Penetrating Injuries to the Abdomen - II
(Editors' note: The review on penetrating abdominal trauma is being posted in two sections. Sections 1.- 4.1.3. were posted February 2009. The remainder of the Review and a pdf of the entire review is being posted in March 2009. The associated online discussion and CME questions will continue throughout this period)
In the literature, penetrating injury to the pancreas is associated with a mortality
rate of 3-32% .  Because of its location the pancreas is rarely injured in isolation. Therefore, the majority of its associated mortality is due to hemorrhage or other associated injuries, while another half of the deaths occur late in the hospitalization due to sepsis or multiple organ failure.,  It has been estimated that, of the late deaths, approximately one third can be directly related to the original pancreatic injury.  The morbidity and mortality related to injuries of the pancreas is almost solely due to pancreatic duct injury. Ductal injuries must be recognized and treated appropriately.
The pancreas is intimately associated with the 1st through 3rd portions of the duodenum and injures often occur to both structures simultaneously (see 5.2. Duodenum).
Pancreatic injuries must always be considered in any penetrating trauma of the upper abdomen or thoraco-abdominal area. Patients can be unstable due to hemorrhage from the gland itself or surrounding vascular structures. Ductal disruption is associated with abdominal pain, tenderness and peritonitis and is present approximately 15% of the time.  The serum amylase is unreliable as a marker of pancreatic injury and its initial value should not be used diagnostically to determine the presence or absence of pancreatic injury. In stable patients and in centers with the resources, ductal disruption can be demonstrated by endoscopic retrograde cholangiopancreatography (ERCP). More recently, in trauma centers, the magnetic resonance cholaniopancreatography (MRCP) has also been utilized to visualize pancreatic injuries in stable patients with persistent amylase elevation or CT scan findings that are unclear.  However, in penetrating trauma, the verification of a pancreatic duct or parenchymal injury is most commonly and appropriately made at the time of exploratory laparotomy.
The pancreas is exposed mainly through the Kocher maneuver (see 5.2. Duodenum) and entry into the lesser sac. The mobilization of the 1st and 2nd part of the duodenum with a Kocher maneuver will allow palpation and visualization of the body, head and uncinate process of the pancreas. The incision of the gastrocolic ligament (best entered on the left side where it is less vascular) will allow entry into the lesser sac and utilizing a large retractor to hold the stomach cephalad the entire anterior surface of the pancreas can be visualized. The superior surface and the splenic artery in this location can also be inspected. Careful division of the retroperitoneal attachments of the pancreas along the inferior border will expose the posterior surface of the pancreas to define any injury.
Further visualization of the pancreas can be achieved by mobilizing the spleen and the tail of the pancreas toward the midline. Penetrating injuries to the head or anywhere along the body of the pancreas must be considered to involve the pancreatic duct. Direct visualization of the duct in the transected or lacerated pancreas can be noted but more commonly, due to the small size of a normal pancreatic duct, an efflux of pancreatic fluid is the main sign of a ductal injury. Depending on the time from injury to laparotomy, there may be noticeable surrounding fat necrosis secondary to the activation and enzymatic digestion of fat by pancreatic lipase as is seen in pancreatitis. If a duct injury is suspected based on the location of the wound, but there are no obvious signs of pancreatic fluid leak, then to confirm a ductal injury an x-ray contrast study as described above to detect common duct injuries can also be performed to detect pancreatic duct injuries, especially in the head of the pancreas (See 4.1.2. above).
Alternatively, a duodenotomy may be performed and direct cannulation of the ampulla of Vater and thereby the pancreatic duct can be achieved using a blunt-tipped fine probe. If the probe is visualized within the pancreatic injury then the diagnosis of ductal involvement is confirmed. If x-rays are available, contrast may be introduced into the ampulla of Vater and contrast extravasation would demonstrate a pancreatic duct injury and its location. As with the common bile duct, the infusion of normal saline slowly with inspection within the penetrating pancreatic wound for pooling or collection of fluid could confirm ductal transection. Accurate determination of associated ductal injury in penetrating trauma to the pancreas has been shown to reduce complications from 55% to 15%.  Therefore, a diligent search for any ductal disruption is warranted.
Pancreatic parenchymal injuries, without pancreatic duct injury, can be treated by surgical hemostasis and adequate drainage with a good outcome. At least two drains, placed within the peritoneal cavity cephalad and caudal or anterior and posterior to the area of injury depending on its location, and should be placed, with closed drainage systems employed whenever available. Attempts at closing lacerations or capsular tears can actually increase bleeding and further damage parenchyma. Injuries left "unsutured" heal mainly without ongoing leakage. However, the ones that do form a pancreatic fistula can be treated nonoperatively with adequate nutrition and drainage (by the originally placed drains as described above). An option is to place a feeding jejunostomy tube for enteral feeding if the pancreatic injury is large enough to be concerning for ongoing leakage and gastric ileus. The drainage from the drains can be sent for amylase measurement if available and when levels fall below or to serum amylase levels and the patient is tolerating an oral diet or full enteral tube feeds, the drains can be removed.
Untreated ductal injuries on the other hand cause major pancreatic fistula that close less often spontaneously and can result in pancreatic ascites. The treatment of a ductal injury is based on its location. Injuries to the left of the mesenteric vessels can be included in a distal pancreatectomy and the remaining pancreas will be at least 50%, minimizing the chance of the development of diabetes or fat malabsorption. The pancreas is divided, at the level of the ductal injury, and the proximal pancreas is closed with nonabsorbable mattress sutures. The duct itself in otherwise healthy trauma patients is usually too small to be visualized and closed. However, if it is visible then it should be closed separately from the parenchyma using a nonabsorbable suture. All patients should also receive a feeding jejunostomy tube to allow for early enteral feeding reducing septic complications postoperatively.
When ductal injury occurs in the head of the pancreas or to the right of the mesenteric vessels then a Whipple's procedure is often required, especially if there is involvement of the ampulla of Vater, the common bile duct or the duodenum. Whipple's operation after pancreatic trauma is required about 3% of the time. If the Whipple's operation or pancreaticoduodenectomy is delayed to the second or third day post-injury after an initial operation (damage control operation) during which excessive hemorrhage and bile leakage are controlled, the outcome is similar to elective Whipple resection for cancer.
Unlike, blunt trauma, penetrating injury to the spleen is usually dealt with operatively. In the patient who is hemodynamically stable, the commonest reason to operate on a penetrating splenic injury is related to a concern over other organ injury and involvement of the left diaphragm. A splenic injury is often associated with a hole in the left hemi-diaphragm that is not otherwise visible except through laparotomy (or diagnostic laparoscopy if the equipment and expertise is available). Asymptomatic diaphragmatic injuries upon presentation can develop late complications of herniation of the spleen, stomach or colon if left untreated. Late diaphragmatic hernias post-trauma have been reported even decades after the initial traumatic event.
A splenic injury should always be suspected in any penetrating injury in the left upper quadrant or left flank. If a patient does not have any signs or symptoms of bleeding then assessment of the patient should follow the algorithm set out in Figure 1. Additional investigations can also include chest x-rays, diagnostic peritoneal lavage and abdominal CT scan, if available. Chest x-rays may be helpful in suggesting a splenic injury if there is a left hemothorax or an elevated left diaphragm but a normal chest x-ray does not rule it out. Diagnostic peritoneal lavage can also be used and a positive result (>10 ml of aspirated blood; fluid has RBC count > 10,000/mm3 or you cannot read newsprint through the fluid) warrants exploratory laparotomy and splenorraphy or splenectomy.
