The majority of trauma patients require standard resuscitation and clinical decision-making. Improvements in pre-hospital care and resuscitation have increased the number of severely-injured patients reaching hospital alive. There is a small group of patients with severe injuries, which are accompanied by hypovolaemic shock leading to the physiological derangement of hypothermia, acidosis, and coagulopathy. A conventional management approach is associated with high mortality in this group of patients. These patients require expedient and intensive concomitant resuscitation and surgery to reduce morbidity and mortality. This is when a damage control approach is appropriate, analogous to that described by the US navy in relation to battleships: ‘… the capacity of a ship to absorb damage and maintain mission integrity’.
The primary aim of damage control is to restore normal physiology before normal anatomy. Damage control surgery is defined by an abbreviated resuscitative procedure consisting of rapid control of haemorrhage and contamination, temporary closure, resuscitation to normal physiology in an intensive care unit, then return to the operating theatre for definitive repair. The initial surgery is limited to enable the patient to reach the intensive care unit alive, reduce complications, and live to fight another day.
Aggressive correction of hypothermia, coagulopathy, and acidosis are essential to reduce the morbidity and mortality associated with post-traumatic multi-organ failure.
Four stages of damage control have been described:
• Stage 1: recognition and resuscitation.
• Stage 2: initial operation.
• Stage 3: restoration of physiology.
• Stage 4: definitive surgery.
Stage 1: recognition and resuscitation
Pre-hospital management in severely-injured, bleeding patients should be prompt, and on scene times short, in an effort to reduce the time to definitive surgical haemorrhage control. Damage control is aimed at reducing the incidence of the lethal triad of acidosis, hypothermia, and coagulopathy. Once this lethal triad is established, haemorrhage control and correction of coagulopathy become more difficult.
The patient requiring damage control is identified by a combination of their mechanism of injury, and a clinical presentation of severe or progressive shock. Once the patient with severe bleeding is identified, the decision to perform damage control surgery must be made early, and must be communicated rapidly to the resuscitation and operating team. They should then move the patient to theatre immediately, expedite preparation, and facilitate abbreviated surgery with simultaneous resuscitative transfusion.
When the source of bleeding is clinically apparent then no investigations are required. When the possibility of multiple sources of bleeding are present, simple investigations to help decision-making, regarding which body cavity to enter first include chest radiograph, pelvic radiograph, FAST scanning, and insertion of intercostal drains if massive haemothorax is suspected. If a major pelvic fracture is present, then a pelvic belt or sheet should be applied to help stabilize the bones and provide a counter weight if pelvic packing is subsequently required. These manoeuvres should be immediately available in the resuscitation room, and should be carried out as the patient is being prepared for movement to theatre. They should not delay the required surgery. To achieve this, well organized systems for trauma care are required.
Damage control resuscitation
Damage control resuscitation begins alongside initial assessment, and continues into the operating theatre and intensive care unit. Large volume fluid resuscitation following haemorrhagic trauma is associated with a transient rise in blood pressure, which may remove the clot, cause dilutional coagulopathy, dilutional anaemia, and further cooling. There is also an increased risk of pulmonary complications and multi-organ failure.
The evidence for permissive or hypotensive resuscitation in penetrating trauma, although not conclusive, is more compelling than that for blunt injury. Resuscitation targets are controversial; for penetrating injuries it is reasonable to aim at a systolic blood pressure associated with cerebration or between 70–90 mmHg, until haemostasis is achieved. It may be reasonable to aim for similar figures in blunt trauma for short periods if the patient is to be transferred to theatre or angiography and there is no associated traumatic brain injury.
The benefits of hypovolaemic resuscitation must be balanced against the risks of organ hypoperfusion and the resuscitation targets adjusted if there is no rapidly identifiable surgically correctable bleeding point. Once haemostasis has been achieved, resuscitation aims to restore tissue perfusion, of which there is currently no ideal measure.
The treatment of coagulopathy has evolved with the greater understanding of the early coagulopathy associated with trauma. For the 10% of patients who are the most severely-injured, blood and plasma have been found to be the best resuscitation fluids. By using point of care testing, warming devices, multiple blood products, and haemostatic adjuncts, all efforts are directed towards normalizing clotting, base deficit, and temperature. Clot preservation is maximized with hypotensive resuscitation, and a ratio of blood to blood products of 1:1 has been found to be associated with reduced mortality in the most severely injured. Fresh frozen plasma and packed red blood cells should be administered in approximately equal amounts during a damage control approach to resuscitation. Having a massive transfusion protocol in place ensures that blood and products are available as soon as possible.
