Objectives

1. Understand the nomenclature of sepsis.
2. Review the epidemiology and pathophysiology of severe sepsis.
3. Review anti-inflammatory and antithrombotic treatments for severe sepsis.
4. Identify supportive therapy strategies demonstrated to reduce mortality.

Key words

antithrombin III; protein C; sepsis; septic shock; severe sepsis; SIRS; systemic inflammatory response syndrome

Abbreviations

ACCP = American College of Chest Physicians; APACHE = acute physiology and chronic health evaluation; SCCM = Society of Critical Care Medicine; SIRS = systemic inflammatory response syndrome

Severe sepsis remains a common and serious problem. It has been estimated that there are more than 750,000 cases of severe sepsis per year in the United States, resulting in 225,000 deaths annually. The financial cost is estimated to be $16.7 billion/yr.¹ Although varying definitions have made it difficult to follow the changing incidence of this disease, it appears to be on the rise. Between 1979 and 1989, the Centers for Disease Control estimated that the incidence rose from 73.6 to 175.9 per 100,000 Americans.² There are multiple reasons for this rise, including an increase in the average age of the population, improved survival of critically ill neonates, an increase in the immunocompromised population, and more invasive health care. The most common sites for infection leading to severe sepsis include the lungs, the abdominopelvic region, and the urinary tract.^3,4 Gram-positive cocci infections and Gram-negative bacilli infections contribute equally to the incidence of this disorder, with fungi making a lesser but growing contribution to the mix.^3,4

In 1992, the American College of Chest Physicians (ACCP) and the Society of Critical Care Medicine (SCCM) issued a consensus statement regarding the definitions for sepsis and organ failure.⁵ This was necessary to encourage accurate tracking of the incidence and prevalence of these disorders, as well as to facilitate the study of innovative treatments. The consensus committee settled on a tiered system of progressively more severe inflammatory states: (1) the systemic inflammatory response syndrome (SIRS); (2) sepsis; (3) severe sepsis; and (4) septic shock.

SIRS involves two or more of the following findings: (1) temperature > 38°C or < 36°C; (2) heart rate > 90 beats/min; (3) tachypnea, with a respiratory rate > 20 breaths/min or PaCO₂ < 32 mm Hg (4.25 kPa); and (4) WBC count > 12 x 10⁹/L or < 4 x 10⁹/L, or > 10% band forms on a peripheral blood smear. Sepsis is defined as SIRS plus compelling evidence of infection. Severe sepsis is present in a patient who has sepsis with at least one end organ dysfunction. Septic shock is defined as severe sepsis in which one of the organ dysfunctions is hypotension refractory to volume expansion.

In 1995, Rangel-Frausto and colleagues⁶ reviewed the ability of the consensus conference definitions to describe progressively smaller subsets of more seriously ill inpatients. They found that the mortality rate for patients with SIRS was 7%, rising to 16% for those with sepsis, 20% for severe sepsis, and 46% for septic shock. Fortunately, septic shock accounted for only 4% of admitted patients with an inflammatory state.

Pathophysiology

The ACCP/SCCM consensus definitions describe a continuum of the inflammatory state. The pathophysiologic underpinnings of this are just becoming known to us. In 1996, Bone⁷ described homeostasis of inflammation as involving (1) an appropriately localized response to a localized insult, and (2) appropriate down-regulation of response as the threat is eliminated.

The continuum of sepsis represents a loss of a balanced response with uncontrolled, intertwined, and self-amplifying cascades of coagulation, abnormal fibrinolysis, and inflammation.

This process begins with an initial host immune response to a stimulus such as endotoxin. In turn, this leads to cytokine production, expression of selectins on the endothelial surface, and resulting neutrophil adhesion to the vessel surface. As further mediators of inflammation are elaborated, vascular endothelial injury and tissue factor release result, activating the coagulation cascade. In particular, activation of thrombin plays a central role as it reinforces both the clotting cascade and inflammation. Unfortunately, as homeostasis is lost, components of the counterregulatory response are down-regulated. This includes decreased expression of thrombomodulin, protein C, protein C receptor, and consumption of antithrombin III. The proinflammatory and prothrombotic process is further aided by down-regulation of fibrinolysis, with increased expression of plasminogen activator inhibitor, or PAI-1, and thrombin activatable fibrinolysis inhibitor, or TAFI. As inflammation and thrombosis advance, there is progressive compromise of the microcirculation, resulting in end organ ischemia and dysfunction. Effective therapy aimed at this disordered response has only recently become available.

