Version May 2024
INTRODUCTION — Gastrointestinal injury of the stomach, small bowel, colon, or rectum can be due to blunt or penetrating trauma. Injuries range from minor bruising to complete devascularization for blunt injuries and small perforations to destructive injuries for penetrating mechanisms. Associated solid organ injury is common.
The duodenum is frequently injured in association with the pancreas, and the management of these combined injuries is complex. The diagnosis and management of duodenal injuries is reviewed separately. (See "Management of duodenal trauma in adults" and "Management of pancreatic trauma in adults".)
The diagnosis and management of gastrointestinal injury (except duodenum) due to trauma is reviewed here. The general approach to the abdominal trauma patient is discussed elsewhere. (See "Initial evaluation and management of blunt abdominal trauma in adults" and "Initial evaluation and management of abdominal stab wounds in adults" and "Initial evaluation and management of abdominal gunshot wounds in adults".)
ANATOMY
Stomach — The stomach is located in the left upper quadrant of the abdomen but can occupy other areas of the abdomen depending upon its degree of distention, phase of diaphragmatic excursion, and the positioning of the individual at the time of the injury. Anteriorly, the stomach is adjacent to the left lobe of the liver, diaphragm, colon, and anterior abdominal wall. Posteriorly, the stomach is adjacent to the pancreas, spleen, left kidney and adrenal gland, splenic artery, left diaphragm, transverse mesocolon, and colon (figure 1).
The blood supply of the stomach is from the gastric arteries. The left gastric artery, which is derived from the celiac artery, courses along the lesser curvature of the stomach and anastomoses with the right gastric artery, which is a branch of the common hepatic artery (figure 1). The right and left gastroepiploic arteries arise from the gastroduodenal artery and splenic arteries, respectively, and anastomose along the greater curvature. The short gastric arteries arise from the splenic artery and supply the fundus of the stomach.
Small bowel — The small bowel is anatomically divided into three portions: the duodenum, jejunum, and ileum. The anatomy of the duodenum is discussed elsewhere. (See "Management of duodenal trauma in adults", section on 'Duodenal anatomy'.)
The jejunum begins at the ligament of Treitz, and there are no true anatomic boundaries that delineate the jejunum from ileum. The jejunum tends to have more adipose tissue in the associated mesentery, and its vascular arcades are fewer and longer when compared with the ileum (figure 2). The jejunum and ileum measure between 6 and 7 meters long and are attached to the posterior abdominal wall by the mesentery.
The arterial supply, venous drainage, autonomic nerves, lymph nodes, and a variable amount of adipose tissue are contained between the layers of the small bowel mesentery. The superior mesenteric artery (SMA) provides the blood supply to the jejunum and ileum and originates from the aorta approximately 1 centimeter below the origin of the celiac trunk (figure 3). The superior mesenteric vein lies anterior and to the right of the artery and joins the splenic vein to form the portal vein behind the neck of the pancreas.
Colon and rectum — The colon and rectum occupy retroperitoneal and intraperitoneal spaces, in close approximation to solid organs (figure 4). The ascending and descending colon are retroperitoneal, while the transverse colon, which extends from the hepatic flexure to the splenic flexure, is intraperitoneal. The sigmoid colon continues from the descending colon, ending where the teniae converge to form the rectum. The rectum is the distal continuation of the colon, measuring 12 to 15 cm in length. The rectum lies anterior to the three inferior sacral vertebrae, the coccyx, and sacral vessels and is posterior to the bladder in men and the vagina in women. The sigmoid colon and the anterior upper two-thirds of the rectum are located intraperitoneally, and the remainder of the rectum is extraperitoneal.
The blood supply to the colon is derived from the superior (SMA) and inferior mesenteric arteries (IMA), and internal iliac artery (figure 5). The SMA, ileocolic, right colic, and middle colic arteries supply the cecum, right colon, and first portion of the transverse colon, respectively. The venous outflow generally follows the arterial supply (figure 6). The transverse colon derives its blood supply from both the SMA and IMA. The watershed area between these vessels is at risk for ischemia. The descending and sigmoid colon are supplied by the left colic artery and the superior sigmoid artery, branches of the IMA. The rectum is supplied by the superior, middle rectal, and inferior rectal artery. Venous drainage from the rectum is from the superior rectal and middle rectal veins draining to the inferior mesenteric veins, and the inferior rectal veins draining to the internal pudendal veins.
MECHANISM OF INJURY — Gastrointestinal injury can be due to blunt mechanisms (motor vehicle crash, pedestrian injury, falls) or penetrating mechanisms (knife, gunshot, shotgun). Most blunt gastrointestinal injuries are due to motor vehicle crashes with an increased risk for those wearing a seatbelt and/or in the front seat passenger position. A multi-institutional prospective review of 95 trauma centers by the Eastern Association for the Surgery of Trauma (EAST Hollow Viscus Injury in Blunt Trauma Study) identified 275,557 trauma admissions and found that patients involved in a motor vehicle crash are 1.5 times more likely to sustain a gastrointestinal injury compared with other mechanisms of blunt abdominal trauma [1,2].