Stage 2: initial operation
Patients with major exsanguinating injuries will not survive complex procedures since intra-operative metabolic failure, rather than failure to complete operative repair, will result in early mortality. The operating theatre can be considered to be a physiologically unfavourable environment and, therefore, treatment of abnormal physiology should be considered as more important than anatomical correction at this stage.
Communication between the surgeon and anaesthetist regarding monitoring, resuscitation, and operative findings is vital. Good vascular access should be secured, adequate suction available, and plenty of large opened abdominal packs should be to hand. The team should be aware of the clock—true damage control must not take longer than 90 min. This is not the time for complex hepatic resections or pancreatic reconstructions; simple manoeuvres are required to achieve the goals of damage control surgery:
• Haemorrhage control.
• Protection from further injury.
• Prevent contamination.
Adequate exposure is essential to ensuring damage control surgery can be performed correctly and expediently. The worst case scenario should always be prepared for, as this will save crucial time in a life-threatening situation. The patient should be positioned by the surgeon prior to scrubbing up: supine crucifix position with both upper limbs out on arm boards and the skin prepared from clavicles to knees. Patients in severe shock or with multi-system injury may require multiple vascular access and this is facilitated by having the arms in this position, still allowing full access to all body cavities and major limb vessels for proximal control. In multi-system injuries or when junctional injuries are present, the initial incision should be chosen to deal with what is considered to be the greatest threat to life first. If this decision is wrong, it may be necessary to change tack immediately.
The peritoneal cavity should be entered through a long midline incision (xiphisternum to pubis) with a scalpel. Preperitoneal fat is thinnest just cranial to the umbilicus and makes an ideal place to enter the cavity and often can be breached with finger dissection alone. Using a pair of heavy scissors the peritoneum can be opened the full extent of the incision. Beware of iatrogenic injury to the left lobe of the liver, small bowel, and bladder.
The hypotensive trauma patient requires immediate control of large volume intra-abdominal haemorrhage, rather than diathermy of tiny bleeding vessels. Suction should be readily available and the first manoeuvre should be to remove all small bowel loops to allow directed abdominal packing. The easiest way to achieve this is by placing large abdominal packs in the following areas:
• Right upper quadrant: a hand should be placed over the right and left domes of liver. Packs should be placed over the hand above the liver. The sub-hepatic area should be packed to form a ‘sandwich’.
• Right paracolic gutter.
• Left upper quadrant: a hand should be placed above the spleen, which should be pulled gently forward and then a pack applied over the hand and then medially.
• Left paracolic gutter.
The packs should be left in place to allow volume resuscitation to take place before removal. The temptation to ‘pack and peek’ (a dangerous cycle of repeated packing, bleeding and resuscitation) should be avoided. After packing, packs should be removed using a system starting in the area with the least amount of bleeding, working towards the likely site of bleeding. Empirical abdominal packing will arrest all but major arterial bleeding, and will permit a more thorough inspection of the peritoneal cavity.
Following empirical packing the priority is to identify the source of haemorrhage. If packing has failed to control major bleeding, proximal aortic control may be necessary. This is achieved by opening up the lesser omentum, followed by blunt dissection at the diaphragmatic hiatus onto the supracoeliac aorta. Digital pressure is an effective manoeuvre and often safer than formal cross clamping when in extremis.
The aim at this stage is to formally arrest the haemorrhage. Strategy is dependent on the organ. The commonest sites of bleeding are from the liver, spleen, and mesenteric vessels.
During damage control procedures, bleeding from splenic injuries should be treated by splenectomy. There is no place for splenic preservation attempts in patients at the limits of their physiological reserve. Following splenectomy the splenic bed should packed to prevent continuous post-operative oozing of blood.
If a splenic injury is present, but not bleeding, then packing this off may be helpful to prevent re-bleeding prior to the planned re-laparotomy.
