Chapter 80. Hemodynamic Support of the Critically Ill Patient

1. Early and adequate resuscitation of patients with acute circulatory failure is important to restore the balance between oxygen needs and delivery. Effective resuscitation can result in improved outcomes.

2. Fluid resuscitation should be guided by repeated fluid challenges.

3. If fluid administration is insufficient to restore an adequate tissue perfusion pressure, vasopressors may be required; norepinephrine is currently considered the best first-line choice.

4. Inotropes or vasodilator drugs may be needed to improve myocardial contractility and cardiac output. Dobutamine remains the inotropic agent of choice.

5. Hemodynamic support should be titrated to the individual patient according to global parameters of hemodynamic and oxygenation status, supported by regional parameters when available.

The most common cause of organ failure in the critically ill patient is inadequate tissue perfusion related to acute circulatory failure. This may result from persistent fluid deficits and/or alterations in regional blood flow or tissue oxygen utilization. Whatever the cause, early and adequate hemodynamic support of these patients is crucial if organ function is to be preserved and multiple organ failure, a common cause of death in critically ill patients, prevented. In this chapter, we briefly review the main causes and symptoms of acute circulatory failure before focusing on the hemodynamic support of such patients.

Clinical Signs of Shock

Circulatory shock can be considered as a state of generalized circulatory failure resulting in tissue hypoxia. It is a major cause of organ failure. A diagnosis of shock can be based on a combination of various clinical, hemodynamic, and biochemical signs, which can broadly be summarized as follows:

1. Arterial hypotension: Hypotension is perhaps the hallmark of acute circulatory failure, but may be only moderate, especially in patients with chronic hypertension. Usually the systolic arterial pressure is less than 90 mm Hg or the mean arterial pressure is less than 70 mm Hg.

2. Signs of tissue hypoperfusion: These are usually recognized at 3 levels: (a) cutaneous: the skin is usually vasoconstricted, cold, and clammy; (b) renal: a reduction in renal perfusion is manifested in adults by a fall in urine output below 0.5 mL/kg/h and, in more severe cases, below 20 mL/kg/h; (c) neurologic: This can of course be appreciated only in the unanesthetized, unsedated patient. Decreased cerebral perfusion is demonstrated by an altered intellect, with disorientation and confusion, and lack of collaboration; there is often obtundation, but coma develops only in advanced stages of multiple organ failure.

3. Biologic signs of altered cellular oxygen availability: The development of anaerobic metabolism is manifest by the development of hyperlactatemia. The normal blood lactate level is around 1 mEq/L (or 1 mMol/L) but is usually increased above 1.5 mEq/L in acute circulatory failure.

Pathophysiologic Classification of Shock

Shock can essentially be classified according to 4 pathophysiologic mechanisms1: hypovolemic, cardiogenic, obstructive, or distributive. Many patients with acute circulatory failure have a combination of 2 or more of the 4 mechanisms. For example, a patient with severe pancreatitis may have characteristics of hypovolemic, distributive, and even cardiogenic shock. Similarly, patients with severe sepsis or anaphylaxis often have elements of hypovolemic and cardiogenic shock in addition to the baseline distributive mechanism.

Whatever the specific type of acute circulatory failure, the result is an imbalance between oxygen requirements and tissue oxygen availability, resulting in regional tissue hypoxia, an important contributor to the development of organ dysfunction and multiple organ failure. In hypovolemic, cardiogenic, and obstructive types of shock, the primary abnormality is the reduced cardiac output and hence inadequate oxygen transport. However, in septic shock, the main fault is increased oxygen requirements, due to the inflammatory response, in addition to altered oxygen extraction capabilities and myocardial contractility. Hence although cardiac output may be normal or even increased in such patients, it may still be inadequate to provide sufficient oxygen for the cells' increased requirements.

Hypovolemic Shock

This form of shock occurs when the intravascular volume is depleted as a result of internal or external fluid loss. Hypovolemic shock is the most common form of circulatory shock. The most obvious cause of hypovolemic shock is acute hemorrhage, but it can also occur as a result of severe dehydration due to severe vomiting or diarrhea, particularly in children and the elderly. The hemodynamic pattern is characteristically one of a low cardiac output (due to reduced venous return), associated with decreased cardiac filling pressures and increased systemic vascular resistance (SVR).

Cardiogenic Shock

This form of shock is characterized by primary failure of the cardiac pump. Cardiogenic shock is most commonly the result of acute myocardial infarction (MI), but other causes include severe cardiac valvular disease, severe myocarditis, end-stage cardiomyopathy, and severe cardiac arrhythmia. Cardiogenic shock is associated with mortality rates in the region of 75%, the highest of the 4 types of shock. Cardiogenic shock is characterized by a low cardiac output, increased cardiac filling pressures, and increased SVR.

Obstructive Shock

This form of shock is the result of an impediment to the normal flow of blood, either due to an obstruction to the outflow of blood from the heart, such as in massive pulmonary embolism, severe aortic coarctation, or severe aortic stenosis, or due to an increased resistance to cardiac filling during diastole, such as in cardiac tamponade or tension pneumothorax. Obstructive shock is typically characterized by a low cardiac output and increased SVR. Right cardiac filling pressures are increased in pulmonary embolism, and right and left pressures are increased in tamponade. Pulmonary arterial hypertension is also present in cardiogenic shock associated with pulmonary embolism.

Distributive Shock

This type of shock is characterized by an increase in the vascular capacity. It is most commonly due to sepsis, secondary to the release of inflammatory mediators. Other causes include anaphylactic shock, neurogenic shock, and acute adrenal insufficiency. The typical hemodynamic pattern is a normal or high cardiac output, reduced SVR (as vascular tone is reduced), and reduced cardiac filling pressures.

