A differentiated approach to repeat small-bowel anastomoses in patients with postoperative peritonitis: a prospective cohort study

Background Postoperative peritonitis still remains the cause of a high mortality rate in emergency abdominal surgery. Here we aimed to evaluate the efficacy of different surgical strategies for small-bowel perforations that resulted in postoperative peritonitis. Methods Surgical management results for 140 patients with postoperative peritonitis due to small-bowel perforations, necrosis and anastomotic leakage were comparatively analyzed. Using the APACHE-II and MPI scoring systems, different surgeon attitudes were examined in three patient groups (primary anastomosis, delayed anastomosis, and enterostomy). Results The surgical approach in patient group I ( n = 47, APACHE-II 11.7 ± 1.2, MPI 14.7 ± 1.3) involved the closure of small-bowel perforations or small-bowel resection to place primary anastomosis. The mortality rate was 17%. Patient group II ( n = 48, APACHE-II 16.8 ± 0.7, MPI 19.3 ± 0.3) underwent delayed small-bowel anastomosis during planned relaparotomies. The mortality rate was 18.8%. Because patients in patient group III ( n = 45, APACHE-II 22.3 ± 1.3, MPI 24.6 ± 1.2) were in very critical condition, anastomoses were not placed after bowel resection, and the surgical procedure was completed with enterostomy. The highest mortality rate of 37.8% was documented in this patient group. Conclusion The differentiated surgical approach undertaken herein using delayed small-bowel anastomosis in more serious patients with postoperative peritonitis was able to mitigate the risk of recurrent anastomotic leaks and was not accompanied by a considerable rise in mortality. The mortality for primary repair and delayed primary closure was basically the same (17.0% and 18.8%, p = 0.03); however, delayed anastomosis in the patients with postoperative peritonitis at higher APACHE-II and MPI scores for severity of illness showed 15.1% less complications in the form of anastomotic leaks ( p = 0.04).