If the initial packing controls the liver bleeding, this should be formalized. The initial packs should be removed to identify the site of the liver injury. The assistant should then be positioned so as to compress the liver into its original anatomical shape to control bleeding. The falciform ligament and triangular ligaments should be divided to enable formal packing of the involved lobe to maintain its anatomical shape with packs placed superiorly, inferiorly, and anteriorly. In the case of the right lobe, packs may also be placed laterally.
If initial packing has not controlled the bleeding, a Pringle manoeuvre should be performed by manual compression of (or placing an arterial clamp across) the porta hepatis. If this controls the bleeding, it suggests that the hepatic artery or one of its branches has been injured. There are two approaches to this situation, and the decision taken is dependant on the resources immediately available:
• If there is early access (within 30 min) to interventional angiography either on table or in the radiology suite, the Pringle manoeuvre should be maintained and compression, mobilization of the liver and formal packing should take place as above. The patient should then have other sources of bleeding controlled before proceeding directly to interventional angiography with a view to embolization of the hepatic artery and its branches. The clamp on the porta hepatis will need to be removed as close to the time of embolization as possible.
• If immediate access to interventional radiology is not possible, the porta hepatis should remain clamped whilst the packs are removed. The liver should then be compressed and mobilized as previously. The depth of the liver injury is examined whilst the clamp is temporarily released, in an attempt to identify the arterial injury and control it by suture ligation. To minimize blood loss the number of attempts and time taken to achieve this should be limited. In between attempts, the porta hepatis should be re-clamped and the liver should be compressed manually. If successful the liver should again be compressed and packed formally. If unsuccessful in gaining rapid control of liver arterial bleeding then the liver should be compressed and packed followed by formal ligation of the hepatic artery in the porta hepatis.
If initial packing and a Pringle manoeuvre have not controlled bleeding, it is likely that an intra-hepatic caval or hepatic vein injury are the cause. Heroic surgery is usually not possible due to the rapid blood loss and high chance of exsanguination in a short space of time. The triangular ligaments should not be divided and the liver should not be mobilized, as this will open up further potential spaces and remove constraints to bleeding formed by the diaphragm and the triangular ligaments. The liver should be pushed back posteriorly to compress the inferior vena cava and the hepatic veins, which should slow the bleeding. The anaesthetist should be informed prior to carrying out this manoeuvre, so they are prepared for the decreased cardiac return and subsequent cardiac output in a patient with an already compromised circulation. Packs should now be placed anteriorly and inferiorly to maintain this position of the liver compressing the site of haemorrhage.
Bleeding from knife or bullet tracks in the liver are best managed by balloon tamponade. The easiest way to achieve this is by inserting a Sengstaken tube into the track and inflating the balloon with saline until the bleeding is arrested. This can then be left in situ until the subsequent laparotomy.
Mesenteric vessel injury
Injuries to these vessels are managed by clamping and ligation or direct suture ligation. If the viability of the bowel supplied by the injured vessels is doubtful, this is not the time for resection and repair. The bowel either side of the doubtful segment should be stapled off or ligated to reduce possible contamination.
Before entering a retroperitoneal haematoma it is helpful to achieve vascular control first. For a central haematoma the aorta can be controlled at the hiatus as previously described. For lateral haematomas (likely to be due to renal bleeding), if possible control of the appropriate renal artery and vein should be obtained prior to entering the haematoma.
Central retroperitoneal haematoma
If the haematoma is central then the approach is dictated by the most likely source of the bleeding. If it is thought to be arterial, the approach should be via left medial visceral rotation, which will expose the full length of the abdominal aorta and its branches. If major venous bleeding is thought to be the most likely source then right medial visceral rotation should be performed to expose the inferior vena cava and its tributaries.
Once exposed, the source of bleeding should be identified and controlled initially digitally, and then by clamp or control with balloon catheters (Fogarty or Foley depending on vessel size) if arterial. If venous it is usually safer to compress either side of a major venous injury with a swab held in a sponge holding forceps.
A judgement then needs to be made as to the ease of formally controlling the bleeding. If the injured vessel has a simple laceration, direct suture repair is appropriate. If complicated repairs or interposition grafts are required then a choice of inserting a shunt, or ligation is most appropriate.
In general the following vessels can be ligated:
• Coeliac axis and main branches.
• Inferior mesenteric artery.
