Colorectal Trauma
Thomas J. Stahl, MD, FACS
Assistant Professor of Surgery
Georgetown University Medical Center
Introduction
The incidence of colon and rectal trauma as seen in civilian trauma centers constitutes a substantial problem in management and care. Many of the surgical principles currently applied to this problem emanate from experiences in World War II and the Korean War. Obviously much has changed in patient management, anesthesia, infection prevention, and mechanism of injury since those periods. Yet, the surgical care of colorectal trauma has until recently changed very little. This discussion will review the recent literature on the subject, and allow the reader to utilize contemporary surgical principles as they apply to colorectal trauma.
Colon Injuries
The war experience created treatment regimens for intra-abdominal injuries that were predicated on several unique circumstances. Injuries were usually created by high velocity weapons or explosives, extrication and triage times were lengthy, antibiotics were often not available, and in general resuscitation techniques were primitive by contemporary standards. Thus, the overwhelming priority as it specifically applied to colon and rectal injuries was to control and minimize intra-abdominal contamination. As a result of this priority, the creation of fecal diversion for almost all colon and rectal injuries was deemed the only acceptable course. This dictum has remained in force for the better part of half a century, and is still employed by many experienced trauma surgeons. Only in the last 15 to 20 years has this guiding principle been questioned.
In 1951, Oschner was one of the first surgeons to advocate consideration of primary repair for civilian colon injuries1 . His premise was that injuries repaired in a timely fashion in the presence of minimal to moderate contamination were safe, and obviated the need for a colostomy and the subsequent second surgery required for it's closure. To say that his assertion was greeted with less than widespread enthusiasm is a significant understatement. By and large the issue was not seriously contested until the 1980's, when a number of respected trauma surgeons began repairing colon injuries primarily, with acceptable outcomes. The fundamental issue in primary repair is whether it is as safe (or perhaps safer) than diversion. Accepting that this premise may perhaps be true begs the secondary issue of patient selection; which injuries under which circumstances can be safely repaired without increased risk for complications compared to diversion.
Burch et. al.2 published a large series describing colon trauma patients treated at Ben Taub Hospital in Houston during the years 1979 through 1984. This was a retropsective analysis of 727 patients with intra-abdominal injuries, including but not limited to colon trauma. They treated 52.4% of colon injuries with primary repair. Primary repair included debridement and closure of an injury as well as resection with anastamosis. No segment of colon was excluded from consideration for repair, although left colon injuries were more likely to be dealt with by colostomy than right-sided injuries. In terms of both mortality and morbidity, the group of patients undergoing primary repair fared better. This however was most likely related to the fact that patients undergoing diversion had more significant injuries than those patients receiving primary repair. The specific concern over anastamotic leaks was ameliorated with the finding of only 3 patients (0.9%) in the primary repair group having this complication. Of the analyzed variables, those which were most likely to generate an adverse outcome were age over 40, shock, and severe associated injuries.
A prospective study published in 1989 sought aggressively to employ repair for nearly all colon injuries 3. Of 102 such patients, primary repair was performed in 83 patients, resection with anastamosis in 12, and resection with colstomy in 7, yielding an overall repair rate of 93%. Of this group only 1 suture line failure occurred. Logistic regression analysis of these patients identified transfusion over 4 units, significant contamination, high colon injury severity index, and greater than 2 injuries as risk factors for sepsis.
One of the first inclusive prospective randomized studies was published by Chappius and colleagues in 1991 4. Fifty-six patients were randomly allocated to either primary repair or diversion without regard to other variables and with no exclusion criteria. Morbidity rates were essentially identical, and no suture line failures occurred. This is a remarkable result considering that in the repair group 11 of 28 patients had high grade (Grade IV or V) colon injuries. The authors were not able to identify any specific predictors of adverse outcome, acknowledging however that the small number of patients made the likelihood of such factors emerging unlikely. In another subsequent prospective randomized study published in 1995, the complication rate was actually somewhat higher in the diversion group 5. When the morbidity from colostomy reversal was appropriately added to the diversion group, the aggregate morbidity was significantly greater (43% vs. 19%).
