CLINICAL CROSSROADS

CONFERENCES WITH PATIENTS AND DOCTORS

CLINICIANS CORNER

Management of Sepsis
A 47-Year-Old Woman With an Indwelling Intravenous Catheter and Sepsis
Derek C. Angus, MD, MPH, FRCP, Discussant DR REYNOLDS: Ms C is a 47-year-old woman with a history of short gut syndrome, dependence on total parenteral nutrition (TPN), and numerous infectious and thrombotic complications of her indwelling catheters who was transferred because of sepsis. Ms C had been discharged 10 days prior to the current admission. At the time of the previous discharge, she was given a planned 2-week antibiotic course for a line infection. Seven days later, she developed nausea and fatigue. The next morning, after a surveillance blood culture was reported positive, she called an ambulance and was transported to the hospital. Ms Cs medical history is significant for Gardner syndrome diagnosed at age 23 years. She was treated with colectomy in 1985 and repeated small bowel resections. She has been dependent on TPN since 1987. She has had numerous bacterial and fungal line infections and catheterrelated venous occlusions. Her current left femoral catheter is positioned in her last remaining access site. If she is unable to keep her left groin line open, she will require canalization of the portal system for TPN infusion. Ms Cs father and 6 siblings also have Gardner syndrome. Ms C previously worked as a nanny but is currently living with her mother, who helps provide her care along with personal care assistants several days a week. Ms C does not smoke, drink alcohol, or use drugs. She enjoys gardening and seeing her nieces and nephews. Ms Cs medications include clopidogrel, a fentanyl patch, ondansetron, lorazepam, and linezolid. She has numerous allergies and adverse reactions to medications. On arrival by ambulance, Ms C was vomiting. She reported chills, lower abdominal pain, hip pain, and myalgias. On physical examination, her blood pressure was 69/45 mm Hg; heart rate, 102/min; temperature, 37.2C; respiratory rate, 15/min; and oxygen saturation, 100% on nasal cannulae. She was thin and appeared chronically ill. Her jugular venous pressure was difficult to assess. Her cardiac and pulmonary examination results were normal. Her abdomen was scaphoid, with numerous surgical scars, and an
CME available online at www.jamaarchivescme.com and questions on p 1494.
2011 American Medical Association. All rights reserved.

Severe sepsis is the term used to describe the host response to infection when complicated by acute organ dysfunction. Severe sepsis occurs in more than 750 000 individuals in the United States each year, with a hospital mortality of about 30%. Although the classic presentation is of florid shock with frank hypotension, fever, and elevated white blood cell count, many patients can present with cryptogenic shock (shock without hypotension) with more subtle signs of vital organ compromise. Using the case of Ms C, a 47-year-old woman with short gut syndrome and an indwelling intravenous catheter who developed an episode of severe sepsis secondary to a central line infection, treatment of sepsis is discussed. Management consists of prompt intervention with broad-spectrum antibiotics and fluid resuscitation, even in the absence of hypotension, and institution of a variety of strategies in the emergency setting to prevent development or worsening of vital organ dysfunction. Although advances in understanding the host immune response have fueled considerable interest in immunomodulatory therapy, the role of such agents in clinical practice remains limited and controversial.
JAMA. 2011;305(14):1469-1477 www.jama.com

ostomy bag was in place without visible skin breakdown. She had moderate tenderness to palpation across her abdomen but no rebound. The tunneled Hickman catheter in her left groin had no tenderness or exudates. Her extremities were warm. She responded to questions appropriately. Abnormal laboratory values included an elevated white blood cell count of 18 700/L; bicarbonate, 16 mEq/L; creThe conference on which this article is based took place at the Medicine Grand Rounds at Beth Israel Deaconess Medical Center, Boston, Massachusetts, on April 9, 2009. Author Affiliation: The CRISMA Center, Department of Critical Care Medicine, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania. Corresponding Author: Derek C. Angus, MD, MPH, FRCP, the CRISMA Center, Department of Critical Care Medicine, University of Pittsburgh School of Medicine, 614 Scaife Hall, 3550 Terrace St, Pittsburgh, PA 15261 (angusdc@upmc.edu). Clinical Crossroads at Beth Israel Deaconess Medical Center is produced and edited by Risa B. Burns, MD, series editor; Tom Delbanco, MD, Howard Libman, MD, Eileen E. Reynolds, MD, Marc Schermerhorn, MD, Amy N. Ship, MD, and Anjala V. Tess, MD. Clinical Crossroads Section Editor: Margaret A. Winker, MD, Deputy and Online Editor. JAMA, April 13, 2011Vol 305, No. 14 1469

Downloaded from jama.ama-assn.org at Capes Consortia on April 13, 2011

CLINICAL CROSSROADS

atinine, 2.0 mg/dL (baseline, 0.5 mg/dL); aspartate aminotransferase, 76 U/L (baseline, 16 U/L); alanine aminotransferase, 101 U/L (baseline, 31 U/L); and lactate, 31.5 mg/dL (normal, 0-18 mg/dL). Cortisol, drawn on her last admission, was 35.8 g/dL (normal, 2-20 g/dL). Her urinalysis was normal and her chest radiograph and electrocardiogram were without change. Blood cultures drawn at home had grown Klebsiella pneumoniae. MS C: HER VIEW I started getting sick on Monday afternoon. I woke up from a nap and was very nauseated. I thought that maybe if I fell back to sleep, it was just like an afternoon nausea kind of thing, like a bug. Then when I woke up a couple of hours later, I was fine. I didnt have a temperature and I didnt have the chills or anything that usually runs along with being sick. Tuesday afternoon, my nurse was coming in to draw laboratories and decided, since I did show something on Monday, that she would do cultures. She drew cultures on Tuesday afternoon and I was pretty much fine on Tuesday. Then, Wednesday morning, I woke up and I was terribly sick. I had the rigors. The bed was shaking. My muscles and joints from my hips to my feet were constantly hurting and I was vomiting; I knew it wasnt just a little bug. My body ached with deep chills. When I got to the hospital, people asked me questions, but I didnt want to answer them because I was just too much in pain. I figured that they could get all the other information off of my last admission, when I was here for my infection, or call somebody that was on my list. On a scale of 1 to 10 for pain, it was a 15. I would say sepsis is like having the flu 100 times. It can affect your muscles. It can affect your joints. It can affect anything. I just said, Let me sleep. Knock me out and let me sleep. Dont wake me up until my white count and everything else is back to normal. Iknowmybloodpressurewasdown.Sotheyputmeonmedication to bring my blood pressure up. They also gave me widespectrum antibiotics. Because they werent sure what the bug was, they gave me this wide spectrum to kill just about anything. After 3 days, they finally reduced my medication to 1 antibiotic, and I think that is really when I saw the light and saw happiness. It was much less painful. I could converse with the nurses who were in the room and the medical staff who were coming in to ask me questions. But up until then I just said, Let me sleep. There are some things I do not want done for me. I dont want to be on a breathing tube. If its a medical mistake of too much pain medication, too much antibiotic, too much of this or whatever, and its the nurses or the doctors mistake, then they can bring me back. But if they come walking in the room and Im blue, then just let me be. I just have been through so much in the last 30 years. I would ask the expert if there is anybody doing any more researchthere are not many antibiotics out there now. Ive taken 3 of the antibiotic bigwigs, but Im beginning to develop a tolerance of them. For example, I cant take vanco1470 JAMA, April 13, 2011Vol 305, No. 14