• Internal iliac arteries.
• Renal arteries.
• All major named veins.
If the superior mesenteric artery trunk is injured and a complex repair is required, flow needs to be preserved and shunting is appropriate. If one or more of the branches is injured then these can be ligated and the affected segment of bowel will need to be isolated by stapling or ligating either end of the affected segment.
The common and external iliac arterial flow should be maintained if possible by simple repair or shunting. In extremis, they can be ligated and a plan made for an early return to theatre within 6 h to complete a definitive repair to restore blood supply to the involved leg. Fasciotomies of the involved leg should be performed at the initial operation.
Lateral retroperitoneal haematoma
A lateral haematoma that is not expanding should be packed off, since entering the haematoma is likely to end in a nephrectomy. Angiography with a view to embolization is appropriate following the operation.
An expanding lateral haematoma is usually due to a renal injury. If this requires surgery for control, a nephrectomy should be expected, as there is no place for renal preservation in extremis. Before entering the haematoma, the presence of a kidney on the contra-lateral side should be confirmed. Ideally, control of the renal artery on the affected side should be gained by approaching the anterior of the aorta by mobilizing the fourth part of the duodenum to expose the aorta at the level of the renal arteries. The left renal vein traversing the aorta marks the level of the renal arteries.
If the haematoma is encroaching across the aorta an alternative approach is to use medial visceral rotation from the opposite side leaving that kidney in situ to access the aorta.
Once the renal artery is controlled, the colon should be mobilized from the haematoma, and the haematoma and perinephric tissues should be entered to gain access to the kidney. If control of the renal vessels is difficult then a direct approach similar to that for splenectomy is appropriate. The perinephric fascia is entered and the kidney is held in one hand to compress the bleeding and is rapidly mobilized. The renal artery and vein should be double-ligated. The ureter should be located and ligated, and the kidney removed. The renal bed should be packed to reduce continued blood loss.
On occasion a lateral retroperitoneal haematoma is due to severe soft tissue injury of the lateral and posterior abdominal wall. In this situation the affected area should be packed off and post-operative angiography, and embolization considered if packing does not give complete control.
Ideally a pelvic fracture will have been identified and a pelvic sling applied pre-operatively. The pelvis should be entered in the pre-peritoneal plane anterior to the lower abdominal peritoneum. This gives direct access to the true pelvis and this should be packed. Consideration should be given to proceeding to angiography and embolization if bleeding from the pelvis is not completely controlled by packing.
Subsequent abdominal exploration must be thorough, systematic, and complete to include inspection of:
• Inframesocolic compartment: transverse colon to rectum (careful attention should be paid to the posterior surface as injuries here are often missed).
• Supramesocolic compartment: liver, gallbladder, right kidney, stomach (up to oesophagogastric junction), duodenum (Kocher’s manoeuvre is required to visualize fully), spleen, left kidney, both hemidiaphragms.
• Lesser sac (blunt dissection through greater omentumbetween stomach and transverse colon): posterior wall of stomach, body and tail of pancreas.
Hollow organ injuries
Simple bowel injuries can be repaired with a single layer of sutures, or skin staples with bowel mucosal inversion can be utilized to achieve this quickly. With multiple perforations, or damage to a significant length of bowel, control of contamination is best achieved through bowel interruption without resection using a linear stapler or cotton tape to ligate the bowel proximal and distal to the injury. Stomas do not need to be fashioned during the first procedure and should be avoided as the abdominal wall is likely to become oedematous placing the stoma at risk. Primary anastomosis should be avoided and reserved for a later date when bowel wall oedema has subsided and gut perfusion can be assured to reduce the risk of anastomotic failure.
Biliary contamination is reduced by drainage either by placing large drains to the area of injury, placing a t-tube in an easily accessible common bile duct injury, or placing a cholecystomy tube in a gall bladder injury. Injuries to the pancreas should be drained. Bladder injuries can be drained or primarily repaired if limited in size.
Abdominal compartment syndrome is common in these patients secondary to massive fluid shifts and bowel oedema. Most patients, especially those with packs in situ, will require a re-look procedure and, therefore, a form of temporary closure is the most appropriate technique. Laparostomy with saline bag cover or vacuum-pack closure devices are both effective in containing and protecting abdominal viscera. The advantage of the vacuum dressing is that it collects intra-abdominal fluid and keeps the patient dry, but unlike the saline bag does not afford visualization of the bowel to assess viability. A formal note of all packs, shunts, and instruments needs to be included on the operation note.
