RESULTS
The study population consisted of 40 patients. Mean age was 52.3 ± 12.3 years. The most common type of malignancy was ovarian cancer (70.0%). 78.5% of the patients were operated for primary debulking purpose. 12.5% (5/40) of the patients suffered an AL. The clinical characteristics of the study population were shown in Table 1.
The most common operation was rectosigmoid resection (67.5%). The mean operation time was 367.6 ± 122.3 minutes. A protective ostomy was opened in 7.5% (3/40) of the patients. Anastomosis was most frequently performed with stapler devices (33/40, 82.5%). In addition to intestinal anastomosis; peritonectomy, splenectomy, appendectomy, and lymph node (pelvic and/or paraaortic) resection were also performed in 57.5%, 10.0%, 50.0%, 67.5% of the patients, respectively (Table 2).
In both groups, PCT values ​​tended to decrease until POD 7. Although the PCT values ​​started to increase from the POD 7 through POD 8-9 in the leak group, they remained stable in the no leak group. The albumin decrease starting from POD 1 to POD 4 was observed in the leak group while it remained nearly stable in the no leak group through the first 10 days postoperatively. The change of mean PCT and albumin values ​​in anastomotic leak and no leak groups within 10 days postoperatively was shown in Figures 1 and 2.
Postoperative 10-day PCT, albumin and platelet values ​​were compared in the group with and without AL. There was a statistically significant difference between the albumin values ​​on POD 3 (p=0.028), on POD 4 (p=0.045) and the platelet values on POD 1 (p<0.001). The mean albumin values on POD 3-4 and the mean platelet values on POD 1 were lower in the AL group. Although it was not statistically significant (p>0.05), median PCT values (ng/mL) on POD 8, 9 and 10 were higher in the AL group compared with no leak group (Table 3). The best cutt-off point for PCT on POD 9 was determined to be 0.11 ng/mL (AUC: 0.917, Sensitivity = %100.0, specifity = %66.7, positive predictive value = %66.7, negative predictive value = %100.0). Postoperative 10-day CRP, leukocyte, neutrophil, leukocyte ratio neutrophil (L/N) and fever values ​​were also compared in the group with and without AL, and no statistically significant difference was found (p> 0.05) (not shown in the table).
In 4 of 5 patients with AL, peritonectomy was added to the debulking procedure. Due to severe peritonitis, 4 patients were re-operated on POD 8-10 and the leakage was classified as major. The other patient was discharged on POD 10 before the diagnosis of AL, and re-hospitalized on POD 15 for a rectovaginal fistula. That patient was followed conservatively up to 2 months untill the fistula closed on its own which provided to classify the leakage as minor. One of the patients with AL died of multiorgan failure on POD 20. That patient was operated for endometrial carcinosarcoma and in the removal of omental cake, transverse colectomy and anastomosis was performed with handsewn technique. Clinical and surgical details of the 5 patients with AL were detailed in Table 4.