success rate of biliary cannulation between the three groups. Overall, complications occurred in 54 patients (19.7 %), of which 33 (25.6 %) were with NKP, 6 (7.7 %) with SPF and 15 (22.4 %) with TPS, respectively, a difference statistically significant favoring the SPF group (p = 0.006). Post-procedure acute pancreatitis was developed in 27 cases (20.9 %) with NKP, compared to two cases (2.6 %) with SPF and 15 cases (22.4 %) with TPS, a difference statistically significant favoring the SPF group. No difference was observed between the groups with regard to the occurrence of post-procedure hemorrhage and perforation. CONCLUSIONS: The three types of precut sphincterotomy have no different overall CBD cannulation rates; SPF reduces post-ERCP pancreatitis risk.

Postpolypectomy bleeding: incidence, risk factors, prevention, and management.

Endoscopic polypectomy is at the forefront of colorectal cancer (CRC) prevention. However, endoscopic polypectomy is not completely free of complications, with bleeding being one of the most common complications encountered. In view of the ongoing campaign to introduce colorectal cancer screening to the population, addressing the issue of colonoscopic complications, and postpolypectomy bleeding (PPB) in particular is becoming more important. Despite the fact that the overall incidence of PPB is low, predisposing factors need to be elucidated to further decrease the frequency of this complication. Furthermore, the role of various techniques of PPB prophylaxis remains controversial. We review recent studies on the incidence, risk factors, prophylaxis, and management of PPB.

Efficacy of colchicine in the treatment of mesenteric panniculitis in a young patient.

Mesenteric panniculitis (MP) is a rare inflammatory and fibrotic disease of the mesentery of unknown etiology. It has various clinical and radiological manifestations, posing a diagnostic challenge for clinicians. Its diagnosis is indicated via radiologic imaging and is usually confirmed via peritoneal biopsies. We describe a case of a patient with histopathologically proven MP, in which steroid dependence was successfully managed with colchicine.

A prospective analysis of factors influencing fluoroscopy time during therapeutic ERCP.

BACKGROUND: Fluoroscopy time (FT) in endoscopic retrograde cholangiopancreatography (ERCP) has a linear relationship with radiation exposure to endoscopist, personnel and patients. The aim of this prospective study was to investigate the factors influencing the FT during ERCP. PATIENTS AND METHODS: Between January 2010 and August 2011, patients with naïve papilla undergoing therapeutic ERCP were included in the study. Patient and procedural factors affecting fluoroscopy duration were investigated. RESULTS: During the study period 549 ERCP records were included in the final analysis. The mean procedural time and FT were 19.53±7.61 min and 48.82±26.43 sec, respectively. There was no effect of age or gender on FT. Univariate analysis showed choledocholithiasis (+17.92 sec; 95%CI: 12.73-23.11, p<0.001), multiple stones (+21.21 sec; 95%CI: 14.31-30.35, p<0.001), stone size >10 mm (+27.514 sec; 95%CI: 16.62-35.71; p<0.001), precut technique (+12.46 sec; 95%CI: 6.32-18.60; p<0.001), periampullary diverticulum (+33.36 sec; 95%CI: 28.49-38.23; p<0.001), mechanical lithotripsy (+31.14 sec; 95%CI: 24.67-37.61; p<0.001) and mechanical lithotripsy plus stent placement (+42.41 sec; 95%CI: 31.93-52.89; p<0.001) to be associated with longer FT. Multivariate analysis identified choledocholithiasis (+13.24 sec; 95%CI: 4.44-22.04; p=0.003), multiple stones (+19.51 sec; 95%CI: 11.72-26.78; p<0.001), stone size >10 mm (+23.95 sec; 95%CI: 14.35-29.45; p<0.001), needle-knife papillotomy (+17.26 sec; 95%CI: 7.77-26.75; p<0.001), periampullary diverticulum (+21.99 sec; 95%CI: 17.81-26.16; p<0.001) and mechanical lithotripsy plus stent placement (+20.39 sec; 95%CI: 7.38-33.40; p=0.002) to prolong FT. CONCLUSIONS: The identified factors influencing the FT may help endoscopists take appropriate precautions during ERCP to significantly decrease FTs.

