Emergency Department Care: The treatment for sepsis has evolved considerably over the past 10 years as it has transitioned from a disease that primarily concerned only critical care physicians to one that has a major impact in the emergency department. Early recognition and early aggressive therapy for patients with sepsis have a significant impact on mortality.
Rivers et al brought this issue to the forefront with their landmark article in the New England Journal of Medicine in 2001, where they instituted a treatment protocol for patients with septic shock, termed Early Goal Directed Therapy (EGDT). EGDT emphasizes early recognition of patients with potential sepsis in the ED, early broad-spectrum antibiotics, and a rapid crystalloid fluid bolus, followed by goal-directed therapy for those patients who remain hypotensive or severely ill after this initial therapy. Those patients who did not respond to an initial fluid bolus and antibiotics received a CV catheter in the internal jugular or subclavian vein to measure central venous pressure (CVP) and an arterial catheter to directly measure arterial blood pressure.
EGDT is basically a three-step process aimed at optimizing tissue perfusion.
The first step involves titrating crystalloid fluid administration to CVP, or administering 500 mL boluses of fluid until the CVP measures between 8 and 12 mm Hg. CVP is a surrogate for intravascular volume, as excess circulating blood volume is contained within the venous system. Only after the CVP is greater than 8 mm Hg should vasopressors be considered.
The second step, if the patient has not improved with fluid alone, is to administer vasopressors to attain a mean arterial pressure (MAP) greater than 65 mm Hg.
The third step is to evaluate the central venous oxygen saturation (SvO2). This is obtained from the CV line, which, in turn, is a surrogate for peripheral tissue oxygenation and cardiac output. An SvO2 of less than 70% is considered abnormal and indicative of suboptimal therapy. In this case, the hematocrit is checked and blood transfused until a hematocrit greater than 30% is attained. Once this is attained and the SvO2 is still low, dobutamine is initiated to increase cardiac output.
Rivers et al were able to enroll 263 patients who met criteria for septic shock:
Suspected infection
2 of the 4 SIRS criteria
Persistent systolic blood pressure <90 mm Hg after initial fluid bolus or lactate >4 mmol/L
These patients were randomized to EGDT versus "standard" therapy, which still included placement of a CV line and arterial catheter (both relatively aggressive measures and probably not standard in most EDs). Despite this, he found an absolute mortality benefit of 16% with EGDT (30% mortality with EGDT vs 46% mortality with standard therapy).
When the data were examined closely, it was found that the patients in the EGDT group received, on average, more crystalloid fluid (5.0 L vs 3.5 L) and a much higher percentage of patients received blood transfusion (64% vs 18%). The resulting average SvO2 measured after therapy was 95% for the EGDT group versus 60% in the standard group. These data attest to the fact that sepsis is likely grossly undertreated in the average ED setting.
Address A and B of the ABCs: Supplemental oxygen should be administered to all patients with suspected sepsis. Early intubation and mechanical ventilation should be strongly considered for patients with an oxygen requirement, dyspnea or increased respiratory rate, hypotension, or those with evidence of poor peripheral perfusion.
Patients with suspected sepsis should receive an initial crystalloid fluid bolus of 20-30 mL/kg (1-2 L) as well as broad-spectrum antibiotics.
Selection of particular antibiotic agents is empiric based on an assessment of the patient's underlying host defenses, the potential sources of infection, and the most likely responsible organisms. Antibiotics must have a broad spectrum and cover gram-positive, gram-negative, and anaerobic bacteria because all classes of these organisms produce identical clinical pictures. Antibiotics must be given parenterally in doses adequate to achieve bactericidal serum levels. Many studies have revealed that clinical improvement correlates with the achievement of serum bactericidal levels rather than with the number of antibiotics given.
Antistaphylococcal coverage is recommended in patients with a history of intravenous drug use or in those with indwelling vascular catheters or devices. Coverage directed against anaerobes should be included in patients with intra-abdominal or perineal infections. Antipseudomonal coverage is indicated in patients with neutropenia or burns. Immunocompetent patients usually can be treated with a single drug with broad-spectrum coverage such as a third-generation cephalosporin. Immunocompromised patients usually require dual-antibiotic coverage with broad-spectrum antibiotics that have overlapping coverages. Within these general guidelines, no single combination of antibiotics is clearly superior to others.
