Primary repair of 58 consecutive penetrating injuries of the colon: should colostomy be abandoned?

Although primary repair of penetrating colon injuries in patients with low injury severity is now widely accepted, several "risk factors" continue to be viewed as relative contraindications to this method of management. The purpose of this study was to evaluate the septic complications and leak rate in a series of consecutive penetrating colon injuries managed exclusively with primary repair. The records of 58 consecutive patients with penetrating intra-abdominal colon injuries managed at an urban Level I trauma center from July 1991 to December 1995 were reviewed. Patients were stratified for injury severity using the Penetrating Abdominal Trauma Index (PATI), and the presence of "risk factors" and septic abdominal and wound complications were recorded. All 58 patients were managed with primary repair. There were 48 gunshot wounds (72%), 7 shotgun wounds (12%), and 9 stab wounds (16%) with a mean PATI of 26.7 +/- 15.2 standard deviation. Seven patients (12.1%) developed intra-abdominal abscess, and all were managed by CT-guided percutaneous drainage. Five patients (8.6%) developed bacteremia, and eight patients (13.8%) developed fascial dehiscence. Three patients (5.2%) underwent abdominal re-exploration in the postoperative period, but none of these was related to failure of the colonic repair. There were no clinically apparent leaks or fistulae and no deaths. The presence of "risk factors" appeared to identify more severely injured patients as indicated by a higher mean PATI score and a higher incidence of intra-abdominal abscess, when compared to patients in whom the "risk factor" was absent. Primary repair can safely be used for virtually all penetrating colon injuries, as clinical leaks are rare, even in patients with "risk factors". Intra-abdominal abscess and other septic complications are dependent on the overall severity of the intra-abdominal injuries and probably result from contamination occurring at the time of injury rather than from postoperative leak from the primary repair.