Continuing Medical Education Article

Concise Definitive Review R. Phillip Dellinger, MD, FCCM, Section Editor

Intensive care unit management of the trauma patient

Edwin A. Deitch, MD; Saraswati D. Dayal, MD

LEARNING OBJECTIVES
On completion of this article, the reader should be able to:
1. Describe successful resuscitation of the patient with multiple injuries.
2. Explain the management of the patient with traumatic brain injury.
3. Use this information in the clinical setting.
Dr. Deitch has disclosed that he is/was the recipient of grant/research funds from Celgene. Dr. Dayal has disclosed that she
has no financial relationships with or interests in any commercial companies pertaining to this educational activity.
Lippincott CME Institute, Inc., has identified and resolved all faculty conflicts of interest regarding this educational activity.
Visit the Critical Care Medicine Web site (www.ccmjournal.org) for information on obtaining continuing medical education
credit.

Objective: The goal of this concise review is to provide an operative and nonoperative therapy. Although progress in the care of
overview of some of the most important intensive care unit issues the injured has been made, death due to uncontrolled bleeding,
and approaches that are unique to trauma patients as compared severe head injury, or the development of multiple organ dysfunction
with the general intensive care unit population. syndrome remains all too common in this patient population. Fur-
Study Selection: Clinical trials in trauma patients focusing on thermore, due to the potential nature of the injuries, the conundrum
hemorrhage control, issues in resuscitation, staged operative not infrequently arises that the optimal treatment for one injury or
repair of multiple injuries, the diagnosis and therapy of the organ system, such as preoperative permissive hypotension in ac-
abdominal compartment syndrome, and the treatment of trau- tively bleeding patients, may result in suboptimal or even deleterious
matic brain injury were identified on PubMed. therapy in the presence of another injury, such as traumatic brain
Conclusions: The intensive care unit care of the trauma patient injury. (Crit Care Med 2006; 34:2294–2301)
differs from that of other intensive care unit patients in many ways, KEY WORDS: trauma; brain injury; hemorrhagic shock; resusci-
one of the most important being the need to continuously integrate tation; organ failure; abdominal compartment syndrome

I
njuries are the leading cause of death unique aspects of intensive care unit (ICU) fusion, and the prevention/limitation of
in patients !45 yrs of age and the care of the trauma patient are focused on ischemia–reperfusion injury. Recogni-
third leading cause of death overall in head injury, hemorrhage control, resuscita- tion that shock causes a global ischemia–
the United States, claiming "100,000 tion, and the staged treatment of multiple reperfusion injury, which directly and in-
lives in 2002 (1). In fact, each year, "1.5 injuries, we will primarily focus on these top- directly leads to cellular and hence organ
million people are hospitalized as a result of ics rather than on issues that are common to injury, has led to an increasing emphasis
acute injury, and injury-related disability is a all ICU patients, such as ventilatory strategies, on the adequacy of volume resuscitation
major public health problem (2), with head infection, and nutrition. As in other areas of and a search for more effective resuscita-
injury, hemorrhage, and sepsis/multiple or- ICU care, management of patients in a trau- tion fluids. Although the colloid– crystal-
gan failure (MOF) being the three major ma-specific ICU or a closed unit is associated loid controversy continues and remains
causes of death and disability in this patient with improved clinical outcome (4). to be resolved (5), there is increasing data
population (3). Because the majority of the Generally speaking, the ICU care of the suggesting that Ringer lactate is proin-
trauma patient has overlapping treatment flammatory and thus may exacerbate the
phases, which include the resuscitative and inflammatory response and contribute to
operative phases. However, for clarity, each of the development of organ injury in shock
Professor and Chair, Department of Surgery, New
these phases will be discussed separately.
Jersey Medical School–University of Medicine and states (6 –9). Given these concerns, plus
Dentistry of New Jersey, Newark, NJ (EAD); Trauma/ the recent recognition that large-volume
Critical Care Surgeon, Hackensack University Medical Resuscitative Phase
Center, Westwood, NJ (SDD). resuscitation with crystalloid solutions
Copyright © 2006 by the Society of Critical Care The resuscitative phase has as its cen- contributes to the development of the
Medicine and Lippincott Williams & Wilkins tral goals the restoration of an effective abdominal compartment syndrome (10),
DOI: 10.1097/01.CCM.0000233857.94604.73 blood volume, optimization of tissue per- attention has refocused on the early re-