Blunt injury — The mechanism of blunt gastrointestinal injury is typically due to crushing of the bowel between the solid structures of the body such as the spine or pelvis and steering wheel, seatbelt, or handle bar (bicycle, motorcycle) [3]. In the EAST Hollow Viscus Injury Study, the incidence of gastrointestinal injury among patients receiving an abdominal workup for trauma was 3.1 percent [2]. In order of frequency, gastrointestinal injury occurred more commonly in the small bowel (jejunum/ileum) followed by colon/rectum, duodenum, stomach, and appendix. The risk for full-thickness injury followed the same pattern. The mechanism of injury to specific organs following blunt trauma is as follows:
●Stomach – Rupture of the stomach is typically uncommon because of its relatively protected anatomic location. The incidence of stomach perforation following blunt trauma ranges from 0.02 to 1.7 percent [4-8]. When rupture does occur, the anterior wall is more commonly affected, and approximately 95 percent of stomach injuries are associated with injuries to other solid and hollow organs [5,8]. The risk of perforation may be higher for patients with preexisting gastric distention, such as following a meal [4]. Increasing severity of abdominal injury is associated with a greater risk for gastric injury. Compared with patients who have small bowel and/or colon injuries, patients with blunt gastric injuries have a higher Injury Severity Score (ISS; 22 versus 17) and a higher percentage of patients with a chest Abbreviated Injury Scores (AIS) >2 (36 versus 12 percent) [8].
●Small bowel – Small bowel injury is associated with injury mechanisms that include front seat passenger position in motor vehicle crashes, falls, and pedestrians struck by automobiles [2]. Small bowel injuries associated with blunt trauma range from contusions, limited to the bowel serosa, full thickness perforations, to severe mesenteric injuries that result in a devascularized bowel segment. These devascularizing or "bucket handle" injuries often have a delayed presentation, if not identified on immediate operation, and can lead to full thickness necrosis and intestinal perforation several days post injury [9]. Penetrating injuries typically present with perforation; however, stab wounds, particularly those with thin blades may produce subtle injuries that are easily missed without careful attention to detail during operative intervention.
●Colon and rectum – Colonic and rectal injuries are relatively uncommon. Blunt injury to the mesentery occurs commonly at transition points where the mobile portions of the colon become fixed retroperitoneally, such as in the region of the ileocecal valve and sigmoid colon. The rectum can be injured in association with pelvic fractures. Colonic injury can be particularly difficult to diagnose due to the retroperitoneal location of much of the colon [10].
Penetrating injury — Any implement or missile that enters the abdomen can injure the bowel. Depending upon the girth of the victim and force applied, even a short impalement that penetrates the abdominal wall can cause gastrointestinal injury, particularly in patients with prior abdominal surgery and bowel adhesions.
The risk of specific organ injury from penetrating trauma depends upon the instrument used (eg, blunt or sharp), site of penetration, and trajectory and velocity of the missile or object. Gunshot or shotgun wounds are more likely to result in multiple injuries to the hollow viscera compared with stab wounds because of their high energy. A blast injury can also result surrounding the tract of a projectile, particularly when fired from a high-velocity weapon. The blast injury can result in devascularization that may evolve over several days to become a perforating wound.
Associated injuries — Injuries associated with lap/seatbelt use, such as flexion-distraction fractures of the spine (Chance fractures) (image 1), cervicothoracic injury [11], and injury to solid organs (eg, kidney, pancreas) should prompt an evaluation for an associated gastrointestinal injury [4-8,12,13]. (See "Spinal column injuries in adults: Types, classification, and mechanisms", section on 'Flexion-distraction (lap belt) injuries'.)
Because of the close anatomic proximity, pelvic and genitourinary injuries are associated with blunt trauma to the distal colon and upper rectum [14]. (See "Blunt genitourinary trauma: Initial evaluation and management" and "Penetrating trauma of the upper and lower genitourinary tract: Initial evaluation and management".)
TRAUMA EVALUATION — We perform the initial resuscitation, diagnostic evaluation, and management of the patient with blunt or penetrating injury based upon protocols from the Advanced Trauma Life Support (ATLS) program, established by the American College of Surgeons Committee on Trauma. The initial resuscitation and evaluation of the patient with blunt or penetrating abdominal or thoracic trauma is discussed elsewhere.
●(See "Initial evaluation and management of blunt abdominal trauma in adults".)
●(See "Initial evaluation and management of abdominal stab wounds in adults".)
●(See "Initial evaluation and management of abdominal gunshot wounds in adults".)
Hemodynamically unstable trauma patients should be transferred immediately to the operating room for evaluation and management. If the clinical setting allows, a Focused Assessment with Sonography for Trauma (FAST) exam may be performed [15,16]. (See "Initial evaluation and management of blunt abdominal trauma in adults", section on 'Ultrasound'.)
Although the FAST examination serves as a screening tool for the detection of hemoperitoneum in blunt abdominal trauma, FAST is not sensitive for detecting gastrointestinal injury unless there is significant fluid volume of spilled enteric contents from a perforated viscus, or blood from a mesenteric laceration [17-19]. For patients with blunt abdominal trauma and gastrointestinal injury, the positive predictive value of FAST is low at approximately 40 percent [2]. Although FAST has little role in the initial evaluation of penetrating abdominal trauma, it may be helpful for decision making in multi-cavitary penetrating trauma. In a prospective study of 75 consecutive patients with penetrating trauma to the abdomen, flank, or back, the overall sensitivity of FAST was 46 percent, the specificity was 94 percent, and the accuracy was 68 percent [20]. Thus, patients with a negative FAST, particularly if there is a high clinical suspicion of gastrointestinal injury, should be further evaluated using additional imaging, such as a computed tomography or abdominal exploration by laparoscopic or open technique. Previously described diagnostic peritoneal lavage (DPL) is no longer routinely performed. (See 'Diagnosis' below.)
In the setting of a negative or equivocal FAST and hypotension, however, a diagnostic peritoneal aspiration (DPA), is an option to determine if there is significant intra-abdominal bleeding. (See "Diagnostic peritoneal lavage (DPL) or aspiration (DPA)".)
History and physical — The trauma history should assess the risk for gastrointestinal injury based upon mechanism. Injury mechanisms associated with gastrointestinal injury are discussed above. (See 'Mechanism of injury' above.)
