Role of Intraoperative Esophagogastroenteroscopy in Minimizing Gastrojejunostomy-Related Morbidity: Experience with 2,311 Laparoscopic Gastric Bypasses with Linear Stapler Anastomosis

Ashraf Haddad Nicholas Tapazoglou Kuldeep Singh Andrew Averbach 0 ) Saint Agnes Hospital , Baltimore, MD, USA Background Anastomotic leaks and strictures of the gastrojejunostomy are a cause of major morbidity following laparoscopic Roux-en-Y gastric bypass (LRYGB). Reported rates of leaks vary between 0 and 5.2 %. This has led bariatric surgeons to use a variety of intraoperative methods to detect incompetent suture lines. The aim of the study was to evaluate the role of intraoperative endoscopy in reducing the rate of postoperative anastomotic complications. The setting of this study is in a community teaching hospital. Methods Medical records of 2,311 patients who underwent a LRYGB from 2002 to 2011 were retrospectively reviewed utilizing the hospitals' bariatric surgery database. Demographics, weight, body mass index, intraoperative endoscopy results, and postoperative outcomes within 90 days after surgery were analyzed. Results Endoscopy was attempted in 2,311 patients and completed in 2,308 (99.9 %). Intraoperative leak was detected in 80 (3.5 %) patients; suture line was reinforced in 46 patients (2 %), while in the other 34 patients the leak was transient at only high insufflation pressure. Postoperative clinical leaks were detected in four cases (0.2 %) two of which had initial leaks intraoperatively. In two cases, the anastomosis was too tight and required reconstruction. Twenty-five patients (1.1 %) developed early postoperative strictures requiring endoscopic dilatation within 90 days. Three patients (0.1 %) had iatrogenic injury at the time of intraoperative endoscopy, all three healed without delayed morbidity. Conclusions The routine use of intraoperative endoscopy in LRYGB with the linear stapler anastomosis technique is associated with a complication/failure rate of 0.3 % and low gastrojejunostomy-related morbidity after LRYGB within 90 days (leak rate of 0.2 % and stricture rate of 1.1 %). - Laparoscopic roux-en-y gastric bypass (LRYGB) is a technically challenging procedure, which is performed frequently in the USA and is becoming more popular in other countries [1]. There are several techniques of gastrojejunal anastomosis (GJA) construction with linear stapler and partially handsewn anastomosis being one of them. Anastomotic leak is one of the most serious potentially preventable complications with reported rates of 15 % [2]. Current guidelines of the American Society of Metabolic and Bariatric Surgery (ASMBS) (2009) state that the vast majority of gastrointestinal leaks occur in the absence of technical error and no highquality clinical evidence exists that intraoperative technique is able to eliminate or substantially decrease the incidence of leaks as a complication of gastric bypass. Nevertheless, GJA leaks remain a cause of major morbidity and every effort to decrease it is justified [3, 4]. In this retrospective study, an attempt was made to evaluate the role of the routine use of endoscopy in reducing GJA-related morbidity. Patients and Methods A retrospective review involved analysis of the medical records of 2,311 consecutive patients with LRYGB performed in 2001 to 2011. Patients database was reviewed for sex, weight, body mass index (BMI), postoperative complications, and length of hospital stay. The database included follow-up data from the surgeons offices on 90days morbidity, mortality, and therapeutic interventions. Of the 2,311 patients 1,849 were females (80 %) with an average age for both sexes of 4410.3 years. The average BMI was 49.88.3. Demographic data are summarized in Table 1. All procedures were performed in integrated minimally invasive operating rooms with laparoscopic and endoscopic capabilities by a bariatric surgeon (KDS or AMA) with the assistance of a senior surgical resident and one surgical assistant. Standardized procedure technique involved construction of the proximal gastric pouch, followed by construction of the Roux limb in a retrocolic, retrogastric position and a gastrojejunostomy between the posterior wall of the proximal gastric pouch and the antimesenteric aspect of the Roux limb. Prior to GJA construction, the Roux limb was secured to the proximal gastric pouch with the blind end facing left and checked for axial rotation from within the omental bursa and from the infracolic aspect. The gastrojejunostomy was fashioned in two layers with the outer posterior layer created first with running 2.0 Surgidac Endo Stitch and the inner layer partially stapled with EndoGIA 45 and 2-cm length of the staple line. The inner layer was completed with running 2.0 Surgidac Endo Stitch with purse-stringing of the anastomosis to a diameter of about 15 mm. At completion, it assumed an oval to round shape. Following that, the anterior sero-muscular layer was finished with running 2.0 Surgidac Endo Stitch. Upon completion of the anastomosis, the Roux limb was clamped with a bowel clamp about 5 cm distally. The table was leveled and the left subdiaphragmatic space was filled with sterile normal saline to cover the proximal pouch and anastomosis. The area was irrigated and aspirated repeatedly until the irrigant became clear from blood and debris. Intraoperative endoscopy was performed by the attending surgeon Table 1 Demographic data of patients % of all patients/averageSD (standard deviation) or senior resident. The gastroscope was advanced with the controls in the unlocked position posterior to endotracheal tube at 5:30 or 6:30 oclock position and was introduced under digital control without force across the superior esophageal sphincter. Occasionally a jaw thrust maneuver provided by the anesthesiologist was required to assist in advancement of the instrument. In case of persistent difficulty with insertion of the endoscope, superior laryngeal structures were visualized with insufflation for appropriate guidance of the instrument. Subsequent advancement of the instrument was done under direct visualization. The proximal pouch was examined and then the gastroscope was negotiated across the anastomosis into the Roux limb. The gastroscope was pulled back into the proximal pouch and the anastomosis re-inspected with continuous insufflation. Following that, the jejunum was accessed again and then all compartments were desufflated while withdrawing the gastroscope making sure no substantial amount of air is left. In case of persistent air leak, the gastroscope was left in position and the procedure was repeated after repair or reinforcement of gastrojejunostomy suture line. After completion of endoscopy, the Roux limb was secured in the mesocolic window and Petersons defect was closed with two running 2.0 Surgidac Endo Stitches. A Jackson Pratt drain was routinely placed and positioned posterior to the anastomosis. An upper gastrointestinal (GI) gastrograffin imaging was performed on the first postoperative day. Patients were started on clear liquid diet and advanced to a pureed diet on the second day. The drain was re (...truncated)