2294 Crit Care Med 2006 Vol. 34, No. 9

suscitation of trauma patients with hy- especially when blood older than 2 wks is resuscitated patients (10). A third issue
pertonic (7.5%) saline. Hypertonic saline administered (26, 27). This observation, also related to volume resuscitation is the
is a promising initial resuscitation op- plus the fact that blood is immune sup- end point of resuscitation. In the severely
tion, and multiple investigators have pressive (28, 29) and that ICU patients injured trauma patient, blood pressure
shown experimentally that at low vol- can be safely managed with hemoglobin may be restored by vasoconstriction;
umes (4 mL/kg), it is more effective at levels in the range of 7g/dL (30, 31), has thus, restoration of blood pressure or
restoring the extravascular volume, car- led to the emergence of a selective trans- even urine output does not ensure that
diac output, and organ perfusion than fusion policy, in which prophylactic organ blood flow and tissue perfusion
large-volume resuscitation with Ringer transfusions are no longer routinely ad- have been adequately restored. In this
lactate (11–14). Another important as- ministered. The mechanisms by which setting, arterial base deficit or serum lac-
pect of hypertonic saline is its ability to “old” blood potentiates organ failure has tate have been found to be better indica-
limit hemorrhagic shock-induced immu- been investigated in animal models and tors of the adequacy of tissue perfusion
nosuppression and organ injury in clini- seems to be related to the proinflamma- and hence volume resuscitation than
cally relevant animal models (15–17), al- tory nature of plasma and lipid factors blood pressure or urine output (37). A
though there is some evidence that the released/produced during the storage pe- worsening base deficit or serum lactate
administration of hypertonic saline after riod (32) and to the inability of stored red has been shown to correlate with ongoing
the initial resuscitation period may be blood cells to traverse the microcirculation, blood loss or inadequate volume resusci-
deleterious due to its ability to stimulate resulting in microcirculatory dysfunction tation, whereas improvements in these
neutrophil function (18). The largest clini- (33). Because large-volume blood trans- variables are indicative of adequate vol-
cal trial of hypertonic saline vs. Ringer lac- fusions are necessary in many trauma ume resuscitation. In severely injured pa-
tate tested hypertonic saline’s ability to im- patients with hemorrhagic shock, one po- tients, because the period of volume re-
prove survival and limit organ injury when tential solution is the use of blood sub- suscitation may last 24 – 48 hrs, serial
administered in the field (19). Overall sur- stitutes (34). However, until recently, en- measurements are important. The resus-
vival was similar between the two groups, thusiasm for stroma-free hemoglobin citative goal should be to reduce and keep
although survival was increased in the sub- solutions has been low based on the re- the base deficit at less than #2 mmol/L
group of trauma patients requiring emer- sults of a phase III clinical trial, in which and serum lactate at !1.5 mEq/L. In fact,
gency surgery. In addition, the hypertonic the administration of diaspirin cross- prospective studies have documented
group had a decreased prevalence of acute linked hemoglobin to trauma patients both a correlation between the magni-
respiratory distress syndrome, renal failure, was associated with almost a three-fold tude of the initial base deficit and survival
and coagulopathy. Nonetheless, at the cur- increase in mortality (46% vs. 17%) (35). and between survival and clearance of the
rent time, due to a paucity of clinical trials, In contrast to this earlier study, recent base deficit; those patients who cleared
there are not enough data to determine clinical trials with human polymerized their base deficits within 2 days had a
whether initial hypertonic resuscitation is hemoglobin (PolyHeme) are encouraging high survival rate, whereas only 13% of
superior to standard crystalloid resuscita- (36). This difference in clinical outcome those who had not cleared their deficit by
tion of the trauma patient (20). seems to be related to the fact that this 2 days survived (38, 39). As an aside, a
Another encouraging approach is the newer polymerized hemoglobin-based persistently elevated base deficit, despite
use of resuscitation fluids containing an- blood substitute, in contrast to the older what should be adequate volume resusci-
tioxidants. Three clinical trials have been products, does not avidly bind nitric ox- tation, indicates ongoing hemorrhage in
published showing that splanchnic- ide and hence does not act as a vasocon- the trauma patient population and may
directed antioxidant therapy helps pre- strictive agent. Thus, although hemo- be an indication for urgent surgery (40).
vent MOF in trauma patients (21–23). globin solutions can potentially limit Lastly, identifying all significant inju-
These trials were based on experimental blood transfusion, caution must be ex- ries is a key part of the resuscitative phase
studies indicating that splanchnic blood ercised because other polymerized he- of trauma care. Missed injuries are not
flow is disproportionately decreased after moglobin solutions seem to manifest uncommon in trauma patients (41– 43),
injury, stress, or shock in trauma and ICU some degree of vasoconstrictive activity and the consequences of missed injuries
patients and in clinical gastric tonometry (34), and the administration of a resus- may be significant because many of these
studies indicating that gut ischemia is a citative fluid that has vasoconstrictive missed injuries require prompt operative
better predictor of the development of properties to a hypovolemic patient could treatment, and deaths from missed inju-
acute respiratory distress syndrome and be catastrophic. ries are not rare (43). The highest preva-
multiple organ dysfunction syndrome A second issue related to volume re- lence of missed significant injuries oc-
than global indices of oxygen delivery suscitation of the trauma patient is the curs in trauma patients with a decreased
(24). In addition, based on an enlarging development of abdominal compartment sensorium, hemodynamic instability, or
investigative effort, other novel resuscita- syndrome (ACS). Because the treatment substance abuse.
tive fluids, such as ethyl pyruvate (25), of ACS is surgical, this topic will be dis-
that have pharmacologic and volume re- cussed below in the operative phase sec- Operative Phase
storative effects are being actively tested tion of the review. However, it is impor-
in preclinical studies. tant to stress that there is a correlation Although the specifics of the operative
The role of blood transfusions in trauma between the magnitude of crystalloid care of the trauma patient are beyond the
patients has also undergone an intense re- fluid resuscitation and the development scope of this review, certain aspects are
evaluation based on clinical studies show- of ACS, and patients receiving supranor- important for ICU management. An ex-
ing that blood transfusions are an indepen- mal resuscitative regimens were found to ample is when the initial operation must
dent predictor of the development of MOF, have twice the rate of ACS as normally be terminated before the definitive repair

