Effectiveness of Laboratory Markers in Determining Serious Bacterial Infection in Children with Fever without Source

GUNDUZ A., Tekin M. , Konca C., Turgut M.

JOURNAL OF PEDIATRIC INFECTIOUS DISEASES, cilt.13, ss.287-292, 2018 (SCI İndekslerine Giren Dergi) identifier identifier

  • Cilt numarası: 13 Konu: 4
  • Basım Tarihi: 2018
  • Doi Numarası: 10.1055/s-0038-1673666
  • Sayfa Sayıları: ss.287-292


We assessed the effectiveness of laboratory markers in determining the presence of serious bacterial infection (SBI) in well- appearing children aged 3 months to 17 years with fever without source (FWS). One hundred eighteen patients aged between 3 months and 17 years presenting to the pediatric department due to fever and with no focus of fever determined through history and physical examination were included in this prospective study. SBI was determined in 14 (11.9%) of the 118 patients. No significant difference was determined between the two groups in terms of age, sex, degree of fever, duration of fever, and mean platelet volume. The neutrophilto- lymphocyte ratio (NLR), C- reactive protein (CRP), and procalcitonin (PCT) were identified as independent predictive markers in a multivariate logistic regression analysis. PCT above 0.58 ng/ mL emerged as the most sensitive and specific test at receiver operating curve analysis (sensitivity: 92.8%, specificity: 90.3%, area under the curve: 0.908, and p < 0.001). The method with the highest positive predictive value, 69.2, was a laboratory score consisting of CRP, PCT, and complete urine analysis 3 (95% confidence interval [CI]: 38.6- 90.7), while the marker with the highest negative predictive value, 98.9, was PCT > 0.58 ng/ mL (95% CI: 94.3- 99.8). The NLR, CRP, and PCT were independent predictive markers in determining the presence of SBI in wellappearing children with FWS. PCT was the marker with the highest sensitivity and specificity in determining presence of SBI. The method with the highest positive predictive value was laboratory score, while PCTwas the marker with the best negative predictive value.