|
Eight Strategies for Managing Severe Sepsis
Severe sepsis kills up to half of all victims—and the incidence is rising. The authors present eight strategies to help emergency physicians fight sepsis early, before it overwhelms the patient.
By Michael H. Catenacci, MD, and Kaira King, MD
The incidence of severe sepsis is increasing in the United States. Approximately 750,000 cases occur every year, with an anticipated annual increase of 1.5%. The rising incidence may be due to the aging of the U.S. population, proliferating antimicrobial resistance, more invasive procedures being performed, and more prevalent immunocompromised states. Estimates of mortality range from 20% to 50%, depending on the stage of the disease studied. The costs are astounding: An estimated $22,000 is spent on every case and $16.7 billion is spent annually.
Emergency department visits have also increased over the past two decades, especially by populations at risk for sepsis, such as the elderly. Angus noted a 100-fold increased incidence of sepsis in the elderly compared with other age groups. Higher patient volume has been associated with an increase in the severity of illness. Lambe reported a 59% increase in critically ill patients presenting to emergency departments between 1990 and 1999. In addition to higher patient volumes and acuity levels, the emergency department physician is challenged by overcrowding from the “boarding” of patients awaiting admission. Analyzing national data from 2001 to 2004, Wang recorded a mean emergency department length of stay of 4.7 hours for severely septic patients, with 20.4% of patients staying longer than six hours.
The rising incidence of severe sepsis, increased emergency department visits by high-risk populations, and protracted time spent in the emergency department awaiting admission, puts emergency medicine practitioners in a position to make a significant contribution in the fight against sepsis. This article will define sepsis and review eight strategies that may improve outcomes.
WHAT IS SEPSIS?
The word sepsis is derived from the Greek word sepein, meaning to putrefy or make rotten. In the past, physicians disagreed on definitions for sepsis, septicemia, and septic shock, making clinical diagnosis, research, and communication difficult. Then in 1992, the American College of Chest Physicians and the Society of Critical Care Medicine established some common ground. They issued a consensus statement, defining the systemic inflammatory response syndrome (SIRS) as the body’s physiologic response to a variety of clinical insults or multiple stressors: infectious, toxicologic, traumatic, ischemic, or immunologic. The syndrome is manifested by a temperature above 100.5°F, heart rate greater than 90, respiratory rate of more than 20 or a PaCO2 below 32, and a white blood cell count above 12,000, below 4000, or more than 10% bands.
The consensus statement went on to define sepsis as SIRS resulting from a documented or presumed infection and to define infection as a pathologic process caused by the invasion of normally sterile tissues by pathogenic organisms. Severe sepsis was defined as sepsis complicated by either hypotension before a fluid challenge, organ dysfunction, or a lactate level equal to or above 4 mmol/L. Sepsis with persistent hypotension was described as septic shock. The definition of persistent hypotension was a systolic blood pressure below 90 mm Hg or more than 40 mm Hg below baseline, or a mean arterial pressure below 70 mm Hg despite adequate fluid resuscitation.
Emergency department physicians should keep these definitions in mind as they consider the following eight strategies to help improve the outcomes of patients with sepsis.
NO. 1: INCREASED PROVIDER AWARENESS
Increasing provider awareness is essential in improving outcomes in patients with sepsis. The Surviving Sepsis Campaign (SSC) was formed under the administration of the Society of Critical Care Medicine, the European Society of Intensive Care Medicine, and the International Sepsis Forum. Since its inception, eight more organizations have joined the effort, including the American College of Emergency Physicians.
Phase one of the SSC began in October 2002 with the Barcelona Declaration at the 15th Annual Congress of the European Society for Intensive Care Medicine. The declaration called for increased public and provider awareness of sepsis and improvement in its diagnosis and treatment, with the goal of decreasing mortality by 25% within five years. Phase II began in 2003 and culminated in updated guidelines for sepsis management published in 2004. Phase III began in 2005 and resulted in updated guidelines published earlier this year.
Phase III of the SSC is almost complete. The aim is to create a focused plan to help clinicians apply the guidelines at the bedside. Web-site-based tools are available to track improvements in guideline application and morbidity and mortality.