Crit Care Med 2006 Vol. 34, No. 9 2295

of all injuries because of hemodynamic and retroperitoneal edema associated in patients with long-bone fractures.
instability or because of the need to limit with massive fluid resuscitation in pa- Thus, an important aspect of ICU care is
the development of coagulopathy, acido- tients with hemorrhagic shock (56). The the integration and timing of secondary
sis, and hypothermia—the so called dam- diagnosis of ACS is made by measuring operations in the therapy of the multiply
age-control laparotomy. The concept of the abdominal pressure through a Foley injured patient.
damage control is to control hemorrhage catheter placed in the bladder. ACS is de- Lastly, the ICU team must recognize
by repair or packing and to control fecu- fined as the combination of 1) a urinary the potential need for prompt operations
lent contamination of the peritoneal cav- bladder pressure of "25 mmHg, 2) pro- or reoperations on properly selected pa-
ity. This concept was learned from the gressive organ dysfunction (urinary output tients who are showing continued signs
care of trauma patients with multiple in- of !0.5 mL·kg#1·hr#1 or PaO2/FIO2 of of organ dysfunction before MOF be-
juries, in which utilization of a damage- !150 or peak airway pressure of "45 cm comes established. This concept is espe-
control approach was shown to improve H2O or cardiac index of !3 L·min#1·m#2 cially important in patients who have un-
survival (44 – 46). The rationale behind a despite resuscitation), and 3) improved or- dergone previous abdominal surgery, in
damage-control laparotomy is the clinical gan function after decompression. Un- whom the abdomen may be the source of
observation that prolonged attempts at treated, ACS is lethal, with a mortality rate infection or have retained necrotic tis-
definitive control of intraabdominal inju- of 100% documented in subsets of patients sues, large hematomas, or fluid collec-
ries can result in hemodynamic instabil- with ACS who did not undergo abdominal tions that are exacerbating the inflamma-
ity, acidosis, and surgically uncontrolla- decompression (54). Thus, a high index of tory response.
ble coagulopathic bleeding. If the patient suspicion leading to an early diagnosis is
survives the operation, the prevalence of critical because the treatment of ACS with Traumatic Brain Injury
postoperative MOF is high. The damage- a decompressive laparotomy improves or-
control approach allows for the rapid gan dysfunction. In fact, surgical treatment Management of the patient with trau-
transport of the patient to the ICU for of increased intraabdominal pressure leads matic brain injury (TBI) is a major por-
continued resuscitation and correction of in most instances to a rapid and profound tion of the ICU care of the multiple-
metabolic abnormalities. A planned reop- correction of the physiologic abnormalities. trauma patient because approximately
eration is safer and easier in patients who Return of urine output is almost immedi- 500,000 patients in the United States sus-
have been warmed, fully resuscitated, and ate, as is an increase in cardiac output and tain a TBI, of whom about 10% (50,000)
who have had their coagulopathy and ac- a decrease in peak airway pressures (55). die, making the brain one of the most
idosis corrected. In patients with ongoing The abdomen may be opened in the ICU or frequently and important parts of the
surgical bleeding after damage-control in the operating room, depending on the body injured in polytrauma patients (63,
laparotomy (patients with major liver in- patient’s condition. Once a patient has an 64). Mortality due to TBI occurs through-
juries or pelvic fractures), angiographic open abdomen, whether due to ACS or as a out the postinjury period (3), with the
control of bleeding has been shown to be consequence of a damage-control laparot- most severely injured patients (Glasgow
successful (47–50). Although the data are omy, the exposed abdominal viscera must Coma Score of !8) requiring intubation
limited, in patients with medical (coagu- be managed. Several techniques are avail- at the scene of the accident or in the
lopathic) bleeding who do not respond to able, including the use of vacuum-assisted emergency department (65). It is now
standard therapy, there is an emerging closure (57). Regardless of the method used well established that episodes of hypoxia
role for factor VIIa therapy (51). to manage the exposed abdominal viscera, or hypotension exacerbate the extent of
A second aspect to be cognizant of in the the goals of management include preven- the brain injury and are associated with a
ICU management of the trauma patient is tion of evaporative water loss, limitation of worsened long-term outcome (66, 67). In
ACS. ACS can be viewed as a reversible bacterial contamination, drainage of ab- this context, failure to maintain a systolic
mechanical cause of MOF that is related to dominal fluid, prevention of abdominal vis- blood pressure of "90 mm Hg and a PaO2
increased intraabdominal pressure (52, 53). ceral desiccation and fistula formation, and of "60 mm Hg is associated with in-
As the intraabdominal pressure increases, optimization of the abdominal cavity for creased morbidity and mortality (68, 69).
abdominal visceral perfusion decreases, secondary abdominal wall closure. Thus, avoidance of hypoxia and hypoten-
ventilation is impaired, and cardiac out- In addition to the other surgical is- sion are critical during the resuscitative
put declines (53–55). Clinically, this sues, early fixation of long-bone fractures and ICU phases of care. Because cerebral
manifests as a decreasing urine output, in severely injured patients seems to have edema is a contributing factor to the pro-
inadequate ventilation associated with el- resulted in a dramatic reduction in the gression of TBI and hypertonic saline has
evated peak airway pressures, and hypo- prevalence of acute respiratory distress been shown to reduce intracranial pres-
tension. Patients sustaining multiple syndrome and MOF as compared with sure (ICP) in experimental brain injury
trauma, massive hemorrhage, prolonged those patients treated with prolonged (70), several studies have investigated the
operations with massive volume resusci- traction (58 – 60). However, there is some prehospital use of small-volume hyper-
tation, and those requiring intraabdomi- evidence that patients with significant tonic saline in these patients (19, 71). In
nal packing to control bleeding or who thoracic injuries or head trauma may be fact, a meta-analysis published in 1997 of
have massively distended bowel are at the harmed by early operative fracture fixa- eight prospective trials indicated that hy-
highest risk of developing the ACS. ACS tion due to a higher prevalence of acute pertonic saline– dextran resuscitation in-
can also develop in patients in severe respiratory distress syndrome and sec- creased survival from 27% to 38% in pa-
hemorrhagic shock without an abdomi- ondary brain injury, respectively (61, 62). tients with very severe TBI (72). However,
nal or retroperitoneal injury. This phe- Despite these subgroups, most evidence a recent, large, controlled trial in hypo-
nomenon is known as secondary ACS and supports early fracture fixation as an ef- tensive TBI patients did not show any
is due to progressive abdominal visceral fective method of reducing organ failure early or long-term benefits of hypertonic

