Purpose: This retrospective study analyzes the adequacy of the empirical antimicrobial treatment for sepsis at the emergency department according to the results of the microbiological isolations and describes the associations between microbiological variables and mortality. Values for several clinical and microbiological variables were prospectively collected for a total of 101 patients. Cases of inadequate antimicrobial treatment as defined in the text were further studied.
Results: Fifty-two patients had microbiological confirmation of infection (51.4%), constituting 62 different isolates. Escherichia coli represented half of the cases. One third of the microbiological isolates were not adequately covered by initially chosen treatment, mostly due to microbiological resistances. Having a respiratory source was the only variable associated to inadequate coverage (p=0.05). Lack of adequate empirical microbiological coverage was not associated with mortality (p=0.16). In a multivariate test, “respiratory source” (OR=46.6 [2.2-972; 95%CI] and “severity” (OR=42.5 [1.2- 1456; 95%CI] remained significantly associated to mortality.
Conclusions: Lack of empirical coverage for microbiological agents in sepsis is not uncommon. Institutional efforts are needed to improve the empirical use of antimicrobials for sepsis in the Emergency Department.