Anticoagulant/Anti-inflammatory Treatment

Recombinant Activated Protein C [Drotrecogin Alfa (Activated)]

The use of exogenous activated protein C is aimed at remedying the intrinsic deficiency of protein C seen in severe sepsis. Increasing degrees of deficiency of activated protein C have been associated with the risk of death.^8,9 Based on the results of studies demonstrating benefit in a baboon model of sepsis¹⁰ and a small placebo-controlled trial in human subjects with severe sepsis,¹¹ the PROWESS (Protein C Worldwide Evaluation in Severe Sepsis) trial was undertaken.⁴

The PROWESS trial was designed as a randomized, double-blind, placebo-controlled, multicenter trial of a 4-day infusion of recombinant human activated protein C [drotrecogin alfa (activated)]. Patients included in the study met criteria for severe sepsis as defined by the ACCP/SCCM consensus conference, with the exception that they met three SIRS criteria rather than two. The primary exclusions of note included increased risk for bleeding, moribund state, and a short expected survival irrespective of the patient’s underlying acute illness. Additional exclusions included a known hypercoagulable state, chronic renal failure, advanced HIV/AIDS, posttransplant status, acute pancreatitis, age < 18 years, and weight > 135 kg. Accepted patients who were randomly assigned to the experimental treatment received 24 µg/kg/h (based on actual body weight) of drotrecogin alfa (activated) for 96 h. The primary end point of the study was 28-day mortality. Although the study was designed to enroll 2,280 patients, it was halted at the second interim analysis when a predefined statistical benefit of study drug over placebo was met.

The results for the experimental group showed an absolute risk reduction in 28-day mortality of 6.1%, representing a relative risk reduction of 19.4%. This benefit of survival has subsequently been shown to extend past 1 year.12 As might be expected, the experimental group showed a trend towards increased risk of serious bleeding (p = 0.06), with an absolute increase in serious bleeding risk of 1.5%. In all, there were four deaths in the study group directly linked to bleeding. In this study,4 as well as in a follow-up compassionate use trial, the greatest risks for life-threatening bleeding have included a platelet count of < 30 x 109/L or meningitis.¹³

A number of subgroups were prospectively identified in an effort to better define those patients most likely to benefit from treatment, including APACHE II score, number of dysfunctional organs or systems, sex, age, site of infection, type of causative organism, and protein C level at enrollment. Analysis showed a consistent benefit in all of the identified subgroups.

In approving recombinant activated protein C [drotrecogin alfa (activated)], the Food and Drug Administration has encouraged limiting use of the agent to those patients who have severe sepsis and are at a particularly high risk of death. Their analysis of the PROWESS data suggested that treated patients in the third and fourth APACHE (acute physiology and chronic health evaluation) II quartiles (score > 25) experienced the greatest degree of benefit in outcomes. Additionally, these quartiles have also shown the most favorable economic outcome per year of life gained.¹⁴ The PROWESS study lacked sufficient statistical power to definitively address the question of benefit in the first and second APACHE II quartiles. A follow-up study of less severely ill patients is planned.

A number of additional concerns regarding the PROWESS trial have been voiced. These include a change in the master cell bank producing recombinant protein C during the study, and an additional exclusion of moribund patients that was approved by the Food and Drug Administration midway through the study. Neither of these factors, however, appears to have been of sufficient weight to counter the measured benefit.

High-Dose Antithrombin III

Shortly after publication of the PROWESS trial, Warren and colleagues³ published the results of a study of high-dose antithrombin III in severe sepsis (the KyperSept Trial Study). The study design was similar to that of the PROWESS trial, and the KyperSept study ultimately enrolled 2,314 patients. The investigators hoped that supplementing deficient levels of antithrombin III would help restore homeostasis to the proinflammatory, procoagulant state.

