Pre-hospital Hypotension!@ traumatic hemmhor management.....

[Ace844]

(Inflammation and the Host Response to Injury @ a Large-Scale Collaborative Project: Patient-Oriented Research Core—Standard Operating Procedures for Clinical Care: III. Guidelines for Shock Resuscitation

[surgical Glue Grant)

Moore, Frederick A. MD; McKinley, Bruce A. PhD; Moore, Ernest E. MD; Nathens, Avery B. MD, PhD, MPH; West, Michael MD, PhD; Shapiro, Michael B. MD; Bankey, Paul MD, PhD; Freeman, Bradley MD; Harbrecht, Brian G. MD; Johnson, Jeffrey L. MD; Minei, Joseph P. MD; Maier, Ronald V. MD

From the Department of Surgery (F.M., B.M.), University of Texas at Houston; Department of Surgery (E.M., J.L.), University of Colorado; Department of Surgery (A.N., R.M.), University of Washington; Department of Surgery (M.W., M.S.), Northwestern University; Department of Surgery (P.B.), University of Rochester; Department of Surgery (B.F.), Washington University; Department of Surgery (B.H.), University of Pittsburgh; Department of Surgery (J.M.), University of Texas at Southwestern; and the Inflammation and the Host Response to Trauma Large Scale Collaborative Research Program*

*Additional participating investigators in the Large Scale Collaborative Research Agreement entitled, “Inflammation and the Host Response to Trauma” include Henry V. Baker, PhD., Timothy R. Billiar, MD, Bernard H. Brownstein, PhD, Steven E. Calvano, PhD, Irshad H. Chaudry, PhD, J. Perren Cobb, MD, Chuck Cooper, MS, Ronald W. Davis, PhD, Adrian Fay, PhD, Robert J. Feezor MD, Richard L. Gamelli, MD, Nicole S. Gibran, MD, Doug Hayden, MS, David N. Herndon, MD, Jureta W. Horton, PhD, John Lee Hunt, MD, Matthew Klein MD, Krzysztof Laudanski MD, MA, James A. Lederer, PhD, Tanya Logvinenko, PhD, John A. Mannick, MD, Carol L. Miller-Graziano, PhD, Michael Mindrinos, PhD, Lyle L. Moldawer, PhD, Grant E. O'Keefe, MD, MPH, Laurence G. Rahme, PhD, Daniel G. Remick, Jr. MD, David Schoenfeld, PhD, Robert L. Sheridan, MD, Geoffrey M. Silver, MD, Richard D. Smith, PhD, Scott Somers, PhD, Ronald G. Tompkins, MD, ScD. Mehmet Toner, PhD, H. Shaw Warren, MD, Steven E. Wolf, MD, Wenzhong Xiao, PhD, Martin Yarmush, MD, PhD, Vernon R. Young, PhD, ScD.

Submitted for publication January 20, 2006.

Accepted for publication April 19, 2006.

Supported by a Large-Scale Collaborative Project Award (U54-GM62119) from The National Institute of General Medical Sciences, National Institutes of Health

Address for reprints: Frederick A. Moore, MD, UTHSC – Houston Medical School, Department of Surgery, 6431 Fannin Street, MSB 4.264, Houston, TX 77030; e-mail: Frederick.A.Moore@uth.tmc.edu.]

Shock frequently accompanies severe trauma. In addition to acute mortality, shock is a predominant risk factor for multiple organ dysfunction syndrome (MODS) and shock resuscitation is an obligatory early intervention. Because both under and over resuscitation contribute to the pathogenesis of MODS, the potential for MODS development can be minimized by developing a guideline to insure early consistent and appropriate resuscitative efforts. The challenges of this guideline include: a) early identification of high risk patients, implementation in environments that are suboptimal for monitoring resuscitation, c) early identification of resuscitation “non-responders” that require more aggressive interventions, and d) avoiding potentially harmful over zealous interventions.

This guideline is based on the best available evidence and expert consensus discussions supported by the Inflammation and Host Response to Injury Large Scale Collaborative Project award from the National Institute of General Medical Sciences, and is being used in the funded clinical studies.1,2 The following section provides a brief overview of the rationale for specific guideline recommendations. This is followed by two algorithms that depict escalation in interventions and monitoring requirements in the subset of patients who do not respond to ongoing volume loading and/or blood transfusions. With multi institutional experience and critical analysis this resuscitation process may be further refined; at this time it is intended to serve as a template for interventional trials and to test the utility of new monitoring technology. This protocol was designed for blunt trauma patients who are presumed not to have a serious concomitant brain injury. Its purpose is to guide resuscitation as soon as feasible after arrival in the Emergency Department (ED) after control of active torso bleeding.

Protocol Rationale

Early Recognition of Shock in the Emergency Department

Recognizing the presence of shock and assessing its severity are key factors in early identification of high risk patients. Shock often can be detected by simple physical examination findings in the ED resuscitation area. Diminished or absent peripheral (radial, pedal) or central (carotid, femoral) pulses, decreased capillary refill associated with pallor or cool clammy extremities may all denote the presence of shock and hypovolemia. The initial blood pressure (BP) measurement should be performed using a manual cuff because automatic cuff BP measurement devices may overestimate systolic BP (SBP) in hypovolemic trauma patients.3 A SBP < 90 mm Hg and/or a heart rate (HR) > 130 bpm is generally considered to be indicative of shock. Some patients (especially the young) compensate for hypovolemia and maintain a normal SBP even in the face of significant ongoing hemorrhage although this is often associated with tachycardia. Additionally, because acute massive blood loss may paradoxically trigger a vagal-mediated bradycardia, the traditional inverse correlation between increased HR and decreased effective blood volume may not hold in the early resuscitation period.4 The initial hemoglobin concentration ([Hb]) is notoriously misleading because there has not been sufficient time for influx of interstitial fluid into the intravascular space and the patient has not yet been volume resuscitated. Therefore, it is important to measure the [Hb] again after the initial 2 L of crystalloid loading, a decrease greater than 2 g/dL is grounds for concern. The magnitude of arterial base deficit (BD) has been shown to be a useful index of the severity of hemorrhagic shock. A BD >= 6 mEq/L is indicative of severe shock.5 Serial BD determinations are important in determining the effectiveness of interventions and lack of response is indicative of a poor prognosis. Other less well studied markers of the severity of shock include venous blood lactate, bicarbonate concentrations and end-tidal CO2 to Paco2 differences.

Volume Loading With Isotonic Crystalloid Fluid

The key step in resuscitation of the injured patient is the control of active hemorrhage. The actively bleeding patient cannot be adequately resuscitated without hemorrhage control. Resuscitation with isotonic crystalloid fluids has been the standard of care in the United States since the late l960s. The laboratory work of Shires and Moyer demonstrated the best survival was achieved with large volume isotonic crystalloid solution. The basic concept is that interstitial fluid moves into both the intravascular and intracellular spaces in response to shock and that adequate resuscitation requires replenishment of both the intravascular and interstitial spaces. Their studies demonstrated that the optimal ratio of isotonic crystalloid infusion to shed blood infusion was 3 to 1. Subsequent studies demonstrated that the optimal ratio for survival after severe shock increases and can be as high as 8 to 1.6,7 Clinical trials were performed in the l970s and 1980s that compared isotonic resuscitation and colloid resuscitation. Individually, these trials were underpowered and reported conflicting results. When subjected to meta-analysis, they have yielded no consistent differences in overall outcome. When the same data were subjected to subgroup analysis, however, the use of isotonic crystalloids in trauma patients was associated with improved survival. A large clinical trial published in 2004 found no differences in outcome between crystalloid and colloid resuscitation in ICU patients, but again, subgroup analysis demonstrated improved outcomes in trauma patients receiving crystalloid.8 Although these subgroup analyses are not definitive, they are consistent with the early laboratory studies, which indicated that survival in hemorrhagic shock is improved with large volume crystalloid resuscitation. However, in recent years, “damage control” surgery combined with prompt ICU resuscitation appears to be salvaging more patients who are arriving with exsanguinating hemorrhage. Unfortunately, over zealous crystalloid infusion appears to have adverse consequences, e.g. cerebral edema (increased ICP), acute lung injury (worsened pulmonary edema), and the abdominal compartment syndrome (primary and secondary).9

Lactated Ringer's is the Preferred Isotonic Crystalloid

Although newer formulations (e.g. Ringer's ethyl pyruvate) are being tested clinically, normal saline (NS) and lactated Ringer's (LR) remain the most commonly used isotonic fluids. In theory, LR is preferable to NS because it provides a better buffer for metabolic acidosis, but to date, investigators have not documented any important differences in outcome. Moreover, the D isomer of lactate may have adverse immunoinflammatory properties. One laboratory study found that NS and LR were equivalent in the setting of moderate hemorrhagic shock but that in the setting of massive hemorrhage, NS was associated with greater physiologic derangement (e.g. hyperchloremic acidosis) and a higher mortality.10 Clinical experience confirms the adverse effects of iatrogenic hyperchloremic acidosis. In addition, the potential benefits of using hypertonic saline (HTS) – rapid blood pressure response, decrease in ICP and improved immunologic status – for resuscitation are unproven but currently in clinical trials.

Blood Transfusion to Maintain Hemoglobin Concentration at 10 g/dL

The optimal [Hb] continues to be a subject of intense debate. Early laboratory studies of shock resuscitation suggested that survival was improved when [Hb] was maintained in the range of 12 to 13 g/dL. Subsequent studies using isovolemic hemodilution models indicated that the optimal [Hb] for maintaining oxygen delivery was 10 g/dL, and, until relatively recently, this value was the recommended level for critically ill patients. Currently, there is a growing recognition that administration of stored packed red blood cells (PRBC) can adversely affect outcome by modulating the inflammatory response (by both amplifying early proinflammation and aggravating late immunosuppression) and by impairing tissue perfusion (limiting access to or obstructing the microcirculation as a consequence of decreased RBC deformability). A 1999 randomized trial found that patients who received transfusions according to a restrictive policy (i.e. transfusion when the [Hb] fell below 7 g/dL) did as well as, and possibly better than, patients who received transfusions on a more liberal basis (i.e. transfusion when [Hb] fell below 10 g/dL).11 However, this study was done in a select group of euvolemic patients in which those with active hemorrhage were excluded; thus, it is not applicable to severely injured trauma patients requiring active shock resuscitation, but is cogent as their clinical course progresses. (See SOP for blood transfusions.) In addition, if blood transfusions are to be restricted during active resuscitation, it is not clear which alternative fluids should be used. Colloid solutions have been associated with complications and indiscriminant use of crystalloid fluid is detrimental.9 Hypertonic saline and hemoglobin based oxygen carriers are attractive alternatives, but additional clinical trials are needed before these could become standard of care. Additionally, maintaining higher [Hb] during active bleeding may facilitate coagulation. Many patients who are resuscitated by this process will also require a massive transfusion (i.e. >10 units PRBC in 24 hour), and are at risk for developing a coagulopathy. This subset of patients may benefit from early fresh frozen plasma (FFP) administration and this should be factored into their volume loading regimen.

Early Central Venous Pressure Monitoring

The most likely etiology of shock following major trauma is hypovolemia secondary to acute blood loss. Therefore, initial volume loading with isotonic crystalloids (1 L boluses in adults and 20 cc/kg in children) is recommended. The response to this empiric volume loading assists in early triage decisions. Prompt correction of abnormal vital signs indicates that a lesser volume deficit (10–20% blood volume) was present and that an expedited trauma evaluation can be safely performed to rule out occult bleeding. Patients who do not respond to empiric volume loading may have severe hypovolemia (30–40% blood volume), cardiogenic shock or neurogenic shock. Given that neurogenic shock is usually well tolerated and typically responds to initial volume loading, the key issue is to quickly distinguish hypovolemic shock from cardiogenic shock. Placement of a catheter that permits reliable measurement of central venous pressure (CVP) following the initial boluses can help differentiate these states. A high CVP (>15 mm Hg) suggests cardiogenic shock (likely etiologies include tension pneumothorax, pericardial tamponade or myocardial contusion/infarction). A low CVP (<5 mm Hg) may indicate acute ongoing blood loss, and mandates focus on identifying occult sources of blood loss. In many instances control of hemorrhage requires operating room (OR) or interventional radiologic (IR) interventions. Patients who initially respond to volume loading, but require ongoing crystalloid volume and/or blood transfusion during expedited trauma evaluation, or who have evidence of severe shock by ABG (i.e. BD >= 6 mEq/L), should have a central venous line placed and have continuous CVP measurements displayed to assist with ongoing resuscitation until the patients arrives in the intensive care unit (ICU). In the ICU, with more intensive monitoring available, the decision needs to be made if escalation to pulmonary artery (PA) catheterization is warranted.

Pulmonary Artery Catheterization in the ICU

The primary goal of shock resuscitation is the early establishment of “adequate” oxygen delivery (DO2) to vital organs. The yet to be resolved controversy is what is “adequate.” The calculated variable DO2 is the product of cardiac output (CO) and arterial oxygen content (Cao2). By convention, CO is indexed to body surface area and expressed as cardiac index (CI), and when multiplied by Cao2 yields an oxygen delivery index (DO2I). Normal DO2I is roughly 450 mL/min/m2. Cao2 and DO2I are calculated as follows:

Cao2 (mL O2/dL) = [Hb] (g/dL) × 1.38 mL O2/g Hb × Sao2 (%) + [Pao2 (mmHg) ×: 0.003 mL O2/mmHg]

DO2I (mL/min/m2) = CI (L/min/m2) ×: Cao2 (mL/dL) × 10 dL/L,

where [Hb] is hemoglobin concentration, Sao2 is hemoglobin O2 saturation, Pao2 is arterial oxygen tension, and 0.003 is solubility of O2 in blood. Thus, there are four variables (i.e. Pao2, Sao2, [Hb], and CI) that determine DO2I. Of the four variables CI is the most difficult to monitor and manipulate. This is the rationale for liberal use of the PA catheter in severely injured patients. Myocardial dysfunction during traumatic shock resuscitation is common, but usually responds to volume loading. However, once a patient has been volume loaded (CVP >15 mm Hg) and has evidence of ongoing shock (e.g. decreased MAP, increased BD or lactate levels), a PA catheter is warranted to better monitor cardiovascular function, especially filling pressures and CI. Once the PA catheter is placed, the key question is what CI is acceptable. Early work from Shoemaker et al. demonstrated that the “survivor” response to traumatic stress is to become hyperdynamic (CI > 4.5 L/min/m2) and consequently have supranormal DO2I (>600 mL/min/m2).12 Supranormal DO2I was, therefore, proposed to be the resuscitation goal. Subsequent prospective randomized controlled trials (RCTs), however, failed to demonstrate improved outcome with goal orientated resuscitation to supranormal DO2I. In fact, several recent studies indicate that this strategy is harmful.13 Recent studies in which a normal DO2I goal of 500 mL/min/m2 was used demonstrated that patients achieve similar hyperdynamic responses to standardized interventions, require less volume loading, and have better outcomes than patients resuscitated to a supranormal DO2I goal.14,15 We recommend using a CI >= 3.8 L/min/m2 as the resuscitation goal for this resuscitation process guideline. During active resuscitation, most severely injured patients will have Sao2 > 92% and [Hb] > 10 g/dL and, therefore, DO2I will approach 500 mL/min/m2. PA catheters capable of continuous monitoring of CO and mixed venous hemoglobin oxygen saturation (SmvO2) are now commonly available and should be utilized. These continuously monitored variables provide rapid feedback that is often necessary to guide effective, timely resuscitation interventions. This approach is based on expert opinion as there are no data to suggest that a PA catheter is either harmful or beneficial in severely injured patients undergoing shock resuscitation.