For patients who have persistent hypotension after this initial treatment, 500 mL boluses (10 mL/kg in children) should be administered every 15 minutes to attain a CVP of 8-12 mm Hg. It is not unusual to require 4-6 L of isotonic crystalloid in order to achieve this goal.
Patients should be monitored for signs of volume overload and need for intubation; these signs include dyspnea, pulmonary rales, or pulmonary edema on the chest radiograph. Improvement, stabilization, and normalization in the patient's mental status, heart rate, BP, capillary refill, and urinary output indicate adequate volume resuscitation.
Colloid resuscitation (with albumin or hetastarch) has no proven benefit compared with isotonic crystalloid resuscitation (isotonic sodium chloride solution or lactated Ringer solution).
Vasopressors should be started once a CVP of 8-12 mm Hg is achieved and the patient remains hypotensive. Vasopressors are titrated to achieve a mean arterial pressure (MAP) greater than 65 mm Hg.
Norepinephrine is now the first agent recommended in treating septic shock refractory to volume resuscitation. Norepinephrine has predominantly alpha-receptor activity, which is effective in vasodilatory shock. The dose ranges from 5 mcg/min up to 20 mcg/min, and it is not based on the weight of the patient.
Dopamine is an agent that is commonly used in the treatment of septic shock. Its beta-receptor effects are equal to its alpha-receptor effect; therefore, it has a greater effect on cardiac contractility than does norepinephrine. This may however have deleterious effects in the setting of sepsis where the heart is already maximally physiologically stimulated. Doses range from 2-20 mcg/kg/min.
The overall treatment goal is to achieve a central venous oxygen saturation (SvO2) of greater than 70%. SvO2 can be measured from a venous blood gas analysis, or it can be measured in a continuous fashion with the use of stand-alone bedside monitors that are now available in many ICUs and in some EDs. Rivers et al utilized such a device in their study. If the above treatments do not achieve an SvO2 greater than 70%, then oxygen carrying capacity and cardiac output must be addressed.
Transfuse packed RBCs until the hematocrit is greater than 30%.
Administer dobutamine to increase cardiac output in order to optimize red cell delivery to peripheral tissues. Dobutamine is a pure beta-agonist, resulting in increased cardiac contractility and heart rate. Doses range from 2-20 mcg/kg/min.
Administration of steroids (eg, methylprednisolone, hydrocortisone, dexamethasone) has theoretical benefits in the setting of severe sepsis by inhibiting the massive inflammatory cascade that is unleashed. Cortisol is a naturally occurring stress hormone that promotes vascular tone and endothelial integrity. It has been shown experimentally to potentiate the effect of vasopressors. A large randomized trial in the 1980s showed a lack of benefit with high-dose methylprednisolone at 30 mg/kg for all patients with septic shock.
More recent data show that steroids can be beneficial for patients with relative adrenal insufficiency and in lower doses. A study in JAMA in 2002 looked at 299 patients with septic shock all who were intubated, persistently hypotensive despite fluids and vasopressors, and had evidence of end-organ failure. Patients were administered a cortisol stimulation test (Cort stim test), which involves measuring cortisol levels before and 30 minutes after administration of cosyntropin (ACTH) 0.25 mg IV. An increase in cortisol less than 10 mcg/dL was considered "nonresponder," and thus adrenally insufficient. Of the 299 patients with septic shock, 77% were nonresponders. All patients were randomized to low-dose steroids (hydrocortisone 50 mg q6h and fludrocortisone 50 mcg daily) versus placebo. For nonresponders, there was an absolute benefit in mortality of 10% (53% vs 43% mortality rate) for those who received steroids.
While performing the Cort stim test in the ED may not be practical due to time and resource constraints, it is worth noting that greater than 75% of patients with vasopressor-refractory hypotension were adrenally insufficient. Therefore, steroids should be empirically administered to this group of patients. A common choice is hydrocortisone 100 mg IV; a good alternative is dexamethasone 10 mg IV. The added benefit of dexamethasone is that it does not interfere with the Cort stim test because it does not affect the hypothalamic-pituitary-adrenal axis.
