Objectives: Our primary goal was to evaluate the impact on in-hospital mortality rate of adequate empirical antibiotic therapy, after controlling for confounding variables, in a cohort of patients admitted to the intensive care unit (ICU) with sepsis. The impact of adequate empirical antibiotic therapy on early (<3 days), 28-day, and 60-day mortality rates also was assessed. We determined the risk factors for inadequate empirical antibiotic therapy.
Design: Prospective cohort study.
Setting: ICU of a tertiary hospital.
Patients: All the patients meeting criteria for sepsis at admission to the ICU.
Measurements and Main Results: Four hundred and six patients were included. Microbiological documentation of sepsis was obtained in 67% of the patients. At ICU admission, sepsis was present in 105 patients (25.9%), severe sepsis in 116 (28.6%), and septic shock in 185 (45.6%). By multivariate analysis, predictors of in-hospital mortality were Sepsis-related Organ Failure Assessment (SOFA) score at ICU admission (odds ratio [OR], 1.29; 95% confidence interval [CI], 1.19–1.40), the increase in SOFA score over the first 3 days in the ICU (OR, 1.40; 95% CI, 1.19–1.65), respiratory failure within the first 24 hrs in the ICU (OR, 3.12; 95% CI, 1.54–6.33), and inadequate empirical antimicrobial therapy in patients with “nonsurgical sepsis” (OR, 8.14; 95% CI, 1.98–33.5), whereas adequate empirical antimicrobial therapy in “surgical sepsis” (OR, 0.37; 95% CI, 0.18–0.77) and urologic sepsis (OR, 0.14; 95% CI, 0.05–0.41) was a protective factor. Regarding early mortality (<3 days), factors associated with fatality were immunosuppression (OR, 4.57; 95% CI, 1.69–13.87), chronic cardiac failure (OR, 9.83; 95% CI, 1.98–48.69) renal failure within the first 24 hrs in the unit (OR, 8.63; 95% CI, 3.31–22.46), and respiratory failure within the first 24 hrs in the ICU (OR, 12.35; 95% CI, 4.50–33.85). Fungal infection (OR, 47.32; 95% CI, 5.56–200.97) and previous antibiotic therapy within the last month (OR, 2.23; 95% CI, 1.1–5.45) were independent variables related to administration of inadequate antibiotic therapy.
Conclusions: In patients admitted to the ICU for sepsis, the adequacy of initial empirical antimicrobial treatment is crucial in terms of outcome, although early mortality rate was unaffected by the appropriateness of empirical antibiotic therapy.
Despite advances in medical practice, mortality rate of sepsis has not declined in the last decades, and this condition is associated with a high cost (1). Several large clinical trials have failed to reduce this mortality rate using different strategies to reduce the inflammatory response (2–4). Interestingly, a recent study demonstrated that recombinant human activated protein C reduced 28-day mortality rate in patients with severe sepsis (5). The kind of patients, source of infection, microbiology, and prognosis clearly differ between intensive care unit (ICU)-acquired episodes and those episodes that give rise to ICU admission (6).
A consensus conference defined systemic inflammatory response syndrome, sepsis, and related clinical entities (7). When systemic inflammatory response syndrome is the result of an infection, it is termed sepsis. However, microbiological documentation is not always feasible and cultures remain negative despite evident infection. Moreover, bacteremia traditionally has been associated with a poor outcome, especially with certain origins and when determined microorganisms are involved in critically ill patients with nosocomial bacteremia (8). Its exact role in the outcome of septic patients is uncertain (2, 3) although bacteremia was a risk factor of early mortality in severe sepsis (9).
Although antibiotic therapy is the cornerstone in the treatment of infections, the influence of adequate antimicrobial therapy on the prognosis of severe sepsis and septic shock has not been clearly proven (8, 10). Early initiation of adequate antibiotic treatment in bacteremic patients was associated with a reduction in mortality rate (11–13), although this benefit was not observed in critically ill patients (8). In these patients, the fatality is probably related to the age of the patients and the presence of underlying diseases and comorbidities (1, 14).
In addition, a great number of patients admitted to the ICU under the diagnosis of sepsis receive a surgical treatment of the focus that is essential and mandatory. In these patients, the usefulness of adequate empirical antibiotic therapy is unknown, and the evolution will probably depend more directly on a correct management of the source of infection (15). To our knowledge, no previous study has addressed the impact of empirical antibiotic therapy on the outcome of patients who are admitted to the ICU with sepsis.
Our hypothesis was that initial antibiotic treatment could influence the evolution of patients admitted to the ICU for sepsis. Hence, the primary goal of this study was to evaluate the impact on in-hospital mortality rate of adequate empirical antibiotic therapy, after controlling for confounding variables. Secondary objectives were to assess the impact of adequate empirical antibiotic therapy on early (3 days), 28-day, and 60-day mortality rates as well as on the length of stay. We also set out to identify the reasons for the administration of inadequate empirical antibiotics on admission to the ICU.