Sengstaken-Blakemore Research Articles

Treatment of acute esophageal variceal bleeding in cirrhotic patients

Esophageal variceal bleeding is the most dangerous complication in cirrhotic patients and is accompanied by high mortality. Treatment strategy involves early diagnosis, maintaining vital body functions and specific therapy aimed at the provision of local hemostasis and reduction of portal pressure. To this end, it is currently recommended to combine vasoactive drug (mainly, terlipressin or somatostatin) therapy with endoscopic methods of hemostasis (sclerotherapy or ligation). The use of Sengstaken-Blakemore tube is appropriate only in cases of refractory bleeding if the above methods cannot be used. An alternative to balloon tamponade may be the installation of self-expandable metal stents. Although transjugular intrahepatic portosystemic shunting is an extremely useful technique for the treatment of acute bleeding from esophageal varices, currently it is viewed as second-line therapy. Urgent surgical intervention is rarely performed and can be considered only in case of failure of conservative and/or endoscopic therapy and being unable to use a transjugular intrahepatic portosystemic shunt for technical or organizational reasons or due to anatomic problems. Among surgical operations described in the literature are various kinds of portocaval anastomoses and azygoportal disconnection procedure. To improve the results of treatment of cirrhotic patients with acute esophageal variceal bleeding it seems important to stratify them by risk groups, which will allow one to tailor therapeutic approaches to the expected results. For example, to initiate early use of more aggressive methods in patients with predictors of poor outcomes, and to protect individuals with a good prognosis from unnecessary invasive procedures. It is hoped that further research will refine this hypothesis.

Pressurized intraluminal aerosol chemotherapy with Dbait in the distal esophagus of swine.

A novel therapeutic concept, pressurized intraluminal aerosol chemotherapy (PILAC), and a corresponding device for distributing drugs to the mucosa and submucosa of the distal esophagus are presented. The endoscopic device that was designed consisted of (i) a double-balloon catheter, similar to a Sengstaken-Blakemore tube; (ii) a carbon dioxide (CO2) line, used to create a gaseous, pressurized environment; and (iii) a micropump, used to generate a therapeutic aerosol. The device was inserted into the distal esophagus in three narcotized Landrace pigs. Dbait (short interfering DNA, or siDNA) was aerosolized under pressure (12 mmHg) in CO2 at 37 °C for 30 minutes. The procedure was well tolerated by all animals. At autopsy, no mucosal or muscular tear was observed. Fluorescence microscopy revealed a homogeneous intramural distribution of Dbait-cyanine 5 in the esophageal wall down to the circular muscular layer (400 - 600 µm). PILAC is feasible in a large animal model and appears to be safe. Therapy of the entire "tissue at risk" for the development of cancer in the distal esophagus is possible without the prior endoscopic identification of diseased tissue.

下行大動脈瘤食道穿破に対して SB チューブ，ステントグラフト内挿術によって救命しえた一例

下行大動脈瘤食道穿破に対して SB チューブ，ステントグラフト内挿術によって救命しえた一例

Catastrophic Hemorrhage Due to Dieulafoyʼs Lesion (DL) Treated Successfully With Sengstaken-Blakemore (SB) Tube Placement and Angiographic Embolization

Case Presentation: A 69-y/o female with a history of congestive heart failure presented with altered mental status, shortness of breath, and nausea. She was found to have acute kidney injury and hyperkalemia. She was admitted to the intensive care unit and underwent emergent dialysis. On third day of admission she complained of vague abdominal pain and nausea. She was pale, tachycardic and abdominal exam revealed a distended abdomen. Her hemoglobin was 6.0 gm/dl and hematocrit was 18%. She became hypotensive requiring three vasopressors. CT abdomen with contrast showed active extravasation at the GE junction and a distended stomach with hypodense material (Figure 1). Following nasogastric aspirate, upper endoscopy revealed active arterial bleeding near the GE junction. A large amount of blood and clot obscured visualization of the underlying lesion. Tamponade was temporarily achieved by insertion of an SB tube and inflation of the gastric balloon. She was transported to Interventional Radiology and underwent left gastric artery embolization. Patient rebled next day. Repeat endoscopy showed evidence of persistent bleeding with a large amount of fresh and old blood in the gastric fundus. The SB tube was reinserted, the gastric balloon inflated and the patient was taken back to Interventional Radiology. Angiography showed evidence of active bleeding from the inferior phrenic artery (Figure 2). This was treated successfully with embolization. Over the next few days her transfusion requirement improved. Repeat upper endoscopy performed on day two following the second embolization showed less amount of old blood in the gastric fundus. The patient's subsequent recovery was uneventful.Figure 1Figure 2Discussion: DLs generally cause brisk arterial bleeding that is identified endoscopically by densely adherent clots with minimal mucosal defect or completely normal underlying mucosa. The proximal stomach is the most common site of DLs, predominantly along the lesser curvature within 5 cm of the gastro-esophageal junction. Our patient was unusual in that, this area was supplied by both the left gastric artery and a branch of the left inferior phrenic artery. Endoscopic hemostasis, when possible, is the modality of choice for the initial treatment of DLs. Angiographic embolization is an alternate modality in patients with active bleeding not amenable to endoscopic therapy. Conclusion: Dieulafoy's lesion is a potentially life-threatening cause of upper gastrointestinal hemorrhage. Diagnosis is often problematic. Our patient's catastrophic hemorrhage secondary to DL required both endoscopic and angiographic modalities to achieve hemostasis.

