Abstract
BACKGROUND: Anastomotic leakage following colorectal resection remains one of the most significant complications with relevant morbidity and mortality. There is evidence that a higher number of stapler firings for rectal division can affect the leak rate in double stapling anastomosis. However, there are no data concerning compression anastomosis. We present our institutional experience addressing this issue. DESIGN: This is a retrospective review of a prospective institutional database of patients undergoing colonic and rectal resection for benign and malignant indications between January 2008 and December 2014 at the surgical department of the St. John of God Hospital, Vienna. Inclusion criteria were rectal division with linear stapling devices and construction of anastomosis to the rectal stump using a circular stapler or compression device. RESULTS: Three hundred eighty two (196 female; 51.3%) patients were included. Mean age was 65.8 years (range: 18-95) Indications for the operation included diverticular disease (44.8%), colorectal carcinoma (51.6%), inflammatory bowel disease (1.8%), and adenoma (1.8%). A laparoscopic approach was employed in 334 cases (87.4%); in 170 patients (44.9%), a compression anastomosis was created. One, two, and three or more stapler cartridges were used for rectal division in 58.4, 33.5, and 8.1%, respectively. Male gender, neoadjuvant therapy, rectal cancer as an underlying disease, laparoscopic surgical approach, and duration of operation longer than 200 min are leading causes for the usage of more than one stapler cartridge. Overall leak rate was 4.7% (18/382). The only factor associated with the occurrence of leakage was the use of three or more stapler cartridges for the closure of the rectal stump (p = 0.002). CONCLUSION: Our data support that multiple stapler firings for rectal division following colorectal resection has a major impact on anastomotic leak rate. Especially in laparoscopic surgery efforts should be made to minimize the number of stapler cartridges used.