Read Core Topics in General & Emergency Surgery: Companion to Specialist Surgical Practice Online
Authors: Simon Paterson-Brown MBBS MPhil MS FRCS
Core Topics in General & Emergency Surgery: Companion to Specialist Surgical Practice (63 page)
As yet, there are few guidelines for exactly when surgical decompression is required in the presence of raised IAP. The indications for abdominal decompression are related to correcting pathophysiological abnormalities as much as achieving a precise and optimum IAP. The abdomen is decompressed, and a temporary abdominal closure (TAC) is achieved. The simplest technique is the Opsite® sandwich technique described above.
Every effort must be made towards the reduction of intra-abdominal pressure, allowing early closure of the abdomen. Clinical infection is common in the open abdomen and the infection is usually polymicrobial. Prophylactic antibiotics are not routinely used.
Patients with raised IAP require close and careful monitoring, aggressive resuscitation and a low index of suspicion for requirement of surgical abdominal decompression.
The American Association for the Surgery of Trauma (AAST) has developed an organ injury scaling system for all major injuries in the body.
51
–
57
These scores not only allow a clear anatomical language when dealing with organ injury but also, since surgical policies are often dictated by the injuries sustained, a degree of consistency in both treatment and prognosis. The scaling system is outlined in the Appendix to this chapter and can be accessed at
http://www.aast.org
on the Internet.
Experience in dealing with injury clearly leads to a better outcome:
Good judgment comes from experience.
Unfortunately, experience often comes from bad judgment!
There is limited evidence-based decision-making in trauma, although certain practice guidelines have been developed, primarily by groups working in the Eastern Association for the Surgery of Trauma (
www.east.org
).
58
Examples of evidence-based practice management guidelines include the following.
58
,
59
Guidelines for patient selection and specific antimicrobial regimens are based on good evidence (
Box 13.1
),
59
,
60
those regarding high-risk patients and duration of therapy less so (
Box 13.2
). The trauma surgery setting is fundamentally different to that of elective surgery, especially with regard to prophylactic antibiotics. Antibiotics are given after the peritoneal insult and possible contamination has occurred (
Boxes 13.3
and
13.4
).
Box 13.1
Evidence-based guidelines for prophylactic antibiotic administration: summary of recommendations
Strong recommendation
Patients with peritoneal contamination due to traumatic bowel injury repaired within 12 hours are not considered to have established intra-abdominal infections, and should be treated with prophylactic antimicrobials for 24 hours or less (
Evidence level I
)
Recommendation based on evidence of effectiveness
Systemic antibiotics should be administered as soon as possible after injury for patients with penetrating trauma requiring surgical intervention (
Evidence level II
)
A single broad-spectrum agent is at least as safe and effective as a double or triple antibiotic therapeutic regimen (
Evidence level II
)
Patients with a fully removable focus of inflammation (e.g. bowel necrosis) should be treated with prophylactic antimicrobials for 24 hours or less (
Evidence level II
)
Recommendation made where there is no adequate evidence as to the most effective practice
Patients with more extensive conditions should be treated as having more extensive infections, and given
therapeutic
antimicrobials for more than 24 hours (
Evidence level III
)
Reproduced from Mazuski JE, Sawyer RG, Nathens AB et al. The Surgical Infection Society Guidelines on antimicrobial therapy for intra-abdominal infections: an executive summary. Surg Infect 2002; 3(3):161–73. Cornwell EE IIIrd, Campbell KA. Trauma. In: Gordon TA, Cameron JL (eds) Evidence-based surgery. Hamilton, Ontario: BC Decker, 2000; pp. 415–28. With permission from Mary Ann Liebert Inc./BC Decker Inc.
