Management of colon trauma underwent remarkable changes during the twentieth century. Practices that were once based on limited experience and the fear of suture line failure have been replaced with superior methods that have decreased morbidity and mortality. The present management of colon trauma was scientifically established over the past 25 years, although the history of this topic dates back to World War I (WWI). This review highlights the major historical aspects of colon trauma and focuses on the current management of nondestructive and destructive colon wounds, rectal injury, the use of perioperative antibiotics, and colostomy closure.

The studies used to support the current recommendations have been selected using a Medline search of the English language since 1979 and are confined to the civilian literature. The historical background comes from review of a wide variety of sources in the surgical literature. Citations supporting current recommendations are prospective, randomized, or prospective cohort studies in most cases. Retrospective data are used to corroborate the findings of prospective studies. All recommendations herein are in agreement with the Eastern Association for the Surgery of Trauma’s current Trauma Practice Guidelines. [1]

Historical Background

Wallace [2], a British military surgeon, provided the first detailed description of managing colon injury in a review of 1200 cases of gunshot wounds to the abdomen during WWI. He reported 155 isolated colon injuries of which 102 (66%) were managed by primary repair (PR). Mortality in the PR group was 50.0%, and it was 73.5% for those undergoing colostomy. Fraser [3], also a British military surgeon, and his colleague Drummond reported the use of PR for an additional 55 colon injuries during WWI with favorable results and recommended the use of PR except for the most extensive colon injuries. In that report, Fraser gave the first account of retroperitoneal sepsis secondary to colon injury for a condition that now bears his name. Thus the initial experience with managing colon injury favored PR.

Few data specifically addressing colon injury can be found during the period between WWI and WWII. Most civilian literature during this time addressed the basic utility of laparotomy for penetrating abdominal trauma [4]. High mortality rates associated with colon trauma during WWII and the lack of experience of many young military surgeons at the time led the U.S. Surgeon General [5] to mandate the routine use of colostomy in all cases of colon injury. High mortality rates seen with colon injury by the British led to a similar proclamation by Ogilvie [6]. Although a number of reports by Hurt [7], Cutler [8], Hamilton [9], Morgan [10], Mason [11], and Colcock [12] echoed the safety of colostomy for colon injury, some military surgeons continued to use PR in selected instances. Imes [13], Taylor and Thompson [4], and Gordon-Taylor [14] all reported favorable experiences with the use of PR in selected cases of colon injury during WWII. Ogilvie himself alluded to the use of PR “for small holes of the colon without peritoneal soiling” in his classic treatise [6].

The inevitable result of the wartime experience was the routine use of colostomy for colon injuries in civilian practice [8, 13, 14]. Woodhall and Ochsner at Tulane in 1951 were the first to challenge the dictum of colostomy after WWII, noting that civilian low velocity gunshot wounds and stabbings were of an entirely different nature than the high velocity devitalizing wounds seen in military combat. They reported mortality rates of 9% for PR and 40% for colostomy in a review series of 55 patients with penetrating colon injury [15]. Isaacson et al. reported further improvements in the management of colon injury at Tulane in 1961, with mortality rates of 2.05% for PR, 0% for exteriorization of the repair and 17% for colostomy in a review series of 128 patients with penetrating colon injury [16]. In 1967 Axelrod and Hanley, also reporting from Tulane [17], showed mortality rates of 0% for PR and 9.3% for colostomy in a review of 103 patients who had sustained colon trauma.

The pioneering work from New Orleans describing selective management of colon injury stimulated increased use of this approach at other institutions. Tucker and Fey [18], Roof et al. [19], Vannix et al. [20], Wolma and Williford [21], Beall et al. [22] and Bartizal et al. [23] subsequently reported retrospective studies demonstrating favorable outcomes after selective management of colon injuries. The consensus of these reviews advocated primary repair of the colonand ushered in a new wave of prospective randomized studies that form the basis for modern management.

