Endoscopic Ultrasound-Guided Coil Embolization With Absorbable Gelatin Sponge Appears Superior to Traditional Cyanoacrylate Injection for the Treatment of Gastric Varices.

INTRODUCTION: Gastric variceal (GV) bleeding is a feared complication of cirrhosis. Traditional endoscopic treatment with cyanoacrylate (CYA) injection can be challenging. Alternatively, endoscopic ultrasound (EUS)-guided delivery of hemostatic coils has shown high therapeutic success without the complications profile of CYA alone. Our aim was to compare the clinical outcomes of EUS-guided coil embolization with endoscopic CYA injection for the treatment of GV. METHODS: We performed a matched cohort study using a prospective registry involving 2 tertiary centers. A total of 10 patients undergoing EUS-based coil therapy were matched in 1:3 fashion to 30 patients who underwent CYA injection. The matching criteria included type of GV, Charlson comorbidity index, and bleeding severity. Primary outcomes were technical success and complications. Secondary outcomes were rebleeding rates, reinterventions rates, total transfusion requirements, and time-to-event analysis (rebleeding, reintervention, and transfusion). RESULTS: Technical success was 100% for EUS coil therapy vs 96.7% for CYA injection (P = 1.0). Complication rates were 10% in the EUS coil group vs 20% in the CYA group; P = 0.65. At 9 months, no EUS coil patient had rebled compared with 38% of the CYA group. No EUS coil patient required blood transfusion for GV rebleed, whereas over 50% of CYA patients did. Ten percent of EUS coil patients required reintervention compared with 60% of CYA patients. The EUS coil group had superior time to reintervention, GV rebleed, and transfusions (all P < 0.05). DISCUSSION: Compared with CYA, EUS-guided coil injection appears superior for the treatment of GV and should be considered initial endoscopic treatment of choice in centers with interventional EUS expertise.

Comparative Study Evaluating Lumen Apposing Metal Stents Versus Double Pigtail Plastic Stents for Treatment of Walled-Off Necrosis.

OBJECTIVES: Lumen-apposing metal stents (LAMSs) are increasingly used for direct endoscopic necrosectomy (DEN) in the management of walled-off necrosis (WON). We compared LAMS and traditional cystoenterostomy in the management of WON. METHODS: Retrospective analysis of patients who underwent DEN for management of WON was performed. Primary outcome was rate of WON resolution. Secondary outcomes included technical and clinical success, time and number of procedures until resolution, requirement for alternative therapy, recurrence, and adverse events. RESULTS: One hundred twelve patients underwent DEN with LAMS (n = 34) or traditional cystoenterostomy (n = 78). Mean WON size was 90.2 × 60.1 mm, and 61.8% had infected necrosis. Overall WON resolution was similar between LAMS and traditional cystoenterostomy (94.1% vs 92.1%, P = 0.510), with similar number of procedures until resolution (1.5 vs 1.5, P = 0.871). The LAMSs were associated with faster resolution (86.9 vs 133.6 days, P = 0.038), lower recurrence (6.3% vs 23.1%, P = 0.032), and decreased requirement for surgery (0% vs 12.8%, P = 0.031) compared with traditional cystoenterostomy, but higher adverse event rates (41.2% vs 7.7%, P < 0.001). CONCLUSIONS: Despite higher adverse event rates, initial LAMS cystoenterostomy for management of WON results in faster resolution, lower recurrence, and decreased requirement for surgery.

Efficacy of a reverse-phase polymer as a submucosal injection solution for EMR: a comparative study (with video).

INTRODUCTION: Reverse-phase polymers undergo a temperature-dependent liquid-to-gel transition that may provide a more durable cushion for EMR and could yield improvements in safety and efficacy. OBJECTIVE: Our purpose was to assess the efficacy of poloxamer solution PS137-25 as a submucosal injection. DESIGN: Comparative study among 3 different solutions for EMR using ex vivo and in vivo porcine models. INTERVENTIONS: A total of 120 gastric submucosal cushions were performed in fresh ex vivo porcine stomachs with saline solution 0.9% (n = 40), hydroxypropyl methylcellulose (HPMC) (n = 40), and PS137-25 (n = 40). Five in vivo porcine colon EMRs were then performed. MAIN OUTCOME MEASUREMENTS: Height and duration of submucosal cushions were measured in the ex vivo model, and the usefulness of reverse-phase polymers was assessed by in vivo en bloc resection. RESULTS: Ex vivo: The height of mucosal elevation was greater with PS137-25 (10.3 +/- 2.2 mm) than with saline solution (8.3 +/- 2.6 mm) or HPMC (9.05 +/- 2.3 mm, not significant). All submucosal elevations with PS137-25 lasted longer (more than 20 minutes) than with saline solution (20.9 +/- 11 minutes, P < .01) or HPMC (89 +/- 32 minutes, P < .01). After 120 minutes, the elevations obtained with PS137-25 showed no differences in size, shape, or consistency from initial measurements. In vivo: All 5 EMRs were successfully performed after 1 injection of PS137-25. No repeat injections were needed. During resection, a large gel cushion was noted. CONCLUSIONS: Mucosal elevation with PS137-25 is more durable than with other substances, with no changes in size or consistency observed over 120 minutes. Additionally, reverse-phase polymers performed well in in vivo colon EMRs. These results suggest that reverse-phase polymers may provide increased safety and efficacy for EMR procedures.