Intensive early resuscitation is essential to restore the imbalance between oxygen requirements and supply in the patient with acute circulatory failure, and therapy should be initiated while investigations are ongoing to determine and correct the underlying cause of the shock. Early resuscitation of patients with severe sepsis and septic shock has been shown to be associated with improved outcomes.2

A useful aid to describe the important steps of hemodynamic resuscitation is the VIP (ventilation, infusion, pump) rule introduced by Weil and Shubin in 1969.3

Ventilatory Support

Oxygen should be started immediately with the aim of increasing oxygen delivery and reducing pulmonary vasoconstriction due to hypoxia. Importantly, oxygen should be administered even if the patient is not very hypoxemic. Although prolonged administration of high inspired oxygen fractions (Fio2) can be toxic, this is not a problem in the acute situation, and once blood gas results are available, oxygen therapy can be adjusted accordingly. If mask ventilation is not possible (eg, due to facial trauma) or provides inadequate oxygenation, mechanical ventilation should be started. In addition to ensuring adequate oxygenation, mechanical ventilation has the additional benefit of reducing left ventricular afterload by increasing intrathoracic pressures, and resting the respiratory muscles, hence reducing oxygen requirements. The aim of oxygen administration should be to maintain Pao2 above 8 kPa (60 mm Hg), and Sao2 above 90%. Although hypoxia should be avoided, hyperoxia can also be harmful, causing peripheral vasoconstriction with reduced regional perfusion and oxygenation.

Fluid Resuscitation

Fluid therapy is an essential part of the treatment of any form of shock.4 The rationale is to improve microvascular blood flow by increasing plasma volume, and to increase cardiac output by the Frank-Starling effect. However, too much fluid also carries risks, principally of pulmonary edema.

Which Fluid?

Many fluids are available for use in resuscitation, and which, if any, is optimal remains controversial.5 Crystalloid solutions (eg, normal saline, Ringer lactate) are inexpensive and well-tolerated, but these can have adverse effects when large volumes are infused. Infusion of large amounts of saline solutions typically results in hypernatremia and, notably, in hyperchloremic acidosis associated with a reduction in the strong ion difference.6 Similarly, infusions of large amounts of the so-called balanced solutions (Ringer lactate, Hartmann solution) can influence electrolyte balance. Moreover, crystalloid solutions leak more into the interstitial space than colloid solutions, thus causing more tissue edema. Increased edema is associated with compromised lung function, reduced systemic oxygen availability, and impaired wound healing, myocardial function, and gut function (Fig. 80-1).7,8 As colloids persist longer in the intravascular space, less colloid solution (eg, albumin, gelatin, hydroxyethyl starch [HES]) than crystalloid is needed to achieve the same hemodynamic goal, and colloids, therefore, have theoretical advantages over crystalloids. However, the clinical relevance of this potential advantage has not been clearly demonstrated, with no study convincingly demonstrating that one fluid type is superior to another. In addition, colloid solutions are more expensive, especially human albumin.

Although there has been some controversy surrounding the use of albumin in intensive care unit (ICU) patients,9 the large randomized controlled Saline versus Albumin for Fluid Resuscitation in the Critically Ill (SAFE) study10 showed no differences in mortality rates for ICU patients who received 4% albumin compared with those who received crystalloid solution as the initial resuscitation fluid. However, hypoalbuminemia is known to be associated with increased morbidity,11 and albumin administration may reduce complications in critically ill patients.12 In the SAFE study, there was a slight, but not significant, decrease in mortality in patients with hypoalbuminemia on admission who received albumin compared with those who received saline (23.7% vs 26.2%, odds ratio: 0.87; 95% confidence interval, 0.73-1.05; p = 0.14).13 Clearly further studies assessing the role of albumin, particularly in certain subgroups of patients (eg, those with hypoalbuminemia) are needed to clarify this issue.

The 3 main types of artificial colloid are gelatin solutions, dextran solutions, and HES solutions. Gelatin solutions are not available in the United States but are still used elsewhere, largely because of their low costs. They have limited oncotic effects and relatively short intravascular persistence. These solutions can induce anaphylactic reactions, and adverse renal effects have been reported. Dextran solutions carry a substantial risk of anaphylactic reaction, so hapten prophylaxis must be given simultaneously. Dextrans also have antihemostatic effects, can cause formation of "rouleaux," which may complicate the type and crossmatch in case of blood transfusions, and may precipitate renal failure. HES solutions have been promoted as having particularly beneficial effects on oxygen delivery to the tissues7 and potentially reducing endothelial activation and inflammation in critically ill patients.14 However, HES solutions have been associated with an increased incidence of acute renal failure in critically ill patients, particularly in those with severe sepsis.15,16

Further study is necessary to clearly define optimal fluid choices because all available fluids have advantages and disadvantages (Table 80-1). Until the results of such studies become available, the choice is best made according to the severity of the circulatory failure, the underlying disease, the type of fluid that has been lost, the serum albumin concentration of the patient, and the risk of bleeding. Using a combination of several fluids rather than excessive amounts of any one will help limit adverse effects.

How Much Fluid?

Having selected the fluid, the physician must then decide how much to give. Precise end points for fluid resuscitation are difficult to define because sensitive tools for monitoring the regional microcirculation and oxygenation are not yet available, and, although systemic parameters may appear to have stabilized, regional tissue perfusion may still be inadequate.17 In addition, the quantity of fluid needed varies among patients and in the same patient over time. A fluid challenge technique18 is the best method of determining a patient's ongoing need for fluids. An example of the fluid challenge technique is provided in Fig. 80-2. The fluid challenge approach incorporates 4 phases18:

• The type of fluid: As previously discussed, the optimal fluid remains controversial and either crystalloid or colloid can be used; the selection is determined on an individual patient basis.
• The rate of fluid administration: It is important to define the amount of fluid to be administered over a defined interval. The Surviving Sepsis Campaign Guidelines for the management of severe sepsis and septic shock recommend at least 1000 mL of crystalloids or 300 to 500 mL of colloids over 30 minutes.19
• The goal to be achieved: The primary defect(s) that prompts the fluid challenge should be identified and quantitated so that a goal can be determined; most commonly this will be restoration of an adequate mean arterial pressure.
• Safety limits: Pulmonary edema due to congestive heart failure is the most serious complication of fluid infusion. A safety limit, generally the central venous pressure, must be set to avoid this complication.

