OBJECTIVES: To assess the diagnostic accuracy of white blood cell count (WBC), absolute neutrophil count (ANC), and C-reactive protein (CRP) in detecting severe bacterial infections (SBI) in well-appearing neonates with early onset fever without source (FWS) and in relation to fever duration. METHODS: An observational study was conducted on previously healthy neonates 7 to 28 days of age, consecutively hospitalized for FWS from less than 12 hours to a tertiary care Pediatric Emergency Department, over a 4-year period. Laboratory markers were obtained upon admission in all patients and repeated 6 to 12 hours from admission in those with normal values on initial determination. Sensitivity, specificity, positive and negative likelihood ratios, and receiver operating characteristic analysis were carried out for primary and repeated laboratory examinations. RESULTS: Ninety-nine patients were finally studied. SBI was documented in 25 (25.3%) neonates. Areas under receiver operating characteristic curves were 0.78 (95% CI, 0.69-0.86) for CRP, 0.77 (95% CI, 0.67-0.85) for ANC and 0.59 (95% CI, 0.49-0.69) for WBC. Sixty-two patients presented normal laboratory markers on initial determination. Of these, 58 successfully underwent repeated blood examination at >12 hours from fever onset. Five of them had an SBI. The area under curve calculated for repeated laboratory tests showed better values, respectively of 0.99 (95% CI, 0.92-1) for CRP, 0.85 (95% CI, 0.73-0.93) for ANC and 0.79 (95% CI, 0.66-0.88) for WBC. CONCLUSIONS: In well-appearing neonates with early onset FWS, laboratory markers are more accurate and reliable predictors of SBI when performed after >12 hours of fever duration. ANC and especially CRP resulted better markers than the traditionally recommended WBC.
OBJECTIVES: To assess the diagnostic accuracy of white blood cell count (WBC), absolute neutrophil count (ANC), and C-reactive protein (CRP) in detecting severe bacterial infections (SBI) in well-appearing neonates with early onset fever without source (FWS) and in relation to fever duration. METHODS: An observational study was conducted on previously healthy neonates 7 to 28 days of age, consecutively hospitalized for FWS from less than 12 hours to a tertiary care Pediatric Emergency Department, over a 4-year period. Laboratory markers were obtained upon admission in all patients and repeated 6 to 12 hours from admission in those with normal values on initial determination. Sensitivity, specificity, positive and negative likelihood ratios, and receiver operating characteristic analysis were carried out for primary and repeated laboratory examinations. RESULTS: Ninety-nine patients were finally studied. SBI was documented in 25 (25.3%) neonates. Areas under receiver operating characteristic curves were 0.78 (95% CI, 0.69-0.86) for CRP, 0.77 (95% CI, 0.67-0.85) for ANC and 0.59 (95% CI, 0.49-0.69) for WBC. Sixty-two patients presented normal laboratory markers on initial determination. Of these, 58 successfully underwent repeated blood examination at >12 hours from fever onset. Five of them had an SBI. The area under curve calculated for repeated laboratory tests showed better values, respectively of 0.99 (95% CI, 0.92-1) for CRP, 0.85 (95% CI, 0.73-0.93) for ANC and 0.79 (95% CI, 0.66-0.88) for WBC. CONCLUSIONS: In well-appearing neonates with early onset FWS, laboratory markers are more accurate and reliable predictors of SBI when performed after >12 hours of fever duration. ANC and especially CRP resulted better markers than the traditionally recommended WBC.