RESUMO

BACKGROUND: A recent expert panel issued recommendations about the technical aspects of direct endoscopic necrosectomy (DEN) for pancreatic walled-off necrosis (WON). However, significant technical heterogeneity still exists among endoscopists. AIM: To report the outcomes of our DEN technique and how it differs from a recent expert consensus statement and previous studies. METHODS: Medical records of patients with WON who underwent DEN from September 2016 - May 2019 were queried for the following information: Age, gender, ethnicity, etiology of acute pancreatitis, WON location and size, DEN technical information, adverse events (AEs) and outcomes. Adverse events were graded according to the American Society of Gastrointestinal Endoscopy Lexicon grading system. Technical success was defined as adequate lumen apposing metal stent (LAMS) deployment plus removal of ≥ 90% of necrosum. Clinical success was defined as complete resolution of WON cavity by imaging and resolution of symptoms at ≤ 3 months (mo) after last DEN. Data analysis was performed using mean and standard deviation for continuous variables, frequency and proportion for categorical variables, and median and range for interval data. RESULTS: A total of 21 patients underwent DEN. Most of them were Hispanic and their mean age was 51 ± 17 years. The majority of the collections were located in the body of the pancreas and the mean size was 13 cm ± 5 cm. The most common indication was persistent vomiting. Antibiotics were administered only in cases of infected necrosis. All LAMS were placed without radiological guidance, dilated the same day of deployment and removed after a mean of 27 ± 11 d. Routine cross-sectional imaging immediately after drainage was not performed. The mean interval between DEN sessions was 7 ± 4 d and the mean number of DEN/patient was 3 ± 2. Technical and clinical success rates were both 95%. AEs were seen in 5 patients and included: Sepsis (2), stent migration (1), stent maldeployment (1), perforation (1). The sensitivity and positive predictive value of an occluded LAMS leading to sepsis was 50% and 0.11 respectively. No fatalities were observed. CONCLUSION: Our DEN technique differed significantly from the one recommended by a recent expert panel and the one published in previous studies. Despite these differences excellent clinical outcomes were obtained.