1116

Abstract

Introduction: Early quantitative resuscitation for sepsis can reduce mortality significantly, therefore early identification and treatment is imperative to improving outcomes. Sepsis is a common cause for clinical deterioration in the admitted patient, and is one of the most common diagnoses identified by Rapid Response Teams (RRTs). Our goal was to determine the accuracy of sepsis recognition and the appropriateness of treatment initiated by the RRT for patients who developed sepsis on the inpatient ward. Methods: Retrospective review of a prospectively collected quality improvement database from January 31, 2012 to February 28, 2013. Descriptive statistics were used in data analysis. Results: 36 patients were identified as septic by the RRT. Patients were 66% male and the average age was 51 years. 92% met SIRS criteria, 11% were in shock, 44% were culture positive, with 28% having positive blood cultures. Overall mortality was 11%. Infectious source was most frequently pulmonary (46%), followed by unknown (18%), urine (14%), catheter-related (11%), abdomen (7%), and primary bacteremia (4%). RRT accuracy in diagnosing sepsis was 78%. The RRT initiated treatment with fluids, antipyretics/cooling, and/or antibiotics in 71% of cases. Blood or urine cultures were sent 69% of the time and additional antibiotics were started 53% of the time with an average time to antibiotics of 180 minutes. Adequate fluid resuscitation (20 ml/kg) was administered only 29% of the time. Conclusions: RRTs can accurately identify and initiate treatment of admitted patients with sepsis. Additional improvement can be made by increasing the quantity of resuscitative fluid, increasing the proportion of patients treated with antibiotics, and reducing time to antibiotics.