Secondary rectal linitis plastica as first manifestation of urinary bladder carcinoma.

Secondary rectal linitis plastica is a very rare malignancy with poor prognosis. Diagnosis is difficult because of nonspecific clinical and endoscopic findings and negative biopsies in most cases owing to the fact that the mucosa is frequently unaffected. We herein describe a 68-year-old man who presented with a six-month history of tenesmus and constipation. Endoscopy revealed a narrow distal rectum with an indurated, cobblestone appearance of mucosa. Multiple biopsies and fine-needle aspiration were negative for malignancy. Abdominal MRI and transrectal ultrasonography showed findings compatible with rectal linitis plastica. He underwent rectal extirpation with total cystectomy and lymph nodes dissection. Histology demonstrated secondary rectal linitis plastica due to a poorly differentiated urinary bladder carcinoma. We emphasize the endoscopic and endosonographic features and the difficulty to establish a preoperative diagnosis of secondary rectal linitis plastica.

Diagnostic yield and clinical impact of wireless capsule endoscopy in patients with chronic abdominal pain with or without diarrhea: a Greek multicenter study.

BACKGROUND: Wireless capsule endoscopy has become the gold standard for the examination of small bowel. However, its role in the evaluation of patients suffering from chronic abdominal pain is not yet clearly defined. We conducted an open-label prospective multi-center study to evaluate the yield and clinical outcome of capsule endoscopy in patients with chronic abdominal pain with/without diarrhea. METHODS: Seventy-two patients with chronic (>3months) abdominal pain with/without diarrhea in whom the underlying pathology could not be diagnosed by conventional modalities, underwent capsule endoscopy in either of the 6 participating centers. Patients were then followed up for clinical outcomes. RESULTS: The overall diagnostic yield of capsule endoscopy was 44.4%. More specifically, its diagnostic yield was 21.4% in patients with abdominal pain and negative inflammatory markers (C-reactive protein and erythrocyte sedimentation rate), 66.7% in patients with abdominal pain and positive inflammatory markers, 0% in patients with abdominal pain, diarrhea and negative inflammatory markers, and 90.1% in patients with abdominal pain, diarrhea and positive inflammatory markers. Both univariate and multivariate regression analyses showed that abnormal C-reactive protein and erythrocyte sedimentation rate were significant factors related with positive capsule endoscopy findings. CONCLUSIONS: Chronic abdominal pain with/without diarrhea should be accompanied by elevated inflammatory markers to be regarded as a valid indication for capsule endoscopy. The yield of capsule endoscopy in such patients is reasonably high and clinical outcomes of patients treated with capsule endoscopy findings as a guide are significant.

Double probe pH-monitoring findings in patients with benign lesions of the true vocal folds: comparison with typical GERD and the effect of smoking.