The trauma patient may complain of abdominal pain, but this symptom is not specific for gastrointestinal injury. Physical examination findings that suggest a possible gastrointestinal injury include abdominal wall ecchymosis, abdominal distention, abdominal tenderness, or signs of peritoneal irritation, but these signs are also not specific for gastrointestinal injury [2]. However, finding no abdominal pain or tenderness significantly decreases the likelihood of finding a gastrointestinal injury that would require surgery. In the Eastern Association for the Surgery of Trauma (EAST) Multi-Institutional Hollow Viscus Injury in Blunt Trauma study, patients with nonperforated small bowel injuries or no small bowel injury were significantly less likely to have abdominal tenderness compared with patients with perforation (53 percent for nonperforated injury, 23 percent no small bowel injury, 72 percent for perforation) [2]. Abdominal distention or peritoneal signs were observed in 4 and 10 percent, respectively, of those who did not have gastrointestinal injury. Thus, the majority of patients who do not demonstrate these signs will not have a gastrointestinal injury.
A finding of gross blood on digital rectal examination is a sign of possible rectal injury, particularly in patients with open pelvic fractures. A sigmoidoscopy, which is typically performed under anesthesia, is needed to definitively evaluate or exclude this injury.
Laboratory studies — There are no specific laboratory tests diagnostic for gastrointestinal injury. An initially elevated white blood cell count in the trauma patient is common and frequently related only to the physical stress of trauma. In one study, there were no significant differences in initial red blood cell and white blood cell counts for patients with perforated small bowel injury, nonperforated small bowel injury, or no small bowel injury [2]. However, an increasing or persistently elevated white blood cell count in the patient with abdominal injury may indicate a developing intra-abdominal process, alerting the clinician to a potential gastrointestinal injury. (See 'Approach to management' below.)
DIAGNOSIS — A definitive diagnosis of traumatic gastrointestinal injury is made at abdominal exploration; however, physical exam and imaging studies, specifically computed tomography (CT) of the abdomen, are important adjuncts that may influence the decision to explore the abdomen.
Plain abdominal films may indicate a traumatic gastrointestinal injury overtly as with a finding of pneumoperitoneum (free air), but the trajectory of a missile (ie, gunshot), or presence of a foreign body (eg, bullet, shrapnel) may strongly suggest that a gastrointestinal injury is likely to have occurred. Plain films are not needed if abdominal CT will be performed; however, trajectory mapping may be useful in operative planning or to help determine the need and extent of further imaging.
CT of the abdomen is the test of choice for identifying specific intra-abdominal injuries in hemodynamically stable patients with blunt injury and is the most sensitive noninvasive imaging test for identifying gastrointestinal injury [2,21]. After blunt trauma, CT of the abdomen has also been shown to be highly specific for ruling out clinically significant injury, particularly in asymptomatic patients [22,23]. The role of CT scanning in hemodynamically stable patients with suspected penetrating injury is less well defined. CT may be useful to differentiate patients for whom abdominal exploration is needed from those with injuries that can be appropriately managed nonoperatively, such as patients with a low/transpelvic trajectory not involving the rectum, a trajectory across the superior abdomen suggestive of an isolated solid organ injury, and possibly those with a lateral trajectory (ie, "love handles" hit) that may have missed the abdomen altogether. For single penetrating wounds, the sensitivity of CT with trajectography was 96 percent in one study; however, it was only 80 percent when there were multiple wounds [24]. (See "Initial evaluation and management of abdominal gunshot wounds in adults" and "Initial evaluation and management of abdominal stab wounds in adults".)
Findings on CT scan that suggest a gastrointestinal injury are listed below [21,25-40]:
●Pneumoperitoneum (free, retroperitoneal)
●Mesenteric air
●Discontinuity of the hollow viscus wall
●Extraluminal enteric contrast
●Free intra-abdominal fluid in the absence of solid organ injury
●Extravasated intravenous contrast
●Bowel wall thickening or edema
●Mesenteric hematoma
Observational studies report widely disparate efficacies of CT scanning in the diagnosis of blunt intestinal injury [2,21,33,34,36,37,41-44]. Some report an accuracy of nearly 100 percent while others have shown that as many as 20 percent of blunt intestinal injuries can be missed by CT. In a systematic review of multidetector CT scan (MDCT) for surgically important bowel and/or mesenteric injury after blunt trauma, sensitivities and specificities ranged from 87 to 95 percent and 48 to 84 percent, respectively [21]. Various investigators have sought to identify which combinations of CT findings may improve the accuracy of CT scanning for blunt injury, but no sign or combination of signs is infallible, and thus, CT scan findings should always be evaluated in the context of the patient's clinical condition.
●In the systematic review, the absence of intraperitoneal free fluid on CT scan all but excluded surgically significant injury, while the presence of intraperitoneal free air, vascular beading, abrupt vessel termination, or the presence of extraluminal contrast were highly specific for injury [21].
●In a retrospective review of 9900 patients with blunt traumatic bowel or mesenteric injuries, the authors noted that the proportion of true positive scans was low in patients with only one positive CT scan finding but increased significantly with multiple radiographic findings consistent with bowel injury [33].
●In the EAST multi-institutional gastrointestinal injury study, gastrointestinal injury was found in 84 percent of patients with free intra-abdominal fluid on CT and no evidence of solid organ injury, and in 92 percent of patients with pneumoperitoneum [2]. The presence of free intra-abdominal fluid, pneumoperitoneum, or small bowel wall thickening was 75 percent sensitive and 79 percent specific for gastrointestinal injury.
