Colon & Rectal Trauma
Jose R. Cintron, MD, FACS, FASCRS
Associate Professor and Chief
Division of Colon and Rectal Surgery
College of Medicine at Chicago
The colon ranks second to the small bowel in frequency of injury in penetrating trauma. Injury to the colon is uncommon in blunt trauma averaging only about 2-5% in most series. When other associated abdominal injuries are present the morbidity and mortality is always increased when colonic injuries are present. However, isolated colonic injuries are rarely fatal. Although the management of colonic injuries was quite debatable during the 20th century, there has been growing consensus in its current management.
There are essentially three therapeutic strategies in the surgical management of colonic injuries. Primary repair, colostomy, and lastly, exteriorized repair. Colostomy and exteriorized repairs minimize the risk of leakage at the expense of requiring a second operation. Primary repairs are desirable provided they do not leak. Balancing the risks versus benefits of each type of repair can be controversial.
It was not until WWI that there was sufficient data available looking at the outcomes of patients with colonic injuries. The data that is frequently cited comes from Cuthbert Wallace who documented a mortality of 59% for isolated colonic injuries. This was in large part due to infrequent experience with intestinal surgery in civilian practice, lack of intravenous fluid and blood transfusions, no available antibiotics, a delay in surgical intervention greater than 6 hours after wounding, and avoidance of certain technical maneuvers such as medial rotation of the right and or left colon. Towards the end of WWI three British surgeons (Wallace, Fraser, Gordon-Taylor) concluded that most wounds could be safely closed primarily, that resection should be avoided and that proximal diversion is beneficial for extensive or left sided injuries.
Ogilvie, a senior British Surgeon during WWII advocated colostomy for all colon injuries. Interestingly enough the mortality rate for colonic injuries reported in his series was 53%. In 1943 the Office of the Surgeon General mandated that all colonic injuries be treated by colostomy. Mortality declined from 60% to 20% and many British and American surgeons were convinced that diversion was responsible for this improvement despite multiple series comparing primary repairs and colostomy showing lower mortality rates with primary repair.
During the Korean and
After WWII Ochsner and Woodhall questioned military dogma for the management of civilian injuries and demonstrated that colostomies were not essential for favorable results in many patients. They reported an 8.3% mortality with primary repair vs. 35% with colostomy. Since the publication of this sentinel paper, primary repair has been used with increasing frequency in many major trauma centers. Exteriorized repairs which peaked in the 1970s have since waned and have been largely replaced by primary repair.
Most colonic injuries are secondary to penetrating trauma. Firearms account for approximately 75% of these injuries and stab wounds for 20% of the injuries. Since most assailants are right-handed facing their victims, this explains the slightly higher prevalence of left sided penetrating injuries. The majority of blunt injuries to the colon are secondary to motor vehicle accidents. Patients usually present with either large blow-outs of the colon or with mesenteric avulsions. These blunt injuries are evenly distributed throughout the colon. Transanal trauma is either iatrogenic or secondary to diverse sexual activities. Iatrogenic injuries include colonoscopy, biopsies, polypectomies, coagulation, or barotrauma from overinsufflation and occur in approximately 0.1% of examinations. Sexually related injuries result from the transanal insertion of foreign bodies.
Injuries of the colon are usually readily apparent at the time of laparotomy. Pneumoperitoneum on an upright chest xray suggests hollow viscus injury and gross blood or hemoccult positive test on digital rectal exam also is suggestive of viscus injury. Failure to diagnose a colonic injury is mosty likely to occur in small caliber gunshot wounds or in stiletto type stab wounds. Furthermore, injuries in the region of the splenic flexure or rectosigmoid region further potentiate the possibility of missing an injury. Signs of potential colonic injury include feculent odors, and blood staining of the colonic wall or mesentery. If these are present further exploration is indicated. Proper examination of any area in question requires full mobilization. Hematomas of the colonic wall or mesentery adjacent to the colon mandate exploration and may require division of one or two vasa recta in order to properly expose the injury. Gently milking intraluminal contents to the presumed area of injury can be performed as a final diagnostic maneuver. In patients with a penetrating wound of the flank and no presumed intraperitoneal injury, a gastrograffin enema or triple-contrast CT may assist in diagnosis.
Generally speaking there are three therapeutic options in the management of colonic injuries. Primary repair involves either the suturing of a perforation or injury in nondestructive colon wounds or resection and anastomosis in destructive colon wounds. Colostomy usually involves resection and or repair and diversion. Exteriorized repair includes suture repair with exteriorization of the repair or resection and anastomosis with exteriorization of the anastomosis. Exteriorized repairs are not so desirable when injuries are located in the cecum, ascending colon or descending colon. Their popularity has waned significantly since the 1980’s.
