Abstract
BACKGROUND: The influence of pulmonary surfactant administration, long- term dexamethasone therapy, and presence of central venous catheters (CVCs) on microbial colonization of the trachea and neonatal septicemia is largely unknown. We examined these relationships using a case-control design. MATERIALS & METHODS: We retrieved data from the medical record and reviewed tracheal aspirate (TA) and blood (BL) cultures from 251 infants born at University Hospital. These infants, who were intubated for an average of 21 days, provided 607 TA (52% positive) and 470 BL (10% positive) cultures. Surfactant (Exosurf) was given to 158 patients; of these, 42 were matched by birthweight, race, and gender to 42 non-treated patients. Thirty-three of 67 dexamethasone-treated in- fants and 28 of 44 patients requiring CVCs were also similarly matched to con- trols. RESULTS: Tracheal colonization increased from 26% at 1 week to 94% at 4 weeks. Gram-negative flora increased from 10% of the first week's positive cul- tures to 30% of the fifth week's positive cultures, then declined. Yeast decreased from 10% of the first week's positive cultures to 5% of the positive cultures taken at 4 weeks. TA Gram-positive flora remained unchanged in composition through- out the period of observation-Staphylococcus epidermidis 34%, Staphylococcus haemolyticus 17%, Streptococcus species 13%, Staphylococcus aureus 5%, and mixed organisms 30%. In 36 patients, BL and TA samples, taken within 48 hours of each other, grew the same organism 13 times. Percent positive TA and BL cul- tures, types of pathogens, and pattern of tracheal colonization were similar for all study groups. Positive BL cultures occurred in 34% of the 44 patients with CVCs and in 9% of the remaining 207 infants. CONCLUSIONS: Neither surfactant nor dexamethasone alter tracheal colonization patterns nor increase the incidence of neonatal septicemia. Central venous catheterization does not alter tracheal colo- nization but along with birthweight and gestational age poses significant risk for neonatal septicemia (p = 0.004 and 0.04, respectively). Weekly TA cultures detect the introduction of pathogens capable of causing nosocomial infection but, in this group of patients, were of limited value in predicting late-onset septicemia.
Get full access to this article
View all access options for this article.