mycin anymore, and that is a bigwig. So I want to know if there is anybody out there researching other, stronger antibiotics. AT THE CROSSROADS: QUESTIONS FOR DR ANGUS What are the definition and incidence of sepsis? What is the pathophysiology of sepsis? How often is an etiology found? What are the treatment options for sepsis and what are the outcome data for each? Of the available treatment options, which should be started empirically at presentation? How is the standard of care for sepsis different now compared with 5 years ago? What are the outcomes of sepsis and what is the mortality rate? Do patients return to their premorbid baseline? What indicates futility and how can intensive care unit (ICU) teams determine when to stop escalating care? What do you suggest for Ms C?
Definition and Incidence of Sepsis and Related Syndromes

DR ANGUS: Sepsis and septicemia are old, broad terms describing patients who are sick secondary to an infection, typically of bacterial origin. Since 1992, these terms have been replaced with the more specific terms of severe sepsis and septic shock.1,2 Severe sepsis is the syndrome defined as infection complicated by acute organ dysfunction, such as the development of altered mental status, acute lung injury, acute kidney injury, or ileus. When the affected organ system is the cardiovascular system, as evidenced by hypotension or occult hypoperfusion, such as elevated serum lactate level, the syndrome is called septic shock. Shock is termed as refractory when it is unresponsive to fluid resuscitation and maintenance of normal blood pressure requires the use of vasopressors despite adequate fluid resuscitation. Ms C, who has acute organ dysfunction and an elevated lactate level, would be classified as having septic shock. In preparing my review, I searched PubMed using the terms sepsis, sepsis clinical trials, and septic shock with no restrictions on year. When discussing current treatment guidelines, I also searched under specific treatment and management topics, in particular to identify clinical trials published since the latest international guidelines.3 In the United States, more than 750 000 individuals develop severe sepsis each year.4 Typically, about 2 cases per 100 hospital admissions develop severe sepsis, and about 10% of all ICU patients have severe sepsis on admission or during their ICU stay.5,6 The incidence around the world varies significantly, in part because it is not typically measured outside ICUs, and the number of ICU beds per 100 000 population varies considerably.7 Thus, the incidence might be better described as the treated incidence. In countries with fewer ICU beds, a third or more of ICU patients may have severe sepsis.
Pathophysiology

Sepsis has a complex pathophysiology. In response to microbial challenge, the host mounts an innate immune response consisting of both humoral and cellular compo2011 American Medical Association. All rights reserved.

Downloaded from jama.ama-assn.org at Capes Consortia on April 13, 2011

CLINICAL CROSSROADS

nents.8,9 This innate immune response is an old and necessary response that is remarkably well-preserved across species. However, in sepsis, the response goes awry (FIGURE). Typically, the inflammatory response is overly exuberant and characterized by high systemic levels of inflammatory proteins and broad white blood cell activation. This overly exuberant response may persist or may be followed by a relative immune paralysis.10 An exuberant response promotes vascular leak, vasodilation, and absolute or relative blood volume depletion. Untreated, this depletion leads to organ hypoperfusion, with downstream ischemia and ischemia reperfusion. Activation of secondary pathways such as coagulation and endothelial injury exacerbate the situation. Relative immune paralysis is characterized by anergy (a de-

creased ability to mount further immune responses), with a consequent increased risk of secondary infections. The exact course of this pathophysiologic response is both heterogeneous and hard to predict. Some patients mount very aggressive inflammatory responses, while others mount a far milder response or develop immunoparesis, and neither the source of infection nor clinical characteristics of the patient appear to be strongly predictive of the nature and scope of the underlying host response.11 One possible explanation is that genetic determinants govern the host response. In a Danish study of adoptees, Srensen et al12 demonstrated that premature death due to infection of the biological parent was strongly predictive of premature death in the adoptee, providing compelling evidence of a strong

Figure. Immune Response, Clinical Course, and Management of Infection and Severe Sepsis
Baseline health with no symptoms of infection Normal immune homeostasis Microbial challenge Fever, tachycardia, increased white blood cell count, general malaise Early appropriate immune response 1 Symptom relief; appropriate antibiotics; control of source infection as indicated
Clinical features Exaggerated immune response Normal immune response Immunoparesis

Management

Uncomplicated infection; no severe sepsis Rapid immune and inflammatory response followed by quick resolution Continue initial treatment

Early severe sepsis with evidence of compromised vital organ function, such as confusion and agitation; impaired pulmonary gas exchange; hemodynamic instability; and decreased urine output Dysregulation of immune response with aberrant systemic inflammation and with coagulopathy, and altered blood flow, compromising normal organ function Continue initial treatment Commence fluid resuscitation Admit to a monitored environment that permits rapid screening and treatment of any worsening organ function

Full recovery

Clinical challenges
1

Extremely difficult to differentiate between patients who will have a normal, brief acute immune response and patients who will develop severe sepsis. No immunomodulatory therapies are available that specifically target the unwanted components of the immune and inflammatory response. No clinical parameters allow determination of whether the underlying immune response remains exaggerated, has become excessively downregulated, or has returned to normal immune homeostasis. No specific therapies are available to promote recovery from organ dysfunction. Pathophysiology underlying residual sequelae is unknown; therefore, specific management options are limited.

Established severe sepsis with multiorgan dysfunction (coma, acute lung injury, acute kidney injury, septic shock) Continued exaggerated immune response or Return to normal immune homeostasis or Immunoparesis

3 4

Complete initial resuscitation Institute and maintain organ support as necessary Consider corticosteroids for vasopressor-resistant shock Maintain close monitoring for further deterioration Provide maintenance feeding and fluids Wean organ support when possible Monitor for potential complications, such as secondary infections

Full recovery Normal termination of inflammatory response

Partial recovery with sequelae (eg, neurocognitive dysfunction, muscle weakness, residual organ dysfunction, mood disorders) Immune function in patients with partial recovery is unknown and could include chronic low-grade inflammation, normal immune homeostasis, or mild immunosuppression. 5 Monitor and treat sequelae (eg, neurocognitive and physical rehabilitation, psychiatric care for mood disorders).