Colon Injury Grading Scales

Two grading scales have been devised to stratify injuries to the colon. Flint et al. first introduced a scale [24] that can be summarized as follows: grade 1: minimal contamination, minimal delay to operation, no associated injuries, and minimal shock; grade 2: through-and-through perforations or lacerations with associated injuries; and grade 3: severe tissue loss, heavy contamination, and deep shock. The American Association for the Surgery of Trauma (AAST) developed the Colon Injury Scale (CIS) [25]: grade I, serosal injury; grade II, single wall injury; grade III, < 25% wall involvement; grade IV, > 25% wall involvement; grade V, circumferential colon wall, vascular injury or both. These scales are used in the subsequent analysis of destructive and nondestructive injuries.

Nondestructive Colon Wounds

A nondestructive colon wound is an injury to the colon that is amenable to primary suture repair with limited amounts of débridement. These wounds include Flint grades 1 and 2 and CIS grades I to III. The first randomized prospective study to evaluate primary repair of such injuries was performed by Stone and Fabian in 1979 [26]. A total of 139 patients were randomized for PR or colostomy if they had an absence of shock, limited associated injury, limited fecal peritoneal soilage or delay to operation of less than 8 hours after injury. Statistically significant lower rates of intraabdominal infection occurred in the PR group (15%) versus those undergoing a colostomy (29%). There was one fecal fistula in the PR group, which healed spontaneously with local wound care. There was no mortality in this study.

Management of colon injury became progressively liberalized following Stone and Fabian’s results by expanding the entry criteria for PR. Chappuis et al. [27], Sasaki et al. [28], and Gonzalez et al. [29] randomized a total of 208 patients to receive PR or colostomy irrespective of shock, contamination, time from injury, or number of associated injuries. Overall complications were similar between groups (17.1% vs. 25.7%), but a significantly decreased intraabdominal abscess rate was observed in the PR group (5.0% vs. 15.6%) (Table 1). There were no deaths attributable to the type of management employed during treatment of the colon injury and no suture line failures. In addition, there were six (5.5%) colostomy-related complications including bowel obstruction (n = 3), stomal prolapse (n = 1), stomal necrosis (n = 1), and peristomal abscess (n = 1).

Table 1. Comparison of primary repair and colostomy for nondestructive colon injury in prospective and prospective randomized trials.
Full size table

George et al. [30], Demetriades et al. [31], and Ivatury et al. [32] prospectively evaluated a total of 282 patients with PR versus colostomy in a nonrandomized fashion. Reasons for stomal formation with nondestructive lesions included delayed presentation (> 24 hours), gross fecal contamination, and damage control procedures. Patients were generally not excluded for shock, blood loss, duration from injury, or number of associated injuries. Overall complication rates were 16.6% for PR and 39.1% for colostomy; the incidence of intraabdominal abscess was 5.5% for PR and 17.2% for colostomy. Three fecal fistulas occurred in the PR group, all of which healed spontaneously. There were no deaths related to the management of colon injuries [30, 31, 32].

Analysis of 20 retrospective studies [33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52] of PR for colon injuries versus colostomy (2516 patients) demonstrates an overall complication rate of 14% for PR and 31% for colostomy (Table 2). Intraabdominal abscesses occurred in 5% of those who underwent PR and 12% of those with a colostomy. The suture line failure rate was 1.6% for PR and 1.3% for colostomy. Mortality was 0.11% for PR and 0.14% for colostomy when the cause of death was directly attributable to the type of repair selected.

Table 2. Comparison of primary repair and colostomy for nondestructive colon injury in retrospective series.
Full size table

Two centers have now reported institutional protocols in which all patients with colon injuries undergo PR regardless of injury type or associated risks [53, 54]. A total of 160 patients, including 33 with destructive colon wounds, have now had PR under these protocols. There was one fecal fistula and one anastomotic breakdown, for a leak rate of 1.3%; 12 intraabdominal abscesses (7.5%); and an overall complication rate of 61%.