It seems therefore that the ability to safely repair most colon injuries is not in question. The difficulty arises in deciding under which circumstances this would be dangerous compared to diversion. Since no 2 trauma patients are alike, and the variables to consider numerous, attempts have been made to specifically identify those circumstances that would mitigate against primary colon repair. A recent study by Durham et. al.6 from St. Louis University examined numerous injury related variables. In analyzing 130 consecutive patients with a colon injury, stepwise regression analysis was performed on 13 separate variables. Of these only the Colon Injury Scale (CIS), Abdominal Trauma Index (ATI), and gross fecal contamination predicted wound or intra-abdominal complications. Applying these variables to their treated patients led them to conclude that only in the most severely injured patients should a colostomy be routinely employed for colonic injuries.
In many respects however, common sense in the treatment of trauma patients yields similiar conclusions. Unstable patients with significant blood loss, multiple severe injuries, and significant fecal contamination represent high risk patients under the best of circumstances, and any factor that might possibly contribute to subsequent complications - such as an anastamosis - should be avoided. Nevertheless, it is readily apparent that primary repair can therefore be applied to 70 or 80% of colon injuries, based on numerous recent studies.
Certain specific components of colon injury have been examined which deserve mention. The length of time from injury to surgical intervention is one such issue. Although 6 hours has commonly been felt to be that turning point where the risk of infection becomes prohibitive for considering injury repair 7, this has only recently been examined in a definitive way. A study from 1991 specifically examined the issue of timing, and found that primary repair could safely be performed up to 12 hours after injury without any increase in morbidity or mortality 8. The role of colonic lavage as an adjunct to primary repair of colon injuries was examined by a group of British trauma surgeons in 1990, and they found no benefit from such an effort in reducing infectious complications, regardless of the type or severity of colonic injury 9. What to do with the bullet represents another interesting issue. For those patients in whom the bullet traverses the large bowel and lodges within the body, the bullet serves as a contaminated foreign body prone to infection 10. It is reasonable to attempt extrication of the bullet if doing so is relatively simple and safe.
A final consideration in primary repair of colon injuries is what is gained by doing so. The primary gain is obviously the avoidance of a colostomy with it's numerous disadvantages and the requisite need for a second operation. The reversal of a colostomy adds to the total cost of care, total time out of work, patient pain and suffering, and adds an additional source of surgical morbidity. The quantification of these d isadvantages was recently published by Drs. Pachter et. al. 11.
Rectal Injury
Part of the challenge in treating rectal injuries is in their identification. As opposed to intra-abdominal colon injuries where all colon surfaces can be assessed with relative ease, the extra-peritoneal rectum is far more difficult to evaluate. Even utilizing rectal examination, proctoscopy, and intra-operative evaluation of accessible proximal rectum, 15 to 20% of rectal injuries may still be missed 12. Therefore a high index of suspicion must prevail, and any penetrating wound that might have injured the rectum should be presumed to have done so, and treatment should be applied accordingly. As with colonic injuries, treatment protocols in large part emanate from war-time experiences. The components of rectal injury management have historically been diversion, repair if possible, pre-sacral drainage, and rectal irrigation or "wash-out". An examination of these 4 components separately for their individual significance in predicting outcome in rectal trauma found that only diversion and pre-sacral drainage were important 12, with irrigation and repair exerting little effect on mortality or morbidity. Although most surgeons would advocate repairing the rectal injury if it is readily accessible, additional pelvic dissection to identify and repair an extra-peritoneal injury is not justified. Rectal resection in turn is only indicated as a debridement maneuver in massive perineal and pelvic injuries where the bulk of the rectum has been irretrievably destroyed.
