Key Concept: “I thought I would get a quick CT scan to rule out other injuries” is a commonly heard refrain at M&M conferences when explaining why a patient with operative abdominal injuries died in the CT scanner.

The sharp decline in civilian penetrating trauma volume and the increase in nonoperative management of most injuries have inadvertently created a widespread disease pathology among surgeons and surgical trainees that we have previously described as catatonia [26]. Defined as “the inability to make definitive management decisions without the use of detailed computed tomography imaging, coupled with a fear of the exploratory operation,” this is a frequent cause of unnecessary delay to definitive operative intervention and a cause of preventable morbidity or mortality. An oft-heard excuse when explaining a major delay in operative intervention for a perforated colon is that the surgeon wanted “more information” about the injury or that the patient “needed resuscitation” before going to the operating room. In this chapter we emphasize the “golden rule” of the acute abdomen; any patient with an acute abdomen or peritonitis, or obvious signs of a bowel injury on initial exam or plain films, belongs in the operating room immediately. No prolonged imaging workup is required, and resuscitation can be done just as well in the operating room as it can anywhere else in the hospital. In the stable patient this approaches may save you time and resources; in the unstable or deteriorating patient with a colorectal injury, these approaches may save his or her life.

Once the decision has been made for operative intervention for a known or suspected colorectal injury, all of the basic principles of emergency surgery and ATLS apply. Although attention should be paid to hypothermia prevention and management with active warming devices (heated forced-air blanket), one exception to common practice that we believe should be omitted is the warming of the operating room to excessive ambient temperatures (>75 °F). This has been shown to be wholly ineffective at treating or preventing hypothermia and only results in discomfort of the operative team and likely additional contamination of the operative field due to sweat [27]. All patients should be expected to have a full stomach, so aspiration precautions during rapid sequence intubation are critical. Bleeding and ongoing blood loss should be anticipated, so emergency-release blood products (type O for packed cells and type AB for plasma) should be standing by and a type and cross should be performed as soon as possible. A Foley catheter should be placed barring any signs of urethral injury, although if there is a suspicion for bladder injury, then we recommend prepping in the genitals and placing a sterile catheter that can be accessed by the surgeon. This can be invaluable for manually distending the bladder with fluid and/or dye to identify injuries and to test for leaks after a repair.

It is important to optimize the operative exposure and visualization, and a few minutes to set up a self-retaining retractor (Bookwalter, Omni, etc.) will pay dividends in decreased overall operative times and frustration levels. This is particularly important for difficult exposure situations such as obesity or for pelvic dissections (sigmoid, rectum). This will also free the hands of the surgeon and assistant to perform more critical tasks. For the unstable or actively bleeding patient, a generous midline incision should be made at the start to allow rapid access to all quadrants of the abdominal cavity. For the stable patient, a smaller laparotomy incision may be made and then extended if necessary based on the injuries identified. Most right-sided colon injuries can be managed through a supra-umbilical midline incision, but left-sided or rectal injuries require inferior extension below the umbilicus toward the pubic symphysis. Descending or sigmoid colon injuries that require resection with anastomosis typically require the longest incisions to adequately mobilize the left colon (including the splenic flexure) and then perform an anastomosis in the lower abdomen or pelvis. If a large amount of hemoperitoneum is discovered on entering the abdominal cavity (even in a “stable” patient), then abandon any attempt at “minilaparotomy” and widely extend the incision from the xiphoid to several centimeters above the pubic symphysis.

One of the most important factors that led to a revolution in the performance of colorectal surgery was the ability to decrease the risk of postoperative infectious complications. The high bacterial content of the colon has long been recognized as an area of concern with any intraoperative spillage, and the concept of increasing bacterial concentrations in the distal colon was a principal factor in promulgating the belief that left-sided injuries should be treated differently than right-sided injuries. For elective surgery, the standard of care has traditionally been to administer an oral mechanical bowel prep (with or without oral antibiotics) before surgery to both remove fecal material and to decrease the bacterial load in the event of intraoperative spillage [28]. In cases of emergent trauma with colorectal injury, the obvious lack of bowel prep before injury has been cited as one reason to treat these differently than elective operations and to forego primary repair or anastomosis in favor of diversion. Over the past decade, a large body of high-quality evidence has amassed that refutes the standard beliefs about the value of mechanical bowel preparation, with multiple randomized trials and several meta-analyses finding no benefit or even evidence of increased morbidity with bowel prep [29, 30].

