Destructive colon injuries requiring resection: Is colostomy ever indicated?

Background: The management of destructive colon injuries requiring resection has shifted from mandatory diverting stoma to liberal use of primary anastomosis. Various risk criteria have been suggested for the selection of patients for primary anastomosis or ostomy. At our center, we have been practicing a policy of liberal primary anastomosis irrespective of risk factors. The purpose of this study was to evaluate the colon-related outcomes in patients managed with this policy.

Methods: This retrospective study included all colon injuries requiring resection. Data collected included patient demographics, injury characteristics, blood transfusions, operative findings, operations performed, complications, and mortality.

Results: A total of 287 colon injuries were identified, 101 of whom required resection, forming the study population. The majority (63.4%) were penetrating injuries. Furthermore, 16.8% were hypotensive on admission, 40.6% had moderate or severe fecal spillage, 35.6% received blood transfusion of >4 U, and 41.6% had Injury Severity Score of >15. At index operation, 88% were managed with primary anastomosis and 12% with colon discontinuity, and one patient had stoma. Damage-control laparotomy (DCL) with temporary abdominal closure was performed in 39.6% of patients. Of these patients with DCL, 67.5% underwent primary anastomosis, 30.0% were left with colon discontinuity, and 2.5% had stoma. Overall, after the definitive management of the colon, including those patients who were initially left in colon discontinuity, only six patients (5.9%) had a stoma. The incidence of anastomotic leaks in patients with primary anastomosis at the index operation was 8.0%, and there was no colon-related mortality. The incidence of colon anastomotic leaks in the 27 patients with DCL and primary anastomosis was 11.1%, and there was no colon-related mortality. Multivariate analysis evaluating possible risk factors identified discontinuity of the colon as independent risk factor for mortality.

Conclusion: Liberal primary anastomosis should be considered in almost all patients with destructive colon injuries requiring resection, irrespective of risk factors.

Level Of Evidence: Therapeutic/Care Management; Level IV.