Assessment of Pre-Load by PCWP and the use of the “Starling Curve” Intervention

The traditional variable used to assess volume status is the pulmonary capillary wedge pressure (PCWP). It is important to recognize that significant hypovolemia can exist despite reasonable PCWP. Peripheral vasoconstriction of less acutely essential organs (e.g. kidney, gut, muscle, and skin) results in blood volume shifts to maintain the central circulation and perfusion of more essential organs (e.g. heart and brain). With a low CI and a low PCWP (i.e. <10 mm Hg), volume loading should be undertaken. After the PCWP increases to >=15 mm Hg, the benefits of increasing CI by the Frank Starling mechanism should be considered. If additional CI is needed then a stepwise incremental volume loading intervention is used to identify the optimal CI-PCWP “operating point.” Because the shape and position of the Frank Starling curve is dependent on left ventricular contractility, compliance and afterload, it is difficult to identify optimal or plateau PCWP during volume loading without frequent sequential measurements.16 In a recent study, one-third of high risk patients had persistently low CO, high systemic vascular resistance (SVR) and high BD despite volume loading to a PCWP >= 15 mm Hg. However, these patients did respond well to the “Starling curve” intervention, by increasing CO, decreasing SVR and decreasing BD. This observation is consistent with a recent RCT in which patients who did not respond to initial volume loading (to presumed euvolemia) were randomized to additional volume loading or to a vasodilating inotropic agent.17 Patients who received further volume loading were found to have a better resuscitation response than those treated with inotropic agents. This favorable response to additional pre-load must, however, be weighed against the risks of increasing hydrostatic pressure leading to increase pulmonary edema, especially in patients with leaky endothelium (e.g. pulmonary contusion). Increasing PCWP >25 mm Hg in attempts to create a “Starling curve” should not be done because of the potential of causing or worsening pulmonary edema.16

Use of Vasoactive Agents in Non-Responders

The primary early problem in shock resuscitation is decreased preload with resulting decreased CO. In this setting, the normal “survivor” response is to become hyperdynamic with volume loading. However, as time and severity of shock increases, a complex pathophysiologic interaction evolves that limits CO. Important factors include relative hypovolemia, primary or secondary myocardial dysfunction and excessive peripheral vasoconstriction. With ineffective intervention, shock will ultimately progress into pathologic vasodilation which heralds irreversible shock. Unfortunately, in the later phases of shock, patients who do not respond to volume loading also do not consistently respond to vasoactive agents. Therefore, a PA catheter should be placed to more accurately characterize their hemodynamic profile and monitor their response to vasoactive agent administration. A specific agent with known physiologic actions should then be administered and titrated to a desired effect. (See Table 1.)

--------------------------------------------------------------------------------

[Help with image viewing]

[Email Jumpstart To Image] Table 1 Vasoactive Agents Commonly Used in Shock Resuscitation

--------------------------------------------------------------------------------

For the typical non responding patient (i.e. low CI, adequate PCWP and high SVR), a vasodilating inotrope such as dobutamine (beware of hypotension) or dopamine (beware of tachycardia) is recommended. Patients with a particularly high SVR may respond better to simple afterload reduction with nitroprusside. However, additional volume loading may be needed to maintain an adequate PCWP and oxygenation may worsen due to loss of hypoxic pulmonary vasoconstriction (i.e. worsened V/Q mismatch). For patients with low CO and normal SVR, dopamine or lower dose norepinephrine are reasonable choices. For patients who have low SVR and are thus unable to maintain an adequate MAP (>= 60 mm Hg), higher doses of norepinephrine should be used. It is also important to rule out relative adrenal insufficiency in patients requiring higher doses of norepinephrine.18 Additionally, low dose vasopressin (as replacement therapy) may reduce the need for norepinephrine in patients exhibiting impending irreversible shock.19 Of note, there are no data that demonstrate that use of a specific vasoactive agent in non-responding patients improves outcome. Several studies have demonstrated improved hemodynamic responses to specific agents, and extrapolate this response to an improved outcome.20–22

Protocol Summary

Initial Resuscitation

Figure 1 depicts initial ED resuscitation and the following text includes explanatory annotations lettered A through F. Variables that drive decision making include SBP, HR, BD, [Hb], CVP and clinical judgment (ever present). Additionally, stopping ongoing active hemorrhage is paramount to the survival of these patients.

--------------------------------------------------------------------------------

[Help with image viewing]

[Email Jumpstart To Image] Fig. 1. Initial resuscitation.

--------------------------------------------------------------------------------

A. Major trauma patients arriving in shock (SPB <90 mm Hg and/or HR >130 bpm) are initially managed by using Advanced Trauma Life Support (ATLS).22 Routine monitoring includes frequent vital signs (minimum q 15 minutes), continuous ECG and pulse oximetry (SpO2) and core body temperature. A data flow sheet is necessary to trend physiologic indices, laboratory test results and fluid volume/blood transfusion administration. During the initial ED evaluation an ABG analysis should be obtained in all patients presenting in traumatic shock.

B. Major torso trauma patients who have evidence of shock (documented by early SBP <90 mm Hg and/or a BD >=6 mEq/L), and who require ongoing resuscitation, should have a central venous line (via subclavian or internal jugular vein) placed in ED. CVP measurements should be used to differentiate the type of shock and to assist with subsequent monitoring of shock resuscitation.

C. Early CVP >15 cm mmHg (before extensive volume loading) suggests cardiogenic shock.

D. Differential diagnosis of cardiogenic shock following blunt trauma includes: 1) tension pneumothorax, 2) myocardial contusion/infarction, 3) pericardial tamponade (uncommon) and 4) air embolus (rare). Specific diagnosis and treatment is beyond the scope of this protocol. ATLS guidelines should be followed.23

E. CVP <10 mm Hg despite volume loading indicates persistent hypovolemia and this most likely reflects ongoing bleeding. The endpoint of initial resuscitation is controversial and the algorithm statement “resuscitate until stable” is intentionally vague (i.e. requires clinical judgment). The crux of the issue is whether it is preferable to administer fluids to restore DO2 to the vital organs (risking hemodilution and disruption of early hemostatic clots) or to withhold fluid resuscitation until control of hemorrhage (risking prolonged cellular shock to the extent that it becomes irreversible by the time hemorrhage control is accomplished). At present, the rationale compromise is hypotensive resuscitation (SBP >90 mm Hg and HR <130 bpm) with moderate volume loading until hemorrhage control is accomplished. This approach is becoming the standard of care for penetrating trauma victims. It is most likely safe for blunt torso trauma patients who do not have significant concomitant brain injuries that could be worsened by permissive hypotension.

F. LR boluses should continue and, when LR infusion exceeds 30 mL/kg, blood should be administered. Earlier empiric blood transfusion is indicated in patients (especially the elderly) who arrive in severe shock or who have injuries associated with significant bleeding (e.g. vertical shear pelvic fracture or bilateral femur fracture.) Protocols for massive transfusion should be established with the blood bank to ensure prompt availability of blood products for patients arriving with ongoing life-threatening hemorrhage. Among the most devastating complications of massive blood and fluid administration is a coagulopathy. Stored blood is deficient in factors V and VIII and platelets. Timely administration of FFP and platelets will minimize risk of coagulopathy after massive transfusion. Presumptive factor replacement is usually not indicated in the early phase of resuscitation, but may be appropriate in patients with massive hemorrhage caused by significant intracavitary bleeding or an unstable pelvic fracture.24,25

ICU Resuscitation

Hemorrhage control is of paramount importance in the initial management of major torso trauma patients arriving in shock. It is assumed that this issue will have been addressed in the vast majority of patients by the time the patient is admitted to the ICU. The priorities in early ICU care are to: a) optimize resuscitation, correct hypothermia, coagulopathy and acidosis and c) monitor for ongoing bleeding requiring OR or IR intervention. Figure 2 depicts ICU resuscitation and the following text includes explanatory annotations lettered A through H. An important early decision is whether to escalate monitoring interventions (i.e. placement of arterial and PA catheters). Variables that drive decisions in the PA catheter algorithm include CI, [Hb], PCWP, BD and clinical judgment.

--------------------------------------------------------------------------------

[Help with image viewing]

[Email Jumpstart To Image] Fig. 2. ICU resuscitation.

--------------------------------------------------------------------------------

A. When the patient arrives in the ICU, the physician needs to decide whether to continue resuscitation using serial vital signs, CVP, [Hb] and BD determinations (i.e. CVP Algorithm.) Most patients can be managed using this process, but close observation by a physician at bedside is required because CVP is a very indirect monitor of hemodynamic function.

B. For patients who are not responding to ongoing volume loading/blood transfusion (i.e. low MAP or persistently high BD) and/or are demonstrating secondary organ dysfunctions (i.e. worsening oxygenation or decreased urine output), pulmonary artery catheterization is warranted. The possibility of an impending abdominal compartment syndrome needs to be considered if crystalloid fluid volume loading exceeds 10 L or PRBC transfusion exceeds 10 units. Periodic urinary bladder pressure measurements should be obtained to monitor for onset of abdominal compartment syndrome. Urinary bladder pressure >=25 mm Hg indicates significant abdominal hypertension and need for bedside assessment for possible surgical intervention.9,14

C. It is assumed that most patients being treated by this algorithm will be intubated. If not intubated and requiring ongoing volume loading, intubation needs to be considered, because worsening pulmonary function is likely. If intubated and PEEP >=12 cm H2O, the effects of high mean airway pressures on cardiac function needs to be considered, as does use of PA catheter.

D. If the patient meets the CI goal of 3.8 L/min/m2, then the patient should be monitored as depicted in this arm of algorithm. Hemodynamic variables should be assessed hourly, and possibly more frequently. Laboratory variables including [Hb] and ABG (BD) should be determined every 4 hours, and possibly more frequently until the patient is fully resuscitated and stable. Coagulation variables and urinary bladder pressure measurements should be monitored as deemed necessary.

E. Most young patients easily exceed the CI goal of 3.8 L/min/m2 with modest crystalloid volume loading and blood transfusion. There is no need to increase PCWP to high levels in responding patients, but [Hb] should be maintained >=10 g/dL during acute resuscitation to assure a safety margin in the event of occult or recurring bleeding. However, once the BD has been normalized and the need for ongoing volume loading has resolved, a lower [Hb] is acceptable. (See SOP for transfusion.)

F. PCWP may not accurately reflect left ventricular end diastolic volume and increasing PCWP to >=15 mm Hg may enhance cardiac performance. The optimal relationship between PCWP and CI can be determined by incrementally increasing left ventricular preload (PCWP) and then measuring cardiac output (CI) in response, i.e. by generating a “Starling curve” to determine the optimal CI-PCWP operating point.10 This should only be done in patients who are not meeting the CI goal and who have evidence of ongoing shock (i.e. persistently elevated BD.)

G. After obtaining optimal PCWP, if CI <3.8 L/min/m2, then infusion of a vasodilating inotropic agent should be started. Dobutamine is recommended as the preferred inotropic agent. Dobutamine infusion should be started at a dose rate of 5.0 µg/kg/min and increased in increments of 2.5 µg/kg/min to a maximum of 20 µg/kg/min to increase CI. If the patient does not tolerate the vasodilation, dopamine should be considered with progression from low to mid- to high dose, while monitoring for excessive tachycardia.

H. Occasionally, an inotropic agent with vasoconstrictive effects may be needed to maintain MAP >=60 mm Hg, enhance myocardial contractility and maintain coronary perfusion pressure. These agents decrease peripheral perfusion at the microcirculatory level by [alpha]1 vasoconstriction of metarterioles thereby prolonging or exacerbating the effects of shock, and are therefore an intervention of last resort. Norepinephrine is recommended as the preferred agent. Norepinephrine infusion should be started at 0.05 µg/kg/min and increased in increments of 0.05 µg/kg/min as needed to obtain CI >=3.8 L/min/m2, and maintain MAP >=60 mm Hg. The maximum recommended norepinephrine dose rate is 0.2 µg/kg/min. Adrenal insufficiency in patients requiring norepinephrine to maintain MAP needs to be ruled out. Low dose vasopressin may decrease the required dose of norepinephrine.