Death Possible causes include ongoing inflammatory storm, immunosuppression with secondary infection, overwhelming burden of organ damage despite improvement in immune function, and other complications.

2011 American Medical Association. All rights reserved.

JAMA, April 13, 2011Vol 305, No. 14 1471

Downloaded from jama.ama-assn.org at Capes Consortia on April 13, 2011

CLINICAL CROSSROADS

hereditary component to death due to infection. In addition, many studies have reported that genetic variants in genes encoding for proteins in the innate immune response are associated with either susceptibility to or adverse outcome from severe sepsis and septic shock.13,14 However, these association studies have not always been replicated and it seems likely that genetic variation is multifactorial and entwined in complex gene-environment interactions.
Etiology

Nearly all microbes can cause sepsis. In immunocompetent hosts, typical causes are gram-negative and grampositive bacteria, such as Escherichia coli, Pseudomonas species, Staphylococcus aureus, Streptococcus pyogenes, and Streptococcus pneumoniae.9,15 In immunocompromised hosts, sepsis can develop secondary to less pathogenic, more opportunistic organisms. Of note, severe sepsis can often progress despite appropriate treatment with antibiotics, emphasizing that the pathophysiology of sepsis in many respects is a problem of the host and not of the invading pathogen. The most frequent type of infection that leads to severe sepsis is pneumonia (44%), accounting for up to half of cases in most series, followed by primary bacteremia (17%), genitourinary infection (9%), abdominal infection (9%), and, less commonly, skin infections, meningitis, etc.4 Even in prospective series and clinical trials with careful attempts to identify a causative organism, as many as a third of patients are culture-negative.16-18 Although bacteria are the classic causes of severe sepsis, fungi are an important cause and viruses are an increasing focus of attention, most notably influenza A(H1N1).19-21 Ms C developed septic shock secondary to a central line infection. Patients with indwelling catheters are at longterm elevated risk of systemic infection, both because the catheter represents a foreign body violating a normally sterile space and because patients who require these catheters usually have chronic illness and may be otherwise debilitated, and thus may have a dampened immune response. Although there are typical causes of central line infections, such as coagulase-negative staphylococci, it is important to note that many other organisms, including gram-negative pathogens, can be responsible, especially in settings such as that of Ms C, with a long-term indwelling line.
Treatment

Guidelines for the treatment of severe sepsis and septic shock have been developed under the aegis of the Surviving Sepsis Campaign, a collaboration of multiple infectious disease and critical care professional societies. The latest guidelines, published in 2008, were developed by an international body of sepsis experts using the GRADE system to summarize and rank both the strength of the recommendations and the strength of evidence supporting each recommendation.3 Many of the key studies underpinning the guidelines are summarized in the TABLE.
1472 JAMA, April 13, 2011Vol 305, No. 14

The first priorities for severe sepsis are to treat the underlying infection with appropriate antibiotics and drainage of pus, resuscitate any overt or cryptic shock with intravenous fluids and pressors if necessary, and institute emergency supportive care for any life-threatening organ dysfunction, such as initiation of mechanical ventilation for severe respiratory failure.2,3,22 Failure to institute any of these steps in a timely manner is frequently fatal. Acute organ dysfunction should usually precipitate emergency admission to an ICU; Ms C was appropriately triaged to the ICU. The exact strategy by which a patient should be resuscitated remains controversial. One study reported a marked reduction in mortality from 46% to 31% with strict use of a timed protocol for the first 6 hours, with monitoring of central venous oxygen levels and instructions for the use of fluids, pressors, inotropes, packed red blood cell transfusions, and mechanical ventilation.23 However, this protocol is invasive, the study was conducted at a single center, and the uptake of the protocol has been variable.24,25 There are currently 3 large government-funded trials testing this resuscitation protocol in the United States, Europe and Australasia.26-28 Nevertheless, there is broad agreement that shock can often be cryptic (no hypotension), and careful attention to the rapid resolution of shock, as evidenced by improvement in signs of organ hypoperfusion, remains a cornerstone of good care. Current appropriate standard of care include initial resuscitation of about 20 mL/kg of crystalloid or colloid intravenous fluid, followed by titration of subsequent fluids depending on ongoing signs of hypoperfusion, such as persistent elevated lactate level.29 If hypotension is not rapidly responsive to fluids, then one would typically start a vasopressor to maintain systolic blood pressure above 90 mm Hg or mean blood pressure above 60 mm Hg. Of the numerous vasopressors available, norepinephrine is probably the best first-line choice based on its low cost, its familiarity to clinicians, and its superior safety profile to dopamine as shown in a multicenter trial.30 In addition, broadspectrum antibiotics should be started as soon as possible, ideally within 1 hour of a preliminary diagnosis.31 After these initial management steps, if it is clear that the patient has continuing acute organ dysfunction, the next considerationiswhethertoadministeranimmunomodulatingtherapy. Several clinical trials have tested the efficacy of agents targeting specific receptors and pathways in the host response, such asantitumornecrosisfactormonoclonalantibodies.32 Although the pooled odds ratio of all specific anti-inflammatory agents suggested a slight benefit (odds ratio of 28-day mortality, 0.91; 95% confidence interval [CI], 0.82-0.99; P=.03), the pooled odds for each individual agent was not significant.32 However, 2 therapies have entered clinical practice: corticosteroids and activated protein C. Corticosteroids, typically administered as 50 to 100 mg of hydrocortisone 3 or 4 times daily for 1 week, appeared to be beneficial in several small trials and 1 large French trial.33 Steroids are pleiotropic drugs, and the exact mechanism by which they may help in sepsis is unclear. However, the 2 most broadly accepted
2011 American Medical Association. All rights reserved.