The SSC has made an impact. In a recent study, Ferrer showed that implementing two treatment “bundles” based on SSC guidelines (see box above) decreased mortality from 44% to 39.7% in 59 ICUs in Spain. The resuscitation bundle consisted of six tasks to be completed over the first six hours after patient presentation, while the management bundle consisted of four tasks to be completed over the first 24 hours of admission.
NO. 2: EARLY ANTIBIOTIC ADMINISTRATION
The early administration of appropriate antibiotics is essential in septic patients. If targeted antibiotics are given early, pathogenic organisms will be destroyed faster, toxin production will be limited, and the detrimental effects of a robust inflammatory response may be curbed.
Several studies have looked at the timing of antibiotics with regard to clinical outcomes. Kumar retrospectively studied 2731 patients admitted to 14 ICUs in North America with a diagnosis of septic shock and found a strong correlation between the speed of antibiotic administration and patient outcome. Giving antibiotics within the first hour of the onset of hypotension resulted in an overall survival rate of 79.9%. For each hour that antibiotics were delayed, mortality increased an average of 7.6%. In multivariate analysis, time to initiation of effective antibiotic therapy was the single strongest predictor of outcome. Similar results were shown in a study investigating mortality in patients with cancer and septic shock. Larche noted a seven-fold increase in mortality in patients who received antibiotics more than two hours after admission.
Antibiotics should be given without delay to the critically ill patient. Additionally, empiric antibiotics need to be selected appropriately, anticipating the most likely causative organism. In a large cohort of patients with microbiologically confirmed severe sepsis, Harbarth showed that inappropriate empiric antibiotic therapy was associated with increased mortality (odds ratio 1.8). In this study, inappropriate antibiotics were defined as those given within 24 hours of the diagnosis of sepsis that did not match the in vitro susceptibility of the pathogen eventually isolated.
In another retrospective study of 2158 bacteremic patients by Leibovici, appropriate empirical antibiotic treatment resulted in a decrease in absolute mortality of 14%. The odds ratio for fatality with inappropriate antibiotics and septic shock was 1.6. Ibrahim conducted a prospective study of 492 patients with bacteremia who were admitted to the ICU and showed that mortality rates for those treated with inappropriate antibiotics was 61.9% compared to 28.4% for appropriately treated patients.
The rule in selecting antibiotics for patients with sepsis is to treat the most likely organisms, even if data are limited. Important factors in this process include assessments of the most likely site of infection, host predispositions to infection, and local antibiotic resistance patterns. Secondary factors include drug potency, cost, and pharmacodynamics. The box above lists suggested empiric antibiotic regimens for adult life-threatening sepsis with an unclear source.
NO. 3: SOURCE CONTROL
Identifying and removing infected devices or tissues will eliminate continued exposure to pathogenic microorganisms. Detection through history, physical examination, and rapid diagnostic imaging is important.
Infected prosthetic devices may need to be removed in the emergency department. Research shows that bacteria resist destruction by antimicrobials by adhering to medical devices in protective biofilms. Types of devices that may be removed include indwelling urinary catheters, vascular access devices (such as central venous catheters, peripherally inserted central venous catheters, and hemodialysis catheters), peritoneal dialysis catheters, and vaginal foreign bodies, such as tampons.
The lung is the most commonly identified source of sepsis in both trauma and nontrauma patient populations. As a general rule, thoracentesis of a parapneumonic effusion should be attempted if pleural fluid thickness is greater than 10 mm on a lateral decubitus chest x-ray film. Thoracostomy tube placement is generally indicated for gross purulence aspirated on thoracentesis, fluid pH less than 7.2, pleural fluid glucose below 40 mg/dl, positive gram stain or culture of fluid, and loculated fluid aspirated on thoracentesis.
