Candidemia is a major infectious complication in neonatal patients. The isolation of yeasts from blood is still the "gold standard" for its diagnosis, but other laboratory markers (i.e., circulating antigens) have been studied with varying specificities and sensitivities. The aim of this study was to evaluate the role of procalcitonin for the diagnosis of candidemia in neonatal patients at high risk. To verify if the use of different commercial methods can highlight dissimilar results of sensitivity and/or specificity, the determination of procalcitonin serum levels was estimated by two systems. Overall, 90 patients from a Neonatal Intensive Care Units were enrolled, of whom six developed Candida bloodstream infection. Four of six infants with candidemia had slight increase of procalcitonin values (0.5-1 ng/mL). Only one baby showed very high levels but he had fungal and bacterial sepsis at the same time, while no elevation was observed in the sixth patient. No statistically significant difference was observed between two different methods at the time of monitoring (p>0.643). Both methods showed a sensitivity of 83.3 % at diagnosis, while the specificity was 73.8 and 63.1 % by methods A and B, respectively. In the light of the low sensibility and specificity of this assay, we can assume that the determination of procalcitonin would not seem to play a significant role in the diagnosis of fungal infection in neonatal patients.