Prospective analysis of coagulase-negative staphylococcal infection in hospitalized infants

• Published in: The Journal of pediatrics Vol. 125; no. 5; pp. 798 - 804
• Main Authors: Nataro, James P., Corcoran, Linda, Zirin, Sharon, Swink, Sharon, Taichman, Norton, Goin, James, Harris, Mary Catherine
• Format: Journal Article
• Language: English
• Published: United States Mosby, Inc 01.11.1994; Mosby-Year Book, Inc
• Subjects: Bacteremia - diagnosis; Bacteremia - epidemiology; Bacteremia - microbiology; Catheterization, Central Venous - adverse effects; Coagulase - metabolism; Colony Count, Microbial; Cross Infection - diagnosis; Cross Infection - epidemiology; Cross Infection - microbiology; Humans; Incidence; Infant; Odds Ratio; Parenteral Nutrition - adverse effects; Prospective Studies; Regression Analysis; Risk Factors; Sensitivity and Specificity; Staphylococcal Infections - diagnosis; Staphylococcal Infections - epidemiology; Staphylococcal Infections - microbiology; Staphylococcus epidermidis - enzymology; Staphylococcus epidermidis - growth & development; Staphylococcus epidermidis - isolation & purification; Staphylococcus epidermidis - pathogenicity; Virulence; CONS; CFU
• ISSN: 0022-3476; 1097-6833
• DOI: 10.1016/S0022-3476(06)80186-1

Coagulase-negative staphylococci are the major cause of late-onset nosocomial neonatal sepsis. We prospectively examined all infants less than 6 months of age hospitalized at Children's Hospital of Philadelphia from whom at least one of two or more blood cultures grew coagulase-negative staphylococci. We considered as infections only those episodes in which multiple blood cultures grew identical isolates. Among 59 episodes that yielded specimens meeting study criteria, 25 were considered infection and 34 contamination. Cultures from infected infants yielded significantly higher numbers of coagulase-negative staphylococci than cultures representing contamination (p = 0.001). Infected infants were more likely to have central venous lines (p = 0.009), and to have received any parenteral nutrition (p = 0.002) or lipids (0.017). Hematologic values were not helpful in distinguishing between infected and uninfected infants. Isolates representing true infection were not different from contaminants in the frequency of positivity for putative virulence phenotypes. Our data corroborate previous studies indicating risk factors and predictors of coagulase-negative staphylococcal infection. (J PEDIATR 1994;125:798-804)