One of the first studies to address the safety of primary repair was the randomized, prospective study of Stone and Fabian. Patients were randomized to primary repair vs. colostomy. However, 48% of the patients were excluded from randomization and underwent colostomy as the investigators thought they were at too high a risk for suture line failure. The exclusionary criteria used by Stone and Fabian included: shock BP<80/60; blood loss>1000cc; .2 organ systems injured; significant contamination; delay. Other authors such as Shannon and Moore utilized the Abdominal Trauma Index, a Colonic Injury Severity Score and hemodynamic status in order to stratify the treatment for their patients. They performed primary repair in 49% of their patients.
The organ injury scaling committee of the American Association for the Surgery of Trauma developed a grading system for colonic injuries. This grading system is similar to the colonic injury severity score used by Shanon and Moore, but the primary purpose of this grading system is to enable comparison of results between institutions and for the coding of medical records.
I Hematoma (contusion w/o devascularization)
Laceration (partial thickness no perforation)
II Laceration <50% circumference
III Laceration >50% circumference
IV Laceration = transection of colon
V Laceration = transection with segmental tissue loss
In 1991 Burch et al. designed a study to identify independent risk factors predictive of adverse outcomes in patients with colonic injuries regardless of the method of treatment. The independent risk factors in Burch’s study predictive of an adverse outcome included: blood loss; solid abdominal organ injury; fecal contamination; mechanism of injury; and age. None of these factors can be specifically predictive of suture line failure.
George and colleagues reported their results with primary repair and were able to perform primary repair in 93% of their patients. They advocated that primary repair should be the standard approach in the non military setting regardless of risk factors. However, four years later they identified a subset of patients with nearly a 42% incidence of anastomotic failure. These were patients that were undergoing damage control laparotomies, had massive transfusion requirements, or serious associated medical conditions
There have been at least four comparable randomized prospective studies comparing primary repair with colostomy. The results of these studies are similar to prior retrospective and prospective studies indicating an increase in septic complications with the use of a colostomy. What is more important is that these studies clearly demonstrate the safety of primary repair in comparison to colostomy.
PRIMARY REPAIR VERSUS COLOSTOMY
Adapted from: Maxwell RA, Fabian TC. Current management of colon trauma. World J Surg 2003;27(6):632-9
Although never attributable to the colonic injury itself, the most common cause of death in patients with colonic injuries is due to exsanguination. On the other hand, the second most common cause of mortality in patients with colonic injuries is sepsis and multiple system organ failure (MSOF) which may very well be attributable to the colonic injury. Nonetheless, as can be seen in the table below, the mortality from sepsis and MSOF is below 5% in more recent series. One can also see from the table that mortality is consistently lower in those series in which primary repair was performed. It is unclear whether patient selection bias is attributable for these differences. Intraabdominal abscess is the most frequent infectious complication and occurs anywhere from 5-15%. Fistulas occur approximately 1-2% of the time in primary repairs and less often in diversions. Stoma complications (necrosis, obstruction, peristomal abscess, etc.) occur in up to 5% of patients. Obstruction is the most common stomal complication of exteriorized repairs.
|Author||Year||#pts||% 10 repair||%abcess||%fistula||%stoma comp||%MOF|
MOF = Multiple organ failure NA= not available
Adapted from: Maxwell RA, Fabian TC. Current management of colon trauma. World J Surg 2003;27(6):632-9
After Stone and Fabian’s work published in 1979, the management of colon injuries has become progressively more liberalized as seen in the table below. Chappius, Sasaki, and Gonzalez randomized a total of 208 patients between primary repair 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 there was a significantly decreased intraabdominal abscess rate observed in the primary group (5% vs. 15.6%). George, Demetriades, and Ivatury prospectively evaluated 282 patients comparing primary repair vs. colostomy. Their exclusion criteria included delayed presentation (>24 hr), gross fecal contamination, and need for damage control. Their overall complications were 16.6% for primary repair and 39.1% for colostomy. The incidence of intraabdominal abscess was 5.5% for primary repair and 17.2% for colostomy.
OC=overall complications, PR=primary repair, DC=diverting colostomy, IAA=intraabdominal abscess. Adapted from: Maxwell RA, Fabian TC. Current management of colon trauma. World J Surg 2003;27(6):632-9
Maxwell and Fabian analyzed 20 retrospective studies comparing primary repair versus colostomy totaling 2516 patients. There was an overall complication rate of 14% for primary repair, and 31% for colostomy. Intraabdominal abscess occurred in 5% of primary repair versus 12% with colostomy. Suture line failure rate was 1.6% for primary repair and 1.3% for colostomy. Mortality was less than 1% for both treatment groups. These studies further support the superiority of primary repair for nondestructive colon injuries.