The main principle in patients with vascular trauma is life before limb preservation. Injuries to major vessels may be compressible or non-compressible depending on the body region injured. For compressible injuries usually involving the limbs or zone 2 of the neck, temporary proximal or direct control can be carried out by pressure or by tourniquet. Recent use of tourniquets in current conflicts in the Middle East has given favourable results in single and multiply-injured patients. Incompressible injuries need to be quickly recognized and surgical control rapidly achieved. Temporary control of incompressible bleeding at the root of the neck and in the groin may be possible with balloon tamponade by inserting a foley catheter in a wound track and inflating the balloon.
Formal control of bleeding due to vascular injuries requires proximal and distal control. Incisions should be longitudinal and extensile. A combined cavity and limb approach may be required for junctional injuries; for example, laparotomy for iliac vessel control followed by longitudinal groin and upper thigh incisions for femoral vessel control for injuries involving the groin, or median sternotomy followed by neck extension for carotid injuries in zone 1 of the neck.
The damage control procedures available to the surgeon include simple direct repair, ligation, and shunting. Direct repair should be reserved for simple lacerations that can be sutured without compromising the lumen of the vessel. The choice of shunt is dependant on availability and vessel size. If commercial shunts are not available then appropriate diameter and lengths of infusion giving sets, naso-gastric feeding tubes and thorocostomy tubes may be appropriate. Prior to insertion the shunt should be clamped in the middle. They should be inserted gently and under vision so as not to raise an intimal flap. The distal end should be inserted first as this is the most likely source of an intimal flap. Once positioned the shunt clamp should be released to ensure back flow, and then the proximal end inserted. They can be secured in position by nylon tapes before finally releasing flow.
Complicated primary repairs, such as vein patching and interposition grafts should be reserved for those with extensive vascular surgical experience and in patients with sufficient physiological reserve, and an otherwise limited injury burden. Distal fasciotomy should be carried out readily and pre-emptively to avoid limb compartment syndrome.
This may be performed as a stand-alone procedure, or combined with a laparotomy or vascular procedure. In these cases the patient is usually supine and so traditional thoracotomy approaches are limited. If it is certain that the thoracic injury is limited to the heart and proximal great vessels then a median sternotomy can be performed alone or as an extension of a laparotomy.
More usually the requirement for thoracotomy becomes apparent due to large volumes of blood in one or both pleural cavities. In this situation the simplest and most expedient approach is via an anterior-lateral approach, which can readily be extended across the sternum using a heavy pair of scissors to a clamshell incision, which has unrivalled access to thoracic structures. The pericardium should be inspected and opened to exclude tamponade. The pleural cavities should be examined to assess the source of the greatest threat to life and this should be dealt with first. Proximal vascular control of great vessel injuries can be carried out either by manual pressure or instrumentation.
Resuscitative thoracotomy is required when the patient is peri-arrest due to cardiac tamponade. It should be considered for all penetrating wounds to the chest, epigastrium, and the root of the neck. The most straightforward approach is via a left fifth intercostal space anterior thoracotomy and then the pericardium should be opened longitudinally, avoiding the phrenic nerve, to relieve the tamponade. Depending on the position of the cardiac wound the incision may be adequate to open up the pericardium further to gain better access to the wound, or if better access is required this can be achieved by extending to a clamshell thoracotomy.
Temporary manoeuvres to control cardiac wounds include digital pressure and foley catheter insertion with gentle traction once the balloon is inflated in the cardiac chamber (remembering to occlude the lumen to prevent blood loss). Depending on the site of the wound it may be repaired by simple suturing, or if the wound is adjacent to a coronary artery then horizontal mattress sutures should be used to avoid incorporation of the coronary artery. If a coronary vessel is injured and the heart is still viable, simple ligation is the best course of action.
The posterior of the heart should be examined by gentle lifting to examine for a through and through injury. The anaesthetist needs to be forewarned when this manoeuvre is carried out.