Patients randomly assigned to receive high-dose antithrombin III received 6,000 IU over 30 minutes followed by a continuous infusion of 6,000 IU/d for 4 days. There was no benefit in the overall study population for the primary end point of 28-day mortality from all causes. However, a trend towards benefit (p = 0.08) was noted in a prospectively identified subgroup of patients receiving no parenteral heparin of any type, at any dose. At 90 days, this subgroup did demonstrate a statistical survival benefit (p = 0.03). At this secondary end point, the absolute risk reduction for death was 7.6%, and the relative risk reduction was 14.5%. The absolute increase in risk of bleeding in this study for the experimental group was 10.3%, which was statistically significant (p < 0.001).

Despite the lack of benefit in terms of the primary end point, this study raises several intriguing questions. A similar protocol either utilizing a lower dose of antithrombin III or actively excluding all heparin use would seem to be merited. Additionally, the potential for positive interactions between various anticoagulants cannot be dismissed. Indeed, a pending study of activated protein C will stratify patients according to use of heparin for deep venous thrombosis prophylaxis.

Supportive Modalities

Drotrecogin alfa (activated)and, to a lesser extent, antithrombin III have garnered great interest as the first anticoagulant/anti-inflammatory agents demonstrated to improve mortality. However, a number studies have demonstrated similar survival advantages using modifications of existing supportive therapies.

Intensive Glycemic Control in Critically Ill Patients

While an earlier study had demonstrated a benefit to maintaining glucose levels of < 215 mg/dL in the setting of acute myocardial infarction,¹⁵ the benefit of tighter control had not been rigorously studied. Van den Berghe and colleagues¹⁶ undertook a study of intensive insulin therapy with a goal of tight glycemic control to answer this question. The trial was prospective, randomized, and controlled. A total of 1,548 surgical ICU patients receiving mechanical ventilation were enrolled. Subjects were randomly assigned to a treatment strategy with a goal of maintaining serum glucose between 80 and 110 mg/dL, or to a control group with a target glucose level of 180 to 200 mg/dL. The primary end point was ICU mortality, and the study was stopped at the fourth interim analysis when a statistically significant benefit was identified in the tight control arm of the study.

Tight control was associated with an absolute risk reduction of ICU death of 3.4% (p < 0.04) in those patients who spent > 5 days in the ICU. Patients with multiple organ dysfunction and a proven septic focus represented the subgroup with the greatest mortality benefit. In addition to a reduction of mortality, patients with tight glycemic control had a statistically significant reduction in bloodstream infection, acute renal failure, total units of packed RBCs transfused, and critical illness polyneuropathy. There was also a reduction in the peak serum bilirubin value. Although 39 patients in the experimental arm experienced hypoglycemia with a serum glucose of < 40 mg/dL, only two patients were symptomatic. No long-term sequelae of hypoglycemia were reported.

Goal-Directed Therapy in Resuscitation of Septic Shock

A number of techniques exist for judging the adequacy of hemodynamic resuscitation in patients with severe sepsis and septic shock. These range from the simple, such as ability to mentate and produce urine, to more complex, unproven interventions such as pulmonary artery catheterization. Rivers and colleagues¹⁷ undertook a study of a simplified hemodynamic resuscitation protocol that encouraged rapid intervention. This protocol made use of readily available monitoring techniques and unambiguous decision points in order to facilitate rapid initial treatment in the emergency department. This study was conducted in a single center and was both randomized and controlled. Patients met simplified defining criteria for septic shock. These included a minimum of two of the four SIRS criteria and a systolic blood pressure of ≤ 90 mm Hg after a fluid bolus or a lactate dehydrogenase level of ≥ 4 mmol/L. The study arms included “standard therapy,” which was not well defined, and the experimental protocol. The experimental protocol involved initial assurance of a stable airway and oxygenation, sedation as needed for those patients requiring mechanical ventilation, and placement of an arterial line and an oximetric central venous catheter.

The experimental protocol permitted the use of pressors only after a central venous pressure of 8 to 12 cm H_2O was obtained. After achieving this goal for filling pressure and a mean arterial pressure of 65 to 90 mm Hg, peripheral oxygen delivery was assessed. Central venous oxygen saturation > 70% was used as the marker of adequate oxygen delivery. Patients who had saturation levels below this value and a hematocrit of < 30% received transfusions of packed RBCs. If the hematocrit was > 30%, inotropic support was provided.

