Sepsis has been defined as “a systemic response to infection where there is fever or hypothermia, tachycardia, tachypnea, and evidence of decreased blood flow to internal organs”1 although others have been more concise, defining it as “the systemic host response to infection.”2
Despite advances in almost every aspect of critical care medicine, the mortality from sepsis is approximately 50 percent, and has not changed significantly. The annual health care cost of sepsis in the United States is more than $16 billion, with an average per-patient cost of $22,100.3 As a result of many years of both basic science and clinical research, there are several new ideas about the treatment (one of which has already come to market). Researchers also have been revisiting older therapies, and new developments give the promise of others on the horizon.
However one defines it clinically, sepsis usually causes fever, (although sometimes hypothermic), hypotension, and often altered sensorium. The infectious causes include bacteria, viruses, fungi, and parasites. There are noninfectious causes that mimic sepsis, and that list includes acute pancreatitis, drug or toxin ingestions, massive myocardial infarction or pulmonary embolism, thyroid storm, and neuroleptic malignant syndrome. The etiology in an individual patient is sometimes not clear, and a more empiric approach is often necessary, especially early in the management.
The approach to the severely septic patient begins with attention to the airway, breathing, and circulation. Mechanical ventilation is used to decrease oxygen consumption by the respiratory muscles and increase oxygen availability for other tissues. If ARDS is present, adding PEEP (10–15cm H20) improves oxygenation. The initial approach to improving hypotension is volume. If the patient remains hypotensive despite a reasonable fluid challenge (2 liters crystalloid), then vasopressor therapy is the next step.
Dopamine, dobutamine, norepinephrine, phenylephrine, and epinephrine all have specific advantages and limitations, which are beyond the scope of this article. One new trend is the more frequent use of norepinephrine as a result of a 1998 study that showed no damage to end-organs in healthy volunteers.4
After appropriate cultures — blood and urine always, and if appropriate, sputum, joint and/or wound are obtained — antibiotics should be administered. The choices are either site-specific or empiric, and the possibility of a hospital-acquired infection should be considered. The immune status of the patient (HIV disease, neutropenia, and medications), allergies, and hospital formularies play into the final selection. The table gives reasonable antibiotic choices for a variety of situations. Note that in most causes there are several choices, and individual alternatives must be tailored to the patient. The next step, if the site is identified as a source amenable to surgical drainage, is to perform the procedure or obtain surgical consultation.
Pathophysiology of Sepsis
For years there has been interest in determining the pathophysiology of sepsis and directing therapy to specific sites in the cascade. In 1972 Lewis Thomas developed a theory that sepsis was a result of uncontrolled inflammation. As a result of that theory, therapy was aimed at attenuating the patient's immune response. Antiendotoxin, interleukin-1 receptor antagonism, tumor necrosis factor antibodies, and cyclooxygenase inhibitors have not been successful in significantly reducing mortality in severe sepsis.
There also have been many studies looking at corticosteroids, and most have shown no decrease (and in some an increase) in mortality. More recently, a 2001 study by Annane5 indicated that “physiologic” doses of corticosteroids may be beneficial in patients who, despite persistent ventilator support and vasopressors, remain in severe shock. In another study by Annane,6 patients given hydrocortisone and fludrocortisone for seven days showed improved survival over controls. The theory supporting these results is that these patients, although having normal adrenal function, have a relative adrenal insufficiency.
It is also thought that corticosteroids increase expression of adrenergic receptors.7 Note that the doses are considered low-dose; no studies have shown a benefit in high-dose, and the second study has yet to be reproduced. It is important to remember that corticosteroids have been shown beneficial in Pneumocystis carinii pneumonia, children with meningitis (especially H. influenza), severe typhoid fever, and adrenal insufficiency.
Another aspect of sepsis that has been studied recently is insulin therapy and close glucose control. A 2001 study8 demonstrated that keeping blood glucose between 80–110 mg/dl resulted in lower mortality in septic patients compared with blood glucose between 180–200 mg/dl; whether the patient was previously diabetic did not matter. The reason for the results is not known, but theories include improved neutrophil function with normal blood sugar, and cell-protective effects of insulin itself.7
The latest theories on the sepsis cascade relate to the link between a generalized inflammatory and procoagulant response by the host as a result of exposure to either the cell wall and/or DNA from the infecting bacteria, virus, fungi, or parasite.9 Even with appropriate antibiotics and supportive treatment, the intensive host response leads to extensive endothelial damage, fibrin deposition in the smaller blood vessels, organ hypoperfusion then dysfunction, and eventually host death.
