Mo1260 Eus-Directed Transgastric ERCP (EDGE) Vs Laparoscopy-Assisted ERCP (LA-ERCP) for Roux-En-Y Gastric Bypass (RYGB) Anatomy: A Multicenter Early Comparative Experience of Clinical Outcomes

Mo1260 Eus-Directed Transgastric ERCP (EDGE) Vs Laparoscopy-Assisted ERCP (LA-ERCP) for Roux-En-Y Gastric Bypass (RYGB) Anatomy: A Multicenter Early Comparative Experience of Clinical Outcomes

Abstracts collections can be drained percutaneously, endoscopically or surgically. Endoscopic necrosectomy has been shown on to be safe during short ...

728KB Sizes 1 Downloads 67 Views

Abstracts

collections can be drained percutaneously, endoscopically or surgically. Endoscopic necrosectomy has been shown on to be safe during short term follow up, but data for long term follow up is limited particularly with lumen apposing self expanding metal stents (LASEMS). Methods: Charts of patients who underwent endoscopic necrosectomy for WOPN at our institution from 2012 to 2016 were reviewed. There were 14 patients who met inclusion criteria. Frequency distributions were analyzed. Results: 14 patients were included with long term follow up in 13 patients. Median follow up duration was 15months (range 3-28). Etiology of pancreatitis was gallstones 78%, alcohol 14% and idiopathic 7%. 71% of patients had no prior treatment. Two patients had previous percutaneous drainage of WOPN. Median size of the fluid collection was 12.3cm(6-18.4cm). Median time from pancreatitis to necrosectomy was 13 weeks. LASEMS were used in all patients. Stent placement was successful in 93% while in one patient the gastric and cyst wall separated during index procedure. A median of 3 endoscopic necrosectomies were performed per patient (range 1-8). Stent was in place for a median of 8 weeks (3-24weeks). 28% required additional drainage with nasocystic tube (14%) and surgery (14%). Initial drainage complications (during management of fluid collections) occurred in 36% including bleeding around the stent site(nZ2), stent migration(nZ3), stent lumen occlusion(nZ1), perforation(nZ1), aspiration(nZ1). Median time to resolution of necrosis was 7 weeks (range3-28). Long term complications occurred in 93% of patients (12/13) including intra-abdominal abscess from recurrent pancreatitis(nZ1), bleeding from gastric varices(nZ1), disconnected pancreatic duct syndrome (DPDS)(nZ5), pancreatico-colonic fistula(nZ1), malabsorption(nZ4), failure to thrive and significant weight loss(nZ5), chronic abdominal pain(nZ11), chronic pancreatitis(nZ9), need for TPN/J-tube(nZ5), venous thrombosis(nZ10), duodenal obstruction(nZ2), biliary obstruction(nZ5), pancreatic ascites and pleural effusion(nZ2), pancreatic diabetes(nZ2), recurrent hospitalizations(nZ10). There was recurrence of fluid collections in 5 patients presenting as DPDS, 2 of which required transgastric double pig tail stent placement and 3 required surgery, one with preceding LASEMS which did not resolve collection. One patient died from metastatic pancreatic adenocarcinoma. Conclusion: Walled off necrosis despite successful endoscopic management has multiple long term complications. A multidisciplinary approach and long term follow up is needed. Many patients need ongoing management of complications. Additional multicenter data is being collected to report a larger series of patients.

Patient demographics, abscess location, etiology, and size Age/Sex

58/F

Abscess location (cm proximal to anal verge) 12

50/F 73/F

8 10

78/F

11

46/F

10

53/F

10

27/F

5

59/F

10

19/M

10

53/M

8

49/M

10

Etiology

Abscess Size (mm)

LAMS size (mm)

post-operative following sigmoidectomy uterine fibroids post-operative following sigmoidectomy

53 x 46

10

80 x 51 100 x 61

NA

15 10 and 10 x 5 mm double pigtail stent through LAMS 10

60 x 50

15

71 x 42

10

73 x 40

15

54 x 45

10

65 x 55

15

70 x 55

10

70 x 25

15

sigmoid colon cancer, colouterine fistula post-operative following hysterectomy post-operative following sigmoidectomy post-operative following ovarian cystectomy perforated appendicitis perforated appendicitis perforated gastric ulcer tuba-ovarian abscess

