Injury, Int. J.

Care Injured (2004) 35, 713—722

Damage control surgery and intensive care

Michael J.A. Parra,b,*, Tareq Alabdia

a
Department of Intensive Care, Liverpool Hospital, Sydney, Australia
b
University of New South Wales, Sydney, Australia

KEYWORDS Summary The introduction and establishment of the ‘damage control surgery’ con-
Damage control surgery cept has led to increasing numbers of severely injured and unstable patients being
(DCS); presented to Intensive Care Units (ICU) for ongoing resuscitation. These patients
Hypovolaemia; present many challenges for the Intensive Care team and emphasise the need for a
Intensive Care Unit multidisciplinary approach to optimise trauma patient management. Multiple issues
need to be addressed simultaneously while the overall aim is to rapidly achieve a
physiological environment that will allow the best possible recovery. The ‘lethal triad’
of hypothermia, acidosis, and coagulopathy due to initial hypovolaemia require
aggressive correction. From the outset ICU management must also attempt to minimise
the complications of these injuries and the resuscitative process. This review will
address some of the key issues relating to the care of these patients in the ICU.
ß 2004 Elsevier Ltd. All rights reserved.

Introduction contributing to the significant improvements in out-

come from DCS in recent years.33
The introduction and establishment of the damage
control surgery (DCS) has led to severely injured and
unstable patients being presented to Intensive Care The concepts of damage control
Units for ongoing resuscitation. These patients pre- applied to intensive care
sent many challenges for the Intensive Care team
and emphasise the need for a multidisciplinary The principles at the core of damage control phi-
approach to optimise patient management. Multi- losophy are not foreign to current ICU practice. The
ple issues need to be addressed simultaneously concept of delay in definitive therapy while systems
while the overall aim is to rapidly achieve a phy- support is maintained is in fact the core principle of
siological environment that will allow the best pos- management for many intensive care patients. The
sible recovery. The ‘lethal triad’ of hypothermia, quote from Voltaire, ‘‘The art of medicine consists
acidosis, and coagulopathy due to initial hypovo- of amusing the patient while nature curses the
laemia require aggressive correction. From the out- disease’’ remains not too far from the truth. Often
set ICU management must also attempt to minimise there is no definitive procedure that is curative and
the complications of these injuries and the resusci- there is great reliance on the system and organ
tative process. This review will address some of the support therapy available in intensive care while
key issues related to the care of these patients in the initial inflammatory, infective or traumatic pro-
the ICU. Improved ICU management is credited as cess resolves. Clearly the level of system or organ
support can be very intensive or even complete and
*Corresponding author. Tel.: þ612-9828-3000;
as with DCS is aimed at rapidly reversing abnormal
fax: þ612-9828-3551. physiology to produce the best possible environ-
E-mail address: [email protected] (M.J.A. Parr). ment for optimal recovery. Within this philosophy
0020–1383/$ — see front matter ß 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.injury.2004.03.010
714 M.J.A. Parr, T. Alabdi

there are clear circumstances when, as with DCS, Upon arrival in the ICU, a brief primary survey,
procedures and treatments may be delayed and confirming the ABCD status of the patient should be
deferred until more stability or improvement has repeated. These patients are often unstable during
been achieved. The following will serve as exam- transfer and serious things can happen in transit
ples: the trauma patient with a severe brain injury that require immediate attention on arrival in the
will not be put through extensive radiology to ICU. Following this, there are a number of key issues
attempt to clear the thoracolumbar spine when to be identified early and addressed during ICU
the priority of management is to achieve a stable management (Table 1).
intra-cranial pressure. Field cannulae and resusci-
tation cannulae are not changed and procedures
that may involve bleeding are delayed until coagu- Identifying patients for surgical re-
lopathy is fully reversed. For the medical hypother- exploration or angiography-guided
mic patient, all non-life-saving procedures are haemorrhage control procedures
delayed until effective rewarming has taken place.
In essence the principles proposed by the DCS The principles of damage control surgery dictate
approach are fully consistent with current high that there is no further benefit to the patient from
quality Intensive Care Medicine. continued operative intervention and definitive sur-
gical procedure should be carried out after correc-
tion of the serious metabolic abnormalities. There
Handover to the intensive care team may however be situations where these patients do
require further surgical or radiology-guided hae-
DCS patients will have been in the operating room morrhage control procedures.
for a significant period of time and the decision to Occasionally a patient who has been returned to
adopt a DCS strategy should be communicated to the ICU for correction of acidosis, hypothermia and
the ICU team prior to patient arrival in the ICU. This coagulopathy will suddenly haemorrhage from a
will provide the opportunity for the ICU team to vessel that was not identified at the initial opera-
prepare for optimal care of the patient. In alerting tion. This may be revealed as haemodynamic
the ICU team, details of the trauma, initial resusci- instability with rapid filling of drains left at opera-
tation and surgical intervention will be described. tion or a distending abdomen. This may be in rela-
The extent of the acidosis, coagulopathy and tion to vessels not identified as bleeding due to
hypothermia can be discussed and the strategies vasospasm and the poor perfusion at operation
to reverse these can then be planned. This may which now with improving temperatures and reper-
include such issues as preparing an isolation room fusion are revealed as active bleeding sites. Urgent
to high ambient temperature, preparing warming re-exploration to stop the haemorrhage may be
devices, obtaining warm fluids, preparing special lifesaving. Usually however all surgically controlla-
equipment (e.g. ventilators and renal replacement ble bleeding will have been dealt with at the time of
therapy (RRT) machines) and alerting the Blood operation and the ongoing haemorrhage is likely to
bank to the likely on-going requirement for blood represent bleeding from a site that cannot be man-
products. Early discussion may also help in identify- aged in the operating room. Some of these patients
ing patients that require angiography with a view to may benefit from interventional radiology with
embolisation of identified active haemorrhage. angiography followed by embolisation of identified