We conducted a pH-monitoring study to determine the prevalence of pathologic gastroesophageal reflux (GER+) and laryngopharyngeal reflux (LPR+) in patients with resected benign true vocal fold lesions (TVFLs) and positive reflux finding score (RFS). We compared our findings with those of patients with typical GER disease (GERD) symptoms and normal laryngoscopy. In the group of patients with TVFLs, we compared the pH-monitoring findings of smokers with those of non-smokers. Seventy-two [females 32, mean (SD) age 49.3 (13.1) years] patients with resected TVFLs (polyps: 32, nodules: 20, Reinke's edema: 12, granulomas: 4, leukoplakia: 4) and 24 [females 14, mean (SD) age 42.2 (13.4) years] patients with typical GERD symptoms, who served as controls for the hypopharyngeal measurements, underwent 24-h double probe, hypopharyngeal and distal esophageal, ambulatory pH monitoring. Thirty-eight (52.8%) patients with TVFLs had GER+ and 52 (72.2%) had LPR+. More laryngopharyngeal reflux episodes (LPREs) were detected in patients with TVFLs compared to those with GERD (P < 0.001). With respect to the specific TVFLs, 12 (37.5%) patients with polyps had GER+ and 24 (75%) had LPR+, 6 (30%) patients with nodules had GER+ and 12 (60%) had LPR+, 6 (50%) patients with Reinke's edema had GER+ and 8 (66.7%) had LPR+ and all the patients with granuloma or leucoplakia had both GER+ and LPR+. Twenty (55.6%) of the 36 smokers and 32 (88.9%) of the 36 non-smokers with TVFLs had LPR+ (P = 0.003), while GER+ was recorded in 16 (44.4%) smokers and 22 (61.1%) non-smokers (P = 0.238). Smokers had significantly less LPREs (P < 0.001). In conclusion, 24-h double probe pH monitoring may detect GER+ and/or LPR+ in a substantial proportion of patients with resected TVFLs and positive RFS. Our study suggests that LPR+ is more prevalent in patients with TVFLs compared with typical GERD patients and that non-smokers with TVFLs are more likely to have LPR+ than smokers with TVFLs.

Endoscopic management of occluded biliary uncovered metal stents: a multicenter experience.

AIM: To compare diverse endoscopic interventions in the management of occluded uncovered self-expanding metal stents (SEMSs) that had been placed for palliative treatment of unresectable malignant biliary obstruction. METHODS: A retrospective review was undertaken in 4 tertiary endoscopic centers to determine optimal management of different types of occluded SEMSs. The technical success of performed treatment in occluded SEMSs, the patency of the stent, the need for re-intervention and the financial costs of each treatment were analyzed. RESULTS: Fifty four patients were included in the analysis; 21 received Hanaro, 19 Wallstent and 14 Flexus. For the relief of obstruction, a plastic stent was inserted in 24 patients, a second SEMS in 25 and mechanical cleaning was performed in 5 patients. The overall median second patency rates between second SEMSs and plastic stents did not differ (133 d for SEMSs vs 106 d for plastic stents; P=0.856). Similarly, no difference was found between the overall survival of SEMS and plastic stent groups, and no procedure-related complications occurred. Incremental cost analysis showed that successive plastic stenting was a cost-saving strategy at least in Greece. CONCLUSION: Insertion of uncovered SEMSs or plastic stents is a safe and effective treatment for occluded uncovered SEMSs; insertion of plastic stents appears to be the most cost-effective strategy.

Chronic pharyngitis is associated with severe acidic laryngopharyngeal reflux in patients with Reinke's edema.

OBJECTIVES: We evaluated the association between pathological acidic laryngopharyngeal reflux (LPR) events and chronic pharyngitis in patients with Reinke's edema. METHODS: We performed a prospective controlled study in 20 consecutive patients with Reinke's edema without pathological acidic LPR events (group A) and 40 consecutive patients with Reinke's edema with both clinical symptoms and 24-hour pH-metry suggesting acidic LPR (group B). The severity of acidic LPR was assessed by use of the Reflux Finding Score (RFS), the Reflux Symptom Index (RSI), and dual antimony probe 24-hour pH-metry. The patients were evaluated for the presence of chronic pharyngitis by clinical examination and biopsy specimens taken from the posterior pharyngeal wall. The Chi2 test was used to compare the groups for the presence of pharyngitis. In group B, the RSI, the RFS, and the total duration and number of acidic LPR events on 24-hour pH-metry were compared between patients with and without concomitant pharyngitis by use of the Mann-Whitney test. RESULTS: Five patients of group A and 20 patients of group B had chronic pharyngitis. Therefore, more patients with Reinke's edema and clinical signs of LPR tended to have chronic pharyngitis than did those with Reinke's edema and no clinical signs of LPR, but the difference was not statistically significant (p = 0.064; odds ratio, 3.0; 95% confidence interval, 0.9 to 9.8). Among group B patients, those with pharyngitis had significantly more acidic LPR events (p < 0.001) and a greater exposure time to gastric fluid (p = 0.008) than did those without pharyngitis. Their RFS and RSI did not differ significantly (p = 0.692 and p = 0.914, respectively). CONCLUSIONS: Only in the subgroup of patients with Reinke's edema and LPR was there a statistically significant correlation between the pH probe results and the incidence of clinical pharyngitis. Awareness should increase among physicians about addressing chronic pharyngitis in therapy for acidic LPR and/or Reinke's edema.