Activated protein C (APC) is an endogenous protein that modulates inflammation and coagulation. Specifically, it inhibits TNF-alpha, IL-1, and IL-6, the mediators thought to play a major role in initiating the inflammatory response seen in sepsis. In addition, it inhibits monocyte and neutrophil adhesion to endothelial cells, and it inhibits thrombin and fibrin production, and thus prevents microvascular thrombi. APC levels have been shown to be low in sepsis.
A study appeared in the New England Journal of Medicine in 2001 looking at 1,690 patients with sepsis and organ dysfunction who were randomized to recombinant human APC (drotrecogin alpha) or placebo. Of note, this study excluded patients who were expected to die within 28 days (eg, end-stage cancer), those with end-stage renal disease or cirrhosis, and those with HIV and a CD4 count less than 50. There was an absolute benefit of 6% in mortality at 28 days with the administration of APC (25% mortality rate vs 31% in the placebo group). They reported a 13% benefit in the sickest patients (817 patients with an APACHE II score > 25) and an 18% benefit for the sickest patients with pneumonia (317 patients with APACHE II score > 25 and pneumonia).
The drawback to APC is an increased incidence of bleeding complications because it inhibits coagulation. Overall bleeding complications were 3.5% in the APC group versus 2.0% in the placebo group (n=0.06). For this reason, APC is contraindicated in patients with a known hypercoagulable condition, recent major surgery, intracranial surgery or stroke within 3 months, any history of arteriovenous malformation (AVM), and cerebral aneurysm or mass.
This is a very expensive therapy, on the order of thousands of dollars, and is best instituted in the ICU under the care of a critical care physician. Nevertheless, it is good to keep this in mind in the ED and identify patients who may benefit, especially those who are most critical and those with pneumonia.
SECOND REGIMEN :
Medical Care
The treatment of patients with septic shock consists of the following 3 major goals: (1) Resuscitate the patient from septic shock using supportive measures to correct hypoxia, hypotension, and impaired tissue oxygenation. (2) Identify the source of infection and treat with antimicrobial therapy, surgery, or both. (3) Maintain adequate organ system function guided by cardiovascular monitoring and interrupt the pathogenesis of multiorgan system dysfunction.
The principles in the management of septic shock, based on current literature, include the following components:
Early recognition
Early and adequate antibiotic therapy
Source control
Early hemodynamic resuscitation and continued support
Corticosteroids (refractory vasopressor-dependent shock)
Drotrecogin alpha (Severely ill if APACHE II > 25)
Tight glycemic control
Proper ventilator management with low tidal volume in patients with ARDS
General supportive care: Initial treatment includes support of respiratory and circulatory function, supplemental oxygen, mechanical ventilation, and volume infusion. Treatment beyond these supportive measures includes antimicrobial therapy targeting the most likely pathogen, removal or drainage of the infected foci, treatment of complications, and interventions to prevent and treat effects of harmful host responses. Source control is essential for the following reasons:
Identifying and obtaining source control is an essential component of sepsis management.
In general, the source of sepsis needs to be removed, drained, or otherwise eradicated.
Administer supplemental oxygen to any patients with sepsis who also have hypoxemia or are in respiratory distress.
If the patient's airway is not secure, the gas exchange or acid-base balance is severely deranged, and if evidence of respiratory muscle fatigue exists or if the patient appears markedly distressed, perform an endotracheal intubation.
Patients in septic shock generally require intubation and assisted ventilation because respiratory failure either is present at the onset or may develop during the course of the illness.
Correction of shock state and abnormal tissue perfusion is the next step in the treatment of patients with septic shock.
Hemodynamic support of septic shock
Shock refers to a state of inability to maintain adequate tissue perfusion and oxygenation, ultimately causing cellular, and then organ system, dysfunction. Therefore, the goals of hemodynamic therapy are restoration and maintenance of adequate tissue perfusion to prevent multiple organ dysfunction.