Repeated fluid challenges will enable the physician to continuously reassess a patient's ongoing fluid needs and limit the risks of adverse effects.

Blood Transfusions

Blood transfusions should be given as necessary, although there is again some debate as to what triggers should be used to determine the need for a blood transfusion. The aim of transfusion is to improve oxygen delivery and thereby to limit tissue hypoxia. However, although oxygen delivery is improved, tissue oxygenation or oxygen use do not necessarily increase.20-22 In addition, risks are associated with blood transfusions including transmission of microorganisms and prions, transfusion-related acute lung injury (TRALI), transfusion-related immunomodulation, which may increase the risk of infections, and administration errors, including wrong type and crossmatch and incorrect patient identification, which can cause hemolytic reactions.

In 1999, Hebert et al23 published the results of the Transfusion Requirements in Critical Care trial, which encouraged many to rethink their transfusion practice. This landmark study enrolled 838 critically ill patients with euvolemia who had hemoglobin concentrations of less than 9.0 g/dL and randomly assigned them to a restrictive strategy of transfusion, in which red cells were transfused if the hemoglobin concentration decreased below 7.0 g/dL, or a liberal strategy, in which transfusions were given when the hemoglobin concentration fell below 10.0 g/dL. The results suggested that the restrictive strategy, maintaining a hemoglobin of 7 to 9 g/dL, was adequate for most critically ill patients,23 with the possible exception of patients with acute MI and unstable angina.24

Since that study, several epidemiologic studies have shown that patients who receive blood transfusions in the ICU have increased mortality rates.25,26 In addition, studies in human volunteers have shown that isovolemic hemodilution to a hemoglobin of 5 g/dL or less does not result in biochemical evidence of anaerobic metabolism27 and studies in Jehovah's Witness patients have shown that survival is possible with markedly decreased hemoglobin concentrations; one case study reported survival of a patient with a hemoglobin concentration of only 1.8 g/dL.28

Current recommendations for the management of patients with severe sepsis support a transfusion trigger of 7 g/dL.19 However, the Sepsis Occurrence in Acutely ill Patients (SOAP) study, which evaluated data on 3147 patients in 198 ICUs across Europe in May 2002, reported that, unlike earlier studies, blood transfusion was not associated with an increased mortality in multivariate analysis or by propensity case matching.29 In addition, the study by Rivers et al2 on early goal-directed therapy (EGDT), showed that patients managed according to the EGDT protocol, who had lower mortality rates than patients receiving standard therapy, received more transfusions, suggesting that at least some patients may benefit from receiving more blood transfusions.

The pendulum would therefore seem to be swinging in favor of blood transfusions once again, and this apparent change may be related to the widespread introduction of leukoreduction in recent years. This technique is a process in which the white cells in blood units are intentionally reduced in number using centrifugation or filtration, with the aim of reducing some of the inflammatory response to transfusion and the transmission of infections. Leukocyte counts can be reduced by more than 99%, and the technique is effective in reducing the transmission of cell-associated viruses, especially cytomegalovirus, herpes viruses, and Epstein-Barr virus.30 It may also reduce parasite and prion transmission, transfusion-related febrile reactions, and TRALI. In a before-after cohort study of 14786 patients who received red blood cell transfusions following cardiac surgery or repair of hip fracture, or who required intensive care following a surgical intervention or multiple trauma, transfusion of leukoreduced blood was associated with fewer febrile reactions and reduced posttransfusion antibiotic use.31 However, the evidence supporting the benefits of leukoreduction is not yet completely clear cut. In a meta-analysis of 14 randomized controlled trials comparing standard blood with leukoreduced or autologous blood, Vamvakas32 reported no consistent effect of leukoreduction on long-term mortality, whereas in another meta-analysis of 10 randomized controlled trials, the authors concluded that "patients who were transfused leukoreduced red blood cells might benefit from a decrease in post-operative infections."33 More recently, in a meta-analysis restricted to patients who received transfusions, Blumberg et al reported that leukoreduced blood significantly reduced the odds of postoperative infection by about 50%.34 Many countries have now adopted leukoreduction as routine, although leukoreduced blood is more expensive, and it is not clear whether it is necessary in all patients.35 (See Chapter 83 for a detailed discussion of blood transfusion.)

Vasopressors

If hypotension persists despite fluid administration, the use of vasopressors will be required. In severe conditions, a vasopressor could be administered early in combination with fluids, but it should be discontinued as soon as the hypovolemia has been corrected.

Adrenergic agonists represent the first type of drug to be considered because of their potency and their short half-life, which allows easy titration. A range of drugs is available and can be considered for this purpose (Fig. 80-3). The specific effects of the various agonists are largely determined by the extent to which they act on α, β, and dopaminergic receptors (Table 80-2), although the precise action of any catecholamine varies among individuals. In addition, chronic sympathetic stimulation and the presence of inflammatory mediators, such as those that occur in sepsis, can reduce receptor response to stimulation.

There are 2 forms of α receptor: α1, found primarily on the smooth muscle cells of arterioles and veins, and α2, found at the presynaptic terminals of adrenergic nerves Stimulation of α1 receptors usually causes smooth muscle contraction with vasoconstriction. Stimulation of α2 receptors decreases the subsequent release of adrenergic transmitter.