The Use of “Through-the-Scope” Fully Covered Self-expanding Metal Stent to Stop Refractory Variceal Bleeding in a Patient With Hemorrhagic Shock

Introduction: A Sengstaken-Blakemore tube is often utilized in cases of severe variceal bleeding. Limitations for this treatment included a high rate of bleeding recurrence upon removal, and increased rate of esophageal perforation with placement. Esophageal fully covered self-expanding metal stents (FCSEMS) have been proposed as a possible treatment option for severe variceal bleeding in place of a Sengstaken-Blakemore tube. Case: Mr. S a 49 year old male with history of cirrhosis from alcohol and hepatitis C who presented with bright red blood per rectum. Two weeks prior, anesophagogastroduodenoscopy (EGD) demonstrated grade 2 esophageal varices for which he underwent banding. Hematemesis resolved and Hgb stabilized and he was discharged home. Vital signs including orthostatic vitals were unremarkable. His mucous membranes were moist without pallor and abdomen was soft/nontender with minimal fluid wave. Laboratory data demonstrated Hgb 10.8 mg/dL, platelets 98 u/L, INR 1.9 and total bilirubin 1.9 mg/dL. Liver enzymes were within normal limits. Proton pump inhibitor and octreotide drips were initiated for treatment. EGD identified a new varix closely associated with an ulceration from previous banding. An attempt was made to band this new varix, but with reinsertion of the endoscope the varix ruptured. Exsanguination ensued and banding nor sclerotherapy could be performed secondary to inability to visualize the bleeding vessel. He progressed to hemorrhagic shock and then pulseless electrical activity. Cardiac resuscitation was performed. A Sengstaken-Blakemore tube was unsuccessful secondary to inability to pass the tube. An 18mm x 120 mm “through the scope” (TTS) FCSEMS (Taewoong Medical Co.) was advanced into the gastric lumen under direct endoscopy and withdrawn into the esophagus. Exsanguination ceased after stent deployment. A 3 cm laceration was noted in the anterior esophagus from the Sengstaken-Blakemore tube placement attempt and a second FCSEMS was placed. His pulse returned after cessation of bleeding and rapid transfusion of red blood cells. However, he sustained severe anoxic brain injury and after several days care was withdrawn. Discussion: Patients with exsanguination from ruptured varices may benefit from placement of TTS esophageal FCSEMS instead of a Sengstaken-Blakemore tube. TTS esophageal FCSEMS allow for direct endoscopic placement, limited risk of perforation, and longer treatment time to stabilize patients for a more definitive treatment, such as TIPS.Figure 1Figure 2Figure 3

Peri-esophageal Pseudoaneurysms: Unique Cause of Refractory Arterial Bleeding Treated With Novel Technique

A 43 year old female with SLE and dysplastic Barrett's esophagus (Prague class C8 M9) treated with radiofrequency ablation presented with hematemesis. An EGD showed esophageal ulcers without any active bleeding. After 2 days of conservative treatment, the patient had an episode of hematochezia and syncope. Colonoscopy showed fresh blood throughout the colon as well as terminal ileum without any source. EGD showed a large esophageal ulcer with active bleeding. Endoscopic clips did not stop the bleeding. A Sengstaken Blakemore tube was placed. Patient was taken to Interventional radiology suite where brisk active bleeding into the distal esophagus from esophageal branches of the left gastric artery was seen. The arteriogram showed saccular dilation/pseudo-aneurysms that were the source of bleeding (arrow, Figure 1). Gel-foam embolization of the left gastric and esophageal branches, coil embolization of the gastrohepatic trunk and particle embolization of the pseudo-aneurysms stopped the bleeding. Esophageal brushings showed active cytomegalovirus infection along with low complement levels consistent with active vasculitis from SLE and mixed connective tissue disease. The patient eventually underwent exploratory laparotomy with resection of a necrotic esophago-gastric junction as well as thoracic esophagectomy. Pathological evaluation showed submucosal hemorrhage, esophagitis with dysplastic Barrett's mucosa and an ulcer containing cytomegaloviral inclusions. It is probable that the ulcer from cytomegalovirus eroded into the pseudo-aneurysm formed by the persistent SLE mediated vasculitis and lead to the refractory bleeding.Figure 1We report the first case of arterial bleeding from peri-esophageal pseudo-aneurysms present in the aortoesophageal branches in the left gastric arcade. We also report the use of gel-foam and particle embolization for arterial bleeding in the esophagus.