In penetrating injury, the splenic injury is more often an arterial vessel that is not amenable to splenic salvage maneuvers. If the spleen is not actively hemorrhaging and the laceration is repairable then splenic salvage is warranted even in penetrating trauma. Any injury to the spleen can be associated with injuries to surrounding structures: stomach, pancreas, transverse and left colon, left kidney and the left lobe of the liver. All surrounding structures should be closely inspected intra-operatively. When splenic repair is being considered then the spleen should be mobilized medially with the tail of the pancreas, to minimize iatrogenic pancreatic injury. This provides the best exposure to the splenic hilum. As in all trauma laparotomies, the midline incision should be used and be extended to the left of the xiphisternum, if necessary. Upon entry into the abdomen, if hemoperitoneum is discovered, then even when splenic injury is suspected; all four quadrants of the abdomen should be packed with lap pads. Then, careful removal of the packs one quadrant at a time, inspecting all structures in each quadrant assures no missed injuries. The spleen should be mobilized gently as an injury can be easily worsened, changing a potential splenorraphy into a splenectomy. The spleen should be elevated from lateral to medial, taking down the lateral wall attachments first, followed by the splenophrenic and splenorenal ligaments. The spleen should then be lifted medially by placing the surgeon's hand posteriorly between the spleen and the left kidney. The tail of the pancreas should be raised with the spleen as one unit. The short gastric vessels are the last to be ligated and depending on the location and severity of the injury, they may be left in-situ. Injury to the greater curvature of the stomach should be avoided during short gastric vessel ligation. The short gastric vessels can be very short and any suspect injury to the stomach wall no matter how small should be over sewn with a Lambert suture. Delayed gastric rupture has occurred from even a small area of stomach wall that has been necrosed in a clamp or the tie of a gastric vessel. As the spleen is elevated, lap sponges can be placed posteriorly to improve visibility. Lastly, the splenocolic ligaments between the spleen and the left colon should be divided. While mobilization is being undertaken, hemorrhage from the spleen can be controlled by direct finger compression or if severe bleeding by clamping the splenic hilum gently with a noncrushing clamp. The splenic fosssa should be packed with lap sponges.
After mobilization of the spleen as described above, the injured area can be inspected. If the splenic injury is no longer actively bleeding, does not involve the hilum of the spleen, and there are no other associated serious intra-abdominal or thoracic injuries requiring immediate attention, then splenorraphy can be considered rather than proceeding with splenectomy. The surgical options and approaches to splenorraphy are similar to the previously described techniques for hepatic repair: 1) suture repair with a large blunt needle and chromic catgut placing interrupted simple or mattress sutures across lacerations; 2) omental patch placed into lacerations or defects and sutured in place with a few interrupted absorbable sutures such as vicryl; 3) placement of hemostatic agents such as surgical or fibrin glue if available; 4) completion partial splenectomy where a transected portion of the spleen is removed and the open parenchymal surface is sutured closed with chromic catgut and reinforced with an omental patch and lastly, 5) if omentum is unavailable, the spleen can be "wrapped" in a vicryl mesh (absorbable mesh) which can act to oppose lacerated and open parenchymal surfaces. There is no evidence to suggest that one method is more successful than the others and choice is usually based on the surgeon's previous experience and the injury s/he is faced with. If splenorraphy is performed the patient should be watched closely post-operatively for any signs or symptoms of recurrent hemorrhage which would mandate reoperation and splenectomy.
Splenectomy is completed by ligating and transecting the hilum. It is preferable to ligate the vein and artery separately if possible to prevent the possible development of arteriovenous fistula. Once the spleen is removed, the greater curve of the stomach, the tail of the pancreas, the left kidney and ureter and the splenic fossa should all be carefully inspected for iatrogenic injury, urine, pancreatic leak or ongoing bleeding. Unless there is concern over a potential pancreatic or urine leak, no drain should be left after an uncomplicated splenectomy or splenorraphy.
Antibiotics should be given at incision but there is no need to continue prophylactic antibiotics post-operatively because of a splenic injury alone. In a stable patient, pieces of spleen may be transplanted into pockets within the greater omentum in an attempt to preserve splenic function and reduce the risk of postsplenectomy sepsis. This severe and often fatal sepsis is more common in the pediatric population, but has occurred at all ages. All post-operative splenectomy patients with a fever should receive antibiotics early and empirically to cover all encapsulated bacterial organisms. All splenectomy patients should receive vaccines to cover hemophilus, pneumococcus and meningococcus 2 weeks post-operatively preferably or prior to discharge, if available.
5. Hollow Viscus Injury
Patients with penetrating injuries to the stomach almost always present with peritoneal signs. The leakage of the low pH content of the stomach causes a rapid, severe peritoneal irritation that almost universally results in significant clinical signs of peritonitis. Bloody aspirate from a nasogastric tube can be a sign of a gastric injury, but it is not specific or sensitive enough as a single test to define a gastric injury.
GI contamination from the stomach can be controlled with a running absorbable suture temporarily until definitive repair is undertaken later in the operation. Once ongoing hemorrhage and GI contamination is controlled, the full anterior and posterior aspect of the stomach must be evaluated. The stomach should be decompressed with a nasogastric tube as soon as possible. The posterior stomach is accessed by entering the lesser sac through an avascular window in the left lateral aspect of the greater omentum between the stomach and the left transverse colon. This is the best location as it minimizes injury to the transverse mesocolon and potential injury to the middle colic vessels. The gastrocolic portion of the greater omentum can be ligated toward the left and right aspect of the stomach, being careful to preserve the gastroepiploic vessels on the greater curve of the stomach in order to improve visualization of the posterior aspect of the stomach, if necessary. The gastroesophageal junction and proximal lesser curvature are also difficult areas of the stomach to completely visualize. The left triangular ligament of the left lobe of the liver often must be divided to allow exposure of the full stomach. Remember the proximal greater curvature has the short gastric vessels, as well as the splenic capsule that can be injured with minimal traction resulting in unnecessary bleeding. If there is an injury in this portion of the stomach, the short gastric vessels can be ligated to improve exposure.
Anytime one hole is found, usually an anterior one, a search must be made for a second hole, often a posterior hole. Perforations and lacerations on the lesser and greater curvature of the stomach are often difficult to find and any hematoma or area of contusion needs to be carefully inspected, including dissection of the associated area of stomach mesentery to expose the stomach wall and confirm its integrity. If a second perforation is not found, then the stomach can be insufflated with air (place a clamp on the proximal duodenum) and saline can be placed in the peritoneal cavity filling over the stomach. Any bubbling of air through the stomach can be used to localize the area of an unseen injury. Alternatively, after clamping the stomach it can be filled with methylene blue mixed in saline - lap sponges placed around the stomach will be stained blue by any leak of the methylene- saline mixture from the stomach. Lastly, the one discovered hole can be extended to allow inspection of the stomach from the interior to discover the second wound.
