C. Berger et al., COMPARISON OF C-REACTIVE PROTEIN AND WHITE BLOOD-CELL COUNT WITH DIFFERENTIAL IN NEONATES AT RISK FOR SEPTICEMIA, European journal of pediatrics, 154(2), 1995, pp. 138-144
We prospectively compared the diagnostic value of C-reactive protein (
CRP) and white blood cell counts for detection of neonatal septicaemia
. Sensitivity and specifity in receiver operating characteristics, and
positive and negative predictive value of CRP and white blood cell co
unt were compared in 195 critically ill preterm and term newborns clin
ically suspected of infection. Blood cultures were positive in 33 case
s. During the first 3 days after birth CRP elevation (sensitivity 75%,
specifity 86%), leukopenia (67%/90%), neutropenia (78%/80%) and immat
ure to total neutrophil count (I/T) ratio (78%/73%) were good diagnost
ic parameters, as opposed to band forms with absolute count (84%/66%)
or percentage (79%/71%), thrombocytopenia (65%/57%) and toxic granulat
ions (44%/94%). Beyond 3 days of age elevated CRP (88%/87%) was the be
st parameter. Increased total (84%/66%) or percentage band count (79%/
71%) were also useful. Leukocytosis (74%/56%), increased neutrophils (
67%/65%), I/T ratio (79%/47%), thrombocytopenia (65%/57%) and toxic gr
anulations had a low specifity. The positive predictive value of CRP w
as 32% before and 37% after 3 days of age, that of leukopenia was 37%
in the first 3 days. Conclusion During the first 3 days of life CRP, l
eukopenia and neutropenia were comparably good tests while after 3 day
s of life CRP was the best single test in early detection of neonatal
septicaemia. Serial CRP estimations confirm the diagnosis, monitor the
course of infection and the efficacy of antibiotic treatment.