Early appropriate antibiotic therapy is an important component of the Surviving Sepsis Guidelines bundles that are associated with decreased in-hospital mortality. National antibiotic guidelines for the treatment of sepsis in Sweden have been available since 2008. Compliance with these guidelines is largely unknown, and whether it translates to improved patient outcome has not been studied.
To assess mortality and its relationship to compliance with Swedish antibiotic guidelines. A secondary aim was to assess the effect of timing of antibiotic administration and mortality.
A registry-based, retrospective cohort study. Registry data were supplemented by manual extraction of data on antibiotic treatment from patient charts. The association between guideline compliance and mortality was evaluated using multivariable analysis. Three levels of compliance were predefined: full compliance – correct antibiotics and dose; partial compliance – correct antibiotic but wrong dose and/or wrong initial antibiotic but corrected within 24 h and/or wrong combination in a combined regime that is at least one antibiotic not in line with the national antibiotic guideline; no compliance – incorrect antibiotic.
Two general ICUs in Sweden between 1 January 2011 and 31 December 2015.
Seven hundred and thirteen patients over the age of 18 with severe sepsis or septic shock identified through the Swedish ICU Registry.
MAIN OUTCOME MEASURES
The primary outcome was 30-day mortality.
Full compliance was observed in 47.0% of patients, partial compliance in 36.0%, and no compliance in 17.0%. Lack of compliance was independently associated with increased risk of 30-day mortality: the adjusted hazard ratio was 1.86 (95% CI 1.34 to 2.58 P
< 0.001) for partial compliance and 2.18 (95% CI 1.34 to 3.40 P
< 0.001) for no compliance. The time to first antibiotic administration was not associated with mortality.
Less than half of the patients with severe sepsis and septic shock received antibiotics according to Swedish national guidelines. Full compliance with the guidelines was associated with decreased mortality. The results of this study show that a strict approach to guideline compliance seems to be beneficial: half measures and inadequate doses should be avoided.