Evaluation of serum procalcitonin (PCT) levels has been suggested for diagnosis of infection, precise medical decision making and guidance for prescribing antibiotics in critically-ill patients. The aim of this study was to assess the effectiveness of PCT to shorten antibiotic treatment in critically-ill patients with bloodstream infections. Furthermore, the mortality and ICU length of stay (LOS) in such patients were secondary outcomes. Medline/PubMed, EMBASE, Scopus and Cochrane Databases were searched from January 1, 2007 to September 1, 2018. Randomized controlled trials (RCTs) on using PCT to guide antibiotic therapy compared with routine treatments for administration of antibiotics in critically-ill adult patients published in English were included. Two reviewers assessed the methodology of the studies included and extracted their data using the CONSORT checklist. Inverse-variance weighting and fixed and random effects meta-analyses were performed using the length of antibiotic treatment, LOS in an intensive care unit (ICU) and all-cause mortality. No significant reduction was found in the length of antibiotic treatment: although the cut-off point of 0.25<PCT<0.5ng/mL resulted in the reduced length of antibiotic treatment, this effect was not significant. Moreover, there was no significant reduction in ICU LOS and mortality. The analysis showed the effectiveness of the PCT cut-off level of 0.25<PCT<0.5ng/mL in decreasing the length of antibiotic treatment and ICU LOS, although this effect was not significant. Further studies are required to evaluate the results of this study on patients with recurrent infections, super-infections and also multidrug-resistant infections.