Life-Threatening Variceal Bleed Caused by Secondary Portal Hypertension Controlled by Emergent Splenopneumopexy

Introduction: Variceal bleeding is a common and feared complication of portal hypertension. In cases of recurrent bleeding despite pharmacologic therapy and endoscopic variceal ligation, portal decompressive therapy is often pursued. We present a case of a life-threatening variceal bleed caused by portal hypertension secondary to extensive pre-sinusoidal thrombosis as a result of schistosomiasis. The patient's bleeding was managed with emergent splenopneumopexy, a procedure that created a portopulmonary shunt via anastomosis of the splenic veins and pulmonary veins. Case Description: A 49-year-old male with past medical history of schistosomiasis initially presented to a community hospital with hematemesis and hematochezia. An EGD revealed multiple grade 3 esophageal varices that were banded with resolution of bleeding. Three weeks later, the patient had recurrent hematemesis requiring hospitalization. CT of the abdomen/pelvis revealed occlusion of portal, splenic, and superior mesenteric veins. An EGD was repeated but no targets for band ligation were identified due bleeding and presence of clots. A TIPS procedure was attempted but was unsuccessful due to completely occluded vasculature. The patient experienced ongoing hematemesis and hematochezia with hypotension requiring resuscitation. He was stabilized and transferred to our hospital for higher level of care. Upon arrival, he was intubated for airway protection. A Blakemore tube was placed for balloon tamponade and he was taken to the OR for emergent surgery. The patient underwent splenopneumopexy for creation of a portopulmonary shunt as well as transection of the esophagus with primary esophageal anastomosis. The patient did well post-operatively, remaining hemodynamically stable and did not have any further bleeding. The etiology of his pre-sinusoidal thrombosis was thought to be secondary to schistosomiasis given prior history of infection. A hypercoagulable workup was unremarkable and liver biopsy revealed normal hepatic parenchyma. The patient received empiric treatment for schistosomiasis and was discharged 14 days post operatively. Six months after discharge, the patient had not had any further bleeding. Discussion: Splenopneumopexy was first developed in Japan in the 1960s for treatment of portal hypertension in patients with Budd-Chiari Syndrome. The operation induces a portopulmonary shunt via anastomosis of the superior pole of the spleen and exposed pulmonary veins in the left lower lobe. In our patient's case, extensive pre-sinusoidal thrombosis from schistosomiasis precluded successful TIPS procedure and in the setting of refractory variceal bleeding, the patient's life was ultimately saved by emergency decompressive surgery with splenopneumopexy.Figure 1Figure 2

Esophageal Perforation Caused by Misplaced Minnesota Tube

48-year-old man with decompensated alcoholic Acirrhosis (Model for End-Stage Liver Disease 14) was admitted to an outside hospital with a 1-day history of hematemesis and melena. Upon arrival, the patient was hemodynamically unstable and his laboratory test results showed a hemoglobin level of 5.9 g/dL. Initial management included resuscitation with packed red blood cells and fresh-frozen plasma, in addition to combined continuous infusion of proton-pump inhibitor and octreotide therapy. An emergent esophagogastroduodenoscopy (EGD) was performed, which showed bleeding from distal esophageal varices. Four bands were placed successfully for endoscopic variceal ligation. Shortly thereafter, the patient had recurrent hematemesis and a repeat EGD proved unsuccessful in banding of a distal esophageal varix. The patient then was intubated for airway protection and a Minnesota tube was placed before transfer to our facility for further care. On arrival to our hospital, the patient’s respiratory status deteriorated acutely and a chest radiograph showed a misplaced Minnesota tube with the gastric balloon inflated within the distal esophagus (Figure A). Both gastric and esophageal balloons subsequently were deflated and the Minnesota tube was removed. A chest computerized tomography showed a new, large, left-sided pleural effusion concerning for hemothorax. Emergent chest tube placement was performed. An EGD showed a large esophageal perforation that tracked into the pleural cavity, extending from 29 to 39 cm from the incisors. Endoscopic placement of 2 fully covered (23mm 15 cm) esophageal metal stents (Boston Scientific, Galway, Ireland) was performed and a follow-up esophagogram showed no contrast extravasation (Figure B). Subsequently, a 10 mm 8 cm Viatorr (GORE, Flagstaff, AZ) transjugular intrahepatic portosystemic shunt was placed by interventional radiology. After a prolonged hospital course, the patient’s condition improved. Two months after presentation, a repeat EGD permitted removal of all stents and fluoroscopy confirmed complete closure of the esophagopleural fistula (Figure C). To prevent this complication, we recommend the following: (1) passage of the tube to at least 50 cm from the incisors, (2) injection of air into the gastric aspiration port while auscultating over the stomach, and (3) radiographic confirmation before complete inflation of the gastric balloon whenever possible.