Downloaded from jama.ama-assn.org at Capes Consortia on April 13, 2011

CLINICAL CROSSROADS

Table. Key Studies Informing Current Practice Guidelines for Severe Sepsis
Source Resuscitation Rivers et al, 200123 Comparison Design Population Sample Size 263 Outcome Primary Effect Adverse Effects Findings

Jones et al, 201029

De Backer et al 201030

EGDT guided by Unblinded Adults singlepresenting continuous center to ED central RCT with septic venous O2 vs usual care shock EGDT guided by Unblinded Adults serial lactate 3-center presenting vs EGDT RCT to ED guided by with septic continuous shock central venous O2 Dopamine vs Multicenter Adult ICU norepinRCT with shock ephrine

Hospital 30.5% vs 46.5%; RR None reported EGDT superior to mortality with EGDT, 0.58; usual care 95% CI, 0.38-0.87; P = .009 300 Hospital 17% (11%-24%) vs None reported Serial lactate mortality 23% (17%-30%); equivalent to mean difference, central venous 6%; 95% CI, 3% O2 monitoring to 15%; noninferiority threshold 1679 (1044 28-d 52.5% vs 48.5%; OR Higher frequency Norepinephrine with septic mortality with dopamine, of arrhythmias and dopamine shock) 1.17; 95% CI, with appear similarly 0.97-1.42; P = .10 dopamine efficacious but norepinephrine has a more favorable adverse effect profile 2154 Hospital Adjusted OR, 1.12 per None reported mortality 1-h delay; 95% CI, 1.1-1.14; P .001 Every hour of delay before initiating antibiotics is associated with higher mortality

Early antibiotics Kumar et al, 200622

Timing from onset of shock to initiation of appropriate antibiotics

Multicenter Adult ICU retrowith septic spective shock cohort study

Corticosteroids Annane et al 200233

Low-dose Multicenter Adult ICU corticosteroids blinded with septic vs placebo RCT shock

Sprung et al, 200834

Low-dose Multicenter Adult ICU corticosteroids blinded with septic vs placebo RCT shock

Annane et al, 200935

Corticosteroids vs placebo

MetaAdult ICU analysis with septic shock

300 (229 with 28-d HR, 0.67; 95% CI, No differences Steroids improve relative mortality 0.47-0.95; P = .02 reported mortality after adrenal in those with septic shock insufficiency) relative adrenal insufficiency 499 28-d 34.3% (95% CI, Superinfection, Steroids do not mortality 28.3%-40.2%) vs hyperglycemia, improve 31.5% (95% CI, hypernatremia mortality after 25.6%-37.3%); septic shock P = .51 2138 in 17 28-d Any dose: RR, 0.84; Hyperglycemia, In aggregate, RCTs (any mortality 95% CI, hypernatremia steroids dose); 0.71-1.00; improve 1228 in 12 P = .05; low dose: outcome after RCTs (low RR, 0.84; 95% CI, septic shock dose) 0.72-0.97; P = .02 28-d 30.8% vs 24.7%; RR Bleeding mortality reduction, 19.4%; 95% CI, 6.6%-30.5%; P = .005 28-d 18.5% vs 17.0%; RR, Bleeding mortality 1.08; 95% CI, 0.92-1.28; P = .34 aPC improves outcome in severe sepsis

Activated protein C aPC vs placebo Bernard et al, 200118

Multicenter Adult ICU 1690 blinded with severe RCT sepsis

Abraham et al, 200538

aPC vs placebo

Multicenter Adult ICU 2640 blinded with severe RCT sepsis and low risk of death Multicenter RCT Adult ICU with ARDS 861

aPC does not improve outcome in lower-severity severe sepsis

Organ support ARDS Network, 200040

Low vs high tidal volume

60-d 31% vs 39.8%; mean hospital difference, 8.8%; mortality 95% CI, 2.4% to 15.3%; P = .007

Bellomo et al, 200941 Girard et al, 200845

Brunkhorst et al, 200849

Low- vs Multicenter high-intensity RCT dialysis Daily combined Multicenter cessation of RCT sedation and SBT vs SBT alone Intensive insulin Multicenter therapy vs RCT usual care

Adult ICU 1508 with acute renal failure Adult ICU on 336 mechanical ventilation

Adult ICU

537

Low tidal volume superior to high tidal volume in ARDS 90-d 44.7% vs 44.7%; OR, HypoNo benefit to mortality 1.0; 95% CI, phosphatemia higher-intensity 0.81-1.33; P = .99 dialysis Ventilator14.7 vs 11.6 d; mean Self-extubation SBT more effective free difference, 3.1 d; when days 95% CI, 0.7-5.6 d; combined with P = .02 sedation cessation 28-d 24.7% vs 26.0%; Hypoglycemia Intensive insulin mortality P = .74 therapy worsens outcome in sepsis

None reported

Abbreviations: aPC, activated protein C; ARDS, acute respiratory distress syndrome; CI, confidence interval; ED, emergency department; EGDT, early goal-directed therapy; HR, hazard ratio; ICU, intensive care unit; OR, odds ratio; RR, relative risk; RCT, randomized clinical trial; SBT, spontaneous breathing trial.

2011 American Medical Association. All rights reserved.

JAMA, April 13, 2011Vol 305, No. 14 1473

Downloaded from jama.ama-assn.org at Capes Consortia on April 13, 2011

CLINICAL CROSSROADS

roles are promotion of hemodynamic stability and dampening of excessive inflammation. The French study was based on the hypothesis that steroids would be particularly helpful when the adrenal axis was relatively suppressed, as evidenced by a rise of 9 g/dL or less in response to an corticotropin stimulation test. Nevertheless, a follow-up trial in Europe showed no benefit, both overall and in the subset of patients with relative insufficiency.34 A meta-analysis suggested a benefit of steroid treatment (risk ratio for 28-day mortality for treated vs control patients in randomized trials, 0.84; 95% CI, 0.71-1.00),35 but their use is controversial.36 Thus, the latest international guidelines caution against broad use, given the well-known adverse effects,3 but do recommend institution of steroids when hypotension cannot be reversed despite adequate fluids and high doses of vasopressors. Current guidelines do not tie steroid treatment to an corticotropin stimulation test.21 If a test is performed, the steroids could presumably be delayed by 30 to 60 minutes until the test is executed, but there is no rationale for delaying steroids further until the results of the test are available. If steroids are given, the evidence supports a 7-day course, but rapid clinical recovery and an intact adrenal axis may be reasonable arguments for earlier termination, especially if there are contraindications for their use. Use of steroids for Ms C would have been a difficult decision. She is weak and at risk of recurrent infection, risks that could be exacerbated by steroid myopathy or immunosuppression. Her baseline cortisol level was high, but it is not known if she had relative insufficiency and, as discussed herein, only the French study supports the notion that steroids should be restricted to this subset of patients. If her blood pressure remained extremely low despite adequate resuscitation, I probably would have started steroids despite some relative contraindications. However, my hope would be to treat Ms C without steroids if possible. Activated protein C is a naturally occurring protein with pronounced anticoagulant properties but also with immunomodulatory effects. One large trial of a recombinant version showed a significant reduction in mortality for patients with severe sepsis (28-day mortality for control vs treated groups, 30.8% vs 24.7%; relative risk reduction, 19.4%; 95% CI, 6.6%30.5%; P=.005), leading to approval by most national regulatory authorities including the US Food and Drug Administration for use in severe sepsis, especially when severity of illness is high.18 However, bleeding is an important adverse effect and follow-up studies in less sick patients and in children showed no benefit, raising controversy over its use.37,38 An international trial of recombinant activated protein C for treatment of septic shock in ongoing, and the results of this trial will hopefully provide clearer recommendations for use.39 The other key tenets of severe sepsis care relate to the use of organ support strategies. Acute lung injury should be managed with mechanical ventilation, and there is broad agreement that ventilation should be provided using a low-tidalvolume strategy to minimize alveolar overdistension.3,40
1474 JAMA, April 13, 2011Vol 305, No. 14