Identified stomach perforations and lacerations should have the traumatized edges resected. As the stomach is very vascular it should be closed in two layers, to ensure hemostasis: a full thickness inner layer of an absorbable running suture followed by an interrupted Lembert or seromuscular absorbable suture. If there is excessive hemorrhage then the inner running suture layer may also be locked. In the pyloric area, a pyloroplasty should be performed to avoid pyloric stenosis. If there is extensive destruction then a partial gastrectomy may be performed with a Billroth I anastomosis if there is enough duodenum or a Billroth II, gastrojejunostomy if the proximal duodenum is also damaged. Hematomas of the wall should be evacuated; hemostasis achieved and closure with absorbable Lembert sutures only if no full thickness injury is encountered. If a gastric injury is associated with a diaphragmatic injury especially if there is extensive contamination, then the chest cavity should be extensively irrigated to prevent potential empyema. Drainage of primary repairs or anastomotic lines on the stomach is not usually necessary.
Injuries to the duodenum often take on increased importance because of its proximity to many life-sustaining structures such as the pancreas, portal vein, common bile duct, pancreatic duct and superior mesenteric vein and artery. Also, the duodenum is partially a retroperitoneal structure such that isolated injuries can have an occult initial presentation. Its posterior boundaries include the kidney, inferior vena cava and aorta. Therefore, location of the duodenal injury and inspecting all its neighboring structures is paramount to minimizing missed injuries in this area from a penetrating injury.
The majority of duodenal injuries are due to penetrating injuries and these should always be considered in any penetrating wound to the upper abdomen or thoraco-abdominal area or when there is a midline upper abdominal hematoma discovered during exploratory laparotomy.  Any signs of blood, bile or air in the retroperitoneum, mandate a complete exploration of the duodenum in its entire length. The initial approach to the patient with a duodenal injury does not differ from the principles set out in 2. Initial Assessment and Figure 1. An unstable patient should always be taken directly to the operating room. If the patient is stable, initial chest and abdomen x-rays utilizing paper clips taped to cover the penetrating wounds can assist in location of the wounds on x-rays. This may allow an increased index of suspicion for the potential of injuries to the duodenum and surrounding structures. The duodenum is a retroperitoneal structure and, therefore, may not present with peritonitis. Air in the retroperitoneal space around the kidney (mainly right) or along the psoas muscle shadow, though rare, is a classic sign of retroperitoneal duodenal injury. Both CT scans and even contrast duodenography have in the order of up to a 20% false negative rate - specific findings but not very sensitive for injury.  Therefore, they are not relied upon to definitively exclude duodenal injury.
At exploratory laparotomy, the duodenum requires careful inspection both anteriorly and posteriorly, mandating the performance of a Kocher maneuver. This is performed by sharply dividing the peritoneal attachments to the "C" or second portion of the duodenum starting from cephalad to caudal on the anti-mesenteric side of the duodenum. This allows the duodenum and the head of the pancreas to be elevated together, palpated anterior to posterior, and visualize all surfaces for injury as well as excluding surrounding named vascular injury. The hepatic flexure of the colon (right colon) also requires mobilization in order to complete the opening of the retroperitoneal attachments of the duodenum around the 3rd portion and up to the mesenteric vessels. If the injury involves the fourth portion of the duodenum, then careful division of the Ligament of Trietz may be necessary, but care must be taken to avoid the inferior mesenteric vein which runs laterally to the 4th part of the duodenum in this ligament. The duodenum can now be mobilized from left to right and the body of the pancreas can also be better visualized. The small bowel can also be reflected superiorly and with careful incision of the right side of the root of the mesentery at the base of the transverse colon, the third portion of the duodenum is exposed.
Once the location of the injury is verified through appropriate exposure, the treatment decision-making is dependent on the extent of injury and associated pancreatic injury involvement. The extent of duodenal injuries include: partial to full thickness lacerations or perforations, complete transection with loss of continuity and length, lacerations involving the ampulla of Vater, devascularization and finally, intramural hematomas without laceration or perforation. Lacerations to the duodenum from partial to complete transection can be repaired primarily if there is no significant devascularization or repair does not involve excessive tension (after excising all nonviable and damaged edges). A single layer interrupted repair with, preferably absorbable suture such as PDS, or alternatively any monofilament absorbable suture can be performed. Repairs performed in two layers must be certain to maintain an adequate lumen size. Repair is usually performed in the direction of the injury, with single layer interrupted repair, even with longitudinal lacerations; the concern of luminal compromise is no longer warranted.
When there is a laceration on the mesenteric or pancreatic side of the duodenum full mobilization of the area for adequate closure may be impossible without devascularization or injury to the pancreas. In these cases, repair can be performed from the luminal side of the laceration through an anti-mesenteric duodenotomy.
Many trauma centers, advocate placement of a feeding jejunostomy tube after repair of a duodenal injury for enteral feeding, along with nasogastric suction. With the more extensive injuries, especially if there was significant tissue destruction or pancreatic injury, healing may require longer than the usual 5-7 days before return to adequate GI integrity, and therefore, the jejunostomy tube allows concurrent enteral feeding minimizing septic complications. However, in small (<2 cm) lacerations involving only one duodenal surface, especially with minimal tissue destruction such as in stab wounds, nasogastric suction without a jejunostomy feeding tube may be acceptable.
Complete transection of the duodenum can be repaired primarily in an end-to-end fashion as long as the edges are viable and reach without tension after minimal mobilization. This is often more difficult to achieve in the 2nd and 3rd portions of the duodenum because of its attachment to the pancreas. Injuries due to gunshot wounds, involvement of the 2nd portion of the duodenum, injuries greater than 75% of the circumference of the wall, and repairs attempted greater than 24 hours from the time of injury were found to be major determinants of a poor outcome, if primary repair was attempted. 
When primary repairs are undertaken in the presence of any of the factors mentioned above or in association with an injury, then protection of the repair is performed using a pyloric exclusion technique. One of the most commonly used techniques was described by Vaughn et al., where the stomach is opened anteriorly on the greater curve of the stomach near the pylorus and the pylorus is over sewn with a running nonabsorbable suture, such as prolene. The gastrotomy is then incorporated into a side-to-side loop gastro-jejunostomy, performed to allow alternative gastric drainage until the duodenum heals and the pylorus spontaneously opens. A feeding jejunostomy tube is usually also placed away from the limbs of the gastrojejunostomy. The pylorus opens in 6-12 weeks so a vagotomy to prevent jejunal ulceration in the gastro-jejunostomy is unnecessary. Severe duodenal injuries, even repaired with minimal tension and adequate vascularization have a higher rate of breakdown and fistula formation than the other parts of the small bowel. Therefore, the pyloric exclusion technique allows conservative nonoperative treatment of the majority of duodenal fistulae that may occur. These will eventually close if there are no associated abscesses and the patient's nutrition is maintained by oral feeding or via the jejunostomy enteral feeding tube.
In cases, where the patient is unstable and there is extensive tissue destruction, then over-sewing the proximal open end of the duodenum, placing a lateral (or anti-mesenteric) duodenostomy tube and a retrograde jejunostomy tube has been reported.  Alternatively, over-sewing the duodenal injury to minimize ongoing bile leakage and then placing extraluminal drainage with two or more Jackson Pratt or similar closed drainage systems. The duodenal injury can then be repaired at a later date when the patient has stabilized. The procedure of choice in these instances is excision of the damaged duodenum and reconstruction with a loop of jejunum passed through the transverse mesocolon. If the injury involves the area of the ampulla of Vater (but it is still intact) then a side-to-end duodenojejunostomy with Roux-en-Y repair should be performed. If the injury is below and away from the ampulla of Vater then a direct end-to-end anastamosis can be performed. In injuries where there is complete disruption of the ampulla of Vater with or without common bile duct injury, initial control of the bile leakage by the methods mentioned above and then a reassessment for the need to perform a Whipple's operation should be made after the patient has stabilized in 24-48 hours. The undertaking of a Whipple's procedure at the initial operation has been discouraged in more recent years as the associated mortality rate is extremely high.
