The trusted source for
healthcare information and
Should Patients at Increased Risk of Bleeding Receive Activated Protein C?
Abstract & Commentary
By Andrew M. Luks, MD, Pulmonary and Critical Care Medicine, University of Washington, Seattle, is Associate Editor for Critical Care Alert.
Dr. Luks reports no financial relationship to this field of study.
Synopsis: This retrospective medical record review demonstrated that patients with severe sepsis who received recombinant activated protein C in the face of bleeding precautions had higher rates of bleeding and death compared to those without bleeding precautions.
Source: Gentry CA, et al. Adverse outcomes associated with the use of drotrecogin alfa (activated) in patients with severe sepsis and baseline bleeding precautions. Crit Care Med 2009;37:19-25.
Despite concerns that recombinant human Activated Protein C (rhAPC) is associated with an increased risk of bleeding, the FDA failed to list several of the bleeding-related exclusion criteria used in the PROWESS trial as contraindications to use of this agent.1 Instead, criteria such as presence of an intracranial aneurysm, use of aspirin within the preceding 7 days, or ischemic stroke within 3 months were labeled as "precautions" and use of the medication under such conditions is still permitted. Gentry and colleagues sought to determine whether the use of rhAPC in patients with such precautions was, in fact, associated with an increased risk of bleeding and other adverse events.
To investigate this issue, the authors conducted a retrospective review of all adult patients who received rhAPC at 2 tertiary medical centers over a 4-year period. For each identified patient, the authors reviewed whether the patients had baseline contraindications and precautions for use of the medication based on the official product labeling. Patients were labeled as having "baseline bleeding precautions" if they had either a contraindication or a precaution for use of rhAPC that would have excluded them from the PROWESS trial. The authors then compared outcomes between patients with and without baseline bleeding precautions with the primary outcomes being the incidence of serious bleeding events and mortality at 30-days post-discharge. Serious bleeding events were defined as an acute fall > 2 g/dL in hemoglobin concentration (except for intracranial bleeding), a need to transfuse > 4 units of packed red blood cells in 48 hours, objective evidence of bleeding that led to prolonged hospitalization or death, and a progress note documenting a diagnosis relevant to bleeding.
Complete data were available for a total of 73 patients who received rhAPC at the 2 institutions during the study period. Twenty (27%) of these patients had baseline bleeding precautions that would have excluded them from the PROWESS trial. Serious bleeding events occurred in 7 of the 20 patients with baseline bleeding precautions (35%) vs only 2 of the 53 patients who lacked such precautions (3.8%; P < 0.0001) The bleeding events included gastrointestinal bleeds (6), hemothorax (1), hemorrhagic stroke (1), and subarachnoid hemorrhage (1). An additional 5 gastrointestinal bleeds, 4 of which occurred in the group with baseline bleeding precautions, occurred but did not meet criteria for serious bleeding events, thereby bringing the total bleeding rate to 55% in the group with baseline bleeding precautions and 5.6% in the group without such precautions.
Overall, 26 of the 73 patients (35.6%) died within 30 days of discharge with mortality being significantly higher (65% vs 24.5%; P = 0.015) in the group with baseline bleeding precautions. Among patients with APACHE II scores > 25, mortality was 73% in the group with baseline bleeding precautions vs 33% in the group without bleeding precautions. In multivariate analysis, the presence of a baseline bleeding precaution was the only variable associated with serious bleeding events but was one of several variables, along with APACHE II score and the presence of bloodstream infection, associated with an increased mortality risk.
Aside from its high cost, the primary concern regarding the use of rhAPC in patients with severe sepsis has been the increased risk of significant bleeding. Given this concern, one would expect that substantial effort would be made to avoid the use of this agent in patients thought to be at increased risk for bleeding complications. Although the original PROWESS investigators employed strict criteria limiting use of rhAPC in these situations, the FDA failed to list many of these criteria as explicit contraindications. Instead, they labeled many study exclusion criteria as warnings, thereby permitting physicians to use the medication in these situations if the benefits were thought to outweigh the risks. The data from Gentry and colleagues clearly show an increased risk of bleeding events and mortality in patients with such baseline bleeding precautions and, as a result, suggest we may be doing our patients harm by using the more liberal FDA criteria.
The data in this study must be viewed with some degree of caution, as the number of patients in each group was very small. Data were also collected in a retrospective manner and, as a result, the two groups are likely not as well matched as if the data had been collected prospectively. If Gentry and colleagues' results differed widely from those of other studies, these issues would severely limit the applicability of the study's findings. However, multiple other studies using the package labeling as entry and exclusion criteria have also shown an increased risk of bleeding compared to trials that used the more strict PROWESS criteria.2,3 This growing body of data suggests that it may be time to review our hospital protocols for the use of this medication. In addition to limiting its use to those patients at high risk of death (APACHE II > 25), we should be adhering to the bleeding-related exclusion criteria in the original trials and not to the more liberal FDA specifications.