REFERENCES

1. Moore FA, McKinley BA, Moore EE. The next generation in shock resuscitation. Lancet. 2004;363:1989–96. [Context Link]

2. Moore FA, Moore EE. Ch 1 Initial Management of Life-Threatening Trauma. Ch 6 Trauma and Thermal Injury. ACS Surgery: Principles and Practice. Souba WW, Fink MP, Jurkovich GJ, et al, eds. Web MD Inc, New York, 2005. [Context Link]

3. Davis JW, Davis IC, Bennick LD, et al. Are automated blood pressure measurements accurate in trauma patients? J Trauma. 2003;55:860–863. Ovid Full Text Bibliographic Links [Context Link]

4. Victorino GP, Battistella D, Wisner DH. Does tachycardia correlate with hypotension after trauma? J Am Coll Surg. 2003;196:679–684. Bibliographic Links [Context Link]

5. Rutherford EJ, Morris JA, Reed GW, Hall KS. Base deficit stratifies mortality and determine therapy. J Trauma. 1992;33:417–423. Bibliographic Links [Context Link]

6. Cervera AL, Moss G. Progressive hypovolemia leading to shock after continuous hemorrhage and 3:1 crystalloid replacement. Am J Surg. 1975;129:670–674. Bibliographic Links [Context Link]

7. Healey MA, Samphire J, Hoyt DB, et al. Irreversible shock is not irreversible: A new model of massive hemorrhage and Resuscitation. J Trauma. 2001;50:826–834. Ovid Full Text Bibliographic Links [Context Link]

8. The SAFE Study Investigators. A comparison of albumin and saline for fluid resuscitation in the intensive care unit. N Engl J Med. 2004;350:2247–2254. [Context Link]

9. Balogh Z McKinley BA, Holcomb JB, et al. Both primary and secondary abdominal compartment syndrome can be predicted early and are harbingers of multiple organ failure. J Trauma. 2003;54:848–859. Ovid Full Text Bibliographic Links [Context Link]

10. Healey MA, Davis RE, Liu FC, et al. Lactated Ringer's is superior to normal saline in a model of massive hemorrhage and resuscitation. J Trauma. 1998;45:894–899. Ovid Full Text Bibliographic Links [Context Link]

11. Herbert PC, Wells G, Blajchman MA, et al. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. N Engl J Med. 1999;340:409–417. [Context Link]

12. Shoemaker WC, Appel PL, Kram HB, et al. Prospective trial of supranormal values of survivors as therapeutic goals in high-risk surgical patients. Chest. 1988;94:1176–1186. Bibliographic Links [Context Link]

13. Gattinoni L, Brazzi L, Pelosi P, et al. A trial of goal-orientated hemodynamic therapy in critically ill patients. N Engl J Med. 1995;333:1025–1032. [Context Link]

14. Balogh Z, McKinley BA, Cocanour CS, et al. Supranormal trauma resuscitation causes more cases of abdominal compartment syndrome. Arch Surg. 2003;138:637–642. Ovid Full Text Bibliographic Links [Context Link]

15. Velmahos GC, Demetriades D, Shoemaker WC, et al. Endpoints of resuscitation of critically injured patients: normal or supranormal? A prospective randomized trial. Ann Surg. 2000;232:409–418. Ovid Full Text Bibliographic Links [Context Link]

16. Marr AB, Moore FA, Sailors RM, et al. Preload optimization using “Starling Curve” generation during shock resuscitation: can it be done? Shock. 2004;21:300–305. Ovid Full Text Bibliographic Links [Context Link]

17. Miller PR, Meredith JW, Chang MC. Randomized, prospective comparison of increased preload versus inotropes in the resuscitation of trauma patients: effects on cardiopulmonary function and visceral perfusion. J Trauma. 1998;44:107–113. Ovid Full Text Bibliographic Links [Context Link]

18. Cooper MS, Stewart PM. Corticosteroid nsufficiency in acutely ill patients. N Engl J Med. 2003;348:727–734. [Context Link]

19. Landry DW, Oliver JA. Mechanisms of disease: the pathogenesis of Vasodilatory Shock. N Engl J Med. 2001;345:588–595. Ovid Full Text Bibliographic Links [Context Link]

20. Abou-Khalil B, Scalea TM, Trooskin SZ, et al. Hemodynamic responses to shock in young trauma patients: need for invasive monitoring. Crit Care Med. 1994;22:633–639. Bibliographic Links [Context Link]

21. Chang MC, Martin RS, Scherer LA, et al. Improving ventricular-arterial coupling during resuscitation from shock: effects on cardiovascular function and systemic perfusion. J Trauma. 2002;53:679–685. Ovid Full Text Bibliographic Links [Context Link]

22. McKinley BA, Marvin RG, Cocanour CS, et al. Nitroprusside in resuscitation of major torso trauma. J Trauma. 2000;49:1089–1095. Ovid Full Text Bibliographic Links [Context Link]

23. Shock. Advanced Trauma Life Support Manual. Chicago:American College of Surgeons; 2004. [Context Link]

24. Hirshberg A, Dugas M, Banez EI, et al. Minimizing dilutional coagulopathy in exsanguinating hemorrhage: A computer simulation. J Trauma. 2003;54:454–463. Ovid Full Text Bibliographic Links [Context Link]

25. Biffl WL, Smith WR, Moore EE, et al. Evolution of a multidisciplinary clinical pathway for the management of unstable patients with pelvic fractures. Ann Surg. 2001;233:843–850. Ovid Full Text Bibliographic Links [Context Link]

[Ace844]

(Early Massive Trauma Transfusion: State of the Art: EDITORS’ INTRODUCTION

[Article)

Holcomb, John B. MD, FACS; Hess, John R. MD, MPH, FACP, FAAAS

From United States Army Institute of Surgical Research, Fort Sam Houston, (J.B.H.), San Antonio, Texas; Department of Pathology and Medicine, University of Maryland School of Medicine (J.H.), Baltimore, Maryland.

Submitted for publication November 18, 2005.

Accepted for publication November 28, 2005.

Address for reprints: John Holcomb, MD, 3400 Rawley E. Chambers Avenue, Bldg. 3611, Fort Sam Houston, Texas 78234; email: John.Holcomb@amedd.army.mil.]

A report on a symposium held at the U.S. Army Institute of Surgical Research, 26–27 May 2005.

Injury is the most important cause of years of productive life lost world-wide. Uncontrolled hemorrhage causes much of that morbidity and mortality and is the critical factor most susceptible to treatment in injured patients. In highly developed trauma systems, supported by the best in blood-banking, early massive transfusion buys time for surgeons and interventional radiologists to get bleeding stopped. In recent years, however, the epidemiologic association of massive transfusion with poorer outcomes has raised important questions about the meaning of this statistical association and its possible physiologic basis. Given the importance of early interventions in the care of the critically injured, understanding the physiology of and true indications for early massive transfusion in trauma care has the potential to save many lives. The U.S. Army Institute of Surgical Research (USAISR) symposium on early massive trauma transfusion focused on what is known and what is unknown about all components of massive transfusion, to aid in the design of standardized research protocols and clinical practice guidelines in hemostasis and transfusion. Forty-six surgeons, anesthesiologists, hematologists, transfusion medicine specialists, epidemiologists, basic scientists, regulatory experts and administrators from Europe and North America were identified by their previous contributions to the literature or their ability to comment critically based on their current position and knowledge. They were divided into teams of two to four, assigned one of twelve topics to research, and asked to pay special attention to the quality of the supporting evidence. Each team produced a manuscript that was circulated to discussants in advance. The group was then brought to San Antonio, TX, site of the USAISR, and met to present and discuss the papers in plenary sessions. Discussion was lively but resulted in broad consensus.

The first presentation, by Kauvar, Levering, and Wade described the epidemiology of injury and the general clinical issues associated with presentation and treatment of massive hemorrhage in trauma. The paper and the resulting discussion set out the key issues that would guide the group over the next day and a half: the international burden of trauma mortality and morbidity; the critical importance of coordinated, early, intervention; the differing concerns but tight links between initial trauma presentation and care and, in those who survive the immediate effects of their injuries, subsequent complications and outcome; the risks and benefits of various approaches to blood and blood product use.

This was followed by Hess and Lawson discussing the coagulopathy of trauma: the roles of blood loss, dilution, hypothermia, acidosis, component consumption, and fibrinolysis in coagulopathic bleeding and the pathologic dissemination of coagulation. The coagulopathy of trauma was compared with disseminated intravascular coagulation (DIC) from brain, fat, amniotic fluid, and diffuse tissue injury in trauma patients and DIC in non-trauma patients.

In the next session, Eastridge, Malone, and Holcomb examined the literature for published data-based predictors of the need for massive transfusion and of mortality in trauma patients. In large, retrospective studies, pre-hospital and presentation physiologic markers, measures of oxygen debt, coagulopathy, and hypothermia predict the need for transfusion. The association between transfusion in the first 24 hours and subsequent multiple organ failure (MOF) is clear in these data. This was followed by Napolitano’s review of published literature on cumulative risks of early red blood cell (RBC) transfusion as demonstrated in retrospective surgical cohort and basic immunologic studies. Blajchman, in turn, presented data from the trauma subset of the prospective Trauma Requirements in Critical Care (TRICC) study, which show no additional risk from single RBC transfusions to hemodynamically stable trauma patients in the critical care phase.

Carson and Dutton then presented data on the indications for early RBC transfusion, emphasizing the mortality seen with increasing anemia in surgery on Jehovah’s Witnesses, and the contribution of anemia to acidosis.

The next two sessions addressed the use of non-RBC blood products. First, MacLennan and Norda reviewed the risks of plasma and platelet transfusions and the U.K. and European organizational responses to these risks. Ketchum, Hess, and Hiippala examined the benefits of and rationale for earlier and more aggressive use of these products than is now generally practiced. Cumulative risk of further injury, inflammatory and infectious, from plasma and platelets as measured in the European hemovigilance studies appears to be at least 2 orders of magnitude below the risk of developing coagulopathy in massively injured and massively transfused individuals.

Military, civil, regulatory, and technical issues that impact the safety of massive transfusion were addressed in the next four sessions. Repine, Perkins, Blackbourne, and Kauvar described the use of fresh whole blood as a field expedient to treat trauma coagulopathy in casualties in Iraq. Lefering, Dutton, and Lynn presented data on risk factors for mortality observed in large civilian trauma system databases. Holness and Vostal from the U.S. Food and Drug Administration (FDA) discussed regulatory aspects of blood safety and blood product licensure, using transfusion-related lung injury and pathogen-reduced platelets as recent examples. Blajchman reviewed the data on the contribution of universal leukoreduction of RBC and platelets to transfusion safety.

Finally, Malone, Hess, and Fingerhut reviewed massive transfusion protocols from well-developed trauma systems in Denver, Houston, Helsinki, Sydney, and Baltimore. Despite superficial differences, all deliver similar amounts of the various blood components in similar circumstances and with similar triggers. This group then presented a massive transfusion protocol based on the best data from their review.

Over the day and a half of meetings, the most contentious issue was the significance of the identification of blood transfusion as an independent risk factor for multiple organ failure and death in multivariate analysis of large retrospective series of trauma patients. Some felt strongly that the association suggested causation. Others noted that the most severely injured both required the most blood and had the highest risk of bad outcomes. The statistical independence of transfusion and injury severity scores in predicting bad outcome may mean that blood use is more linearly related to injury severity than the quadratically modified injury severity scores. In addition, in the literature reviewed, blood transfusion was not separable from blood loss. All agreed that better evidence is needed to explore these issues.

Despite this controversy, general consensus was reached that, in the most severely injured patients, early use of RBC, plasma, and platelets still offers the best chance of limiting the coagulopathy of trauma in early phases of care. Further, the practical problems of initiating venous access, delivering initially uncross-matched RBC, and obtaining further RBC, thawed plasma, and platelets from a transfusion service mean that most of these patients will be receiving these components on an approximately 1:1:1 ratio in an effort to make up early deficits that occur when only crystalloid and RBC are available. These facts allow standardized guidelines for massive transfusion that use the 1:1:1 ratio and accept the number of units of transfused RBC as a surrogate for the amount of blood lost to become guidelines for clinical practice and clinical research.

The conveners wish to thank all of the participants for their hard work, thoughtful comments and attention to detail. We also thank the sponsors, NovoNordisk and the U.S. Army for the support that allowed the assembly of this symposium and the production of this supplement to The Journal of Trauma. We hope that the results of this effort are better understanding and more generally applicable protocols for research and improved care of the injured.

The opinions or assertions contained herein are the private views of the authors and are not to be construed as official or reflecting the views of the Department of Defense or the United States Government. One of the authors is an employee of the U.S. Government. This work was prepared as part of his official duties and, as such, there is not copyright to be transferred.

(Impact of Hemorrhage on Trauma Outcome: An Overview of Epidemiology @ Clinical Presentations, and Therapeutic Considerations

[Article)

Kauvar, David S. MD; Lefering, Rolf PhD; Wade, Charles E. PhD

From the United States Army Institute of Surgical Research (D.S.K., C.E.W.), Fort Sam Houston, San Antonio, Texas; and the University of Cologne (R.L.), Cologne, Germany.

Submitted for publication November 18, 2005.

Accepted for publication November 28, 2005.

The opinions or assertions contained herein are the private views of the authors and are not to be construed as official or reflecting the views of the Department of Defense or United States Government. The authors are employees of the U.S. government. This work was prepared as part of their official duties and, as such, there is no copyright to be transferred.

Address for reprints: Charles E. Wade, PhD, USAISR, 3400 Rawley E. Chambers Avenue, Fort Sam Houston, TX 78234; email: charles.wade@cen.amedd.army.mil.]

Abstract

The world-wide impact of traumatic injury and associated hemorrhage on human health and well-being cannot be overstated. Twelve percent of the global disease burden is the result of violence or accidental injury. Hemorrhage is responsible for 30 to 40% of trauma mortality, and of these deaths, 33 to 56% occur during the prehospital period. Among those who reach care, early mortality is caused by continued hemorrhage, coagulopathy, and incomplete resuscitation. The techniques of early care, including blood transfusion, may underlie late mortality and long-term morbidity. While the volume of blood lost cannot be measured, physiologic and chemical measures and the number of units of blood given are readily recorded and analyzed. Improvements in early hemorrhage control and resuscitation and the prevention and aggressive treatment of coagulopathy appear to have the greatest potential to improve outcomes in severely injured trauma patients.

--------------------------------------------------------------------------------

Trauma accounts for a significant proportion of annual mortality world-wide. The World Health Organization (WHO) estimates that in the year 2000, 5 million people died of injuries, accounting for 9% of global annual mortality. That same year, 12% of the global burden of disease resulted from injury. Over 90% of the world’s trauma mortality occurs in low- and middle-income nations, with those in Eastern Europe having the highest rates. Almost 50% of those who die are between 15 and 44 years of age, with males accounting for twice as many deaths as females; death due to traumatic injury is, therefore, the leading cause of life years lost.1 Violence—self-inflicted, interpersonal, and war-related—accounts for half of trauma mortality, with 1.6 million deaths in the year 2000. Road traffic accounts for the next largest proportion, roughly 1.2 million deaths, per year, 2.1% of overall mortality. An additional 20 to 50 million people are injured annually in road traffic incidents.2,3

Though most of the world’s trauma mortality occurs in developing countries, trauma is a significant cause of morbidity and mortality in industrialized nations as well. In the United States in 2003, over 29 million people, more than 10% of the population, suffered nonfatal injuries. Injury was the third leading cause of death overall and the leading cause of death among those aged 1 to 44 years. In the U.S., nearly 30% of years of potential life lost before age 65 results from traumatic injury, the largest contribution of any cause of death and nearly twice that of the next leading cause, cancer.4

The direct economic burden of trauma care is also considerable. The youth of the affected population and the potential chronicity of disease and complications contribute greatly to the resulting social and economic burdens. A number of studies have documented the lasting impact of trauma-related morbidity and its effect on quality-of-life,5–7 and the WHO estimates that nations can spend up to 2% of their gross domestic product caring for patients injured as the result of road traffic incidents alone.1 The Centers for Disease Control and Prevention estimate that in the U.S. $117 billion was spent on medical care for injuries in the year 2000, representing approximately 10% of national health care spending.8 The economic burden of trauma is also felt indirectly as lost work hours and productivity among injured patients and their caregivers. A regional study of trauma patient recovery conducted 18 months following hospital discharge revealed a 16% decrease in a standard measurement of functional well-being among trauma patients with a mean Injury Severity Score (ISS) of 13.5 In a recent 24 month follow-up study of German multi-system trauma victims with mean ISS of 23 conducted by one of the authors (RL), the return-to-work rate was only 50%. A similar study in Spain revealed a return-to- work rate of 58% at two years.7

The Epidemiology of Hemorrhage in Trauma

Independent of the mechanism of injury, hemorrhagic shock consistently represents the second-leading cause of early deaths among the injured, with only central nervous system (CNS) injury consistently more lethal (Table 1).9–15 Severe CNS injury is devastating and has a high rate of prehospital mortality; and there are few interventions offering hope for survival and functional recovery.16 In contrast, hemorrhage and hemorrhagic shock, which account for 30 to 40% of trauma deaths, are more amenable to interventions to reduce mortality and morbidity. Furthermore, about 25% of CNS injuries are complicated by shock.15,17 Among those with multiple injuries, brain injury remains the primary cause of death, but hypotension increases mortality in this group two- to three-fold.17,18 The significant contribution of hemorrhagic shock to brain injury mortality further illustrates the role of hemorrhage control in reducing traumatic mortality.