Table 1 Key issues for early ICU management

Identifying patients who may benefit from surgical re-exploration or

radiology/angiographic guided haemorrhage control procedures
Correction of hypothermia
Correction of coagulopathy
Correction of acidosis
Fluid and blood product therapy
Monitoring resuscitation end points
Assessing for missed injuries (tertiary survey)
Planing return to OR
Initiating specific therapy to reduce complications (ACS, peptic ulceration, thrombo-prophylaxis, protective lung
ventilation, infection control and appropriate antimicrobial therapy)
Maintaining effective communication
Damage control surgery and intensive care 715

actively bleeding vessels. This strategy has been Abdominal compartment syndrome
credited with improved outcomes.15,27,36 This
approach of urgent post-operative angiography Patients managed by damage control laparotomy
should be considered for patients with complex are at high risk of intra-abdominal hypertension and
hepatic injury (e.g. for deep or transhepatic abdominal compartment syndrome. 30 For this rea-
GSW), when intrahepatic arterial bleeding or an son the abdomen is only partially (or temporarily)
AV fistula can be embolised/occluded. Hepatic closed. Despite this, intra-abdominal hypertension
angiography appears to be a safe adjunct to the and abdominal compartment syndrome may still
principles of damage control.32 Angiography should occur because of increasing visceral swelling,
also be considered for patients with significant expanding haematomata, and the use of abdominal
retroperitoneal, pelvic or deep muscle injuries packs. Signs of a distended abdomen, increased
identified at surgery. A contrast blush seen at angio- ventilator inspiratory pressure requirement, raised
graphy indicates active arterial bleeding and the intracranial pressure, oliguria progressing to anuria,
need for embolisation. Angiography before damage a decreased cardiac output and hypotension may
control laparotomy may also be indicated if there is occur insidiously and can also be associated with
a very high suspicion of a bleeding lesion that may other pathologies than intra-abdominal hyperten-
be amenable to embolisation. sion and abdominal compartment syndrome.48
Clearly, safe patient transfer to the angiography An objective assessment of intra-abdominal pres-
suite will require a coordinated approach between sure is therefore required. Intra-abdominal pres-
intensive care, surgical and radiology staff. The sure can be estimated most conveniently from
monitoring, ongoing resuscitation, and medical the transduced pressure of an indwelling urinary
and nursing supervision of these patients in the catheter.29 This is usually done by intermittent
radiology department needs to be of the same measurement but it is also possible to monitor
standards as the ICU and OR for optimal outcome continuously using a three-way irrigation cathe-
and serves to emphasise the importance of seeing ter.35 A pressure >30 mmHg confirms ACS and
‘intensive/critical care as a process and not a loca- requires return to OR for initial or further decom-
tion’. Therefore, patients should be transferred pression. Decompression should lead to improved
with full monitoring in place and with the equip- visceral perfusion, cardiac function, and ventilatory
ment to provide optimal care including a high capa- mechanics. Occasionally complex pelvic injuries
city fluid warmer and a forced air heating device may need fascial closure to achieve pelvic tampo-
that can be used in the radiology department. This nade at the cost of renal dysfunction and this may
will require a large team of individuals to coordinate be an indication for early RRT to optimise the
safe transfer. metabolic and fluid status of the patient.

Return to operating room Resuscitation: physiological

optimisation
Planning for return to the OR for definitive care
follows successful resuscitation, when the patient During the ICU resuscitation phase, procedures and
is normothermic, with corrected coagulation and interventions should be limited to those that are
platelet count, and resolved acidosis. This resus- essential for the correction of the triad of acidosis,
citation may occasionally be achieved over a few hypothermia and coagulopathy while ensuring
hours but is usually optimal at 24—48 h. Compro- patient stability. The components of the triad
mising and returning to the OR before adequate should not be looked at in isolation and require
resuscitation will leave the patient less likely to simultaneous management. Exposure of the patient
withstand the prolonged procedure of definitive should be kept to absolute minimum to allow effec-
surgery. Early re-operation may, however, be tive rewarming. Non-essential mobile X-rays (e.g. to
appropriate if gross contamination was initially clear the thoraco-lumbar spine) should be post-
present, to reduce the incidence of subsequent poned.
infection. Prior to returning to the OR, the route The underlying principle of the ICU resuscitation
for on-going feeding should be considered as it is to provide physiological optimisation that will
may be appropriate to place a nasojejunal or allow the best chance for recovery. At the centre
enterostomy feeding tube at the time of definitive of this principle is the aim of reversing hypovolae-
surgery. Abdominal closure is performed when mia which then allows adequate cardiac output and
oedema has resolved enough to allow closure with- oxygen delivery with the resultant correction of
out tension. metabolic acidosis, coagulopathy and hypothermia.
716 M.J.A. Parr, T. Alabdi