SENGSTAKEN-BLAKEMORE tube can prevent whole liver twisting the right hepatic fossa improving venous outflow post-orthotopic liver transplantation

Introduction: Reduced-size liver transplantation in combination with an unusual situation poses an increased risk for venous outflow obstruction. Reduced-size liver transplantation using ex-vivo graft reduction or splitting has become standard procedure in children, with some of these cases described in literature after the successful management use of Sengstaken-Blakemore, although used in orthotopic whole liver transplantation (OLT) has never been described. The aim was to report the successful management of acute venous obstruction after OLT by using a Sengstaken-Blakemore tube within the hepatic fossa in a 51-year-old male patient. Case Report: A 51-year old male patient with hepatitis C cirrhosis and alcoholism, C Child-Pugh, 24 MELD score was submitted to OLT by using the piggy-back technique. The donor and recipient body-weight ratio was 1.38 (65/90kg). The liver recipient weight was 1,152g and the donor liver was 992g. After reperfusion time, the patient showed a venous outflow obstruction in supra-hepatic anastomosis with total resolution after clockwise rotation of the graft, but reduced venous outflow was observed after liver twist to the right hepatic fossa and mean arterial blood pressure (MAP) was reduced to 30 mmHg . After five attempts, a Sengstaken-Blakemore tube was inserted into the right subphrenic space and the gastric and oesophageal balloons were inflated to maintain the graft at the midline position with normalization of the MAP and venous outflow. Although data on hemodynamic measurements were maintained during the ICU period, a Doppler ultrasonography (DUS) was carried out daily to ensure that venous outflow was maintained. On 4th day, the gastric balloon was deflated; after 2 hours the oesophageal balloon was deflated and no systemic abnormality was detected. The Sengstaken-Blakemore tube with deflated balloons was removed on 5th day using slight traction. Later, DUS was carried out after it was shown that venous outflow was maintained and the liver was fixed in a hemodynamically correct position without interfering in the mechanical outflow. The patient was discharged, and he is in good condition. Conclusion: The Sengstaken -Blakemore tube is a good tool to improve liver venous outflow due to mechanical problems such as twist of the right hepatic fossa and preventing re-operations.

PTH-006 Experience in the use of self-expandable metal stents for the management of variceal haemorrhage

Introduction Self-Expandable Metal Stents (SEMS) have been reported in small case series to be effective in controlling bleeding from oesophageal varices. A NICE technology appraisal published in 2008 (updated 2011) approved their use, providing arrangements were in place to audit practice. We describe our experience of SEMS in patients presenting with variceal haemorrhage. Method The Royal Derby Hospital has 24/7 provision of emergency endoscopy by a consultant gastroenterologist and TIPSS insertion available within normal working hours. A SEMS (SX-ELLA Danis) was inserted in 15 patients (M:F 12:3, mean age 56, range 29–73) presenting with variceal haemorhage between 2009 and 2014. All stents were inserted under endoscopic guidance. Portal hypertension was due to alcohol induced cirrhosis (n = 14) and Budd Chiari (n = 1). The median Child Pugh score was 11 (B=2/ C =13) and MELD 21 (range 12–35). All patients had band ligation therapy prior to SEMS insertion, with 6 patients having two attempts at endoscopic treatment and 8 patients having a Sengstaken-Blakemore Tube (SBT). The indication for SEMS insertion was ongoing variceal haemorrhage (n = 4), bleeding from band ligation ulceration (n = 7), a combination of the above (n = 2) or complication of SBT insertion (n = 2). SEMS was used as a bridge to TIPSS in 3 patients (though unsuccessful in 2), but TIPSS was felt to be contraindicated in 11 patients due to the severity of hepatic decompensation and in 1 patient due to portal vein thrombosis. Results SEMS insertion was successful and immediate haemostasis achieved in all 15 patients. One patient developed aspiration pneumonia, but patients were subsequently nursed at 45° and received a prokinetic, without further incidence of aspiration. Stent migration into the stomach was observed in 3 patients, with partial slip across the gastro-oesophageal junction seen in a further 2 patients. The SEMS were removed from 9 patients after a median of 11 days (range 1–18) without complication. The remaining 6 patients died prior to removal. No patients had evidence of ongoing bleeding while SEMS in situ , but 33% had rebleeding within 30 days of removal. A median of 7 units of blood were transfused pre-SEMS insertion and 2 units post. Patient survival was 40% at 30 days and 27% at 1 year. Amongst survivors at 1 year the median Child Pugh/ MELD score at SEMS insertion was 8.5/ 14 compared to 11/ 24 in those who died. Conclusion Our experience shows that SEMS can be safely inserted and effectively achieve haemostasis in patients with refractory haemorrhage from oesophageal varices and in those with bleeding from band ligation induced ulceration. Patient survival was poor, however, due to the severity of hepatic decompensation in the study population. Disclosure of interest None Declared.