●By contrast, a multicenter study of 2299 patients with blunt abdominal trauma identified free intra-abdominal fluid without solid organ injury on CT scan in 90 patients, and of these, only 8 percent had a gastrointestinal injury identified at laparotomy, indicating that free fluid alone as a CT scan finding should not mandate laparotomy [41]. CT scan detected 22 of 25 blunt intestinal injuries in this series [34]. Abdominal tenderness and bruising were present in 61 percent of all patients, but only 22 percent had a positive CT scan. The negative predictive value of a CT scan was 99.6 percent.
●In a prospective single-center study, the seatbelt sign was associated with hollow viscus injury in only 7 percent of patients and, consistent with contemporary evaluations of the utility of CT scan, no patients with a negative CT scan, even in this population with seatbelt sign, had a hollow viscus injury [45].
For penetrating injury, the accuracy of CT for the evaluation of hemodynamically stable patients without obvious indications for immediate exploration is less studied. In a systematic review of seven studies involving 575 patients with anterior abdominal stab wounds, a report of a negative CT was associated with a prevalence approaching 9 percent for injuries requiring therapeutic laparotomy [46]. Almost one half of identified injuries involved the small intestine. Thus, in this patient population, serial physical examination remains central to early diagnosis.
Another approach has been to use scoring systems based on radiographic and/or clinical features to predict the likelihood of gastrointestinal injury [47,48]. While such scoring systems may be useful for evaluating characteristics that influence management systems, they have not been validated prospectively. Our recommendation would be to rely on physical examination to determine operative need in patients without any other clear indication for surgical exploration.
INJURY GRADING — The American Association for the Surgery of Trauma (AAST) Organ Injury Scales are the most widely used injury classification of traumatic injury including gastrointestinal injury (table 1) [49-51]. Although injury management does not correlate exactly with grade, injury scales provide a practical means to describe the severity of injury and can help guide decisions regarding primary repair and/or resection of gastrointestinal injury [49].
APPROACH TO MANAGEMENT — Patients with traumatic gastrointestinal injury that is immediately suspected on clinical presentation should undergo abdominal exploration. Absolute indications for immediate operative intervention include hemodynamic instability, diffuse abdominal pain and/or peritonitis on clinical examination, or findings on initial imaging studies consistent with gastrointestinal perforation (pneumoperitoneum, spilled intraluminal contrast, bowel infarction). In hemodynamically stable patients with abdominal trauma without peritonitis or clear radiographic evidence of gastrointestinal injury, selective nonoperative management is an alternative.
For patients with blunt abdominal injury, observational studies support nonoperative management in the setting of a benign abdominal exam and negative or equivocal computed tomography (CT) scan findings [16,52,53]. Mesenteric hematomas identified on CT scan due to blunt injury that are not associated with active contrast extravasation or signs of bowel compromise can also be observed if there are no other operative injuries identified. Many of these hematomas will resorb spontaneously and will not cause any clinical sequelae. However, it is important to note that patients with mesenteric hematoma can develop delayed peritonitis from bowel ischemia that may take several days to develop. Thus, the traditional "24 hour" observation rule discussed below does not apply to patients with potential devascularization injuries. (See 'Control of mesenteric bleeding' below.)
In addition, there is a growing body of evidence to support selective nonoperative management of penetrating abdominal injury. In a large retrospective review, 792 hemodynamically stable patients with abdominal gunshot wounds (no peritoneal signs, reliable clinical examination) were observed with serial abdominal exams and laboratory studies [54]. Of these, 10 percent developed delayed symptoms requiring laparotomy. The rate of nontherapeutic laparotomy was 14 percent but would have been 47 percent if routine laparotomy had been performed. Complications attributed to delay in operation were observed in 0.3 percent with no increase in mortality. Although the precise duration of observation should be tailored for each individual case based on the clinical course, a minimum observation time of 12 hours for abdominal stab wounds and 24 hours for gunshot wounds in the clinically stable patent has been recommended [55,56].
Selective observation of patients with abdominal injury consists of serial abdominal examination, repeat laboratory studies, and repeat imaging (usually CT scan), as needed. The main component is frequent serial abdominal examinations, ideally conducted hourly by a member of a dedicated team. For this reason, patients who are unable to provide a reliable physical exam, such as those with coexistent severe brain injury, spinal cord injury, altered mental status, and patients undergoing anesthesia (nonabdominal injury), are not candidates for observation. Clinical signs during the period of observation that suggest the presence of a bowel injury include progressively worsening abdominal pain, hemodynamic instability, or intolerance of oral intake. Monitoring the white cell count may contribute to the clinical picture if significantly or persistently elevated. However, no conclusive data regarding the utility of this parameter exist [57]. White blood cell count should be obtained on admission and serial counts obtained every six hours until resolution or the need for surgical exploration reveals itself. When indicated, surgical intervention should be undertaken expeditiously to minimize morbidity and mortality [18,58].
Repeat CT scanning among those who are initially managed conservatively may be associated with an increased sensitivity and negative predictive value for detection of injury requiring surgical repair. In a retrospective review, 34 of 62 patients with suspected blunt gastrointestinal injury who were observed following an initial CT had a repeat CT performed within 48 hours [59]. Compared with initial CT, the early repeat CT diagnostic performance showed an increase in sensitivity (from 63.6 to 91.7 percent) and in negative predictive value (from 77.4 to 94.7 percent). In cases of equivocal physical examination findings, this method of repeat cross-sectional imaging may be of benefit.
