Objectives: This study aimed to compare the efficacy ojumbilical cord blood levels oj procalcitonin (PCT) and C-reactive protein (CRP) as early predictors ojearly-onset sepsis (within 72 hours since delivery) in premature neonate admitted to NICU.
Patients & Methods: The study included 88 preterm neonates with mean gestational age oj 33.8±3; range: 29-38 weeks and mean birlh weight oj 1955±234: range: 1480-2350 gm with a mean 5-min Apgar score was 7.7±1.1; 7 neonates were small:for-gestational age and 23 neonates required resuscitation at birth. Neonates were categorized according to the presence ojsepsis into two groups: Irifected neonates had clinical manifestations oj sepsis and positive blood culture and Non-injected neonates showed no clinical manifestations and had negative blood culture at 72 hours since delivery. Two blood samples were obtained: umbilical cord blood samples obtained at time ojadmisSion to NICUJor estimation oj serum CRP and plasma PCT and a venous blood sample was obtained either at time ojdevelopment ojclinical signs ojsepsis or at 72 hours since delivery in non-injected groups Jor blood culture and complete blood count (CBC) to assure the clinical diagnosis ojinjected cases.
Results: Sixty neonates (68.2%) developed clinical signs ojsepsis and proved by blood culture to be injected. The mean levels ojCRP and PCT estimated in umbilical blood sample obtained at time oj admission to NICU were significantly higher (p0.6 ng / ml as the appropriate value for exclusion of neonatal sepsis with 100% sensitivity and negative predictive value (NPV) and specifICity rate of 93% and accuracy of diagnosis with rate of 97. 7%. Comparison of the diagnostic validity characters of umbilical cord plasma PCT (at cutoffpoint of>0.6 ng/ml) and umbilical cord serum CRP (at cutoffpoint of>10 mg/l) as an early predictor ofdevelopment of neonatal sepsis showed a signifICant difference in favor of plasma PCT. (X2= 3.19. p0.6 ng/ ml is appropriate for identification ofneonates at risk ofdeveloping sepsis with 100% sensitivity and negative predictive value. |