Duodenal hematomas are less common in penetrating injury but can occur. They typically present after 48 hours after the injury with signs and symptoms of gastric outlet obstruction. Initial management should entail nasogastric suction and IV hydration and observation for 5-7 days as the majority resolve with time. However, if symptoms persist then operative evacuation may be necessary. If a duodenal intramural hematoma is found during an exploratory laparotomy and there is no associated perforation, most trauma surgeons would advocate not evacuating the lesion and treating the patient with nasogastric suction and hydration as most resolve spontaneously. However, if the hematoma distorts anatomy significantly, either evacuation with careful primary repair may be undertaken or placement of a feeding jejunostomy tube to allow enteral nutrition until the hematoma resolves are alternatives for these larger lumen-compromising hematomas.
5.3. Small Bowel
In penetrating injuries, small bowel injury is relatively common especially in comparison to the incidence found in patients who have sustained a blunt force trauma. With gun shot wounds to the abdomen that penetrate the fascia, the incidence approaches 80% while after a stab wound it is closer to 30%. The operative management of small bowel injuries is relatively uncomplicated and therefore, the main factor determining a successful patient outcome is the timely diagnosis of the injury. With the increasing application of nonoperative management for penetrating injury as well as blunt in the hemodynamically stable patient, the possibility of a small bowel injury must always be paramount in the surgeon's mind. Missed injuries can lead to abdominal sepsis and death.
As discussed in 2. Initial Assessment, penetrating injuries can be treated nonoperatively if the patient is completely symptom-free and has no distracting injuries or intoxication. However, this is the safest and most successful only for stab wounds, while the nonoperative management of gun shot wounds requires more experience, close clinical observation and in most instances is not recommended. Any patient with diffuse abdominal tenderness, peritonitis, active hemorrhage or hemodynamic instability, should be taken immediately to the operating room. Also, evisceration after a stab wound warrants a laparotomy as it is associated with a 75% chance of an intra-abdominal injury, even without any clinical signs or symptoms.  Small bowel injuries, unlike gastric or colonic injuries, can lack obvious peritoneal signs early after an injury. This is due to the fact that small bowel succus entericus has a neutral pH and is sterile in the normal individual. Therefore, ascertaining the time of injury is important in considering the presence or absence of small bowel injuries. Also, depending on the mechanism, stab versus gun shot wound, and the number and size of bowel "holes" there may be a limited amount of succus spillage initially masking early symptoms. Except for patients on high dose steroids or who are severely immunocompromised, the majority of small bowel injuries become clinical symptomatic with positive physical examination findings within 24 hours of injury.
Abdominal examination to detect injury to the small bowel is accurate more than 90% of the time and especially when performed serially by the same clinician. Laboratory tests such as hematocrit, white blood cell count and serum amylase are not useful in the initial evaluation of a patient with small bowel injury. However, the development of fever (temperature greater than 38.5 Celsius), a rising white blood cell count with a shift to the appearance of immature cells or "bands" in the count, a rising amylase, development of metabolic acidosis, unexplained tachycardia or hypotension are suggestive of a small bowel injury.
A midline incision should be utilized to enter the abdomen. If a gunshot wound has made a large abdominal defect, it may be used and extended to explore the abdomen in some cases. If there is significant bleeding upon entry, after packing, GI contamination should be controlled with Babcock clamps until life-threatening bleeding injuries are controlled. Alternatively, a running suture with catgut or vicryl can be placed over intestinal perforations to temporarily control GI contamination. This can also be useful if there is ongoing bleeding from the perforated intestinal wall.
The small bowel should be eviscerated and carefully inspected from the ligament of Trietz to the ileo-cecal valve on at least two occasions during the same operation. Penetrating injuries are often multiple and can be surprisingly small and inconspicuous and require careful circumferential inspection of the bowel wall including the mesenteric side. Any hematomas must be opened, evacuated and the adjoining bowel wall examined for its integrity. After all areas of injury on the bowel are found then the appropriate type of repair can be planned. Primary repair after resection of traumatized edges is the preferred method of repair. Single layer closure with interrupted absorbable suture or two layer closure, have equal effectiveness. If available, a longer lasting absorbable suture such as PDS is preferred especially if a single layer closure is completed. If the areas of injury are multiple or there is extensive damage, then resection of all involved area in continuity is necessary. However, if resection may compromise adequate length for future absorption, then every attempt should be made to primarily repair the bowel. The direction of repair is most often transversely to minimize the potential for stenosis but long anti-mesenteric injuries may have to be closed following the long axis. This type of repair may be closed with one layer to minimize narrowing of the lumen. The direction of the repair will also be determined by assuring a tension-free repair as well as an adequate lumen size. If >50% of the circumference of the bowel is damaged, then usually resection is required to prevent clinically significant luminal narrowing. The proximal intestine has the largest diameter which narrows as it goes distal into the ileum, and as such, primary repair are more often possible in proximal injuries. Areas of small bowel, where the mesentery has been damaged and the bowel has been devascularized, must be resected and re-anastomosed. The main two choices of method of repair are either hand-sewn or stapled anastamosis. There has been controversary over this choice with retrospective trials trending toward a lower morbidity rate after hand-sewn anastamoses.  At present, the decision to hand sew or to use a stapler is based on availability of stapling equipment, surgeon choice and with similar complication rates. The technique of closure is not as important as following the basic principles of all intestinal anastomoses: tension-free, good blood supply, adequate lumen size, and water-tight closure.
Mesenteric hematomas should be re-inspected throughout the operation to detect any expansion. If the hematoma is seen to be expanding, it should be opened, evacuated and hemostasis achieved. The ligament of Trietz may be divided to access proximal jejunal injuries but care must be taken to avoid injuring the inferior mesenteric vein. If necessary, the vein may be ligated to improve visualization and repair of the proximal bowel. Bleeding at the base of the mesentery must be done under direct visualization of the bleeding vessel, if possible, as damage to the main artery; especially since injury to the superior mesentery artery at its root can compromise the blood supply of the entire midgut.
Small bowel anastomoses have low leak rates in the nontrauma setting. But in trauma, intestinal anastomoses have an increased risk of leak and breakdown if any of the following factors are present: shock with massive fluid and blood resuscitation, associated pancreatic injury, or the development of an abdominal compartment syndrome. ,  If a patient is severely injured and has developed acidosis, hypothermia and is coagulopathic, then damage control measures mandate limiting definitive surgical repairs. Damaged bowel may be resected and the ends stapled or sewn shut and left in the abdomen without immediate re-anastomosis. After 24-48 hours of intensive resuscitation and restoration of more near normal physiology, the patient can be brought back to the operating room to have any intestinal anastomosis performed at that time. Any patient who has blind-end loop of bowel requires a nasogastric tube to intermittent suction to minimize bowel distension.