Assessment of the adequacy of the circulating Table 2 Strategies to correct hypothermia

volume accompanies active rewarming and correc-
Warm room
tion of coagulopathy. If these issues are correctly
Room T > 28 8C, may not be feasible in many ICUs
addressed the metabolic acidosis will gradually
improve. Cover/insulate patient
Reduces convection, conduction and radiant
heat loss
Prevent unnecessary exposure
Rewarming
Dry patient
Temperature control is dependent on a balance Remove any wet sheets/clothing to reduce
between heat generation, CNS temperature control evaporative heat loss
and heat loss by conduction, convection, evapora- Active warming
tion, and radiation. Heat loss commences at the site External
of trauma and is compounded by the degree of Forced air warming device (e.g. Bair HuggerTM)
reduced perfusion, prolonged exposure, immobility, Warm water blanket
resuscitative fluid and procedures. The very young Radiant heaters
and very old are particularly at risk. In the absence Internal
of preemptive treatment, this process continues in Use pre-warmed fluids
the emergency department, where the patient is High capacity fluid warmer (e.g. Level 1TM,
uncovered and exposed to a large patient—room Rapid Infusion System [RIS])
temperature gradient. This is continued into the Warmed ventilator gases (warm water
reservoir/HME)
OR where general anaesthesia, blood loss, and
Peritoneal or pleural lavage
further exposure, now with open body cavities,
Extracorporeal rewarming
rapidly results in severe hypothermia.22
Core temperature decreases precipitously for 1 h
and then decreases slowly for 2—3 h in a predictable
manner following the induction of general anaes- They result in less core temperature after-drop
thesia in normal circumstances, which emphasises and a 6—10 fold faster rate of warming when com-
the need for awareness and the use of preventative pared with passive and inhalation warming techni-
strategies.54 A temperature <35 8C is associated ques.21 The majority of patients can be rapidly
with reduced cardiac output, increased SVR, risk rewarmed by using these techniques and ensuring
of arrhythmia, cold diuresis, metabolic acidosis, all fluids administered are warm.
left-shift of the oxygen-haemoglobin dissociation Cold temperature IV fluid administration is a very
curve and coagulopathy in it’s own right. These rapid way of accelerating hypothermia. Therefore, all
patients have increased fluid, transfusion, vaso- fluids administered should be warm and there are
pressor and inotropic requirements, and more organ numerous devices on the market to provide consistent
dysfunction, mortality, and prolonged intensive warming of fluids at both low and high infusion rates.
care unit stay.37,51 A temperature <34 8C is further High capacity fluid warmers (e.g. Level 1 and rapid
associated with inhibition of platelet aggregation infusion system (RIS)) are capable of warming fluid
and dysfunction of intrinsic and extrinsic coagula- from 4 8C to body temperature at flow rates up to
tion factors and is used as an indicator for DCS.19 0.5 l/min (Level 1) and 1.5 l/min (RIS). These devices
These processes emphasize the need for prevention have been credited with improved resuscitation of
and effective warming strategies (Table 2). major trauma, reduced requirements for fluid/blood
Maintaining a room temperature >28 8C may be products, less coagulopathy, more rapid correction of
impractical for multi-bed space ICUs but feasible for acidosis and hypothermia, and fewer hospital com-
ICUs with single rooms. Keeping the patient covered plications when compared to conventional methods
and dry needs particular emphasis if effective of fluid/blood product administration .5,14 However,
warming is to be achieved. Ventilated patients there may also be potential for overly aggressive fluid
should have gases humidified and warmed and in administration that may worsen outcome .23
this situation the heated water bath humidifiers will There may be potential for an adverse response to
be more efficient than heat and moisture exchange surface warming when cold, acidotic blood returns
(HME) devices. Warming gases is however, an inef- to the central circulation further lowering core
ficient method of warming hypothermic patients temperature, and precipitating life-threatening
compared with forced air warming devices. Forced arrhythmias. With severe hypothermia (T < 33 8C)
air warming devices are an increasingly popular and there is an increasing risk of cardiac depression and
efficient method of warming from hypothermia. arrhythmias and the more extreme the hypothermia,
Damage control surgery and intensive care 717