There are 3 major forms of β receptor: β1 found in heart muscle and in the kidney, β2, found in smooth muscle and metabolic tissue, and β3, found in adipose tissue. Stimulation of β1 receptors results in increased heart rate and contractility, whereas β2 receptor stimulation causes vasodilation in skeletal and cardiac muscle, bronchodilation, and decreased gastrointestinal motility. β3 activation stimulates lipolysis.

Stimulation of each group of receptors therefore has potentially beneficial and potentially harmful effects. For example, β-adrenergic stimulation increases blood flow but can also increase heart rate and carry a risk of ischemia, whereas α-adrenergic stimulation increases blood pressure but may decrease cardiac output and cause peripheral vasoconstriction, with decreased renal and hepatosplanchnic blood flow.

Phenylephrine

Although not a catecholamine, phenylephrine is an almost pure α-adrenergic agent, except at very high doses where some β activity is seen. Phenylephrine is a very powerful vasoconstrictor, but this carries the risk of decreasing blood flow and reducing tissue perfusion. In patients with septic shock, phenylephrine was associated with reduced splanchnic blood flow and oxygen delivery.36 However, a small randomized controlled trial reported no differences in terms of cardiopulmonary performance, global oxygen transport, or regional hemodynamics when phenylephrine was administered instead of norepinephrine in the initial hemodynamic support of septic shock.37 Nevertheless, in the ICU, this drug is currently reserved for the occasional management of severe refractory hypotension.

Epinephrine

Epinephrine is an endogenous catecholamine secreted by the adrenal medulla with potent α and β1 activity and moderate β2 effects. At lower doses, β effects predominate with α effects becoming more significant at higher doses. In acute hypotension, epinephrine is sometimes preferred to norepinephrine due to its stronger β-adrenergic effects that are useful to maintain or increase cardiac output. It is also the drug of choice in cardiac arrest, where it can be administered via the endotracheal tube if intravenous (IV) access is difficult, and in acute anaphylaxis (0.1-0.5 mg subcutaneously). However, epinephrine is associated with a decrease in splanchnic blood flow or gastric intramucosal pH (pHi).38,39 Epinephrine treatment is also associated with an increase in cellular metabolism mediated by the increase in intracellular cyclic adenosine monophosphate (cyclic AMP) and leading to increased blood lactate levels.38,40 A study comparing epinephrine with norepinephrine in patients with shock reported no differences in outcomes between the 2 groups, although epinephrine was associated with the development of significant tachycardia, lactic acidosis, and increased insulin requirements for the first 24 hours of the study, which resulted in withdrawal of 13% of the patients from the study.40 Epinephrine is therefore best avoided in critically ill patients except in the specific circumstances mentioned here, but in countries where more expensive vasopressors are not available, epinephrine represents a reasonable alternative.

Norepinephrine

Norepinephrine is an endogenous amine secreted by the adrenal medulla and the terminal endings of postganglionic nerve fibers. Norepinephrine has predominantly α-adrenergic properties, although its weak β-adrenergic effects can help to maintain cardiac output. Norepinephrine administration thus generally results in a clinically significant increase in mean arterial pressure, with little change in heart rate or cardiac output. The normal dose-range is 0.05 to 2 μg/kg/min IV. The increased afterload due to the vasoconstriction caused by norepinephrine can increase myocardial workload, and norepinephrine can precipitate cardiac failure, myocardial ischemia, and pulmonary edema. Although there have been concerns that excessive vasoconstriction with norepinephrine may have negative effects on blood flow, particularly in the hepatosplanchnic and renal circulations, studies have suggested that it can successfully increase blood pressure without causing any deterioration in organ function, particularly in the presence of decreased vascular tone, as in septic shock.39,41

Dopamine

Dopamine is another naturally occurring compound (the precursor of norepinephrine) which also combines α- and β-adrenergic properties, but it has several specific features. First, dopamine also has dopaminergic effects, which predominate at very low doses (<3 mcg/kg/min IV), and may dilate the hepatosplanchnic and renal circulations, thereby selectively increasing flow in these important regions. Second, the adrenergic effects of dopamine vary with the dose; at lower doses (3-10 mcg/kg/min IV), β-adrenergic effects predominate, so that blood flow may increase together with blood pressure. At higher doses, α-adrenergic effects become increasingly powerful, which may be necessary in more severe cases of hypotension. Importantly, these dose ranges are not cutoff values at which one set of receptors are activated at the expense of another, but are ranges in which the effects of one group of receptors predominate over another. Dopamine increases arterial pressure primarily by increasing cardiac index, due to an increase in stroke volume and, to a lesser extent, to increased heart rate, with minimal effects on SVR. However, dopamine also has drawbacks. First, it is a relatively weak agent, so that norepinephrine or epinephrine must often be added to control hypotensive states. Second, although dopamine may increase blood flow more effectively than other vasopressors, it also increases heart rate. A study in patients undergoing coronary artery bypass grafting with cardiopulmonary bypass suggested that dopamine, even at low doses, is associated with an increased risk of developing atrial fibrillation.42 Third, the advantage of the dopaminergic effects may be more theoretical than practical. Thus the routine administration of low dose dopamine to prevent renal failure is not recommended.43 Finally, dopaminergic stimulation may have undesired endocrine effects on the hypothalamopituitary axis, resulting in immunosuppressant effects, primarily by a reduction in prolactin release.44

In a recent randomized trial, there were no overall significant differences in mortality rates between patients in shock who were treated with dopamine or norepinephrine as the first-line vasopressor agent, but the use of dopamine was associated with a greater number of adverse events, notably arrhythmias.45 In subgroup analyses, dopamine administration was associated with increased 28-day mortality rates in patients with cardiogenic shock. It would thus seem reasonable to recommend norepinephrine rather than dopamine as the first-line vasopressor in patients with shock.

