Key points are not available for this paper at this time.
To determine whether changes in procalcitonin (PCT) could be used to predict the onset of sepsis, daily PCT levels were monitored in 20 burned children. Analysis indicated a PCT rise of 5 ng/ml or greater as the best indication of sepsis. We compared the surgeons' determination of sepsis, which was based on changes in platelet count, C-reactive protein (CRP), and other clinical manifestations, with the prediction of sepsis from PCT. There were 26 septic episodes and 36 nonseptic episodes in the 20 patients. PCT results were classified as to true positives, false positives, true negatives, and false negatives. As an indicator of sepsis, the PCT assay had a sensitivity of 42%, a specificity of 67%, and an efficiency of 57%. Even when the assay correctly identified sepsis, the determination was made an average of 0.8 days after the surgeon had already made the diagnosis based on CRP and/or platelet count. We conclude that PCT is not as effective as CRP and/or platelet count in the early detection of sepsis in burned children.
Neely et al. (Thu,) studied this question.