Optimal management of acute kidney injury is controversial, with no clear evidence on when to institute dialysis, what dose to use, and whether intermittent dialysis is better or worse than continuous methods.41-43 Altered mental status is common and is potentially exacerbated by the use of certain sedative agents. Therefore, a strategy to minimize the use of sedatives, such as daily awakening, is preferable.44 Such a strategy also appears to facilitate weaning from mechanical ventilation.45 Optimal nutrition also remains unclear. Frank starvation is not desirable, and undue reliance on parenteral nutrition is probably dangerous due to increased risk of secondary infection and liver damage. Thus, use of enteral feeding is encouraged in patients (unlike Ms C) who are able to be fed enterally. However, the benefits of early feeding, the exact caloric requirements, and the value of additives, such as glutamine or arginine, all remain speculative and controversial.46,47 Patients frequently develop hyperglycemia, and there was great enthusiasm several years ago for tight glucose control with insulin infusions.48 However, subsequent trials have demonstrated that tight control can in fact be harmful in sepsis, and current recommendations promote moderate control (eg, 150 mg/dL).3,49,50
Outcome

Severe sepsis, untreated, is fatal. However, in modern ICUs, with adequate resuscitation and organ support and appropriate use of antibiotics, hospital mortality can reasonably be expected to be 25% to 35%.4,6 Although there have been few broad, explicit breakthroughs in treatment over recent years, most series are reporting lower mortality rates than previously. Indeed, the mortality rate was often 50% or higher despite intensive care in series from 15 to 20 years ago.51 The exact reasons for improved outcomes are unclear but probably attributable to better awareness of the need for prompt resuscitation and initiation of appropriate antibiotics and to improvements in acute organ support, such as the use of low-tidal-volume ventilatory strategies. Unfortunately, less is known of the fate of those who are never referred to an ICU, and it is likely that many millions may fare far more poorly, especially in developing nations. Even with modern intensive care, survival to hospital discharge does not guarantee full recovery. There is increasing awareness that survivors of severe sepsis remain at considerable risk of late mortality and morbidity.52-55 The exact causes are unclear but include residual vital organ compromise; the debilitating effects of prolonged bed rest and immobilization; recurrent infection, perhaps due to impaired immune function; and precipitation of preexisting chronic disease.56 As short-term mortality improves, researchers are focusing increasingly on strategies to improve the chances of long-term recovery.
Futility and De-escalation of Care

Although the initial focus of care is on prevention, treatment, and resolution, severe sepsis still carries one of the
2011 American Medical Association. All rights reserved.

Downloaded from jama.ama-assn.org at Capes Consortia on April 13, 2011

CLINICAL CROSSROADS

highest mortality rates of all diseases, and thus often requires palliation and end-of-life care. In some instances, the initial shock is so severe that a patient dies within hours despite heroic resuscitation efforts. In other instances, severe sepsis arises in a patient already laden with extremely serious underlying disease, in which case the patients preferences should be determined as soon as possible and care titrated appropriately. In other instances, a patient is successfully resuscitated but then languishes in the ICU with multiple organ system failure. This third instance is perhaps most difficult. Often, after several days or even weeks of supportive care, the patient will gradually recover, but the illness trajectory cannot be predicted with certainty. Accurate prognostication of which patients with multisystem failure will recover and which patients will succumb would be very valuable, but unfortunately there are no reliable tools. Certainly, if a patient develops recurrent episodes of shock refractory to treatment, with rising lactate level, or develops a devastating secondary complication, such as massive intracranial hemorrhage secondary to disseminated intravascular coagulation, then the prognosis for meaningful recovery is poor. Otherwise, however, recovery is hard to predict, and patients who die in these situations often die simply because the physicians and families decide together that prolonged life support would be against the patients wishes and therefore withdraw support.57 Ms C is in a unique situation in that she has had repeated episodes of sepsis and has recovered to hospital discharge from each of her previous episodes. However, with each episode, she is likely to become increasingly debilitated and, absent an opportunity to fully rehabilitate, her chance of dying would therefore increase. In addition, patients are likely to have strong preferences about repeated episodes of aggressive resuscitation and intensive care. As noted above, Ms C has already expressed that she does not want to be intubated again unless the cause is iatrogenic and likely reversible, such as a sedation overdose. Future care planning should incorporate this preference. RECOMMENDATIONS FOR MS C Ms C presented with a short history of chills, fever, myalgia, and malaise. She has a chronic indwelling venous catheter and positive blood cultures. At presentation, she was frankly hypotensive with elevated blood lactate level and acidosis despite fluid resuscitation at the outlying hospital. Therefore, she has classic features of septic shock, and the likely culprit is a line infection. My approach to her care generally reflects principles espoused in the recent Surviving Sepsis Campaigns guidelines.3 Initial management includes further resuscitation with fluids, vasopressors, and admission to the ICU, and these steps were all executed appropriately. She also began antibiotic treatment titrated to her positive blood cultures, which was correct. (She was initially given cefepime, daptomycin, and metronidazole. Metronidazole was stopped after it was clear that a gram-negative rod was growing, and daptomycin
2011 American Medical Association. All rights reserved.