Mo1259 EUS Guided Drainage of Pelvic Fluid Collections With Lumen-Apposing Metal Stents Amar Manvar*1, Amit H. Sachdev2, Sammy Ho1, Amit P. Desai3, Kunal Karia1, Safeera Javed1, Tamas A. Gonda3, Nadia Ansari3, Amrita Sethi3 1 Montefiore Medical Center, New York, NY; 2Downstate Medical Center, Brooklyn, NY; 3Columbia University Medical Center, New York, NY Background: Endoscopic ultrasound (EUS) drainage of pelvic fluid collections is a non-invasive alternative to percutaneous or surgical drainage. EUS-guided pelvic abscess drainage has historically been performed using double pigtail plastic stents. Recent developments of EUS-guided procedures using lumen apposing metal stents provide an alternative to plastic stents in the management of pelvic fluid collections. In this multicenter study, we describe our experience using lumen-apposing metal stents (LAMS) in the drainage of pelvic fluid collections. Patients and Methods: From May 2015 to October 2016, all data from patients undergoing endoscopic drainage of pelvic fluid collections at two academic medical centers were retrospectively reviewed. All procedures were performed by three expert endosonographers using LAMS. Cross-sectional imaging was reviewed in conjunction with radiologists to evaluate the feasibility of EUS-guided drainage. Technical success was defined by visualization of purulent drainage into the colonic lumen via proximal flange after stent deployment. Surveillance scans were obtained to ensure radiographic resolution prior to stent removal. A descriptive analysis was then performed. Design: Retrospective chart review. Results: Eleven pelvic fluid collections were drained endoscopically. Specific etiologies are listed in Table 1. The mean size of the lesions was 70 x 50.9 mm. Technical success rate was 100%. Clinical success, defined by clinical and radiographic resolution, was 100%. The cautery-enhanced LAMS was used in 64% (nZ7) of cases. No complications were noted. Conclusions: Endoscopic management of pelvic fluid collections using LAMS is a safe and effective alternative to conventional management including percutaneous, surgical and endoscopic drainage via plastic stents. Further research needs to be conducted to determine the comparable efficacy of this technique.

Pre-stent and Post-stent radiographic views of peri-rectal abscess, EUS image of peri-rectal abscess, and endoscopic image of post-stent deployment with drainage of purulent debris from abscess.

Mo1260 Eus-Directed Transgastric ERCP (EDGE) Vs LaparoscopyAssisted ERCP (LA-ERCP) for Roux-En-Y Gastric Bypass (RYGB) Anatomy: A Multicenter Early Comparative Experience of Clinical Outcomes Prashant Kedia*2, Paul R. Tarnasky2, Stephen L. Steele2, Ali Siddiqui3, Ming-ming Xu1, Amy Tyberg1, Reem Z. Sharaiha1, Monica Gaidhane1, Michel Kahaleh1 1 Weill Cornell Medical Center, New York, NY; 2Gastroenterology, Methodist, Dallas, TX; 3Gastroenterology, Thomas Jefferson, Philadelphia, PA Introduction: Gastroenterologists are frequently encountered by the challenge of managing pancreaticobiliary disease in altered Roux-en-Y gastric bypass (RYGB) anatomy. The standard of care for biliary access in these patients is laparoscopyassisted ERCP (LA-ERCP), but is limited by cost and adverse events. Recently a completely endoscopic approach using endoscopic ultrasound (EUS) to create a gastrogastric fistula, using a lumen-apposing metal stent (LAMS), through which an

AB480 GASTROINTESTINAL ENDOSCOPY Volume 85, No. 5S : 2017

www.giejournal.org

Abstracts

ERCP can be performed has been described. The benefits of EUS-directed transgastric ERCP (EDGE) are that it is minimally invasive and performed by a single team. The clinical outcomes of this new technique have not been compared to the surgical standard. Methods: Charts of patients from May 2005 to October 2016 with RYGB anatomy having undergone ERCP at four tertiary care centers were captured in a dedicated registry. Any RYGB patient having undergone either LA-ERCP or EDGE was included in the study. Patient demographics along with procedural and clinical outcomes were measured for each group. Chi Square and T tests were conducted to compare demographics and outcomes for both groups. Results: A total of 71 patients (n Z 28 EDGE, nZ43 LA ERCP) were included in this study. There was no significant difference in the baseline demographics between the two groups (age, sex), or the indication for ERCP. There was no significant difference in the technical success of EDGE gastrogastric fistula (96.4%) vs LA-gastrostomy creation (100%). The success rate of achieving therapeutic ERCP (EDGE 96.4% vs LA-ERCP 97.8%) and number of ERCP (EDGE 1.2 vs LA-ERCP 1.02) needed to achieve clinical resolution was similar between both groups. The overall adverse event rate for EDGE, 21% (6/28) and LA-ERCP, 18% (8/43) was similar. The total procedure time (79 vs 183.5 min; p value <.00001) and length of hospital stay (0.7 vs 2.65 days p value Z 0.0003) was significantly shorter for EDGE compared to LA-ERCP. The overall weight change after EDGE was -1.4lbs at an average 20 week follow-up. Conclusions: Pancreaticobiliary therapy in RYGB is a growing challenge. The optimal biliary access procedure would be one that can be performed by a single team in a minimally invasive fashion. This comparative analysis suggests that the EDGE procedure has similar technical success and adverse events compared to LA-ERCP with the benefit of significantly shorter procedure times and hospital stay. Compared to LA-ERCP, EDGE may offer a minimally invasive, effective option, with less resource utilization, and without significant weight gain for pancreaticobiliary disease in RYGB patients. Prospective studies are needed to confirm this notion.