SURGERY — When gastrointestinal injury is identified or confirmed at abdominal exploration, the injured bowel can be repaired primarily or resected with an immediate or delayed anastomosis. The choice depends upon the clinical status of the patient, the severity of the specific injury, site of injury, and number and relationship of hollow viscus injuries to each other. The decision also needs to take into account the presence of other traumatic injuries (eg, liver, spleen). At times, a damage control approach may be needed and is accomplished by quickly securing the injured stomach or bowel to stop bleeding and/or gastrointestinal leakage. Definitive repair or resection can be delayed for up to 24 hours [60]. (See "Overview of damage control surgery and resuscitation in patients sustaining severe injury".)
Preparation
●Antibiotic prophylaxis – Prophylactic intravenous antibiotics should be given to all patients who require trauma laparotomy. Antibiotic prophylaxis should be as specific as possible and directed at the site of injury (upper versus lower gastrointestinal tract) [61,62]. However, in patients suspected of having a gastrointestinal injury, the grade and site of injury cannot be known for certain. If upper and lower tract injuries are suspected, or the site and severity are unknown, broad-spectrum coverage is appropriate. Antibiotic choices are given in the table (table 2), and perioperative coverage is usually sufficient. It is important to be familiar with and consult the antibiogram specific to the facility in which the patient is being treated. (See "Antimicrobial prophylaxis for prevention of surgical site infection in adults" and "Antimicrobial prophylaxis for prevention of surgical site infection following gastrointestinal procedures in adults".)
●Thromboprophylaxis – Patients sustaining trauma are at high risk for venous thromboembolism (VTE). Early prophylaxis is recommended for trauma patients to reduce their risk. Therapies for the prevention of VTE are discussed in detail elsewhere. (See "Overview of inpatient management of the adult trauma patient", section on 'Thromboprophylaxis' and "Prevention of venous thromboembolic disease in adult nonorthopedic surgical patients" and "Venous thromboembolism risk and prevention in the severely injured trauma patient", section on 'Thromboprophylaxis'.)
Abdominal exploration — Hemodynamically unstable patients and those with multiple injuries should be explored though a generous midline abdominal incision. Trauma laparotomy should be performed in a standard fashion by packing the four quadrants and evaluating the intra-abdominal organs in a systematic fashion and, when indicated, exploring the retroperitoneum.
Although laparoscopy is a safe and effective diagnostic tool that has become commonplace in general surgery, it is less commonly needed in the setting of trauma. Laparoscopy is best described for evaluating for evaluating diaphragmatic injuries [63,64]; however, it may also play a role in the diagnosis and potentially treatment of stable patients with equivocal signs or symptoms concerning for possible intra-abdominal injury. (See "Recognition and management of diaphragmatic injury in adults".)
The classic recommendation to use laparoscopy to evaluate for peritoneal penetration in penetrating trauma is no longer supported and is not recommended [65-68]. In stable patients with penetrating trauma to the abdomen, the decision for operative exploration is based on stability, imaging, and physical examination [69-71]. (See 'Approach to management' above.)
In accordance with accepted trauma protocols, injuries to the gastrointestinal tract should be evaluated and repaired in a systematic manner. Control of intra-abdominal hemorrhage (spleen, liver, mesentery) should be the first priority to minimize the need for transfusion and minimize fluid requirements [72], followed by control of gastrointestinal contamination [2]. Patients who are hemodynamically stable with limited other injuries can undergo definitive management of their bowel injuries at the time of the initial exploration [73,74]. In the setting of damage control, repair of gastrointestinal injury can be performed at the index operation or with severe trauma, delayed until after hemodynamic stabilization, which is typically within 24 hours of the injury. Repair should not be delayed more than 24 hours because bowel distention can extend the injury. (See "Overview of damage control surgery and resuscitation in patients sustaining severe injury".)
●The anterior and posterior surfaces of the stomach should be inspected for signs of contusion or laceration. The posterior surface can be examined after opening the lesser sac. Ligating a few of the short gastric arteries will facilitate exposure. Small gastric perforations can be identified by injecting air into the nasogastric tube to insufflate the stomach and then filling the abdomen with saline to cover the stomach while observing for air bubbles. Alternatively, methylene blue can be instilled into the stomach via the nasogastric tube and the stomach inspected for leakage.
●The entire bowel and mesentery, beginning from the ligament of Treitz, should be examined. All abnormalities should be thoroughly evaluated and tagged (eg, bowel clamp), but definitive repair should not be undertaken until the entire length of bowel has been examined. (See 'Management of intestinal injuries' below.)
●Evaluation of duodenal injury requires mobilizing the duodenum from its retroperitoneal attachments, and the pancreas, which is commonly injured as well, should also be examined. Duodenal and pancreatic injuries are discussed in detail separately. (See "Management of duodenal trauma in adults" and "Management of pancreatic trauma in adults".)
●All hematomas in the small bowel should be explored if discovered at operation because full thickness injury may be present. If there is evidence of large bowel injury (eg, overlying hematoma), the involved region of the colon should be fully mobilized to allow inspection of the colon circumferentially.
Control of mesenteric bleeding — Mesenteric bleeding and mesenteric hematomas identified intraoperatively can be managed using straightforward techniques. Active mesenteric arterial bleeding can usually be controlled with simple ligation. Due to the rich collateral blood supply to most areas of the small intestine, limited ligation of mesenteric arterial vessels will not result in bowel compromise, but multiple ligations, proximal arterial branch ligation (figure 2), or mesenteric resection may necessitate resection of the associated bowel. Once the injury to the mesentery has been managed, the viability of the bowel should be assessed. (See 'Evaluating bowel viability' below.)
Most mesenteric venous injuries can be ligated without sequelae. Although ligation of the inferior mesenteric vein is well tolerated, the superior mesenteric vein (SMV) should be preserved whenever possible. If ligation of the SMV is necessary, the surgeon should consider temporary abdominal closure and "second-look operation" given the potential for massive bowel edema. (See 'Abdominal closure' below.)
