Source: Marti-Carvajal AJ, Salanti G, Cardona-Zorrila AF et al. Human recombinant activated protein C for severe Sepsis. Cochrane Database Syst Rev 2008. Issue 1. Art. No.: CD004388. DOI: 10.1002/14651858.CD004388.pub3.
Efficacy Endpoints: Mortality
Harm Endpoints: Major bleeding event (intracranial bleeding, fatal bleeding, or bleeding requiring transfusion)
Narrative: Severe sepsis is a common condition, with an incidence of 176 per 100,000 in the United States according to a Centers for Disease Control and Prevention study from 20011. It is also often fatal, with mortality ranging from 20-80%. Current treatment practices including early goal directed therapy have had a significant impact, however, new therapies are needed to further decrease mortality.
This review examined the effect of human recombinant activated protein C (APC), brand name Xigris, in the treatment of severe sepsis. Severe sepsis was defined as known or suspected infection accompanied by acute organ dysfunction, organ hypoperfusion or hypotension. The authors included four trials involving 4911 participants (4434 adults and 477 pediatric patients) who received either APC or placebo in addition to conventional treatment. Mortality and hospital mortality were the primary endpoints.
APC yielded no benefit. There was no statistically significant difference in mortality between the control group and those who received APC, regardless of the severity of the sepsis. Use of APC was, however, associated with a higher risk of serious bleeding. There was an absolute risk increase of 1.4%, or one in every 70 patients treated with APC. However, in the pediatric subgroup there was no increase in serious bleeding events. Removing pediatric patients from the analysis, the absolute risk increases to 1.9%, or 1 in 55 adult patients.
Caveats: Subgroup analysis performed in the Cochrane review suggested a benefit for APC when used in patients with three or more organ system dysfunctions (ARR 9%, NNT = 11), however this is data from a single study that was not powered for subgroup analysis.2 This study remains the only one of four major trials to demonstrate a benefit, and this benefit appears to have been dependent upon a modified protocol that was instituted at the midpoint of the trial, excluded patients in septic shock, and changed the placebo from 0.9% sodium chloride to 0.1% human serum albumin. These changes were not reported in the original publication. The data from this study were also used as the basis for 22 additional publications, 21 of which concluded that APC was beneficial, an impressive example of data duplication.
Author: Dan Runde, MD
Published/Updated: August 27, 2010
The title bar is color-coded with our overall recommendation.
If you have suggestions, requests, or questions about a particular NNT review, please send us a message and we’ll try to address it as soon as possible.