was stopped after sensitivities were available. She switched to once-daily ceftriaxone at discharge about 1 week after admission.) The other key feature initially is to conduct a thorough clinical examination and initiate tests such as chest x-ray to rule out other potential sources of infection. Assuming no other source was found, a working diagnosis of central line associated bloodstream infection is appropriate. Often, management of a central lineassociated bloodstream infection involves removal of the central line. In this instance, however, removal of this last line could have very serious consequences. It was tunneled and could therefore not practically be rewired. Instead, Ms C would then need portal venous access for future TPN. Thus, if she rapidly stabilized with resuscitation, I might have tried to nurse her through the infection without removal of the line, keeping an extremely close eye on her vital organ function in the ICU over the first few hours and days. There are other considerations specific to recurrent line infections. First, the patient may be infected by organisms resistant to traditional antibiotics, necessitating more tailored selection. Second, one must consider concomitant venous thrombosis, which may require imaging studies and institution of anticoagulant therapy or mechanical intervention. Third, depending on the organism, some patients also have associated endocarditis, which, if suspected, should also be investigated with additional imaging. Aside from line management, another controversial issue is the use of steroids. Ms C has septic shock with an adequate baseline cortisol level. I would have liked to have known whether her cortisol level increased secondary to corticotropin. If it did not, and if hypotension persisted despite fluids and vasopressors, I would consider administering hydrocortisone. The evidence for such a strategy is not clear-cut, and I do not use steroids routinely in severe sepsis. However, I equally would consider its use in preference to having a patient die secondary to intractable hypotension. Although Ms C had profound hypotension and shock at presentation, she improved quickly with rapid intervention. Following her recovery, the major challenges shift to rehabilitation and her long-term horizon. Her long-term dependence on TPN puts her at risk of recurrent bouts of sepsis, and it is therefore important to help Ms C address and express her preferences regarding aggressive care in the future. QUESTIONS AND DISCUSSION QUESTION: What is the future of targeted therapies? Do you think that a lot of the negative trials in the past have resulted because we have focused on the innate immune response? Do you think that the future is in highly stratified trials, like the PROWESS Shock trial?39 DR ANGUS: There are 2 ways to think about targeted therapy. One is using a narrow-target therapy, blocking a particular pathway, and the other is targeting patients, selecting only a specific subset of patients. The problem is that neither premise is a proven strategy in human sepsis. In the first premise, there
JAMA, April 13, 2011Vol 305, No. 14 1475

Downloaded from jama.ama-assn.org at Capes Consortia on April 13, 2011

CLINICAL CROSSROADS

is great redundancy in the innate immune response, especially by the time the patient is manifestly sick, and so it may be unlikely that a therapy targeting one specific pathway can have any meaningful effect on patient-centered outcomes.11 In the second premise, we can narrow the selection criteria, but if we narrow without knowing what we are targeting, we may narrow without narrowing in. In other words, we may miss the population of interest altogether.58 QUESTION: In the Spanish article published in 200859 that showed improvement in outcome with protocolized use of a bundle of severe sepsis treatments, how can one separate whether the improvement was just because protocolizing care makes a big impact from whether there was a specific component of care that led to benefit? How do you as a scientist separate those? Is it the actual interventions or is it just attention to detail of some sort that makes a difference? DR ANGUS: This question is a classic criticism of all multicomponent interventions, which are typical in many health services research studies. There are drawbacks to both approaches. If you test one component at a time, you may never find an overall difference. If you test a package or suite of activities, you may show benefit but you cannot work out why. A strong reason to favor testing multiple strategies is the taking your best shot argument. If you test an approach most likely to succeed and show a benefit, you can establish that a protocolized approach with a multicomponent bundle is superior to unprotocolized usual care. This information can generate a change in background care, and future trials can start parsing out which components of the bundle are most necessary, particularly testing the value of the more difficult or expensive components. QUESTION: Going back to the patient, I have a number of patients who have short gut syndrome or a requirement for chronic TPN. They are guaranteed to get sepsis again. And these indwelling lines become infected. Each time the patients come in, they are slightly worse, with worse kidney function, etc. What can you suggest to prevent the sepsis episodes in the future? DR ANGUS: There have been great efforts to control central line infections.60 However, certain patients, such as Ms C, remain susceptible because of their many risk factors, and many of the usual protocols may be of limited effectiveness. Nevertheless, the classic principles hold true, such as adherence to good catheter care in the first place and using the ports as infrequently as possible. In addition, careful attention to minimize exacerbation of her chronically frail state (through attention to nutritional status, rehabilitation, etc) to keep her as healthy as possible are all key. Despite these efforts, you are right that Ms C will likely continue to have recurrent bouts of severe sepsis, which raises the issue of futility and her preferences for continued aggressive care, especially since her underlying health status will likely decline. For example, every time she has acute renal failure, her creatinine level may bounce back somewhat, but her underlying renal function may not return to baseline.
1476 JAMA, April 13, 2011Vol 305, No. 14

Ms C, when interviewed about these recurrent bouts, stated: Well, if it is a doctors mistake and it is just short term, then I dont mind. But I dont want to live on a breathing tube. This is a classic response that we probably all have. However, it describes a relatively rare situationthat of a permanent vegetative state. The far more likely scenario is that the requirement for life support would be shorter days or weeks. In that case, would one still want support? Clearly, since Ms C has recurrent bouts and is coherent in between events, there is an opportunity to engage her in discussion about her end-of-life preferences. One strategy might be to avoid talking about whether life support should be initiated since, at the time of initiation, it may be impossible to tell whether the duration of required support will be short or very long. Instead, the discussion could be reframed around limitations on a trial of organ support. One might ask, for example, If you were to get sick again, and we agreed to put you on life support, during which time you may be unable to participate in further discussion, would you want us to discontinue life support if we could not wean you within a certain time period, and if so, how long? The key point is to make sure an opportunity to have these discussions is not missed and to make sure that the discussions involve education in both directions of all salient points.
Conflict of Interest Disclosures: Dr Angus has completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and reports that he has served as investigator for Eisai Inc in a long-term follow-up of a phase 3 trial of an antiTLR4 agent in severe sepsis; that he has been a data and safety monitoring board member for Eli Lilly in a multicenter trial of aPC for septic shock; and that he has served as a consultant for Eisai Inc (anti-TLR4 therapy for severe sepsis), ALung (extracorporeal gas exchange for respiratory disorders), ZD Associates (sepsis and acute illness), and Idaho Technology (sepsis biomarkers). Disclaimer: Dr Angus, Contributing Editor with JAMA, had no role in the evaluation of or decision to publish this article. Additional Contributions: We thank the patient for sharing her story and for providing permission to publish it. REFERENCES: 1. Bone RC, Balk RA, Cerra FB, et al; ACCP/SCCM Consensus Conference Committee. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. Chest. 2009;136(5)(suppl):e28. 2. Levy MM, Fink MP, Marshall JC, et al; SCCM/ESICM/ACCP/ATS/SIS. 2001 SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference. Crit Care Med. 2003;31(4):1250-1256. 3. Dellinger RP, Levy MM, Carlet JM, et al; International Surviving Sepsis Campaign Guidelines Committee; American Association of Critical-Care Nurses; American College of Chest Physicians; American College of Emergency Physicians; Canadian Critical Care Society; European Society of Clinical Microbiology and Infectious Diseases; European Society of Intensive Care Medicine; European Respiratory Society; International Sepsis Forum; Japanese Association for Acute Medicine; Japanese Society of Intensive Care Medicine; Society of Critical Care Medicine; Society of Hospital Medicine; Surgical Infection Society; World Federation of Societies of Intensive and Critical Care Medicine. Surviving Sepsis Campaign: international guidelines for management of severe sepsis and septic shock: 2008. Crit Care Med. 2008; 36(1):296-327. 4. Angus DC, Linde-Zwirble WT, Lidicker J, Clermont G, Carcillo J, Pinsky MR. Epidemiology of severe sepsis in the United States: analysis of incidence, outcome, and associated costs of care. Crit Care Med. 2001;29(7):1303-1310. 5. Sands KE, Bates DW, Lanken PN, et al; Academic Medical Center Consortium Sepsis Project Working Group. Epidemiology of sepsis syndrome in 8 academic medical centers. JAMA. 1997;278(3):234-240. 6. Linde-Zwirble WT, Angus DC. Severe sepsis epidemiology: sampling, selection, and society. Crit Care. 2004;8(4):222-226. 7. Wunsch H, Angus DC, Harrison DA, et al. Variation in critical care services across North America and Western Europe. Crit Care Med. 2008;36(10):27872793.