The commonest penetrating injury to the colon is direct perforation. However, serosal injuries or partial thickness injuries can occur, especially after gunshot wounds, due to a "blast effect" of the passing bullet. This blast effect can cause contusion, ischemia and eventual perforation of the colon wall without entering the colon. Blast effect can also occur to the retroperitoneal colon from bullets that pass through the soft tissues of the back, not entering the colon but passing close enough to affect this damage. The diagnosis of the patient with a colon injury should follow the initial steps discussed in Section 2. Initial Assessment and the algorithm in Figure 1. However, the main stay of diagnosis of a colonic injury is by physical examination: peritonitis with or without signs of abdominal tenderness. Patients who are initially completely asymptomatic and hemodynamically stable, should be followed closely for increasing symptoms of abdominal pain or increasing signs of fever, white blood cell count or worsening base deficit. Patients with retroperitoneal injuries often present initially asymptomatically. In a review of penetrating flank wounds in our institution, we found that all initially asymptomatic retroperitoneal injuries developed signs and symptoms within 18 hours from the time of admission. This time lapse between injury and definitive repair in retroperitoneal injuries is not associated with an increase in complications.
During an exploratory laparotomy for trauma, any hematoma, contusion, discoloration of the colonic wall should be inspected and dissected to assure an intact, viable colonic wall underneath this injury. Any hematoma in the mesentery should also be explored to ascertain that no significant vessel is only temporarily being tamponaded. Also, ligation of any mesenteric vessels requires close observation of the associated colon for the duration of the laparotomy to ensure that portion of the colon has enough collateral circulation to remain viable. Any devascularized or compromised colon should be resected as a postoperative leak in a multi-system trauma patient is associated with a high rate of morbidity and mortality.
The traditional treatment for any colonic injury is complete resection of the area of damaged colon and the creation of an end colostomy. However, this approach requires the patient to undergo a second operation in the future with an associated complication rate. Therefore, there has been increasing interest in treating colonic injury with one definitive operation at the time of injury. Presently, any patient who presents without shock, extensive fecal contamination, multiple associated injuries, significant blood loss with blood transfusion greater than 6 units, prolonged delay to operation or a left sided colon injury, may be considered for primary repair as opposed to resection and creation of a colostomy. A primary repair of a colonic perforation is usually performed when less than 50% of the circumference of the colon is involved in the injury or in right sided colon injuries, that require resection, an ileo-colonic anastomosis can be performed. As confidence in primary repair has increased some surgeons have broadened the scope to include the primary repair of a left colon lesion where the injury is small (<4 cm), associated with minimal tissue destruction and the other criteria mentioned above are met. Prospective randomized studies from 1979 to present, have shown there is no increased morbidity or mortality to perform a primary repair or resection and anastomosis of a right colon lesion in comparison to diverting colostomy., , and 
The American Association of Surgery for Trauma embarked upon a multicenter prospective noncontrolled study of more severe colon injuries treated by either primary repair or diverting colostomy. Once again there was no difference between the groups in terms of complications and in a multivariate analysis of the data; type of surgical repair (primary versus colostomy) was not an independent predictor of outcome.  Therefore, more and more surgeons are widening their indications for primary repair and resection and anastomosis rather than colostomy for operative management of colonic injuries. When performing a primary repair, controversy persists in regards to stapled (when available) versus hand-sewn and one layer hand-sewn repairs versus two layer hand-sewn, but there is no strong evidence to conclude the preeminence of one of these techniques over the others.  The main tenets of a well-vascularized, tension-free, water-tight anastomotic or repair line are vital to minimizing post-operative complications and mortality. Issues of suture type, amount, and number of layers still remain surgeon choice.
In situations where a colostomy is utilized, there is good reason to consider a preference for a completely diverting loop colostomy rather than an end colostomy. Complete diversion can be achieved by suturing over the distal end of the loop after exteriorization. Almost all trauma colostomies are nonpermanent therefore ease of re-anastomosis at a future operation is a major advantage of loop colostomy over end colostomy. The literature has shown that primary repair does not require drainage and it is not routinely utilized in most trauma centers.  Although there is a wide discrepancy in wound infection rates when primarily closing the skin after colon trauma in the literature, the standard of care in most trauma centers is to leave the skin open and allow healing by secondary intention. Wound infection is a major risk factor for wound dehiscence and necrotizing soft tissue infections and therefore, not closing the skin after colon trauma, remains the prominent approach to postoperative wound care.
The commonest major complication after colon injury is infection, often occurring in greater than 50% of the patients.   Abscess formation is a common sequel and its occurrence is associated with significant mortality. Therefore, there remained debate over the use of prophylactic antibiotics: doses, timing, amount and type. However, recent evidence shows that infection rates after colonic injury are not reduced by the prolonged administration of antibiotics post-operatively. Therefore, patients with colonic injuries should receive a single broad spectrum antibiotic agent for a total of 24 hours post-operatively only.
Any penetrating injury in the region of the pelvis, including the gluteal region should be considered as possibly involving the intra-peritoneal or extra-peritoneal rectum. As has been described above a chest x-ray and abdominal x-ray should be taken (in the stable patient) with paper clips over the penetrating wounds. If the trajectory between two bullet wounds or a single bullet wound is above a line which can be drawn between the greater trochanters on the x-ray then the likelihood of a rectal injury is greatly increased. In our institution, a penetrating wound above this line on the x-ray mandates a rigid sigmoidoscopy examination. The clinical abdominal examination is only useful with intraperitoneal rectal injuries which will present with peritonitis and abdominal tenderness. The posterior 2/3 and the lower 1/3 of the rectum are extra-peritoneal and initial symptoms are usually absent. Therefore, a high index of suspicion is required and should include a digital rectal examination with palpation of the entire mucosal surface for defects and a rigid sigmoidoscopy if available. Either of these examinations is considered positive for rectal injury if blood is seen on the examining glove or through the scope. It is not necessary to delineate the exact injury for it to be recognized as present and be definitively treated. If available, a CT scan of the pelvis is also employed for pelvic penetrating injuries. An alternative could be rectal contrast x-ray to display a rectal injury.
Intraperitoneal injuries are treated as colon injuries discussed in the previous section and increasingly are treated with primary repair without colostomy. For extra-peritoneal injuries the traditional standard of care has been a diverting colostomy with presacral drainage and rectal washout. Rectal washout is no longer performed and it is now believed to increase the rate of infection by "pushing" stool/bacteria into surrounding tissues.  Presacral drainage is also being utilized less and less. A recent randomized control trial found no difference in the number of patients who developed infection between the group with a presacral drain and the group without.  Therefore, presacral drains are used less commonly but diverting colostomy, usually loop colostomy, remains the standard of care for an extraperitoneal rectal injury. If the rectal injury is visible through the anus then a primary repair can be performed. However, primary repair is not essential to a good outcome. Rectal fistula and pelvic sepsis are rare and their occurrence is not correlated to initial primary repair of the defect. If primary repair is performed, a colostomy is still believed to be necessary because of the lack of a serosa in the extraperitoneal rectum and the anatomical difficulty in assuring a water-tight seal. Therefore, overall, diverting colostomy with or without primary repair remains the main treatment for extra-peritoneal rectal injuries.