--------------------------------------------------------------------------------

[Help with image viewing]

[Email Jumpstart To Image] Table 1 Studies of Trauma Mortality

--------------------------------------------------------------------------------

Early Mortality

The majority of trauma deaths occur in the first few hours following injury, often before the injured patient reaches a hospital (Fig. 1).9,13,15 Hemorrhage contributes to death during the prehospital period in 33 to 56% of cases, and exsanguination is the most common cause of death among those found dead upon the arrival of emergency medical services (EMS) personnel.13 Hemorrhage accounts for the largest proportion of mortality occurring within the first hour of trauma center care, over 80% of operating room deaths after major trauma, and almost 50% of deaths in the first 24 hours of trauma care.9,12,13 After the first hours of trauma center care, CNS injury replaces hemorrhage as the leading cause of trauma mortality. Very few hemorrhagic deaths occur after the first day.9,13

--------------------------------------------------------------------------------

[Help with image viewing]

[Email Jumpstart To Image] Fig. 1. Timing and mechanism of traumatic death.Data adapted from Acosta et al.9and Sauaia et al.13 Nearly all deaths due to hemorrhage occur within the first 24 hours of injury. The rate of death from CNS injury remains relatively constant over time. After the first 24 hours, critical care complications such as organ failure and sepsis replace hemorrhage as a major cause of trauma death.* Prehospital data from Sauaia et al.13 only.** Other causes of death include combined hemorrhage/CNS, multiple organ failure, sepsis, and pulmonary embolism.CNS, central nervous system.

--------------------------------------------------------------------------------

Late Mortality and Morbidity

Early hypotension as a marker for late mortality

The presence of hemorrhagic shock is a predictor of poor outcome in the trauma patient, and the volume of hemorrhage is tied to outcome. As the amount of blood loss increases, so do resuscitation requirements and physiologic derangements including hypotension and acidosis. The volume of blood lost has proven impossible to reconstruct, but blood pressure and the number of blood units replaced are readily measured. Hypotension noted in the field or upon initial hospital evaluation is associated not only with late mortality but also specifically with the development of eventual complications including multiple organ failure (MOF) and infections such as pneumonia and sepsis.19,20 The presence of early hemorrhagic shock as defined by a systolic blood pressure less than or equal to 90 mm Hg in the pre-hospital setting or emergency department is associated with high rates of organ failure (24%) and infection (39%).20

Acidosis/Base Deficit

Early acidosis, measured as base deficit in the first hour of admission, is associated with significant hemorrhage, injury severity and hypotension.21 Early and overall red blood cell transfusion requirements increase with increasing base deficit from a mean of 2.6 units (1.4 in the first 24 hours) in patients with mild base deficits (-3 to -5) to 9.7 units (8.3 in the first 24 hours) in patients with severe base deficits (<=-10). Base deficit also predicts the development of coagulopathy, organ failure, and mortality. Patients with mild base deficits have survival rates near those of patients without acidosis (89% versus 94%), while those with severe acidosis have a nearly 50% mortality rate.21

Multi-organ Failure

Not surprisingly, trauma patients die more often of the immediately uncontrollable consequences of their injuries than of late sequelae.9,10,13,15 Delayed death is also most often due to complications developing during care rather than directly to the injuries themselves.9,10,13 Multiple organ failure, the synchronous derangement of more than one critical organ system, is the leading cause of morbidity and mortality in the trauma intensive care unit (ICU).22 The incidence of MOF following injury occurs in a bimodal pattern, with an initial peak within the first three days of hospitalization, and a second between 5 and 7 days.23,24 The combined impact of this delayed mortality is 7 to 9%. The mortality rate for patients who develop organ failure is directly related to the number of involved organ systems and can exceed 50% overall, reaching over 80% fatality with four involved systems.23,25

Despite differences among sources in the definition of failure for individual organ systems, the incidence of MOF following major trauma does appear to have decreased over the past 15 years (Fig. 2).26 In the early 1990’s, among trauma patients with injury severity scores (ISS) 19–25, mortality was reported at 13 to 15%, while by the latter part of the decade, mortality had decreased to 5%.23–25 A recently published, prospective, 12-year single-center study of MOF in trauma patients with a mean ISS of 29 reported nearly half the incidence of MOF in 2003 than in 1992 despite increasing injury severity over the same period.26 Despite the apparently decreasing incidence of MOF, however, the proportion of trauma patients with MOF listed as the cause of death remained essentially unchanged; 7% in the early 1990’s and 9% later in the decade.13,15 Similarly, throughout the follow-up period, essentially half of all trauma patients diagnosed with MOF died.23,25

--------------------------------------------------------------------------------

[Help with image viewing]

[Email Jumpstart To Image] Fig. 2. Multiple organ failure in trauma patients through the 1990s.Data adapted from Sauaia et al. 1994,23 Sauaia et al. 1995,13 Moore et al.,22 Stewart et al.,15 and Durham et al.21 While the rate of MOF has decreased in trauma patients, reflective of improvements in critical care, the proportion with it as a cause of death has not decreased over time. The mortality of MOF remains stable as well. Ciesla et al.20 have recently published their experience with MOF from 1992 through 2003 and reported that these trends continue.MOF, multiple organ failure.

--------------------------------------------------------------------------------

Sepsis

Contaminated and devitalized tissue leading to post- injury derangement of immune function puts trauma patients at high risk for sepsis. The inflammatory modulation and resulting derangement of immunity induced by hemorrhagic shock is similar to that seen in sepsis and MOF and is likely mediated through the activity of similar cytokines and pathways.27 The statistical association of massive transfusion for hemorrhagic shock with the development of MOF and with overall mortality may also be a reflection of immuno-modulation but this link has not yet been fully elucidated.23,25,28,29

Clinical Presentations in Trauma Hemorrhage

Causes of injury

Patients who present with penetrating wounds to the thorax and abdomen are at risk for severe injuries to major vessels, and therefore for massive hemorrhage, and are most likely to die during the acute phase of care.9,13 For these patients, rapid identification and control of hemorrhage is paramount, and they often require immediate surgery, especially if they are in shock.19 The difference in severity of hemorrhage from vascular injuries caused by blunt and penetrating mechanisms is unclear. Major hemorrhage from penetrating injuries is frequently not difficult to localize; however the diagnosis of the source of even severe bleeding in blunt trauma can be more challenging. In the patient with blunt trauma, the localization of hemorrhage frequently requires specialized diagnostic procedures such as computed tomography, ultrasound, and angiography to optimally control bleeding.

Causes of Shock

Hypovolemia from hemorrhage is the most common cause of shock in the trauma patient but not the only possible cause. High spinal cord injuries can cause hypotension, so-called neurogenic shock. Myocardial contusion, as well as intrinsic dysfunctional states such as myocardial infarction or heart failure, can cause cardiogenic shock. In addition, cardiovascular physiology and the response to injury can be affected by exogenous influences, most commonly beta and calcium channel blockers and ethanol.

Causes of Bleeding

Direct injury

The term “surgical bleeding” is not well-defined and not particularly helpful. Traditionally, it was intended to describe vascular and tissue disruption amenable to surgical intervention, that is, direct operative visualization and suture repair. The advent of damage control surgery, interventional radiology with embolization, and novel topical hemostatic agents has blurred the utility of this definition. Regardless of precise definition, most of what has been called surgical bleeding in the past is severe and will be rapidly fatal if not controlled.

Coagulopathy

Overt coagulopathy affects at least 1 in 4 seriously injured trauma patients.30 Etiologies include the direct effects of hemorrhage, hemodilution, hypothermia, and acidosis. The coagulopathy of trauma is directly proportional to injury severity, massive resuscitation and transfusion, and hemorrhagic shock.30–33 The presence of an abnormal prothrombin time (PT) on admission is associated with a tripling of the mortality rate of injured patients, and this mortality tends to occur early.30 Irreversible bleeding due to the coagulopathy of trauma causes the largest proportion of post-operative trauma fatalities and contributes substantially to the overall mortality of trauma.30,32–38

Coagulopathic comorbidities such as cirrhosis and hemophilia that predispose to bleeding diatheses may also be present in trauma patients. Much more commonly however, intentional or unintentional anti-coagulation due to exogenous agents, that is, pharmaceutical anti-coagulants like warfarin or anti-platelet agents like aspirin or ethanol,39 may complicate the coagulation capability of the trauma patient. As examples, clopidogrel (Plavix®) inhibits platelet function and has been demonstrated to nearly double the red blood cell transfusion requirement in cardiac surgery while increasing the platelet requirement by a factor of almost eight.40 Ibuprofen has been shown to increase operative blood loss by nearly 60% in hip arthroplasty.41 Warfarin and aspirin can increase the mortality rate of traumatic brain injury four-to-five fold.42,43

Therapeutic Considerations: Preventing Complications and Improving Outcomes

Critically injured trauma patients are treated in three, often overlapping phases: the initial control and resuscitation phase, when initial hemorrhage control and lifesaving stabilization efforts occur; the interventional phase, when definitive control of bleeding is attained; and the critical care phase, when support and restoration of normal physiology are accomplished.

Initial Control and Resuscitative Phase

Early trauma care focuses on minimizing hemorrhage and resuscitating effectively. There is no debate about the importance of hemorrhage control as a first-line measure. The optimal development and deployment of novel hemostatic agents, dressings and tourniquets are subjects of active research.

Novel Agents for Early Hemostasis

The control of bleeding and limitation of blood loss is the only means of avoiding the problems associated with massive hemorrhage in trauma. Novel methods of early hemorrhage control are under investigation. Recombinant factor VIIa (NovoSeven®, Novo-Nordisk Pharmaceuticals, Inc.) has demonstrated promise in clinical series as an adjunct to traditional measures in controlling hemorrhage in acute, life-threatening traumatic coagulopathy.44,45 Prospective trials investigating the utility of this powerful but expensive agent in traumatic hemorrhage are ongoing. Progress is also being made in the development and testing of novel dressings and dressing-adjuncts for use on externally compressible or visceral hemorrhage. The most promising of these in pre-clinical studies has been the fibrin dressing developed by the American Red Cross which has shown superior hemostatic effect in models of severe arterial (femoral and aortic) and hepatic hemorrhage.46–49 The fibrin dressing is distinguished from other available agents such as poly-N-acetyl glucosamine (chitosan and rapid deployment hemostat) bandages and granular mineral zeolite (QuickClot) in that it contains purified human fibrinogen and thrombin and thus is inherently hemostatic while other products support hemostasis primarily through adherence to and dessication of the bleeding wound, not directly through thrombogenesis.48,50–52

Tourniquets

Though uncommon in civilian trauma, exsanguination from traumatic extremity amputation has historically been a common cause of potentially preventable deaths from combat injuries.53 Tourniquet use in civilian situations is controversial and has been avoided in recent years due to what appears to be primarily a theoretical fear of limb damage or loss,54 however, military medical doctrine has adopted a liberal policy on the prehospital use of field tourniquets to prevent excessive blood loss and mortality from extremity vascular wounds.55,56

Other New Agents and Techniques

Emerging areas of research in early hemostasis for trauma include intra-cavitary agents for non-compressible bleeding and transcutaneous high frequency ultrasound. Intra-cavitary hemostatic agents are foams that can be instilled into a closed abdominal or thoracic cavity and will expand to compress a bleeding vessel, limiting blood loss before definitive control.57 Transcutaneous high frequency ultrasound claims to merge the utility of ultrasound for both the localization and control of hemorrhage by using the same ultrasound probe for the low-frequency localization of internal bleeding followed by the targeted application of high frequency sound waves to the source for coagulation of the bleeding point.58

Resuscitation

In contrast to the obvious logic and relative uniformity of opinion on hemorrhage control, current opinion on resuscitation is not as clear. There is no argument that the hemodynamically unstable patient must be supported; however, the optimal degree and agents of that support remain unclear. In a patient whose bleeding has not been definitively controlled, resuscitation to physiologically normal blood pressure may lead to “popping the clot”, that is, dislodgment of hemostatic thrombus, and further bleeding. Resuscitation to a physiologically adequate but subnormal blood pressure, so-called “hypotensive resuscitation”, before definitive hemorrhage control can be attained, has been used to avoid some of the rebleeding that occurs with resuscitation to conventional degrees.31,59–64 However, in head injury patients, the prevention of re- bleeding may be outweighed by decreased cerebral perfusion: even transient hypotension in patients with combined hemorrhage and brain injury is associated with increased mortality.17,18

Blood Product Use in Resuscitation

Although powerfully intuitive and almost universally practiced, the use of blood products in resuscitation has not been examined by the kinds of controlled clinical trials demanded now for the introduction of new clinical care products and techniques. This is because blood transfusion and resuscitation were synonymous when the practice and the term first became generally accepted in trauma care during World War I.65 However, the increasing number of studies questioning the long-term consequences of early massive transfusion for trauma 28,29 make the planning and implementation of trials to try to answer some of these questions both likely and important.

Operative Phase

Approximately 50% of patients in hemorrhagic shock are taken directly from the emergency department to the operating room.19 Because anatomically defined, so-called “surgical” bleeding, tends to be severe and can only be controlled by specialized intervention, early identification of these injuries is essential. Prompt definitive control of this kind of hemorrhage, by surgical or angiographic embolization techniques, is unarguably essential to preserve life and minimize morbidity. However, in the trauma patient who is cold and hypovolemic and becoming acidotic and coagulopathic, a “damage control” approach is now widely advocated.66 That is, life-threatening injuries, bleeding, and contamination are addressed emergently, and then the patient is taken to the ICU for warming and continued resuscitation with the goal of restoring normal hemostatic physiology before definitive surgery is attempted.

Critical Care Phase

The critical care phase begins after definitive hemorrhage control has been attained and involves completion of resuscitation, intensive monitoring, and optimization of the physiologic milieu for injury recovery. Correction of hypothermia, coagulopathy, and resuscitation to physiologic endpoints occur in this phase. As noted earlier, hemorrhage itself is not a large problem in this phase; however the degree to which massive hemorrhage and/or the blood products used to treat life-threatening hemorrhage set the stage for the principal causes of morbidity and late mortality, that is, sepsis and MOF, are of concern and remain to be adequately explored.

