OBJECTIVES: The aim of the research was to determine the most common predisponing factors, primary sites of infection and causing organisms of sepsis and their antimicrobial susceptibility and to examine promptness and adequacy of antibiotic choice after microbiological determination of the causing microorganism and it's antimicrobial susceptibility. MATERIALS AND METHODS: The study included 106 patients with a discharge diagnosis of sepsis, treated at Clinic for infectious diseases of University Hospital of Split in the year of 2017. Retrospective observational study was performed by insight into the patients' history of ilness and discharge letters. RESULTS: At least one predisponing factor was confirmed in approximately 75% patients; age of 70 years or more in 60%, diabetes mellitus and neurological diseases in 20%, urinary catether in 12% and chronic heart disease in 11% of patients. The most common primary sites of infection were urinary tract infections (UTI), soft tissue infections and organ abscesses. The most common isolated causes were E. coli, K. pneumoniae, P. mirabili, methicillin-sensitive S. aureus and M. morgagnii, while the highest rate of ESBL production and resistance to amoxicillin/clavulanic acid and ciprofloxacin was observed in K. pneumoniae, P. mirabilis and E. coli. In 70% of patients with origin of sepsis in UTI it was not necessary to adjust empirical antibiotic therapy or the adjustment was justified, while in the remaining 30% the empirical therapy was changed or was not changed unjustifiably. In case of sepsis with origin in UTI the most often empirically used antibiotic was amoxicillin/clavulanic acid, followed by ceftriaxone and carbapenems. In case of sepsis with origin in soft tissue infection the most common empirically used antibiotics were carbapenems, with or without clindamycin, in approximately 90% of cases the therapy was adjusted justifiably, and in 10% of cases it was changed unjustifiably. Generally, if the spectrum of antimicrobial activity needed adjustment, it was performed within 3 days of receiving the microbiological laboratory report in approximately 85% cases. Clinical outcome was recovery in approximately 80% and death in 15% of patients, while the outcome remained unknown in 5% of patients. There was statistically significant difference between death and age of 70 years or more, chronic heart disease, presence of 3 or more predisponing factors, infection with MRSA or ESBL-producing bacteria and justified or unjustified change of antibiotic spectrum. CONCLUSION: The risk for developing sepsis is higher if at least one predisponing factor is present, including age of 70 years or more, female gender, diabetes mellitus or neurological illness. Empirical antibiotic therapy is most commonly left unchanged, while if the microbiological laboratory report required it, the change was performed within 3 days. Due to a high rate of antimicrobial resistance of the most common causing microorganisms of sepses with the origin in UTI, first choice antimicrobial therapy should be an antibiotic with a higher probability of empirical efficiency.