Lori J. Herman

In Vivo Full-Thickness Endoluminal Gastroplication Using Tissue Anchors in a Live Pig Model

In Vivo Full-Thickness Endoluminal Gastroplication Using Tissue Anchors in a Live Pig Model Jose G. De la Mora, Elizabeth Rajan, David Rea, Thomas C. Smyrk, Lori J. Herman, Jodie L. Deters, Mary A. Knipschield, Christopher J. Gostout Background: Long-term success of gastric wall apposition performed by flexible endoscopy is dependent on fold permanence. Prior work by our group demonstrated that only full-thickness folds with serosal apposition are durable. Aim: To study feasibility of different tissue anchors to create a full thickness inverted fold and the durability of each single fold plication. Material & Methods: Four 35-45 Kg female pigs were used. Under anesthesia a midline abdominal incision was performed. A 5-cm incision parallel to the greater curvature of the stomach was made. The posterior wall was exposed and longitudinal folds were created by indenting the wall from the serosal side (inverted fold) 1.5 cm in height and 5 cm long. Anchors were deployed to traverse the inverted gastric wall, including apposing serosal surfaces within the fold. Anchors were 1 cm apart with 3-4 of the same type used per fold. 4-6 folds were made in each pig. Four types of paired anchors joined with suture (prolene 2-0) were used: T-bar (T); polypropylene mesh pledget (TM); plastic star-shaped buttons (S) and a self-expanding nitinol basket (B). Suture (vicryl 2-0) for incision closure was used to control for tissue reaction. Follow-up endoscopy was done at 15, 30 and 60 days. Two pigs were sacrificed each at 30 and 60 days. Macroscopic description of the folds was done and samples of the folds sent for histology. Results: Day 15: all folds were still present endoscopically. Day 30: S and B folds were unchanged, TM folds were reduced in height, and T folds had flattened. Day 60: only S & B folds remained. Histologically, all B folds included the muscle layer (30 & 60 day specimens) and one developed serosal fusion (30-day specimen). Only one S fold included the muscle layer with serosal fusion at 60 days. Conclusions: The durability of endoluminally placed full thickness inverted folds remains a challenge. Serosal apposition remains requisite for fold permanence. The use of tissue anchors such as the S and B designs may help achieve greater durability for endoscopic gastric remodeling by tissue apposition. Abstracts

3465 The durability of different solutions injected submucosally in a porcine model.

During endoscopic mucosal resection and polypectomy of sessile lesions an injected submucosal fluid cushion isolates the tissue prior to resection and reduces thermal injury, perforation, and hemorrhage. Aim: To investigate the efficacy of five different solutions in creating a durable submucosal esophageal fluid cushion in a pig model. Methods: Five groups of 5 pigs were studied. In each pig, six submucosal injections of 5 cc of a test solution were performed within the lower third of the esophagus. The time required for the elevated mucosa to flatten completely was recorded after each injection. The solutions used were: normal saline (NS), NS+epinephrine (NSE), dextrose 50% (D50), 10% glycerine/5% fructose in NS (G), and 1% rooster comb hyaluronic acid (HA). Statistical analysis was performed using analysis of variance (ANOVA). NS was used as the standard solution for comparative analysis. Results: The mean and median times (minutes) for each solution to dissipate are provided in the table below. Conclusions: 1. This study provides important and practical information on the durability of available solutions used to create a submucosal fluid cushion during endoscopic mucosal resection and polyectomy of sessile lesions. 2. Hyaluronic acid appears to be the ideal solution for a lasting submucosal cushion during prolonged procedures. 3. Dextrose 50% is superior to normal saline and serves as a practical alternative in terms of availabilty and cost.

3470 Widespread circumferential mucosectomy in the porcine esophagus–a pilot study.

Endoscopic mucosal resection (EMR) is a procedure growing in significance as an alternative to surgery in the treatment of gastrointestinal mucosal lesions. Current EMR techniques are limited to lesions measuring less than 2 cm. Aim: To demonstrate that widespread circumferential mucosectomy up to 5 cm in length can be performed using a stripping method. Method: Eight, 60 kg anesthetized pigs were studied. The distal 5 cms of the esophagus was injected with 50% dextrose to create a lasting submucosal protective cushion. A strip of mucosa approximating 5 cm in length from the esophagogastric junction and 1 cm in width was marginated along its length and proximal end using a porcelain ball-tipped needle knife and prototype teflon backed monopolar scalpel (Olympus America, Inc., NY). The proximal end once undermined was grasped using forceps and stripped off the submucosal cushion in a distal direction. The distal end was then resected using a needle knife from a retroflexed position. This was repeated sequentially until the distal esophagus was circumferentially denuded of mucosa. Four pigs were sacrificed immediately after the procedure while two each of the remaining four animals were followed for two and four weeks. Result: Complete widespread circumferential mucosectomy was achieved in five animals with an average of four strips removed per pig. The average strip length was 5.5 cm(range 5-6 cm) and width 0.9 cm(range 0.6-1.2 cm). There were no procedural complications in the five animals. Mucosectomy was incomplete in the first two animals as the technique of mucosal stripping evolved with another animal experiencing a perforation early in this learning curve. The animals sacrificed immediately post-procedure showed histologically intact muscularis propria with some preservation of the submucosa and minimal associated edema and cautery injury. Repeat endoscopy of the surviving animals (4) showed healing with moderate (1) to tight (3) strictures at both two and four weeks. Conclusion: 1. Widespread circumferential esophageal mucosectomy can be performed endoscopically using a true stripping technique with preservation of the muscularis propria. 2. This technology has exciting potential clinical applications especially in the treatment of Barretts esophagus. 3. The appropriate measures to prevent post-procedural stricture formation need to be further investigated and identified. (The Apollo Group: Sydney Chung, Peter Cotton, Christopher Gostout, Robert Hawes, Anthony Kalloo, Pankaj Pasricha, Thadeus Trus)