Careful clinical and invasive monitoring is required for assessment of global and regional perfusion. A mean arterial pressure (MAP) of less than 60 mm Hg or a decrease in MAP of 40 mm Hg from baseline defines shock at the bedside.
Elevation of the blood lactate level on serial measurements of lactate can indicate inadequate tissue perfusion.
Mixed venous oxyhemoglobin saturation serves as an indicator of the balance between oxygen delivery and consumption. A decrease in maximal venous oxygen (MVO2) can be secondary to decreased cardiac output; however, maldistribution of blood flow in patients experiencing septic shock may artificially elevate the MVO2 levels. An MVO2 of less than 65% generally indicates decreased tissue perfusion.
Regional perfusion in patients with septic shock is evaluated by adequacy of organ function. The evaluation includes evidence of myocardial ischemia, renal dysfunction manifested by decreased urine output or increased creatinine, CNS dysfunction indicated by a decreased level of consciousness, hepatic injury shown by increased levels of transaminases, splanchnic hypoperfusion manifested by stress ulceration, ileus, or malabsorption.
The hemodynamic support in septic shock is provided by restoring the adequate circulating blood volume, and, if needed, optimizing the perfusion pressure and cardiac function with vasoactive and inotropic support to improve tissue oxygenation.
Intravascular volume resuscitation
Hypovolemia is an important factor contributing to shock and tissue hypoxia; therefore, all patients with sepsis require supplemental fluids. The amount and rate of infusion are guided by an assessment of the patient's volume and cardiovascular status. Monitor patients for signs of volume overload, such as dyspnea, elevated jugular venous pressure, crackles on auscultation, and pulmonary edema on the chest radiograph. Improvement in the patient's mental status, heart rate, MAP, capillary refill, and urine output indicate adequate volume resuscitation.
Large volumes of fluid infusions are required as initial therapy in patients with septic shock. Administer fluid therapy with predetermined boluses (500 mL or 10 mL/kg) titrated to the clinical end points of heart rate, urine output, and blood pressure. Continue fluid resuscitation until the clinical end points are reached or the pulmonary capillary wedge pressure exceeds 18 mm Hg. The volume resuscitation can be achieved by either crystalloid or colloid solutions. The crystalloid solutions are 0.9% sodium chloride and lactated Ringer solution. The colloids are albumin, dextrans, and pentastarch. Clinical trials have failed to show superiority of either crystalloids or colloids as the resuscitation fluid of choice in septic shock. However, 2-4 times more volume of crystalloids than colloids are required, and crystalloids take a longer time to achieve the same end points, whereas the colloid solutions are much more expensive.
Data from several studies suggest that formation of pulmonary edema is no different with crystalloids compared to colloids when the filling pressures are maintained at a lower level. However, if the higher filling pressures are required for maintenance of optimal hemodynamics, crystalloids may increase extravascular fluid fluxes because of a decrease in plasma oncotic pressure.
In some patients, clinically assessing the response to volume infusion may be difficult. By monitoring the response of the central venous pressure or pulmonary artery occlusion pressure to fluid boluses, the physician can assess such patients. A sustained rise in filling pressure of more than 5 mm Hg after a volume is infused indicates that the compliance of the vascular system is decreasing as further fluid is being infused. Such patients are susceptible to volume overload, and further fluid should be administered with care.
Early goal-directed management of sepsis: In a study by Rivers et al, 263 patients treated in an emergency department were randomized to either a standard care control group or an aggressive care therapy arm for their initial 8 hours of treatment. Patients in the therapy arm provided aggressive resuscitation via to reach a central venous pressure to 8- 12 mm Hg, organ perfusion pressure maintained by keeping mean arterial pressure (MAP) 65-90mm Hg using either vasopressors or vasodilators, and contractility with dobutamine to keep central venous O2 saturation (ScvO2) greater than 70% after transfusion to hematocrit greater than 30%. This treatment strategy resulted in a 16% improvement in mortality.
