Katherine Manupipatpong, BS, David May, DO, Anthony Petrick, MD, Jon Gabrielsen, MD, William Strodel, MD, Anna Ibele, MD. Geisinger Medical Center, Danville PA.
Introduction: Gastro-gastric fistula (GGF) complicates 11-40% of undivided Roux-en-Y gastric bypasses (RYGB)s and 1-2% of patients with divided gastric staple lines. Repair of GGF represents a technical challenge even to the experienced bariatric surgeon and historically has been performed via open surgical approach. Experienced bariatric centers have increasingly reported series of laparoscopic revision of GGF. We reviewed our clinical series of pouch revisions for GGF with the aim of comparing perioperative outcomes of patients undergoing this technically challenging revision via laparoscopic versus open approach.
Methods and Procedures: We reviewed our IRB-approved prospectively maintained institutional bariatric database of patients who underwent any bariatric operation between 05/2001 and 01/2013. Eighty-five patients with GGF were identified and their electronic medical records were reviewed for preoperative demographics, BMI, indications for surgical revision, surgical approach (open or laparoscopic), perioperative complications, and long term morbidity. Data was analyzed using standard statistical methods with Wilcoxon rank sum test.
Of the 4237 patients in our database, 85 underwent revision for gastrogastric fistula. Sixty-two cases were done open and 23 cases were approached laparoscopically. There were no conversions. Patients did not differ significantly in terms of preoperative comorbidities. The open group had a significantly higher mean BMI prior to revision (43.8 kg/m2 open vs. 34.9 kg/m2 lap, p = 0.0004). Mean operative time was 195 minutes for open revision and 205 minutes for laparoscopic revision (p = 0.25). There was no significant difference between groups in terms of bleeding requiring transfusion (8.7% lap vs. 8.1% open, p = 0.93), leak (8.7% lap vs. 9.7% open, p = 0.89), stricture (0 lap vs. 1.6% open, p = 0.54), pulmonary complications (4.3% lap vs. 14.5% open, p = 0.2), requirement for ICU admission (4.3% lap vs. 12.9% open, p = 0.26), intrabdominal abscess (4.3% lap vs. 4.8% open, p = 0.92), requirement for reoperation (13.0% lap vs. 9.7% open, p = 0.66) or mortality (0 for each group). The open group experienced a 40.3% incidence of wound complications compared to 12.5% in the laparoscopic group (p=0.056). Average length of stay was 4.9 days in the open group and 3.2 days in the laparoscopic group (p = 0.0002). Thirty day readmission rates were 20.1% for the open group and 17.4% for the laparoscopic group (p = 0.72).
Conclusions: Revision of the gastric pouch is technically challenging. Patients undergoing revision of GGF had increased LOS as well as higher reoperation and readmission rates than those reported from COE data and also compared to our own patients undergoing primary RYGB. Comparable short term morbidity and mortality rates to those of open revision suggest that laparoscopic revision of GGF is feasible. LOS for patients undergoing laparoscopic revision of GGF was significantly shorter with a trend toward fewer wound complications. Larger series should clarify whether trends toward better outcomes for laparoscopic revision of GGF are significant.