PWE-423 Human factors training (hft) through simulation based scenarios in endoscopic haemostasis improves teamworking

Introduction In the early 1970s, number of air crashes were attributed to human error. HFT is well recognised for improving safety in aviation. Medical errors rank high amongst the leading causes of death and HFT is increasingly been recognised to reduce errors in CPR and anaesthetics. The value of HFT in endoscopy has not been evaluated. Method We delivered haemostasis and HFT to Gastroenterology Specialty trainee Registrars (StRs) at Lancashire simulation centre. A computer-based life-sized manikin with human physiology emulation capability was used. Its vital signs were controlled remotely. The manikin could talk and was connected to ECG, Pulse rate, SpO2 and BP monitor. Endoscopy equipment and porcine stomach models were used. Haemostatic methods including adrenaline injection, haemoclip placement, Haemospray, Sengstaken tube and Danis stent insertion were practiced. Introductory presentations on Human factors were delivered. 4 scenarios were written to provide ‘high-stress/high-stakes’ events that are encountered infrequently, but where rapid and professional responses are crucial i.e. patients with an upper GI bleed developing periarrest whilst in endoscopy unit. Each scenario involving 5 to 6 StRs was facilitated by Gastroenterologists and Simulation skills educators. Feedback was provided by peer observation. Microteaching was delivered using video assisted debrief. Pre and post course feedback was obtained. Results 23 StRs attended. Prior to this course 55% StRs have had no HFT, 14% had self directed learning, 9% received lectures and 22% had formal workshop based HFT. Data analysis showed improvement in all areas of Human Factor: decision making, situational awareness, team working and leadership skills. On the Likert scale (1 strongly disagree, 5 strongly agree) StRs felt scenarios were useful (Median 5, IQR 1), amount of information was appropriate (Median 5, IQR 1) and hands on training was useful (Median 5, IQR 1). StRs reported increased awareness of managing acutely ill patients in relation to sedation issues (Median 4, IQR 2), principles of resuscitation and fluid management (Median 4, IQR 2), improved leadership, teamwork and communication (Median 5, IQR 1). Qualitative feedback was positive. StRs however recommended smaller groups in future courses. 90% StRs felt that training on Human Factors and their influence on patient safety should be compulsory. Conclusion Human patient simulation provides ‘real life’ environment to observe Human Factors and allows chance to debrief, something that is difficult in the busy workplace. StRs benefit from HFT and report improvement in their skills. We recommend that HFT in endoscopy should be incorporated in competency based training curriculum and rolled out nationally. Disclosure of interest None Declared.

PTH-034 Hemospray may not reliably achieve hemostasis beyond 48 hours in acute upper gastrointestinal bleeding

Introduction Hemospray (Cook labs, Birmingham, USA) has been shown in multiple case series to be a useful and effective tool 1 in the management of acute GI bleeding where existing therapies fail to achieve hemostasis 2 or are impractical. It is as yet unclear whether treatment with Hemospray provides temporary or definitive benefit. Method We prospectively followed 9 consecutive cases in which Hemospray was used to control acute upper GI bleeding. Each case was followed to 30 days at which point the primary end-points of rebleeding and mortality were documented. Data was sourced from case notes and the local endoscopy database. All patients were treated at Southmead Hospital, Bristol, United Kingdom – a large endoscopy unit serving a population of around 620,000. Statistical analysis was by chi-squared and unpaired t-test. Results 9 patients (7 males, 2 females) were included in the study with a mean age of 60.1 years (range 18–86 years). Endoscopic diagnoses included duodenal ulcer (3), gastric ulcer (1), gastric tumour (1), oeosophageal varices (1), Dieulafoy lesion (1), ERCP sphincterotomy site bleed (1) and angiodysplasia (1). Hemospray was used as the only therapeutic modality in 7 cases. Hemostasis was achieved in 6/9 cases at the index endoscopy. In 3/9 cases bleeding continued: 2 patients underwent further endotherapy and 1 patient required a Sengstaken-Blakemore tube and subsequently died. 3/9 patients experienced rebleeding. 1 patient required further endotherapy and 2 patients required laparotomy. 2 patients experienced rebleeding more than 48 h after the index endoscopy. Rebleeding did not vary across the range of endoscopic diagnoses (p = 0.423) or age of patient (p = 0.302). At 30 days only 3/9 patients treated with Hemospray did not require further intervention. 8/9 patients survived to 30 days. Conclusion Our case series indicates that apparent initial hemostasis after application of Hemospray may be a temporary outcome only. In the UK, where physicians are urged to discharge patients soon after clinical stability has been achieved, this is a particularly important message. There may be benefit from relook endoscopy at 48 h or prolonged inpatient observation although this has not been suggested elsewhere. Larger studies are required and a region-wide series is planned. Disclosure of interest None Declared. References Bustamante-Balen M, Plume G. Role of hemostatic powders in the endoscopic management of gastrointestinal bleeding. World J Gastrointest Pathophysiol . 2014;5(3):284–92 Smith LA, Morris AJ, Stanley AJ. The use of hemospray in portal hypertensive bleeding; a case series. J Hepatol . 2014 Feb;60(2):457–60