When combining all prospective and retrospective studies that compared PR and colostomy for management of nondestructive colon wounds, the suture line failure rate was 1.6% for PR. The incidence of intraabdominal abscesses was 4.9% for PR and 12% for colostomy. The overall complication rate was 14% for PR and 30% for colostomy, with mortality rates of 0.11% and 0.14%, respectively. These findings clearly show the superiority of PR for nondestructive colon wounds without the morbidity associated with a colostomy.

Destructive Colon Wounds

Destructive colon wounds encompass those injuries that require segmental resection due to loss of colonic integrity or segmental devascularization due to mesenteric injury (or both). These wounds typically result from high velocity gunshot wounds or close-range shotgun blasts. Occasionally, blunt injuries from lapbelts can cause devitalizing injuries to segments of the rectosigmoid colon or cecum. These wounds include Flint grade 3 or CIS grades IV and V injuries. Management is less clear because these injuries occur less frequently and therefore less information is available.

Among reported patients with colon trauma, two prospective [31, 32] and three prospective randomized [27, 28, 29] studies identified a total of 65 patients who underwent resection and primary anastomosis (PA) for their colon injuries. The overall complication rate was 35%, and intraabdominal abscesses occurred in 23% of patients undergoing resection and PA. The leak rate was 3.1% with no associated deaths. When combined with 142 patients from 10 retrospective studies [35, 38, 40, 41, 47, 49, 50, 51, 52, 55] that also specifically assessed resection and PA for destructive colon injuries, the overall complication rate was 36%, the intraabdominal abscess rate 19%, the leak rate 7%, and the mortality 1.7%, which was secondary to anastomotic failure.

In the largest single institution experience to date, Murray et al. retrospectively evaluated 140 destructive colon wounds, with 112 (80%) patients undergoing resection and PA and 28 (20%) undergoing colostomy [56]. There were 12 (11%) suture line failures for the resection and PA group (3 colonic fistulas and 9 anastomotic leaks), with two deaths attributable to leaks. Leaks were associated with an Abdominal Trauma Index (ATI) higher than 25 and hypotension in the emergency department. For right-sided injuries, there were fewer anastomotic complications when ileocolostomy was used than with colocolostomy.

In a landmark multicenter study, Demetriades et al. prospectively evaluated 297 patients with destructive colon injuries in which 197 underwent resection plus PA and 100 underwent diversion [57]. Patients were not randomized; rather, management of the colon wound was determined by the surgeon at exploration. Not surprisingly, patients with diversion were significantly more injured and ill than those who underwent resection and PA. Thirteen leaks (6.6%) occurred in the resection plus PA group, and there was one leak from the stump of a Hartmann’s pouch. The four deaths were attributed to colon-related morbidity and abdominal sepsis, all occurring in the diversion group. The results of univariate analysis revealed that severe fecal contamination, transfusion requirement of more than 4 units, and single-agent antibiotics placed patients at greater risk for abdominal complications but not for anastomotic leakage. When controlling for all known risk factors, however, multivariate analysis demonstrated no significant difference in mortality or abdominal complications between diversion and resection plus PA. The authors therefore concluded that destructive colon injuries “should be managed by primary repair regardless of risk factors.”

In summary, overall complications, intraabdominal abscess formation, the leak/colocutaneous fistula rate, and mortality are lowest for PR of nondestructive colon wounds, second lowest for colostomy due to nondestructive colon wounds, followed by resection plus PA of destructive colon wounds, and highest for colostomy for destructive colon wounds (Table 3). Factors that have been associated with anastomotic failure in patients undergoing resection and PA include co-morbid immunocompromising disorders such as diabetes mellitus, aquired immunodeficiency syndrome (AIDS), and cirrhosis and a transfusion requirement of more than six units of blood [57]. Other possible risk factors appear to be shock, significant associated injuries, and delay of operation. Patients with destructive colon injuries and any of these underlying factors traditionally would be considered for colostomy, but the latest data support more widespread use of resection and PA.