the more justification for aggressive warming, Dilution of coagulation factors and platelets
including extracorporeal techniques. A variety of occurs early and abnormal prolonged coagulation
invasive techniques of rewarming have been studies may be demonstrated after even small
described including gastric lavage, bladder irriga- volume fluid resuscitation. The clinically observed
tion, and peritoneal or pleural lavage. While all may coagulopathy however, may not be confirmed by
have potential uses, they are often difficult to man- laboratory investigation, demonstrating that ele-
age effectively especially in the multi-trauma situa- ments other than levels of clotting factors and
tion. Extracorporeal rewarming is possible using platelet count are involved in haemostatic failure.
several techniques and is likely to see increased Hypothermia results in dysfunction of intrinsic and
popularity given the increasing availability of simple extrinsic coagulation cascades that will not be
pumps and heparin-bonded circuits. The need for demonstrated by laboratory prolongation of pro-
anticoagulation however, may restrict use. Contin- thrombin time (PT) and activated partial thrombo-
uous arteriovenous rewarming has been achieved plastin time (APTT) because coagulation testing is
using a heparin-bonded circuit without a pump normally performed at 37 8C rather than at the
where flow is driven by the patient’s own blood patient’s core temperature.44,45
pressure. This technique rewarms at a rate of 4— Hypothermia causes a temperature-dependent
5 8C/h, which is more efficient than many other defect in thromboxane B2 production and altered
methods. Because the warmed blood is sent directly enzyme kinetics which delay the initiation and
to organs, continuous arteriovenous rewarming propagation of platelet aggregation.53 Resultant
rapidly increases core temperature without shiver- platelet dysfunction is seen despite adequate repla-
ing .20 Depending on the technique used extracor- cement of platelet number.18 Also the increasing
poreal rewarming can achieve warming rates of 4— use of non-steroidal anti-inflammatory drugs and
50 8C/h. Advocates suggest that if the temperature aspirin impair platelet activity prior to the trauma
drops below 33 8C, aggressive extracorporeal incident.
rewarming (arteriovenous or veno-venous) rewarm- As there is often poor correlation between plate-
ing should be considered.20,26,43 The fact that let count and ability to form clots in patients who
despite aggressive replacement of clotting factors have received massive transfusion, the presence of
and platelets, normal coagulation may not occur continued haemorrhage in this setting is an indica-
until the core temperature exceeds 34 8C adds sup- tion for platelet transfusion even with a ‘‘normal’’
port to this suggested approach.44 platelet count.
Many coagulation factors and enzyme reactions
are pH dependent and the presence of severe meta-
Coagulopathy bolic acidosis may directly contribute to the coa-
gulation failure.8 Citrate present in transfused
The incidence of early coagulation abnormalities blood may decrease calcium concentration and
after trauma is high and they are independent pre- add to the coagulopathy,12 rapid plasma protein
dictors of mortality. An initial abnormal PT administration may also decrease ionised calcium
increases the adjusted odds of dying by 35% and levels, presumably as a consequence of the binding
an initial abnormal APTT increases the adjusted of calcium ions to anionic sites on plasma protein. In
odds of dying by 326%.40 The late coagulopathy most situations this is a transient phenomenon and
associated with DCS is multifactorial in origin dependent on the total dose of citrate administered
(Table 3). and the rate of infusion. Often calcium is given in an
empiric fashion during massive and continuing high
volume transfusion but rational calcium replace-
Table 3 Factors contributing to coagulopathy
ment may be based on direct ionised calcium mea-
Hypothermia surement. Interestingly, Haemaccel, a gelatin
Dilution of coagulation factors colloid solution has a calcium concentration of
Continued loss of coagulation factors from bleeding 6.25 mmol/l, and its use during trauma resuscita-
Thrombocytopaenia tion has been associated with transient hypercal-
Non-functional platelets caemia.17
Acidosis Increased fibrinolysis is seen in hypothermia and
Hypocalcaemia
imbalances in the production and degradation of
Reduced synthesis of coagulation factors due to liver
dysfunction secondary to hypoxic/ischaemic insult
fibrin may also lead to excessive bleeding.16,49
Fibrinolysis Fibrinolysis may also be excessive due to extensive
Disseminated intravascular coagulation (DIC) tissue damage and hypotension, and is particularly
seen in head injury and lung injury and results in a
718 M.J.A. Parr, T. Alabdi

DIC with prolonged coagulation, low fibrinogen, include the provision of up to 10 units of group
elevated D dimers and thrombocytopaenia. O, non-cross-matched packed red blood cells
As conventional laboratory assessments of coa- (PRBCs), 6 units of platelets, and 4 units of stored
gulation function have limitations in the DCS situa- (previously thawed) fresh frozen plasma (FFP) as
tion, thrombelastography has been advocated as a the initial response.33 A further 4 units of FFP should
technique to determine coagulation abnormalities be thawed to replace the dispensed FFP and five
and give information about fibrinolytic activity more units of group O blood should be available on
and platelet function.34,41 Thromboelastography request. The cycle should repeat automatically
assesses the clotting from initial platelet-fibrin until discontinued by the trauma team. The ongoing
interaction, through platelet aggregation, clot availability of blood products needs to be con-
strengthening, and fibrin cross-linking, to clot lysis, firmed, particularly for hospitals that may be large
within around 20 minutes. It is practical for use in distances from a central blood transfusion service.
the ICU and OR, simplifies the diagnosis of coagulo- Prolonged turn around times from request to infu-
pathy, and may be an early predictor of the need for sion of blood products will not help these patients.
transfusion. However, despite the availability of Guidelines for the administration of blood pro-
thromboelastography technology it has not become ducts vary on what is the ideal red blood cell,
popular. Recently, the activated coagulation time platelet and clotting factor replacement regimen
(ACT) has been assessed in relation to global coa- and the ideal endpoints. Again a locally agreed
gulation status. The investigators suggest that an policy is the way to appropriately plan for and
elevated ACT (which only takes a few minutes to provide the optimal chances of survival for these
perform) is an objective indicator that the coagula- patients.28
tion system reserve is near exhaustion and may The rheological effect of anaemia (platelet func-
represent an indication for considering damage tion is inversely related to haematocrit) is com-
control manoeuvres or more aggressive resuscita- pounded by thrombocytopenia < 100  109 l1
tion.3 resulting in an increasingly prolonged bleeding
There are no simple strategies to correct coagu- time. It follows that during continued haemorrhage
lopathy that can be applied to all trauma patients. and instability especially if head injury is present
Computer modelling has generated suggested pre- that the haematocrit should be maintained >30 and
ventive strategies recognising that existing proto- the platelet count > 100  109 l1. These values
cols underestimate the dilution of clotting factors in correlate with the minimum value recommended
severely bleeding patients.25 Again a number of by the British Committee for Standards in Haema-
strategies (Table 4) may need to be implemented tology for intracranial or eye surgery.6
simultaneously and enrolling the help of experi- DCS patients should receive the freshest blood
enced haematologist should be considered at an available and as whole blood if possible. However
early stage. the increasing use of donated blood to provide blood
To optimize early blood product resuscitation, a components has lead to whole blood availability
local ‘trauma exsanguination policy’ needs to be being very limited.
agreed that can be activated to allow rapid blood Platelet counts of < 100  109 l1 can be antici-
product availability during DCS and ICU care. This pated when between 1 and 1.5 blood volumes have
protocol should only require a single phone call for been replaced.9 Initially platelets will be given
activation and should result in the expedited deliv- empirically and repeated according to ongoing
ery of blood and blood products to the OR. A reason- blood loss, transfusion requirement and platelet
able approach for a major trauma centre would counts. In the face of massive transfusion and

