PURPOSE: The mortality of severe sepsis exceeds 40%. Recently, the treatment of severe sepsis has focused on management guidelines which incorporate multiple therapeutic approaches. These guidelines include algorithms for resuscitation, vasopressors, antibiotic administration, blood transfusions, rhAPC, and glucose control. In 2005 our institution adopted the “surviving sepsis guidelines” to treat severe sepsis. When utilizing a multimodality approach, the effect of each individual therapy on outcome is unknown. The purpose of this study was to identify the individual components of the guidelines which might be most responsible for the improved outcome in patients with severe sepsis.
METHODS: Data from all sepsis patients were entered prospectively into an institutional database for assessment. All patients admitted from the ED to the ICU who met the ACCP/SCCM definition of severe sepsis between 7/2005 and 10/2006 were included. Data collected included: demographics, APACHE II, SVO2, glucose control, and the administration of antibiotics, vasopressors, fluids, blood products, and, rhAPC. Multiple stepwise logistic regression was performed with mortality as the dependent variable. All treatment modalities, age, and APACHE were used as predictors for outcome. Historical controls from the preceding 6 months (1/2005 –6/2005) were used to compare mortality prior to initiating the sepsis guidelines.
RESULTS: There were 176 patients with severe sepsis identified. Overall mortality was 31.3%. 118 historical controls were identified. The mortality of historical controls was 55.1%. Compared to historical controls, there was a significant decrease in absolute mortality of 23% (p < 0.01). The logistic regression revealed that APACHE II score on admission and time to antibiotic administration were the only two independent predictors of survival; APACHE (OR 1.13; 95% CI 1.08-1.19; p < 0.001) Antibiotics (OR 1.005; 95% CI 1.00-1.01; p < 0.05).
CONCLUSION: Sepsis management guidelines improve outcome; this improvement demonstrates multimodality synergy. When controlling for other therapies in the sepsis guidelines, time to antibiotic administration was the only independent predictor of mortality in patients with severe sepsis.
CLINICAL IMPLICATIONS: Antibiotic administration should take precedence in treating patients presenting with severe sepsis.
DISCLOSURE: Marc Zubrow, None.