Superficial skin infections are rarely the source of sepsis. However, immediate surgical consultation for operative debridement is indicated if suspicion of a necrotizing skin infection exists. These infections are often polymicrobic and may include necrotizing fasciitis, invasive streptococcal and staphylococcal cellulitis, clostridial myonecrosis, and Fournier gangrene. The most common bacteria isolated are group A beta-hemolytic streptococcal species, staphylococcal species, non-group A streptococcal species, Enterobacteriaceae, and anaerobes. These infections are suggested by the combination of a skin infection with signs of sepsis, pain out of proportion to examination, rapid progression, bullae formation, subcutaneous crepitus, and gas on imaging studies. Elliott found a low sensitivity (85%) in detecting these infections when the criteria for diagnosis included crepitance, blistering, and visible gas on imaging studies. This underscores the importance of consultation or advanced imaging when clinical suspicion exists.
Intra-abdominal sources of infection may be approached through open or percutaneous drainage. There is an expanding body of literature on this topic and recommendations continue to evolve. Recently, CT-guided drainage of intra-abdominal abscesses was shown to be effective in a small European study. Betsch found that 83% of patients with intra-abdominal abscesses who underwent CT-guided drainage had complete resolution with no surgery required. Higher success rates occurred in patients with Acute Physiology and Chronic Health Evaluation (APACHE) III scores under 30 and in patients with smaller, more accessible abscesses.
NO. 4: EARLY GOAL-DIRECTED THERAPY
In 2001, Rivers conducted a landmark study investigating early goal-directed therapy (EGDT) in septic shock. Goal-directed therapy had been used previously in the ICU to match systemic oxygen delivery with demand through manipulating cardiac preload, afterload, and contractility. This study commenced before ICU admission, within six hours of the patients’ arrival at a busy urban emergency department. Patients were randomized to EGDT or usual care and were followed for 72 hours to investigate mortality, end points of resuscitation, and APACHE II scores. The EGDT protocol is shown in the algorithm below.
In the patients who received EGDT, absolute mortality was decreased by 16% compared with the control group (30.5% versus 46.5%), APACHE II scores were lower (6.3 versus 13.0), and end points of resuscitation were improved (lower serum lactate levels, higher serum pH values, and higher central venous oxygen saturation).
After the initial EGDT study was published, questions of external validity as well as the practicality of instituting the protocol in other centers were raised. In a recent review, Rivers identified 11 peer-reviewed publications and 28 abstracts from academic, community, and international settings that have investigated the use of the EGDT protocol since the initial study. These studies provided external validation, showing a decrease in morbidity, mortality, and health care resource consumption.
NO. 5: USE OF CORTICOSTEROIDS
The therapeutic use of steroids in sepsis spans decades of clinical research. Most early trials investigated high doses of steroids and showed no benefit or possible increased harm. Next, low physiologic or replacement doses of steroids were suggested based on the results of cosyntropin stimulation testing. In this test, cosyntropin (a synthetic adrenocorticotropic hormone analog) is given and cortisol levels are measured at 0, 30, and 60 minutes. If the cortisol level increases by less than 9 µg/L, the patient is said to be a nonresponder. If the cortisol levels increase by greater than 9 µg/L, the patient is said to be a responder.
In 2002, Annane examined outcomes in steroid replacement therapy in a placebo-controlled, randomized, double-blinded study in 19 European ICUs. Septic patients with vasopressor-resistant shock (hypotension despite fluid administration and vasopressors) were all given a cosyntropin stimulation test. They were then randomly assigned to receive either a placebo or the combination of 50 mg of IV hydrocortisone every six hours and 50 µg of oral fludrocortisone every day for seven days. The results of the trial showed that cosyntropin nonresponders given steroid replacement had a lower mortality rate than nonresponders given a placebo (53% versus 63%; p = 0.02). They also had had a decrease in total time dependent on vasopressors. Cosyntropin responders had no significant differences in outcome whether they received steroids or placebo. Overall, there were no differences in adverse events such as superinfection, bleeding, and psychiatric disorders in patient receiving placebo or steroids.