Vasopressin and Terlipressin

Vasopressin is an endogenous stress hormone with a wide range of functions including effects on blood osmolality and volume, body temperature, insulin release, corticotropin release, memory, and social behavior; as a vasoconstrictor of vascular smooth muscle, it also has an important role in regulating blood pressure.46,47 Patients with septic shock appear to develop relative vasopressin deficiency, with lower vasopressin levels than patients with cardiogenic shock for the same degree of hypotension.48 This deficiency may be due, in part, to decreased central stores of vasopressin,49 and administration of low-dose vasopressin to restore normal vasopressin levels can result in substantial increases in arterial pressure and reduced requirements for other catecholamines.48,50-55 The 2008 VASST study showed that a combination of low-dose vasopressin and norepinephrine was as safe as norepinephrine alone in the treatment of patients with septic shock.55 Further, the benefit of vasopressin compared with norepinephrine was greater in a subgroup of patients at risk of renal dysfunction, with a 28-day mortality reduction of 43.7% (p = 0.01), but not in those with more advanced renal dysfunction. These results suggest that vasopressin administration should perhaps be administered early in hyperkinetic states of septic shock.56 Interestingly, in a post hoc analysis of this study, patients who received corticosteroids and vasopressin had reduced mortality rates compared with those who received norepinephrine, whereas in patients who did not receive corticosteroids, vasopressin was associated with increased mortality compared with norepinephrine.57

Vasopressin has also been assessed in patients with other forms of vasodilatory shock (eg, after cardiopulmonary bypass). In such patients, administration of low doses of vasopressin increased arterial blood pressure and reduced vasopressor requirements.58,59 Interestingly, Morales et al60 found that giving vasopressin (0.03 U/min IV) prophylactically before cardiopulmonary bypass to avoid vasopressin deficiency resulted in reduced requirements for norepinephrine and a shorter ICU length of stay, again suggesting that early use may be beneficial.

Vasopressin has also been investigated as a potential alternative to epinephrine for use in cardiorespiratory arrest. A randomized controlled study of vasopressin versus epinephrine in in-hospital cardiac arrest reported no advantage of vasopressin over epinephrine,61 although a later, larger study comparing the 2 agonists in out-of-hospital cardiac arrest suggested improved outcomes for patients with asystole who received vasopressin.62 Although a cohort study suggested that a combination of epinephrine and vasopressin may be better, providing the beneficial effects of both drugs but avoiding the harmful effects of excessive doses of either,63 a randomized study in 2894 patients showed that the combination of vasopressin (40 units) and epinephrine (1 mg) was no better than epinephrine alone.64 The risk with vasopressin is a reduction in cardiac output due to excessive vasoconstriction and a selective reduction in hepatosplanchnic blood flow. Studies are in progress to further evaluate the role of vasopressin in various shock states.

Terlipressin is an analog of vasopressin that has been available in Europe for more than 20 years. Terlipressin has a half-life of 6 hours and a duration of action of 2 to 10 hours, whereas the half-life of vasopressin is 6 minutes and its duration of action is 30 to 60 minutes. Terlipressin has been compared with norepinephrine in patients with hyperdynamic septic shock. Both drugs increased mean arterial blood pressure and improved renal function, but terlipressin administration was associated with a decrease in cardiac output.65 The DOBUPRESS study reported that dobutamine administration could reverse terlipressin-induced reductions in cardiac output and SVO2 but high doses of dobutamine were needed; nevertheless, no adverse effects associated with the high doses were reported.66 The 2009 TERLIVAP study, which tested the efficacy and safety of continuous infusions of terlipressin, vasopressin, or norepinephrine in septic shock patients, suggested that terlipressin may be more efficient than vasopressin for shock reversal with no increased incidence of undesirable effects.67 The ongoing TESST-1 (TErlipressin in Septic Shock Trial) is comparing terlipressin at low (2 mcg/kg/min) and ultralow dose (1 mcg/kg/min) to norepinephrine as a first-line agent in septic shock. Terlipressin may also be useful for the treatment of hypotension occurring after induction of anesthesia in patients who have received long-term treatment with renin-angiotensin system inhibitors.68

Restoring and Maintaining Cardiac Output

An adequate cardiac output is essential to ensure sufficient tissue perfusion and oxygen delivery. However, a normal cardiac output is difficult to define, and the adequacy of cardiac output varies among individuals and in the same individual over time. To increase cardiac output, one must remember its 4 determinants (Fig. 80-4).69

• Heart rate: Bradycardia can limit cardiac output, but only when severe, and this is therefore easily recognized at the bedside. The management of bradycardia is beyond the scope of this chapter, but the principles of treatment consist primarily of a pacemaker, possibly with isoproterenol as a temporary measure. Outside these extreme situations, increasing heart rate by changing the rate of an in situ pacemaker usually does not increase cardiac output because stroke volume decreases concurrently. It may even be the opposite: Cardiac output may decrease if cardiac filling is impaired by a too short diastolic time, especially in cases of diastolic dysfunction.
• Preload: Fluid administration should be considered in all cases of inadequate cardiac output, and if uncertain of the need for fluid, a fluid challenge technique must be used as described earlier, in which a given amount of fluids is given rapidly under control of cardiac filling pressures.18
• Contractility: If the increase in cardiac output is limited by impaired contractility, the use of inotropes may be considered. The major risk is increased oxygen demand due to the increased work of the myocardium, which may precipitate myocardial ischemia.
• Afterload: By reducing the factors that oppose ejection of blood by the ventricles, principally using vasodilators, cardiac output may be increased without increasing myocardial oxygen demand. The major limitation of this approach is the risk of decreasing arterial pressure to levels that may compromise tissue perfusion in the peripheral organs as well as in the myocardium, and vasodilators should be avoided in hypotensive states. In addition, all vasodilators may induce some increase in heart rate, especially in the presence of underlying hypovolemia. Such an increase in heart rate should raise the possibility of underlying hypovolemia and suggest the need for a fluid challenge.