Box 13.2
Evidence-based guidelines for duration of antibiotic therapy: summary of recommendations
Recommendation based on evidence of effectiveness
High postoperative septic complications can be expected in patients with gunshot injuries to the colon, high Penetrating Abdominal Trauma Index (PATI) scores, major blood loss, and common need for postoperative care.There is no evidence to date that extending prophylactic antibiotic therapy beyond 24 hours decreases that high risk (
Evidence level II
)
Antimicrobial therapy of most established intra-abdominal infections should be limited to no more than 5 days (
Evidence level II
)
Antimicrobial therapy can be discontinued in patients when they have no clinical evidence of infection such as fever or leucocytosis (
Evidence level II
)
Recommendation made where there is no adequate evidence as to the most effective practice
Continued clinical evidence of infection at the end of the time period designated for antimicrobial therapy should prompt appropriate diagnostic investigations rather than prolongation of antimicrobial treatment (
Evidence level III
)
If adequate source control cannot be achieved, a longer duration of antimicrobial therapy may be warranted (
Evidence level III
)
Box 13.3
Evidence-based guidelines for antimicrobial regimens: summary of recommendations
Strong recommendation
Antimicrobial regimens for intra-abdominal infections should cover common aerobic and anaerobic enteric flora. No regimen has been demonstrated to be superior to another (
Evidence level I
)
Recommendation based on evidence of effectiveness
Once-daily administration of aminoglycoside is the preferred dosing regimen for patients receiving these agents for intra-abdominal infections (
Evidence level II
)
Recommendation made where there is no adequate evidence as to the most effective practice
Aminoglycosides should be avoided in the acute trauma patient due to difficulty in reaching adequate MIC levels without toxicity (
Evidence level III
)
For less severely ill patients with community-acquired infections, antimicrobial agents having a narrower spectrum of activity, such as antianaerobic cephalosporins, are preferable to more costly agents having broader coverage of Gram-negative organisms and/or a greater risk of toxicity (
Evidence level III
)
The routine use of intraoperative cultures is controversial, and there is no evidence that altering the antimicrobial regimen on the basis of intraoperative culture results improves outcome (
Evidence level III
)
Box 13.4
Evidence-based guidelines for antimicrobial therapy in the high-risk patient: summary of recommendations
Strong recommendation
In patients with intra-abdominal infections, treatment failure and death is associated with patient-related risk factors such as advanced age, poor nutritional status, low serum albumin concentration and pre-existing medical disease.A higher APACHE II score is the most consistently recognised risk factor for both death and treatment failure (
Evidence level I
)
Recommendation based on evidence of effectiveness
Disease and treatment-related risk factors, including nosocomial origin of infection, the presence of resistant pathogens, and the lack of adequate source control are associated with treatment failure and death (
Evidence level II
)
Routine addition of an aminoglycoside to other agents having broad-spectrum Gram-negative coverage provides no additional benefit (
Evidence level II
)
Addition of empirical antifungal therapy with fluconazole is reasonable for patients with postoperative intra-abdominal infections at high risk of candidiasis (
Evidence level II
)
Recommendation made where there is no adequate evidence as to the most effective practice
Patients at higher risk of failure (e.g. damage control) should be treated with an antimicrobial regimen having a broader spectrum of coverage of Gram-negative aerobic/facultative anaerobic organisms (de-escalation therapy) (
Evidence level III
)
Abdominal injury is of a blunt or penetrating nature.While it is hard to define precise parameters for the approach to management, the following guidelines are offered.
Repair and resection for treatment of hepatic trauma demands a working knowledge of the anatomy of the liver. Segmental anatomical resection has been well documented but usually is not applicable to injury. The three main hepatic veins divide the liver into four sections: right posterior lateral, right anterior medial, left anterior and left posterior. Each of these sectors receives a portal pedicle.The liver is practically divided into eight Couinaud segments (
Fig. 13.4
).
Figure 13.4
Couinaud's segments of the liver.
Understanding the anatomy also helps to explain some of the patterns of injury following blunt trauma. There are also differences in tissue elasticity, which determine injury patterns. The forces from blunt injury are usually direct compressive forces or shear forces. The elastic tissue within arterial blood vessels makes them less susceptible to tearing than any other structures within the liver. Venous and biliary ductal tissues are moderately resistant to shear forces, whereas the liver parenchyma is the least resistant of all. Therefore, fractures within the liver parenchyma tend to occur along segmental fissures or directly into the parenchyma. This causes shearing of lateral branches to the major hepatic and portal veins. With severe deceleration injury, the origin of the hepatic veins may be ripped from the inferior vena cava, causing devastating haemorrhage. Similarly, the small branches from the caudate lobe entering directly into the cava are at high risk of shearing with linear tears on the caval surface.
Direct compressive forces usually cause tearing between segmental fissures in an anterior posterior sagittal orientation. Horizontal fracture lines into the parenchyma give the characteristic burst pattern to such liver injuries. If the fracture lines are parallel, these have been dubbed ‘bear claw’-type injuries and probably represent where the ribs have been compressed directly into the parenchyma. Occasionally, there will be a single fracture line across the horizontal plane of the liver, usually between the anterior and posterior segments. This can cause massive haemorrhage if there is direct extension or continuity with the peritoneal cavity.
The liver is at risk in any penetrating trauma to the right upper quadrant of the abdomen. CECT has enhanced the diagnosis of significant liver injury in the stable patient. Virtually all penetrating injuries to the abdomen should be explored promptly, especially when they occur in conjunction with hypotension. The surgeon should be aware that penetrating injuries to the right lower chest, presenting with haemothorax, may have penetrated the diaphragm, with the bleeding originating from the liver.