The currently available data has led many surgeons to abandon or to more selectively utilize bowel prep for elective surgery, but it also provides reassurance for trauma surgeons that a preoperative bowel prep is not a requirement for a safe bowel repair or anastomosis. The two primary issues that a classic bowel prep was designed to address are (1) mechanical removal of fecal matter and (2) administration of antibiotic prophylaxis against infections due to fecal flora. For the vast majority of traumatic colorectal injuries, mechanical clearance of the colon is unnecessary and in fact would only serve to delay completion of the procedure. The second goal of infection prophylaxis is achieved with the administration of intravenous antibiotics that cover both aerobic and anaerobic organisms typical of fecal flora. Antibiotics should be administered as soon as there is evidence of the injury or a decision for laparotomy has been made. Although the optimal goal would be 30–60 min prior to the skin incision, this is often not realistic in trauma surgery. Re-dosing of antibiotics should be performed if the surgery is prolonged to near the normal dosing interval for the selected antibiotic or in cases where blood loss and transfusion are approaching one whole blood volume. Antibiotics should be continued for 24 h only, and there is no benefit (and potential harm) of continuing them longer even in the face of large volume contamination [31]. Appropriate single-agent therapy is as effective as multidrug regimens, and most centers use a B-lactam agent such as Unasyn or Zosyn. In class 4 cases with a delayed presentation and established infection, antibiotics should be continued as dictated by the location and nature of the infection.

Finally, an additional well-described option that can be utilized is the “on-table” bowel preparation or lavage [32]. This concept was initially espoused for all emergent colon surgery in order to achieve the same benefits as a standard preoperative bowel prep. However, it has fallen out of favor as the role of mechanical bowel preparation has been questioned and due to the difficulty of adequately lavaging the colon effectively and without spillage. The pros of an on-table lavage include the ability to debulk and clean the colon to facilitate easier repair or anastomosis, decompress a dilated colon, and remove a large stool burden that could cause future constipation or impaction. The cons include the lack of a proven benefit, the possibility of increased spillage and infections, and the need for making additional holes in the colon or small bowel to place the lavage catheter. We recommend consideration of on-table lavage only in situations of major colonic distension with a large amount of solid or liquid matter or significant solid fecal impaction that could impair recovery of motility or anastomotic healing. Some technical tips/tricks include as follows: (1) lavage the entire colon by opening the appendix and inserting a balloon-tipped catheter for instilling the fluid, then perform a standard appendectomy after catheter removal, and (2) insert a sterile ventilator tubing into the open proximal end of the colon to be lavaged and secure it with a large suture or umbilical tape (Fig. 34.2).

Also remember to use warmed fluids for the lavage in order to avoid causing or exacerbating systemic hypothermia.

Key Concept: Most traumatic colorectal injuries will present within the first few hours and a short delay should not impact management decisions. Longer delays with uncontrolled abdominal contamination should push you toward diversion and/or damage control techniques.

Key Concept: Delayed diagnosis of a contained injury with minimal associated symptoms should prompt consideration of nonoperative management – if the patient has already passed this “trial of life,” then CLOSE observation or percutaneous drainage may be all they need.

One of the most important factors in determining the optimal approach to a colorectal injury is the duration of time that has elapsed between the initial injury, recognition of the injury, and intervention for the injury. Fortunately, for the vast majority of traumatic injuries to the colon or rectum, patients are immediately transported to a trauma hospital and the diagnosis is established on the initial evaluation and imaging. It also appears that minor delays in diagnosis or intervention (2–8 h) do not have much impact on postoperative patient outcomes, and this would typically not change any intraoperative decision making. This also applies to patients that present in a delayed fashion due to a partial-thickness injury that then progresses to full-thickness injury with perforation. As long as the change in status is quickly identified and intervention performed in a timely manner, there does not appear to be major added morbidity or mortality.

However, occasionally there may be a significant delay to either diagnose or intervene in a patient with a major colorectal injury, and this can result in severe morbidity or even mortality. Scenarios where this may occur include failure to recognize peritoneal signs or imaging findings at initial presentation, presence of factors that compromise the abdominal exam (such as severe head injury or intoxication), patient delay in seeking medical attention or in transfer to an appropriate facility, and masking of peritoneal signs by medications (i.e., steroids) or other patient factors. Delays of more than 8–12 h associated with fecal contamination will typically significantly alter both the local anatomy and the patient physiology, and as a result, the intraoperative decision making and choice of procedure will frequently have to be adapted based on the findings.