Reproduced from: Maxwell RA, Fabian TC. Current Management of Colon Trauma. World J. Surg. 2003;27:632-39
Nelson and Singer performed a Cochrane review which was published in 2003. They were seeking to determine whether fecal diversion was necessary in penetrating colonic injuries. They performed an extensive search of three medical databases for randomized prospective studies comparing primary repair against fecal diversion in penetrating colonic injuries. The outcomes considered are listed below. 5 studies met the criteria for this review.
Mortality between groups not significant [OR] 1.7, 95% [CI] 0.51-5.66
Total complications (OR 0.28, CI 0.18-0.42)
Total infectious complications (OR 0.41, CI 0.27-0.63)
Intraabdominal infections (OR 0.59, CI 0.31-0.86)
Abd infections ex dehisence (OR 0.52 CI 0.31-0.86)
Wound comp.+ dehisence (OR 0.55, CI 0.34-0.89)
Wound comp.- dehisence (OR 0.43, CI 0.25-0.76)
Penetrating abdominal trauma index as well as mortality did not differ significantly between groups. However, total complications, total infectious complications, total intraabdominal infections, abdominal infections excluding dehisence, and wound complications icluding and excluding dehisence all had odds ratios which significantly favored primary repair. The authors of this Cochrane Review concluded that primary repair of penetrating colon injuries is as safe as fecal diversion and is associated with a lower complication rate.
Colon wounds that are amenable to primary repair are otherwise called nondestructive wounds. This would encompass CIS grade I & II. Destructive colon injuries require segmental resection because of devascularization of the colon or significant loss of tissue integrity. These types of injuries usually are secondary to high velocity missiles or close range shotgun blasts. Additionally, CIS grade III-V injuries as well as multiple close perforations of the colon warrant segmental resection for management. Similar types of injuries can also occur in blunt trauma from a motor vehicle accident secondary to the lapbelts. The rectosigmoid region and the cecum are the most frequent locations that become devitalized during blunt trauma due to lapbelts.
Two prospective and three prospective randomized studies looking at destructive colon injuries identified 65 patients that underwent resection and primary anastomosis. The overall complication rate was 35%, the intraabdominal abscess rate was 23% and the leak rate was 3.1% with no mortality. If combined with 142 patients from 10 retrospective studies the overall complication rate was 36%, intraabdominal abscess rate 19%, leak rate 7% and mortality of 1.7% secondary to anastomotic failure.
The study published by Murray et al in the Journal of Trauma in 1999 is the largest single institution experience published to date on destructive colon injuries. In this retrospective study of 140 destructive colon injuries, 80% or 112 underwent resection and primary anastomosis and 20% underwent resection and colostomy. There was an 11% leak rate which appeared to be associated with patients who had an Abdominal Trauma Index greater than 25 and the presence of hypotension in the trauma bay. There were fewer anastomotic complications in right sided injuries requiring ileocolostomy in comparison to colocolostomies.
In 2001 Demetriades and colleagues published a landmark prospective non-randomized multicenter study on 297 patients with destructive colon injuries. Of these 197 underwent resection and primary anastomosis and 100 underwent resection and diversion. Thirteen leaks or 6.6% occurred in the resection and primary anastomosis group. Univariate analysis showed that severe fecal contamination, transfusion>4 units, and single-agent antibiotics placed the patient at increased risk of intraabdominal complications but not for suture line failure. They concluded that primary anastomosis should be considered in all patients with destructive injuries regardless of risk factors.
Although many contemporary trauma surgeons would agree that colocolostomy is as safe as ileocolostomy there is still controversy among surgeons. The following table shows a number of studies that address this issue.
Adapted from: Burch JM: Injury to the Colon and Rectum, in Moore, Ernest E.; Feliciano, David V.; Mattox, Kenneth L (eds): Trauma, 5th ed., McGraw-Hill Professional, 2003
|1991||Burch et al||36 (1)||14 (1)|
|1994||Stewart et al||16 (3)||24 (3)|
|1999||Murray et al||56 (2)||56 (7)|
|2001||Demetriades et al||96 (4)||101 (9)|
|2002||Miller et al||18 (0)||22 (3)|
|Total||222 (10)=4.5%||217 (25)=11.5%|
The overall suture line failure rate in those patients who underwent ileocolostomy is 4.5%. On the other hand, patients who underwent colocolostomy (anastomosis distal to the middle colic artery) had a suture line failure rate of 11.5%. Although these studies are not strictly comparable, there is a trend towards better results with ileocolostmy vs colocolostmy suggesting that location of the injury may have an influence on suture line failure. Nonetheless, colocolostomy can be performed with reasonably good results and the data continue to support more widespread use of resection and primary anastomosis for stable patients without significant associated injuries.