Table 3. Comparison of primary repair, diverting colostomy and resection, and primary anastomosis for destructive colon injuries.
Full size table

Rectal Wounds

Since WWII rectal wounds have been managed by the basic principles of proximal fecal diversion, PR of the injury when possible, and presacral drainage (PSD) [5, 6]. Only minor refinements have occurred since that time, which include the addition of distal rectal washout (DRWO) [58] and the avoidance of colostomy when PR is possible [59]. Controversies presently exist regarding the efficacy of DRWO and PSD as well as the safety associated with avoiding colostomy when the injuries are repaired. Close analysis of the available data can help clarify these issues.

After Mandell et al. [58] reported successful results with DRWO in combat casualties in the Vietnam War, four subsequent civilian studies found similarly favorable results with this technique [60, 61, 62, 63]. Only Shannon et al. [63], however, demonstrated any statistically significant benefit from DRWO. In the largest series, reported by Burch et al. [64], and in all subsequent series [59, 65, 66, 67, 68], no statistically significant benefit was achieved when DRWO was added to diversion and PSD when patients were evaluated for septic complications. In light of these conflicting findings, DRWO remains an option until further prospective data become available. No evidence to date suggests any harmful effects when DRWO is used. Patients with injuries similar to the combat casualties for which the technique was originally adopted with extensive rectal wall loss are probably the more appropriate candidates for DRWO.

Presacral drainage has been widely used for rectal trauma to prevent pelvic soft tissue infection. Most of the literature in this area advocates routine use of PSD in all instances of rectal injury [60, 62, 63, 64, 65, 66, 68, 69]. However, several authors have found no difference in the incidence of pelvic sepsis when this technique is employed [59, 61, 67]. This disparity cannot be resolved until a prospective randomized study is undertaken. In lieu of such a study, PSD should be routinely performed in all patients who have a suspected rectal injury that cannot be identified or repaired. However, if all injuries are identified and repaired (i.e., intraperitonealized by exposure through the abdomen), PSD probably adds no benefit [68].

Primary repair of rectal wounds without concomitant fecal diversion has been reported in 21 patients with no related complications [59, 65, 67, 68]. Most of these repairs were done transabdominally at the time of laparotomy. Five were done transanally for low-lying injuries without abdominal exploration [67]. The key to selecting patients who need not undergo diversion depends on the anatomy of the injury and whether the injuries can be repaired satisfactorily. Injuries along the anterior and lateral side walls of the upper two-thirds of the rectum are covered with peritoneum and are considered intraperitoneal. These injuries can be managed essentially the same as colon injuries. The distal one-third of the rectum circumferentially and the upper two-thirds of the rectum posteriorly are not covered with peritoneum and are considered extraperitoneal. Extraperitoneal injuries in the upper and middle third of the rectum can usually be dissected out without significant difficulty and repaired. However, only four such injuries are reported in the literature without concomitant diversion (none of which leaked). Therefore PR of extraperitoneal rectal injuries without fecal diversion remains an option and should be determined on a case by case basis.

Additional Management Techniques

Several other techniques have been employed in attempts to lessen the morbidity of colon injury. Prograde colonic lavage and intracolonic bypass have been used to little avail to decrease suture line leaks after PR. Prograde intraoperative lavage has been evaluated in one prospective randomized study in patients with nondestructive colon injuries [70]. With this technique, a saline infusion line is placed through the amputated appendix, and fecal effluent is drained from the rectum using corrugated tubing that has been placed through the anus.

A total of 172 patients were randomized to receive PR alone (n = 81) or prograde colonic lavage and PR (n = 91). There were four intraabdominal abscesses in the lavage group and two in the PR group. This technique has since been abandoned owing to its lack of efficacy. Results led to the same conclusion for destructive wounds.

