Flange Gastroenterostomy Results in Reduction in Delayed Gastric Emptying after Standard Pancreaticoduodenectomy: A Prospective Cohort Study

• Published in: Journal of the American College of Surgeons Vol. 225; no. 4; pp. 498 - 507
• Main Authors: Khan, Adeel S., Williams, Greg, Woolsey, Cheryl, Liu, Jingxia, Fields, Ryan C., Doyle, Majella M.B., Hawkins, William G., Strasberg, Steven M.
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.10.2017
• Subjects: Aged; Female; Gastric Emptying; Gastroenterostomy - methods; Gastroparesis - epidemiology; Gastroparesis - prevention & control; Humans; Incidence; Length of Stay; Male; Middle Aged; Pancreatic Diseases - surgery; Pancreaticoduodenectomy - adverse effects; Postoperative Complications - epidemiology; Postoperative Complications - prevention & control; Prospective Studies; Treatment Outcome; POPF; DGE; MAGS; OSI; PD; SPD; FL-GE; NonFL-GE; ISGPS; ASA
• ISSN: 1072-7515; 1879-1190
• DOI: 10.1016/j.jamcollsurg.2017.06.006

Delayed gastric emptying (DGE) is a common serious problem after pancreaticoduodenectomy (PD). Flange gastrojejunostomy (FL-GE) is a previously described technique that creates an internal flange in a hand-sewn gastroenterostomy. Results of FL-GE on incidence and severity of DGE after PD are presented. Data were extracted from a prospective database of PD. Standard PD with antrectomy were performed with flange gastroenterostomy (FL-GE) or other techniques (NonFL-GE) at a single institution. The International Study Group of Pancreatic Surgery (ISGPS) definition of DGE was used, and DGE severity was graded based on the ISGPS grading system and the Modified Accordion Grading System (MAGS). There were 215 standard PDs performed. Sixty-eight (32%) were FL-GE and 147 (68%) were NonFL-GE. Delayed gastric emptying rates in FL-GE and NonFL-GE were 9% and 23%, respectively (p = 0.012). Differences in severity of DGE were even more prominent: 29% of DGEs in the NonFL-GE group were ISGPS grade C vs 0% in FL-GE. Also, 35% of DGEs in the NonFL-GE group were MAGS 3 vs 0% in FL-GE. Because of some differences in sex and inflammatory complications between groups, a propensity score analysis was performed, creating 57 matched patients in the FL-GE and NonFL-GE groups. The incidence of DGE remained significantly different in the groups (5% in FL-GE vs 18% in NonFL-GE; p = 0.039). In this cohort study, the flange technique was associated with a marked reduction in the incidence of DGE after PD.