The recently published CORTICUS study questioned the use of low-dose steroids in sepsis. The study was conducted with nearly 500 patients as a multicenter, blinded, and placebo-controlled trial. Septic patients with evidence of shock within the previous 72 hours were enrolled and given a cosyntropin stimulation test. They were then randomized to receive either 50 mg of IV hydrocortisone every six hours for five days or a placebo. After 28 days, results showed no statistically significant difference in mortality between the hydrocortisone group and the placebo group in nonresponders (39.2% mortality versus 36.1% mortality; p=0.69). In addition, there were more episodes of superinfection, including new sepsis and septic shock, in the patients who received hydrocortisone.
An important difference between the study by Annane and CORTICUS is that CORTICUS patients were enrolled if their systolic blood pressure was below 90 mm Hg for one hour, regardless of response to vasopressor therapy. This probably enrolled a less severely ill population, as evidenced by lower simplified acute physiology scores (SAPS-II) and lower 28-day mortality rates in CORTICUS’s placebo group (61% in the Annane study versus 32% in the CORTICUS study). A second important difference is that Annane enrolled patients within eight hours of the onset of shock, while CORTICUS allowed for up to 72 hours.
Further studies with larger numbers of patients are required to truly define the role of steroid replacement therapy in septic patients. For now, the 2008 SSC guidelines emphasize the use of intravenous hydrocortisone in adult septic shock patients only if their blood pressure is poorly responsive to fluid resuscitation and vasopressor therapy (as in the patients enrolled by Annane). The guidelines also recommend against routinely using steroids in septic patients without shock, unless warranted by the patient’s history of endocrine or corticosteroid use.
NO. 6: LUNG-PROTECTIVE VENTILATION
Sepsis-induced pulmonary dysfunction is a common source of morbidity and mortality in critically ill patients. Acute lung injury (ALI) and acute respiratory distress syndrome (ARDS) are disorders characterized by diffuse pulmonary parenchymal injury associated with noncardiogenic pulmonary edema. Acute lung injury differs from ARDS only in the degree of hypoxemia (see box below). Both disorders occur with high frequency in patients with sepsis and result in varying degrees of hypoxemic respiratory failure.
 |
Initial studies in the early 1990s by Hickling showed that ventilation using lower tidal volumes reduced mortality in ARDS, and the concept of lung-protective ventilation (LPV) was promoted. It was shown that ARDS patients do not have a homogeneous distribution of infiltration within the pulmonary parenchyma. Overdistension of the remaining normal alveoli was thought to lead to injury. The ARDSNET study was the largest and most successful trial to investigate the use of LPV in patients with ARDS. The study enrolled more than 800 patients in a prospective multicenter design and compared ventilation with tidal volumes of 6 ml/kg and 12 ml/kg using predicted body weight. The results showed that low tidal-volume ventilation was associated with a 9% absolute reduction in mortality when the plateau pressure was kept below 30 cm H2O.
One recent study looked at the time course and risk factors associated with the development of ALI/ARDS in patients with sepsis. Iscimen investigated 160 patients with septic shock and no evidence of lung injury at onset and found 44% developed lung injury at a median of five hours. Risk factors for the development of lung injury included delayed goal-directed resuscitation and delayed administration of antibiotics, emphasizing the time-dependent nature of sepsis treatment.
There are strong recommendations to use reduced tidal volumes for patients with sepsis and evidence of early ALI/ARDS. Practically, when using lower tidal volumes (6 ml/kg of ideal body weight), one must tolerate a degree of hypoventilation and permissive hypercapnea (pH 7.3 to 7.45). Use positive end-expiratory pressure to improve oxygenation and maintain plateau pressures less than 30 cm H2O.
NO. 7: INTENSIVE GLYCEMIC CONTROL
Hyperglycemia is common in the ICU. In the past, it was thought that this stress glycemic response was adaptive. In 2001, Van den Berghe studied a strategy of intensive blood glucose control. This was a prospective, randomized, controlled trial of 1548 mechanically ventilated patients admitted to a single surgical ICU. Patients were randomized to intensive glycemic control (serum blood glucose maintained between 80 and 110 mg/dl) through the use of an insulin infusion protocol, versus conventional care. Results showed a reduction in overall ICU mortality in the group randomized to intensive glycemic control compared to conventional care (4.6% versus 8%, p < 0.04). The greatest reduction in mortality was seen in patients who stayed in the ICU for more than five days (10.6% mortality versus 20.2% mortality) and in patients with a proven septic focus and multiple organ failures.