Penetrating injuries to the torso are highly lethal and are associated with internal injuries in a large number of patients, especially after a gun shot wound. Any patient hemodynamically unstable or with acute peritonitis mandates an operation. However, with the refinement of the field of trauma surgery, alternative options have developed for patients who sustain a penetrating injury but do not require immediate operation. In this review, we have covered the recent treatment approaches utilized based on the patient's presenting condition as well as the operative and nonoperative approach to specific organ injuries. A summary of recommendations from this review include:
Jana B.A. MacLeod MD, MSc, FRCS(C), FACS
Assistant Professor of Surgery, Assistant Director of Trauma
Emory University School of Medicine/ Grady Memorial Hospital,
2. Rozycki GS, Ballard RB, Feliciano D, Schmidt JA. Surgeon –performed ultrasound for the assessment of truncal injuries: Lessons learned from 1540 patients. Ann Surg 1998 Oct. 228(4):557-67. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/68425
3. Sisley AC, Rozycki GS, Ballard RB, Namias N, Salomone JP, Feliciano DV. Rapid detection of traumatic effusion using surgeon-performed ultrasound. J Trauma 1998 Feb. 44(2):291-6. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/68426
4. Boulanger BR, Kearney PA, Tsuei B, Ochoa JB. The routine use of sonography in penetrating torso injury is beneficial. J Trauma 2001 Aug. 51(2):320-5. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/68427
5. Ochsner MG, Knudson MM, Pachter HL, Hoyt DB. Cogbill TH. McAuley CE. Davis FE. Rogers S. Guth A. Garcia J. Lambert P. Thomson N. Evans S. Balthazar EJ. Casola G. Nigogosyan MA. Barr R. Significance of minimal or no intraperitoneal fluid visible on CT scan associated with blunt liver and splenic injuries: A multicenter analysis. J Trauma 2000 Sep. 49(3):505-10. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/68428
6. Ginzburg E, Carrillo EH, Kopelman T, . McKenney MG. Kirton OC. Shatz DV. Sleeman D. Martin LC. The role of computed tomography in selective management of gunshot wounds. J Trauma 1998 Dec. 45(6): 1005-9. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/68429
7. Stafford RE, McGonigal MD, Weigelt JA, Johnson TJ. Oral contrast solution and computed tomography for blunt abdominal trauma: A randomized study. Arch Surg 1999 Jun. 134(6):622-6. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/68430
8. Velmahos GC, Demetriades D, Cornwell EE III, Asensio J. Belzberg H. Berne TV. Gunshot wounds to the buttocks: Predicting the need for operation. Dis Colon Rec 1997 Mar. 40(3):307-11. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/68431
9. Velmahos GC, Gomez H, Falabella A, Demetriades D. Operative management of civilian rectal gunshot wounds: Simpler is better. World J Surg 2000 Jan. 24(1):114-8. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/68432
10. Demetriades D & Velmahos GC. Indications for and Techniques of Laparotomy. In Trauma, 6th Edition. 2008. Pgs 610-11.
11. Demetriades D, Rabinowitz B. Indications for operation in abdominal stab wounds. Ann Surg 1987 Feb. 205(2):129-32. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/68433
12. Shorr RM, Gottlieb MM, WebbK, Ishiguro L. Berne TV. Selective management of abdominal stab wounds. Arch Surg 1988 Sep.123(9):1141-5.
13. Demetriades D, Rabinowitz B, Sofianos C, Charalambides D. Melissas J. Hatzitheofilou C. Da Silva J. The management of penetrating injuries of the back. A prospective study of 230 patients. Ann Surg 1988 Jan. 207(1):72-4. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/68434
14. Granson MA, Donovan AJ: Abdominal stab wound with omental evisceration. Arch Surg 1983 Jan. 118(1):57-9.
15. Udobi Kf, Rodriguez A, Chiu WE, Scalea TM. Role of ultrasonography in penetrating abdominal trauma: A prospective clinical study. J Trauma 2001Mar. 50(3):475-9. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/68435
16. Shah R, Max MH, Flint LM Jr. Negative laparotomy: Mortality and morbidity among 100 patients. Am Surg 1978 Mar. 44(3):150-4.
17. McCarthy MC, Lowdermilk GA, Canal D, Broadie TA. Prediction of injury caused by penetrating wounds to the abdomen, flank and back. Arch Surg 1991 Aug. 126(8):962-5.
18. Moore EE, Moore Jb, Van Duzer-Moore S, Thompson JS. Mandatory laparotomy for gunshot wounds penetrating the abdomen. Am J Surg 1980 Dec. 140(6):847-51.
19. Lowe RJ, Slietta JD, Read DR, Radhakrishnan J. Moss GS. Should laparotomy be mandatory or selective in gunshot wounds of the abdomen? J Trauma 1977 Dec. 17(12):903-7. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/68439
20. Muckart DJ, Abdool-Carrim AT, King B. Selective conservative management of abdominal gunshot wounds: a prospective study.Br J Surg 1990 Jun. 77(6):652-5. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/68438
21. Velmahos GC, Demetraides D, Foianini E, Tatevossian R. Cornwell EE 3rd. Asensio J. Belzberg H. Berne TV. A selective approach to the management of gunshot wounds to the back. Am J Surg 1997 Sep. 174(3):342-6. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/68437
22. Velmahos GC, Demetriades D, Cornwell EE III, Belzberg H. Murray J. Asensio J. Berne TV. Selective management of renal gun shot wounds. Br J Surg 1998 Aug. 85(8):1121-4. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/68436
23. Bickell WH, Wall MJ, Pepe PE, . Martin RR. Ginger VF. Allen MK. Mattox KL. Immediate verus delayed resuscitation for hypotensive patients with penetrating torso injuries. N Engl J Med 1994 Oct 27. 331(17):1105-9. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/50463
24. Dutton RP, Mackenzie DF, Scalea TM. Hypotensive resuscitation during active hemorrhage: impact on in-hospital mortality. J Trauma 2002 Jun. 52(6):1141-6. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/68440
25. MacLeod J. Lynn M. McKenney MG. Jeroukhimov I. Cohn SM. Predictors of mortality in trauma patients. Am Surg 2004 Sep. 70(9):805-10. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/68441
26. Stone HH, Lamb JM: Use of pedicled omentum as an autogenous pack for control of hemorrhage in major injuries of the liver. Surg Gyn & Obstet 1975 Jul. 141(1):92-4.
27. Pachter HL, Spencer FC, Hofstetter Sr, Liang HG. Coppa GF. Significant trends in the treatment of hepatic trauma: experience with 411 injuries. Ann Surg 1992 May. 215(5):492-500. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/68442
28. Feliciano DV, Mattox KL, Jordan GL, Burch JM. Bitondo CG. Cruse PA. Management of 1000 consecutive cases of hepatic trauma (1979-1984). Ann Surg 1986 Oct. 204(4):438-45. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/68443
29. Man K, Fan ST, Ng IO, Lo CM. Liu CL. Wong J. Prospective evaluation of Pringle maneuver in hepatectomy for liver tumors by a randomized study. Ann Surg 1997 Dec. 226(6):704-11. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/68444
30. Fabian TC, Stone HH. Arrest of severe liver hemorrhage by an omental pack. South Med J 1980 Nov. 73(11):1487-90.
31. McClelland RN, Shires T. Management of liver trauma in 259 consecutive patients. Ann Surg 1965 Feb. 161:248-57. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/68445
32. Lee TY. Chen YL. Chang HC. Yang LH. Chan CP. Chen ST. Kuo SJ. Anatomic resection for severe blunt liver trauma. International Surgery 2005 Nov-Dec. 90(5):266-9.