Dobutamine

Dobutamine is the first choice of inotropic agent for patients with low cardiac output who have received adequate fluid resuscitation. A decrease in arterial pressure during dobutamine administration should raise the possibility of underlying fluid deficits requiring another fluid challenge trial. With predominant β-adrenergic properties, dobutamine also possesses some α-adrenergic effects that limit the increase in heart rate seen with pure β-adrenergic stimulation using isoproterenol. An initial dose of just a few mcg/kg/min may increase cardiac output significantly. Doses in excess of 20 mcg/kg/ min IV are seldom used because they add little benefit and may lead to excessive tachycardia. Dobutamine has limited effects on arterial pressure, although arterial pressure may increase slightly if heart failure is the primary abnormality. Dobutamine formed part of the EGDT protocol used by Rivers et al in their study of emergency department patients presenting with septic shock.2 Dobutamine was used in addition to fluids and red cell transfusion to increase central venous oxygen saturation (SCVO2) to more than 70%, and this protocol was associated with improved outcomes compared with standard care.2 However, dobutamine should not routinely be used to increase oxygen delivery to supranormal levels because this approach has been associated with worse outcomes.70 Rather it should be titrated on an individual basis to achieve acceptable oxygenation parameters. Interestingly, using orthogonal polarization spectral imaging, DeBacker et al showed that dobutamine improved capillary perfusion in patients with septic shock, independent of its systemic effects,17 suggesting that it may have additional specific effects on regional blood flow.

Dopexamine

Dopexamine hydrochloride is a newer synthetic catecholamine, structurally similar to dopamine. It has marked β2-adrenergic agonist activity, some dopaminergic receptor activity, weak β1-adrenergic activity, and no direct α-adrenergic effects. It also inhibits the neuronal uptake of endogenous catecholamines. Dopexamine's positive inotropic effects combined with its vasodilating effects make it useful in acute exacerbations of chronic heart failure and in heart failure associated with cardiac surgery. However, its use is limited by the development of marked tachycardia, particularly at higher doses. Although it was suggested that because of its dopaminergic effects it may have beneficial effects on renal and splanchnic blood flow, a meta-analysis of 21 randomized controlled studies found no evidence to support the use of dopexamine to improve hepatosplanchnic or renal blood flow in critically ill patients.71

Phosphodiesterase Inhibitors

The phosphodiesterase inhibitors (PDEs) are a group of enzymes that degrade the cyclic nucleotides cAMP and cGMP. PDE inhibitors can thus prolong or enhance the effects of physiologic processes mediated by cAMP or cGMP. PDE3 inhibitors, such as enoximone and milrinone, have inotropic and vasodilating properties. These drugs may be poorly tolerated by patients with arterial hypotension, and their administration can be difficult due to their long half-life. Intermittent administration may be preferable to continuous infusion. The administration of very small doses of PDEIII inhibitors may reinforce the effects of dobutamine.72 Short-term administration of PDEIII inhibitors has been associated with a high incidence of complications, including arrhythmias, especially in patients with ischemic heart disease, likely related to their effects on cAMP and Ca2+ levels. Dobutamine and milrinone provided equally effective inotropic support in patients awaiting cardiac transplantation with no differences in right heart hemodynamics, death, need for additional vasodilator/inotropic therapy, need for mechanical cardiac support before transplantation, or occurrence of ventricular arrhythmias requiring increased antiarrhythmic therapy. In critically ill patients with catecholamine-dependent heart failure, milrinone improved central hemodynamics and was associated with a reduction in dobutamine dose, although there was a tendency for milrinone-treated patients to require higher doses of vasopressors and more fluids.73 Some studies have suggested that milrinone may have additional anti-inflammatory effects and beneficial effects on hepatosplanchnic perfusion.74,75 Further studies are clearly needed to determine the exact place of these drugs in the management of patients with severe heart failure and shock.

Levosimendan

Levosimendan is a relatively new agent that provides inotropic effects by increasing calcium sensitivity of myocytes by binding to cardiac troponin-C, and vasodilator effects by opening ATP-sensitive potassium channels in vascular smooth muscle.76 Levosimendan has a long half-life that may limit the practicality of its use. Levosimendan may be useful in patients with severe heart failure, where it has been shown to improve hemodynamic performance more effectively than dobutamine and to reduce mortality.77,78 It may also be of use for inotropic support after myocardial ischemia, after myocardial stunning, during and after cardiac surgery, and in patients with right ventricular dysfunction.76,79 However, its place in acutely ill patients has not been well studied and remains poorly defined. In addition, the high costs of the drug limit its use. Several large phase 3 studies are ongoing to assess the use of levosimendan in various conditions.

Sodium Nitroprusside

Sodium nitroprusside has a direct, short-acting relaxing effect on vascular smooth muscle and has been used primarily to reduce afterload, although by reducing venous return it also reduces preload. It has significant effects on arterial pressure, so that it is also used in the management of hypertensive crises. Sodium nitroprusside has the advantage of a short half-life, allowing easy titration. Administration should start at 20 mcg/min IV, and the dose can be progressively increased to 150 to 200 mcg/min. The administration of nitroprusside is complicated by the fact that it is very light sensitive, so that infusion sets must be opaque. Nitroprusside is rapidly metabolized to cyanide and thiocyanate, and accumulation of these metabolites can lead to cyanide or thiocyanate toxicity during prolonged administration, especially in patients with renal failure.

Nitrates

Nitrate products include nitroglycerin or isosorbide dinitrate. They again cause relaxation of vascular smooth muscle and hence reduce afterload. Low doses (30-40 mcg/min IV) predominantly produce venodilatation; high doses (150-500 mcg/min) lead to arteriolar dilatation as well. These drugs are widely used IV in the treatment of recurrent ischemia, hypertensive emergencies, and congestive heart failure associated with MI. Nitroglycerin is the preferred vasodilator in acute MI, especially when infarction is complicated by congestive heart failure.80 Nitrates are also indicated in patients with cardiogenic pulmonary edema for their strong relieving effects on venous congestion. Their dose is similar to that of nitroprusside.

