• D Pérez Solís, J B López Sastre, G D Coto Cotallo, María A Diéguez Junquera, E María Deschamps Mosquera, and M Crespo Hernández.
• Departamento de Pediatría, Hospital Universitario Central de Asturias, Oviedo, Spain.
• An Pediatr (Barc). 2006 Apr 1;64(4):349-53.

BackgroundNosocomial sepsis is a major problem in neonatal units. Because the clinical signs are nonspecific, highly reliable diagnostic markers are required to guide diagnosis. The aim of this study was to evaluate the utility of procalcitonin (PCT) as a diagnostic marker for nosocomial neonatal sepsis, and to compare the results of PCT with those of the most widely used laboratory tests for sepsis.Patients And MethodsTwenty neonates with nosocomial sepsis and 20 controls aged 4-30 days were included in a prospective study performed in a neonatal intensive care unit. PCT, C-reactive protein (CRP), leukocyte count, and the immature-to-total neutrophil ratio (I/T ratio) were measured at onset of signs of infection. The sensitivity, specificity, and likelihood ratio for a positive (LR+) and a negative (LR-) result were calculated.ResultsPCT, CRP, and the I/T ratio discriminated septic from nonseptic patients. Their areas under the ROC curve were 0.849, 0.880, and 0.884, respectively, with no statistically significant differences. Optimal cut-off values were: PCT > or = 0.65 ng/ml (sensitivity 85 %, specificity 80 %, LR 1 4.25, LR- 0.19), PCR > or = 5 .g/ml (sensitivity 80 %, specificity 95 %, LR 1 16, LR- 0.21), and I/T > or = 0.03 (sensitivity 90 %, specificity 75 %, LR 1 3.6, LR- 0.13).ConclusionsPCT may be a useful marker for the diagnosis of nosocomial neonatal sepsis. Studies with larger samples are required to compare the accuracy of PCT with that of other markers of sepsis.

19 keywords...

hide…