2296 Crit Care Med 2006 Vol. 34, No. 9

saline administration (73). Thus, the use degrees to reduce the ICP and fluid brain edema, multiple clinical trials have
of hypertonic saline in these patients re- should be administered to have a mean clearly shown that steroids are not help-
mains controversial, although studies in arterial pressure of 90 mm Hg. In addi- ful (94, 95). On the other hand, there is
pediatric patients indicate that refractory tion, control of fever, analgesia, and se- abundant class I data documenting the
intracranial hypertension can be success- dation and ventilatory management to beneficial effects of a 7-day course of sei-
fully treated with hypertonic saline (74, 75). ensure adequate oxygenation and ventila- zure prophylaxis in patients with severe
Although the role of hypotensive resusci- tion are important adjuvant therapies. Al- TBI (96).
tation in patients with surgical sites of though, in the past, patients with severe
bleeding to limit blood loss before oper- TBI were treated with hyperventilation to Maintenance of Organ Function
ative or angiographic control of the improve cerebral blood flow, prospective and Additional ICU Therapies
bleeding remains controversial (76, 77), randomized trials have shown that hyper-
it is clear that hypotensive resuscitation ventilation (PaCO2, !25–30 mm Hg) Strategies directed at the protection
is absolutely contraindicated in severe worsens the neurologic outcome (85). and maintenance of pulmonary, cardiac,
TBI patients because it is associated with This has resulted in the recommendation intestinal, and renal function are part of
a poor neurologic outcome (77). In addi- that the PaCO2 of ventilated patients be the ICU care of the trauma patient. In
tion, because patients with TBI may have kept in the range of 30 –35 mm Hg (63). fact, many of the ICU therapies shown in
surgically treatable lesions (i.e., subdural Interestingly, although evidence-based prospective, randomized, clinical trials to
or epidural hematomas), it is critical to guidelines exist to direct the manage- reduce mortality in other populations are
obtain a head computed tomographic ment of severe TBI, such as the use of ICP commonly used in the treatment of
scan as early as feasible to guide therapy monitoring, limiting hyperventilation, trauma patients, although they remain to
and serve as a baseline for further scans. maintaining cerebral perfusion pressure, be fully validated in the trauma popula-
In fact, daily head computed tomographic and so on (63), there remains consider- tion. These include: 1) glycemic control
scans are frequently required in the pa- able institutional variation in following (97, 98), 2) early goal-directed hemody-
tient with severe TBI. these guidelines (86). namic therapy beginning in the emer-
In addition to avoiding hypoxia and If the ICP remains at "20 mm Hg, gency department (99), 3) the use of early
hypotension, a major therapeutic goal in despite adequate sedation, then addi- adequate empirical antibiotic use for pa-
the care of the patient with TBI is the tional therapies are utilized. If hyperther- tients with suspected pneumonia (100),
maintenance of adequate cerebral perfu- mia ("38.5°C) is present, especially when 4) the use of lower tidal volumes and
sion. Because increased ICP is associated the ICP is elevated, then it should be limitation of plateau pressures to reduce
with impaired cerebral perfusion, the aggressively lowered because hyperther- pulmonary injury in ventilated patients
placement of an invasive pressure moni- mia in this situation has been associated (101), 5) use of low-dose steroids in pa-
tor to measure ICP is a key element of with an increased mortality rate and a tients with septic shock associated with
care (78, 79). Because patients with se- worsened neurologic outcome (87). In adrenal insufficiency (102), and 6) the
vere TBI (i.e., Glasgow Coma Score of fact, several groups have proposed the creation of a closed ICU (103). A number
!8) have a 60% chance of having an use of deliberate mild hypothermia (33– of other therapeutic organ-directed strat-
increased ICP (80), it has been recom- 35°C) to limit cerebral oxygen demand, egies are important in trauma as in other
mended that ICP monitoring be routinely although its effectiveness remains con- patient populations. Some of these strat-
employed in this patient population (63, troversial (88). In the normothermic pa- egies support several systems, such as
65). The two key hemodynamic factors tient with an increased ICP and a ventric- early enteral alimentation of the trauma
associated with cerebral perfusion are the ulostomy, cerebrospinal fluid drainage is patient, which helps preserve intestinal
ICP and the cerebral perfusion pressure an effective strategy to decrease the ICP function and limit infectious complica-
(cerebral perfusion pressure $ mean ar- (89). If the patient does not have a ven- tions (104). Other strategies, such as el-
terial pressure minus the ICP). Based on triculostomy or if cerebrospinal fluid evation of the head of the bed of venti-
class II data showing that functional out- drainage is ineffective, mannitol should lated patients, seems to both reduce the
come is inversely related to the longer be administered because of its diuretic prevalence of pneumonia and to better pre-
the percentage of time the ICP is "20 effect and its ability to decrease the ICP, serve pulmonary function (105). Methods
mm Hg, it seems that attempts to lower increase the cerebral perfusion pressure, of supporting renal function, such as low-
the ICP should be initiated when it is and thereby improve cerebral blood flow dose dopamine, have not been found to be
"20 mm Hg (63, 68, 81). Likewise, there (90). Although prospective trials have effective (106), and thus, currently, the best
are excellent class II data showing that shown that mannitol is more effective way to limit renal failure is to avoid under-
maintaining a cerebral perfusion pres- than barbiturate therapy in improving resuscitation and to promptly diagnose and
sure of "60 mm Hg is associated with a mortality and neurologic outcome (91), treat infectious complications. Once renal
reduction in morbidity and mortality (82, high-dose barbiturate therapy needs to be failure has occurred, continuous veno-
83) and reduced long-term neurologic considered for intracranial hypertension venous hemodialysis seems to be superior
disability (84). Thus, once an ICP moni- that is refractory to maximal medical and to hemodialysis because it avoids the need
tor is in place, therapy is directed at surgical therapy, although it is associated for systemic anticoagulation and is less
maintaining an adequate cerebral perfu- with substantial complications (92). Like- likely to cause hypotensive episodes (107).
sion pressure, either by reducing the ICP, wise, in desperate situations of refractory Cardiac issues can complicate the re-
increasing the mean arterial pressure, or intracranial hypertension, decompressive covery of trauma and other ICU patients
both. craniotomy should be considered as a last and range from cardiac contusion to the
As a rule of thumb, whenever possible, resort (93). Although it was hoped that consequences of preexistent coronary ar-
the bed should be kept elevated at 30 steroids would be effective in reducing tery disease. Many trauma patients pre-

