New research indicates that not all patients suspected of having sepsis may benefit from receiving early antibiotics.  The study was posted online in the Abstract Issue of the American Journal of Respiratory and Critical Care Medicine. [1]

A research team led by Ian Barbash, MD, MS, assistant professor of medicine at the University of Pittsburgh and associate medical director of the UPMC Presbyterian ICU looked at electronic health records of 10 academic and community hospitals to identify adults admitted to the emergency room with suspected infection within six hours of arrival and two or more instances of organ failure within 24 hours of the suspected infection.  They identified 58,413 patients who met these criteria.

The researchers used statistical methods to determine the likelihood of these patients dying and applied a second statistical model to assess their probability of infection.  They then divided the patients into 100 groups based on their statistical likelihood of dying and infection and determined the likelihood of patients in each group receiving antibiotics within three hours of suspected infection.  They examined the relationships between receipt of early antibiotics, predicted mortality and predicted infection.

Dr. Barbash and colleagues found that patients with a higher probability of infection when arriving at the emergency room, based on bloodwork and vital signs, were more likely to receive antibiotics within three hours of suspected infection than were patients with a lower probability of infection. They also found that receiving early antibiotics was associated with lower mortality in patients with a higher probability of infection, but a higher likelihood of mortality in patients with a lower probability of infection. 

While Dr. Barbash could not offer a definitive explanation of why the team found these seemingly contradictory results due to the study being observational, he hypothesized: “My overall suspicion is that it is the competing effects of the benefits of antibiotics in treating infection, balanced against the indication bias created by physicians giving early antibiotics to patients they are concerned about.  When the probability of infection is high, the beneficial effects of antibiotics win out; when the probability of infection is low, indication bias may play a greater role, because antibiotics are less likely to reduce the risk of mortality in the absence of bacterial infection.”

Current guidelines for suspected sepsis recommend antibiotic administration within specific timeframes.  The researchers conducted this study because there is increasing recognition within the medical community that not all patients with sepsis are the same, and that physicians might be given more discretion in antibiotic administration to account for these differences.

  1. Abstract #12285. “Interaction Between Severity of Illness, Diagnostic Uncertainty and Antibiotic Timing in Patients with Suspected Infection.”