Table 4 Strategies to correct coagulopathy

Reverse hypothermia and maintain an effective circulating blood volume and oxygenation
Preferably use whole blood and/or the freshest available
Give FFP to replace coagulation factors (10—15 ml/kg approx. 2 units will achieve 30% factor activity in adults)
Give platelets
Give cryoprecipitate if fibrinogen level is <1 g l1
Give calcium (10 mmol) to reverse hypocalacaemia
Consider the use of adjuncts to promote coagulation/reduce fibrinolysis (e.g. aprotinin)
Consider the use of rVIIa
Give Vitamin K
Repeat coagulation tests and blood count and modify treatment accordingly
Damage control surgery and intensive care 719

DCS aiming for an early platelet count > 100  tinues despite all attempts at correcting coagulation
109 l1 will provide a margin of safety. With resolu- and all attempts at surgery and embolisation; then
tion of the acidosis, coagulopathy and approaching rVIIa is administered in a dose of 100 mg kg1. Higher
normothermia lower platelet counts (50— doses or repeated doses have been used and remain
75  109 l1) may be a more suitable level to aim an option.42 A haemotologist should be involved in
for. Platelets are presented in different volumes the decision making process.
and platelet yields depend on their method of pre- There is also anecdotal evidence (including the
paration (either single donor, apheresis or pooled). authors experience) that aprotinin, the protease
A standard single donor unit (average platelet inhibitor may have a role as adjunctive therapy in
yield ¼ 55  109 unit) can be expected to raise the presence of ongoing coagulopathy and potential
the platelet count by 5—10  109 l1 in marrow DIC by decreasing fibrinolysis.52
failure patients, but in the DCS patient this level
of effect may be unlikely as they are rapidly con-
sumed. The initial adult dose will be 4 units or one Acidosis
adult therapeutic dose for pooled packs and apher-
esis packs (average platelet yield > 240  109 unit). All patients having DCS have a lactic acidosis that
The requirement for FFP is less predictable as the reflects the degree of anaerobic metabolism in the
labile factors V and VIII have wide normal ranges and presence of inadequate oxygen delivery. The acido-
levels down to 20% of normal may not result in sis also contributes to coagulopathic bleeding
coagulopathy if the haematocrit and platelet count resulting in a vicious cycle of continued blood loss,
are maintained. FFP should also be given initially coagulopathy, hypovolaemia and worsening acido-
empirically in massive transfusion (defined as the sis. The correction of this acidosis therefore
replacement of one blood volume in 24 h) and then requires control of haemorrhage and optimization
laboratory tests should be used to assist with of oxygen delivery, initially by blood and fluid
ongoing requirements. FFP 10—15 ml/kg or an administration. Inotropic/vasopressor should be
empiric approach of giving 4 units initially and then considered if the patient does not respond ade-
at a ratio of 2:5 for FFP:PRBCs is often practiced quately to volume and red cell loading and myo-
once one blood volume has been replaced and cardial depression may be seen at pH < 7:2.
bleeding is not under control. Frequent samples However, the need for inotropes and vasopressors
should be taken for the usual clotting assays includ- correlates with poor outcomes for these patients.
ing fibrinogen. Because FFP may not supply enough The acidosis puts a large burden on the respiratory
fibrinogen in the massive transfusion situation system and continued ventilatory support is
cryoprecipitate (10 units) should be given if the required. The direct use of IV sodium bicarbonate
measured fibrinogen levels is <1 g l1. If the to raise the pH > 7:2 remains an option but may be
hypothermia is not corrected the administered pla- better avoided (particularly for patient who appear
telets and coagulation factors will remain ineffec- to be stable) because of well recognised adverse
tive. Blood components are administered until the PT effects of bicarbonate administration. Some
and APTT reach less than 1.25 times control levels, patients will have established acute renal failure
the platelet count is more than 100  109 l1, and and the early commencement of RRT may benefit
the fibrinogen is greater than 1 g l1. these patients with more rapid correction of acido-
Recently, recombinant activated factor VII sis (particularly if a bicarbonate dialysis fluid is
(rFVIIa) has been reported as useful for adjunctive used) and provision of an optimal metabolic envir-
treatment of coagulopathy in trauma patients and is onment. The persistence of a metabolic acidosis
undergoing trials. rFVIIa was originally developed as and base deficit are negative prognostic indicators
a pro-hemostatic agent for the treatment of hae- in trauma patients10,11,46 and although the initial
mophilia patients who had become sensitised to lactate level may not be predictive, lack of clear-
donated factor VIII or IX. rFVIIa has been successfully ance of lactate within 48 h is strongly predictive of
used, with impressive clinical resolution of bleeding mortality.1 Clearance of lactic acidosis in DCS
in moribund trauma patients in whom standard pro- patients suggests that optimal resuscitation has
cedures had failed. Various dosage regimens have occurred while persistent acidosis is generally a sign
been used and clinical studies are under way.38,39 In of hypovolaemia. Persistent acidosis may also
the absence of definitive studies local agreed poli- represent reduced cardiac output, reduced oxygen
cies on the use of rFVIIa need to be formulated. Our delivery or abnormal oxgen utilisation and empha-
current approach is that if after the administration sises the need for continuous monitoring, repeated
of 10 units of PRBCs, 10 units of platelets, 10 units of objective assessment and setting appropriate resus-
FFP, and 10 units of cryoprecipitate, bleeding con- citation endpoints.
720 M.J.A. Parr, T. Alabdi