The experimental protocol was associated with an absolute risk reduction of in-hospital mortality of 16% (p = 0.009), and a significantly lower APACHE II score 72 h after study entry. The most striking difference in terms of therapy received by the two treatment arms was a temporal shift in volume resuscitation. The experimental group received significantly more volume in the first 6 h. The standard therapy group did not achieve a degree of volume expansion similar to that of the experimental group until the 72-h evaluation point.

The problems with this study include the inability to fully blind such a treatment protocol, the limitations in applying single-center data to other facilities, and a higher than typical mortality seen in the control group. Additionally, the goal values selected for central venous pressure, central venous oxygen saturation, and mean arterial pressure are open to question. In particular, there is little evidence to support the transfusion trigger used in the experimental group. Nonetheless, this study supports the concept that early and rapid hemodynamic resuscitation using simple end points takes precedence over obtaining perfect monitoring data.

Adrenal Dysfunction

The risk of acute adrenal crisis has long been recognized in critically ill patients who have either a history of chronic primary adrenal insufficiency or chronic secondary insufficiency, commonly due to exogenous glucocorticoid therapy. Despite interest in using corticosteroids for severe sepsis dating back to 1940, previous randomized, controlled trials failed to support this approach.^18,19

Recently, there has been renewed study of acute adrenal insufficiency as a manifestation of end organ dysfunction in severe sepsis. An accurate measurement of the incidence of this problem has been hampered by the lack of consensus regarding what constitutes the lower limit of cortisol production in the critically ill.

Annane et al²⁰ have sought to better define the incidence of this disorder, as well as to explore further the therapeutic benefit of corticosteroids in septic shock. They conducted a prospective, randomized, double-blind, placebo-controlled, multicenter study of 300 patients with septic shock.²⁰ Patients randomly assigned to the treatment arm received hydrocortisone 50 mg IV, in a bolus and every 6 hours, along with a daily dose of 50 µg of 9 alpha-fludrocortisone via nasogastric tube. The duration of intervention was 7 days, and all patients underwent a short cosyntropin stimulation test. The results of the stimulation test were not used in the randomization process, but rather for subgroup analysis at the study’s conclusion. The results demonstrated a reduction in 28-day, all-cause mortality in the corticosteroid-treated group. However, the benefit was entirely within the subgroup of the patients who did not respond adequately to the cosyntropin stimulation test. Inadequate response was defined as a rise of < 9 µg/dL. There was a 10% absolute risk reduction and a 17% relative risk reduction of death in treated nonresponders. The only adverse effect of the corticosteroid therapy was a 5% increase in surgical wound infections (p = 0.007).

These results suggest that patients with septic shock should undergo the short cosyntropin test, and treated pending results. Nonresponders would then continue to receive treatment, although the optimum dose and duration would require individualization.

Daily Dialysis

Acute renal failure requiring dialysis occurs in < 5% of patients with severe sepsis, but carries a substantial mortality risk.²¹ The appropriate dose of dialysis for acute renal failure in this setting has remained an open question, with many favoring more intensive support. Continuous renal replacement therapy appears to offer a more physiologic alternative to alternate-day dialysis, but it requires around-the-clock logistical support and has not been shown to offer a survival advantage. Schiffl and colleagues²² studied two alternative intermittent dialysis regimens in ICU patients with acute renal failure. The 160 patients in the study were assigned to receive conventional (alternate-day) dialysis or daily dialysis. Approximately one third of the patients had sepsis as the underlying etiology of their renal dysfunction. The primary end point was survival 14 days after the last dialysis session. The intention-to-treat analysis demonstrated an absolute risk reduction in mortality of 18% (p = 0.01) and a relative risk reduction of 39%. There was also a reduction in the duration of dialysis and in the development of SIRS or sepsis after enrollment.

The study design precluded effective blinding, and patients were assigned to a study arm in an alternating manner rather than via true randomization. Additionally, study assignment took place only after a decision had been made to forgo continuous renal replacement therapy. The reader is not informed of the basis for this decision, but the lower than expected mortality in the conventional treatment group suggests that continuous dialysis was frequently selected for the most ill patients. Despite these concerns, this study supports the contention that more intensive dialysis support is beneficial in the critically ill.