Crit Care Med 2006 Vol. 34, No. 9 2297

senting to the ICU present with tachyar- have included, insulin-like growth factor-1, unnecessarily than to miss or delay the
rhythmias, hypotension, or both. Before insulin, beta-blockade, and the use of ana- diagnosis of a deep infection or necrotizing
treatment is initiated, it is important to bolic androgens with minimal virilizing ac- fasciitis. Once diagnosed, the principles of
differentiate their pathogeneses. The dif- tivity, such as oxandrolone. Although en- therapy for soft-tissue infections are rela-
ferential diagnosis includes, but is not couraging results from prospective tively straight forward and include ade-
limited to, preexisting cardiac dysfunc- randomized trials in burn patients (110 – quate drainage of the site, evacuation of
tion, sepsis, under-resuscitation, anxiety, 112) have resulted in the Food and Drug hematomas if present, removal of devital-
and pain. A specific issue related to Administration approving oxandrolone for ized tissues, and the appropriate institution
trauma patients is blunt cardiac injury use as an anabolic agent in severely stressed of antibiotics followed by aggressive wound
(BCI) (108, 109). Trauma victims who patients, there are no class I or II data in care.
have had direct precordial impact or nonburn trauma patients to support its use
crush injuries are known to be at in- as of yet. Conclusion
creased risk for BCI, formerly known as In addition to organ-directed thera-
myocardial contusion. There is no diag- pies, the prompt and efficient diagnosis of In this article, we have mainly focused
nostic gold standard for the diagnosis of infectious complications is important be- on areas of ICU management that are
BCI. The presentation may vary from mi- cause, after major trauma, infectious unique to trauma patients. Although the
nor electrocardiographic or cardiac en- complications are relatively common and principles of the critical care of the trauma
zyme changes to cardiovascular collapse. are a major cause of morbidity and mor- patient are similar to those of the care of
The right heart is the most common site tality (113). Because many of the nonin- other severely ill patients, the presence of
of BCI secondary to its anatomic position fectious stimuli associated with trauma, injuries to a wide variety of organ systems
under the sternum. Most patients with such as soft-tissue and bony injuries, and the risk and consequences of severe
BCI recover without any sequelae, and traumatic pancreatitis, shock, or brain hemorrhage can complicate diagnosis and
due to its frequent benign nature, the injury, can generate a septic-like state the delivery of necessary care. In addition,
diagnosis of BCI may never even be sus- and even lead to acute respiratory distress the optimal treatment for one injury, such
pected. However, BCI should be sus- syndrome or multiple organ dysfunction as maintaining an elevated mean arterial
pected in patients with chest trauma who syndrome, it can be clinically difficult to pressure in patients with head injuries, can
present with a poor cardiovascular re- accurately differentiate the infected from confound the treatment of other injuries,
sponse to therapy, and in this subgroup the noninfected, seemingly septic trauma such as aortic transections. Likewise, the
of patients, inotropic support may be re- patient without an apparent source of in- optimal timing of the definitive operative
quired. An incompletely resolved issue is fection (i.e., abdomen, wound, or injury treatment of secondary injuries, such as
the use of beta-blockade in the trauma site). The approach to this conundrum fractures, may be influenced by the treat-
patient because its use can be beneficial remains to be fully resolved, and little if ment needed for the patient’s other injuries
or detrimental, depending on the situa- any prospective data are available in this and the patient’s hemodynamic status.
tion. For example, beta-blockers should area. Nonetheless, after doing a “fever Thus, there is an intricate balance between
not be instituted in patients who are ac- workup,” most trauma intensivists would the operative and nonoperative care of the
tively being resuscitated unless the pa- consider starting empirical antibiotics, trauma patient that is observed in few other
tient has previously been on a beta- especially in high-risk patients, and be- patient populations.
blocker for cardiac dysfunction. On the cause of the risk of the emergence of
other hand, if there is clinical and labo- antibiotic-resistant organisms, plan on REFERENCES
ratory evidence of myocardial injury, be- stopping these antibiotics in 3–5 days if
ta-blocker use should be entertained. no site of infection is documented during 1. Anderson RN, Smith BL: Deaths: Leading
In addition to these topics, the notion this time period. In addition to pneumo- causes for 2002. Natl Vital Stat Rep 2005;
of pharmacologic manipulation of the nia, which is common in trauma ICU 53:1– 89
metabolic response to injury has been an patients, as in all mechanically ventilated 2. Mackenzie EJ, Fowler CJ: Epidemiology. In:
area of intense study because it is well rec- ICU patients, trauma patients are at in- Trauma. Fifth Edition. Moore EE,
Feliciano DV, Mattox KL (Eds). New York,
ognized that trauma- or sepsis-induced hy- creased risk of developing wound and
McGraw-Hill, 2004, pp 21–39
percatabolism and muscle wasting is a ma- soft-tissue infections. Because trauma in-
3. American College of Surgeons Committee
jor cause of morbidity and mortality. Based juries can even devitalize tissues, trauma on Trauma: Advanced Trauma Life Support.
on research in this field, two major ap- patients are uniquely susceptible to the Seventh Edition. Chicago, American Col-
proaches have emerged. The first is nutri- development of soft-tissue and wound lege of Surgeons, 2004, pp 1–10
ent pharmacology, in which specific nutri- sepsis at both operative and nonoperative 4. Carson SS, Stocking C, Podsadecki T, et al:
ents and growth factors are administered injury sites, and such infections occur in Effects of organizational change in the
for their ability to support the physiologic about 10% of trauma ICU patients (113). medical intensive care unit of a teaching
function and repair of certain organs and For this reason, daily examination of all hospital: A comparison of open and closed
tissues and to augment the immune system incisions and sites of injury must be per- formats. JAMA 1996; 276:322–328
5. Roberts I, Alderson P, Bunn F, et al: Col-
and promote wound healing (110). A sec- formed, looking for signs of infection, and
loids versus crystalloids for fluid resuscita-
ond approach involves the use of hormones in patients who are septic and deteriorat-
tion in critically ill patients. Cochrane Da-
or anabolic agents to improve wound heal- ing, all closed wounds, no matter how tabase Syst Rev 2004; 4:CD000567
ing, improve immune function, or support seemingly innocuous, must be considered a 6. Rhee P, Burris D, Kaufmann C, et al: Lac-
organ function in addition to limiting the potential septic source and thoroughly in- tated Ringer’s solution resuscitation causes
hypermetabolic response. Anabolic agents vestigated. In essentially all cases, it is bet- neutrophil activation after hemorrhagic
used to improve the metabolic response ter to locally explore a wound or injury site shock. J Trauma 1998; 44:313–319