Monitoring of resuscitation end points Table 6 Strategies to reduce complications

Measurement of IAP
The response to therapy is monitored initially by
Peptic ulceration prophylaxis
observing vital signs and urine output. However, Thromboprophylaxis
even in young trauma patients, these signs may Protective lung ventilation
be unreliable and fail to demonstrate significant Infection control and appropriate antimicrobial
cardiac depression.47 These patients may therefore therapy
benefit from invasive (PA catheter) or non-invasive Early nutritional support (preferably enteral)
(echocardiography) repeated assessments of car-
diac filling and contractility. A number of resusci-
tative end points have been proposed that go
beyond conventional ‘‘vital signs’’, including serum but may be better delayed until the acidosis is
lactate and base deficit as stated above, mixed corrected and splanchnic perfusion is deemed to
venous oxygen saturation in conjunction with oxy- be adequate. Compression stockings should be
gen delivery and consumption4 and gastric mucosal applied initially and calf compressors added at an
pH31. No particular strategy has been shown to be early stage, heparin will be withheld until the coa-
universally applicable and all have potential limita- gulopathy is corrected. Low molecular weight
tions or risks that need to be considered but aggres- heparin may be an option but renal impairment
sive therapy, continuous assessment and repeated and repeated surgery restrict is use in the early
laboratory evaluations are well justified. The abso- days post injury. These patients are at high risk
lute level of haemoglobin required in the acute of acute respiratory distress syndrome (ARDS)
setting is controversial and while much has been because of chest trauma, aspiration, hypotension
written on the tolerance of restrictive transfusion and large IV fluid administration or transfusion
strategies in critically ill patients 24, the unstable associated lung injury (TRALI), in which case a lung
DCS patient at high risk of organ failure is not a protective ventilatory strategy will be appropriate.2
suitable candidate for this strategy in the immedi- Scrupulous infection control is required and the
ate ICU phase. Maintaining a haemoglobin level rational use of antibiotics is particularly important.
>10 g dl1/haemotocrit >30% will provide a margin Positioning the patient head up will reduce the rate
of safety in the initial ICU phase. of nosocomial pneumonia. Antimicrobial prophy-
laxis should be limited and proven infection should
be appropriately targeted. The indisciminant use of
Specific strategies to reduce braod spectrum antimicrobials will promote bacter-
complications ial resistance and must be avoided. General nursing
and pressure area care must be meticulous and
Given the critical condition of DCS patients it is not neuromuscular blockade should be avoided. Ade-
surprising that the overall mortality rate is high quate sedation and analgesia will be required. A
(12—67%) and complications are frequent50. How- tertiary survey should be performed within 24 h to
ever many of the complications are predictable ensure that all injuries have been identified and the
(Table 5) and strategies can be commenced early priorities of interventions and ongoing management
in the ICU phase to reduce the risks (Table 6). are agreed. It is not uncommon for these patients to
All patients should have IAP measured and pro- have been incompletely assessed on admission
phylaxis from peptic ulceration should be com- because of haemodynamic instability requiring
menced on admission (H2 blocker or proton pump immediate surgery. It may be only after resuscita-
inhibitor)7. Early enteral nutrition is also protective tion in the ICU that the patient is deemed stable
enough to be transferred for further radiology ima-
Table 5 Predictable complications seen in DCS ging if needed.
patients
ACS
Peptic ulceration
Communication
Venous thromboembolism
ARDS Given the complexity of the management of DCS
Nosocomial infection patients it is not surprising that one of the crucial
Intra-abdominal infection, fistula, dehiscence factors in optimising care is communication. This is
Nutritional failure often overlooked.13 Whether it is the timely avail-
Critical illness myoneuropathy ability of blood products, the comprehensive
assessment of all injuries, the coordinated return
Damage control surgery and intensive care 721

