Using Biomarkers to Optimize Antibiotic Strategies in Sepsis

The goal of this project is reduce unnecessary use of antibiotics in the ICU. The purpose of Phase I of the study is to identify the biomarker, or combination of biomarkers, that provides optimal test characteristics in identifying adults and children/neonates with presumed sepsis who have a very low likelihood of bacterial infection. Results of Phase I will result in development of a biomarker-based algorithm to inform need for antibiotic use in ICU patients. In Phase II, the impact of this biomarker-based algorithm on reducing antibiotic use in the ICU will be determined. Costs or savings associated with the algorithm will also be assessed.

Inclusion criteria

• SIRS Criteria

SIRS is considered to be present when patients have more than one of the following clinical findings:

• body temperature >38°C or <36°C
• heart rate >90 min-1
• respiratory rate of >20 min-1 or a Paco2 of <32 mm Hg
• and a white blood cell count of >12,000 cells µL-1 or <4,000 µL-1
• new empiric antibiotic therapy is initiated, indicating the suspicion of infection. Accepted criteria for SIRS will be used for the Medical Intensive Care Unit and Surgical Intensive Care Unit populations, with appropriate age-specific vital signs definitions to help make the definitions relevant for the Pediatric Intensive Care Unit population.

Exclusion criteria

• a code status of "do not resuscitate"
• absence of initiation or expansion of antibiotic therapy despite meeting criteria for sepsis
• presence of an immunocompromising condition.

An immunocompromising condition will be defined as one of the following:

• human immunodeficiency virus (HIV) infection with a t-helper cell (CD4) count <200 cell/mm3; 2) immunosuppressive therapy after organ transplantation
• neutropenia (<500 neutrophils/mm3)
• active chemotherapy within the 3 months preceding eligibility or
• diagnosis of cystic fibrosis.

These criteria all represent conditions in which antibiotic use is much less likely to be decreased regardless of the results of a biomarker and are consistent with exclusion criteria used in past studies of the impact of biomarkers.