Angiotensin-Converting Enzyme Inhibitors

Angiotensin-converting enzyme (ACE) catalyses the conversion of angiotensin I to angiotensin II, a potent vasoconstrictor, and it is also involved in the inactivation of bradykinin, a potent vasodilator. ACE inhibitors therefore cause reduced formation of angiotensin II and increased levels of bradykinin, thus reducing vascular resistance. ACE inhibitors are widely used in the treatment of chronic heart failure and may be helpful in the acute care setting as well. However, they have 2 limitations in the acute setting. One is the risk of renal failure that can be precipitated by ACE inhibition especially in the presence of fluid shifts that may result in relative hypovolemia. The second is the lack of availability of an IV preparation. Enalapril is available as the only parenteral preparation, but its administration results in marked decreases in arterial pressure, so its use is very limited.

Hydralazine

In contrast to nitrates, hydralazine has greater effects on the arteriolar side of the vasculature and is therefore more likely to increase cardiac output by decreasing SVR. It also increases heart rate, so may be useful in patients with a relatively slow heart rate. Its use requires close cardiovascular monitoring. Long-term administration may be complicated by the development of iatrogenic lupus. Hence hydralazine is not often prescribed today for prolonged therapy of heart failure because there are other drugs with a better benefit-to-risk ratio. However, it can still be considered as an alternative to ACE inhibitors in patients with heart failure.

Indices of Global Perfusion

Mean Arterial Pressure

Arterial hypotension represents a marker of serious disease, and a systolic pressure of less than 90 mm Hg (or a mean arterial pressure less than 70 mm Hg) or a decrease in mean arterial pressure greater than 30 mm Hg are indications of acute circulatory failure. Noninvasive measurement of blood pressure with the sphygmomanometer is perhaps one of the most widely used procedures in clinical medicine. However, in critically ill patients, especially those with significant hypotension, invasive intra-arterial monitoring of pressure is more accurate and continuous,81 allowing rapid assessment of response to treatments.

Cardiac Output

A so-called normal cardiac output is difficult to define because what is normal or adequate may vary widely among patients and even in the same patient over time.69 For example, one cardiac output value may be perfectly adequate for an anesthetized, mechanically ventilated patient but inadequate should that same patient develop severe sepsis with increased oxygen requirements. The measurement of cardiac output provides an assessment of global systemic blood flow but provides no information on regional flow. Cardiac output can be judged inadequate when tissue perfusion is altered, as shown clinically by cutaneous vasoconstriction, decreased urine output attributed to kidney underperfusion, or altered mental status. Biochemical signs of reduced tissue perfusion include decreased SvO2 or its surrogate, ScvO2, or increased blood lactate concentrations (see later).

Cardiac output can be measured using various techniques, each of which has its own benefits and drawbacks. For many years, the thermodilution cardiac output measured using a pulmonary artery catheter (PAC) was the gold standard, but with concerns about the safety and overall benefits with PAC insertion and use,82 less invasive techniques, including echocardiography and Doppler measurements, pulse contour analysis, partial carbon dioxide rebreathing, and electrical bioimpedance, are being used more commonly. Nevertheless, although the PAC is not indicated in all patients, it remains valuable in more complex cases, allowing almost continuous measurement not only of cardiac output, but also of other important parameters including cardiac filling pressures and SVO2. Accurate absolute measures of cardiac output are perhaps less important than monitoring the trends in cardiac output in response to treatment and time, and if one wants to define the adequacy of cardiac output accurately, measures of SVO2 are essential.83 Importantly, targeting predefined cardiac output values for all patients or even for groups of patients is not advisable because adequacy of cardiac output varies among patients and in the same patient over time.

Mixed Venous Oxygen Saturation

SvO2 can be measured using a PAC. The normal value is 70% to 75%. SVO2 depends on cardiac output, oxygen demand, hemoglobin, and arterial oxygen saturation. Measurements of SVO2 are very useful for the correct interpretation of cardiac output (Fig. 80-5). SVO2 is typically decreased in low flow states, but normal or high in inflammatory conditions.

SCVO2, measured in the superior vena cava using a central venous catheter, may represent a useful surrogate for patients who do not require a PAC. However, it must be remembered that as SCVO2 reflects the oxygenation of the venous blood from the upper half of the body and not of the lower half of the body, it is an estimate of the true SVO2, which measures whole body venous oxygen saturation. Physiologically, SCVO2 is slightly less than SVO2, but in critically ill patients it is often greater. This effect is explained by increased oxygen extraction in the kidneys (where blood flow is often reduced) and in the gut (where oxygen consumption is often increased), whereas oxygen extraction in the liver is more constant. A study has suggested that early therapy in patients with severe sepsis or septic shock targeting an SCVO2 of 70% or higher may be associated with improved outcomes.2 As with many other parameters, trends in SCVO2 over time are of more use than individual values, and values must be interpreted in the light of clinical assessment and other hemodynamic and oxygenation parameters.

Blood Lactate Levels

Blood lactate levels provide a useful indication of the presence of anaerobic metabolism due to hypoperfusion, and, although changes in blood lactate levels occur too slowly to guide therapy, nevertheless repeating measurements every hour or so can provide useful information on the ongoing adequacy of global tissue oxygenation. In patients with sepsis, the interpretation of blood lactate levels is not always straightforward, in that increased lactate levels may result from cellular metabolic failure as well as global hypoperfusion. However, the prognostic value of increased blood lactate levels is well established in septic shock patients, particularly if the high levels persist.84

Indices of Regional Perfusion

Although measurement and monitoring of cardiac output, SvO2, and blood lactate levels provide valuable information regarding global adequacy or perfusion and oxygenation, they do not provide detail on the adequacy of regional microcirculatory blood flow or oxygen delivery and utilization, which is believed to be a crucial factor in the development of multiple organ failure in patients with shock. Measuring regional parameters is of particular importance because they can remain inadequate even when global parameters have returned to normal. However, although considerable progress is being made in this field, there is as yet no ideal technique that can be used in critically ill patients to assess regional blood flow or oxygenation.