Vasopressor supportive therapy
If the patient does not respond to several liters of volume infusion with isotonic crystalloid solution (usually 4 L or more) or evidence of volume overload is present, the depressed cardiovascular system can be stimulated by inotropic and vasoconstrictive agents. When proper fluid resuscitation fails to restore hemodynamic stability and tissue perfusion, initiate therapy with vasopressor agents. These agents are dopamine, norepinephrine, epinephrine, and phenylephrine. These agents are vasoconstricting drugs that maintain adequate blood pressure during life-threatening hypotension and preserve perfusion pressure for optimizing flow in various organs.
The mean blood pressure required for adequate splanchnic and renal perfusion (MAP of 60 or 65 mm Hg) is based on clinical indices of organ function. Dopamine is the most commonly used agent for this purpose. Treatment usually begins at a rate of 5-10 mcg/kg/min IV, and the infusion is adjusted according to the blood pressure and other hemodynamic parameters. Often, patients may require high doses of dopamine (as much as 20 mcg/kg/min). Presently, norepinephrine is the preferred drug because dopamine is known to cause unfavorable flow distribution.
If the patient remains hypotensive despite volume infusion and moderate doses of dopamine, a direct vasoconstrictor (eg, norepinephrine) should be started at a dose of 0.5 mcg/kg/min and titrated to maintain a MAP of 60 mm Hg. While potent vasoconstrictors (eg, norepinephrine) traditionally have been avoided because of their adverse effects on cardiac output and renal perfusion, data from animal and human studies reveal that norepinephrine can reverse septic shock in patients unresponsive to volume and dopamine. These patients require invasive hemodynamic monitoring with arterial lines and pulmonary artery catheters. Vasopressors may cause more harm than good if administered to patients whose inadequate intravascular volume is not restored (ie, a patient "whose tank is not filled").
The following is a brief review of the mechanism of action and utility of drugs used for hemodynamic support of septic shock:
Dopamine: A precursor of norepinephrine and epinephrine, dopamine has varying effects according to the doses infused. A dose of less than 5 mcg/kg/min results in vasodilation of renal, mesenteric, and coronary beds. At a dose of 5-10 mcg/kg/min, beta1-adrenergic effects induce an increase in cardiac contractility and heart rate. At doses of about 10 mcg/kg/min, alpha-adrenergic effects lead to arterial vasoconstriction and elevation in blood pressure. Dopamine is effective in optimizing MAP in patients with septic shock who remain hypotensive after volume resuscitation. The blood pressure increases primarily as a result of inotropic effect and, thus, will be useful in patients who have concomitant reduced cardiac function. The undesirable effects are tachycardia, increased pulmonary shunting, potential to decrease splanchnic perfusion, and increase in pulmonary arterial wedge pressure.
Norepinephrine
This agent is a potent alpha-adrenergic agonist with minimal beta-adrenergic agonist effects. Norepinephrine can increase blood pressure successfully in patients with sepsis who remain hypotensive following fluid resuscitation and dopamine. The dose of norepinephrine may vary from 0.2-1.5 mcg/kg/min, and large doses as high as 3.3 mcg/kg/min have been used because of the alpha-receptor down-regulation in sepsis.
In patients with sepsis, indices of regional perfusion (eg, urine flow) and lactate concentration have improved following norepinephrine infusion. Two recent trials have shown that a significantly greater proportion of patients treated with norepinephrine were resuscitated successfully, as opposed to the patients treated with dopamine. Therefore, norepinephrine should be used early and should not be withheld as a last resort in patients with severe sepsis who are in shock.
The concerns about compromising splanchnic tissue oxygenation have not been proven; the studies have confirmed no deleterious effects on splanchnic oxygen consumption and hepatic glucose production, provided adequate cardiac output is maintained.
Epinephrine: This agent can increase MAP by increasing cardiac index and stroke volume, along with an increase in systemic vascular resistance and heart rate. Epinephrine may increase oxygen delivery and oxygen consumption and decreases the splanchnic blood flow. Administration of this agent is associated with an increase in systemic and regional lactate concentrations. The use of epinephrine is recommended only in patients who are unresponsive to traditional agents. The undesirable effects are an increase in lactate concentration, a potential to produce myocardial ischemia, development of arrhythmias, and a reduction in splanchnic flow.