Conclusion

After a long drought, the last 2 years have brought clinicians a number of sepsis therapies that demonstrate a significant survival benefit. Drotrecogin alfa (activated) and perhaps antithrombin III represent a novel approach to disordered homeostasis in sepsis. Additionally, newer supportive strategies are particularly remarkable as they possess the cardinal virtues of simplicity, low cost, and effectiveness. Intensivists are now challenged to systematically implement and integrate these new approaches into their practices and their ICUs.

References

1. Angus DC, Linde-Zwirble WT, Lidicker J, et al. Epidemiology of severe sepsis in the United States: analysis of incidence, outcome, and associated costs of care. Crit Care Med 2001; 29:1303–1310
2. Centers for Disease Control. Increase in national hospital discharge survey rates for septicemia: United States, 1979-1987. MMWR Morb Mortal Wkly Rep 1990; 39:31–34
3. Warren BL, Eid A, Singer P, et al. High-dose antithrombin III in severe sepsis: a randomized controlled trial. JAMA 2001; 286:1869–1878
4. Bernard GR, Vincent JL, Laterre PF, et al. Efficacy and safety of recombinant human activated protein C for severe sepsis. N Engl J Med 2001; 344:699–709
5. American College of Chest Physicians–Society of Critical Care Medicine Consensus Conference. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. Crit Care Med 1992; 20:864–875
6. Rangel-Frausto MS, Pettit D, Costigan M, et al. The natural history of the systemic inflammatory response syndrome (SIRS): a prospective study. JAMA 1995; 273:117–123
7. Bone RC. Sir Isaac Newton, sepsis, SIRS, and CARS. Crit Care Med 1996; 24:1125–1128
8. Yan SB, Helterbrand J, Hartman DL, et al. Low levels of protein C are associated with poor outcome in severe sepsis. Chest 2001; 120:915–922
9. Mesters RM, Helterbrand J, Utterback BG, et al. Prognostic value of protein C concentrations in neutropenic patients at high risk of severe septic complications. Crit Care Med 2000; 28:2209–2216
10. Taylor FB, Chang A, Esmon CT, et al. Protein C prevents coagulopathic and lethal effects of Escherichia coli infusion in the baboon. J Clin Invest 1987; 79:918–925
11. Hartman DL, Bernard GR, Helterman JD, et al. Recombinant human activated protein C (rhAPC) improves coagulation abnormalities associated with severe sepsis [abstract]. Intensive Care Med 1998; 24(suppl 1):S77
12. Angus DC, Laterre PF, Helterbrand J, et al. The effects of drotrecogin alpha (activated) on long-term survival after severe sepsis [abstract]. Chest 2002; 122(suppl):51S
13. Bernard GR, Macias WL, Vincent JL, et al. Drotrecogin alpha (activated) cumulative safety update data [abstract]. Chest 2002; 122(suppl):50S
14. Manns BJ, Lee H, Doig CJ, et al. An economic evaluation of activated protein C treatment for severe sepsis. N Engl J Med 2002; 347:993–1000
15. Malmberg K, Ryden L, Efendic S, et al. A randomized trial of insulin-glucose infusion followed by subcutaneous insulin treatment in diabetic patients with acute myocardial infarction (DIGAMI study): effects on mortality at one year. J Am Coll Cardiol 1995; 26:57–65
16. Van den Berghe G, Wouters P, Weekers F, et al. Intensive insulin therapy in critically ill patients. N Engl J Med 2001; 345:1359–1367
17. Rivers E, Nguyen B, Havstad S, et al. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med 2001; 345:1368–1377
18. The Veterans Administration Systemic Sepsis Group. Effects of high-dose glucocorticoid therapy in patients with clinical signs of sepsis. N Engl J Med 1987; 317:659–665
19. Bone RC, Fisher CJ, Clemmer TP, et al. A controlled clinical trial of high-dose methylprednisolone in the treatment of severe sepsis and septic shock. N Engl J Med 1987; 317:653–656
20. 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. JAMA 2002; 288:862–871
21. Bernard GR, Wheeler AP, Russell JA, et al. Effects of ibuprofen on the physiology and survival of patients with sepsis. N Engl J Med 1997; 336:912–918
22. Schiffl H, Lang SM, Fisher R. Daily hemodialysis and the outcome of acute renal failure. N Engl J Med 2002; 346:305–310