to the OR for definitive care or the urgent inter- 11. Davis JW, Shackford SR, Mackersie RC, Hoyt DB. Base deficit
as a guide to volume resuscitation. J Trauma 1988;28:
ventions required to save life, the need for clear and
1464—7.
effective communication between emergency phy- 12. Denlinger JK, Nahrwold ML, Gibbs PS, Lecky JH. Hypocal-
sicians, surgeons, anaesthetists, intensivists, radi- caemia during rapid blood transfusion in anaesthetized
ologists, haematologists, nurses, laboratory staff, man. Br J Anaesth 1976;48:995—1000.
and all others involved should be at the forefront of 13. Donchin Y, Gopher D, Olin M, et al. A look into the nature
management. This is often an area of trauma care and causes of human errors in the intensive care unit.
Quality and safety in health. Care 1995;12:143—7.
that can easily be improved. 14. Dunham CM, Belzberg H, Lyles R, et al. The rapid infusion
system: a superior method for the resuscitation of hypovo-
lemic trauma patients. Resuscitation 1991;21:207—27.
Conclusion 15. Dutton RP, Mackenzie CF, Scalea TM. Hypotensive resusci-
tation during active hemorrhage: impact on in-hospital
mortality. J Trauma 2002;52:1141—6.
Damage control surgery is a significant advance in 16. Enderson B, Chen J, Robinson R, et al. Fibrinolysis in multi-
trauma patient management. The central principle system trauma patients. J Trauma 1991;31:1240—6.
of damage control surgery is that patients are more 17. Evans PA, Madira W, Riyatt MS, et al. Changes in plasma
likely to die from a triad of coagulopathy, hypother- ionised calcium within 24 h of trauma in patients infused
mia and metabolic acidosis than from a failure to with the calcium containing colloid Haemaccel during fluid
resuscitation. J Acc Emerg Med 1997;14:73—5.
complete operative repairs. If the patient is to 18. Ferrara A, MacArthur J, Wright H, et al. Hypothermia and
survive, definitive surgery must wait until hypother- acidosis worsen coagulopathy in the patient requiring
mia, acidosis and coagulopathy are corrected. massive transfusion. Am J Surg 1990;160:515—8.
These corrections should occur over a period of 19. Garrison J, Richardson D, Hilakos A, et al. Predicting the
hours in the ICU and result in rapid and complete need to pack early for severe intra-abdominal haemorrhage.
J Trauma 1996;40:923—9.
reversal of metabolic failure. A multidisciplinary 20. Gentilello I, Rifley. Continuous arteriovenous rewarming:
coordinated approach is needed to optimize out- report of a new technique for treating hypothermia. J
come. Trauma 1991;31:1151—54.
21. Goheen MS, Ducharme MB, Kenny GP, et al. Efficacy of
forced-air and inhalation rewarming by using a human model
for severe hypothermia. J Appl Physiol 1997;83: 1635—40.
References 22. Gregory J, Flancbaum L, Townsend M, et al. Incidence and
timing of hypothermia in trauma patients undergoing
1. Abramson D, Scalea TM, Hitchcock R, et al. Lactate operations. J Trauma 1991;31:795—800.
clearance and survival following injury. J Trauma 23. Hambly PR, Dutton RP. Excess mortality associated with the
1993;35:584—9. use of a rapid infusion system at a level 1 trauma center.
2. Anonymous. Ventilation with lower tidal volumes as Resuscitation 1996;31:127—33.
compared with traditional tidal volumes for acute lung 24. Hebert PC, Wells G, Blajchman MA, et al. A multicenter,
injury and the acute respiratory distress syndrome. The randomised, controlled clinical trial of transfusion require-
Acute Respiratory Distress Syndrome Network. NEJ Med ments in critical care. NEJ Med 1999;340:409—17.
2000:342;1301—8. 25. Hirshberg A, Dugas M, Banez EI, et al. Minimizing dilutional
3. Aucar JA, Norman P, Whitten E, et al. Intraoperative coagulopathy in exsanguinating hemorrhage: a computer
detection of traumatic coagulopathy using the activated simulation. J Trauma 2003;54:454—63.
coagulation time. Shock 2003;19:404—7. 26. Hirshberg A, Walden R. Damage control for abdominal
4. Bishop MH, Shoemaker WC, Appel PL, et al. Prospective trauma. Surg Clin North Am 1997;77:813—20.
randomized trial of survivor values of cardiac index, oxygen 27. Hoey BA, Schwab CW. Damage control surgery. Scand J Surg
delivery and oxygen consumption as resuscitation endpoints 2002;91:92—103.
in severe trauma. J Trauma 1995;38:780—7. 28. Horsey PJ. Multiple trauma and massive transfusion.
5. Booke M, Sielenkamper A. Massive transfusion with the Anaesthesia 1997;52:1027—9.
Rapid Infusion System. Its effect on core body temperature. 29. Iberti TJ. A simple technique to accurately determine intra-
German Anaesthesist 2001;50:926—9. abdominal pressure. Crit Care Med 1987;15:1140—2.
6. British Committee for Standards in Haematology. Guidelines 30. Ivatury RR, Diebel L, Porter JM, et al. Intra-abdominal
for platelet transfusions. Transfus Med 1992;2:311—8. hypertension and the abdominal compartment syndrome.
7. Cook DJ, Reeve BK, Guyatt GH, et al. Stress ulcer Surg Clin North Am 1997;77:783—800.
prophylaxis in critically ill patients. Resolving discordant 31. Ivatury RR, Simon RJ, Islam S, et al. A prospective,
meta-analyses. JAMA 1996;275:308—14. randomized study of endpoints of resuscitation after major
8. Cosgrif N, Moore EE, Sauaia A, et al. Predicting life- trauma: global oxygen transport indices versus organ-
threatening coagulopathy in the massively transfused specific gastric mucosal pH. J Am Coll Surg 1996;183:
patient: hypothermia and acidoses revisited. J Trauma 145—54.
1997;42:857—62. 32. Johnson JW, Gracias VH, Gupta R, et al. Hepatic angio-
9. Counts RB, Haisch C, Simon TL, et al. Hemostasis in graphy in patients undergoing damage control laparotomy.
massively transfused trauma patients. Ann Surg 1979;190: J Trauma 2002;52:1102—6.
91—9. 33. Johnson JW, Gracias VH, Schwab CW, et al. Evolution in
10. Davis JW, Kaups KL. Base deficit in the elderly: a marker of damage control for exsanguinating penetrating abdominal
severe injury and death. J Trauma 1998;45:873—7. injury. J Trauma 2001;51:261—9.
722 M.J.A. Parr, T. Alabdi