Phenylephrine: This agent is a selective alpha1-adrenergic receptor agonist that is used primarily in anesthesia to increase blood pressure. Although studies are limited, phenylephrine increased MAP in patients who were septic hypotensive with increased oxygen consumption. However, the concern remains about its potential to reduce cardiac output and lower heart rate in patients with sepsis. Phenylephrine may be a good choice when tachyarrhythmias limit therapy with other vasopressors.
Inotropic therapy: Although myocardial performance is altered during sepsis and septic shock, cardiac output generally is maintained in patients with volume-resuscitated sepsis. Data from the 1980s and 1990s suggest a linear relationship between oxygen delivery and oxygen consumption (pathologic supply dependency), indicating that the oxygen delivery likely was insufficient to meet the metabolic needs of the patient. However, recent investigators have challenged the concept of pathologic supply dependency, suggesting that elevating cardiac index and oxygen delivery (hyperresuscitation) was not associated with improved patient outcome. Therefore, the role of inotropic therapy is uncertain, unless the patient has inadequate cardiac index, mean arterial pressure, mixed venous oxygen saturation, and urine output despite adequate volume resuscitation and vasopressor therapy.
Renal-dose dopamine: In the setting of circulatory shock of any etiology, several well-designed clinical trials have failed to demonstrate any beneficial effects of low dose dopamine to improve renal blood flow and support renal function. Dopamine at a dose of 2-3 mcg/kg/min is known to initiate diuresis by increasing renal blood flow in healthy animals and volunteers. Multiple studies have not demonstrated a beneficial effect of prophylactic or therapeutic low-dose dopamine administration in patients with sepsis who are critically ill. Considering the real side effects of dopamine infusion, the use of renal dose dopamine should be abandoned.
Empirical antimicrobial therapy
Initiate this therapy early in patients experiencing septic shock. However, antibiotics have little effect on the clinical outcome for at least 24 hours. The selection of appropriate agents is based on the patient's underlying host defenses, the potential sources of infection, and the most likely culprit organisms. If the patient is "antibiotic experienced," strongly consider the use of an aminoglycoside rather than a quinolone or cephalosporin for gram-negative coverage. Knowing the antibiotic resistance patterns of both the hospital itself and its referral base (ie, nursing homes) is important. Antibiotics must be broad-spectrum agents and must cover gram-positive, gram-negative, and anaerobic bacteria because the different classes of these organisms produce an identical clinical picture of distributive shock.
Administer the antibiotics parenterally, in doses adequate to achieve bactericidal serum levels. Many studies find that the clinical improvement correlates with the achievement of serum bactericidal levels rather than the number of antibiotics administered.
Include coverage directed against anaerobes in patients with intra-abdominal or perineal infections. Antipseudomonal coverage is indicated in patients with neutropenia or burns or in patients who acquired sepsis while hospitalized. Patients who are immunocompetent usually can be treated with a single drug with broad-spectrum coverage, such as a third-generation cephalosporin. Patients who are immunocompromised typically require dual broad-spectrum antibiotics with overlapping coverage. Within these general guidelines, no single combination of antibiotics is clearly superior to others.
The following points should always be kept in mind:
Early, empiric antibiotic coverage is essential with narrowed spectrum when culture results are available.
Waiting until cultures are back is an invalid reason to withhold antibiotics.
Only 30% of patients with presumed septic shock have positive blood cultures.
Twenty-five percent of presumed septic shock patients remain culture negative from all sites, but mortality with culture positive counterparts is similar.
Recombinant human activated protein C
The inflammatory mediators are known to cause activation of coagulation inhibitors of fibrinolysis, thereby causing diffuse endovascular injury, multiorgan dysfunction, and death. Activated protein C is an endogenous protein that not only promotes fibrinolysis and inhibits thrombosis and inflammation but also may modulate the coagulation and inflammation of severe sepsis. Sepsis reduces the level of protein C and inhibits conversion of protein C to activated protein C. Administration of recombinant activated protein C inhibits thrombosis and inflammation, promotes fibrinolysis, and modulates coagulation and inflammation.