In 2005, the same researchers repeated the study in a population of medical ICU patients. This was a prospective, randomized, controlled study of 1200 mechanically ventilated patients admitted to a single center’s MICU. Results were different this time, showing that intensive glycemic control did not reduce overall mortality (37.3% in the tightly controlled group versus 40% in the conventional group; p=0.33) and resulted in more cases of hypoglycemia. However, intensive glycemic control did reduce morbidity by preventing newly acquired kidney injury, accelerating weaning from mechanical ventilation, and hastening discharge from both the ICU and hospital. Additionally, although length of stay in the ICU could not be predicted at admission, patients who stayed in the ICU for more than three days and underwent intensive insulin therapy had decreased mortality compared to those whose stays were shorter (43% mortality versus 52.5% mortality; p=0.009).
Why septic patients seem to benefit more from intensive glycemic control with prolonged ICU stays is not clear at this time. Additional large, multicenter studies are needed to reveal the best strategy for glycemic control in sepsis. The 2008 SSC guidelines stress the reduction in morbidity and mortality of intensive glycemic control with longer ICU stays and recommend that patients with severe sepsis and hyperglycemia receive intravenous insulin therapy to a targeted glucose level of less than 150 mg/dl.
NO. 8: RECOMBINANT HUMAN ACTIVATED PROTEIN C
In sepsis, a complex host response to infection occurs that involves significant interplay between inflammatory cytokines and coagulation mediators. In severe sepsis, the end result of these interactions may be diffuse endothelial injury, disseminated intravascular coagulation, multiple organ dysfunction, and death. Activated protein C (APC) is an endogenous mediator in the coagulation cascade that inhibits thrombosis, inflammation, and apoptosis. In 2001, Yan showed that reduced levels of APC are found in most patients with sepsis and are associated with an increased risk of death.
The efficacy of recombinant human APC (rhAPC) in severe sepsis was investigated in the PROWESS study. This was an international, randomized, placebo-controlled study of 1690 patients. Results showed that mortality at 28 days was reduced from 30.8% in the placebo group to 24.7% in the treatment group. Major bleeding was 3.5% in the rhAPC group and 2.0% in the placebo group (p=0.06). Mortality benefit was greatest in patients with higher severities of illness, as reflected in higher APACHE II scores and greater numbers of sepsis-induced organ dysfunctions.
Since the PROWESS study, other trials investigating rhAPC have been conducted. The administration of drotrecogin alfa (activated) in the Early Stage Severe Sepsis Trial showed that rhAPC in less severely ill patients (APACHE II score of less than 25 or single-organ dysfunction) confers no benefit. The ENHANCE study was a global, unblinded, single-arm trial that showed a mortality benefit similar to PROWESS (25.3% at 28 days) but increased rates of bleeding (6.5% at 28 days). The study also showed a greater benefit if rhAPC was started within 24 hours from the first sepsis-induced organ dysfunction (22.9% mortality if started between zero to 24 hours versus 27.4% mortality if started after 24 hours).
 |
Indications for rhAPC use in the emergency department are limited. As demonstrated by the ENHANCE data, there may be a role for early therapy. However, based on the risks of bleeding and cost, rhAPC should probably be considered only after hemodynamic optimization is achieved, appropriate antibiotics are given, and if there is significant delay in admission to the ICU. Contraindications to rhAPC use are listed in the box above.
ATTAINABLE GOAL
Research on the pathophysiology and management of sepsis has made great strides over the past two decades. While many questions still need to be answered and complexities unraveled, one thing is certain: morbidity and mortality can be decreased with timely intervention. Because of the rising incidence of sepsis, the increasing use of the emergency department by at-risk populations, and the increase in time patients spend in the emergency department awaiting hospital admission, emergency medicine practitioners are poised to make a significant difference in the fight against sepsis. By administering appropriate antibiotics in a timely fashion, removing possible sources of infection, matching systemic oxygen delivery to demand, using lung-protective ventilation strategies, and judiciously using corticosteroids and intensive insulin therapy, the goal of reducing mortality in sepsis can be attained.
