The trusted source for
healthcare information and
Selenium and Sepsis. It's Not Your Average Once-a-Day Vitamin
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: A prospective randomized, placebo-controlled, multi-center trial demonstrates that a prolonged course of intravenous selenium improves mortality in patients with severe sepsis and septic shock and is associated with minimal to no side effects.
Source: Angstwurm MW, et al. Selenium in Intensive Care (SIC): results of a prospective randomized, placebo-controlled, multiple-center study in patients with severe systemic inflammatory response syndrome, sepsis, and septic shock. Crit Care Med. 2007; 35(1):118-126.
Building on the results of previous pilot studies from their laboratory and a meta-analysis1 that showed a tendency toward improved mortality in critically-ill patients following selenium administration, Angstwurm and colleagues used a randomized, placebo-controlled multi-center trial to determine whether intravenous administration of selenium could improve outcomes in severe sepsis and septic shock. They recruited patients from 11 centers across Germany and included those men and women above the age of 18 with Acute Physiology and Chronic Health Evaluation (APACHE) III scores > 70 and at least two other markers of sepsis, such as elevated or decreased body temperature, tachycardia, tachypnea, leukocytosis, leucopenia or thrombocytopenia. Patients were excluded if they were pregnant, or had concomitant disease with expected mortality within 2 months, hypoxic-ischemic encephalopathy, primary malignant disease or hemorrhagic pancreatitis without infection.
Subjects were randomized to receive either 1000 micrograms of selenium as an initial bolus injection followed by a 14-day continuous infusion at a rate of 1000 micrograms per 24 hours or an equal volume of normal saline. The authors did not specify whether the infusion was continued if the patients recovered prior to the end of the 14-day period. The primary end-point was 28-day mortality. Multiple secondary endpoints such as the number of days of vasopressor therapy, mechanical ventilation and hemodialysis were examined.
After excluding 60 patients due to protocol violations and other factors, the authors found that 28-day mortality among the 92 patients in the selenium group was 42.4% compared with 56.8% among the 97 patients in the placebo group. This corresponds to a number needed to treat of seven and the authors estimate that the cost per life saved is roughly 1050 Euro (about $1400 USD). Among pre-defined subgroups, mortality was improved in patients with sepsis and disseminated intravascular coagulation, patients with APACHE III scores above 102, and those patients receiving intensive insulin therapy. Mortality was inversely correlated with whole blood selenium concentrations; mortality rates as low as 23-24% were found when drug levels were among the upper two-thirds of measured concentrations in the study. There were no major differences between the selenium and placebo groups with regard to any of the secondary endpoints and there was no difference in the incidence of adverse events between the two treatment arms.
Given that the mortality from severe sepsis ranges as high as 50%, identifying a cost-effective, low risk, simple-to-use therapy that could reduce mortality would be of great benefit. On the surface, the trial presented by Angstwurm and colleagues seems to provide promise in this regard. In a clinical trial that is methodologically sound apart from a lower number of subjects than one might expect in an 11-center study, the authors demonstrate improvements in mortality using an intervention that is safe, relatively inexpensive and associated with a low number needed to treat to save one life.
While these are appealing factors, enthusiasm for the results must be tempered by a simple fact: the proposed treatment presents important logistical problems, the most important of which is the long duration over which the therapy must be administered. The selenium infusion lasted 14 days and the authors provide no information as to their protocol in the event patients improved before the two weeks had elapsed. As a result, it is not clear if the mortality benefit extends to patients who stop the infusion early and/or are converted to oral administration after their illness resolves. Given the increasing demands on hospital resources, keeping patients in the hospital for a two-week period is infeasible and exposes the patient to risks such as nosocomial infection. The fact that 49 of their original 238 patients were dropped from the study due to protocol violations involving drug administration should also serve as a warning that the protocol may, in fact, not be an easy one to follow.
With these problems in mind, critical care providers would do well to avoid the quick urge to adopt interventions as standard practice after only one positive trial, as happened with activated protein C, intensive insulin therapy, and corticosteroids in refractory septic shock; and should instead wait for further trials on selenium with perhaps more feasible drug administration protocols.