Endoscopic ultrasound-guided celiac plexus neurolysis using a reverse phase polymer.

AIM: To assess the feasibility of endoscopic ultrasound (EUS)-guided celiac plexus neurolysis (CPN) using a poloxamer. METHODS: In this prospective evaluation, six Yorkshire pigs underwent EUS-guided CPN. Three received an injection of 10 mL of 0.25% Lidocaine plus methylene blue (group 1) and three received an injection of 10 mL of 0.25% Lidocaine plus blue colored poloxamer (PS137-25) (group 2). Necropsy was performed immediately after the animals were sacrificed. The abdominal and pelvic cavities were examined for the presence of methylene blue and the blue colored poloxamer. RESULTS: EUS-guided CPN was successfully performed in all 6 pigs without immediate complication. Methylene blue was identified throughout the peritoneal and retroperitoneal cavity in group 1. The blue colored poloxamer was found in the retroperitoneal cavity immediately adjacent to the aorta, in the exact location of the celiac plexus in group 2. CONCLUSION: EUS-guided CPN using a reverse phase polymer in a non-survival porcine model was technically feasible. The presence of a poloxamer gel at the site of the celiac plexus at necropsy indicates a precise delivery of the neurolytic agent.

Endoscopic evaluation of the defunctionalized stomach by using ShapeLock technology (with video).

BACKGROUND: Patients with Roux-en-Y gastric bypass (RYGB) present a unique problem if they require diagnostic or therapeutic interventions for which the pancreatobiliary limb or the defunctionalized stomach must be accessed. Novel shape-locking guides have been reported in the literature to reduce looping during colonoscopy, and a new guide is now available to assist with enteroscopy. OBJECTIVE: To use ShapeLock technology to permit evaluation of the defunctionalized stomach. DESIGN: Observational case series. SETTING: Tertiary-care center. PATIENTS: Nine patients with a history of RYGB referred for repeat endoscopic evaluation after initial enteroscopy failed to reach the excluded stomach. INTERVENTIONS: After achieving appropriate levels of sedation, a standard enteroscope was back-loaded with the ShapeLock enteroscopy guide and was inserted through the mouth. The device was moved through the gastrojejunal (GJ) anastomosis, along the Roux limb, and into the distal pancreatobiliary limb. The device was then locked, which allowed the enteroscope to be advanced to the defunctionalized stomach. RESULTS: The ShapeLock guide was able to be advanced to the excluded stomach and perform a thorough examination of the pancreatobiliary limb in 8 of 9 patients, without complications. In 1 patient, the diameter of the GJ anastomosis prevented passage of the device. CONCLUSIONS: The ShapeLock enteroscopy guide can allow access to the upper-GI tract in patients after RYGB, provided the GJ anastomosis is of adequate diameter. This study suggested that the technique is safe and has the potential to allow therapeutic interventions in the defunctionalized stomach and duodenum, including ERCP.

Peroral transgastric organ resection: a feasibility study in pigs.

BACKGROUND: The feasibility of peroral transgastric peritoneoscopy, liver biopsy, and tubal ligation has been demonstrated in prior animal studies. This approach has the potential to reduce postoperative morbidity. OBJECTIVE: To explore the technical challenges and complications of performing a transgastric organ resection. DESIGN: Two-week animal survival study. INTERVENTIONS: Five female Yorkshire pigs underwent peroral transgastric partial hysterectomy. With the animals under general anesthesia, a sterile esophageal overtube was placed and a gastric antibiotic lavage was performed. Subsequently, a needle-knife and through-the-scope dilating balloon were used to make an anterior gastric wall incision through which a therapeutic gastroscope was advanced into the peritoneal cavity. Endoscopes and instruments previously had been disinfected in a 2.4% glutaraldehyde bath or were disposable. An endoloop and polypectomy snare were used to resect a portion of the uterus, which was removed through the mouth. Gastric incision closure was attempted with endoclips. After 2 weeks of observation, necropsy was performed. RESULTS: Transgastric partial hysterectomy was performed on all 5 animals. One animal became lethargic and febrile, so necropsy was performed on day 4. An incompletely closed gastric incision and suppurative peritonitis were found. A second animal developed a fever; necropsy performed on day 14 revealed a gastric abscess at the incision site and scattered abdominal pus. No complications were encountered in the 3 remaining animals, and necropsy revealed well-healed gastric incisions and no evidence of peritonitis. CONCLUSION: Endoscopic transgastric partial hysterectomy is technically feasible in a porcine model. Strict sterility seems mandatory but cannot always be assured. Incomplete closure of the gastric incision may lead to significant complications.