2298 Crit Care Med 2006 Vol. 34, No. 9

 7. Koustova E, Stanton K, Gushchin V, et al: Antioxidant therapy in the prevention of Lactate clearance and survival following in-
Effect of lactated Ringer’s solutions on hu- organ dysfunction syndrome and infectious jury. J Trauma 1993; 35:584
man leukocytes. J Trauma 2002; 52:872– 878 complications after trauma: Early results of 40. Davis JW, Mackersie RC, Holbrook TL, et al:
8. Alam HB, Sun L, Ruff P, et al: E- and P- a prospective randomized study. Am Surg Base deficit as an indicator of significant
selectin expression depends on the resusci- 1999; 65:478 – 483 abdominal injury. Ann Emerg Med 1991;
tation fluid used in hemorrhaged rats. 23. Nathens AB, Neff MJ, Jurkovich GJ, et al: 20:842– 844
J Surg Res 2000; 94:145–152 Randomized, prospective trial of antioxi- 41. Enderson BL, Reath DB, Meadors J, et al:
9. Rhee P, Wang D, Ruff P, et al: Human dant supplementation in critically ill surgi- The tertiary trauma survey: A prospective
neutrophil activation and increased adhe- cal patients. Ann Surg 2002; 236:814 – 822 study of missed injury. J Trauma 1990; 30:
sion by various resuscitation fluids. Crit 24. Marik PE: Gastric intramucosal pH: A better 666 – 669
Care Med 2000; 28:74 –78 predictor of multiorgan dysfunction syn- 42. Houshian S, Larsen MS, Holm C: Missed
10. Balogh Z, McKinley BA, Cocanour CS, et al: drome and death than oxygen-derived vari- injuries in a level I trauma center. J Trauma
Supranormal trauma resuscitation causes ables in patients with sepsis. Chest 1993; 2002; 52:715–719
more cases of abdominal compartment syn- 104:225–229 43. Brooks A, Holroyd B, Riley B: Missed injury
drome. Arch Surg 2003; 138:637– 643 25. Sappington PL, Fink ME, Yang R, et al: in major trauma patients. Injury 2004; 35:
11. Kramer GC: Hypertonic resuscitation: Physi- Ethyl pyruvate provides durable protection 407– 410
ologic mechanisms and recommendations for against inflammation-induced intestinal ep- 44. Hirschberg A, Wall MJ Jr, Mattox KL:
trauma care. J Trauma 2003; 54(5 Suppl): ithelial barrier dysfunction. Shock 2003; 20: Planned reoperation for trauma: A two year
S89 –S99 521–528 experience with 124 consecutive patients.
12. Kreimeier U, Messmer K: Small-volume re- 26. Moore, Fa, Moore EE, Sauaia A: Blood J Trauma 1994; 37:365–369
suscitation: From experimental evidence to transfusion: An independent risk factor for 45. Rotondo MF, Schwab CW, McGonigal MD,
clinical routine. Advantages and disadvan- postinjury multiple organ failure. Arch et al: ‘Damage Control’: An approach for
tages of hypertonic solutions. Acta Anaes- Surg 1997; 620:620 – 625 improved survival in exsanguinating pene-
thesiol Scand 2002; 46:625– 638 27. Zallen G, et al: Age of transfused blood is an trating abdominal injury. J Trauma 1993;
13. Kien ND, Antognini JF, Reilly DA, et al: independent risk factor for postinjury mul- 35:375–382
Small-volume resuscitation using hypertonic tiple organ failure. Am J Surg 1999; 178: 46. Johnson JW, Gracias VH, Schwab CW, et al:
saline improves organ perfusion in burned 570 –572 Evolution in damage control for exsanguinat-
rats. Anesth Analg 1996; 83:782–788 28. Nielsen HJ: Detrimental effects of perioper- ing penetrating abdominal injury. J Trauma
14. Mouren S, Delayance S, Mion G, et al: ative blood transfusion. Br J Surg 1995; 2001; 51:261–271
Mechanisms of increased myocardial con- 82:582–587 47. Kushimoto S, Arai M, Aiboshi J, et al: The
tractility with hypertonic saline solutions in 29. Offner PJ, Moore EE, Biffl WL, et al: In- role of interventional radiology in patients
isolated blood-perfused rabbit hearts. creased rate of infection associated with requiring damage control surgery. J Trauma
Anesth Analg 1995; 81:777–782 transfusions of old blood after severe injury. 2002; 54:171–176
15. Angle N, Hoyt DB, Coimbra R, et al: Hyper- Arch Surg 2002; 137:711–717 48. Johnson JW, Gracias VH, Gupta R, et al:
tonic saline resuscitation diminishes lung 30. Consensus conference: Perioperative red Hepatic angiography in patients undergo-
injury by suppressing neutrophil activation blood cell transfusion. JAMA 1988; 260: ing damage control laparotomy. J Trauma
after hemorrhagic shock. Shock 1998; 2700 –2703 2002; 52:1102–1106
9:164 –170 31. Hebert PC, Wells G, Blajchman MA, et al: A 49. Richardson JD, Franklin GA, Lukan JK, et
16. Pascual JL, Khwaja KA, Ferri LE, et al: multicenter, randomized, controlled clini- al: Evolution in management of hepatic
Hypertonic saline resuscitation attenuates cal trial of transfusion requirements in crit- trauma: A 25-year experience. Ann Surg
neutrophil lung sequestration and transmi- ical care. N Engl J Med 1999; 39:665– 670 2000; 232:324 –328
gration by diminishing leukocytes-endothe- 32. Silliman CC, Voelkel NF, Allard JD, et al: 50. Miller PR, Moore PS, Mansell E, et al: Ex-
lial interactions in a two-hit model of hem- Plasma and lipid from stored packed red ternal fixation or arteriogram in bleeding
orrhagic shock and infection. J Trauma blood cells cause acute lung injury in an pelvic fracture: Initial therapy guided by
2003; 54:121–132 animal model. J Clin Invest 1998; 101: markers of arterial hemorrhage. J Trauma
17. Shields CJ, Winter DC, Manning BJ, et al: 1458 –1467 2003; 54:437– 443
Hypertonic saline infusion for pulmonary 33. Marik PE, Sibbald WJ: Effect of stored-blood 51. Dutton RP, McCunn M, Hyder M, et al:
injury due to ischemia-reperfusion. Arch transfusion on oxygen delivery in patients Factor VIIa for correction of traumatic co-
Surg 2003; 138:9 –14 with sepsis. JAMA 1993; 269:3024 –3029 agulopathy. J Trauma 2004; 57:709 –718
18. Murao Y, Hoyt DB, Loomis W, et al: Does 34. Moore EE: Blood substitutes: The future is 52. Balogh Z, McKinley BA, Holcomb JB, et al:
the timing of hypertonic saline resuscita- now. J Am Coll Surg 2003; 196:1–17 Both primary and secondary abdominal
tion affect its potential to prevent lung 35. Sloan EP, Koenigsberg M, Gens D, et al: compartment syndrome can be predicted
damage. Shock 2000; 14:18 –23 Diaspirin cross linked hemoglobin (DCLHb) early and are harbingers of multiple organ
19. Mattox KL, Maningas PA, Moore EE, et al: in the treatment of severe traumatic hem- failure. J Trauma. 2003; 54:848 – 859
Prehospital hypertonic saline/dextran infu- orrhagic shock: A randomized controlled 53. Burch JM, Moore EE, Moore FA, et al: The
sion for post-traumatic hypotension: The efficacy trial. JAMA 1999; 282:1857–1864 abdominal compartment syndrome. Surg
USA multicenter trial. Ann Surg 1991; 213: 36. Gould SA, Moore EE, Hoyt DB, et al: The Clin North Am 1996;76:833– 842
482– 491 life sustaining capacity of human polymer- 54. Cullen DJ, Coyle JP, Telpick R, et al: Car-
20. Bunn F, Roberts I, Tasker R, et al: Hypertonic ized hemoglobin when red cells might be diovascular, pulmonary, and renal effects of
versus near isotonic crystalloid for fluid resus- unavailable. J Am Coll Surg 2002; 195: massively increased intra-abdominal pres-
citation in critically ill patients. Cochrane Da- 445– 452 sure in critically ill patients. Crit Care Med
tabase Syst Rev 2004; 3:CD002045 37. Davis JW, Shackford SR, Mackersie RC, et 1989; 17:118 –121
21. Barquist E, Kirton O, Windsor J, et al: The al: Base deficit as a guide to volume resus- 55. Shelly MP, Robinson JW, Hesford JW, et al:
impact of antioxidant and splanchnic- citation. J Trauma 1988; 28:1464 Haemodynamic effects following surgical
directed therapy on persistent uncorrected 38. Davis KW, Kaups KL: Base deficit in the release of increased intra-abdominal pres-
gastric mucosal pH in the critically injured elderly: A marker of severe injury and death. sure. Br J Anaesth 1987; 59:800 – 805
trauma patient. J Trauma 1998; 44:355–360 J Trauma 1998; 45:873– 877 56. Maxwell RA, Fabian TC, Croce MA, et al:
22. Porter JM, Ivatoury RR, Azimuddin K, et al: 39. Abramson D, Scalea TM, Hitchcock R, et al: Secondary abdominal compartment syn-