Intracolonic bypass is used for destructive colon wounds. A soft latex tube is placed through the proximal end of the resected colon and sewn in place with absorbable sutures. The other end is brought through the distal colon and positioned at the anus. The anastomosis is then completed over the top of the tube. The net result is exclusion of the anastomosis from feces while healing occurs. The tube passes spontaneously within the first postoperative month. Two studies, one randomized prospective [71] and one prospective [72], have evaluated this technique with a total of 20 patients. There were no reported leaks or complications related to the bypass tube. There was one instance of Clostridium difficile colitis in a bypass patient. The number of patients is small, and meaningful conclusions cannot be drawn at this time.

Exteriorization of the repaired segment so the suture line can be directly observed for leakage is another technique that has been utilized in the past. One randomized [70], one prospective [73], and eleven retrospective [35, 36, 38, 40, 41, 41] studies have evaluated this technique in 441 patients (Table 4). Suture line failure occurred in 31% and intraabdominal abscess occurred in 4%. Reasons for suture line failure included leakage, obstruction, and gangrenous changes of the exteriorized segment. Exteriorizing the suture line subjects it to desiccation, luminal compromise, venous outflow obstruction, and tension, leading to excessively high failure rates. This technique currently enjoys little support.

Table 4. Results of exteriorization of primarily repaired colon injuries.
Full size table

Retained bullets or bullet fragments that have passed through the lumen of the colon are associated with an approximately 10% incidence of soft tissue infection [75]. Therefore recovery of these fragments is recommended if possible and if the patient’s condition otherwise permits it.

Associated Risk Factors

A number of risk factors have been associated with increased morbidity and mortality after colon injury: hypotension or shock, interval of injury to operation, amount of fecal contamination, associated organ injury, number of transfusions, co-morbid disease. Although some of these factors may help stratify the overall risk, none has been found to increase the risk of suture line failure when PR is used for nondestructive colon injuries [29, 30, 31, 32, 33, 34, 35, 36, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 55, 76]. Regarding destructive injuries, the risk factors for suture line failure remain controversial, with conflicting reports. The following section reviews these data in greater detail.

Perioperative shock has been shown to increase the overall incidence of postoperative complications and intraabdominal abscesses in one prospective randomized study [29] and four retrospective studies [41, 46, 48, 51] for both destructive and nondestructive colon injuries. Several retrospective studies have also reported increased mortality [35, 51] and multiple organ failure [40] when hypotension was present on admission. There must be adequate resuscitation with crystalloid fluids and blood products, and hemorrhage must be controlled to lessen the impact of shock on the outcome.

Delay of operation for 4 to 6 hours has not been shown to be a significant risk factor for nondestructive or destructive colon injury [27, 28, 29] in prospective randomized trials. In most modern-day trauma systems delays longer than 4 to 6 hours are unlikely, but data addressing this issue are sparse. Nonetheless, delays longer than that interval in the presence of destructive injuries should be considered contraindications to resection and PA due to the potential increased risk of anastomotic failure especially in the face of large amounts of contamination or peritonitis. Patients with delays of more than 6 hours with nondestructive injuries are appropriate for PR if peritonitis has not advanced. Patients who present more than 12 hours after injury with nondestructive wounds andsevere contamination or hemodynamic instability should be assessed on an individual basis for possible diversion.

The amount of fecal contamination has been associated with an increase in septic complications and intraabdominal abscesses in four retrospective studies [30, 42, 44, 51]. However, two prospective studies that specifically assessed the amount of peritoneal contamination at the time of surgery [29, 33] showed no difference in septic complications when mild amounts (localized to the area immediately surrounding the colon wound) and moderate amounts (confined to one abdominal quadrant) of contamination were compared with severe contamination (more than one abdominal quadrant). The reason cited for improved results in these prospective studies was the use of copious irrigation with thorough cleansing of the abdominal cavity. Demetriades et al. reported a higher incidence of intraabdominal abscess with severe fecal contamination in destructive wounds but no increased risk of leakage [57]. Therefore the amount of fecal contamination alone should not influence the choice of the repair or diversion technique for any colon injury.