PTH-010 Uk training in therapeutic endoscopy – are we achieving the basics? results of the bsg national training survey

Introduction Colonoscopic polypectomy and endoscopic treatment of acute upper gastrointestinal bleeding (AUGIB) are key skills for the majority of consultant gastroenterologists. Competence in polypectomy is assessed by direct observation of polypectomy skills (DOPyS) with 4 competent level 2 (>1 cm) polypectomies required for full colonoscopy certification. Measures for defining competence in endoscopic treatment of AUGIB are in development. There is evidence of slow rates of progression to achieve basic endoscopic skills amongst gastroenterology trainees due to other service and training pressures. 1 Method The BSG national training survey was open to all UK gastroenterology trainees from June–August 2014. Data on exposure to polypectomy and endoscopic treatment of AUGIB was analysed as well as trainees desire to train in advanced endoscopic techniques. Results 32.6% (263/806) of UK trainees completed the survey. 68% of ST6 (year 4) trainees and 38% of final year trainees have performed senior trainees ) wish to formally train in endoscopic mucosal resection (EMR), with corresponding figures of 40% for ERCP, 29% for EUS, and 33% for enteroscopy. 28% of senior trainees report having the opportunity to perform endoscopy in a patient with AUGIB either monthly or less often (less than once per month/never). >95% of senior trainees feel they will be competent to perform variceal and non-variceal AUGIB endoscopic treatment as a newly qualified consultant. 91% feel they will be competent to place a Sengstaken-Blakemore tube if required. Conclusion Exposure to level 2 polypectomy amongst senior gastroenterology trainees is limited, which may explain slow rates of progression to full colonoscopy certification. 1 Over a quarter of senior trainees have limited opportunity to perform endoscopy in cases of AUGIB. Virtually all trainees feel they will be competent in the endoscopic management of AUGIB as a new consultant, but the documented lack of exposure for some trainees suggests this may not be the case. Despite slow progression to basic endoscopy competencies, large numbers of trainees wish to train in advanced endoscopic techniques. These results suggest the need for focused training lists to develop key therapeutic endoscopic skills such as removal of level 2 polyps, and systems to guarantee dedicated supervised exposure to AUGIB cases. Disclosure of interest None Declared. Reference Neale JR, Basford PJ. General medical training in gastroenterology: views from specialist trainees on the challenges of dual accreditation. Clin Med. 2015;15(1):35–9

Novel Use of Glidescope Indirect Laryngoscopy for Insertion of a Minnesota Tube for Variceal Bleeding

BackgroundWith improvements in endoscopic and interventional radiologic therapies, insertion of gastroesophageal balloon tamponade catheters, commonly known as Sengstaken-Blakemore or Minnesota tubes, is a rarely performed procedure for esophageal or gastric variceal bleeding. In small hospitals or freestanding emergency departments, endoscopic or interventional radiology (IR) therapies might not be available, so patients with exsanguinating variceal bleeding must be stabilized or temporized for transport to larger hospitals. Occasionally, tamponade devices are necessary as a rescue therapy for failed endoscopic or IR therapies or can be used as definitive therapy in select cases. In addition to being rarely performed, there are multiple technical complications associated with blind insertion of tamponade catheters. DiscussionWe describe a novel use of indirect laryngoscopy using a Glidescope for assisting in placement of a Minnesota tube in 4 patients with exsanguinating esophageal bleeding. Conclusions: Insertion of a Minnesota tube for bleeding esophageal or gastric varices is an uncommon, technically challenging procedure that can be lifesaving, and is something emergency physicians, intensivists, and gastroenterologists should be capable of performing. Addition of indirect laryngoscopy may help to improve rapid, safe, and successful placement of these devices.