In instances of massive blood loss or an air leak that precludes adequate oxygenation, damage control procedures should be carried in preference to formal repair or resection. The initial step should be to gain access to the lung hilum by dividing the inferior pulmonary ligament, taking care not to injure the inferior pulmonary vein. Hilar control can most easily be achieved by squeezing digitally. Other options for hilar control include tape ligation, lung twist, or clamping. Once hilar control is achieved, the lung can be assessed for the degree of injury and options for management decided upon.
Simple linear surface lacerations can be managed by suture ligation of obvious bleeding vessels and air leaks. Through and through injuries following penetrating trauma should be managed by tractotomy to gain access to the bleeding vessels along the course of the wound track. A linear stapler is used to open up the length of the track, followed by suture ligation of the bleeding vessels and leaking airways. If a stapler is not available the same procedure can be performed by opening up the track between two straight vascular clamps, suturing vessels in the base and under-running the lung tissue controlled with the clamps.
If sections of a lung lobe are severely injured these areas can be controlled by performing non-anatomical resections with a linear stapler or applying vascular clamps and under-running the resected margins.
The hilar control is released once the repair has been completed, and the resection margins and sutured wounds should be inspected for active bleeding or large air leak. If present, these require further simple sutures for control.
If the injury involves the majority of a lung, then control of the hilum should be maintained and a decision made as to whether to proceed to immediate pneumonectomy, or to leave this until further resuscitation has been achieved and the patient is in a better physiological condition. If there are other injuries present, delaying this is usually the best choice.
Chest wall injury
Following penetrating injuries it is easy to overlook a lacerated intercostal or internal thoracic artery injury, particularly when shock is present. A point should be made to inspect for these injuries along the course of the wound track and ligate them if an injury is identified.
Massive bleeding from a major blunt chest wall injury can be very difficult to deal with. Bleeding can be from the ribs, thoracic vertebrae, muscles, or intercostal vessels. The best approach is to pack off the injured area and apply pressure to control the bleeding, and allow resuscitation with blood and clotting factors. If bleeding is controlled then the packs should be left in situ.
If bleeding is not controlled, the packs should be gently removed to inspect for active arterial bleeding from muscular branches or intercostals arteries. If identified they should be ligated. Once all easily controlled arterial bleeding has been addressed the injured area should be packed again and the lung replaced in the chest to help provide further tamponade. Post-operative selective embolization (if readily accessible) is worth considering.
Injuries involving the trachea are best dealt with by inserting a tracheal tube into the site of injury or if lower down by passing a tracheal tube across the defect. For proximal bronchial defects, one lung ventilation achieved by passing the endotracheal tube into the uninjured main bronchus can be life-saving. If difficult due to bleeding and location, this can be enabled by a joint approach by the anaesthetist and operating surgeon. Injuries of the main bronchial branches, or more distally in the lung hilum, are best dealt with by hilar clamping or non-anatomical resections when the patient is in extremis.
These should be sought with penetrating injuries involving the posterior mediastinum and with severe blunt injury. Control of contamination is obtained by placement of suitable large drains and spillage can be prevented by insertion of a large bore T-tube in the oesophageal defect.
Prior to closure, chest, and pericardial drains should be placed as appropriate. Rapid closure of a thoracotomy can be achieved by interrupted sutures around adjacent ribs and mass closure of the muscles. Take care to ensure that injured or divided inter-costal and internal thoracic arteries are identified and ligated as severe bleeding can occur post-operatively if these are not ligated, once the circulation is restored.
If thoracic closure is not possible because of packing or a distended heart following repair, temporary closure can be achieved using a plastic bag technique similar to that for laparotomy.
In patients with pelvic and long bone fractures with associated other injuries (particularly head, chest, and abdominal), a damage control approach to fracture fixation leads to better outcomes. This is likely to be due to the lesser physiological insult to the patient at this critical time when they are already close to their physiological limit. Temporary fixation is obtained expeditiously by external fixation with less operating time and lower blood loss. Associated wounds are cleaned and debrided.
Initial control of bleeding from severe and mangled limb injuries can be achieved by application of a tourniquet. This can be applied pre-hospital by trained personnel, in the emergency department or in the operating theatre. The identification and ligation of major vessels can then take place. Initial soft tissue debridement should be kept to devitalized or heavily contaminated tissues, while maintaining length of skin and muscle flaps. The wounds should be left open for later review. On release of the tourniquet, it should be ensured that bleeding from larger vessels is controlled by ligation and bleeding from the soft tissues is controlled by compression bandages.
