AT THE AAST ANNUAL MEETING
WAIKOLOA, HAWAII (FRONTLINE MEDICAL NEWS) – The risk of anastomotic failure among emergency general surgery patients requiring bowel resection and anastomosis stands at 12.5% and is similar between stapled and hand-sewn techniques, results from a multicenter analysis demonstrated.
Surgeons participating in the study, known as Stapled vs. Handsewn: A Prospective Emergency Surgery Study (SHAPES), “appear to be performing hand-sewn techniques in patients who have a higher burden of disease,” Brandon R. Bruns, MD, said at the annual meeting of the American Association for the Surgery of Trauma. “Without adjustment and despite being performed in a more ill population, there was no difference in failure rate between hand-sewn and stapled techniques. After modeling, only being managed with an open abdomen and contamination at initial operation were associated with anastomotic failure.”
For SHAPES, which is the largest study of its kind and was sponsored by the AAST Multi-Institutional Studies Committee, Dr. Bruns and his associates set out to prospectively evaluate anastomotic failure rates for hand-sewn and stapled anastomoses in patients undergoing urgent/emergent operations. A 1999 study by Seattle researchers found that stapled techniques seemed to be associated with anastomotic failure ( J Trauma. 1999;47:500-08 ). Two years later, the same researchers pooled 4-year retrospective data from four trauma centers and concluded that again, hand-sewn techniques appeared to be superior to stapled techniques after traumatic bowel resection and anastomosis ( J Trauma. 2001;51:1054-61 ). Also in 2001, AAST sponsored a multi-institutional study that examined the same question, this time in penetrative colonic injury. Investigators found no difference in complications between the two groups ( J Trauma. 2002;52:117-21 ). They did, however, find a 22.7% overall incidence of colon-related complications.
“With this background we hypothesized that anastomotic failure rate would be high for emergency general surgery patients undergoing bowel resection and anastomosis,” said Dr. Bruns, an acute care surgeon at the R. Adams Cowley Shock Trauma Center at the University of Maryland Medical Center, Baltimore. “We also hypothesized that failure rates would be higher for stapled techniques, compared with hand-sewn.”
The SHAPES researchers prospectively enrolled 595 patients at 15 medical centers in the United States who underwent urgent/emergent bowel resection for emergency general surgery pathology between July 22, 2013, and Dec. 31, 2015. The patients were grouped by hand-sewn vs. stapled anastomoses and demographic and clinical variables were collected. The primary outcome was anastomotic failure. As in other studies, anastomotic failure was evaluated at the anastomosis level. Multivariable logistic regression was done, controlling for age and risk factors for anastomotic failure.
Dr. Bruns reported that the 595 patients had 649 anastomoses. Of these, 61% were stapled and 39% were hand-sewn. The mean age of patients was 62 years, 51% were male, and the overall mortality rate was 7.7%. More than two-thirds of the patients (35%) had more than one indication for operative intervention. The most common single indication for operation was small bowel obstruction, at 23%. The majority of the anastomoses were small bowel to small bowel (72%), while 21% were small bowel to large bowel, and 7% were large bowel to large bowel. There were a total of 81 anastomotic failures, for a rate of 12.5%.
When the researchers compared the hand-sewn and stapled groups, they were similar with respect to sex and age, with higher Charlson comorbidity indices in stapled and a higher body mass index in the hand-sewn group. They also observed a lower hemoglobin, higher INR, higher lactate, lower albumin, and worse renal function in the hand-sewn group, compared with the stapled group. Moreover, hand-sewn anastomotic techniques were performed more frequently in patients that were on vasopressor support. “Despite patients receiving hand-sewn techniques, having worse laboratory values, and more intraoperative vasopressors, the two techniques had equivalent failure rates, at 15.4% for hand-sewn and 10.6% for stapled,” Dr. Bruns said. “Patients with hand-sewn techniques had longer hospital length of stay, longer ICU length of stay, and a significantly higher mortality, compared with those who underwent stapled techniques. Interestingly and different from most other studies on the topic, operating time between the two groups was equivalent.”
On multivariate regression, the presence of contamination at initial bowel resection (OR 1.96) and the patient being managed with an open abdomen (OR 2.53) were independently associated with anastomotic failure, while the type of anastomosis (hand-sewn vs. stapled) was not.
The researchers also conducted a subanalysis of 165 patients managed with an open abdomen who had bowel resection and anastomosis. These patients had higher BMIs, higher lactates, higher INRs, and more negative base deficits, compared with those who were not managed with an open abdomen. “Perhaps not unexpectedly, open abdomen patients were more likely to be on vasopressor agents,” Dr. Bruns said. “They had longer hospital lengths of stay, more ICU days, and an 18.2% overall mortality. Overall there was an almost 22% anastomotic failure rate in patients managed with an open abdomen, compared with an 8.5% rate in patients managed with non-open techniques.” Comparing hand-sewn and stapled techniques, there was no difference in failure rate in patients managed with an open abdomen (25.2% vs. 17.5%, respectively; P = .20).
“An overall mortality rate of 8% and an anastomotic complication rate of 12.5% should emphasize the dire needs for these operations and the need for meticulous operative as well as surgically directed perioperative care in these patients,” the invited discussant, Gregory Jurkovich, MD, professor of surgery at the Davis Medical Center, University of California, said at the meeting. “We as surgeons must pay attention to all aspects of care in these patients.”
Dr. Bruns acknowledged certain limitations of the study, including the fact that it was not a randomized, controlled trial. Also, “surgeon preference did dictate the type of anastomosis that was created, and the patient and surgeon populations were heterogeneous,” he said. “The multivariable model was limited by missing laboratory data, likely given the urgent nature of some of the operative procedures.”
He reported having no financial disclosures.