As a general principal, there should be a much more liberal use of proximal diversion rather than primary anastomosis or repair when there has been a long delay to operation in the setting of fecal contamination and peritonitis. Some of the key considerations and technical factors that may alter your approach in these cases include:

In select cases the patient may present with delayed recognition of a colorectal injury and minimal or localized associated signs or symptoms. In this case the patient has already demonstrated tolerance or control of the injury, and temporizing minimally invasive interventions may be all that are required. The classic non-trauma example of this would be perforated diverticulitis or appendicitis with a localized abscess that can be managed with antibiotics and a percutaneous drain. This is infrequently encountered in the trauma setting but may be seen more frequently with iatrogenic colorectal injury from endoscopy, biopsy, or other interventions. In these cases the optimal management strategy is usually to fully characterize the pathology with a high-quality CT scan, cover the patient with intravenous antibiotics, initiate bowel rest and NPO status, and ensure drainage of significant abscesses or fluid collections. This most often can be accomplished with image-guided percutaneous or transrectal drainage but may require laparoscopy or even laparotomy for difficult locations. Percutaneous drain placement may also be considered in the patient with a large amount of free air who is otherwise a good candidate for nonoperative management. Aspiration of the free air with or without placement of a drain can markedly improve the abdominal discomfort and provide an improved baseline for serial physical examinations.

Key Concept: Injury location used to drive management decisions for colon trauma, but this distinction has largely been abandoned and management should be based on the exact colon injury, associated injuries, and patient factors.

Key Concept: Remember to consider ALL of the risks for repair versus diversion, including the morbidity associated with a second surgery for ostomy takedown. This risk analysis should most often favor primary repair or anastomosis, but a diverting or protective ostomy still has a clear role in high-risk injuries or patients.

Once a full exploration has been completed and all bowel injuries identified, the first decision is whether the injury is amenable to primary repair or will require segmental or formal anatomic resection. The standard teaching is to categorize injuries as either destructive (>50 % of bowel circumference or devascularized) or nondestructive, with resection recommended for destructive and primary repair for nondestructive wounds. Although this is a good basic guideline, there are several other important factors to consider and these are outlined in Table 34.2. These include not only the size of the injury but the number and location of injuries as well as the adequacy of primary repair that may be achieved. An important technical point for missile wounds, particularly fragment or high-velocity gunshot wounds, is that the wound edges need to be debrided back to healthy tissue before closure. These missiles will frequently cause extensive tissue damage or even direct thermal injury (Fig. 34.3) and will often break down within several days of closure if not adequately debrided. Thus, a small wound that initially appears appropriate for primary repair may actually require segmental resection once the full extent of tissue injury is identified and debrided. On the other hand, excessive debridement of surrounding tissue for high-velocity injuries has been advocated based on erroneous assumptions about the size and impact of the “temporary cavity” created by the missile. In addition to these overestimates of the size and force of cavitation, it is important to note that elastic tissues such as bowel tolerate stretch forces much better than inelastic tissue such as the liver [33]. Debridement of all clearly devitalized and any questionable tissue should be performed, but do not perform additional debridement of healthy tissues simply based on the injury mechanism or ballistics. The status of the mesentery is also an important factor as it relates to the adequacy of the blood supply to the injured area. A classically described injury seen in blunt trauma with rapid deceleration is a large tear of the mesentery without injury to the colon wall. This “bucket-handle” deformity can be misleading as the bowel will usually appear uninjured but should be resected due to the large area of devascularization (Fig. 34.4).

As discussed above in the historical overview of colorectal trauma, a dogma developed over time of categorizing the wounds into “right”- or “left”-sided injuries and basing the management off of this simple distinction. Right-sided injuries could undergo standard repair or resection with anastomosis, but left-sided injuries required exteriorization or proximal diversion. This was based on microbiological data showing increasing bacterial loads in the left colon and the observation of increased complications with left sided or more distal colon injuries. However, subsequent experience with both traumatic colon resections as well as emergent resection for diverticular disease demonstrated that left-sided repair or resection with primary anastomoses could be performed with an acceptably low risk of anastomotic leak or other major morbidity.

A major potential confounder of studies examining the question of right- versus left-sided colon injuries is the unequal distribution and severity of associated injuries. The right colon generally lies outside of the pelvis and is not intimately associated with any major vessels or organs except for the right kidney. Alternatively, the transverse colon lies over the body of the pancreas and the aorta/vena cava, while the sigmoid colon lies in the pelvis adjacent to the iliac vessels, bladder, ureter, sacral veins, and pelvic bones. Thus, left-sided colon injuries are much more likely to have associated vascular, pancreatic, and genitourinary injuries or a major pelvic fracture. The presence of these associated injuries will naturally increase the risk of blood loss, need for transfusion, duration of surgery, and systemic inflammatory response. As a result, morbidity and mortality is higher among this patient cohort, but most of this increase is likely independent of the colon injury. Some other key technical factors also play a likely role, including the difficulty of working in the narrower pelvis (left sided) versus the mid to upper abdomen (right sided) and the higher likelihood of anastomotic tension or ischemia with a colorectal anastomosis versus an ileocolic anastomosis.

