1040 127 Resections for Pancreatic Neuroendocrine Tumor: Evaluating the Impact of Minimally Invasive and Parenchymal-Sparing Surgical Techniques

Abstract

Background: Gastro-gastric fistulae are a challenging complication of obesity surgery that often requires surgical revision. Performance of endoscopic fistula closure is increasing and may provide a less invasive alternative; nevertheless, the majority of fistulae ultimately require surgical revision. The impact of prior endoscopic intervention on surgical revision outcomes remains unknown. We present the largest series on surgical revision of gastro-gastric fistulae to date and perform cohort analysis to assess the impact of prior endoscopic therapy. Methods: A database of all bariatric surgical revisions performed at a single institution was searched for patients with gastro-gastric fistula. Electronic records and clinic charts were then reviewed. The cohort was divided between patients with attempted endoscopic fistula closure prior to surgical revision, and patients without endoscopic therapy prior to revision. 30-day morbidity and mortality was the primary outcome. Age, sex, initial surgery BMI, revision surgery BMI, type of initial surgery, type of revision surgery, number of prior surgeries, number of endoscopic fistula closure attempts, presence of dilated gastrojejunostomy, number of endoscopic clips placed, number of endoscopic sutures placed, fistula size, smoking status, thyroid disease, presence of diabetes, medical co-morbidities, OR time and 30-day minor and major post-operative complications were collected. Categorical and continuous data were analyzed with Fisher's exact test and Wilcoxon signed-rank test respectively. Results: 68 total cases of surgical revision were reviewed. 35 cases were performed for gastro-gastric fistula. Of 35 cases, 22 had attempted endoscopic closure prior to surgical revision while 13 went directly to surgical revision. In the endoscopy group, 2 minor complications and 7 major complications occurred (total 9/22; 40.9%). In the surgery only group, 3 minor complications and 3 major complications occurred (total 6/13; 46.1%). No deaths occurred. No statistical difference existed in the demographic or primary outcome variables of interest between the two groups. Sub-group analysis of the endoscopy group did suggest a relationship between the number of sutures (p=0.04) and clips p=(0.04) placed at the gastrojejunostomy and major 30-day complications. Conclusion: In the largest study of surgical revision of gastro-gastric fistulae to date, there is no evidence that prior attempts at endoscopic fistula closure lead to increased complications at the time of surgical revision.