Gut Tonometry

The measurement of regional perfusion initially focused on the hepatosplanchnic circulation because this is particularly sensitive to changes in blood flow and oxygenation, and it may have a higher critical oxygen delivery threshold than other organs. Gastric tonometry was introduced to monitor gastric pHi, but this technique is limited by measurement errors and can be unreliable during patient feeding.85 Calculations of pHi were replaced by gastric mucosal PCO2 because this measure is not confounded by arterial bicarbonate, but because gastric mucosal PCO2 is influenced by systemic arterial PCO2, the gastric-arterial PCO2 difference may be of more interest.86 An early trial suggested that tonometry-derived parameters may be useful in guiding therapy,87 but these findings were not confirmed in a later study,88 and many investigators have emphasized the limited sensitivity, and especially specificity, of these measurements. Different vasoactive agents have been shown to have divergent effects on PCO2 and pHi that are neither consistent nor predictable.89

Sublingual Capnography

The methodological limitations of gastric tonometry encouraged researchers to look elsewhere for another easily accessible tissue where local PCO2 could be measured, and the sublingual circulation has been suggested for this purpose. Several clinical studies have suggested that sublingual PCO2 (PslCO2) is a reliable marker of tissue hypoperfusion. However, as with gut PCO2, PslCO2 may also be influenced by arterial carbon dioxide and the gradient (ie, gap) between PslCO2 and PaCO2 may be more specific for tissue hypoperfusion.90 In a recent study, the sublingual PCO2 gap was found to correlate with the alterations in microcirculatory blood flow seen in patients with severe sepsis.91 Further work is clearly needed to define normal and pathologic sublingual PCO2 levels and to establish whether targeting treatments to restore a normal sublingual PCO2 can improve outcomes.

Microcirculatory Imaging

The development of handheld orthogonal polarization spectral (OPS)/sidestream dark field (SDF) imaging techniques is providing a new means of directly visualizing the microcirculation and evaluating the effects of interventions on microcirculatory flow in easily accessible epithelial surfaces, such as the sublingual area.92 Microcirculatory changes have been identified in various disease processes including sepsis, cardiogenic shock, heart failure, and after cardiac surgery.93,94 The observed alterations include decreased capillary density and a reduced proportion of perfused capillaries, and increased heterogeneity of blood flow; persistent alterations are associated with worse outcomes.95 Various therapeutic interventions have been shown to have an impact on the microcirculatory changes, but further studies are needed to determine whether therapy guided by OPS/SDF monitoring can improve outcomes.

Near-Infrared Spectroscopy

Near-infrared spectroscopy (NIRS) is a technique that uses near-infrared light to measure chromophores (oxy- and deoxyhemoglobin, myoglobin, and cytochrome aa3) in tissues. The tissue oxygen saturation (StO2) is calculated from the fractions of oxy- and deoxyhemoglobin. However, the StO2 represents the sum of oxygen saturations in the volume under the probe and can be unreliable in conditions of heterogeneous flow. Analysis of the changes in StO2 during a brief episode of forearm ischemia enables quantification of microvascular dysfunction.96 Again, recent studies have shown that patients with severe sepsis frequently have profound alterations in NIRS-derived dynamic variables and that these alterations are associated with a poor outcome.97 Further studies are again needed to determine whether therapy guided by NIRS can improve outcomes.

Multiple organ failure is a serious problem in ICU patients, resulting in high morbidity and mortality and associated resource use and costs. Multiple organ failure can arise from many causes, but a key factor in its development appears to be an imbalance between tissue oxygen requirements and uptake. Tissue oxygen uptake can be limited by inadequate oxygen delivery and availability, and also by altered oxygen extraction ability. Appropriate hemodynamic support can improve both global and regional blood flow, and hence oxygen delivery. Early studies suggested that those patients with greater cardiac index and oxygen delivery values were more likely to survive conditions associated with the development of multiple organ failure.98,99 This observation was the rationale behind several studies in the early 1990s, which assessed the benefits of increasing oxygen delivery to supranormal values using inotropes and fluids. Although this approach was shown to improve outcomes in some groups of critically ill patients, notably surgical and trauma patients,100,101 routine application to all critically ill patients does not confer a uniform survival benefit and may worsen outcomes in some patients.70,102 However, a protocol targeting adequate oxygen transport as reflected by the SCVO2, in patients at risk of multiple organ failure, has been shown to result in improved survival,2 and adequate oxygenation clearly remains an important parameter to target. Unfortunately, at present there is no clear marker of adequate tissue oxygenation, and we must therefore rely on surrogate markers of tissue perfusion and oxygenation as discussed earlier. Adequate oxygen delivery is likely to vary among patients and in the same patient over time, so that hemodynamic support must be titrated and adjusted carefully based on each individual's response to the therapy.

• Dellinger RP, Levy MM, Carlet JM, et al. Surviving Sepsis Campaign: international guidelines for management of severe sepsis and septic shock: 2008. Crit Care Med. 2008;36: 296-327.
• Hollenberg SM, Ahrens TS, Annane D, et al. Practice parameters for hemodynamic support of sepsis in adult patients: 2004 update. Crit Care Med. 2004;32(9):1928-1948.
• Vincent JL, Weil MH. Fluid challenge revisited. Crit Care Med. 2006;34:1333-1337.