Surviving Sengstaken

Aim of the StudyTo report the outcomes of children who underwent Sengstaken–Blakemore tube (SBT) insertion for life-threatening haemetemesis. MethodsSingle institution retrospective review (1997–2012) of children managed with SBT insertion. Patient demographics, diagnosis and outcomes were noted. Data are expressed as median (range). Main Results19 children [10 male, age 1 (0.4–16) yr] were identified; 18 had gastro-oesophageal varices and 1 aorto-oesophageal fistula. Varices were secondary to: biliary atresia (n=8), portal vein thrombosis (n=5), alpha-1-anti-trypsin deficiency (n=1), cystic fibrosis (n=1), intrahepatic cholestasis (n=1), sclerosing cholangitis (n=1) and nodular hyperplasia with arterio-portal shunt (n=1). Three children deteriorated rapidly and did not survive to have post-SBT endoscopy. The child with an aortooesophageal fistula underwent aortic stent insertion and subsequently oesophageal replacement. Complications included gastric mucosal ulceration (n=3, 16%), pressure necrosis at lips and cheeks (n=6, 31%) and SBT dislodgment (n=1, 6%). Six (31%) children died. The remaining 13 have been followed up for 62 (2–165) months; five required liver transplantation, two underwent a mesocaval shunt procedure and 6 have completed endoscopic variceal obliteration and are under surveillance. ConclusionsSBT can be an effective, albeit temporary, life-saving manoeuvre in children with catastrophic haematemesis.

A Blakemore for the Bile Duct

Case Report: An 86-year-old woman presented with 1 day of epigastric pain, vomiting, and chills. On exam, she was normotensive, afebrile, and had epigastric tenderness. Laboratory findings demonstrated: TBili 3.4 mg/dL, DBili 2.6 mg/dL, SGOT 410, SGPT 316, WBC 9.3 K/uL (90% neutrophils). Ultrasound showed gallstones, gallbladder wall thickening, and a common bile duct (CBD) measuring 8-9 mm. CT revealed gallbladder wall thickening. Choledocholithiasis was suspected and ERCP was performed. Cholangiogram revealed dilated CBD with a small stone, and sphincterotomy was performed without bleeding. With multiple balloon sweeps, 1 gallstone was extracted from the CBD. During extraction, fluoroscopy revealed extravasation of dye outside the distal CBD. Shortly thereafter, copious blood was seen in the duodenum. A 10 x 60-mm Boston Scientific® Wallflex fully covered stent (SEMS) was then deployed into the CBD under fluoroscopic and wire guidance. No source of bleeding was identified with either duodenoscope or gastroscope. After resuscitation, an emergency CTA was performed and showed portal venous gas, portal vein thrombus, and biliary varices. No further bleeding episodes occurred this hospitalization. No attempt was made to remove the SEMS out of concern of rupturing additional varices. Discussion: Choledochal varices form from venous dilation in the setting of increased extrahepatic portal pressures. The extrahepatic bile duct is surrounded by 2 different venous systems. Dilation of the paracholedochal veins of Petren, which run parallel to the bile duct, usually causes extrinsic compression of the bile duct wall. With dilation of the epicholedochal venous plexus of Saint, which line the outer surface of the CBD, bleeding and mechanical obstruction may occur1. Although choledochal varices are uncommon, they can lead to life-threatening consequences during endoscopy or surgery. The proposed mechanism is squeezing of the dilated vascular channels proximally resulting in rupture of distal channels2,3. Jacobson et al first described rupture of a choledochal varix following balloon extraction of a stone, resulting in massive bleeding, and subsequently placing a plastic stent.1 Sharma et al described 3 similar cases3. While choledochal variceal rupture following balloon extraction has been described (4-7), this is the first reported case of choledochal variceal bleeding secondary to balloon extraction and successful treatment with the placement of a fully covered SEMS. The proposed mechanism is tamponade of the choledochal varices, similar to the effect of a Blakemore tube on esophageal varices. Future prospective randomized trials are required to explore the use of SEMS for bleeding choledochal varices.

Su1506 Systematic Evaluation of the Complication Rate Following Endoscopic Biopsy of the Major Papilla