A recent publication by the Recombinant Human Activated Protein C Worldwide Evaluation in Severe Sepsis (PROWESS) study group demonstrated that the administration of recombinant human activated protein C (drotrecogin-alpha, activated) resulted in lower mortality rates (24.7% vs 30.8%) in the treated group compared with placebo. Treatment with activated drotrecogin-alpha was associated with reduction in the relative risk of death by 19.4% (95% CI, 6.6-30.5) and an absolute reduction in risk of death by 6.1%, (P=.005).
Corticosteroids: Although theoretical and experimental animal evidence exists for the use of large doses of corticosteroids in those with severe sepsis and septic shock, all randomized human studies (except 1 from 1976) found that corticosteroids did not prevent the development of shock, reverse the shock state, or improve the 14-day mortality rate. Therefore, no support exists in the medical literature for the routine use of high doses of corticosteroids in patients with sepsis or septic shock. A meta-analysis of 10 prospective, randomized, controlled trials of glucocorticoid use did not report any benefit from corticosteroids. Therefore, high-dose corticosteroids should not be used in patients with severe sepsis or septic shock.
Although further studies await further confirmation, current recommendations are as follows:
Drotrecogin alpha (activated protein C) is the only widely accepted drug specific to the therapy of sepsis.
Drotrecogin alpha should be considered for patients with APACHE II scores greater than 25.
The main side effect of Drotrecogin alpha is bleeding.
Stress-dose glucocorticoids: Recent trials (Briegel, 1999; Cartlet, 1999) demonstrated positive results of stress-dose administration of corticosteroids in patients with severe and refractory shock. Although further confirmatory studies are awaited, stress-dose steroid coverage should be provided to patients who have the possibility of adrenal suppression.
The following key points summarize use of corticosteroids in septic shock:
Older, traditional trials of corticosteroids in sepsis were unsuccessful likely because of high doses and poor patient selection.
Recent trials with low-dose (physiologic) dosages in select patient populations (vasopressor dependent and possibly relative adrenal insufficiency) have resulted in improved outcome.
Corticosteroids should be initiated for patients with vasopressor-dependent septic shock.
A cosyntropin stimulation test may be performed to identify patients with relative adrenal insufficiency defined recently as failure to increase levels > 9 mcg/dL
Tight glycemic control:
Tight glycemic control has recently become a prominent emphasis in the care of critically ill patients, and recent data has been extrapolated to potentially apply to septic populations. A 2001 Belgian study of surgical intensive care unit (ICU) patients that remained in the ICU for more than 5 days showed a 10% mortality benefit in those with tighter glycemic control. The glucose levels for these patients were maintained from 80-110 mg per dL through the use of intensive insulin therapy. The benefit of glycemic control appears to result more from aggressive avoidance of the detrimental effects of hyperglycemia rather than the potential therapeutic effect of insulin.
Based on the current evidence, the Surviving Sepsis Campaign recommends maintaining a glucose level of less than 150 mg/dL, although the logic behind choosing this level is unclear (Dellinger, 2004). Van den Berge documented benefit only once glucose levels were maintained below 110 mg/dl, with increased mortality when blood glucose levels were allowed to reach 130-150 mg/dl. This same group recently finished a large prospective study in medical patients (NEJM, 2006) documenting similar benefits in these patients.
Tight glycemic control has been shown to improve mortality in both postoperative surgical patients including, and particularly, those with sepsis and in-medical ICU patients.
The Surviving Sepsis Campaign recommends that glucose levels in the septic patient should be kept at less than 150 mg/dL although the published evidence supports controlling blood glucose between 80 and 110 mg/dL.
Tight glycemic control is not without risks. In the elderly (>75 years of age) and in those patients with liver failure, excessive hypoglycemic reactions limits its use. Furthermore, to be effective, glycemic control needs to be protocol driven and run by the bedside caregiver, usually the bedside nurse.
Experimental and other therapies include nonadrenergic vasopressors and inotropes. The clinical utility of several of these agents remains unproven despite several studies indicating their beneficial effect on hemodynamic instability.