33. Ginzburg E, Klein Y, Sutherland M, Levi D. Nery JR. Sleeman D. Shatz DV. Lynn M. Goldberg D. Cohn SM. Prolonged clamping of the liver parenchyma: a salvage maneuver in exsanguinating liver injury. J Trauma 2004 Apr. 56(4):922-3. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/68446
34. Poggetti RS, Moore EE, Moore FA, Mitchell MB. Read RA. Balloon tamponade for bilobar transfixing hepatic gunshot wounds. J Trauma 1992 Nov. 33(5):694-7. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/68447
35. Beal SL. Fatal Hepatic Hemorrhage: an unresolved problem in the management of complex liver injuries. J Trauma 1990 Feb. 30(2):163-9. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/68448
36. Caruso DM, Battistella FD, Owings JT, Lee SL. Samaco RC. et al. Perihepatic packing of major liver injuries. Arch Surg 1999 Sep. 134(9):958-62. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/68449
37. Feliciano DV, Pachter HL. Hepatic Trauma revisited. Curr Prob Surg 1989 Jul. 26(7):453-524.
38. Krige JE, Bornman PC, Terblanche, J. Therapeutic perihepatic packing in complex liver truma. Br J Surg 1992 Jan. 79(1):43-6. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/68450
39. Cohn SM, Feinstein AJ. Nicholas JM. McKenney MA. Sleeman
D. Ginzburg E. Shatz DV. Kirton OC. Byers PM. Augenstein JS. Recipe for
Poor man’s Fibrin Glue. J Trauma 1998 May. 44(5):907.
40. Sattler S, Gentilello LM. The liver bag: report of a new technique for treating severe, exsanguinating hepatic injuries. J Trauma 2004 Oct. 57(4):884-6. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/68452
41. Aaron S, Fulton RL, Mays ET. Selective ligation of the hepatic artery for trauma of the liver. Surg Gynecol Obstet 1975 Aug. 141(2):187-9.
42. Lucas CE, Ledgerwood AM. Liver necrosis following hepatic artery transection due to trauma. Arch Surg 1978 Sep. 113(9):1107.
43. Buckman RF, Miraliakbari R, Badellino MM. Juxtahepaitc venous injuries: a critical review of reported management strategies. J Trauma 2000 May. 48(5):978-84. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/68453
44. Chen RJ, Fang JF, Lin BC, Jeng LB. Chen MF. Surgical management of juxtahepatic venous injuries in blunt hepatic trauma. J Trauma 1995 Jun. 38(6):886-90. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/68454
45. Richardson DJ, Franklin GA, Lukan JK. Carrillo EH. Spain DA. Miller FB. Wilson MA. Polk HC Jr. Flint LM. Evolution in the management of hepatic trauma: a 25-year perspective. Ann Surg 2000 Sep. 232(3):324-30. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/68455
46. Fabian TC, Croce MA, Stanford GG. Factors affecting morbidity following hepatic trauma. A prospective analysis of 482 injuries. Ann Surg 1991 Jun. 213(6):540-7. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/68456
47. Cue JI, Cryer Hg, Miller FB, et al. Packing and a planned reexploration for hepatic and retroperitoneal hemorrhage: critical refinements of a useful technique. J Trauma 1990 Aug. 30(8):1007-11. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/68457
48. Noyes LD, Doyle Dj, McSwain NE Jr. Septic complications associated with the use of peritoneal drains in liver trauma. J Trauma 1988 Mar. 28(3):337-46. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/68458
49. Jurkovich GJ, Hoyt DB, Moore FA, . Ney AL. Morris JA Jr. Scalea TM. Pachter HL. Davis JW. Portal triad injuries. J Trauma 1995 Sep. 39(3):426-34. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/68459
50. Feliciano DV. Biliary injuries as a result of blunt and penetrating trauma. Surg Clin N Amer 1994 Aug. 74(4):897-907.
51. Ivatury RR, Rohman M, Nallathambi M, Rao PM. Gunduz Y. Stahl WM. The morbidity of injuries of the extrahepatic biliary system. J Trauma 1985 Oct. 25(10):967-73. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/68460
52. Voyles CR, Blumgart LH. A technique for the construction of high biliary-enteric anstomoses. Surg Gynecol Obstet 1982 Jun. 154(6):885-7.
53. Sheldon G, Lim R, Yee E, et al. Management of injuries to the porta hepatic. Ann Surg 1985. 202:539-44.
54. Fish JC. Reconstruction of the portal vein. Case reports and literature review. Am Surg 1966 Jul. 32(7):472-8.
55. Stone HH, Fabian TC, Turkelson M. Wounds of the portal venous system. World J Surg1982 May. 6(3):335-41.
56. Ivatury RR, Nallathambi M, Gaudino J, Rohman M. Stahl WM. Penetrating duodenal injuries: An analysis of 100 consecutive cases. Ann Surg 1985 Aug. 202(2):153-8. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/68461
57. Cogbill TH, Moore E, Morris J Jr, Hoyt DB. Jurkovich GJ. Mucha P Jr. Ross SE. Feliciano DV. Shackford SR. Distal pancreatrectomy for trauma: A multicenter experience. J Trauma 1991 Dec. 31(12):1600-6. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/70781
58. Stone HH, Fabian TC, Satiani B, Turkleson ML. Experiences in the management of pancreastic trauma. J Trauma 1981 Apr. 21(4):257-62. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/70784
59. Werschky LR. Jordan GL Jr. Surgical management of traumatic injuries to the pancreas. AmJ Surg 1968 Nov. 116(5):768-72.
60. Brownstein MR, Bunting, T, Meyer AA, Fakhry SM. Diagnosis and management of blunt small bowel injury: A survey of the membership of the American Association for the Surgery of Trauma. J Trauma 2000 Mar. 48(3):402-7. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/70785
61. Flint LM, Cryer HM, Howard PA, Richardson JD. Approaches to the management of shotgun injuries. J Trauma 1984 May. 24(5):415-19. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/70786
62. Nagy K, Roberts R, Joseph K. An G. Barrett J. Evisceration after abdominal stab wounds: Is laparotomy required? J Trauma 1999 Oct. 47(4):622-4. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/70787
63. Salim A. Teixeira PG. Inaba K. Brown C. Browder T. Demetriades D. Analysis of 178 penetrating stomach and small bowel injuries. World J Surg 2008 May. 32(3):471-5. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/70788
64. Brundage SI, JurkovichGJ, Grossman DC. Tong WC. Mack CD. Maier RV. Stapled versus sutured gastrointestinal anastomoses in the trauma patient. J Trauma 1999 Sep. 47(3):500-7. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/70789
65. Witzke JD, Kraatz JJ, Morken JM, Ney AL, West MA, Van Camp JM, Zera RT, Rodriguez JL. Stapled versus hand sewn anastomosis in patients with small bowel injury: a changing perspective. J Trauma 2000 Oct. 49(4):660-5. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/70790
66. Kirkpatrick AW, Baxter KA, Simons RK, Germann E. Lucas CE. Ledgerwood AM. Intra-abdominal complications after surgical reapir of small bowel injuries: an international review. J Trauma 2003 Sep. 55(3):399-406. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/70791
67. Brundage S, Kirilcuk N, et al. Complications associated with small bowel resections: Concurrent injuries are more relevant to morbidity than method of Gastrointestinal Anastomosis. J Trauma 60:249, 2006.
68. Timaran CH, Daley BJ, Enderson BL. Role of duodenography in the diagnosis of blunt duodenal injuries. J Trauma 2001 Oct. 51(4):648-51. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/70793
69. Pietzman AB, Makaroun MS, Slasky BS, Ritter P. Prospective study of computed tomography in initial management of blunt abdominal trauma. J Trauma 1986 Jul. 26(7):585-92. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/70794
70. Greenlee T. Murphy K. Ram MD. Amylase isoenzymes in the evaluation of trauma patients. Am Surg 1984 Dec. 50(12):637-40.