Unfortunately, iatrogenic injury is a well-described cause of late morbidity and mortality in trauma patients. Infections of central venous catheters are common, causing over 40% of episodes of bacteremia in patients with organ failure and much attention has been paid to preventing these infections.23 Adult respiratory distress syndrome (ARDS) occurs in between 70 and 80% of patients with MOF.24 Protective ventilatory strategies in patients with ARDS have improved outcomes.67,68

SUMMARY AND CONCLUSIONS

Injury is a world-wide problem with severe and far-ranging consequences. Much of the mortality and morbidity resulting from injury arises from hemorrhage, but many of the problems associated with severe traumatic hemorrhage are potentially solvable. Improving outcomes will require improved early hemorrhage control, resuscitation procedures, and more complete understanding of the patho-physiology of the coagulopathy of trauma, sepsis, and MOF. If the physiologic derangements of injury can be minimized with better hemorrhage control measures early in care, it seems likely that the rates of late complications and mortality will be decreased and outcomes improved.

REFERENCES

1. Peden M, McGee K, Sharma G. The injury chart book: a graphical overview of the global burden of injuries. Geneva: World Health Organization 2002. [Context Link]

2. Peden M (ed.). World report on road traffic injury prevention: summary. Geneva: World Health Organization 2004. [Context Link]

3. Krug E, Dahlberg L, Zwi A, Mercy J, Lozano R (eds.). World report on violence and health. Geneva: World Health Organization. 2002. [Context Link]

4. CDC: Web-based Injury Statistics Query and Reporting System (WISQARS). In: U.S. Department of Health and Human Services, CDC, National Center for Injury Prevention and Control. 2002. [Context Link]

5. Holbrook TL, Anderson JP, Sieber WJ, Browner D, Hoyt DB. Outcome after major trauma: 12-month and 18-month follow-up results from the Trauma Recovery Project. J Trauma. 1999, 46:765–771; discussion 771–763. [Context Link]

6. Shackford SR, Mackersie RC, Hoyt DB, et al. Impact of a trauma system on outcome of severely injured patients. Arch Surg. 1987;122:523–527. Bibliographic Links [Context Link]

7. Vazquez Mata G, Rivera Fernandez R, Perez Aragon A, Gonzalez Carmona A, Fernandez Mondejar E, Navarrete Navarro P. Analysis of quality of life in polytraumatized patients two years after discharge from an intensive care unit. J Trauma. 1996;41:326–332. Ovid Full Text Bibliographic Links [Context Link]

8. Medical expenditures attributable to injuries–United States, 2000. MMWR Morb Mortal Wkly Rep. 2004;53:1–4. Bibliographic Links [Context Link]

9. Acosta JA, Yang JC, Winchell RJ, et al. Lethal injuries and time to death in a level I trauma center. J Am Coll Surg. 1998;186:528–533. Bibliographic Links [Context Link]

10. Baker CC, Oppenheimer L, Stephens B, Lewis FR, Trunkey DD. Epidemiology of trauma deaths. Am J Surg. 1980;140:144–150. Bibliographic Links [Context Link]

11. Goris RJ, Draaisma J. Causes of death after blunt trauma. J Trauma. 1982;22:141–146. Bibliographic Links [Context Link]

12. Hoyt DB, Bulger EM, Knudson MM, et al. Death in the operating room: an analysis of a multi-center experience. J Trauma. 1994;37:426–432. [Context Link]

13. Sauaia A, Moore FA, Moore EE, et al. Epidemiology of trauma deaths: a reassessment. J Trauma. 1995;38:185–193. Ovid Full Text Bibliographic Links [Context Link]

14. Shackford SR, Mackersie RC, Holbrook TL, et al. The epidemiology of traumatic death. A population-based analysis. Arch Surg. 1993;128:571–575. Bibliographic Links [Context Link]

15. Stewart RM, Myers JG, Dent DL, et al. Seven hundred fifty-three consecutive deaths in a level I trauma center: the argument for injury prevention. J Trauma. 2003; 54:66–70; discussion 70–61. [Context Link]

16. Bouillon B, Raum M, Fach H, et al. The incidence and outcome of severe brain trauma - Design and first results of an epidemiological study in an urban area. Restor Neurol Neurosci. 1999;14:85–92. [Context Link]

17. Manley G, Knudson MM, Morabito D, Damron S, Erickson V, Pitts L. Hypotension, hypoxia, and head injury: frequency, duration, and consequences. Arch Surg. 2001;136:1118–1123. Ovid Full Text Bibliographic Links [Context Link]

18. Chesnut RM, Marshall SB, Piek J, Blunt BA, Klauber MR, Marshall LF. Early and late systemic hypotension as a frequent and fundamental source of cerebral ischemia following severe brain injury in the Traumatic Coma Data Bank. Acta Neurochir Suppl (Wien). 1993;59:121–125. [Context Link]

19. Franklin GA, Boaz PW, Spain DA, Lukan JK, Carrillo EH, Richardson JD. Prehospital hypotension as a valid indicator of trauma team activation. J Trauma. 2000;48:1034–1037; discussion 1037–1039. [Context Link]

20. Heckbert SR, Vedder NB, Hoffman W, et al. Outcome after hemorrhagic shock in trauma patients. J Trauma. 1998;45:545–549. Ovid Full Text Bibliographic Links [Context Link]

21. Davis JW, Parks SN, Kaups KL, Gladen HE, O’Donnell-Nicol S. Admission base deficit predicts transfusion requirements and risk of complications. J Trauma. 1996;41:769–774. Ovid Full Text Bibliographic Links [Context Link]

22. Marshall JC, Cook DJ, Christou NV, Bernard GR, Sprung CL, Sibbald WJ. Multiple organ dysfunction score: a reliable descriptor of a complex clinical outcome. Crit Care Med. 1995;23:1638–1652. Ovid Full Text Bibliographic Links [Context Link]

23. Durham RM, Moran JJ, Mazuski JE, Shapiro MJ, Baue AE, Flint LM. Multiple organ failure in trauma patients. J Trauma. 2003;55:608–616. Ovid Full Text Bibliographic Links [Context Link]

24. Moore FA, Sauaia A, Moore EE, Haenel JB, Burch JM, Lezotte DC. Postinjury multiple organ failure: a bimodal phenomenon. J Trauma. 1996;40:501–510; discussion 510–502. [Context Link]

25. Sauaia A, Moore FA, Moore EE, Haenel JB, Read RA, Lezotte DC. Early predictors of postinjury multiple organ failure. Arch Surg. 1994;129:39–45. Bibliographic Links [Context Link]

26. Ciesla DJ, Moore EE, Johnson JL, Burch JM, Cothren CC, Sauaia A. A 12-year prospective study of postinjury multiple organ failure. Arch Surg. 2005;140:432–440. Ovid Full Text Bibliographic Links [Context Link]

27. Angele MK, Faist E. Clinical review: immunodepression in the surgical patient and increased susceptibility to infection. Crit Care. 2002;6:298–305. Bibliographic Links [Context Link]

28. Malone DL, Dunne J, Tracy JK, Putnam AT, Scalea TM, Napolitano LM. Blood transfusion, independent of shock severity, is associated with worse outcome in trauma. J Trauma. 2003;54:898–905; discussion 905–897. [Context Link]

29. Moore FA, Moore EE, Sauaia A. Blood transfusion. An independent risk factor for postinjury multiple organ failure. Arch Surg. 1997; 132:620–624; discussion 624–625. [Context Link]

30. MacLeod JB, Lynn M, McKenney MG, Cohn SM, Murtha M. Early coagulopathy predicts mortality in trauma. J Trauma. 2003;55:39–44. [Context Link]

31. Bickell WH, Wall MJ Jr., Pepe PE, et al. Immediate versus delayed fluid resuscitation for hypotensive patients with penetrating torso injuries. N Engl J Med. 1994;331:1105–1109. Ovid Full Text Bibliographic Links [Context Link]

32. Brohi K, Singh J, Heron M, Coats T. Acute traumatic coagulopathy. J Trauma. 2003;54:1127–1130. Ovid Full Text Bibliographic Links [Context Link]

33. MacLeod J, Lynn M, McKenney MG, Jeroukhimov I, Cohn SM. Predictors of mortality in trauma patients. Am Surg. 2004;70:805–810. Bibliographic Links [Context Link]

34. Lynn M, Jeroukhimov I, Klein Y, Martinowitz U. Updates in the management of severe coagulopathy in truama patients. Intensive Care Med. 2002; 28 (suppl):S241–S247. Bibliographic Links [Context Link]

35. Cosgriff N, Moore EE, Sauaia A, Kenny-Moynihan M, Burch JM, Galloway B. Predicting life-threatening coagulopathy in the massively transfused trauma patient: Hypothermia and acidoses revisited. J Trauma. 1997;42:857–861. Ovid Full Text Bibliographic Links [Context Link]

36. Ferrara A, MacArthur JD, Wright HK, Modlin IM, McMillen MA. Hypothermia and acidosis worsen coagulopathy in the patient requiring massive transfusion. Am J Surg. 1990;160:515–518. Bibliographic Links [Context Link]

37. Staab DB, Sorensen VJ, Fath JJ, Raman SBK, Horst HM, Obeid FN. Coagulation defects resulting from ambient temperature-induced hypothermia. J Trauma. 1994;36:634–638. Bibliographic Links [Context Link]

38. Watts DD, Trask A, Soeken K, Perdue P, Dols S, Kaufmann C. Hypothermic coagulopathy in trauma: Effect of varying levels of hypothermia on enzyme speed, platelet function, and fibrinolytic activity. J Trauma. 1998;44:846–854. Ovid Full Text Bibliographic Links [Context Link]

39. Rubin R, Rand ML. Alcohol and platelet function. Alcohol Clin Exp Res. 1994;18:105–110. Bibliographic Links [Context Link]

40. Chen L, Bracey AW, Radovancevic R, et al. Clopidogrel and bleeding in patients undergoing elective coronary artery bypass grafting. J Thorac Cardiovasc Surg. 2004;128:425–431. Bibliographic Links [Context Link]

41. Slappendel R, Weber EW, Benraad B, Dirksen R, Bugter ML. Does ibuprofen increase perioperative blood loss during hip arthroplasty? Eur J Anaesthesiol. 2002;19:829–831. Bibliographic Links [Context Link]

42. Mina AA, Bair HA, Howells GA, Bendick PJ. Complications of preinjury warfarin use in the trauma patient. J Trauma. 2003;54:842–847. Ovid Full Text Bibliographic Links [Context Link]

43. Mina AA, Knipfer JF, Park DY, Bair HA, Howells GA, Bendick PJ. Intracranial complications of preinjury anticoagulation in trauma patients with head injury. J Trauma. 2002;53:668–672. Ovid Full Text Bibliographic Links [Context Link]

44. Dutton RP, McCunn M, Hyder M, et al. Factor VIIa for Correction of Traumatic Coagulopathy. J Trauma. 2004;57:709–719. Ovid Full Text Bibliographic Links [Context Link]

45. Geeraedts LM Jr., Kamphuisen PW, Kaasjager HA, Verwiel JM, van Vugt AB, Frolke JP. The role of recombinant factor VIIa in the treatment of life-threatening haemorrhage in blunt trauma. Injury. 2005;36:495–500. Bibliographic Links [Context Link]

46. Holcomb JB, Pusateri AE, Harris RA, et al. Effect of dry fibrin sealant dressings versus gauze packing on blood loss in grade V liver injuries in resuscitated swine. J Trauma. 1999;46:49–57. Ovid Full Text Bibliographic Links [Context Link]

47. Larson MJ, Bowersox JC, Lim RC Jr., Hess JR. Efficacy of a fibrin hemostatic bandage in controlling hemorrhage from experimental arterial injuries. Arch Surg. 1995;130:420–422. Ovid Full Text Bibliographic Links [Context Link]

48. Pusateri AE, Modrow HE, Harris RA, et al. Advanced hemostatic dressing development program: animal model selection criteria and results of a study of nine hemostatic dressings in a model of severe large venous hemorrhage and hepatic injury in Swine. J Trauma. 2003;55:518–526. Ovid Full Text Bibliographic Links [Context Link]

49. Sondeen JL, Pusateri AE, Coppes VG, Gaddy CE, Holcomb JB. Comparison of 10 different hemostatic dressings in an aortic injury. J Trauma. 2003;54:280–285. Ovid Full Text Bibliographic Links [Context Link]

50. Alam HB, Burris D, DaCorta JA, Rhee P. Hemorrhage control in the battlefield: role of new hemostatic agents. Mil Med. 2005;170:63–69. Bibliographic Links [Context Link]

51. Alam HB, Uy GB, Miller D, et al. Comparative analysis of hemostatic agents in a swine model of lethal groin injury. J Trauma. 2003;54:1077–1082. Ovid Full Text Bibliographic Links [Context Link]

52. King DR, Cohn SM, Proctor KG. Modified rapid deployment hemostat bandage terminates bleeding in coagulopathic patients with severe visceral injuries. J Trauma. 2004;57:756–759. Ovid Full Text Bibliographic Links [Context Link]

53. Bellamy RF. The causes of death in conventional land warfare: implications for combat casualty care research. Mil Med. 1984;149:55–62. [Context Link]

54. Walters TJ, Mabry RL. Issues related to the use of tourniquets on the battlefield. Mil Med. 2005;170:770–775. Bibliographic Links [Context Link]

55. Lakstein D, Blumenfeld A, Sokolov T, et al. Tourniquets for hemorrhage control on the battlefield: a 4-year accumulated experience. J Trauma. 2003;54(5 Suppl):S221–S225. Ovid Full Text Bibliographic Links [Context Link]

56. Lounsbury DE, Bellamy RF (eds.). Emergency War Surgery, Third edn. Washington, DC: Borden Institute, Walter Reed Army Medical Center. 2004. [Context Link]

57. Kheirabadi B, Klemcke HG. Hemostatic Agents for Control of Intracavitary Non-Compressible Hemorrhage: An Overview of Current Results. Presented at the Combat Casualty Care in Ground Based Tactical Situations Symposium, 16–18 August. 2004. [Context Link]

58. Cornejo CJ, Vaezy S, Jurkovich GJ, Paun M, Sharar SR, Martin RW. High-intensity ultrasound treatment of blunt abdominal solid organ injury: an animal model. J Trauma. 2004;57:152–156. Ovid Full Text Bibliographic Links [Context Link]

59. Burris D, Rhee P, Kaufmann C, et al. Controlled resuscitation for uncontrolled hemorrhagic shock. J Trauma. 1999;46:216–223. Ovid Full Text Bibliographic Links [Context Link]

60. Capone AC, Safar P, Stezoski W, Tisherman S, Peitzman AB. Improved outcome with fluid restriction in treatment of uncontrolled hemorrhagic shock. J Am Coll Surg. 1995;180:49–56. Bibliographic Links [Context Link]

61. Kowalenko T, Stern S, Dronen S, Wang X. Improved outcome with hypotensive resuscitation of uncontrolled hemorrhagic shock in a swine model. J Trauma. 1992;33:349–353; discussion 361–342. [Context Link]

62. Mapstone J, Roberts I, Evans P. Fluid resuscitation strategies: a systematic review of animal trials. J Trauma. 2003;55:571–589. Ovid Full Text Bibliographic Links [Context Link]

63. Sondeen JL, Coppes VG, Holcomb JB. Blood pressure at which rebleeding occurs after resuscitation in swine with aortic injury. J Trauma. 2003;54(Suppl):S110–S117. Ovid Full Text Bibliographic Links [Context Link]

64. Stern SA. Low-volume fluid resuscitation for presumed hemorrhagic shock: helpful or harmful? Curr Opin Crit Care. 2001;7:422–430. Ovid Full Text Bibliographic Links [Context Link]

65. Cushing H. From a surgeon’s journal. Boston: Little, Brown; 1936. [Context Link]

66. Rotondo MF, Schwab CW, McGonigal MD, et al. ‘Damage control’: an approach for improved survival in exsanguinating penetrating abdominal injury. J Trauma. 1993;35:375–382; discussion 382–373. [Context Link]

67. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. The Acute Respiratory Distress Syndrome Network. N Engl J Med. 2000;342:1301–1308. [Context Link]

68. Gajic O, Dara SI, Mendez JL, et al. Ventilator-associated lung injury in patients without acute lung injury at the onset of mechanical ventilation. Crit Care Med. 2004;32:1817–1824. Ovid Full Text Bibliographic Links [Context Link]

DISCUSSION

Dr. Jeff Johnson: The experienced trauma clinician understands that all bleeding stops. However, as Dr. Kauvar has pointed out, the demise of the patient too often precedes the cessation of bleeding. Through a series of well-thought-out experiments Carl Wiggers nicely described that the depth of shock and the duration of shock are the primary determinants of outcome. While we now have more elegant ways of describing the depth of shock and studying oxygen debt and oxygen delivery, we don’t yet understand the best way to reverse shock.