Although some similarities exist regarding colonic injuries and rectal injuries there are differences which are unique to the rectum. Anatomically, two thirds of the rectum is located in an extraperitoneal position. In conjunction with being surrounded by the bony confines of the pelvis, the extraperitoneal rectum does not lend itself to easy exposure and or reliable repair. On the other hand, it does provide some accessibility via the anus.
Approximately 80% of rectal injuries are attributable to firearms and less than 3% are secondary to stab or impalement etiologies. Less than 10% of rectal injuries are blunt in nature as a result of falls, motor vehicle accidents or pelvic fractures. The remaining rectal injuries are secondary to transanal trauma from autoeroticism, sexual assault, self induced, and iatrogenic from enema tips and thermometers.
First and foremost in the diagnosis of rectal injuries is to have a high index of suspicion especially under specific circumstances. Patients with gunshot wounds to the trunk, buttocks, perineum or upper thigh should raise suspicion for rectal injuries. Additionally, stab or impalement injuries of the buttocks, perineum, or lower abdomen as well as any history of anal manipulation and lower abdominal pain should also raise suspicion of rectal injury. Digital rectal examination should be performed noting sphincter tone, sphincter defects, mucosal and or rectal wall lacerations, the presence of any foreign body, and gross blood on the examining finger. This should be followed by rigid proctoscopy looking for gross blood, lacerations, or hematomas of the rectal wall.
Diagnostic xrays are useful for tracking missile trajectories, looking for foreign bodies and assessing for pelvic fractures. Water soluble contrast enemas may be useful when conflicting information is present. An attempt should be made to grade the injury followed by treatment of the rectal injury.
Although there is some controversy regarding the colostomy type, the drainage method, the need for distal washout, or the need to repair the rectal wound, most trauma surgeons would agree on the need for diversion and drainage in the management of extraperitoneal rectal injuries. Colostomies should be placed as close to the injury as technically possible. As mentioned, the drainage of extraperitoneal rectal injuries is well established and is performed by making a curvilinear incision posterior to the anus, entering the presacral space, and then placing either passive drains or closed suction drains close to the site of injury. Most extraperitoneal rectal injuries do not require closure. Intraperitoneal rectal injuries do not require drainage.
1. Adkins RB,Jr,
2. Burch JM, Brock JC and Gevirtzman L, et al. The injured colon. Ann Surg 1986;203(6):701-11.
3. Burch JM, Feliciano DV and Mattox KL. Colostomy and drainage for civilian rectal injuries: is that all? Ann Surg 1989;209(5):600,10; discussion 610-1.
4. Burch JM, Martin RR, Richardson RJ, Muldowny DS, Mattox KL and Jordan GL,Jr. Evolution of the treatment of the injured colon in the 1980s. Arch Surg 1991;126(8):979,83; discussion 983-4.
5. Chappuis CW, Frey DJ, Dietzen CD, Panetta TP, Buechter KJ and Cohn I,Jr. Management of penetrating colon injuries. A prospective randomized trial. Ann Surg 1991;213(5):492,7; discussion 497-8.
6. Cook A, Levine BA, Rusing T, Sirinek KR and Gaskill HV,3rd. Traditional treatment of colon injuries. An effective method. Arch Surg 1984;119(5):591-4.
7. Demetriades D, Murray JA and Chan L, et al. Penetrating colon injuries requiring resection: diversion or primary anastomosis? An AAST prospective multicenter study. J Trauma 2001;50(5):765-75.
8. Falcone RE, Wanamaker SR, Santanello SA and Carey LC. Colorectal trauma: primary repair or anastomosis with intracolonic bypass vs. ostomy. Dis
9. George SM,Jr, Fabian TC, Voeller GR, Kudsk KA, Mangiante EC and Britt LG. Primary repair of colon wounds. A prospective trial in nonselected patients. Ann Surg 1989;209(6):728,33; 733-4.
10. Gonzalez RP, Merlotti GJ and Holevar MR. Colostomy in penetrating colon injury: is it necessary? J Trauma 1996;41(2):271-5.
11. Ivatury RR, Licata J, Gunduz Y, Rao P and Stahl WM. Management options in penetrating rectal injuries. Am Surg 1991;57(1):50-5.
12. Kudsk KA, Hanna MK. Management of complex perineal injuries. World J Surg 2003;27(8):895-900.
13. Levison MA, Thomas DD, Wiencek RG and Wilson RF. Management of the injured colon: evolving practice at an urban trauma center. J Trauma 1990;30(3):247,51; discussion 251-3.