Which Is the Safer Anastomotic Method for Colon Surgery? – Ten-year Results

Patients and Methods

Patients. From July 2003 to June 2013, 1,324 patients underwent colorectal surgery with primary intestinal reconstruction at the Jikei University Hospital, Tokyo, Japan. A retrospective comparative investigation was carried out to study the influence of the methods of intestinal anastomosis on anastomotic complications using the data from 684 cases, after exclusion of the following cases, some with multiple overlaps: emergency surgery (29 cases), rectal anastomosis (594 cases), use of a circular stapler (627 cases) and creation of a defunctioning stoma (113 cases). The patients were classified into three groups by the type of anastomosis as follows: HS (n=93), FEEA (n=255) and TRI (n=336) (Table I). Methods of intestinal anastomoses were selected by the preference of each surgeon. The diagnosis of anastomotic leakage was defined based on imaging studies and clinical signs, such as fever >38.5°C, leukocytosis, elevated serum C-reactive protein, drainage of intestinal content from the drain or computed tomography findings of abscess formation around the anastomosis. The diagnosis of anastomotic stricture was defined based on imaging studies demonstrating intestinal distension starting from oral side of the anastomotic site. Anastomotic bleeding was defined by melena or endoscopic findings within seven days after surgery.

As patient's factors, we classified tumor locations into two groups: right-sided colon (cecum, ascending colon and right transverse colon) vs. left-sided colon (left transverse colon, descending colon and sigmoid colon).

We investigated the relation between anastomotic complications (anastomotic leakage, stricture and bleeding) and clinical factors, such as age, gender, etiology (benign or malignant), location (right-sided colon or left-sided colon), anastomosis (ileo-colostomy or colo-colostomy), approach (open or laparoscopic), operation time, estimate of blood loss and anastomotic method (HS, FEEA or TRI). This study was approved by the Institutional Review Board (27-2s83 8168).

Surgical techniques. (i) TRI: In case of ileo-colostomy, antimesenteric intestinal wall was incised vertically from the wall edge to compensate for the difference in diameter after colectomy. Since 2008, the incision of the intestinal wall edge in colo-colostomy has been carried out to enlarge the anastomosis diameter. The circumference of the walls of the intestinal tract was divided into three parts and a linear stapler (TA™ 60-3.5 with DST Series™ Technology; Medtronic Inc., Minneapolis, MN, USA or PROXIMATE® Reloadable Linear Stapler (TX) 60; Ethicon Inc., Somerville, NJ, USA) was used three times to anastomose each part. At the posterior one-third of the circumference of the mesentery (posterior wall) side, transmural stay sutures were placed in an inverted manner in five sites (Figure 1a). By pulling up stay sutures, the posterior wall was anastomosed from the mucosal side, including the five stay sutures using a linear cutter (one-third of the circumference) (Figure 1b). Then, corner stay sutures were placed on both ends of the first anastomosis to evert both sides of the anastomosis and, then, another whole-layer stay suture was placed in the center of the residual two-third of the circumference. Two more sutures were similarly added in an everted manner between these sutures (Figure 1c). Subsequently, one-half of the residual two-third of the circumference was anastomosed in two steps with a linear stapler. Finally, the remaining one-third of the circumference was also anastomosed in an everted manner (Figure 1d). The intersection of staple lines was not reinforced by seromuscular sutures.

(ii) FEEA: The anastomotic technique was basically adapted by the method originally described by Steichen (11) with minor modifications advocated by Chassin et al. (12). The sites of division of the colon or ileum were selected and the mesentery was divided. Proximal and distal intestines were separated using a linear stapler (DST Series™ GIA™ Stapler, Endo GIA™ Ultra Universal Short Stapler; Medtronic Inc. and Proximate® Linear Cutter 75 or ECHELON FLEX™ ENDOPATH® Stapler 60mm; Ethicon Inc.) or a knife at the division sites. For the use of a linear stapler, a small hole, large enough for insertion of the fork of the stapling device, was made in the wall at the antimesenteric border of each intestine. The forks were inserted in full length and fired along the antimesenteric border to perform a side-to-side anastomosis. After the stapling instrument was removed, the stump of the intestinal tract was grasped with Allis clamps in an everted manner while avoiding overlapping of the staple line edges of side-to-side anastomosis and closed using another linear stapler. Finally, two or three reinforcing sutures were placed at the distal end of the longitudinal (first) staple line between the intestines.

(iii) HS: An absorbable suture was used for anastomosis using the Albert-Lembert method or the layer-to-layer method.

Statistical analysis. The retrospective analysis was performed to study postoperative complications at the anastomotic site, i.e. anastomotic leakage, stricture and bleeding. Univariate analysis of complications at the anastomotic site was performed using the Student's t-test or the Chi-squared test. In addition, a backward elimination stepwise approach was utilized for multivariate analysis. In either statistical analysis, p<0.05 was considered statistically significant. The statistical analysis was performed using a statistics analysis SPSS version 22 software (IBM Inc., Chicago, IL, USA).