The number of associated organs injured, the ATI, and the Penetrating Abdominal Trauma Index (PATI) have all been associated with increased risk of both infectious and noninfectious complications [28, 30, 32, 35, 38, 40, 41, 42, 44, 46, 48, 51, 52, 55, 56]. However, like fecal contamination, the number of associated injuries, ATI, or PATI alone has not been predictive of suture line failure for any type of colon injury, destructive or nondestructive. Therefore it appears that these factors need not dictate the type of repair chosen.

Transfusion requirement has not been found to be a risk factor for suture line failure in any study evaluating nondestructive injuries. However six retrospective studies have shown transfusion requirement to increase the risk of septic and overall complication rates for all types of colon injury [30, 41, 42, 44, 48, 51]. Stewart et al., in a review of destructive colon injuries, showed that the need for more than six units of blood did increase the risk of anastomotic leak [55]. Demetriades et al. prospectively showed an increased rate of intraabdominal infection in patients receiving more than four units of blood during the first 24 hours but no increased risk of leak in destructive wounds [57]. Although still controversial, support tends to favor more liberal use of resection and primary repair, even in face of significant transfusion requirements

Perioperative Antibiotics

The use of perioperative antibiotics has been shown to be effective in reducing infectious complicationswith most types of gastrointestinal surgery. However, the literature regarding the appropriate choice has not demonstrated any agent or combination that appears superior for colon injuries. The data available clearly indicate that appropriate choices cover aerobic and anaerobic organisms with broad coverage of gram-negative species. The least expensive, most commonly available agents meeting these criteria are second-generation cephalosporins (e.g., cefoxitin), which have been shown to be effective in a number of studies [77, 78, 79, 80, 81, 82, 83, 84]. Other appropriate choices include a β-lactam penicillin [85, 86, 87, 88] or dual therapy with an aminoglycoside or aztreonam and a specific anaerobic agent such as clindamycin or metronidazole [80, 83, 87, 88, 89, 90, 91, 92, 93, 94]. Use of newer broad- spectrum agents such as imipenem cilastin or “triple” antibiotics is not necessary for coverage of normal colon flora, nor is the use of these agents likely to be cost-effective.

Another pertinent area with respect to antibiotic therapy for colon injury is the duration of treatment. Some controversy appears in the literature regarding this issue. However, review of the data clearly defines 12 to 24 hours of antibiotic coverage to reduce the number of infectious complications compared to more lengthy courses of therapy. Early studies by Rowlands et al. [95] and Griswold et al. [96] advocated extended periods of antibiotic coverage (5 days), but the studies lacked a good design and have subsequently been outdated by better studies. In randomized prospective trials, Oreskovich et al. [97] and Dellinger et al. [98] reported no difference in the infectious complication rate (including intraabdominal abscesses) when antibiotics were given for 12 hours or 5 days. In a prospective randomized blinded trial, Fabian et al. [99] found similar rates of major abdominal infection (MAI) when colon wounds were treated for 24 hours or 5 days with antibiotics. Interestingly, patients with ATIs higher than 25 treated for 5 days had higher rates of MAI (30% vs. 17%) than those treated with 24 hours of antibiotics. Treatment with 24 hours of perioperative antibiotics (typically 24 hours of a second-generation cephalosporin) is now the acceptable level of care in patients with colon trauma.

Colostomy Closure

Most colostomies performed after colon injury are traditionally closed 3 months after the initial operation when the patient’s general medical condition otherwise permits. Morbidity has been reported at 4.9% to 26.3% with essentially no mortality [100, 101, 102, 103, 104, 105, 106, 107, 108, 109]. Berne et al. [110] reported increased complication rates in patients with colostomy closure after colonic injury (55%) than for rectal injuries (12.5%) (p < 0.05). They concluded that such high complication rates further supported use of resection and PA over colostomy and that diversion for extraperitoneal rectal injuries was still a good choice. Areas of present interest are early colostomy closure and whether a barium enema is necessary prior to closure.