The treatment of severe liver injuries begins with temporary control of haemorrhage. Most catastrophic bleeding from hepatic injury is venous in nature and can therefore be controlled by liver packs. If there is bright red blood pouring from the parenchyma, it is then appropriate to apply a vascular clamp to the porta hepatis (Pringle's manoeuvre), via the gastrohepatic ligament. If this controls the bleeding, the surgeon should be suspicious of hepatic arterial or possible portal venous injury. While control is being obtained, it is important to establish more intravenous access lines and other monitoring devices as needed. (The treatment of bleeding is to stop the bleeding!) Hypothermia should be anticipated and corrective measures taken.
After haemostasis and haemodynamic resuscitation has been achieved, any packs in the two lower abdominal quadrants are removed. If there is abdominal contamination, it is appropriate to control this as rapidly as possible. The packs in the left upper quadrant are then removed. If there is an associated spleen injury a decision must be made either to remove it promptly or to clamp the hilum of the spleen temporarily with a vascular clamp to reduce further bleeding. Finally, the packs are removed in the right upper quadrant and the injury to the liver rapidly assessed.
If there is bleeding from the porta hepatis, careful exploration for a portal vein injury should be carried out, with repair or shunting. Traction on the dome of the liver, which produces a sudden gush of retrohepatic blood, should make the surgeon suspicious of injury to the posterior hepatic veins or inferior vena. The options for hepatic vein and cava injuries include direct compression (packing), extension of the laceration and direct control, atrial–caval shunt, non-shunt isolation (Heaney technique) and veno–veno bypass. Liver packs are increasingly used as definitive treatment, particularly when there is bilobar injury, or they can simply buy time if the patient develops a coagulopathy, hypothermia or there are no blood resources.
Blunt pancreatic trauma can be subtle and presentation is often that of associated injuries. The pancreas is a retroperitoneal organ and there may be no intraperitoneal signs. History can be helpful, for example if there is a history of epigastric trauma. The physical examination is often misleading. Amylase and blood count are non-specific, and may be normal; CT is 85% accurate, but does not always help with the grading of the injury, and endoscopic retrograde cholangiopancreatography (ERCP) can be helpful in selected stable patients.
Penetrating pancreatic trauma is usually obvious since the patient will almost invariably have been explored for other injuries.
61
Once the retroperitoneum has been violated it is imperative for the surgeon to do a thorough exploration in the central region. This includes:
an extended Kocher manoeuvre to examine the entire duodenum;
a right medial visceral rotation to examine the back of the head of the pancreas;
division of the ligament of Treitz to examine the front of the pancreas;
division of the gastrocolic omentum to examine the top of the pancreas;
a left medial visceral rotation to examine the anterior and posterior surfaces of the tail of the pancreas as it extends towards the splenic hilum.
Injury to the main pancreatic duct appears to increase the pancreatic-specific morbidity and mortality.
62
Any parenchymal haematoma of the pancreatic head should be thoroughly explored and should include irrigation of the haematoma and adequate drainage with a suction drain (e.g. a Blake® drain).
Injuries to the tail and body of the pancreas can usually either be drained or, if a strong suspicion for major ductal injury is present (Grade III injury), resection can be carried out as a distal pancreatectomy. In young patients, the duct is not usually visualised following trauma and so cannot specifically be tied off.
The injuries that vex the surgeon most, however, are those to the head of the gland, particularly those with juxtaposed injuries or also involving the duodenum. Resection (Whipple's procedure) is usually reserved for those patients who have destructive injuries or those in whom the blood supply to the duodenum and pancreatic head has been embarrassed. This procedure should only be embarked upon as a planned procedure in the stable patient, with consideration being given to appropriate damage control procedures in all others. The remainder are usually treated with variations of drainage and pyloric exclusion. This would include tube duodenostomy and extensive closed drainage around the injury site. Common duct drainage is not indicated.
63
There are limited guidelines for the management of major pancreatic trauma (
Box 13.5
).
Box 13.5
Evidence-based guidelines for diagnosis and management of pancreatic trauma: summary of recommendations
Recommendations made where there is no adequate evidence as to the most effective practice
Delay in the recognition of main pancreatic duct injury causes increased morbidity (
Evidence level III
)
CT scan is suggestive, but not diagnostic of, pancreatic injury (
Evidence level III
)
Amylase/lipase levels are suggestive, but not diagnostic of, pancreatic injury (
Evidence level III
)
Grade I and II injuries can be managed by drainage alone (
Evidence level III
)
Grade III injuries can be managed with resection and drainage (
Evidence level III
)
Closed suction is preferred to sump suction (
Evidence level III
)
Aortic and caval injuries are primarily a problem of access (rapid) and control of haemorrhage. If the surgeon opens the abdomen and there is extensive retroperitoneal bleeding centrally, there are two options. If the bleeding is primarily venous in nature, the right colon should be mobilised to the midline, including the duodenum and head of the pancreas, using a right medial visceral rotation. This will expose the infrarenal cava and infrarenal aorta. It will also facilitate access to the portal vein.
