Objective Anastomotic leak remains the main concern after colorectal anastomosis in ovarian cancer. Our objective was to compare the use of three different management approaches after colorectal resection and anastomosis in patients with ovarian cancer.

Methods Between January 2010 and June 2018, a total of 133 patients with International Federation of Gynecology and Obstetrics (FIGO) stage II–IV ovarian cancer who underwent colorectal resection and anastomosis were included. According to the approach followed after colorectal anastomosis and during the post-operative period, patients were stratified into three groups: conservative management and observation, diverting ileostomy, or ghost ileostomy technique. Univariate analyses were performed for quantitative variables by applying Student’s t test or Mann-Whitney U test and for qualitative variables by using the χ² test (or Fisher’s test according to the sample size).

Results A total of 145 patients underwent colorectal resection during cytoreduction for FIGO stage II–IV ovarian cancer. Twelve patients were excluded because a colostomy was required. Thus, 133 patients were included in the final analysis. Modified posterior pelvic exenteration was performed in 121 (91%) patients and recto-sigmoid resection in 12 (9%) patients with relapse. The approach after anastomosis was wait-and-see in 72 patients (54.1%), diverting ileostomy in 19 patients (14.4%), and ghost ileostomy in 42 patients (31.5%). There were no differences in diagnosis, age, body mass index, ECOG (Eastern Cooperative Oncology Group), histology, tumor grade, FIGO stage, or type of surgery between the groups. No differences were found regarding the anastomosis leak related factors or the rate of anastomotic leak between the three groups (5.6% vs 5.3% vs 4.8%; p=0.98). Two patients died because of the anastomotic leak in the wait-and-see group, and none died in the diverting ileostomy or ghost ileostomy group. In the diverting ileostomy group, a higher number of patients had complications compared with the ghost ileostomy group (78.9% vs 7.1%; p<0.01). Four patients (21.1%) developed dehydration due to high output stoma (>1500 mL) causing electrolyte imbalance in the diverting ileostomy group, and one patient (2.4%) in the ghost ileostomy group (p=0.03). The stoma reversal rate was 73.7% for the diverting ileostomy group and 100% for the ghost ileostomy group.

Conclusions There were no differences found in the rate of anastomotic leak among the three groups of patients. The use of ghost ileostomy avoids the drawbacks of diverting ileostomy and seems to have advantages over routine diverting ileostomy and wait-and-see approaches for ovarian cancer patients undergoing colorectal anastomosis. Rates of stoma reversal are lower after diverting ileostomy when compared with ghost ileostomy.