Editor’s note: Portions of this article are adapted from: Catenacci MH, King K. “Severe Sepsis and Septic Shock: Improving Outcomes in the Emergency Department.” Emergency Medicine Clinics of North America.
Suggested Reading
Acute Respiratory Distress Syndrome Network. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med 342(18):1301, 2000.
Angus D, et al.: Epidemiology of severe sepsis in the United States: Analysis of incidence, outcome, and associated costs of care. Crit Care Med 29(7):1303, 2001.
Annane D, et al.: Effect of treatment with low doses of hydrocortisone and fludrocortisone on mortality in patients with septic shock. JAMA 288(7):862, 2002.
Bernard GR, et al.: Efficacy and safety of recombinant human activated protein C for severe sepsis. N Engl J Med 344(10):699, 2001.
Betsch A, et al.: CT guided percutaneous drainage of intra-abdominal abscesses: APACHE III score stratification of 1-year results. Acute Physiology, Age, Chronic Health Evaluation. Eur Radiol 12(12):2883, 2002.
Catenacci MH and King K: Severe sepsis and septic shock: improving outcomes in the emergency department. Emerg Med Clin North Am 26(3):603, 2008.
Elliott DC, et al.: Necrotizing soft tissue infections. Risk factors for mortality and strategies for management. Ann Surg 224(5):672, 1996.
Ferrer R, et al.: Improvement in process of care and outcome after a multicenter severe sepsis campaign in Spain. JAMA 299(19), 2294, 2008.
Gilbert DN, et al.: The Sanford Guide to Antimicrobial Therapy 2007, 37th ed, Sanford, 2007.
Harbarth S, Garbino J, Pugin J, et al. Inappropriate initial antimicrobial therapy and its effect on survival in a clinical trial of immunomodulating therapy for severe sepsis. Am J Med 115(7):529, 2003.
Ibrahim EH, et al.: The influence of inadequate antimicrobial treatment of bloodstream infections on patient outcomes in the ICU setting. Chest 118(1):146, 2000.
Iscimen R, et al.: Risk factors for the development of acute lung injury in patients with septic shock: an observational cohort study. Crit Care Med 36(5):1518, 2008.
Kumar A, et al.: Duration of hypotension before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock. Crit Care Med 34(6):1589, 2006.
Lambe S, et al.: Trends in the use and capacity of California’s emergency departments, 1990-1999. Ann Emerg Med 39(4):389, 2002.
Larche J, et al.: Improved survival of critically ill cancer patients with septic shock. Intensive Care Med 29(10):1688, 2003.
Leibovici L, et al.: The benefit of appropriate empirical antibiotic treatment in patients with bloodstream infection. J Intern Med 244(5):379, 1998.
Rivers E, et al.: Early goal-directed therapy in severe sepsis and septic shock: a contemporary review of the literature. Curr Opin Anesthesiol 21(2):128, 2008.
Rivers E, et al.: Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med 345(19):1368, 2001.
Sprung CL, et al.: Hydrocortisone therapy for patients with septic shock. N Engl J Med 358(2):111, 2008.
Van den Berghe G, et al.: Intensive insulin therapy in critically ill patients. N Engl J Med 345(19):1359, 2001.
Van den Berghe G, et al.: Intensive insulin therapy in the medical ICU. N Engl J Med 354(5):449, 2006.
Vincent JL, et al.: Drotrecogin alfa(activated) treatment in severe sepsis from the global open-label trial ENHANCE: further evidence for survival and safety and implications for early treatment. Crit Care Med 33(10):2266, 2005.
Wang HE, et al.: National estimates of severe sepsis in United States emergency departments. Crit Care Med 35(8):1928, 2007
Yan SB, et al.: Low levels of protein C are associated with poor outcome in severe sepsis. Chest 120(3):915, 2001. |
|