Stage 3: restoration of physiology
The patient should be moved to the intensive care unit once all life-threatening bleeding has been controlled. The circulation should be restored rapidly to normal volumes with a combination of blood, blood products (including plasma, platelets and cryoprecipitate), and crystalloid solutions. Crystalloid use should be restricted to that required to make up for using packed red blood cell transfusions and daily fluid requirements so as to minimize pulmonary complications, reduce bowel oedema, and the risk of abdominal compartment syndrome. The restoration of normal coagulation should be guided by laboratory testing and may be aided by thromboelastography. During this resuscitative process, fluids should be warmed, and warming blankets applied to restore normal temperature.
If a secondary survey has not been completed this should be performed at this stage. The timing of further imaging is dependent on the physiology of the patient and the degree of urgency required for the treatment of the suspected injury. Transfers to the radiology department should be kept to a minimum, as any transfer is a further challenge to the physiology of the patient. If all bleeding is controlled and a major head injury is suspected, this would justify an early move for CT scanning, compared with a suspected spinal injury or simple fracture, both of which can be managed with a period of immobilization without major consequences for the patient.
A tertiary survey needs to be performed in the first 48 h to identify any missed injuries. A careful plan then needs to be made for the management of all the identified injuries and for the completion of surgery.
Stage 4: return to theatre
An unplanned return to theatre may be required before the patient has been fully resuscitated if there is ongoing bleeding, re-bleeding, or thrombosis or dislodgement of a vascular shunt. In these circumstances, a damage control approach should again be used.
If the initial packing has failed to control bleeding, a further attempt is justified. The original packs should be removed and an assessment made as to whether they were sited appropriately, the organ involved was not compressed adequately or the bleeding was arterial in origin. In the former situations, an attempt at further packing has a good chance of success. In the case of arterial bleeding, if ligation is a straightforward option then this should be performed. If access is difficult or is likely to be prolonged, for example, with liver injuries, then packs should be replaced and the patient taken for selective embolization.
If a shunt has been dislodged this can be re-sited, or if the patient is in a good physiological condition and an appropriate surgeon is present formal repair can be considered at this time. If a shunt has thrombosed and the limb is threatened a similar approach can be taken depending on the condition of the patient and surgeon availability.
The return to theatre is ideally on a planned basis and should be carried out by the best available surgical teams. One or more procedures may be carried out during this session depending on the injury burden and the physiological reserve of the patient.
The time for this procedure varies between 6 and 72 h following the initial surgery, depending on the patient and the original injuries. For example, a shunted vascular injury, or cross-clamped or twisted lung hilum would justify an earlier return than a well controlled packed liver injury. For an early planned return this should occur once the patient’s temperature is greater than 35°C, the serum lactate is less than 2.5 mmol/L, and there is normal coagulation by laboratory testing or thromboelastography. For surgery at 24 h or later, the patient’s physiology should be as close to normal as possible.
Blood, plasma, platelets, and cryoprecipitate should be available for the re-operation. The temporary closure should be removed and an assessment made as to a plan of attack. Definitive repairs that are straightforward and easily accessible should be performed prior to pack removal, as they would not get done if severe bleeding were to reoccur on pack removal. Examples include a limb vascular repair, or small bowel anastomosis prior to removal of abdominal packs. If access is difficult with packs in situ then they should be removed. Packs should be removed following soaking in saline, by gently teasing them apart and removing them. Packs should be counted against those recorded as being inserted at initial operation.
Once the packs have been removed, the packed area should be inspected for bleeding. Frequently, minor surface bleeding occurs and this may be controlled by a short period of direct pressure. If a vessel is identified as the source it can be ligated. Occasionally, massive bleeding recurs and, in this situation, the packs should be replaced and left for a further period. The development of a subsequent plan of attack may be aided by angiography and CT investigation.
Bowel continuity should be restored at the first re-operation if possible. Colostomies and ileostomies should be avoided, particularly if abdominal closure is unlikely due to bowel oedema or re-packing.