Evaluating bowel viability — The integrity of the bowel should be carefully inspected, particularly in areas of mesenteric injury. Areas of bowel for which the viability is uncertain should be observed throughout the course of the operation. A warm (not hot) laparotomy pad can be placed over the region to promote blood flow. The patient's hemodynamic status should be re-evaluated and hypotension corrected, avoiding vasoconstricting agents if possible.
Bowel viability is generally assessed in the operating room using a combination of clinical experience and judgment, a handheld continuous wave Doppler to assess for audible arterial signals within the mesentery and bowel wall, and intravenous fluorescein administration [75-81]. Within a few minutes of giving fluorescein (1 to 2 ampules, 500 mg each), examination of the bowel with ultraviolet light (ie, Woods lamp) will show perfused regions as bright green while areas of ischemia will appear darker by comparison. Another method to assess viability uses fluorescence angiography with intravenous indocyanine green and a commercially available detection system [82]. This system has become more readily available in many hospital settings.
If warming and improvement in the patient's blood pressure do not improve intestinal perfusion, resection of the affected segment may be required. Areas of bowel that are potentially viable should not be resected immediately; rather, a second-look operation should be planned for no later than 24 hours after the initial procedure.
Management of intestinal injuries — Whether a specific injury should be repaired or resected depends upon the grade of bowel injury (table 1), the presence of other closely associated hollow viscus injuries, the adequacy of perfusion to the segment of bowel, and the patient's overall clinical status. Immediate resection is indicated for overtly infarcted segments of bowel and devascularized bowel segments. Resection is frequently necessary for multiple injuries affecting the same bowel segment. Resected segments of stomach, small bowel, colon, and intraperitoneal rectum can be reapproximated primarily without proximal diversion (ostomy creation) under most circumstances. For patients with severe and multiple injuries, a damage-control approach involves resecting the affected segments of bowel using a gastrointestinal stapler and delaying restoration of gastrointestinal continuity until the patient has stabilized.
Repair techniques — Options for intestinal repair include using a one-layer or two-layer hand-sewn technique or a stapled technique. The choice of technique depends upon the location of the injury, the nature and extent of the patient's other injuries, and surgeon experience and preference.
A stapled technique can be liberally applied to small and large bowel anastomoses when managing traumatic gastrointestinal injuries when there is a minimal size discrepancy between the ends of the bowel. Hand-sewn and stapled techniques have been shown in observational studies to have similar complication rates (eg, anastomotic leak, infection) in the trauma population [83-87]. However, it is important to remember that bowel edema from the traumatic injury can prevent proper closure with a gastrointestinal stapler [86].
The stomach is highly vascular and is usually closed using two layers to provide better hemostasis. In a two-layer closure, the inner layer is a running, absorbable suture and the outer layer is interrupted, permanent (eg, silk) Lembert sutures. An alternative to a hand-sewn closure for a gastric laceration is closure using a gastrointestinal anastomosis (GIA) stapler.
Intestinal anastomoses can be performed using a one-layer or two-layer hand-sewn or stapled technique. Single-layer hand-sewn anastomosis is constructed using a running 3-0 permanent (eg, polypropylene) suture while the two-layer technique uses a running, absorbable suture for the inner layer and interrupted, permanent (eg, silk) Lembert sutures for the outer layer. In a randomized trial of 125 patients undergoing elective surgery, no differences in complication rates or lengths of hospital stay were found for a one-layer versus a two-layer technique for intestinal anastomosis [88].
Repair by injury grade — Low-grade injuries (grade I, II, or III) (table 1) of the stomach, small bowel, colon, and rectum can usually be repaired primarily, whereas higher grade injuries are generally not amenable to primary repair and will require resection. Techniques for intestinal repair are discussed elsewhere. (See 'Repair techniques' above.)
●Grade I – Partial-thickness lacerations should be repaired primarily in one or two layers. Intramural hematomas encountered intraoperatively are typically opened, evacuated, and the bowel wall inspected and repaired. The management of duodenal hematomas is discussed elsewhere, although simple repairs are almost always sufficient [89] (See "Management of duodenal trauma in adults" and "Management of pancreatic trauma in adults".)
●Grade II – The edges of small, full-thickness lacerations should be debrided to fresh margins and repaired as discussed above. Small bowel defects should be closed transversely (aligned with mesenteric/antimesenteric axis) to avoid narrowing the lumen or compromising bowel vascularity.
●Grade III – Many larger lacerations of the small bowel, colon, and rectum can also be repaired with debridement and primary closure. However, lacerations located in certain regions require a more specific approach.
•A gastric laceration near the pylorus should be repaired transversely, such as when performing a pyloroplasty, to maintain a wide gastric outlet.
•An injury at the gastroesophageal junction should be repaired in one or two layers over a nasogastric tube with closed-suction drainage and consideration of a fundoplication to buttress the repair.
●Grade IV and V – Extensive injury or devascularized segments of small or large bowel require resection and anastomosis as appropriate. Such injuries to the stomach and/or duodenum often require complex repair or reconstruction depending upon the location and extent of injury as well as the patient's overall clinical status. (See "Management of duodenal trauma in adults" and "Management of pancreatic trauma in adults".)
Colorectal injuries — Following debridement or resection, observational studies comparing outcomes for colorectal repair with historical controls that favored ostomy creation support primary repair or primary anastomosis for the majority of colorectal injuries [90-95].
For patients with nondestructive colon and intraperitoneal rectal injuries, primary closure is associated with fewer complications [90]. For colon injuries not amenable to primary repair, colon resection with primary anastomosis can be performed in most cases [90-95]. Risk factors that have previously been associated with breakdown of colon repairs have included significant comorbid conditions, >6 units of blood transfusion, shock, delay in operation, and heavy fecal contamination [96,97]. However, a prospective, multicenter trial undertaken by the American Association for Surgery in Trauma (AAST) that included 297 patients found that the surgical method of colon repair did not influence the rate of abdominal complications, regardless of associated risk factors [92].