Although still occasionally taught, the trauma community has largely abandoned the management of colon injuries according to the simplified right-left dichotomy. However, there is now a commonly held misunderstanding that the injury location is not important and that all colon injuries can be treated in a similar fashion regardless of location. Although the left-right distinction is no longer of primary importance, it is critical for the surgeon to appreciate the key issues and technical factors that will be encountered in different injury locations. Table 34.3 shows some of these important considerations according to the exact area of injury from the cecum to the rectum. Each of these individual colon and rectal injury types and locations has an associated “textbook” answer for what operation should be performed, but these do not take into account the wide variations in disease presentation, presence of multiple bowel injuries, presence of associated injuries, and the patient physiology at the time of operation. There are multiple standard and nonstandard surgical options that can be used alone or in combination, and it is critical to tailor your approach primarily to the patient and not only to the disease. We recommend a surgical approach based on the following general categorizations of the patient and the injuries present:

For low-risk patients, a standard primary repair or segmental resection should be performed, and continuity restored without the need for a colostomy or ileostomy. For moderate-risk patients, the majority can also safely undergo primary repair or resection with anastomosis and will not benefit from fecal diversion. In addition, placement of an ostomy will also subject them to the risks of a second surgery for ostomy takedown or the well-described possibility that they will never have their ostomy reversed [34]. A recent randomized trial in perforated diverticulitis demonstrated that only 58 % of end colostomies were eventually reversed [35]. For patients in the high-risk group, the decision for ostomy versus anastomosis can usually be deferred in favor of a damage control approach (see following section “Damage control laparotomy”). However, whether the decision is made initially or at a subsequent laparotomy, this is the patient population where a diverting ostomy should still be strongly considered. Although there is no level 1 evidence to support these decisions, a temporary diverting ostomy to allow both healing of the colon repair and recovery of the patient can be lifesaving. In addition, with more distal pelvic anastomoses (colorectal, ileorectal), the margin of error and tolerance to a leak is significantly decreased and could result in the need for a permanent ostomy. There are multiple intraoperative decisions and considerations that need to be made quickly and decisively, and that can have significant impact on both short- and long-term outcomes. Table 34.4 provides a partial list of these key decisions with associated factors and technical advice that should be considered.

If fecal diversion is felt to be indicated, then two additional (and in our opinion underutilized) options are available to the surgeon as a “compromise” solution between no diversion and standard ostomy placement. The first is to proceed with placement of a diverting ostomy with a plan for early reversal within 1–2 weeks of the initial surgery. This method allows for initial healing of the colon repair or anastomosis, and then a contrast study should be performed to demonstrate adequate healing without leak or stricture. This also allows time for the natural sorting-out process whereby patients will either progress appropriately following their trauma or develop additional complications or physiologic decline and thus eliminate them from consideration of an early ostomy reversal. A prospective randomized trial of this approach in patients with colon injuries demonstrated that early colostomy closure was safe and resulted in shorter operative times and less blood loss than delayed closure [36]. The second compromise option is to perform the definitive repair or anastomosis and then create a separate more proximal protective diverting colostomy or ileostomy. This concept has long been recognized in colorectal surgery for protection of low pelvic anastomoses or ileal pouch procedures and is also applicable to protecting the high-risk anastomosis in the trauma setting. The choice of the type and location of the ostomy should be mainly based on (1) completeness of fecal diversion, (2) ease of placement, and (3) ease of reversal. Other factors that should be considered include the body habitus and location of abdominal wall wounds and incisions that could alter the viable options available for siting the ostomy. Multiple studies have demonstrated that loop ostomies are clearly superior to end stomas in ease of placement and reversal and also provide adequate fecal diversion [36–38]. A recent prospective randomized trial in patients with feculent or purulent perforated diverticulitis demonstrated the superiority of this approach (primary colorectal anastomosis with protective loop ileostomy) versus a standard Hartmann’s procedure [35]. The protective loop ileostomy approach was proven safe and was associated with significantly reduced major complications, hospital stay, and costs. In addition, the choice of ileostomy versus colostomy appears to be equivalent in terms of the adequacy of diversion [38]. As the ileum is typically more mobile and of smaller caliber than the colon, a loop ileostomy is the technically easier option and provides more options for external siting. However, the surgeon must be cognizant that the risk of fluid and electrolyte problems due to high stoma output will be higher with ileostomy versus colostomy, and this can be particularly difficult to manage in elderly or debilitated patients. Our recommendation is that if a temporary protective ostomy is indicated, then a loop ileostomy (or end-loop ileostomy as shown in Fig. 34.5) is the best option and can easily be reversed without the need for laparotomy in most patients. An excellent alternative option, particularly in the patient who may need longer-term or even permanent diversion, is a loop transverse or sigmoid colostomy. Prior to ostomy reversal, all patients should undergo contrast imaging to document anastomotic healing and should have documentation of adequate anal sphincter function for fecal continence.