The main objective was to evaluate if consensus “bundle” measures to optimize the treatment of sepsis have been integrated in our routine practice. We also tried to identify variables significantly associated to mortality. An observational, unblinded study of those patients who, according to their physicians, met criteria for sepsis under its different stages of severity was conducted. Six items as proposed by surviving sepsis campaign were evaluated: (1) measurement of blood lactate; (2) obtaining blood samples for culture before use of antibiotics; (3) early use of antibiotics (>3h); (4) fluid replacement with at least 20 mL/kg of crystalloids in the first hour, unless contraindicated, if hypotension or blood lactate >4 mmol/L; (5) use of vasoactive drugs, other than dopaminergic doses of dopamine, if hypotension or shock; (6) measurement of central venous pressure and central venous O 2 saturation in shock. Data concerning several aspects of diagnosis, treatment, and consultation to ICU were collected. Compliance with published guidelines is poor after evaluating the 6 items proposed: 12.5% of shocked patients had determinations of blood lactate; 15% of septic patients had no blood cultures drawn; 32% had received their first dose of antibiotics in the first 3 hours after admission; 46.6% of the cases of severe sepsis or shock received a fluid aggressive therapy; 43.3% of the patients with an indication for vasoactive drugs received them; no patient had central venous pressure monitoring at the ED. In addition, intensive care specialists were seldom consulted (17%). Having used dopaminergic doses of dopamine and having a respiratory focus as a source of infection were independently associated to mortality, respective OR: 21 (1.7-254.9; 95% CI) and 9.6 (1.7-52; 95% CI). The “bundle” measures proposed in the surviving sepsis campaign seem not to have had enough impact in our ED. This is the start point of a “plan-do-study-act” process directed to improve the outcome of patients with sepsis at our institution.