The use of endoclips in the treatment of nonvariceal gastrointestinal bleeding.

Acute nonvariceal gastrointestinal bleeding is the most common emergency managed by endoscopists and the endoscopic therapy has generally been recommended as the first-line treatment. Traditionally, endoscopic treatment included injections of epinephrine and sclerosing solutions or the use of thermocoagulation. In the last decade with the introduction of hemoclips and band ligators, we have witnessed a significant improvement in the clinical outcome of nonvariceal gastrointestinal bleeding. Endoclipping is a safe and effective technique that contributes to hemostasis of bleeding lesions of the gastrointestinal tract.

Successful treatment of intractable bronchobiliary fistula using long-term biliary stenting.

A bronchobiliary fistula (BBF) is an uncommon entity with bilioptysis being a pathognomonic sign. We describe the case of a 41-year-old man who had recurrent BBF, 6 months after resection of the anterior segment of the right lower pulmonary lobe and repair of a BBF due to hepatic hydatid disease. Magnetic resonance cholangiography revealed a communication between the biliary tree and the lower lobe of the right lung. Endoscopic biliary sphincterotomy and repeated insertion of large size biliary plastic stents led to a successful resolution of the symptoms and closure of the fistula.

Aggressive treatment of acute anal fissure with 0.5% nifedipine ointment prevents its evolution to chronicity.

AIM: To investigate the efficacy of topical application of 0.5% nifedipine ointment in healing acute anal fissue and preventing its progress to chronicity. METHODS: Thirty-one patients (10 males, 21 females) with acute anal fissure from September 1999 to January 2005 were treated topically with 0.5% nifedipine ointment (t.i.d.) for 8 wk. The patients were encouraged to follow a high-fiber diet and assessed at 2, 4 and 8 wk post-treatment. The healing of fissure and any side effects were recorded. The patients were subsequently followed up in the outpatient clinic for one year and contacted by phone every three months thereafter, while they were encouraged to come back if symptoms recurred. RESULTS: Twenty-seven of the 31 patients completed the 8-wk treatment course, of them 23 (85.2%) achieved a complete remission indicated by resolution of symptoms and healing of fissure. Of the remaining four unhealed patients (14.8%), 2 opted to undergo lateral sphincterotomy and the other 2 to continue therapy for four additional weeks, resulting in healing of fissure. All the 25 patients with complete remission had a mean follow-up of 22.9 +/- 14 (range 6-52) mo. Recurrence of symptoms occurred in four of these 25 patients (16%) who were successfully treated with an additional 4-wk course of 0.5% nifedipine ointment. Two of the 27 (7.4%) patients who completed the 8-wk treatment presented with moderate headache as a side effect of nifedipine. CONCLUSION: Topical 0.5% nifedipine ointment, used as an agent in chemical sphincterotomy, appears to offer a significant healing rate for acute anal fissure and might prevent its evolution to chronicity.

Endoscopic mucosal resection of large sessile colorectal polyps with submucosal injection of hypertonic 50 percent dextrose-epinephrine solution.