Crit Care Med 2006 Vol. 34, No. 9 2299

 drome: A underappreciated manifestation of tion of shock and brain injury: Short- and logic outcome in head-injured patients.
severe hemorrhagic shock. J Trauma 1999; long-term effects. J Trauma 1997; 42: J Neurotrauma 1992; 9:S349 –S358
47:995–999 592– 601 85. Muizelaar JP, Marmarou A, Ward DJ, et al:
57. Barker DE, Kaufman HJ, Smith LA, et al: 71. Vassar MJ, Fischer RP, O’Brien PE, et al: A Adverse effects of prolonged hyperventila-
Vacuum pack technique of temporary ab- multicenter trial for resuscitation of injured tion in patients with severe head injury: A
dominal closure: A 7-year experience with patients with 7.5% sodium chloride: The randomized control trial. J Neurosurg 1991;
112 patients. J Trauma 2000; 48:201–207 effect of added dextran 70. The Multicenter 75:731–739
58. Riska EB, von Bonsdoff H, Hakkinen S, et Group for the Study of Hypertonic Saline in 86. Bulger EM, Nathens AB, Rivara FP, et al:
al: Prevention of fat embolism by early in- Trauma Patients. Arch Surg 1993; 128: Management of severe head injury: Institu-
ternal fixation of fractures in patients with 1003–1011 tional variations in care and effect on out-
multiple injuries. Injury 1976; 8:110 –116 72. Wade CE, Kramer GC, Grady JJ, et al: Effi- come. Crit Care Med 2002; 30:1870 –1876
59. Johnson KD, Cadmabi A, Siebert GB: Inci- cacy of hypertonic 7.5% saline and 6% dex- 87. Manno EM, Farmer JC: Acute brain injury: If
dence of adult respiratory distress syn- tran-70 in treating trauma: A meta-analysis hypothermia is good, then is hyperthermia
drome in patients with multiple musculo- of controlled studies. Surgery 1997; 122: bad? Crit Care Med 2004; 32:1489 –1495
skeletal injuries: Effect of early operative 609 – 616 88. Yamamoto T, Mori K, Maeda M: Assessment
stabilization of fractures. J Trauma 1985; 73. Cooper DJ, Myles PS, McDermott FT, et al: of prognostic factors in severe traumatic
25:375–384 Prehospital hypertonic saline resuscitation brain injury patients treated by mild thera-
60. Seibel R, LaDuca J, Hassett JM, et al: Blunt of patients with hypotension and severe peutic cerebral hypothermia therapy. Neu-
multiple trauma (ISS 36), femur traction traumatic brain injury. JAMA 2004; 291: rol Res 2002; 30:1630 –1635
and pulmonary failure-septic state. Ann 1350 –1357 89. Bullock MR, Chesnut RM, Clifton GL, et al:
Surg 1985; 202:283–295 74. Peterson B, Khanna S, Fisher B, et al: Pro- Critical pathway for the treatment of estab-
61. Pape HC, Auf’m’kolk M, Paffrath TR, et al: long hypernatremia controls elevated intra- lished intracranial hypertension. J Neuro-
Primary intramedullary femur fixation in cranial pressure in head-injured pediatric pa- trauma 2000; 17:537–547
multiple trauma patients with associated tients. Crit Care Med 2000; 28:1136 –1143 90. Roberts I, Schierhout G, Wakai, A: Mannitol
lung contusion: A cause of posttraumatic 75. Khanna S, Davis D, Peterson B, et al: Use of for acute traumatic brain injury. Cochrane
ARDS? J Trauma 1993; 35:540 –547 hypertonic saline in the treatment of severe Database Syst Rev 2003; 2:CD001049
62. Townsend RH, Lheureau T, Protech J, et al: refractory posttraumatic intracranial hyper- 91. Schwartz ML, Tator CH, Rowed DW: The
Timing fracture repair in patients with severe tension in pediatric traumatic brain injury. University of Toronto Head Injury Treat-
brain injury (Glasgow Coma Score !9). Crit Care Med 2000; 28:1144 –1151
ment Study: A prospective randomized
J Trauma 1998; 44:977–982 76. Bickell WH, Wall MJ Jr, Pepe PE, et al:
comparison of pento-barbital and mannitol.
63. Brain Trauma Foundation, American Associ- Immediate versus delayed fluid resuscita-
Can J Neurol Sci 1984; 11:434 – 440
ation of Neurological Surgeons, Congress of tion for hypotensive patients with penetrat-
92. Roberts I: Barbiturates for acute traumatic
Neurological Surgeons, Joint Section on Neu- ing torso injuries. N Engl J Med 1994; 331:
brain injury. Cochrane Database Syst Rev
rotrauma and Critical Care: Guidelines for the 1105–1109
2000; 2:CD000033
Management of Severe Traumatic Brain In- 77. Dutton RP, Mackenzie CF, Scalea TM: Hy-
93. Sahuquillo J, Arikan F: Decompressive crani-
jury: Cerebral Perfusion Pressure. New York, potensive resuscitation during active hem-
ectomy for the treatment of refractory high
Brain Trauma Foundation, March 14, 2003 orrhage: Impact on in-hospital mortality.
intracranial pressure in traumatic brain in-
64. Valadka AB: Injury to the cranium. In: J Trauma 2002; 52:1141–1146
jury. Cochrane Database Syst Rev 2006;
Trauma. Fifth Edition. Moore EE, Feliciano 78. Chestnut RM, Marshall LF, Klauber MR, et
1:CD003983
DV, Mattox KL (Eds). New York, McGraw- al: The role of secondary brain injury in
94. Saul TG, Ducker TB, Salcman M, et al:
Hill, 2004, pp 385– 406 determining the outcome from severe head
Steroids in severe head injury: A prospec-
65. Procaccio F, Stocchetti N, Citerio G, et al: injury. J Trauma 1993; 34:216 –222
Guidelines for the treatment of adults with 79. Bullock MR, Chestnut RM, Clifton GL, et al: tive randomized trial. J Neurosurg 1981;
severe head trauma (part I): Initial assess- Indications for intracranial pressure moni- 54:596 – 600
ment; evaluation and pre-hospital treat- toring. J Neurotrauma 2000; 17:479 – 491 95. Alderson P, Roberts I: Corticosteroids for
ment; current criteria for hospital admis- 80. Narayan RK, Kishore PR, Becker DP, et al: acute traumatic brain injury. Cochrane Da-
sion; systemic and cerebral monitoring. Intracranial pressure: To monitor or not to tabase Syst Rev. 2005; 1:CD000196
J Neurosurg Sci 2000; 44:1–10 monitor? A review of our experience with 96. Temkin NR, Dikmen SS, Wilensky AJ, et
66. Chesnut RM, Marshall LF, Klauber MR, et severe head injury. J Neurosurg 1982; 56: al: A randomized, double-blind study of
al: The role of secondary brain injury in 650 – 659 phenytoin for the prevention of post-
determining outcome from severe head in- 81. Eisenberg H, Frankowski R, Constant C, et traumatic seizures. N Engl J Med 1990;
jury. J Trauma 1993; 34:216 –222 al: High dose barbiturate control of elevated 323:497–502
67. Wald SL, Shackford SR, Fenwick J: The intracranial pressures in patients with se- 97. Van Den Berghe G, Wouters P, Weekers F,
effect of secondary insults on mortality and vere head injury. J Neurosurg 1998; 69: et al: Intensive insulin therapy in critically
long-term disability after severe head injury 15–23 ill patients. N Engl J Med 2001; 345:
in a rural region without a trauma system. 82. Kiening KL, Hartl R, Unterberg AW, et al: 1359 –1367
J Trauma 1993; 34:377–381 Brain tissue pO2-monitoring in comatose 98. Laird AM, Miller PR, Kilgo PD, et al: Relation-
68. Marmarou A, Anderson RL, Ward JD: Im- patients: Implications for therapy. Neurol ship of early hyperglycemia to mortality in
pact of ICP instability and hypotension on Res 1997; 19:233–240 trauma patients. J Trauma 2004; 56:
outcome in patients with severe head 83. Juul N, Morris GF, Marshall SB, et al: In- 1058 –1062
trauma. J Neurosurg 1991; 75:159 –166 tracranial hypertension and cerebral perfu- 99. Rivers E, Nguyen B, Havstad S, et al: Early
69. Fearnside MR, Cook RJ, McDougall P, et al: sion pressure: Influence on neurological de- goal-directed therapy in the treatment of
The Westmead Head Injury Project out- terioration and outcome in severe head severe sepsis and septic shock. N Engl J Med
come in severe head injury: A comparative injury. The Executive Committee of the In- 2001; 345:1368 –1377
analysis of pre-hospital, clinical and CT ternational Selfotel Trial. J Neurosurg 2000; 100. Garnacho-Montero J, Garcia-Garmendia JL,
variable. Br J Neurosurg 1993; 7:267–279 92:1– 6 Barrero-Almodovar A, et al: Impact of inad-
70. Anderson JT, Wisner DH, Sullivan PE, et al: 84. Robertson CS, Contant CF, Narayan RK, et equate empirical antibiotic therapy on the
Initial small-volume hypertonic resuscita- al: Cerebral blood flow, AVDO2, and neuro- outcome of patients admitted to the inten-

