mortality. Previous reports suggested that transgastric endoscopic necrosectomy is a safe and effective treatment. However, some pancreatic abscesses are located far away from the stomach. Aim: To evaluate if percutaneous endoscopic therapy offers a safe and more effective alternative treatment for pancreatic abscess. Methods: A retrospective analysis of the clinical outcome of 5 consecutive patients treated with the percutaneous approach because the pancreatic abscesses were located away from the gastric wall. Patients were either unfit for surgery or refused surgical treatment. They received treatment with Carbapenems and Fluoroquinolones or other broad-spectrum antibiotics. Percutaneous drainage of the abscess was performed using abdominal ultrasound-guided or CT-guided puncture and placement of a 12Fr PTCD drainage tube into the abscess cavity. The tube was kept in place for 1 to 2 weeks to allow the percutaneous sinus track to form. The track was then dilated and a flexible endoscope (regular gastroscope, nasal endoscope or choledochoscope) was inserted via the sinus track into the abscess. Pus was suctioned through the biopsy channel and a basket was used to remove the necrotic tissue. A double lumen tube (or two tubes) was inserted over a guide wire into the abscess cavity to provide (irrigation and) drainage. Results: Pancreatic abscesses were successfully drained in all 5 patients thus avoiding emergency surgery. Subsequent surgery was not necessary in 4 patients during a median follow-up of 7.6 months (range 5-12 months). Combined surgery and endoscopic therapy was performed in 1 patient because the necrotic tissue could not be removed completely with endoscopy. There were no complications. Conclusion: Preliminary study showed that debridement of pancreatic abscess with percutaneous flexible endoscopic therapy was feasible and safe. Further work is in progress. Figure 1 A Abdominal wall puncture B Gastroscope inserted into the abscess C Necrotic tissue in abscess cavity D The abscess after debridement Figure2 A X-ray view of endoscope in abscess B Removed necrotic tissue C Double lumen-tube drainage D Postoperative CT scan. Sa1427 Duodeno-Jejunal Bypass Liner for the Treatment of Obesity and Type 2 Diabetes: 1-Year Treatment With 6-Months Post-Removal Follow-up Eduardo G. De Moura* 1 , Bruno C. Martins 1 , Guilherme Sauniti 1 , Ivan R. Orso 1 , Suzana L. De Oliveira 1 , Marco AuréLio Santo 2 , Manoel P. Galvao Neto 3 , Almino C. Ramos 3 , Marcio C. Mancini 4 , Arthur B. Garrido 2 , Alfredo Halpern 4 , Paulo Sakai 1 , Ivan Cecconello 2 1 Gastroenterology Department - University of Sao Paulo, Endoscopy Unit - Hospital das Clinicas, Sao Paulo, Brazil; 2 Gastroenterology, Gastrointestinal Surgery - University of Sao Paulo - School of Medicine, Sao Paulo, Brazil; 3 Gastroenterology, Gastro Obeso Center, Sao Paulo, Brazil; 4 Endocrinology, University of Sao Paulo - School of Medicine, Sao Paulo, Brazil Background: endoscopic duodeno-jejunal bypass liner (DJBL) promotes functional exclusion of duodenum and initial jejunum. Previous studies have shown effective weight loss and improvement of type 2 diabetes mellitus (T2D). However, post-removal results are lacking. The aim of this study was to evaluate the efficacy and safety of DJBL, maintained for 12 months, on weight loss and T2D control, and the results after its removal. Methods: open-label, cohort, prospective study, including 22 patients with obesity (BMI  35 kg/m2) and T2D, candidates for bariatric surgery, submitted to DJBL implant maintained for 12 months, and with 6-months follow-up after removal. Results: DJBL was successfully implanted in all patients. Thirteen patients completed 12 months with device and 12 returned 6 moths after removal. Intention-to-treat analysis was done including all patients with device implanted (n = 22). Comparing baseline with last observation carried forward (LOCF), there was significant reduction of weight (119,4  23 kg vs. 102,1  18,7 kg - p  0,001), BMI (45,3  7,1 kg/m2 vs. 38,7  5,7 kg/m2 - p  0,001), fasting glucose (179,4  68,8 vs. 139  58,1 mg/dL - p  0,001) and HbA1c (8,9  1,7% vs. 6,9  1,2% - p  0,001). On per protocol analysis, the 12 patients who returned 6 months after device removal were considered. Weight loss and glycemic control obtained with DJBL were maintained at the end of follow-up (weight, BMI and HbA1c - p  0,001 compared with baseline; fasting glucose - p= 0,001 compared with baseline). There was also significant reduction of triglycerides, cholesterol, insulin and C-peptide at the LOCF (ITT analysis). Conclusion: DJBL maintained for one year is safe and efficient for weight loss and T2D control. Benefits are maintained 6-months after removal. Sa1428 Improvement of Insulin Resistance and Reduction of Cardiovascular Risk Among Obese Patients With Type 2 Diabetes With the Duodenojejunal Bypass Liner Eduardo G. De Moura* 1 , Ivan R. Orso 1 , Bruno C. Martins 1 , Guilherme Sauniti 1 , Suzana L. De Oliveira 1 , Marco AuréLio Santo 2 , Marcio C. Mancini 4 , Manoel P. Galvao Neto 3 , Almino C. Ramos 3 , Arthur B. Garrido 2 , Alfredo Halpern 4 , Paulo Sakai 1 , Ivan Cecconello 2 1 Gastroenterology Department - University of Sao Paulo, Endoscopy Unit - Hospital das Clinicas, Sao Paulo, Brazil; 2 Gastroenterology, Gastrointestinal Surgery - University of Sao Paulo - School of Medicine, Sao Paulo, Brazil; 3 Gastroenterology, Gastro Obeso Center, Sao Paulo, Brazil; 4 Endocrinology, University of Sao Paulo - School of Medicine, Sao Paulo, Brazil Background: To evaluate the effectiveness of the Duodeno Jejunal Bypass Liner (DJBL) in the improvement of resistance to insulin and reduction of cardiovascular risk among morbidly obese patients with type 2 diabetes mellitus (T2DM), using the Triglycerides/HDL cholesterol ratio (TG/HDL), the percentage of weight loss, and glycemic control. Methods: This study included 54 patients implanted with the DJBL and followed for a period of 6 months. All patients had a TG/HDL ratio above 3.5, suggesting greater insulin resistance and lipid profile consistent with increased cardiovascular risk. The initial value of the ratio was compared with the value obtained 6 months after device implantation, in order to assess whether this value decreased, indicating an improvement in insulin resistance and cardiovascular risk reduction. We also evaluated the improvement of glycated hemoglobin (HbA1c) levels and the weight loss resulted from the use of the device, and we correlated that with the improvement of the TG/HDL ratio. Results: All patients implanted with the DJBL presented a statistically significant reduction of the HbA1c levels, with the majority of the patients (70.3%) obtaining diabetes control with HbA1c levels lower than 7% at the end of the study. All patients also presented a significant weight reduction, with an average loss of 12.6% of their initial weight. We observed an important improvement in insulin resistance and metabolic syndrome, with a significant reduction of the TG/HDL Abstracts www.giejournal.org Volume 77, No. 5S : 2013 GASTROINTESTINAL ENDOSCOPY AB201