PURPOSE: Before endoscopic mucosal resection of large sessile colorectal polyps, injection of solution into submucosa cushions and isolates the tumor, although there is little information as to which solution with optimal effect should be used. This study investigated the effectiveness of endoscopic mucosal resection by use of a hypertonic dextrose plus epinephrine solution for large sessile colorectal polyps. METHODS: We removed 59 large sessile colorectal polyps in 59 patients by introducing an endoscopic submucosal hypertonic dextrose plus epinephrine injection technique. Endoscopic evaluations were repeated at 3, 6, and 12 months or longer. If no residual tumor was observed endoscopically and histologically at one year or more, the patient was considered to be "cured." The main outcome measurements were the mean amount of solution injected, mean disappearance time of solution, safety, complications, and recurrence at follow-up. RESULTS: Of the 59 large sessile colorectal polyps, 23 (39 percent) were resected en bloc and 36 (61 percent) piecemeal. The mean amount of hypertonic dextrose plus epinephrine solution injected was 24.42 +/- 17.52 ml, and its mean disappearance time was 13.61 +/- 5.21 (range, 7-21) minutes. Of the 36 patients treated with piecemeal resection, 18 (50 percent) required additional endoscopic interventions. In patients who entered the follow-up surveillance protocol for one year or longer, the cure rate by en bloc resection was 100 percent (23/23) and that by piecemeal intervention was 96.78 percent (30/31). Four patients (6.8 percent) had local bleeding after endoscopic mucosal resection that was mainly controlled endoscopically. CONCLUSIONS: Endoscopic mucosal resection after submucosal hypertonic dextrose plus epinephrine solution injection, with an intensive follow-up program, seems to be a safe and effective treatment for large sessile colorectal polyps.

Endoscopic mucosal resection of lateral spreading tumors of the colon using a novel solution.

Lateral spreading tumors (LSTs) of the colon are lesions over 10 mm in diameter that are low in height and grow superficially. They are increasingly being diagnosed in Western cohorts. The aim of this study was to investigate the safety and efficacy of dextrose 50% solution in the endoscopic mucosal resection (EMR) of LSTs. The study population consisted of 21 patients with LSTs of the colorectum. The mean size of the LSTs was 23.52+/-13.60 mm. Dextrose 50% solution was injected, via a variceal needle, into the submucosa to lift up the LST sufficiently from the proper muscle layer. Subsequently, a snare was positioned around the lesion and then closed while being pressed against the mucosa, with suction being applied to draw the lesion into the snare. Blended current was used for resection. If necessary, a piecemeal technique was used to achieve complete resection. Immediate and delayed complications were recorded. After the EMR, patients were followed up at 3, 6, and 12 months or later, using total colonoscopy. Endoscopic resection was completed in all LSTs. Of the 21 LSTs, 15 (71.4%) were resected en bloc and 6 (28.6%) piecemeal. The mean amount of injected dextrose 50% solution was 14.86+/-9.13 mL. One patient (4.78%) had immediate bleeding after EMR, which was stopped endoscopically. Histologic examination of resected LSTs showed adenoma with high-grade dysplasia 9 (42.9%), adenoma with low-grade dysplasia 10 (47.6%), and invasive carcinoma 2 (9.5%). Twenty patients were followed up for 37.9+/-24.03 months. Local recurrent disease was detected in 4 patients (20%), all within 6 months of the index EMR. These recurrent lesions were completely resected endoscopically. The contribution of submucosal injection of dextrose 50% is significant for a safe and efficient EMR of LSTs of the colorectum.

The role of endoscopic treatment in postoperative bile leaks.

