OBJECTIVE: To identify the best combination of serum cytokines and clinical parameters to diagnose rapidly early-onset neonatal infection (EONI) in critically ill preterm infants. At birth, most critically ill neonates are receiving broad-spectrum antibiotics pending bacterial culture results, because distinguishing infected from noninfected infants at birth is difficult. DESIGN: Prospective study. SETTING: Neonatal intensive care unit in a tertiary care hospital. PATIENTS: Two hundred thirteen infants, born before 33 wks' gestation, admitted to the neonatal intensive care unit within 6 hrs of life with a presumptive diagnosis of EONI. INTERVENTION: A presumptive diagnosis of EONI was associated with a 300-μL blood sample to measure six cytokine (interleukin [IL]-1β, IL-6, IL-8, IL-10, IL-12, tumor necrosis factor-α) concentrations, using the cytometric bead array technique. MEASUREMENTS AND MAIN RESULTS: Of the 213 infants included, 31 had a definite or possible EONI and 182 were not infected. Concentrations of IL-6, IL-8, and IL-10 were significantly increased in infected neonates, in comparison with infants without EONI. In contrast, IL-1β, IL-12, and tumor necrosis factor-α concentrations were not. Logistic regression analyses were performed to construct multivariate predictive models that could distinguish infected from noninfected infants at birth. A clinical score was based on three parameters independently associated with EONI (i.e., interval of >12 hrs between the membranes rupture and delivery, prenatal maternal colonization and mechanical ventilation at birth). This score was compared with scores including clinical parameters and serum cytokines, alone or in combination. The best predictive model combined the three clinical parameters, IL-6 (positive threshold, 300 pg/mL) and IL-8 (positive threshold, 300 pg/mL) concentrations. CONCLUSION: A predictive model combining serum IL-6 and IL-8 measurements and selected clinical variables could distinguish infected from noninfected preterm infants at birth and should help the clinician in reducing or shortening the unnecessary use of antibiotics.
OBJECTIVE: To identify the best combination of serum cytokines and clinical parameters to diagnose rapidly early-onset neonatal infection (EONI) in critically ill preterm infants. At birth, most critically ill neonates are receiving broad-spectrum antibiotics pending bacterial culture results, because distinguishing infected from noninfected infants at birth is difficult. DESIGN: Prospective study. SETTING: Neonatal intensive care unit in a tertiary care hospital. PATIENTS: Two hundred thirteen infants, born before 33 wks' gestation, admitted to the neonatal intensive care unit within 6 hrs of life with a presumptive diagnosis of EONI. INTERVENTION: A presumptive diagnosis of EONI was associated with a 300-μL blood sample to measure six cytokine (interleukin [IL]-1β, IL-6, IL-8, IL-10, IL-12, tumor necrosis factor-α) concentrations, using the cytometric bead array technique. MEASUREMENTS AND MAIN RESULTS: Of the 213 infants included, 31 had a definite or possible EONI and 182 were not infected. Concentrations of IL-6, IL-8, and IL-10 were significantly increased in infected neonates, in comparison with infants without EONI. In contrast, IL-1β, IL-12, and tumor necrosis factor-α concentrations were not. Logistic regression analyses were performed to construct multivariate predictive models that could distinguish infected from noninfected infants at birth. A clinical score was based on three parameters independently associated with EONI (i.e., interval of >12 hrs between the membranes rupture and delivery, prenatal maternal colonization and mechanical ventilation at birth). This score was compared with scores including clinical parameters and serum cytokines, alone or in combination. The best predictive model combined the three clinical parameters, IL-6 (positive threshold, 300 pg/mL) and IL-8 (positive threshold, 300 pg/mL) concentrations. CONCLUSION: A predictive model combining serum IL-6 and IL-8 measurements and selected clinical variables could distinguish infected from noninfected preterm infants at birth and should help the clinician in reducing or shortening the unnecessary use of antibiotics.
Authors: Thor A Wagner; Courtney A Gravett; Sara Healy; Viju Soma; Janna C Patterson; Michael G Gravett; Craig E Rubens Journal: J Glob Health Date: 2011-12 Impact factor: 4.413