The New England Journal of Medicine: Septic Shock — Evaluating Another Failed Treatment

Published on May 22, 2012
New England Journal of Medicine (
Septic Shock — Evaluating Another Failed Treatment

By Richard P. Wenzel, M.D., and Michael B. Edmond, M.D., M.P.H.

Sepsis, severe sepsis, and septic shock are progressively severe stages of the hosts systemic inflammatory response to infection. The latter stages carry increasing rates of end-organ failure and death. The spectrum of the sepsis syndrome remains a leading cause of death in the United States, and early intervention with appropriate antibiotics (matching the antibiogram susceptibilities to the drugs administered) saves lives.

Specific to septic shock, a large retrospective cohort study involving 2731 adults in 14 critical care units showed an absolute decrease of 7.6 percentage points in the survival rate for each hour of delay in antimicrobial administration from the onset of hypotension. Early recognition of severe sepsis or septic shock and the immediate initiation of effective antibiotic therapy are essential.

In the past decade, many clinicians have embraced additional strategies — with various levels of evidence to support their use — to improve the outcomes of severe sepsis and septic shock: the rapid infusion of intravenous fluids to reverse hypotension from reduced vascular resistance, the use of low-dose glucocorticoids (100 mg of hydrocortisone every 8 hours) in septic shock, low-tidal-volume ventilation (tidal volume, 6 ml per kilogram of ideal body weight) in patients with the acute respiratory distress syndrome, continual renal-replacement therapy in patients with kidney failure, and maintenance of blood glucose levels to less than 150 mg per deciliter (8.3 mmol per liter).

Since the inflammatory response to infection is complex, independent efforts have been made in the past two decades to block the activity of such likely biochemical triggers as endotoxin, tumor necrosis factor α, interleukin-1, and others. All clinical trials of agents that are designed to interfere with these pathways have not shown a benefit. However, in 2001, Bernard and colleagues reported a modest improvement in all-cause mortality at 28 days associated with adjunctive therapy with recombinant human activated protein C, or drotrecogin alfa (activated) (DrotAA).

The investigators found a death rate of 24.7% in treated patients versus 30.8% in controls, an absolute reduction of 6 percentage points and a relative reduction of 19%.3 The efficacy of DrotAA was biologically plausible, because the drug has effects on anticoagulant activity that are thought to be important in reversing the adverse microvascular effects of sepsis. Nevertheless, the initial success could not be replicated in patients with a low risk of death or in children with severe sepsis.

The outcomes of the trial were also clouded by a lack of consensus within the advisory panel of the Food and Drug Administration that evaluated evidence supporting the proposal to license DrotAA, as well as by the sponsors use of a new master lot of cells to produce the drug after the trial had begun, uneasiness about serious bleeding in patients who received the drug, and ethical questions surrounding the sponsors hiring of a public relations firm to assemble a task force (consisting of many members with conflicts of interest) to promote sepsis-treatment bundles that would include the drug — despite a single positive study and lingering controversies.

Because a subgroup analysis of the first trial suggested a greater mortality benefit among patients who were more seriously ill, a new trial was proposed targeting patients at highest risk for death. Ranieri and colleagues9 now report in the Journal the results of that trial, in which the use of DrotAA, as compared with placebo, had no significant effect on all-cause mortality among patients with septic shock.9 The absolute between-group differences in all-cause mortality at 28 and 90 days were 2.2 and 1.4 percentage points, respectively — both trends slightly favoring placebo over DrotAA.

The strengths of the study include a focus on patients with the most advanced stage of sepsis — those with sustained shock and organ dysfunction. In addition, the definitions regarding the patients conditions were precise; patients, investigators, treating clinicians, and representatives of the sponsor were all unaware of study-group assignments; and the study groups appeared to be similar after randomization.

A weakness of the study is that it did not achieve its stated statistical power of 80%, for the following reasons: The sample-size calculation was based on a rate of death of 35% in the placebo group. However, the end point of all-cause mortality was composed of the rate of death from patients underlying disease plus the rate directly attributable to infection. Thus, if we assume that the 35% baseline rate of death included a 10% component from underlying disease and 25% from sepsis, any adjunctive therapy for sepsis would affect only the component of mortality from sepsis. Even with appropriate antibiotic treatment, there will be some residual deaths from sepsis.

Theoretical Effect of a Drug for Septic Shock on All-Cause Mortality at 28 Days, According to Revised Estimates of Residual Attributable Mortality.).  The stated goal of treatment with DrotAA was a relative reduction of 20% and an absolute reduction of 7% percentage points in all-cause mortality. Since DrotAA could have an effect only on the residual attributable mortality, a 20% relative reduction in the residual attributable mortality (from 25% to 20%) would result in a decrease in all-cause mortality from 35% to 30% — an absolute difference of 5 percentage points rather than the targeted 7 percentage points. Many more study patients would have been required to provide a power of 80% to detect a decrease of only 5 percentage points. Stated another way, in order to achieve an absolute reduction of 7 percentage points in all-cause mortality, the residual attributable mortality would have needed to fall from 25% to 18% — a relative reduction of 28%, not 20%.

Despite its limitations, the large and well-conducted study by Ranieri et al. should end any further pursuit of a niche for DrotAA in the treatment of sepsis. The investigators findings provide a sad chapter in the noble quest for a truly effective adjunct for the treatment of septic shock. This setback should inspire a redoubling of efforts to seek new approaches to treatment that are based on a more crystalline view of the biology of sepsis.

Disclosure forms provided by the authors are available with the full text of this article at


From the Department of Internal Medicine, Medical College of Virginia, Virginia Commonwealth University, Richmond.