The morbidity and mortality associated with colorectal surgery results partly from anastomotic leakage. Animal experiments have shown some advantages of sutureless anastomosis over conventional colorectal anastomosis. Compression anastomosis follows the same biological principles as sutureless anastomosis.
The compression anastomosis AKA-2 was evaluated in a prospective consecutive audit of 442 patients between September 1989 and August 1998.
Anastomoses were performed in 372 elective and 70 emergency situations. The indication for operation was colorectal cancer (56·3 per cent) and diverticulitis (23·5 per cent). A defunctioning colostomy was performed in 110 patients (24·9 per cent). Fourteen patients died (3·2 per cent). Death was related to anastomotic complications in three patients (0·7 per cent). Twenty-four patients (5·4 per cent) developed intra-abdominal complications. There were 11 symptomatic (2·5 per cent) and six asymptomatic (1·4 per cent) leakages. Anastomoses that were more than 10 cm from the anal verge leaked in seven (2·4 per cent) of 291 cases, while anastomoses between 5 and 10 cm leaked in three (2·6 per cent) of 116 cases and those less than 5 cm from the anal verge leaked in one (3 per cent) of 35 cases.
The low incidence of anastomotic complications demonstrates good biological healing of compression anastomoses. The compression anastomosis AKA-2 is safe in both high and low anterior resection and can therefore be recommended for use in colorectal surgery. © 2000 British Journal of Surgery Society Ltd