EDGE vs LA-ERCP demographic and procedure outcomes Number Age (years) Gender (M) Indication (biliary/pancreatic) Technical success of gastric access Technical success of therapeutic ERCP Adverse Events Total # of ERCP for clinical success Total Procedure Time (min) Hospital Length of Stay (Days)

EDGE 28 56 (35-82) 4/28 23/5 96.4% (27/28) 96.4% (27/28) 21.4% (6/28) 1.2 (1-3) 79 (24-230) 0.7 (0-5)

LA-ERCP 43 55 (3-80) 7/43 36/7 100% (43/43) 97.8% (42/43) 18.6% (8/43) 1.04 (1-2) 183.5 (55-393) 2.65 (1-12)

P-value P Z 0.72 P Z 0.82 P Z 0.86 P Z 0.21 P Z 0.75 P Z 0.64 P Z 0.16 P < 0.00001 P < 0.0003

Mo1261 Stent Patency in Endoscopic Ultrasound Guided Transmural Versus Transpapillary Biliary Drainage Ravindra Gaadhe1, Amol Bapaye*1, Nachiket Dubale1, Mahesh Mahadik1, Rajendra Pujari1, Harshal P. Gadhikar1, Suhas Date1, Jay A. Bapaye2 1 Shivanand Desai Center for Digestive Disorders, Deenanath Mangeshkar Hospital and Reseach Center, Pune, Maharashtra, India; 2 Smt. Kashibai Navale Medical College, Pune, Maharashtra, India Background and Aim: Endoscopic ultrasonography guided biliary drainage (EUS-BD) is an alternative to failed ERCP. It is performed either by transmural (EUS-choledocoduodenostomy, EUS-CDS or hepatogastrostomy, EUS-HGS; EUS-BD-TM) or transpapillary (antegrade or rendezvous; EUS-BD-TP) techniques. Comparative data regarding efficacy of these modalities is limited. This retrospective study compares long term stent patency after EUS-BD via TM or TP approach using metal stents in inoperable malignant biliary obstruction (MBO). Methods: Data of patients undergoing EUS-BD using metal stents for unresectable MBO after failed ERCP from 2011 to 2015 was retrieved from a prospectively maintained database. Parameters included etiology and disease stage, laboratory, imaging, reason for failed ERCP, EUS-BD technique (TM / TP). Follow up was obtained and one of the following endpoints was identified - recurrence of jaundice, re-intervention or death, whichever was earlier. Statistical analysis was performed using Kaplan-Meier graph and logrank test. P<0.05 was considered significant. Results: Total ERCP’s Z 4064; EUS-BD Z 108 (2.6%); EUS-BD using metal stents for MBO Z 71; follow up available Z 56 (79%). EUS-BD-TM Z 33 & EUS-BD-TP Z 23. Both groups comparable for demographic and clinical characteristics. Overall median stent patency Z 77 days (IQR 48 – 228); median patency in TM group Z 69 days (IQR 51 – 240) vs. TP Z 87 days (IQR 42 – 213) (p Z 0.35). Stent related adverse events TP Z 8 vs. TM Z 4 (pZ0.080); stent migration – TP Z 0 vs TM Z 2; stent occlusion or cholangitis (TP Z 8 vs TM Z 2). Conclusions: Outcomes of EUS-BD by TM or TP approach are comparable in terms of stent patency and frequency of adverse events.

www.giejournal.org

Mo1262 Cholecystectomy After Endoscopic Ultrasound Guided Gallbladder Drainage? Absolutely! Monica Saumoy*1, Amy Tyberg1, Elizabeth Brown1, Soumitra R. Eachempati1, Michael Lieberman1, Rastislav Kunda2, Natalie Cosgrove3, Apeksha Shah3, Ali Siddiqui3, Monica Gaidhane1, Reem Z. Sharaiha1, Michel Kahaleh1 1 Weill Cornell Medical Center, New York, NY; 2Gastroenterology, Aarhus University Hospital, Aarhus, Denmark; 3Gastroenterology, Thomas Jefferson, Philadelphia, PA Background: Endoscopic ultrasound guided gallbladder (EUS-GLB) drainage is becoming a widely accepted alternative for patients who are not eligible for cholecystectomy (CCY). However, once a patient’s medical comorbidities improve, they can become surgical candidates for CCY. There is a theoretical concern that after EUS-GLB drainage, the resulting fistula will lead to more surgical complications because the gallbladder is adherent to the gastric wall. But CCY has been successfully preformed after percutaneous gallbladder (PTC-GLB) drainage despite the resulting cutaneous fistula. Therefore the aim of this study was to report feasibility and outcomes of CCY after EUS-GLB drainage in comparison with PTC-GLB drainage. Methods: This was a multi-centered cohort study of patients who underwent successful EUS-GLB drainage, or PTC-GLB drainage, all of whom subsequently underwent a surgical CCY after resolution of cholecystitis. Baseline patient characteristics

Volume 85, No. 5S : 2017 GASTROINTESTINAL ENDOSCOPY AB481