71. Harrell DJ, Viatle GC, Larson GM. Selective role for endoscopic retrograde cholaniopancreatography in abdominal trauma. Surg Endosc 1998 May. 12(5):400-4. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/70795
72. Soto JA, Alvarez O, Munera F, Yepes NL. Sepulveda ME. Perez JM. Traumatic disruption of the pancreatic duct: Diagnosis with MR pancreatography. Am J Roentgenol 2001 Jan. 176(1):175-8. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/70796
73. Berni G, Bandyk D, Oreskovich M, et al. Role of intraoperative pancreatography in patients with injury to the pancreas. Am J Surg 1982. 143:602-5.
74. Ascensio J, Feliciano D, Britt L, Kerstein MD. Management of duodenal injuries. Curr Probl Surg 1993 Nov. 30(11):1023-93.
75. McInnis W, Aust J, Cruz A, Root HD. Traumatic injuries of the duodenum: A comparison of primary closure and the jejuna patch. J Trauma 1975 Oct. 15(10):847-53. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/70797
76. Kashuk JL, Moore EE, Cogbill TH. Management of the intermediate severity duodenal injury. Surgery 1982 Oct. 92(4):758-64.
77. Snyder W, Weigelt J, Watkeins W, Bietz DS. The surgical management of duodenal trauma. Precepts based on a review of 247 cases. Arch Surg 1908 Apr. 115(4):422-9.
78. Vaughn GD 3rd, Frazieer OH, Graham DY, Mattox KL. The use of pyloric exclusion in the management of severe duodenal injuries. Am J Surg 1977 Dec. 134(6):785-90.
79. Stone HH, Fabian TC. Management of duodenal wounds. J Trauma 1979 May. 19(5): 334-9. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/70799
80. Jewett TC Jr, Caldarola V, Karp MP, Allen JE. Cooney DR. Intramural hematoma of the duodenum. Arch Surg 1988 Jan. 123(1):54-8.
81. Touloukian RJ. Protocol for the nonoperative treatment of obstructing iintramural duodenal hematoma during childhood. Am J Surg 1983 Mar. 145(3):330-4.
82. Nowak MM, Baringer DC, Ponsky JL. Pancreatic injuries. Effectiveness of debridement and drainage for nontransecting injuries. Am Surg 1986 Nov. 52(11):599-602.
83. Innes JT, Care LC. Normal pancreatic dimensions in the adult human. Am J Surg 1994 Feb. 167(2):261-3. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/70801
84. Oreskovich MR, Carrico CJ. Pancreaticoduodenectomy for trauma: A viable option? Am J Surg 1984 May. 147(5):618-23.
85. Feliciano DV, Martin T, Cruse P, Graham JM. Burch JM. Mattox KL. Bitondo CG. Jordan GL Jr. Management of combined pancreatoduodenal injuries. Ann Surg 1987 Jun. 205(6); 673-80. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/70802
86. Sharma OP, Oswanksi MF, White PW. Injuries to the colon from blast effect of penetrating extra-peritoneal thoracabdominal trauma. Injury 2004 Mar. 35(3):320-4. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/70803
87. Macleod JB, Freiberger D, Lewis F & Feliciano DV. What is the optimal observation time for a penetrating wound to the flank? Am Surg 73(1):25-31. Jan2007 http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/70804
88. Velmahos GC, Demetrades D, Toutouzas KG, Sarkisyan G. Chan LS. Ishak R. Alo K. Vassiliu P. Murray JA. Salim A. Asensio J. Belzberg H. Katkhouda N. Berne TV. Selective nonoperative management in 1,856 patients with abdominal gunshot wounds: Should routine laparotomy still be the standard of care? Ann Surg 2001 Sep. 234(3):395-402. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/70805
89. Maxwell RA, Fabian TC. Current management of colon trauma. World J Surg 2003 Jun. 27(6):632-9. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/70806
90. Stone HH, Fabian TC. Management of perforating colon trauma: randomization between primary closure and exteriorization. Ann Surg 1979 Oct. 190(4):430-6. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/70807
91. Gonzalez RP, Merlotti FJ, Holevar MR. Colostomy in penetrating colon injury: Is it necessary? J Trauma 1996 Aug. 41(2):271-5. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/70808
92. Kamwendo NY, Modivba MC, Matlala NS, Becker PJ. Randomized clinical trial to determine if delay from time of penetrating colonic injury precludes primary repair. Br J Surg 2002 Aug. 89(8):993-8. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/70809
93. Demetriades D, Murray JA, Chan LC, Ordonez C. Bowley D. Nagy KK. Cornwell EE 3rd. Velmahos GC. Munoz N. Hatzitheofilou C. Schwab CW. Rodriguez A. Cornejo C. Davis KA. Namias N. Wisner DH. Ivatury RR. Moore EE. Acosta JA. Maull KI. Thomason MH. Spain DA. Committee on Multicenter Clinical Trials. American Association for the Surgery of Trauma. Penetrating colon injuries requiring resection: Diversion or primary anastomosis? An AAST prospective multicenter study. J Trauma 2001 May. 50(5):765-75. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/70810
94. Lustosa SA, Matos D, Atallah AN, Castro AA. Stapled versus handsewn methods for colorectal anastomosis surgery. Cochrane Database of Systematic Reviews (3):CD003144, 2001. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/70812
95. Burch JM, Francoise RJ, Moore EE, Biffl WL. Offner PJ. Single layer continuous versus two layer interrupted intestinal anastomosis: A prospective randomized trial. Ann Surg 2000 Jun. 231(6): 832-7. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/70814
96. Jesus EC, Karlicek A, Matos D, Castro AA. Atallah AN. Prophylactic anastomotic drainage for colorectal surgery. Cochrane Database for Systematic Revews (4):CD002100, 2004. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/70816
97. Murray JA, Demetraides D, Colson M, Song Z. Velmahos GC. Cornwell EE 3rd. Asensio JA. Belzberg H. Berne TV. Colonic resection in trauma: Colostomy versus anastamosis. J Trauma 1999 Feb. 46(2):250-4. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/70818
98. Velmahos GC, Toutouzas KG, Sarkisyan G, Chan LS. Jindal A. Karaiskakis M. Katkhouda N. Berne TV. Severe trauma is not an excuse for prolonged
antibiotic prophylaxis. Arch Surg 2002 May. 137(5);537-41. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/70819
99. Fabian TC, Croce MA, Payne LW, Minard G. Pritchard FE. Kudsk KA. Duration of antibiotic therapy for penetrating abdominal trauma: a prospective trial. Surgery 1992 Oct. 112(4):788-94.
100. Ivatury RR, Licata J, Gunduz Y, Rao P. Stahl WM. Managment options in penetrating rectal injuries. Am Surg 1991 Jan. 57(1):50-5.
101. Gonzalez RP, Falimirski ME, Holevar MR. The role of presacral drainage in the management of penetrating rectal injuries. J Trauma 1998 Oct. 45(4): 656-61. http://simplelink.library.utoronto.ca.myaccess.library.utoronto.ca/url.cfm/70820
102. Miller FB. Cryer HM. Chilikuri S. Creech P. Richardson JD. Negative findings on laparotomy for trauma. Southern Medical Journal 1989 Oct. 82(10):1231-4.