BACKGROUND: Procalcitonin (PCT) has been suggested as a marker of bacterial infection with systemic manifestation. The purpose of this study was to determine whether PCT level can be used to discriminate between the sepsis or septic shock and localized infection or non-infection. MATERIALS AND METHODS: This was a prospective study involving 71 patients who presented with fever. The final diagnosis was inflammation without infection in 16, localized infection in 25, sepsis in 15, and septic shock in 15. We compared the different parameters of infection- erythrocyte sedimentation rate (ESR), C- reactive protein (CRP), interleukin-6 (IL-6), and PCT - by comparing the area under the receiver operating characteristic curves (AUCs) of sepsis/or septic shock and localized infection/or non-infection. We also determined the predictive power of PCT in detecting sepsis. RESULTS: The median PCT concentrations were 0.85 (range, 0.747-1.57) ng/mL for non-infection, 1.28 (0.73- 2.33) ng/mL for localized infection, 2.59 (1.87-9.0) ng/ mL for sepsis, and 23.9 (23.1-126.1) ng/mL for septic shock. PCT exhibited the highest discriminative value, with an AUC of 0.889 (95% CI, 0.81-0.97), followed by IL-6 (0.779; CI, 0.65-0.90), CRP (0.642; CI, 0.51-0.78), and ESR (0.412; CI, 0.28-0.56). At a cutoff value of 1.73 ng/mL, PCT showed sensitivity of 86.5%, specificity of 75.6%, positive predictive value of 72.2%, and negative predictive value of 88.6%. CONCLUSION: The PCT concentrations could be used to help discriminate sepsis in newly admitted febrile patients.