2011 American Medical Association. All rights reserved.

Downloaded from jama.ama-assn.org at Capes Consortia on April 13, 2011

CLINICAL CROSSROADS
8. Riedemann NC, Guo R-F, Ward PA. Novel strategies for the treatment of sepsis. Nat Med. 2003;9(5):517-524. 9. Wheeler AP, Bernard GR. Treating patients with severe sepsis. N Engl J Med. 1999;340(3):207-214. 10. Hotchkiss RS, Karl IE. The pathophysiology and treatment of sepsis. N Engl J Med. 2003;348(2):138-150. 11. Kellum JA, Kong L, Fink MP, et al. Understanding the inflammatory cytokine response in pneumonia and sepsis: results of the Genetic and Inflammatory Markers of Sepsis (GenIMS) study. Arch Intern Med. 2007;167(15):1655-1663. 12. Srensen TIA, Nielsen GG, Andersen PK, Teasdale TW. Genetic and environmental influences on premature death in adult adoptees. N Engl J Med. 1988; 318(12):727-732. 13. Angus DC, Wax RS. Epidemiology of sepsis: an update. Crit Care Med. 2001; 29(7)(suppl):S109-S116. 14. Lin MT, Albertson TE. Genomic polymorphisms in sepsis. Crit Care Med. 2004; 32(2):569-579. 15. Vincent JL, Abraham E. The last 100 years of sepsis. Am J Respir Crit Care Med. 2006;173(3):256-263. 16. Angus DC, Birmingham MC, Balk RA, et al; E5 Study Investigators. E5 murine monoclonal antiendotoxin antibody in gram-negative sepsis: a randomized controlled trial. JAMA. 2000;283(13):1723-1730. 17. Abraham E, Reinhart K, Opal S, et al; OPTIMIST Trial Study Group. Efficacy and safety of tifacogin (recombinant tissue factor pathway inhibitor) in severe sepsis: a randomized controlled trial. JAMA. 2003;290(2):238-247. 18. Bernard GR, Vincent JL, Laterre PF, et al; Recombinant Human Protein C Worldwide Evaluation in Severe Sepsis (PROWESS) Study Group. Efficacy and safety of recombinant human activated protein C for severe sepsis. N Engl J Med. 2001; 344(10):699-709. 19. Kumar A, Zarychanski R, Pinto R, et al; Canadian Critical Care Trials Group H1N1 Collaborative. Critically ill patients with 2009 influenza A(H1N1) infection in Canada. JAMA. 2009;302(17):1872-1879. 20. Domnguez-Cherit G, Lapinsky SE, Macias AE, et al. Critically ill patients with 2009 influenza A(H1N1) in Mexico. JAMA. 2009;302(17):1880-1887. 21. Webb SA, Pettila V, Seppelt I, et al; ANZIC Influenza Investigators. Critical care services and 2009 H1N1 influenza in Australia and New Zealand. N Engl J Med. 2009;361(20):1925-1934. 22. Kumar A, Roberts D, Wood KE, et al. Duration of hypotension before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock. Crit Care Med. 2006;34(6):1589-1596. 23. Rivers E, Nguyen B, Havstad S, et al; Early Goal-Directed Therapy Collaborative Group. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med. 2001;345(19):1368-1377. 24. Carlbom DJ, Rubenfeld GD. Barriers to implementing protocol-based sepsis resuscitation in the emergency departmentresults of a national survey. Crit Care Med. 2007;35(11):2525-2532. 25. Reade MC, Huang DT, Bell D, et al; British Association for Emergency Medicine; UK Intensive Care Society; UK Society for Acute Medicine; Australasian Resuscitation in Sepsis Evaluation Investigators; Protocolized Care for Early Septic Shock Investigators. Variability in management of early severe sepsis. Emerg Med J. 2010; 27(2):110-115. 26. ClinicalTrials.gov. Protocolized Care for Early Septic Shock (ProCESS). http: //www.clinicaltrials.gov/ct2/show/NCT00510835?term=ProCESS&rank=1. Accessed December 2010. 27. ISRCTN. The clinical and cost-effectiveness of early, goal-directed, protocolised resuscitation for emerging septic shock (ProMISe). http://www.controlled-trials .com/ISRCTN36307479/ProMISe. Accessed December 2010. 28. Australian New Zealand Clinical Trials Registry. Australasian Resuscitation in Sepsis Evaluation Randomised Controlled Trial (ARISE). http://www.anzctr.org .au/trial_view.aspx?ID=82563. Accessed December 2010. 29. Jones AE, Shapiro NI, Trzeciak S, Arnold RC, Claremont HA, Kline JA; Emergency Medicine Shock Research Network (EMShockNet) Investigators. Lactate clearance vs central venous oxygen saturation as goals of early sepsis therapy: a randomized clinical trial. JAMA. 2010;303(8):739-746. 30. De Backer D, Biston P, Devriendt J, et al; SOAP II Investigators. Comparison of dopamine and norepinephrine in the treatment of shock. N Engl J Med. 2010; 362(9):779-789. 31. Kumar A, Ellis P, Arabi Y, et al; Cooperative Antimicrobial Therapy of Septic Shock Database Research Group. Initiation of inappropriate antimicrobial therapy results in a fivefold reduction of survival in human septic shock. Chest. 2009; 136(5):1237-1248. 32. Deans KJ, Haley M, Natanson C, Eichacker PQ, Minneci PC. Novel therapies for sepsis: a review. J Trauma. 2005;58(4):867-874. 33. Annane D, Sebille V, Charpentier C, et al. Effect of treatment with low doses of hydrocortisone and fludrocortisone on mortality in patients with septic shock [published correction appears in JAMA. 2008;300(14):1652]. JAMA. 2002; 288(7):862-871. 34. Sprung CL, Annane D, Keh D, et al; CORTICUS Study Group. Hydrocortisone therapy for patients with septic shock. N Engl J Med. 2008;358(2):111-124. 