⁎4518 Variable rigidity colonoscopy: a prospective randomized pilot study.

Looping can prolong colonoscopy, cause discomfort, and preclude a complete examination. A colonoscope with variable insertion tube rigidity may facilitate colonoscopy. Aim: To determine if VRC can facilitate colonoscopy by reducing insertion time (IT) and improving patient acceptability (PA). Methods: Fifty patients were randomized to undergo conventional (CC; Olympus CF-140) or variable insertion tube rigidity colonoscopy (VRC; distal/insertion tube O.D. 13.2/12.9 mm, working length 133 cm, instrument channel I.D. 3.7 mm, view angle 140°, range of motion 180° up/down, 160° right/left; Olympus America, Inc, NY) by the primary investigator. A rotary dial on the handle adjusted the insertion tube rigidity. After reaching the splenic flexure without increased rigidity, maximum rigidity was maintained to the cecum. PA was assessed on a visual-analogue scale(0=most acceptable/no discomfort). Medication dosage(mg midazolam and/or meperidine) was documented, as was the frequency of abdominal pressure (AP), or repositioning (RP). Statistical analysis was performed by the twosample Wilcoxon rank sum test and an extension of Fisher s Exact test (K exact). Results: The groups were comparable in age (median 63 and 65 years VRC and CC, respectively), gender (14 male VRC, 12 male CC), and sedation required (median midazolam/meperidine dose: 4/75 mg, both groups). The cecum was reached in all 25 VRC cases; one patient had failed CC previously. The cecum was not reached in 4 patients in the CC group (2 poor prep, 2 looping). The median IT was 9 min (10.6±1.6) in the VRC group and 10 min (10.6±1.7) in the CC group (p=0.97). PA was better in the VRC group (0.4±0.2 vs. 1.3±0.6, p=0.14 and less AP (0.3±0.1 vs. 1.1±0.4, p=0.05) or RP (0.4±0.1 vs. 1.2±0.4, p=0.46) was required. VRC patients reported less abdominal distention (0.9+.3 vs. 2.2+.5, p=0.05) and had less pain observed by the endoscopist (2.0+.4 vs. 3.2+.4, p=0.06) and the nurse (1.9+.4 vs. 3.3+.4, p=0.05). No complications occurred. Conclusion: VRC facilitated colonoscopy by reducing abdominal pressure and patient repositioning. Although insertion time was not significantly reduced, patient acceptability was better.

7040 A developmental endoscopy unit: paving the way for the future of endoscopy.

Endoscopy volumes continue to grow despite steady cut-backs in reimbursement. Significant physician and paramedical personnel time is needed to meet this growth providing less time for endoscopic research and development. AIM: To create an environment that facilitates endoscopic research and development. METHODS: A physician and paramedical personnel core group was assembled to include a director, lead gastrointestinal assistant, study coordinator, and research fellow. Dedicated space and equipment were provided. Contacts were established with industry to seek mutual areas of equipment, accessories, and technique development with an emphasis on confidentiality. A system of record keeping using the latest information technologies was established with a monthly and yearly reporting schedule for all activities. The working unit created was entitled the Developmental Endoscopy Unit (DEU). RESULTS: The DEU began operation on 8/1/98. The activities which evolved include: clinical research, prototype endoscope and accessory testing, animal research of new endoscopic techniques, the performance of complex clinical cases using cutting edge technology, and the invention of new equipment. The DEU physician endoscopist staff has grown from one to six. Planned personnel growth includes the addition of a gastrointestinal assistant and research coordinator. A total of 14 new research protocols were initiated the first year by the original core group. Projects include the study of high resolution and high magnification endoscopy, endoscopic anti-reflux methods, and tissue resection techniques. CONCLUSIONS: 1. Given the demanding pace of clinical endoscopy, it is possible to successfully assemble a working unit dedicated to promote the advancement of endoscopic care. 2. The DEU core group was able to efficiently establish a large number of both clinical and animal research projects. 3. The DEU structure enhanced communication and a developmental liaison with industry.