A review was undertaken of 360 patients undergoing elective left-sided colonic or rectal resections with primary anastomosis, under the care of one surgeon, over a nineteen year period. The incidence, aetiology and management of anastomotic leaks and strictures was studied and the role of proximal diverting colostomy considered. Perioperative mortality was 2.7%. The incidence of anastomotic leaks was 24.4%. Leaks were more common when anastomoses were low, were sutured or were constructed by trainees. Strictures developed in 5.8%. Local recurrence of tumour was the cause of 25% of these strictures. Anastomotic leakage was the principal cause of benign strictures; those developing in association with leaks were more likely to require surgical intervention. There was no evidence that delay in colostomy closure contributed to the development of benign anastomotic strictures. It was not possible to determine whether the presence of a colostomy affected the incidence of leaks but the local effects of such leaks were mitigated in patients with colostomies. Where a minor leak had occurred it was not necessary to wait for complete anastomotic healing before closing the colostomy. After major leaks, colostomy closure before complete healing was associated with further anastomotic problems in 16.0% of cases.

In a randomized prospective clinical trial and layer and two layer techniques have been compared in 92 patients undergoing colorectal anastomosis. The results were assessed radiologically by barium studies on the tenth postoperative day. There was no significant difference in the incidence of anastomotic breakdown with either technique when the anastomosis was performed above the pelvic peritoneal reflection. When the anastomosis was situated below the pelvic peritoneum the incidence of dehiscence was significantly greater when a two layer technique was used.

One hundred patients undergoing elective left-sided colonic or rectal resections were randomly allocated to have an anastomosis performed either with the EEA stapling gun or by hand suture using a single layer of interrupted sutures. In six patients the anastomosis could not be performed with the stapling gun. Clinical leakage occurred in two of the remaining 94 cases; both had hand-sutured anastomoses. Radiological leakage was demonstrated in 13 further cases (7 stapled, 6 sutured); there was no statistical difference in leakage rate with stapled and hand-sutured anastomoses.

No abstract

Background:We describe an afferent loop obstruction caused by an adhesion band in a case of distal gastrectomy with Roux-en-Y end-to-side jejunal anastomosis for cancer.Methods:An initial clinical presentation of acute pancreatitis was ruled out by a computed tomography scan, which revealed intestinal obstruction; it was then confirmed on laparoscopy. Definitive treatment was laparoscopic adhesiolysis. A complete review of the literature concerning afferent loop obstructions is presented.Results:The treatment was successful, with minimal postoperative pain, and the 5-day hospital stay was uncomplicated. The patient remains asymptomatic at 1-year follow-up.Conclusions:The authors advocate minimally invasive surgery as a complete diagnostic and therapeutic alternative to emergency laparotomy in cases where afferent loop syndrome is suspected, and acknowledge that prompt surgery has a higher rate of success and reduces operative morbidity and mortality.