We sought to determine whether the implementation of a sepsis protocol in a Canadian emergency department (ED) improves care for the subset of patients admitted to the intensive care unit (ICU).

After implementing a sepsis protocol in our ED we used an ICU database and chart review to compare various time-dependent end points and outcomes between a historical control year and the first year after implementation. We reviewed the charts of all patients admitted to the ICU within 24 hours of ED admission with a primary or other diagnosis of sepsis, severe sepsis or septic shock, who met criteria for early goal-directed therapy within the first 6 hours of their ED stay.

We compared 29 patients from the control year with 30 patients from the year after implementation of our sepsis protocol. We found that patients treated during the postintervention year had improvements in time to antibiotics (4.2 v. 1.0 h, difference = –3.2 h, 95% CI –4.8 to –2.0), time to central line placement (above the diaphragm) (11.6 v. 3.2 h, difference = –8.4 h, 95% CI –12.1 to –4.7), time to arterial line placement (7.5 v. 2.3 h, difference = –5.2 h, 95% CI –7.4 to –3.0), and achievement of central venous pressure and central venous oxygen saturation goals (11.1 v. 5.1 h, difference = –6.0 h, 95% CI –11.03 to –1.71, and 13.1 v. 5.5 h, difference = –7.6 h, 95% CI –11.97 to –3.16, respectively). There were no statistically significant differences in ICU length of stay, hospital length of stay or mortality (31.0% v. 20.0%, difference = –11.0%, 95% CI –33.1% to 11.1%).

Implementation of an ED sepsis protocol improves care for patients with severe sepsis and septic shock.