BACKGROUND/AIMS: Bile leak is among the most common and serious complications of biliary tract surgery. The aim of this non-randomized study was to evaluate the role of endoscopic intervention as an accepted treatment for this complication. METHODOLOGY: An endoscopic retrograde cholangiopancreatography (ERCP) database was reviewed retrospectively to identify all cases of bile leak related to cholecystectomy (laparoscopic or open). Patients' records and endoscopy reports were reviewed. Moreover, structured telephone interviews were conducted to collect data. RESULTS: Twenty-four patients, 4 males and 20 females, with a median age of 54 (range 28-76 years) with suspected postcholecystectomy bile leaks were referred for ERCP performed 3-73 days after operation (mean 9.5 days). All but one case had high-grade bile-like liquid outflowing from the original drainage tubes or the fistulous tract of T-tube. One patient presented with bilious ascites, 17 patients had sudden or gradual abdominal pain, 3 jaundice, 2 abdominal pain with fever, and 1 nausea and vomiting. ERCP was successful in all cases, and revealed leakage from the cystic stump in 10 cases, from a common bile duct (CBD) defect in 6, from a common hepatic duct defect in 3, from the gallbladder bed in 2, from a T-tube track in 1, and complete CBD transection in 2 patients. Seventeen patients were successfully treated by endoscopic sphincterotomy (ES) plus endoprosthesis, 3 by stent placement without sphincterotomy, 2 with complete transection by proximal hepaticojejunostomy, and 2 patients with leakage from the cystic stump and a CBD defect were operated after unsuccessful endoscopic intervention. CONCLUSIONS: ERCP is recommended as a safe and efficacious intervention to detect and treat postoperative bile leaks. ES plus endoprosthesis is effective for the treatment of bile leakage. Endoscopic stenting without sphincterotomy may be offered as a primary option in young patients with postoperative bile leaks.

Efficacy and safety of therapeutic ERCP in patients 90 years of age and older.

BACKGROUND: Therapeutic ERCP has an established role in the treatment of pancreatobiliary diseases, but little information is available on the outcomes of this procedure in patients 90 years of age and older. OBJECTIVE: To evaluate the efficacy and the safety of therapeutic ERCP in an extremely elderly cohort. DESIGN: Retrospective study. SETTING: Two Greek cohorts of patients > or =90 and 70 to 89 years of age who underwent therapeutic ERCPs. PATIENTS: Sixty-three patients aged 90 years and older (group A) and 350 patients 70 to 89 years of age (group B). INTERVENTIONS: A retrospective review of therapeutic ERCPs was performed between 1994 and 2000 on both groups, identified by using a database linked to the endoscopy reporting system in our department. MAIN OUTCOME MEASUREMENTS: Efficacy and safety of therapeutic ERCPs. Concomitant diseases, complications, and outcome were also evaluated. RESULTS: Group A patients had a higher incidence of concomitant diseases than group B patients (100% vs 72.8%, respectively). The rate of post-ERCP early complications was low in both groups: 6.3% in group A and 8.4% in group B. The frequency of ERCP-related mortality was 1.6% (1 patient) in group A and 0.6% (2 patients) in group B. Group A required endoscopic sessions for stone clearance and mechanical lithotripsy more frequently than group B (20.6% vs 11.4% and 17.5% vs 10.3%, respectively). No patient in either group experienced subjective deterioration in mental status, and the 3 patients who died required ventilatory support before death. Late complications occurred in 2.3% of patients in group B. CONCLUSIONS: Therapeutic ERCP is safe and effective for the treatment of pancreatobiliary diseases in extremely elderly patients, and advanced age per se should not impinge on decisions relating to its use.

Severe acute haemorrhagic gastritis controlled by hydrogen peroxide.

A 92-year-old woman presented with severe acute haemorrhagic gastritis due to abuse of non-steroidal anti-inflammatory drugs (NSAIDs). She was treated with instillation of 150 ml 3% hydrogen peroxide (H2O2) every 2 h via a nasogastric tube. The copious amount of bright red blood through the nasogastric tube started to decline substantially after the first administration of H2O2 and continued to reveal clear material during the second and third instillation of H2O2. The total amount of H2O2 administered was 600 ml. No rebleeding and only a few flame-shaped intramucosal haemorrhages were observed on the following four consecutive daily endoscopic evaluations. These are promising observations which will have to be confirmed with respect to the safety and efficacy of H2O2 treatment by further controlled studies.