I have three general questions. The first relates to the argument that limited resuscitation strategies should prevent or limit further hemorrhage. The human data for this are largely derived from a study of patients with penetrating torso injuries who are very close to definitive care of their injuries. Looking at blunt or mixed trauma populations, limited resuscitation, while apparently doing no harm, appears to be no better. In recent US military operations, with smaller units in more dispersed areas and using body armor, extremity injuries are what we’re seeing. My question is: How are we going to assess the efficacy of limited resuscitation in this kind of setting? What kind of data are needed to see whether limited resuscitation is effective in an extremity injury, distant from definitive care? Further, does this concept apply to the patient with a successfully applied tourniquet or a successfully applied hemostatic bandage?

My second question relates to the “bloody vicious cycle”. As the authors point out, coagulopathy, hypothermia and acidosis can doom either operative or non-operative management. My question relates to the timing of intervention. With the inherent delay in laboratory testing, are we identifying these patients quickly enough? What is the best way to predict the patient who will suffer that problem? If you had one test at the arrival of the patient to help predict that, what would it be?

Lastly, a question for the trauma surgeons in the group. Hemorrhage control is no longer just about stitches. It is a multi-disciplinary project. Should we as surgeons abdicate our role as those who are the champions of hemostasis, or shall we make sure that we are adequately trained in techniques of angio-embolization, hemorrhage control, transfusion medicine, and component therapy?

Dr. Angus Wells: I would like to make some comments about the epidemiology of trauma from a blood bank point of view. The international and national data that you used to illustrate the problem of trauma is very illuminating. We’ve collected population-based data on blood use in the north of England for some years now. The area has a population of about 2.9 million and we’ve consistently found that trauma needs 6% of our blood supply, less than the estimates for America. Yet within that, half is for fractured neck or femurs, that is, frail elderly patients who have perhaps been topped up before surgery. Hospital episode statistics for the National Health Service in England, show only 2.4% of admissions were for trauma. And within that group, only 0.28% were major or multiple trauma. Therefore I am suggesting that there is a very small group of heavily transfused trauma patients who don’t use a huge amount of the blood supply but, as individuals, are a high-volume, high-risk group.

Dr. David Kauvar: Dr. Johnson asked about limited resuscitation. ISR researchers have been working on animal models of limited resuscitation or hypotensive resuscitation, and their shock models show that hypotensive resuscitation prevents re-bleeding and limits further blood loss before definitive control can be attained. Human studies are limited. One problem in implementing hypotensive resuscitation strategies, whether in a study or if this becomes accepted practice, is coordination, because you have to start in the initial, or control, phase. Do you have anything to add, Dr. Wade?

Dr. Charlie Wade: Well, the question basically is how low for how long? This is the axiom that Drs. Sondeen, Dubick and Holcomb of our group here at ISR have come up with. We’re still trying to understand those parameters. We have an idea, as mentioned, that the penetrating trauma patient is probably the place to apply this strategy. But then how long can you sustain that patient in the hypotensive state? In the animal models, it looks on the order of about four hours at 80 mm Hg systolic arterial pressure. This still has to be addressed in the clinical setting.

Dr. Rick Dutton: When we did a hypotension trial in Maryland, we used definitive control of hemorrhage as an end point. It’s usually easy to tell clinically when a person has stopped bleeding. Extremity injuries have usually stopped bleeding by the time they arrive at the trauma center because hemostatic mechanisms are very effective in controlling the bleeding. It isn’t until we start working on the patient that we make them re-bleed, whether through fluid resuscitation or surgery. I think that is going to be a very important point, in defining how we approach this and define future studies.

Dr. Harvey Klein: Can you define what you mean by limited resuscitation? Are we talking just about hypotensive resuscitation, no matter what you use, or are there other aspects to that? Are you using blood pressure as your measurement?

Dr. Dutton: When all you have is a hammer, everything looks like a nail. We use blood pressure because that’s what we have. But obviously there’s a huge difference whether you’re resuscitating with water or with blood products. And there’s a huge difference whether you’re vasoconstricted or vasodilated. So anesthetic state is important as well.

Dr. Fred Moore: I moved from Denver General, which was a knife and gun club, to Memorial Hermann Hospital in Houston, which is basically a blunt trauma hospital. And my second round of education as a trauma surgeon began, because a blunt trauma patient is very complex. And probably the biggest problem is assessing the severity of shock when the patient arrives. It’s hugely variable, and it’s not that easy. If we just had something we could slap on somebody, and say, this patient is really in shock, so we can’t be wasting time. Or, this patient isn’t in very bad shock, so we can let their blood pressure sit at 80 while we call in the OR team, that would be real nice. But we don’t have that right now.

Dr. John Holcomb: I think we’ve identified several issues here so far. One is that only a small proportion of trauma patients receive blood. That small group of patients gets a large amount of blood, so that’s our study target for just about anything that deals with shock and resuscitation.

Dr. Mauricio Lynn: To be a little bit controversial, why do we need to focus on doing pre-hospital clinical trials on hypotensive resuscitation if the time frame in major urban centers, which is most of our trauma patients, from injury to arrival at the trauma center, is extremely short?

Dr. Holcomb: I think we must do prehospital studies. As many hemorrhagic deaths occur prehospital as occur early in hospital care. Most of us in this room represent the in-the-hospital care group. But you have an equal number of patients who do poorly prehospital. It is less controlled and more difficult, but it’s just as big a problem. And if you’re going to apply your solution, whatever that is, your studies probably need to start prehospital.

Dr. Jeffrey Lawson: I want to debate one of the paper’s theses, that stopping bleeding directly relates to changes in multi-system organ failure. I think that’s implied data. Coagulation is an inflammatory pathway; it’s a host-defense system set off by a number of different mechanisms. Overdriving coagulation is probably the same as overdriving inflammatory pathways. So one needs to be very careful, especially with the systemic biologics that are becoming available. Assuming that using potent, potentially inflammatory, molecules is also going to limit multi-system organ failure, when in fact it might drive systemic inflammation, is somewhat improbable, given what we already know.

Dr. Kauvar: It’s overwhelmingly intuitive, though, that stopping the blood loss and decreasing the early physiologic derangement that occurs as a result of the blood loss will improve the outcome later on.

Dr. Lawson: My fear is that if we think of stopping bleeding as a uni-directional path that we will drive microvascular thrombosis systemically into inflammatory pathways.

Dr. Uri Martinowitz: The multicenter trauma trial, which was published in J Trauma found the opposite (Boffard KD et al. Recombinant factor VIIa as adjunctive therapy for bleeding control in severely injured trauma patients: two parallel randomized, placebo-controlled, double-blind clinical trials. J Trauma. 2005;59:8–18). Earlier control of hemorrhage showed a trend toward reducing ARDS and multi-system organ failure. We were very concerned that the complex formation between rFVIIa and tissue factor can increase inflammation, and apparently it is the opposite. And the same was in the brain: it decreases not only the hematoma but also edema.

I would like to make a comment. It’s time to realize that there is no such thing as coaguloapathic versus surgical bleeding, unless you cut the vessel. But after half an hour this also becomes coagulopathy because these patients are also coagulopathic. Every trauma patient is coagulopathic; it’s just our primitive tests that cannot measure most of the coagulopathy. We just miss 90% of the coagulopathies.

Dr. John Owings: I’m going to cautiously agree with Dr. Martinowitz. Coagulopathy means you have derangement of the coagulation cascade. We had the opportunity to study about 200 trauma patients and do a battery of coagulation assays from the moment they hit the emergency department. They were all coagulopathic, that is, the patients that come in at least sick enough to activate a physiologic or a mechanistic criterion. And those are the ones you care about, the injured patients who, as a rule, all have up-regulation of their coagulation cascade. The question is whether they go off the end of that spectrum into the disorganized coagulopathy that gets called medical bleeding. It’s all exactly the same thing and it comes from the same place.

Dr. Martinowitz: We have to be very careful in interpreting the tests. Like with TEG tests, you demonstrate the shortening of the clotting time and you interpret it as hypercoagulation. First, it is not necessarily hypercoagulation. And second, there are about 50 different changes in the coagulation system. So you cannot take one or two or three simple measurements and decide about the global response of the coagulation system. Now, every hypothermic patient is coagulopathic by definition. We usually underestimate the effect of hypothermia since we take the blood samples at the low temperature of the patient and the samples are warmed to 37°C. In addition we are measuring a test tube phenomenon. Nobody is checking what’s actually happening in the body. Another example: acidosis. The acidotic patient is severely coagulopathic, there is 70% inhibition in thrombin generation at pH 7.2, but we don’t measure it. Another example: hyperfibrinolysis. Do we measure fibrinolysis? So I think we should really start to think that every massively bleeding patient is coagulopathic. It’s just a matter of timing. As Marcel Levi demonstrated, the patients are hypo-coagulopathic initially, and the next day they start to be more hypercoagulopathic.

Dr. Holcomb: I think many of you can now understand why it was so difficult to design some of the controlled studies that we’ve been working on for several years. The definitions are not consistent, the epidemiology is unclear, the intended study group is small, the tests are not very good and the variation is very broad.

I have a question for Dr. Moore, or anyone else in the group about improved hemostasis preventing multi-organ failure. That’s been discussed extensively in a lot of small groups around the country. Do you believe that implied the link of the data?

Dr. Moore: It depends on how you’re controlling it. I’m concerned that coagulopathy may just be a marker of adverse outcome. If you treat the coagulopathy with whatever hammer you have, you might hurt patients. A good example is blood transfusions. I was trained to transfuse liberally but the most recent data (some of which I generated) strongly suggests blood transfusions are harmful and may be a contributing factor in MOF pathogenesis.

Dr. Holcomb: Dr. Pruitt has made the point that all bleeding will stop within the first 12 to 24 hours, one way or the other. So our timeline to intervene is very short. And if we’re not successful, our patients die.

[Ace844]

(Timing of Fluid Resuscitation Shapes the Hemodynamic Response to Uncontrolled Hemorrhage: Analysis Using Dynamic Modeling

[Original Articles)

Hirshberg, Asher MD; Hoyt, David B. MD; Mattox, Kenneth L. MD

From the Michael E. DeBakey Department of Surgery, Baylor College of Medicine, Houston, Texas (A.H., K.L.M.); and the Department of Surgery, Division of Trauma, Burns and Critical Care, University of California San Diego School of Medicine, San Diego, California (D.B.H.).

Submitted for publication August 27, 2004.

Accepted for publication April 21, 2005.

Address for Reprints: Kenneth L. Mattox, MD, Michael E. DeBakey Department of Surgery, Baylor College of Medicine, One Baylor Plaza, Houston, Texas 77030; email: kmattox@bcm.tmc.edu.]

Abstract

Background: Timing of fluid resuscitation with respect to intrinsic hemostasis is an unexplored aspect of uncontrolled hemorrhage, because most animal models do not allow direct monitoring of blood loss. The aim of this study was to define how timing of crystalloid administration affects the bleeding patient’s hemodynamic response to fluids, using a computer model of blood volume changes during uncontrolled hemorrhage.

Methods: A multi-compartment lumped-parameter deterministic model of intravascular volume changes in a bleeding adult patient was developed and implemented. The model incorporates empirical mathematical descriptions of intrinsic hemostasis and rebleeding.

Results: The predicted hemodynamic response to uncontrolled hemorrhage closely corresponds to that seen in animal studies. A 2-L crystalloid bolus given during ongoing hemorrhage increases blood loss by 4 to 29%, an effect that is inversely related to the initial bleeding rate. A similar bolus given after intrinsic hemostasis may trigger rebleeding if given when the hemostatic clot is mechanically vulnerable. This period of clot vulnerability (ranging from 0–34 minutes) changes with both the initial bleeding rate and the rate of fluid administration.

Conclusions: The timing of crystalloid administration with respect to intrinsic hemostasis shapes the bleeding patient’s hemodynamic response. An early bolus delays hemostasis and increases blood loss, while a late bolus may trigger rebleeding. These observations provide valuable insight into the hemodynamic response to fluid resuscitation.

--------------------------------------------------------------------------------

Timing of fluid resuscitation with respect to intrinsic hemostasis is an unexplored aspect of uncontrolled hemorrhage. Animal studies of crystalloid resuscitation for hemorrhagic shock have traditionally focused on either fluid volume and composition, or on the target blood pressure.1,2 In these studies, fluid resuscitation was empirically begun 15 to 30 minutes from injury, roughly imitating “early” or “delayed” resuscitation. However, it is unknown whether fluids were given while hemorrhage was still ongoing or after bleeding has ceased because of intrinsic hemostasis.3–5

If uncontrolled hemorrhage is conceptualized as a pressure-driven outflow from a container (a “leaky bucket”) checked by intrinsic hemostasis, it is intuitively obvious that identical fluid boluses given either during hemorrhage or after cessation of bleeding may have different hemodynamic effects. This crucial aspect of timing, rather than fluid volume or composition, has not been systematically addressed in animal studies of uncontrolled hemorrhage.

There is good reason for this ambiguity. Most animal models of internal hemorrhage are based on bleeding inside a closed abdomen,3,6–8 without direct monitoring of blood loss and hemostasis. Even in recent models that allowed such monitoring 9–11 or in models of external hemorrhage,12,13 fluids were given at predetermined times, regardless of the onset of intrinsic hemostasis.

Research questions not easily accessible to traditional experimentation can sometimes be addressed using other quantitative tools such as dynamic computer modeling.14 Lewis 15 pioneered the use of computer modeling to explore the effects of fluid resuscitation during uncontrolled hemorrhage and showed that a computer model is a tightly controlled experimental system that allows simulation of a wide range of fluid resuscitation scenarios. The methodology involves constructing a set of mathematical equations to capture the key elements of a problem, translating them into a computer program and solving them numerically. The result is a dynamic simulation that often provides surprisingly useful insight into the behavior of the system with time. Uncontrolled hemorrhage is particularly well suited to this approach because there are good mathematical descriptions of the key variables, and the feedback loops are well defined. Most importantly, the temporal relationship between fluid administration and intrinsic hemostasis, so difficult to control in animal models, can be easily manipulated in a mathematical model.

The aim of this study was to explore how timing of crystalloid administration affects the patient’s hemodynamic response to fluids across a range of bleeding scenarios, using a computer model of blood volume changes during uncontrolled hemorrhage.

MATERIALS AND METHODS

A simple multi-compartment deterministic model of intravascular volume changes in a bleeding 70-kg adult patient was developed and implemented using the graphical modeling language STELLA 8.0 (High Performance Systems, Hanover, N.H.). The model consists of a lumped-parameter mathematical description of the pressure-volume relationship of the systemic circulation. Intravascular volume loss is because of pressure-driven bleeding and volume gain comes from crystalloid administration and transcapillary refill. Figure 1 is a STELLA block diagram of the key model components, and the entire set of model equations is given in the Appendix.

--------------------------------------------------------------------------------

[Help with image viewing]

[Email Jumpstart To Image] Fig. 1. A diagram of the STELLA model of intravascular volume changes in a bleeding adult patient. The intravascular volume (accumulation) is represented by a rectangle, while bleeding, rebleeding, transcapillary refill and crystalloid administration are control variables (flows) that update the intravascular volume every dt. Factors affecting the flows are marked by circles.

--------------------------------------------------------------------------------

A simple empirical continuous function was used to translate the intravascular volume (baseline of 5 L) into a corresponding mean arterial pressure (MAP). The function was calibrated to produce hypotension (MAP <60 mm Hg) with acute volume loss of 30%, and MAP of 0 mm Hg with volume loss of more than 50%. These parameters correspond to the clinical description of hemorrhagic shock in the Advanced Trauma Life Support course of the American College of Surgeons and are compatible with previously published large animal and mathematical models of hemorrhagic shock.15–17 Pressure-driven hemorrhage was modeled across a fixed resistance,16 simulating hemorrhage from a side hole in a large artery. Different initial bleeding rates represent varying laceration sizes.

The mathematical description of intrinsic hemostasis was developed from the swine aortotomy model of Bickell et al.,18 which contains detailed blood pressure measurements during the first few minutes after a major abdominal vascular injury. This study demonstrated a sharp exponential decline in MAP with time, slowing of hemorrhage as the MAP approached 25 mm Hg, and complete cessation of bleeding when bleeding was very slow. In addition, Shaftan and Harris 19,20 showed that arterial hemorrhage is controlled by formation of an extraluminal clot, a time-dependent process that occurs when the animal is hypotensive but not when the animal is normotensive. Based on all these experimental data, we modeled intrinsic hemostasis as both a flow-dependent and a time-dependent process. Flow-dependent hemostasis was represented as a step function that stops hemorrhage for bleeding rates below 0.005 L/min (5 mL/min). Time dependent clot formation was represented mathematically as a linear function that operates between the values of 1 (no clot) and 0 (full clot formation, no bleeding). This gradual extraluminal clot formation is triggered by hypotension (MAP <60), and lasts 10 minutes, which is the mean normal value of whole blood clotting time and also corresponds to the experimental observations of Shaftan et al.19 The resulting blood pressure curve closely corresponds to those obtained from published animal models of major abdominal vascular injury.9,11,18,21

Because setting the rebleeding threshold at a fixed value 11 disregards the process of clot maturation and gradual increase in mechanical strength of the clot, we used a time-dependent linear function that gradually increases the threshold MAP between 50 to 100 mm Hg over a period of 60 minutes.

The model assumes crystalloid infusion at a constant rate, typically 0.1 to 0.2 L/min, representing a rapid bolus given through 1 to 2 large bore peripheral intravenous lines. The infused volume undergoes an exponential decay with a half-life of 17 minutes,15,22 yielding a retained intravascular crystalloid fraction of approximately 1:4 for the time frame simulated in the study. This also corresponds to the experimental observations and mathematical model of crystalloid retention by Cervera and Moss.23

Transcapillary refill was computed from the area under the pressure deficit curve (the difference between the baseline and instantaneous MAP), using a modification of the approach of Simpson et al.,22,24 up to a maximum of 15% of the initial blood volume.15 Thus the rate of transcapillary filling is directly proportional to the volume deficit. This yielded a gradual increase in MAP to values between 40 to 50 mm Hg at 60 minutes, closely corresponding to those observed in animal models of major hemorrhage.7,9,11,18,21

The model equations were solved numerically using Euler’s integration method for a dt of 0.01 minute. Each simulation was run for 60 minutes from injury. Crystalloid administration during ongoing hemorrhage was labeled “early” while administration after bleeding has stopped was labeled “late” in all simulation runs.

RESULTS

General Behavior of the Model

The model was designed to simulate hemorrhagic shock in an urban trauma context. The behavior of the model without intravenous fluids for two initial bleeding rates (0.1 and 1.0 L/min) shows a rapid drop in blood pressure until intrinsic hemostasis is activated, followed by gradual restoration of blood pressure through transcapillary refill (Fig. 2).

--------------------------------------------------------------------------------

[Help with image viewing]

[Email Jumpstart To Image] Fig. 2. Model predicted changes in MAP during pressure-driven hemorrhage at initial bleeding rates of 0.1 and 1 L/min with the intrinsic hemostasis activited at a flow rate below 5 mL/min.

--------------------------------------------------------------------------------

The model predicts that an injured patient with a fast initial bleeding rate of 1.0 L/min (representing a major vascular injury) bleeds for 5.5 minutes until intrinsic hemostasis, while a patient with a slower initial rate (0.1 L/min) bleeds for 24.8 minutes. Thus, at approximately 15 minutes from injury (a realistic time frame for a paramedic to arrive in an urban scenario), both the faster and slower bleeders will be hypotensive (MAP <60 mm Hg). However, the former will already be in the “self resuscitation” phase with a gradual increase in MAP after bleeding has stopped, while the latter will still have ongoing blood loss.

Early Versus Late Fluid Administration

The predicted hemodynamic effect of early or late crystalloid administration was compared using a reference scenario of initial bleeding at 0.2 L/min, yielding a blood loss of 1.90 L during 17.4 minutes until intrinsic hemostasis occurs.

The model was used to compare the hemodynamic response to a crystalloid bolus (2 L given over 10 minutes) set to begin either 5 minutes after the onset of bleeding (“early”) or 5 minutes after bleeding has stopped (“late”). The early bolus had a minimal effect on the time to intrinsic hemostasis (18.7 minutes) and increased blood loss by 0.25 L (13%) above that of the reference scenario without fluids. The same crystalloid bolus, begun 5 minutes after intrinsic hemostasis, resulted in an entirely different hemodynamic profile with an episode of rebleeding and additional blood loss of 0.38 L (20% above baseline) (Fig. 3).

--------------------------------------------------------------------------------

[Help with image viewing]

[Email Jumpstart To Image] Fig. 3. Model prediction of the effect of early and late crystalloid bolus on the hemodynamic response to uncontrolled hemorrhage at an initial bleeding rate of 0.2 L/min.

--------------------------------------------------------------------------------

Effect of Early Fluid Administration

The effect of crystalloid volume and initial bleeding rate on blood loss during early fluid administration was analyzed by simulating a rapid fluid bolus that is completed before the onset of intrinsic hemostasis (Table 1). The model predicts that additional blood loss above the baseline (without fluids) is inversely proportional to the initial bleeding rate. Thus, with an initial bleeding rate of 0.1 L/min (which causes hypotension in 15 minutes), a rapid 2-L bolus increases blood loss by 0.48 L (29%) compared with no fluids. With a threefold increase in the initial bleeding rate, the same bolus increases blood loss by only 0.09 L (4%) compared with no fluids. The reason for this difference is that in the slower bleeder a 2-L bolus slows the drop in MAP, delaying the onset of intrinsic hemostasis by 5 minutes as compared with only 0.2 minutes in the faster bleeder.

--------------------------------------------------------------------------------

[Help with image viewing]

[Email Jumpstart To Image] Table 1 Effect of Crystalloid Volume and Initial Bleeding Rate on Blood Loss in Early Fluid Resuscitation

--------------------------------------------------------------------------------

Effect of Late Fluid Administration

When crystalloids are given after the onset of intrinsic hemostasis, rebleeding occurs when the combined effect of IV fluids and transcapillary refill on the MAP exceeds the rebleeding threshold, a linear time-dependent function that operates between 50 to 100 mm Hg. Thus, the model defines a time interval after the onset of intrinsic hemostasis during which a crystalloid bolus will trigger rebleeding because the hemostatic clot is mechanically vulnerable (Table 2). The duration of this vulnerability period is inversely related to the initial bleeding rate, because higher bleeding rates translate into earlier onset of intrinsic hemostasis at a lower MAP, and therefore more time for the clot to gain mechanical strength. With higher initial bleeding rates (0.4 L/min or more) only fluid volumes in excess of 2 L, given at rates over 0.2 L/min will trigger rebleeding.

--------------------------------------------------------------------------------

[Help with image viewing]

[Email Jumpstart To Image] Table 2 Effect of Initial Bleeding Rate and Fluid Infusion Rate on the Period of Clot Vulnerability to Rebleeding in Late Fluid Administration

--------------------------------------------------------------------------------

DISCUSSION

The central message of this modeling study is that crystalloid administration before or after intrinsic hemostasis has entirely different hemodynamic effects. An early bolus increases blood loss and delays intrinsic hemostasis, particularly for slower bleeding rates. A late bolus may trigger rebleeding if given soon after the onset of intrinsic hemostasis, adding to the blood loss.

The unique feature of our model is the incorporation of intrinsic hemostasis and rebleeding, making the model clinically relevant despite its simplicity.25 No previous model has captured these two key elements of the hemodynamic response to uncontrolled bleeding.15,16,22–24,26–28 Our model allows a critical examination of some widely held beliefs about fluid resuscitation for uncontrolled hemorrhage in an urban trauma setting. If the patient is hypotensive when seen by a first responder roughly 15 minutes from injury, the initial bleeding rate must be at least 0.1 L/min to produce a 1.5 L volume loss at 15 minutes. At these high initial bleeding rates, a 2-L bolus, even if administered at a fast rate of 0.2 L/min will not correct hypotension (even transiently) if the patient still has ongoing hemorrhage. The most such an early bolus can do is augment blood loss. Therefore, only a patient who has already stopped bleeding can be a “responder” to a crystalloid bolus. However, our model shows that in such a responder, a standard 2-L bolus may trigger rebleeding if given during the period of mechanical vulnerability of the hemostatic clot. Therefore, a “transient responder,” as defined by the Advanced Trauma Life Support course of the American College of Surgeons, must be either a patient who has already stopped bleeding (and in whom the fluid bolus triggered rebleeding) or a very slow bleeder who will not become hypotensive within 15 or even 30 minutes from injury.

The model also provides new insight into resuscitation-induced rebleeding. It is obvious from our analysis that rebleeding reflects a race between the restoration of MAP (by the combined effects of crystalloids and transcapillary refill) and clot maturation. The model shows that rebleeding depends on fluid administration during a period of mechanical vulnerability of the clot after the onset of intrinsic hemostasis. Although Sondeen et al.11 have recently shown a single reproducible threshold for rebleeding in a pig aortotomy model, their results do not explain why trauma patients (and experimental animals) do not invariably rebleed when their blood pressure returns to normal.

The fundamental difference between early and late fluid administration has implications for the interpretation of experimental 1,3,7,9,12,13,29–39 and clinical 40 studies of fluid resuscitation, because these studies do not effectively control for the timing of fluid administration with respect to intrinsic hemostasis. Not surprisingly, this leads to misinterpretation of experimental results. For example, Bruscagin et al.21 have shown that crystalloid resuscitation does not increase bleeding in an animal model of vascular injury, whereas others 4,7,29 have shown that it does. This contradiction can easily be explained by differences in the timing of crystalloid administration with respect to intrinsic hemostasis, and emphasizes the difficulty in comparing studies if this crucial aspect is disregarded.

Little is currently known about the dynamics of intrinsic hemostasis and rebleeding.11,41 Despite the pioneering work of Shaftan and Harris 19,20 who showed that intrinsic hemostasis after a vascular injury is by means of an extraluminal clot that can be breached by fluid resuscitation or inotropes, the dynamics of intrinsic hemostasis remain ill-defined. This paucity of data stands in sharp contrast to the abundance of animal data on fluid resuscitation.1,2 As a result, we had to use very crude empirical representations of these complex biological processes, guided only by the shape of the blood pressure curve during bleeding from a major vascular injury as seen in animal models. Clearly, better understanding of intrinsic hemostasis and rebleeding in the trauma patient is sorely needed.

Despite the limitations of a theoretical model when compared with a biological one, the modeling perspective has a unique advantage: It forces us to face the implications of our assumptions. While our model still requires external validation by comparing its predictions to independently obtained experimental results, it nevertheless sheds light on the dynamic interactions of hemorrhage, intrinsic hemostasis, fluid resuscitation, and rebleeding. It also emphasizes the need to explicitly incorporate these dynamics into animal and clinical studies of fluid resuscitation in hemorrhagic shock.

APPENDIX: MODEL EQUATIONS AND PARAMETERS

Clot_age_monitor (t) = Clot_age_monitor (t - dt) + (clot_age) * dt

INIT Clot_age_monitor = 0

Clot_age = IF (Flow_dependent_clotting = 0) THEN 1 ELSE 0

Crystalloids_retained (t) = Crystalloids_retained (t - dt) + (Crystalloid_infusion) * dt

INIT Crystalloids_retained = 0

Crystalloid_infusion = Retained_crystalloids

Crystalloids_volume (t) = Crystalloids_volume (t - dt) + (Crystalloid_inflow - Crystalloid_equilibration_rate) * dt

INIT Crystalloids_volume = 0

Crystalloid_inflow = IF (TIME > Time_begin_infusion)AND (TIME < = Time_begin_infusion + Infusion_duration) THEN (Infusion_rate) ELSE 0

Crystalloid_equilibration_rate = (Crystalloids_volume * 0.25 * 0.04) + (Crystalloid_inflow * 0.75)

Intravascular_volume (t) = Intravascular_volume (t - dt) + (Retained_crystalloids + Transcapillary_refill_rate -Bleeding - Secondary_bleeding) * dt

INIT Intravascular_volume = 5.0

Retained_crystalloids = Crystalloid_inflow - Crystalloid_equilibration_rate

Transcapillary_refill_rate = IF (Flow_dependent_clotting = 0) THEN Refill_rate * (1 -(Transcapillary_filling_volume/(0.75))) ELSE 0

Bleeding = Bleeding_rate * Flow_dependent_clotting *Time_dependent_clottting

Secondary_bleeding = IF (Rebleed_signal > 0)THEN (Bleeding_rate) ELSE 0

MAP_monitor (t) = MAP_monitor (t - dt) + (MAP_accumulator) * dt

INIT MAP_monitor = 0

MAP_accumulator = IF (MAP < 60)THEN PULSE (100) ELSE 0

Original_blood_volume (t) = Original_blood_volume (t - dt) + (-Original_volume_loss_rate) * dt

INIT Original_blood_volume = 5

Original_volume_loss_rate = (Bleeding *(Original_blood_volume/Intravascular_volume))

Original_volume_lost (t) = Original_volume_lost (t - dt) + (Original_volume_loss_rate) * dt

INIT Original_volume_lost = 0

Original_volume_loss_rate = (Bleeding *(Original_blood_volume/Intravascular_volume))

Plug_timer (t) = Plug_timer (t - dt) + (Plug_counter) * dt

INIT Plug_timer = 0

Plug_counter = IF (MAP_monitor > 0) AND(Bleeding > 0) THEN (1) ELSE (0)

Primary_blood_loss (t) = Primary_blood_loss (t - dt) + (Bleeding) * dt

INIT Primary_blood_loss = 0

Bleeding = Bleeding_rate * Flow_dependent_clotting *Time_dependent_clotting

Rebleed_signal (t) = Rebleed_signal (t - dt) + (go - stop) * dt

INIT Rebleed_signal = 0

go = Popping_signal

Stop = IF (MAP < (Hemostatic_threshold + 25))THEN (Rebleed_signal/DT) ELSE (0)

Secondary_blood_loss (t) = Secondary_blood_loss (t - dt) + (Secondary_bleeding) * dt

INIT Secondary_blood_loss = 0

Secondary_bleeding = IF (Rebleed_signal > 0)THEN (Bleeding_rate) ELSE 0

Transcapillary_filling_volume (t) = Transcapillary_filling_volume (t - dt) + (Transcapillary_filling_counter) * dt

INIT transcapillary_filling_volume = 0

transcapillary_filling_counter = Transcapillary_refill_rate

Bleeding_rate = IF ((MAP - Hemostatic_threshold > 1))THEN (MAP/R) ELSE ((MAP - Hemostatic_threshold)/R)

Flow_dependent_clotting = IF (TIME < 3) OR(Primary_blood_loss - DELAY (Primary_blood_loss, 1)) >0.005 THEN 1 ELSE 0

Infusion_duration = 20

Infusion_rate = Infusion_volume/Infusion_duration

Infusion_volume = 2

Initial_bleeding_rate = 0.4

Percent_original_volume_loss = IF (Total_blood_loss = 0)THEN 0.001 ELSE(Original_volume_lost/Total_blood_loss) * 100

Popping_signal = IF (Flow_dependent_clotting = 0) AND(MAP > Rebleeding_threshold)THEN (PULSE(1, 0)) ELSE 0

R = 90/Initial_bleeding_rate

Refill_rate = 0.00010 * 5 * (90 - (MAP))

Time_begin_infusion = 15

Total_blood_loss = Primary_blood_loss + Secondary_blood_loss

Hemostatic_threshold = GRAPH (Initial_bleeding_rate)

(0.00, 35.0), (0.1, 25.0), (0.2, 22.5), (0.3, 21.3), (0.4, 20.5),(0.5, 20.3), (0.6, 20.1), (0.7, 20.1), (0.8, 20.0),(0.9, 20.0), (1, 20.0)

MAP = GRAPH (Intravascular_volume)

(2.40, 0.00), (2.60, 15.0), (2.80, 24.0), (3.00, 33.0),(3.20, 40.0), (3.40, 48.0), (3.60, 60.0), (3.80, 75.0),(4.00, 84.0), (4.20, 86.0), (4.40, 88.0),(4.60, 90.0), (4.80, 90.0), (5.00, 90.0)

Rebleeding_threshold = GRAPH (Clot_age_monitor)

(0.00, 50.0), (15.0, 62.5), (30.0, 75.0), (45.0, 87.5), (60.0, 100)

Time_dependent_clottting = GRAPH (Plug_timer)

(0.00, 1.00), (1.00, 0.9), (2.00, 0.8), (3.00, 0.7),(4.00, 0.6), (5.00, 0.495), (6.00, 0.4), (7.00, 0.3),(8.00, 0.2), (9.00, 0.095), (10.0, 0.00) [Context Link]

REFERENCES

1. Mapstone J, Roberts I, Evans P. Fluid resuscitation strategies: a systematic review of animal trials. J Trauma. 2003;55:571–589. Ovid Full Text Bibliographic Links [Context Link]

2. Stern SA. Low-volume fluid resuscitation for presumed hemorrhagic shock: helpful or harmful? Curr Opin Crit Care. 2001;7:422–430. Ovid Full Text Bibliographic Links [Context Link]

3. Bickell WH, Bruttig SP, Millnamow GA, O’Benar J, Wade CE. The detrimental effects of intravenous crystalloid after aortotomy in swine. Surgery. 1991;110:529–536. Bibliographic Links [Context Link]

4. Owens TM, Watson WC, Prough DS, Uchida T, Kramer GC. Limiting initial resuscitation of uncontrolled hemorrhage reduces internal bleeding and subsequent volume requirements. J Trauma. 1995;39:200–207. Ovid Full Text Bibliographic Links [Context Link]

5. Stern SA, Dronen SC, Birrer P, Wang X. Effect of blood pressure on hemorrhage volume and survival in a near-fatal hemorrhage model incorporating a vascular injury. Ann Emerg Med. 1993;22:155–163. Bibliographic Links [Context Link]

6. Krausz MM, Bashenko Y, Hirsh M. Crystalloid or colloid resuscitation of uncontrolled hemorrhagic shock after moderate splenic injury. Shock. 2000;13:230–235. Bibliographic Links [Context Link]

7. Matsuoka T, Hildreth J, Wisner DH. Uncontrolled hemorrhage from parenchymal injury: is resuscitation helpful? J Trauma. 1996;40:915–922. Ovid Full Text Bibliographic Links [Context Link]

8. Riddez L, Johnson L, Hahn RG. Early hemodynamic changes during uncontrolled intra-abdominal bleeding. Eur Surg Res. 1999;31:19–25. Bibliographic Links [Context Link]

9. Burris D, Rhee P, Kaufmann C, et al. Controlled resuscitation for uncontrolled hemorrhagic shock. J Trauma. 1999;46:216–223. Ovid Full Text Bibliographic Links [Context Link]

10. Holmes JF, Sakles JC, Lewis G, Wisner DH. Effects of delaying fluid resuscitation on an injury to the systemic arterial vasculature. Acad Emerg Med. 2002;9:267–274. Bibliographic Links [Context Link]

11. Sondeen JL, Coppes VG, Holcomb JB. Blood pressure at which rebleeding occurs after resuscitation in swine with aortic injury. J Trauma. 2003;54:S110–S117. Ovid Full Text Bibliographic Links [Context Link]

12. Capone AC, Safar P, Stezoski W, Tisherman S, Peitzman AB. Improved outcome with fluid restriction in treatment of uncontrolled hemorrhagic shock. J Am Coll Surg. 1995;180:49–56. Bibliographic Links [Context Link]

13. Krausz MM, Bar-Ziv M, Rabinovici R, Gross D. “Scoop and run” or stabilize hemorrhagic shock with normal saline or small-volume hypertonic saline? J Trauma. 1992;33:6–10. Bibliographic Links [Context Link]

14. Hannon B, Ruth M. Dynamic Modeling. New York: Springer-Verlag; 2001:3–23. [Context Link]

15. Lewis FR Jr. Prehospital intravenous fluid therapy: physiologic computer modelling. J Trauma. 1986;26:804–811. Bibliographic Links [Context Link]

16. Barnea O, Sheffer N. A computer model for analysis of fluid resuscitation. Comput Biol Med. 1993;23:443–454. Bibliographic Links [Context Link]

17. Guyton AC, Hall JE. Textbook of Medical Physiology, 10th edition. Philadelphia: W.B. Saunders Company; 2000:254–255. [Context Link]

18. Bickell WH, Bruttig SP, Wade CE. Hemodynamic response to abdominal aortotomy in the anesthetized swine. Circ Shock. 1989;28:321–332. Bibliographic Links [Context Link]

19. Shaftan GW, Chiu CJ, Grosz CS, Dennis C. The effect of transfusion and of certain hemodynamic factors on the spontaneous control of arterial hemorrhage. J Cardiovasc Surg (Torino). 1964;13:251–256. Bibliographic Links [Context Link]

20. Shaftan GW, Chiu CJ, Dennis C, Harris B. Fundamentals of physiologic control of arterial hemorrhage. Surgery. 1965;58:851–856. Bibliographic Links [Context Link]

21. Bruscagin V, de Figueiredo LF, Rasslan S, Varicoda EY, Rocha e Silva M. Fluid resuscitation improves hemodynamics without increased bleeding in a model of uncontrolled hemorrhage induced by an iliac artery tear in dogs. J Trauma. 2002;52:1147–1152. Ovid Full Text Bibliographic Links [Context Link]

22. Simpson SH, Menezes G, Mardel SN, Kelly S, White R, Beattie T. A computer model of major haemorrhage and resuscitation. Med Eng Phys. 1996;18:339–343. Bibliographic Links [Context Link]

23. Cervera AL, Moss G. Crystalloid distribution following hemorrhage and hemodilution: mathematical model and prediction of optimum volumes for equilibration at normovolemia. J Trauma. 1974;14:506–520. Bibliographic Links [Context Link]

24. Mardel SN, Simpson SH, Kelly S, Wytch R, Beattie TF, Menezes G. Validation of a computer model of haemorrhage and transcapillary refill. Med Eng Phys. 1995;17:215–218. Bibliographic Links [Context Link]

25. Lewis FR. Pressure-driven hemorrhage does not model uncontrolled hemorrhage. Shock. 1996;6:307. Bibliographic Links [Context Link]

26. Gyenge CC, Bowen BD, Reed RK, Bert JL. Preliminary model of fluid and solute distribution and transport during hemorrhage. Ann Biomed Eng. 2003;31:823–839. Bibliographic Links [Context Link]

27. Pirkle JC Jr, Gann DS. Restitution of blood volume after hemorrhage: mathematical description. Am J Physiol. 1975;228: 821–827. Bibliographic Links [Context Link]

28. Sheffer N, Hirshberg A, Barnea O. Myocardial O2 balance during fluid resuscitation in uncontrolled hemorrhage: computer model. J Trauma. 1997;42:647–651. Ovid Full Text Bibliographic Links [Context Link]

29. Abu-Hatoum O, Bashenko Y, Hirsh M, Krausz MM. Continuous fluid resuscitation and splenectomy for treatment of uncontrolled hemorrhagic shock after massive splenic injury. J Trauma. 2002;52:253–258. Ovid Full Text Bibliographic Links [Context Link]

30. Doucet JJ, Hall RI. Limited resuscitation with hypertonic saline, hypertonic sodium acetate, and lactated ringer’s solutions in a model of uncontrolled hemorrhage from a vascular injury. J Trauma. 1999;47:956–963. Ovid Full Text Bibliographic Links [Context Link]

31. Dronen SC, Stern SA, Wang X, Stanley M. A comparison of the response of near-fatal acute hemorrhage models with and without a vascular injury to rapid volume expansion. Am J Emerg Med. 1993;11:331–335. Bibliographic Links [Context Link]

32. Gross D, Landau EH, Klin B, Krausz MM. Treatment of uncontrolled hemorrhagic shock with hypertonic saline solution. Surg Gynecol Obstet. 1990;170:106–112. Bibliographic Links [Context Link]

33. Healey MA, Davis RE, Liu FC, Loomis WH, Hoyt DB. Lactated ringer’s is superior to normal saline in a model of massive hemorrhage and resuscitation. J Trauma. 1998;45:894–899. Ovid Full Text Bibliographic Links [Context Link]

34. Kowalenko T, Stern S, Dronen S, Wang X. Improved outcome with hypotensive resuscitation of uncontrolled hemorrhagic shock in a swine model. J Trauma. 1992;33:349–353. Bibliographic Links [Context Link]

35. Riddez L, Drobin D, Sjostrand F, Svensen C, Hahn RG. Lower dose of hypertonic saline dextran reduces the risk of lethal rebleeding in uncontrolled hemorrhage. Shock. 2002;17:377–382. Ovid Full Text Bibliographic Links [Context Link]

36. Solomonov E, Hirsh M, Yahiya A, Krausz MM. The effect of vigorous fluid resuscitation in uncontrolled hemorrhagic shock after massive splenic injury. Crit Care Med. 2000;28:749–754. [Context Link]

37. Stern SA, Dronen SC, Wang X. Multiple resuscitation regimens in a near-fatal porcine aortic injury hemorrhage model. Acad Emerg Med. 1995;2:89–97. Bibliographic Links [Context Link]

38. Stern SA, Wang X, Mertz M, et al. Under-resuscitation of near-lethal uncontrolled hemorrhage: effects on mortality and end-organ function at 72 hours. Shock. 2001;15:16–23. Bibliographic Links [Context Link]

39. Talmor D, Merkind V, Artru AA, et al. Treatments to support blood pressure increases bleeding and/or decreases survival in a rat model of closed head trauma combined with uncontrolled hemorrhage. Anesth Analg. 1999;89:950–956. Ovid Full Text Bibliographic Links [Context Link]

40. Bickell WH, Wall MJ Jr, Pepe PE, et al. Immediate versus delayed fluid resuscitation for hypotensive patients with penetrating torso injuries. N Engl J Med. 1994;331:1105–1109. Ovid Full Text Bibliographic Links [Context Link]

41. Milles G, Koucky CJ, Zacheis HG. Experimental uncontrolled arterial hemorrhage. Surgery. 1966;60:434–442. Bibliographic Links [Context Link]