Success with early colostomy closure during the same admission has been demonstrated by Velmahos et al. in a randomized prospective trial [100]. Altogether, 19 of their patients underwent colostomy closure during the second week after injury. A barium enema was performed prior to surgery to establish that the injury had healed. Other criteria for inclusion required that patients had to be recovering satisfactorily from their injuries, displaying healing wounds and being sepsis-free during the second week. Otherwise they were excluded from further analysis. The patients undergoing early revision were compared to 20 patients undergoing routine colostomy closure at 3 months. There were no differences in leakage rates (one in each group) or other complications. The early group reportedly was technically easier to revise than the late group owing to less adhesion formation. Early colostomy closure can therefore be considered an option in the management of colon injury after fecal diversion. Colostomy revision at 2 to 6 weeks corresponds to the period of maximum adhesion formation and should be avoided.

Crass et al. [105] and Atweh et al. [107] reported that barium enemas demonstrated no findings that changed the course of therapy in 159 patients undergoing colostomy closure an average of 3 months after their injury. Barium enemas therefore may be considered unnecessary for general evaluation prior to colostomy closure. The exception to this policy would be patients with rectal injuries in the distal segment that could not be identified or repaired. Additionally, patients with unexplained heme-positive stool, obstructive symptoms, or other indication of problems should also have a barium enema or colonoscopy prior to colostomy closure.

Conclusions

The general principles of managing nondestructive colon trauma went full circle during the twentieth century and now clearly support the use of PR in most cases. Exceptions to the rule are patients with destructive colon injuries and associated risk factors. Risk factors for increased rates of complications or death, but not necessarily the leakage rate, have been identified as shock, duration between injury and operation, associated organ injury, transfusion requirement, and co-morbid disease. The mainstays of treating rectal injury remain diversion for extraperitoneal injuries and PSD when the injury cannot be identified or repaired. All patients requiring laparotomy for abdominal trauma should be given perioperative antibiotics. A second-generation cephalosporin for 24 hours has proved adequate. Early colostomy closure, a relatively new endeavor, is a possible option for management. Barium enema is not a mandatory part of the preoperative workup for traditional colostomy take-down procedures.

Areas that require future research include contraindications to resection and PA for destructive colon injuries and the utility of DRWO and PSD for rectal injuries. Delayed colonic anastomosis during a take-back laparotomy after a primary damage control procedure deserves investigation. Avoidance of colostomy in patients with extraperitoneal rectal injuries who undergo PRrequires further investigation. Early colostomy closure, within the first 2 weeks of injury, is an appealing technique but should undergo further evaluation before it can be considered more than an option. The answers to most of these questions require multicenter trials to accrue enough patients for statistical validity.

Résumé.

Cet article offre une revue compréhensive du traumatisme du côlon depuis la première guerre mondiale jusqu’à présent. En utilisant la médecine basée sur la preuve, on a analysé les données des 25 dernières années et on a défini les standards des soins dans ce domaine. Là où les données sont moins convaincantes, les recommandations sont fournies tout comme les suggestions pour la recherche future. Les sujets abordés sont les lésions du côlon «destructives» ou «non-destructives», les lésions du rectum, la préparation colique «sur table», les stents, les facteurs de risque, les antibiotiques en période périopératoire et la fermeture des colostomies.

Resumen.

El presente artículo ofrece una revisión comprensiva del trauma de colon, desde la primera guerra Mundial hasta el presente. El proceso de medicina basada en la evidencia fue utilizado en el análisis de la información correspondiente a los últimos 25 años y para definir estándares de atención. Cuando la información era menos convincente, se optó por hacer recomendaciones y sugerencias sobre futuros campos de investigación. Los tópicos de mayor preeminencia incluyen las lesiones destructivas y no destructivas del colon, las lesiones de recto, el lavado colónico en la mesa de operaciones, los tubos de “bypass” colónico, los factores de riesgo, los antibióticos perioperatorios y el cierre de la colostomía.