If the bleeding is primarily arterial in nature, it is best to approach the injury from the left. Initial control can be obtained by direct pressure at the oesophageal hiatus or via the lesser sac. Additional exposure can be obtained by simply dividing the left crus of the diaphragm. Exposure includes reflecting the left colon medially and mobilising the pancreas, kidney and spleen to the midline. By approaching the aorta from the left lateral position it is possible to identify the plane of Leriche more rapidly than by approaching it through the lesser sac. The problem is the coeliac and superior mesenteric ganglia, both of which can be quite dense and hinder dissection around the origins of the coeliac and superior mesenteric arteries.
Once isolation of the injury has been achieved, treatment of aortic or caval injuries is usually straightforward, either by direct suture or occasionally grafting. Caval injuries below the renal vein, if extensive, can be ligated, although repair is preferred. Injuries above the renal veins in the cava should be repaired if at all possible. If there is injury to the posterior wall of the vein, it is preferable to isolate the segment and repair it from within the vein using an anterior approach.
Many studies have shown primary repair of colonic injuries to be safe in patients at low risk of postoperative complications; however, identification of the high-risk patient in whom avoidance of an intraperitoneal colonic suture line may be beneficial is still controversial. The significant morbidity and financial costs associated with the creation and reversals of colostomy, and the destructive effect of colostomy on the patient's quality of life, have been cited as evidence to support the primary repair of colonic wounds.
60
,
64
–
66
Stone and Fabian excluded patients from primary repair in the presence of shock, major blood loss, more than two organs injured, faecal contamination more than ‘minimal’, delay to repair of more than 8 hours and wounds of the colon or abdominal wall so destructive as to require resection.
67
Murray et al. reviewed 140 patients with colonic injuries that required resection.
68
They suggested that the majority of patients can safely undergo colonic resection with primary colo-colonic anastomosis, even for severe injuries; however, there is a subgroup of critically injured patients at higher risk of anastomotic leakage who may be best treated by colostomy.
In the ‘physiologically challenged’ patient, hypoperfusion of splanchnic tissues leads to local tissue hypoxia, and repair or anastomosis under those circumstances is more likely to fail. This high-risk group should be treated by damage control techniques with restoration of continuity of the bowel once the physiological insult has been corrected. Outcome is determined by the patient's physiological envelope and not by anatomical integrity
69
,
70
(
Box 13.6
).
Box 13.6
Evidence-based guidelines for management of colon injuries: summary of recommendations
Strong recommendation
Primary repair is supportedfor
non-destructive
(involvement of < 50% of the bowel wall without devascularisation) colon wounds in the absence of peritonitis. The risk is not obviated by colostomy, and intra-abdominal sepsis frequently occurs in the absence of suture line disruption. Accordingly, the majority of colon injuries in civilian practice may be managed by primary repair (
Evidence level I
)
Recommendations based on evidence of effectiveness
Patients with penetrating colon wounds that are
destructive
can undergo resection and primary anastomosis if they:
are haemodynamically stable without evidence of shock
have no significant underlying disease
have minimal associated injuries (PATI < 25, ISS < 25)
have no peritonitis
(
Evidence level II
)
Given the small number of patients in high-risk categories with destructive colon injuries requiring resection who were randomised to primary repair, there is still room to consider colostomy in the management of these patients (
Evidence level II
)
Patients with shock, underlying disease, significant associated injuries or peritonitis should have destructive wounds managed by resection and colostomy (
Evidence level II
)
Colostomies performed following colon and rectal trauma can be closed within 2 weeks if contrast enema is performed to confirm distal healing. This recommendation pertains to patients who do not have non-healing bowel injury, unresolved wound sepsis or are unstable (
Evidence level II
)
A barium enema should not be performed to rule out colon cancer or polyps prior to colostomy closure for trauma in patients who otherwise have no indications for being at risk (
Evidence level II
)
Multiple blood transfusions, shock and a high PATI reliably identify patients at high risk of septic complications following penetrating colon injuries (
Evidence level II
)
Reproduced from Cayten CG, Fabian TC, Garcia VF et al. Patient Management Guidelines for penetrating intraperitoneal injuries. EAST Practice Parameter Working Group;
http://www.east.org
.