To minimize the impact of a colostomy, a study was performed examining the feasibility and safety of same-admission colostomy reversal after rectal trauma 13. Using contrast enemas, they were able to demonstrate healing of rectal wounds in 75% of patients after 10 days, and in 87.5% of these patients the colostomy was in turn closed without incident prior to discharge. The complications were secondary to the stoma closure, with no rectal injury related complications in those patients with a radiographically healed rectal wound. It would therefore seem appropriate in properly selected patients to consider colostomy closure prior to discharge. This certainly minimizes the patient's need to contend with a c olostomy for several months, but also reduces the total hospitalization and recuperative period.
In an attempt to specifically address the role of primary repair in extra-peritoneal rectal injuries, Levine and his colleagues examined this issue in their rectal trauma cohort from St. Louis University 14, specifically in the setting of repair without diversion. Although their series was small, encompassing only 6 patients, they found that in these six patients for whom the injury was repaired without diversion, there were no complications or deaths. Five of the six patients had trans-anal repairs of low lying rectal injuries, and the sixth patient had an easily accessible proximal rectal injury discovered at abdominal exploration that was repaired without diversion. In all cases the injuries were comparatively low grade with minimal tissue destruction. This study however does indicate that in selected patients the creation of a colostomy is not categorically necessary. Further study with greater patient numbers is therefore indicated to address this approach further.
In summary, rectal injuries are best treated with diversion and pre-sacral drainage, with rectal irrigation being neither harmful nor particularly advantageous. Primary repair without diversion may be considered under favorable conditions, but repair in conjunction with diversion and drainage does not seem to confer any additional advantage.
References
1. Woodhall JP, Ochsner A. The management of perforating injuries of the colon and rectum in civilian practice. Surgery 1951;29:305-320.
2. Burch JM, Brock JC, Gevirtzman L, et. al. The injured colon. Ann Surg 1989;203:701-711.
3. George SM, Fabian TC, Voeller GR, et. al. Primary repair of colon wounds: A prospective trial in non-selected patients. Ann Surg 1989;209:728-734.
4. Chappius CW, Frey DJ, Dietzen CD, et. al. Management of penetrating colon injuries: A prospective randomized trial. Ann Surg 1991;213:492-498.
5. Sasaki LS, Allaben RD, Golwala R, et. al. Primary repair of colon injuries: A prospective randomized study. J Trauma 1995;39:895-901.
6. Durham RM, Pruitt C, Moran J, et. al. Civilian colon trauma: Factors that predict success by primary repair. Dis Colon & Rectum 1997;40:685-692.
7. Schrock TR: Trauma to the colon and rectum. In Trauma Management; Vol. I Abdominal Trauma. New York: Thieme-Stratton, 1982:183.
8. Martin RR, Burch JM, Richardson R, et. al. Outcome of delayed operation of penetrating colon injuries. J. Trauma 1991;31:1591-1595.
9. Baker LW, Thomson SR, Chadwick SJD. Colon wound management and prograde colonic lavage in large bowel trauma. Br J Surg 1990;77:872-876.
10. Poret HA, Fabian TC, Croce MA, et. al. Analysis of septic morbidity following gunshot wounds to the colon: The missile is an adjuvant for abscess. J Trauma 1991;31:1088-1095.
11. Pachter HL, Hoballah JJ, Corcoran TA, et. al. The morbidity and financial impact of colostomy closure in trauma patients. J Trauma 1990;30:1510-1513.
12. Burch JM, Feliciano DV, Mattox KL. Colostomy and drainage for civilian rectal injuries: Is that all ? Ann Surg 1989;209:600-611.
13. Renz BM, Feliciano DV, Sherman R. Same admission colostomy closure. A new approach to rectal wounds: A prospective study. Ann Surg 1993;218:279-293.
14. Levine JH, Longo WE, Pruitt C, et. al. Management of selected rectal injuries by primary repair. Am J Surg 1996;172:575-579.
Recommended reading
1. Fabian TC, Croce MA. Management of penetrating colon injuries. Perspective Col Rect Surg 1992;5:24-49.