UC San Diego

Abstract

Background

Balancing antimicrobial stewardship with sepsis management is a challenge. At our academic medical center, a “Code Sepsis” was implemented as a nursing driven initiative to improve early recognition and management of sepsis. Per protocol, Code Sepsis is activated in patients who meet two or more systemic inflammatory response syndrome (SIRS) criteria due to a suspected infection to allow for early implementation of the sepsis bundle, which includes laboratory testing, fluid resuscitation, and antibiotic administration (Figure 1). We analyzed the impact that Code Sepsis had on antimicrobial use among hospitalized patients over a six month period.

Methods

We reviewed the electronic medical records of hospitalized patients with Code Sepsis activation between January 1, 2018 and June 30, 2018 to determine whether antibiotics were “escalated” or “not escalated.” Among patients who had antibiotic escalation, escalation was classified as “indicated” or “not indicated” (Figure 2). A logistic regression model was used to identify characteristics, SIRS or organ dysfunction criteria predictive of indicated antimicrobial escalation.

Results

Code Sepsis was activated in 529 patients with antibiotics escalated in 247 (47%) and not escalated in 282 (53%) (Table 1). Among patients whose antibiotics were escalated, 64% (152) had an indication. In 36% (89), escalation was not indicated as Code Sepsis was due to a suspected noninfectious source, known infectious source already on appropriate antimicrobials, or a suspected infectious source in which diagnostic results had already shown the absence of the infection (Figure 2). Odds of indicated antibiotic escalation increased with the number of SIRS and organ dysfunction criteria (Table 2).

Conclusion

In our efforts to improve sepsis outcomes, we focused on early recognition (Code Sepsis) and intervention (sepsis bundle). However, our Code Sepsis inadvertently led to antibiotic overutilization. By refocusing Code Sepsis on early recognition of severe sepsis and septic shock, we hope to optimize resource utilization and improve patient outcomes.

Disclosures

All authors: No reported disclosures.