35. Annane D, Bellissant E, Bollaert PE, et al. Corticosteroids in the treatment of severe sepsis and septic shock in adults: a systematic review. JAMA. 2009; 301(22):2362-2375. 36. Jaeschke R, Angus DC. Living with uncertainty in the intensive care unit: should patients with sepsis be treated with steroids? JAMA. 2009;301(22):2388-2390. 37. Nadel S, Goldstein B, Williams MD, et al; Researching Severe Sepsis and Organ Dysfunction in Children: a Global Perspective (RESOLVE) Study Group. Drotrecogin alfa (activated) in children with severe sepsis: a multicentre phase III randomised controlled trial. Lancet. 2007;369(9564):836-843. 38. Abraham E, Laterre P-F, Garg R, et al; Administration of Drotrecogin Alfa (Activated) in Early Stage Severe Sepsis (ADDRESS) Study Group. Drotrecogin alfa (activated) for adults with severe sepsis and a low risk of death. N Engl J Med. 2005;353(13):1332-1341. 39. ClinicalTrials.gov. Investigation of the Efficacy and Safety of Drotrecogin Alfa (Activated) in Pediatric Severe Sepsis (PROWESS) Shock. http://clinicaltrials.gov /ct2/show/NCT00049764?term=Drotrecogin+Alfa&rank=1. Accessed December 2010. 40. 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. 2000;342(18):1301-1308. 41. Bellomo R, Cass A, Cole L, et al; RENAL Replacement Therapy Study Investigators. Intensity of continuous renal-replacement therapy in critically ill patients. N Engl J Med. 2009;361(17):1627-1638. 42. Palevsky PM, Zhang JH, OConnor TZ, et al; VA/NIH Acute Renal Failure Trial Network. Intensity of renal support in critically ill patients with acute kidney injury. N Engl J Med. 2008;359(1):7-20. 43. Vinsonneau C, Camus C, Combes A, et al; Hemodiafe Study Group. Continuous venovenous haemodiafiltration vs intermittent haemodialysis for acute renal failure in patients with multiple-organ dysfunction syndrome: a multicentre randomised trial. Lancet. 2006;368(9533):379-385. 44. Kress JP, Pohlman AS, OConnor MF, Hall JB. Daily interruption of sedative infusions in critically ill patients undergoing mechanical ventilation. N Engl J Med. 2000;342(20):1471-1477. 45. Girard TD, Kress JP, Fuchs BD, et al. Efficacy and safety of a paired sedation and ventilator weaning protocol for mechanically ventilated patients in intensive care (Awakening and Breathing Controlled Trial): a randomised controlled trial. Lancet. 2008;371(9607):126-134. 46. Doig GS, Simpson F, Finfer S, et al; Nutrition Guidelines Investigators of the ANZICS Clinical Trials Group. Effect of evidence-based feeding guidelines on mortality of critically ill adults: a cluster randomized controlled trial. JAMA. 2008; 300(23):2731-2741. 47. Jones NE, Heyland DK. Implementing nutrition guidelines in the critical care setting: a worthwhile and achievable goal? JAMA. 2008;300(23):2798-2799. 48. van den Berghe G, Wouters P, Weekers F, et al. Intensive insulin therapy in the critically ill patients. N Engl J Med. 2001;345(19):1359-1367. 49. Brunkhorst FM, Engel C, Bloos F, et al; German Competence Network Sepsis (SepNet). Intensive insulin therapy and pentastarch resuscitation in severe sepsis. N Engl J Med. 2008;358(2):125-139. 50. Finfer S, Chittock DR, Su SY, et al; NICE-SUGAR Study Investigators. Intensive versus conventional glucose control in critically ill patients. N Engl J Med. 2009; 360(13):1283-1297. 51. Friedman G, Silva E, Vincent JL. Has the mortality of septic shock changed with time? Crit Care Med. 1998;26(12):2078-2086. 52. Quartin AA, Schein RM, Kett DH, Peduzzi PN; Department of Veterans Affairs Systemic Sepsis Cooperative Studies Group. Magnitude and duration of the effect of sepsis on survival. JAMA. 1997;277(13):1058-1063. 53. Kaplan V, Clermont G, Griffin MF, et al. Pneumonia: still the old mans friend? Arch Intern Med. 2003;163(3):317-323. 54. Weycker D, Akhras KS, Edelsberg J, Angus DC, Oster G. Long-term mortality and medical care charges in patients with severe sepsis. Crit Care Med. 2003; 31(9):2316-2323. 55. Yende S, DAngelo G, Kellum JA, et al; GenIMS Investigators. Inflammatory markers at hospital discharge predict subsequent mortality after pneumonia and sepsis. Am J Respir Crit Care Med. 2008;177(11):1242-1247. 56. Yende S, Angus DC. Long-term outcomes from sepsis. Curr Infect Dis Rep. 2007;9(5):382-386. 57. Cook D, Rocker G, Marshall J, et al; Level of Care Study Investigators; Canadian Critical Care Trials Group. Withdrawal of mechanical ventilation in anticipation of death in the intensive care unit. N Engl J Med. 2003;349(12):11231132. 58. Angus DC, Crowther MA. Unraveling severe sepsis: why did OPTIMIST fail and whats next? JAMA. 2003;290(2):256-258. 59. Ferrer R, Artigas A, Levy MM, et al; Edusepsis Study Group. Improvement in process of care and outcome after a multicenter severe sepsis educational program in Spain. JAMA. 2008;299(19):2294-2303. 60. Pronovost P, Needham D, Berenholtz S, et al. An intervention to decrease catheter-related bloodstream infections in the ICU. N Engl J Med. 2006;355(26): 2725-2732. JAMA, April 13, 2011Vol 305, No. 14 1477

2011 American Medical Association. All rights reserved.

Downloaded from jama.ama-assn.org at Capes Consortia on April 13, 2011