Benefit of Treatment
After unsuccessful attempts at finding therapies for sepsis in the coagulation cascade, the PROWESS (Protein C Worldwide Evaluation of Severe Sepsis) trials found that therapy with activated protein C resulted in reductions of 19.4 percent in relative and 6.1 percent absolute risks of death.10 The mechanism of action in this mortality reduction is thought to be inhibition of thrombosis, decreased inflammation, and endothelial cell death. Activated protein C is marketed under the name drotrecogin alfa (activated) or Xigris.
Patients with either gram-negative or gram-positive sepsis had comparable benefit, and the most ill patients (highest APACHE II scores) had the greatest benefit, especially those with sepsis from pneumonia. Patients received therapy within 48 hours of identification of infection. Predictably, the most serious side effect was serious bleeding (2.5% vs.1.0% in placebo), and most bleeding episodes were procedure-related. The number needed to treat was 16 (16 patients treated to benefit one).
The recommendations for procedures in patients treated with this drug are generally holding the infusion two hours before and restarting the infusion two hours after the procedure is performed. The exceptions are surgery (two hours before and 12 hours after) and epidural catheters (generally not recommended). Drotrecogin alfa raises the abnormally low protein C levels in septic patients, but also further prolongs both the activated partial thromboplastin time (APTT) and prothrombin times (PT); it will also decrease the D-dimer, and in some patients will increase the platelet counts.
Choosing patients who will benefit from this therapy will probably be made by infectious disease/critical care consultants, and should follow the inclusion criteria of the PROWESS study. These criteria are:
- ▪ Presence of known or suspected infection.
- ▪ Presence of at least three systemic inflammatory response syndrome criteria (SIRS).
- ▪ Dysfunction of one or more organs or organ systems.
The SIRS criteria are:
- ▪ Temperature higher than 38°C or lower than 36°C.
- ▪ Pulse greater than 90 beats per minute.
- ▪ Respiratory rate greater than 20 breaths per minute or PaCO2 less than 32 mm Hg.
- ▪ WBC count greater than 12,000 or less than 4,000 or greater than 10 percent bands.2
The dose is 24 mcg/kg/hour over 96 hours, and no allergic reactions have been reported.
Drotrecogin alfa is our most advanced drug therapy to date, but animal studies have shown improvement in survival using interleukin-12, inhibition of macrophage migration factor, and PARP (poly-ADP-ribose polymerase).7 For now, however, good supportive care, appropriate antibiotics, and in the proper patients, possibly using drotrecogin alfa is the best we can do to treat our septic patients.
1. Taber's Medical Dictionary
. F.A. Davis Co. Electronic pages; accessed June 7, 2003.
2. Bone RC, Balk RA, Cera FB, et al. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis: ACCP/SCCM Consensus Conference. Chest
3. Angus DC, Linde-Zwinble WT, Lidicker J, et al. Epidemiology of severe sepsis in the United States: Analysis of the incidence, outcome, and associated cost of care. Crit Care Med
4. Hoogenberg K, Smit AJ, Girbe ARJ: Effects of low-dose dopamine on real and systemic hemodynamics during incremental norepinephrine infusion in healthy volunteers. Crit Care Med
5. Annane D. Corticosteroids for septic shock. Crit Care Med
6. Annane D, Sebille V, Charpenter C, et al. Effect of treatment with low doses of hydrocortisone and fludrocortisone on mortality in patients with septic shock. JAMA
7. Hotchkiss RS, Karl I. The pathophysiology and treatment of sepsis. N Engl J Med
8. Van den Berghe G, Wooters P, Weekers F, et al. Intensive insulin therapy in critically ill patients. N Engl J Med
9. Aderem A, Ulevitch RJ. Toll-like receptors in the induction of the innate immune response. Nature
10. Bernard GR, Vincent J-L, Latterre P-F, et al. Efficacy and safety of recombinant human protein C for severe sepsis. N Engl J Med