34. Kaufmann CR, Dwyer KM, Crews JD, et al. Usefulness of 44. Reed R, Bracey A, Hudson J. Hypothermia and blood
thrombelastography in assessment of trauma patient coa- coagulation: dissociation between enzyme activity and
gulation. J Trauma 1997;42:716—22. clotting levels. Circ Shock 1990;32:141—52.
35. Kron IL, Harman PK, Nolan SP. The measurement of intra- 45. Rohrer M, Natale A. Effect of hypothermia on the coagula-
abdominal pressure as a criterion for abdominal reexplora- tion cascade. Crit Care Med 1992;20:1402—5.
tion. Ann Surg 1984;199:28—30. 46. Rutherford EJ, Morris JA, Reed GW, Hall KS. Base deficit
36. Kushimoto S, Arai M, Aiboshi J, et al. The role of stratifies mortality and determines therapy. J Trauma
interventional radiology in patients requiring damage 1992;33:417—23.
control laparotomy. J Trauma 2003;54:171—6. 47. Sauaia A, Moore FA, Moore EE, et al. Epidemiology of trauma
37. Luna G, Maier R, Pavlin E, et al. Incidence and effect of deaths: a reassessment, J Trauma 1995;38:185—93.
hypothermia in seriously injured patients. J Trauma 48. Schein M, Wittmann D, Aprahamian C, et al. The abdominal
1987;27:1014—8. compartment syndrome: the physiological and clinical
38. Lynn M, Jerokhimov I, Jewelewicz D, et al. Early use of consequences of elevated intra-abdominal pressure. J Am
recombinant factor VIIa improves mean arterial pressure Coll Surg 1995;180:745—53.
and may potentially decrease mortality in experimental 49. Schmidt U, Enderson B, Chen J, et al. D-Dimer levels
hemorrhagic shock: a pilot study. J Trauma 2002;52:703—7. correlate with pathologic thrombosis in trauma patients. J
39. Lynn M, Jeroukhimov I, Klein Y, Martinowitz U. Updates in Trauma 1992;33:312—20.
the management of severe coagulopathy in trauma patients. 50. Shapiro MB, Jenkins DH, Schwab CW, Rotondo MF. Damage
Int Care Med 2002;28:S241—7. control: a collective review. J Trauma 2000;49:969—78.
40. MacLeod JB, Lynn M, McKenney MG, et al. Early coagulopathy 51. Steinemann S, Davis JW, Shackford SR. Implications of
predicts mortality in trauma. J Trauma 2003;55:39—44. admission hypothermia in trauma patients. J Trauma 1990;
41. Mallett SV, Cox DJ. Thrombelastography. Br J Anaesth 30:200—2.
1992;69:307—13. 52. Valentine S, Williamson P, Sutton D. Reduction of acute
42. Martinowitz U, Kenet G, Lubetski A, et al. Posssible role of haemorrhage with aprotinin. Anaesthesia 1993;48:405—6.
recombinant activated factor VII (rFVIIa) in the control of 53. Valeri C, Cassidy G, Khuri S. Hypothermia-induced reversal
hemorrhage associated with massive trauma. Can J Anaesth and platelet dysfunction. Ann Surg 1987;205:175—81.
2002;49:S15—20. 54. Yamakage M, Kamada Y, Honma Y, et al. Predictive
43. Moore EE. Staged laparotomy for the hypothermia, acidosis, variables of hypothermia in the early phase of general
and coagulopathy syndrome. Am J Surg 1996;172:405—10. anesthesia. Anesth Analg 2000;90:456—9.