Dopexamine: This agent has beta 2-adrenergic and dopaminergic effects without any alpha-adrenergic activity and is known to increase splanchnic perfusion. A few small studies have shown that dopexamine increases cardiac index and heart rate and decreases systemic vascular resistance in a dose-dependent manner. The hepatic blood flow and gastric intramucosal pH improve, but results are not reproducible consistently. This drug appears to be promising for patients with sepsis and septic shock, but superiority over the other drugs has not been demonstrated. Dopexamine continues to be an experimental medication in the United States.
Vasopressin: This agent may be useful in patients with refractory septic shock; however, minimal studies have been conducted. In patients with septic shock, infusion of 0.04 U/kg/min of vasopressin resulted in improved MAP secondary to peripheral vasoconstriction.
Phosphodiesterases inhibitors: Inamrinone (formerly amrinone) and milrinone are inotropic agents with vasodilating properties, and each has a long half-life. The mechanism of action occurs via phosphodiesterase inhibition. These medications are beneficial in cardiac pump failure, but their benefit in patients experiencing septic shock is not well established. Furthermore, these agents have a propensity to worsen hypotension in patients with septic shock.
Nitric oxide inhibitor: This agent is a potent endogenous vasodilator. Excessive nitric oxide production, because of the cytokines and other mediators, induces vasodilation and hypotension in patients with sepsis. Nitric oxide is synthesized from endogenous L-arginine by the enzyme nitric oxide synthase. Inhibitors of nitric oxide synthase (N-monomethyl-l-arginine, L-NMMA) in sepsis augment mean arterial pressure, decreased cardiac output, and increased systemic vascular resistance. Inordinate mortality was the cause of early termination of a randomized trial of nitric oxide synthase inhibition with L-NMMA. The clinical benefit of this therapeutic approach in patients with sepsis remains unproven.
Anti-inflammatory therapy: The rationale for anti-inflammatory therapy is that blocking the production of inflammatory mediators may ameliorate the deleterious host inflammatory response and, hence, may limit the tissue injury.
Ibuprofen: Despite promising results in animal studies, the use of ibuprofen has not been proven of any benefit in patients with septic shock.
Antiendotoxin treatment: The insight that endotoxin, a lipid-polysaccharide compound found in the cell wall of gram-negative bacteria, plays a key role in initiating the humoral cascade observed in septic shock led to the hypothesis that neutralizing the circulating endotoxin with IV administration of an antiendotoxin antibody might be beneficial. Several products have been developed and investigated by carefully conducted human trials. To date, no proven benefit to these agents has been observed. Other methods of extracorporeal elimination of endotoxin, polyclonal antiendotoxin antibodies, or monoclonal antiendotoxin antibodies showed neither improvement in short-term survival nor amelioration of sepsis in humans with septic shock. Trials with some of these compounds are ongoing, and, despite a tendency towards benefit, efficacy data are lacking.
Anticytokine treatment: Serum levels of TNF and IL-1 are elevated in patients with septic shock. Both produce hemodynamic effects that duplicate those found in sepsis. Many studies indicate that both the mediators play key roles in sepsis and septic shock, and some think that TNF may be the central mediator in sepsis. As is the case with antiendotoxin antibodies, antibodies to TNF or IL-1 were hypothesized to be useful in patients with septic shock. However, anti-TNF or antiβIL-1 antibodies have yet to be shown to improve the outcome in sepsis or septic shock. Cytokines are the major mediators of inflammatory cascade. Antibodies or blocking medications against TNF, interleukins, and their receptor blockers have been developed and have undergone clinical trials. In 1997, Zeni conducted a meta-analysis and selected 21 trials representing a total of 6429 patients. A small but insignificant beneficial effect was demonstrated.
Miscellaneous treatment: Several other experimental interventions and therapies have undergone clinical trials for sepsis. Although several of these may have shown benefit, no convincing evidence suggests that these therapies are efficacious. A long list of these interventions or therapies exists; the important ones include intravenous immunoglobulins, interferon gamma, antithrombin-3 infusion, naloxone, pentoxifylline, growth hormone, G-CSF, and hemofiltration or extracorporeal removal of endotoxins. None of these agents was efficacious in properly designed controlled clinical trials.