For extraperitoneal rectal injuries, diverting colostomy alone may be adequate for managing penetrating rectal injury (missile, penetrating bony fragment), which has traditionally been managed with primary repair, presacral drainage, rectal washout, and proximal diversion (colostomy) [93,94]. In one small trial of 48 patients with penetrating rectal injuries, no differences in outcome were found whether or not presacral drainage was used [95].
Abdominal closure — Following trauma surgery, a decision to close the abdomen with or without skin closure depends upon the ability to approximate the fascial edges, the amount of intra-abdominal contamination, the potential for anastomotic breakdown, and the need to perform a second-look operation. In patients undergoing damage control surgery and in those with a planned second-look operation to assess bowel viability, the abdomen should be left open and a temporary abdominal closure used. Leaving the abdomen open may also be more prudent in patients who are at risk for abdominal compartment syndrome. Methods of diagnosing abdominal compartment syndrome and the management of the open abdomen are discussed elsewhere. (See "Abdominal compartment syndrome in adults" and "Management of the open abdomen in adults".)
When the abdominal fascia is closed, the question of whether to close the skin is unresolved. A randomized trial of 48 patients with colon injuries found that leaving the skin wound open resulted in significantly fewer wound infections compared with skin wound closure (17 versus 31 patients) [98]. Thus, with injury to the colon, rectum, or in cases of a large volume of material spilled from the stomach or small bowel, we prefer to loosely approximate the skin edges with staples, which can be easily removed should an infection become apparent.
Role of drains — Data are limited regarding the use of intra-abdominal drains in emergency trauma surgery. Extrapolating from data obtained in elective surgery, we prefer not to use drains following repair or resection of most gastrointestinal injuries. One notable exception is stomach injuries requiring total gastrectomy [99]. Although the systematic review identified no specific data, expert consensus concluded that drains should be used following total gastrectomy with esophageal-jejunal anastomosis because of the high mortality rates associated with anastomotic leak [100]. Thus, in the setting of an emergent total gastrectomy with an esophageal anastomosis, we prophylactically drain the repair site.
In elective surgery, drainage of colorectal anastomoses does not influence postoperative morbidity (eg, surgical site infections) or mortality [93,95,100-105]. A systematic review identified eight trials (two emergency surgery) that compared drainage following colorectal resections (primary anastomosis) with no drainage. No significant differences were found in the incidence of complications [100]. Thus, we prefer not to use drains following the repair or resection of colon or intraperitoneal rectum when managing traumatic injury. [100]
POSTOPERATIVE CARE — Postoperatively, the patient is typically cared for in an intensive care unit unless an isolated bowel injury has been repaired. General issues related to the inpatient care of injured patients are discussed in a separate topic review. (See "Overview of inpatient management of the adult trauma patient".)
●Patients with rectal injury who required a diverting colostomy should have frequent evaluations of the ostomy site to ensure viability. The routine care and management of colostomies is discussed elsewhere. (See "Ileostomy or colostomy care and complications".)
●For patients who have undergone abdominal wall closure, we have a low threshold to monitor intra-abdominal pressure. The diagnosis of abdominal compartment syndrome and indications for abdominal decompression are reviewed elsewhere. (See "Abdominal compartment syndrome in adults", section on 'Measurement of intra-abdominal pressure'.)
●The open abdomen is managed with a temporary abdominal closure. The frequency of dressing changes depends upon the intra-abdominal injuries and type of closure used. An ostomy can interfere with the temporary abdominal closure device, and alternative techniques for placement might need to be considered. (See "Management of the open abdomen in adults" and "Management of the open abdomen in adults", section on 'Ostomy interference'.)
COMPLICATIONS — Following management of gastrointestinal injury, the incidence of complications ranges from 22 to 29 percent [2,92]. Common postoperative complications include systemic complications such as pneumonia, sepsis, and renal dysfunction and thromboembolism, and complications specific to abdominal surgery such as surgical site infection, intra-abdominal abscess, and abdominal sepsis [10,18,106]. A multivariate analysis of 102 patients with traumatic bowel injury (all mechanisms) identified time to surgery and colon injury as independent risk factors for morbidity [107].
In the Eastern Association for the Surgery of Trauma (EAST) multi-institutional gastrointestinal injury study, the incidence of complications in patients suffering blunt abdominal trauma with small bowel injury was significantly higher compared with those who did not have small bowel injury (29 versus 13 percent) [2]. A delay in diagnosis of small bowel injury >24 hours was associated with a significantly higher morbidity rate compared with a diagnosis made within 8 hours (55 versus 8 percent) [2]. The exact time cutoff for increased morbidity (and mortality) is unknown.
Surgical site infection is common in patients with gastrointestinal injury. In one review, surgical site infection occurred in 14 percent of patients [108]. In a prospective study of 178 patients with penetrating abdominal injury, the overall incidence of surgical site infection (SSI; wound infection, fascial dehiscence, or intra-abdominal abscess) was 20 percent and the incidence of enterocutaneous fistula was 3.4 percent [106]. Risk factors for SSI included duodenal and colonic injuries. When patients with colon injuries were excluded, the SSI was 16 percent for patients with gastric injuries and 13 percent for those with small bowel injuries. In contrast to the EAST study [1,2], there was no association between SSI and time to operative management.
