Original Article
Before and after standardizing the controversial sepsis resuscitation bundle in a large hybrid academic center
Abstract
Background: Sepsis is a life-threatening disease associated with a high mortality and cost. Studies have shown that implementation of standardized measures in the processes of care can improve patient outcomes in sepsis. However, compliance with bundle recommendations varies across hospital systems. The purpose of this study was to analyze the clinical outcomes of patients before and after implementation of a standardized multidisciplinary sepsis protocol linked to a best practice alert (BPA).
Methods: The study was conducted in a 560-bed hybrid medical staff hospital consisting of both community and academic providers. All sepsis, severe sepsis, and septic shock patients based on ICD-9 coding admitted through the emergency department (ED) over a year were retrospectively reviewed prior to protocol implementation. We developed a multidisciplinary, standardized sepsis protocol based on the 2012 Surviving Sepsis Campaign (SSC) guidelines which included an electronic medical record (EMR) BPA, standardized ED and intensive care unit order sets, and a sepsis alert team. Sepsis alerts were called to streamline ICU admission. Intensive house wide education on the new process was performed and the protocol was implemented. We compared mortality, length of stay (LOS), time to fluids, antibiotics, and blood culture before and after protocol implementation for severe sepsis and shock patients.
Results: Prior to protocol implementation, 1,194 total patients were diagnosed with sepsis, severe sepsis, or septic shock; 134 (11%) initially presented with either severe sepsis or septic shock to the ED. Post implementation, 450 BPAs were triggered over a 1 year period. In total, 116 of these resulted in a “sepsis alert”, indicating severe sepsis or septic shock. Pre-protocol mortality was 56/134 (41.8%) versus 32/116 {27.6%; χ2[1]=5.50, P=0.024} post-protocol. There was a significant decrease in the time to obtaining blood cultures (pre: 219±42 min; post: 40±14 min, P<0.001), time to delivering antibiotics (pre: 273±59 min, post: 98±16 min, P=0.005), and time to fluids (pre: 190±52 min; post: 61±14 min, P=0.017) from initial ED vitals. Cultures, fluids, and antibiotics timing goals (within 3 hours) were achieved in 50.0% post implementation vs. 24.6% prior. LOS did not differ between groups {pre: 11.2±1.3; post: 9.7±0.8, t[247]=1.00, P=0.32}.
Conclusions: Timely sepsis identification, antibiotic delivery, fluid administration, and blood culture drawing significantly improved in our hospital after protocol implementation. There was a significant improvement in mortality with no statistically significant change in LOS. Standardized processes of care with education and practice alerts could improve sepsis outcomes.
Methods: The study was conducted in a 560-bed hybrid medical staff hospital consisting of both community and academic providers. All sepsis, severe sepsis, and septic shock patients based on ICD-9 coding admitted through the emergency department (ED) over a year were retrospectively reviewed prior to protocol implementation. We developed a multidisciplinary, standardized sepsis protocol based on the 2012 Surviving Sepsis Campaign (SSC) guidelines which included an electronic medical record (EMR) BPA, standardized ED and intensive care unit order sets, and a sepsis alert team. Sepsis alerts were called to streamline ICU admission. Intensive house wide education on the new process was performed and the protocol was implemented. We compared mortality, length of stay (LOS), time to fluids, antibiotics, and blood culture before and after protocol implementation for severe sepsis and shock patients.
Results: Prior to protocol implementation, 1,194 total patients were diagnosed with sepsis, severe sepsis, or septic shock; 134 (11%) initially presented with either severe sepsis or septic shock to the ED. Post implementation, 450 BPAs were triggered over a 1 year period. In total, 116 of these resulted in a “sepsis alert”, indicating severe sepsis or septic shock. Pre-protocol mortality was 56/134 (41.8%) versus 32/116 {27.6%; χ2[1]=5.50, P=0.024} post-protocol. There was a significant decrease in the time to obtaining blood cultures (pre: 219±42 min; post: 40±14 min, P<0.001), time to delivering antibiotics (pre: 273±59 min, post: 98±16 min, P=0.005), and time to fluids (pre: 190±52 min; post: 61±14 min, P=0.017) from initial ED vitals. Cultures, fluids, and antibiotics timing goals (within 3 hours) were achieved in 50.0% post implementation vs. 24.6% prior. LOS did not differ between groups {pre: 11.2±1.3; post: 9.7±0.8, t[247]=1.00, P=0.32}.
Conclusions: Timely sepsis identification, antibiotic delivery, fluid administration, and blood culture drawing significantly improved in our hospital after protocol implementation. There was a significant improvement in mortality with no statistically significant change in LOS. Standardized processes of care with education and practice alerts could improve sepsis outcomes.