2300 Crit Care Med 2006 Vol. 34, No. 9

 sive care unit with sepsis. Crit Care Med dominal trauma. Ann Surg 1992; 215: for the Surgery of Trauma (EAST), Practice
2003; 31:2742–2751 503–511 Parameter Workgroup for Screening of
101. Ventilation with lower tidal volumes as 105. Drakulovic MB, Torres A, Bauer TT, et al: Blunt Cardiac Injury. Available at: http://
compared with traditional tidal volumes for Supine body position as a risk factor for www.east.org. Accessed March 7, 2006
acute lung injury and the acute respiratory nosocomial pneumonia in mechanically 110. Wolf SE, Thomas SJ, Dasu MR, et al: Im-
distress syndrome. The Acute Respiratory ventilated patients: A randomized trial. proved net protein balance, lean mass and
Distress Syndrome Network. N Engl J Med Lancet 1999; 354:1851–1858 gene expression changes with oxandralone
2000; 342:1301–1308 106. Friedrich JO, Adhikari N, Herridge MS, et treatment in the severely burned. Ann
al: Meta-analysis: Low-dose dopamine in- Surg 2003; 237:801– 811
102. Annane D, Sebille V, Charpentier C, et al:
creases urine output but does not prevent 111. Demling RH: Comparison of the anabolic
Effect of treatment with low doses of hydro-
renal dysfunction or death. Ann Intern Med effects and complications of human growth
cortisone and fludrocortisone on mortality in
2005; 142:510 –524 hormone and the testosterone analog, ox-
patients with septic shock. JAMA 2002; 288:
107. Swartz RD, Bustami RT, Daley JM, et al: Es- androlone, after severe burn injury. Burns
862– 871 timating the impact of renal replacement 1999; 25:215–221
103. Topeli A, Laghi F, Tobin MJ: Effect of closed therapy choice on outcome in severe acute 112. Herndon DN, Hart DW, Steven EW, et al:
unit policy and appointing an intensivist in renal failure. Clin Nephrol 2005; 63:335–345 Reversal of catabolism by beta-blockade af-
a developing country. Crit Care Med 2005; 108. Schultz JM, Turnkey DD: Blunt cardiac in- ter severe burns. N Engl J Med 2001; 345:
33:299 –296 jury. Crit Care Clin 2004; 20:57–70 1223–1229
104. Kudsk KA, Croce MA, Fabian TC, et al: En- 109. Pasquale MD, Nagy K, Clarke J: Practice 113. Deitch E, Livingston DH, Hauser CJ: Septic
teral versus parenteral feeding: Effects on sep- management guidelines for screening of complications in the trauma patient. Crit
tic morbidity after blunt and penetrating ab- blunt cardiac injury: Eastern Association Care Med 1999; 7:158 –172

Crit Care Med 2006 Vol. 34, No. 9 2301