Any dead or doubtful tissue should be debrided, and abdominal or thoracic closure should be attempted. If closure of the abdomen is not possible, it may be attempted again after a further 48 h, to allow further reduction in oedema and swelling. If closure is not achieved after this, then the wound should be allowed to granulate and a split skin graft be applied. If skin closure only is possible, this can be performed. In both situations a plan for formal abdominal wall reconstruction can be made for 9–12 months later.
Long bone injuries can be definitively fixed and skin cover achieved if appropriate at this stage. For severe soft tissue injuries and traumatic amputations further surgery may be required before formal wound closure is achieved.
Complications of damage control include sepsis and multi-organ failure, wound infections, intra-abdominal abscess, and enterocutaneous fistulae. Abdominal compartment syndrome may occur even when a temporary abdominal closure has been used. Complications should be managed on their merits, taking into account the patient’s condition at the time they present.
Bickell WH, Wall MJ Jr, Pepe PE, Martin RR, Ginger VF, Allen MK, et al. Immediate versus delayed fluid resuscitation for hypotensive patients with penetrating torso injuries. N Engl J Med 1994; 331: 1105–9.Find this resource:
Borgman MA, Spinella PC, Perkins JG. The ratio of blood products transfused affects mortality in patients receiving massive transfusions at a combat support hospital. J Trauma 2007; 63: 805–13.Find this resource:
Brohi K, Singh J, Heron M, Coats T. Acute traumatic coagulopathy. J Trauma. 2003; 54: 1127–30.Find this resource:
Burch JM, Moore EE, Moore FA, Franciose R. The abdominal compartment syndrome. Surg Clin North Am 1996; 76: 833–42.Find this resource:
Roberts CS, Pape H-C, Jones AL, Malkani AL, Rodriguez JL, Giannoudis PV. Damage control orthopaedics. evolving concepts in the treatment of patients who have sustained orthopaedic trauma. J Bone Jt Surg Am 2005; 87: 434–49.Find this resource:
Dutton RP, Mackenzie CF, Scalea TM. Hypotensive resuscitation during active hemorrhage: impact on in-hospital mortality. J Trauma 2002; 52(6): 1141–6.Find this resource:
Giannoudis PV, Pape HC. Damage control orthopaedics in unstable pelvic ring injuries. Injury 2004; 35(7): 671–7.Find this resource:
Hirshberg A, Mattox. KL Top Knife—the Art and Craft of Trauma Surgery. Shrewsbury: TFM Publishing Ltd, 2005.Find this resource:
Hirshberg A, Mattox KL. 'Planned reoperation for severe trauma'. Ann Surg 1995; 222: 3–8.Find this resource:
Holcomb J. Damage control resuscitation. J Trauma 2007; 62: S36–7.Find this resource:
Johnson JW, Gracias VH, Schwab CW, Reilly PM, Kauder DR, Shapiro MB, et al. Evolution in damage control for exsanguinating penetrating abdominal injury. J Trauma 2001; 51(2): 261–71.Find this resource:
Ketchum L, Hess JR, Hiippala S. Indications for early fresh frozen plasma, cryoprecipitate, and platelet transfusion in trauma. J Trauma 2006; 60: S51–8.Find this resource:
Moore EE, Thomas G. Staged laparotomy for the hypothermia, acidosis and coagulopathy syndrome. Am J Surg 1996; 172: 405–10.Find this resource:
Pape HC, Hildebrand F, Pertschy S, Zelle B, Garapati R, Grimme K, et al. Changes in the management of femoral shaft fractures in polytrauma patients: from early total care to damage control orthopedic surgery. J Trauma 2002; 53(3): 452–61.Find this resource:
Rotondo MF, Schwab CW, McGonigal MD, Phillips GR, Fruchterman TM, Kauder DR, et al. 'Damage control—an approach for improved survival in exsanguinating penetrating abdominal injury'. J Trauma 1993; 35: 375–82.Find this resource:
Shapiro MB, Jenkins DH, Schwab CW, Rotondo MF. Damage control: collective review. J Trauma 2000; 49: 969–78.Find this resource:
Smith LA, Barker DE, Chase CW, Somberg LB, Brock WB, Burns RP. Vacuum pack technique of temporary abdominal closure: a four year experience. Am Surg 1997; 63: 1102–7.Find this resource: