http://surgerybook.net/

Recently several investigations have raised concerns about the proinflammatory effects of

resuscitation with lactated Ringer solution. Some commercially available lactated Ringer solutions

contain racemic lactate that is made of equal concentrations of D(−)- and L(−)-isomers of lactate. The

D(−)-isomer has been demonstrated in vitro to result in enhanced production of reactive oxygen species

by neutrophils and inflammatory gene expression by leukocytes.81–82 In addition, increased apoptosis in

both the small intestine and the liver was seen after resuscitation from hemorrhage with lactated Ringer

solution but not with hypertonic solutions or blood resuscitation.83 These findings, however, are not

consistent and do not appear dissimilar from saline resuscitation in other animal models.84 Thus, further

investigations are required to determine the overall inflammatory effect of crystalloid resuscitation and

extent of D(−)-isomer use in current commercial fluids.

Colloids

Colloids have the theoretic advantages of increasing the colloid oncotic pressure and requiring smaller

volumes for resuscitation than crystalloids.85 Colloids commonly used for volume expansion in

hypovolemia include albumin, dextran 70, dextran 40, and hydroxyethyl starch. Although each has

unique individual characteristics, currently there is little justification for the routine addition of colloids

to balanced salt solutions for volume replacement during shock. In fact routine use of colloid such as

hydroxyethyl starch may be associated with increased risk of mortality without any benefit.86

Albumin solutions have been used during resuscitation to increase colloid oncotic pressure and,

hypothetically, to protect the lung from interstitial edema; however, there is a relatively rapid flux of

albumin across the pulmonary capillary membranes and relatively rapid clearance through the

pulmonary lymphatics. In fact, colloid albumin infusion has been demonstrated to prolong the

resuscitation phase and delay postresuscitation diuresis. Additionally, albumin may serve to depress

circulating immunoglobulin levels and suppress albumin synthesis.

Dextran 40 and dextran 70 are polysaccharides with molecular weights of 40 and 70 kD, respectively.

Dextran 40 (10%) is hyperoncotic and initially exerts a volume-expanding effect. However, because of

its lower molecular weight, it is more rapidly excreted. Thus, dextran 40 is commonly used in cases of

peripheral vascular disease and hyperviscosity syndromes. Dextran 70, conversely, is provided as a 6%

solution and does not exert a hyperoncotic effect. The volume expansion is somewhat greater than the

amount infused, and because of its large molecular size the effect is maintained for up to 48 hours. The

dextran preparations, however, cause decreased platelet adhesiveness and decreased factor VIII activity.

They also carry an incidence of allergic reaction of up to 5% and anaphylaxis of 0.6%.85,87

Hydroxyethyl starch is an amylopectin with volume-expanding effects for approximately 36 hours. It

has side effects similar to those of dextran, but with less frequency. The incidence of anaphylaxis is

0.006%. A new hydroxyethyl starch, pentastarch, has a lower molecular weight and fewer hydroxyethyl

groups than hydroxyethyl starch. Pentastarch has a shorter duration of action (2.5 hours) and has been

reported to have even fewer side effects.87–88

The controversy regarding use of crystalloids versus colloids in resuscitation has not been resolved.

Both types of solutions can restore circulating volume. The effects of the solutions on pulmonary

function are at issue and are summarized as follows: (a) the use of crystalloid solutions decreases

plasma oncotic pressure, thereby leading to lung edema at lower microvascular pressures; and (b)

colloids given in the face of pulmonary injury can extravasate, promoting edema because of the reduced

plasma interstitial oncotic gradient. In fact, a previous meta-analysis of colloid versus crystalloid

resuscitation after hemorrhagic shock demonstrated a higher mortality rate among the colloidresuscitated patients, partly because of pulmonary complications.89 Therefore, since colloid infusion has

not demonstrated a significant benefit over crystalloid resuscitation alone, it is not currently

recommended in the management of hypovolemic shock.90–91

Resuscitative Strategy

7 Aggressive fluid resuscitation is clearly a lifesaving modality and a key strategy in the treatment of

shock and prevention of secondary consequences (Algorithm 9-2). However, indiscriminate fluid loading

causes problematic edema in the lungs, gut, brain, and other organs. The amount of fluid used for

resuscitation should be titrated to carefully selected hemodynamic and oxygen transport endpoints.

Solutions currently in development (“artificial blood”) with oxygen transport capabilities may hold the

potential of restoring oxygen transport while minimizing the need for large volumes.

Permissive Hypotension

282

http://surgerybook.net/

Elevation of systemic arterial pressure in patients with disruption of the arterial system or major solid

organ injury, especially after penetrating trauma, may cause acceleration of hemorrhage, disrupt natural

clotting mechanisms, and cause dilution of clotting factors. Laboratory and clinical evidence support

judicious use of intravenous fluids until hemorrhage is controlled by surgery, angiography, or direct

pressure for penetrating trauma.92 Fluid resuscitation for patients with multisystem blunt trauma,

especially with concomitant traumatic brain injury, represents a more complicated decision process, as

maintaining cerebral perfusion pressure is a competing priority. Avoidance of both excessive fluid

administration and prolonged hypoperfusion is best achieved in all patients not by maintenance of a

marginal blood pressure, but by rapid surgical or angiographic intervention to control bleeding.

Transfusion

Anemia prompts clinical concerns because it may signify blood loss or hematologic disease, but it rarely

causes tissue ischemia. The hemoglobin level that causes concern should depend on the adequacy of

other mechanisms involved in oxygen delivery such as arterial oxygen saturation and cardiac output, the

specific clinical situation, and the organ systems most at risk, balanced against the risk of transfusion.

Clinical evidence suggests that hemoglobin values above 7 mg/dL are adequate in most patients,

including the critically ill, but this has not been explored during shock. In one prospective randomized

trial in critically ill patients, it was clearly demonstrated that a reduction in complications and

improvement in survival were noted when lower hemoglobin values were accepted.93 However, this

study excluded patients with hypovolemia, acute coronary syndrome, and sepsis. Although the role of

transfusion during shock remains problematic, it appears in patients without shock these thresholds of

transfusion are even acceptable for elderly patients with significant cardiac disease.94

Table 9-6 Comparison of Blood Availability

Given this limitation, currently it is held that most patients with class I or II shock can be resuscitated

with balanced salt solutions alone. Patients who lose more than 25% to 30% of total blood volume

require blood for resuscitation, as do patients with persistent evidence of inadequate end-organ

perfusion.20 The decision about the extent of blood cross match prior to being transfused is determined

in part by the urgency of the situation. Blood that has been fully typed and cross matched carries the

least risk of transfusion reactions, but it also takes the most time to obtain. Other transfusion options

include the use of type O or type-specific blood (Table 9-6).

Type O Blood

Type O (universal donor) blood is immediately available without a cross match. Because type O blood

contains no AB cellular antigens, administration of packed red blood cells is relatively safe in patients

with any blood type. Males should be transfused type O Rh-positive blood, while prepubescent females

and females of childbearing age should be given type O Rh-negative blood to avoid sensitization that

would complicate future pregnancies. The administration of more than 4 units of type O blood to a non–

O-blood-type patient, however, theoretically can result in an admixture of blood type. A pretransfusion

blood specimen should be sent to the blood bank when the patient is admitted, and type-specific blood

should be transfused as soon as it is available.

Type-Specific Blood

Type-specific blood is available from most blood banks within 5 to 10 minutes of receipt of the blood

specimen, while the patient is being resuscitated with balanced salt solutions. Although not cross

matched, this blood can be administered safely, as demonstrated in both military and civilian

experiences.95

Autotransfusion

Autotransfusion involves collection of the shed blood and its reinfusion through a filter back into the

patient. Autotransfusion can be as simple as aspiration of the blood into a citrate-containing collection

283

http://surgerybook.net/

chamber, followed by reinfusion through a 40-mm filter. A more elaborate system, the Haemonetics

autotransfuser (Haemonetics Corp., Braintree, MA), centrifuges the collected blood and delivers

washed, packed red blood cells for reinfusion. The advantages of autotransfusion include transfusion

with warm, compatible blood without delays and with no risk of transmission of hepatitis, human

immunodeficiency virus, or other bloodborne pathogens.

Autotransfused blood can produce disseminated intravascular coagulation and activation of

fibrinolysis. In addition, collection of blood from the peritoneal cavity after hollow viscus injury, even

with cell washing, may lead to bacterial contamination of the autotransfused blood.96 Successful

autotransfusion of contaminated blood has been demonstrated, but blood obtained from entericcontaminated cavities probably should not be used, except perhaps in extreme circumstances.97 Despite

the potential benefits, investigators have found that the autotransfuser was used in only 26% of the

trauma patients for whom it was prepared.98 Currently, no evidence exists that autotransfusion

improves outcome compared to exclusive homologous blood transfusions in trauma patients.

Endpoints of Resuscitation

Endpoints of resuscitation can be categorized as either global or regional indicators of perfusion. Blood

pressure and pulse are global measures and are relatively poor determinants of the adequacy of tissue

oxygenation. They must also be interpreted in the context of patient age and pre-existing medical

conditions. Tachycardia is a component of SIRS and does not always resolve with increased preload.

Arterial pressure is maintained by myriad compensatory mechanisms, even in the face of a significant

volume deficit, and interpretation is complicated by highly variable baseline pressures, age, and preexisting medical conditions.

Base deficit and serum lactate are also global indicators of perfusion and may help in the detection of

patients who are in otherwise compensated shock. Acidosis arising from regional tissues may not be

apparent in peripheral blood samples, as is frequently the case in patients with intestinal ischemia, in

whom systemic acidosis is a late finding. An elevated base deficit and lactate can be caused by

electrolyte abnormalities, accelerated glycolysis or pyruvate production, and/or decreased clearance by

the liver. They may also reflect dysfunction caused by a period of hypoperfusion that has already

resolved and that does not need further treatment. A positive response toward correction, however, is

indicative of appropriate resuscitation.

A PAC has obvious appeal as a monitor because ensuring adequate oxygen delivery is paramount in

the treatment of shock (Algorithm 9-2). A progressive decline in systemic oxygen delivery (DO2

) results

in an increase in the oxygen extraction ratio, evident as a reduction in pulmonary mixed venous oxygen

saturation. When DO2

is reduced below the level needed to maintain normal tissue metabolic activity,

anaerobic metabolism occurs. This is evident as a decrease in total-body oxygen consumption (VO2

).

Adequate resuscitation requires eliminating any pathologic decrease in VO2 by restoring oxygen

delivery to an adequate level. In clinical practice, however, there is no precise level of VO2

that can be

used as an endpoint, as tissue oxygen needs vary according to the patient’s condition, level of sedation,

body temperature, and other factors, and are affected by endogenous and exogenously administered

catecholamines.

Oxygen consumption may be especially difficult to interpret in patients with late-stage sepsis because

acquired defects in mitochondrial respiration may prevent utilization of oxygen, resulting in decreased

consumption and progressive acidosis despite normal or high DO2

.

The adequacy of resuscitation can also be assessed by measurement of end-organ function and

perfusion, in addition to global measures. Blood flow to the most vital organs (brain and heart) is

preserved during shock at the expense of flow to the skin, muscles, gut, and, ultimately, kidneys.

Detection of ischemia in less vital organs could theoretically identify patients in compensated shock who

have otherwise normal global indicators.

Low urine output (<0.5 to 1.0 mL/kg/hr) is an indicator of inadequate end-organ perfusion, but

inappropriate urine output may initially be maintained by peripheral venoconstriction and maintenance

of cardiac output due to tachycardia. The use of gastric tonometry to measure intramucosal pH has

highlighted the uneven recovery from shock by visceral organs. Persistent visceral hypoperfusion, as

demonstrated by intramucosal acidosis despite correction to normal hemodynamics, is associated with

organ failure and poor outcomes. Unfortunately, direct measurement of visceral hypoperfusion, as well

as hypoperfusion in other regional vascular beds, requires use of technically challenging, labor-intensive

devices that often produce variable unreliable results and has not yet had widespread application.

Presently, the goal of therapy is to restore tissue perfusion, both global and regional as measured by

284

http://surgerybook.net/

organ function, and to normalize cellular metabolism while avoiding excessive use of fluids and

inotropes.

Recently, several biomarkers, in addition to base deficit and lactate, have demonstrated potential

promise. Among the most promising is procalcitonin for the early recognition of sepsis. However, in

addition to being significantly elevated during sepsis, procalcitonin has recently been demonstrated,

similar to lactate and base deficit, to be prognostic of outcome from hypovolemia based on rate of

clearance.99 Thus, this biomarker along with potential others may lead to early recognition and

treatment of shock.

MONITORS

Central Venous Pressure

CVP has been used as a surrogate for venous return to the right heart and extended to represent

preload. Preload is often diminished in the setting of shock and must be optimized in order to maintain

adequate oxygen delivery. The venous system serves a capacitance function containing roughly twothirds of our circulating blood volume. It is a compliant system that, in a healthy state, functions to

maintain adequate venous return to the heart over a variable total blood volume. In states of shock, this

capacitance function may be overcome by too low a circulating volume – as in the setting of

hypovolemic or hemorrhagic shock – or by dysregulation – as in the setting of neurogenic or septic

shock. CVP may be measured directly via an appropriately positioned central venous catheter. The

catheter tip should ideally be positioned at the cavoatral junction – roughly 2 cm caudal to the carina on

chest x-ray.100 Intravenous fluids are administered to achieve adequate preload. Despite challenges for

the ability for CVP to determine preload, CVP still remains used clinically.101 In the surviving sepsis

campaign, CVP is pushed to a goal of 8 to 12 mm Hg, which is thought to optimize preload.20 This

however, has been called into question in recent years with several investigators suggesting that there is

no survival benefit to achieving such high CVPs.20 It has certainly been demonstrated that CVP is not a

perfect measure of venous return to the right heart or cardiac preload as there are several intrinsic

cardiopulmonary factors that influence both CVP and cardiac output. The role of CVP may be most

useful at its extremes and when used within the context of the patients’ clinical situations. A CVP <5

mm Hg in the setting of a young, otherwise healthy trauma patient with ongoing hemorrhage is likely

consistent with hypovolemic shock and warrants volume resuscitation. Alternatively, a CVP of 15 mm

Hg in an elderly patient with known congestive heart failure does not alone provide adequate

information as to a patient’s volume status nor should it guide resuscitation. As such, CVP is one

measure that can help to guide shock resuscitation when evaluated in the proper context.

Pulmonary Artery Catheter

Hemodynamic monitoring utilizing a flow-directed PAC has been a standard in the treatment of shock

for decades and has been considered to be essential for optimal management of certain forms of shock.

However, as a result of a multicenter study that reported that use of a PAC was associated with

increased mortality, the indications for PAC use have recently been questioned.102 PAC patients were

retrospectively compared to matched control patients who were selected using a scoring system. This

scoring system has not been validated, which raises the possibility that physicians inserted a PAC in the

more critically ill patients.

Intensive care unit (ICU) staffing practices may have also been a factor explaining these surprising

results. The study was conducted in “open” ICUs, where any physician on the medical staff could admit

a patient to the unit and insert a PAC. Clinicians may not always interpret PAC data appropriately. In a

multicenter study of physicians’ knowledge and interpretation of PAC data, nearly half (47%) were

unable to appropriately determine the wedge pressure from a clear tracing, and a similar percentage

(44%) could not identify the determinants of oxygen delivery.103 Studies have demonstrated improved

outcome when patients are managed in ICUs staffed by specialists in critical care, despite their more

frequent use of PACs.104

An additional consideration regarding the controversy over the utility of PACs is that currently used

endpoints for PAC-guided resuscitation may be inappropriate. For example, efforts to augment systemic

oxygen delivery have not demonstrated any benefits and, in fact, harm may be caused by this approach,

whereas treatments used to alter the variables ascertained with the catheter may themselves cause harm

(fluid overload, excessive inotropes, blood transfusions). Finally, PACs are associated with specific

285