MORTALITY — Mortality in patients with gastrointestinal injury is related to overall injury severity (ISS) score, associated solid organ and vascular injuries, and medical comorbidities. Mortality related directly to the gastrointestinal injury is most closely related to a delay in initial diagnosis, intra-abdominal sepsis, and development of multiorgan failure. In a retrospective review of 268 patients with gastrointestinal injury, mortality rates were 28 percent for injuries to the stomach, 19 percent for rectal injuries, and 15 percent for small bowel injuries [8].
Mortality rates for patients sustaining a ruptured stomach have been reported between 28 and 66 percent [1,2,8]. The higher mortality rate for stomach injury appeared to be related to the more severe injury mechanism needed to cause gastric perforation, and multiple other associated organ injuries (eg, spleen, diaphragm, lung), as evidenced by higher mean injury severity scores [2].
A delay in diagnosis ≥24 hours in patients found to have isolated blunt small bowel injury was associated with a significantly higher mortality compared with diagnosis at <24 hours (16 versus 4 percent) [2]. In the Eastern Association for the Surgery of Trauma (EAST) Hollow Viscus Injury in Blunt Trauma Study, patients with small bowel injury were significantly more likely to die compared with patients sustaining blunt abdominal trauma who did not have small bowel injury (19 versus 14 percent) [2].
A multicenter study of patients who survived the first 72 hours after penetrating colon injury found an overall colon-injury-related mortality rate of 1.3 percent [92]. All the patients who died were treated with a proximal diversion at the time of initial procedure, and death was ultimately the result of intra-abdominal sepsis and multiorgan failure. The need for colostomy is a marker of severe injury. (See 'Complications' above.)
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Traumatic abdominal and non-genitourinary retroperitoneal injury" and "Society guideline links: Gastrointestinal perforation".)
SUMMARY AND RECOMMENDATIONS
●Traumatic gastrointestinal injury – Gastrointestinal injury refers to injury of the stomach, small bowel, colon, or rectum due to traumatic mechanisms, which can be blunt or penetrating. Gastrointestinal injury is less common than solid organ injury and occurs more often following penetrating than blunt trauma. The nature and severity of the injury depends upon the injury mechanism, with injuries ranging from minor bruising to complete devascularization. (See 'Introduction' above.)
●Trauma evaluation – The clinical evaluation of the trauma patient should assess the risk for gastrointestinal injury based upon injury mechanism. Although physical examination findings (eg, wall ecchymosis, abdominal distention, abdominal tenderness, or signs of peritoneal irritation) may suggest intra-abdominal injury, these signs are not specific for gastrointestinal injury. (See 'Trauma evaluation' above.)
●Hemodynamically stable patients – For hemodynamically stable patients, in conjunction with physical examination, computed tomography (CT) is the test of choice for identifying specific intra-abdominal injuries and is the most sensitive imaging modality for identifying gastrointestinal injury. (See 'Approach to management' above and 'Diagnosis' above.)
•Findings on CT scan that suggest a gastrointestinal injury and the need for exploration include free air (pneumoperitoneum), mesenteric air, extraluminal enteric contrast, discontinuity of the wall of a hollow viscus, free intra-abdominal fluid in the absence of solid organ injury, bowel wall thickening, active bleeding (extravasated intravenous contrast), evidence of bowel infarct, and mesenteric hematoma. A diagnostic peritoneal aspiration (DPA) may be useful if CT scan is equivocal, but, ultimately, abdominal exploration is required to confirm the diagnosis.
•Observation may be appropriate for selected hemodynamically stable patients in the absence of peritonitis when no other operative injuries are present, imaging studies are equivocal for gastrointestinal injury, and a reliable physical exam is possible. Mesenteric and intramural hematomas can resolve without the need for surgery.
●Hemodynamically unstable patients – Patients with abdominal trauma who are hemodynamically unstable, patients with known gastrointestinal injury, and those with other abdominal injuries that require operative intervention should be taken to the operating room for open abdominal exploration. (See 'Approach to management' above.)
●Damage control approach – A damage control approach to bowel injury may be needed when severe associated injuries are present and is accomplished by quickly securing the injured stomach or bowel to stop bleeding and/or gastrointestinal leakage. Definitive repair or resection can be delayed for up to 24 hours. (See 'Surgery' above.)
●Injury grading and repair – The decision to perform a primary repair or resect injured bowel depends upon the severity and location of the injury, and proximity of the injuries to each other. Grading can help guide surgical management of the injury. We use the American Association for the Surgery of Trauma (AAST) Organ Injury system to grade gastrointestinal injuries (table 1). Grade I, II, and III injuries of the stomach, small bowel, colon, and rectum can usually be repaired primarily; however, multiple injuries along the same segment or higher-grade injuries (grade IV, V) typically need resection. (See 'Injury grading' above and 'Management of intestinal injuries' above.)
●Colon and rectal injuries – For most patients with colon or intraperitoneal rectal injuries, we suggest primary repair or resection and re-anastomosis (as appropriate), but without a diverting colostomy (Grade 2C). For most patients with extraperitoneal rectal injuries, proximal diverting colostomy alone is sufficient. (See 'Colorectal injuries' above.)
●Drainage – For patients undergoing total gastrectomy for trauma, we suggest the placement of an intra-abdominal drain at the esophagojejunal anastomosis rather than no drainage (Grade 2C). For other gastrointestinal injuries, we suggest not placing drains following repair or resection (Grade 2C). (See 'Role of drains' above.)
●Open abdomen – Following exploratory laparotomy for trauma, the abdominal wall may need to remain open because of the risk for abdominal compartment syndrome, or the need for a planned second-look procedure (damage control). (See 'Abdominal closure' above and "Management of the open abdomen in adults".)
آیا می خواهید مدیلیب را به صفحه اصلی خود اضافه کنید؟