(SD) and median (range) were computed for continuous variables and frequency distributions were calculated for all categorical variables. All statistical tests were twotailed with p!0.05 considered to be significant. Results: Patients who underwent ETI for airway protection were more likely to have a history of liver cirrhosis (36 [40%] vs. 15 [21.7%], pZ0.017), massive hematemesis (O500 ml of bloody emesis20 [22%] vs.7 [8.7%], pZ0.003], alcoholism (38 [43%] vs. 12 [17.4%], p!0.0001), gastro-esophageal varices (35 [39%] vs. 16 [23.2%], pZ0.03), hepatic encephalopathy (13 [15%] vs. 3 [4.3%], pZ0.03), sepsis (28 [31%] vs. 3 [4.3%], p!0.0001), and placement of Sengstaken-Blakemore (SB) tube (8 [9%] vs. none, pZ0.009). Pulmonary aspiration was more frequent in the intubated group (36 [40%] vs. 4 [5.8%], p!0.0001). A risk score was calculated using seven variables: liver cirrhosis [score of 1], massive hematemesis or frequent blood transfusions (O3 units, score of 1), presence of varices [score of 1], alcoholism [score of 1], encephalopathy [score of 1], sepsis [score of 1] and placement of SB tube [score of 1]. The cumulative risk score among the intubated group was (1.77 vs. 0.71, p!0.0001) -higher than that of the non-intubated group. Conclusion: Pulmonary aspiration is common complication in patients with UGIB admitted to ICU undergoing EGD. It is frequently associated in patients who are cirrhosis (1), varices (1), encephalopathy (1), alcoholism [(1), hematemesis/ repeated transfusions (1), sepsis (1) and placement of SB tube (1). A risk score using these variables might guide a clinician to use ETI to protect the airways. A score of 2 or more indicates higher risk of pulmonary aspiration, while a score of less than 1 indicates lower risk. A large prospective study to identify the amount of risk contributed by each of the above risk factors might be useful in the future.

Esophageal Perforation: A Rare but Fatal Complication of Urgent Sengstaken Blakemore Tube Intubation

1Department of Internal Medicine, Intensive Care Unit, Faculty of Medicine, Marmara University, Istanbul, Turkey 2Department of Infectious Diseases, Faculty of Medicine, Marmara University, Istanbul, Turkey 3Department of Gastroenterology, Faculty of Medicine, Marmara University, Istanbul, Turkey 4Department of Radiology, Faculty of Medicine, Marmara University, Istanbul, Turkey 5Department of Thoracic Intensive Care, Faculty of Medicine, Marmara University, Istanbul, Turkey

Surgical management of acute esophageal necrosis

I read with great interest the article by Drs. M.H. Wu and H.Y. Wu [1] on acute esophageal necrosis (AEN). The authors described two patients with AEN and emphasized the role of esophagectomy in the treatment of this rare syndrome. While I welcome reports of new cases of ‘‘black esophagus’’ that stir scientific discussion about the condition, I disagree with the proposed, rather aggressive, surgical approach to manage AEN. In my original paper, I introduced a staging system that describes the progression of AEN based on the endoscopic and pathologic findings [2]. It should be noted that black esophagus is a striking medical condition that may present with acute gastrointestinal hemorrhage in patients with hemodynamic compromise, debilitated health, sepsis, multiorgan failure, acute gastric outlet obstruction and diabetic ketoacidosis. The mainstay of treatment of AEN is supportive medical therapy with acid suppression, nil-per-os restriction, red blood cell transfusion, and importantly, correction of the underlying medical condition that has led to the development of the AEN. The overall mortality rate in patients with black esophagus is high at 31 %, but the mortality specific to the esophageal disease is only 6 %. To date, surgical resection has been reserved only for cases of esophageal perforation, an event typically caused by external compression by a ruptured thoracic aorta, gastric volvulus or severe thromboembolic disease [3]. It may also be considered in situations with rapid clinical decompensation in the absence of any other explanation. I absolutely disagree with the authors that esophagectomy is indicated for a wide group of stage 1 AEN cases. I further disagree that surgical intervention is necessary for stages 2 and 3 of AEN to prevent the development of esophageal strictures. A subgroup of patients with diabetes mellitus has a low chance of developing stricture or stenotic complications, and all patients generally do well with endoscopic balloon dilatation [2]. In conclusion, the surgical management of AEN should be performed only in select cases of perforation/mediastinitis. The passage of a nasogastric tube, and especially a Sengstaken–Blakemore tube, in AEN patients should be prohibited, as the friability of the esophagus will lead to tragic perforation and death, as was seen in the second patient. In their description of their two cases of AEN, the authors failed to explain the need for rapid surgical treatment of patients with black esophagus, without first attempting conventional medical management.

Successful Bridging Hemostasis Using a Sengstaken-Blakemore Tube in Massive Rectal Variceal Bleeding

Life-threatening rectal variceal bleeding is a rare complication of liver cirrhosis. Various therapeutic interventions including endoscopic variceal ligation and percutaneous transvenous obliteration have been proposed to control significant rectal variceal bleeding. However, these definite hemostasis modalities are not readily available and require an experienced endoscopist or interventional radiologist. Therefore, bridging therapy to control active bleeding is necessary especially in patients with massive bleeding. We report a case of massive rectal variceal bleeding in which a Sengstaken-Blakemore tube was effective at stopping the bleeding until percutaneous transvenous obliteration could be performed.