Friday, March 19, 2010
Grand Hall (Hyatt Regency Atlanta)
Background: Staphylococcus epidermidis is an important healthcare-associated pathogen in premature infants in the NICU, where it is the most common cause of late-onset sepsis and intravascular catheter infections. Hand hygiene of healthcare workers (HCWs) has been scrutinized, but transmission of S. epidermidis from parents to their infants has not been examined. One obstacle to such studies is that current molecular fingerprinting methods are cumbersome, limiting the number of strains that can be examined.
Objective: We sought to determine the sources of S. epidermidis colonization in newborns admitted to a level III NICU to see if they became colonized by particular S. epidermidis clones that might be associated with infections. We also wanted to see if infants acquired any of their S. epidermidis skin flora from their mothers, or if their strains were more often shared with HCWs and other babies. To accomplish this we developed a rapid DNA fingerprinting method to screen large numbers of staphylococcal isolates.
Methods: In a prospective 6-month colonization study, we swabbed the skin (chest and axillae) of 54 infants enrolled at birth and cultured twice weekly until a total of 4 swabs were collected. Nineteen additional infants of various ages were swabbed one time each. Samples from 65 of their mothers (9 had twins) and 42 HCWs were collected once from nares, dominant hand, and side of the neck. Samples were plated on Columbia CNA blood agar, and up to 8 typical (mannitol-negative) colonies were picked for DNA fingerprinting, using a novel highly variable single locus method. About 1500 isolates were screened, of which 1230 were S. epidermidis. Selected strains were further characterized by Multiple-Locus Variable number of tandem repeat Analysis (MLVA). We also examined 72 blood culture isolates from 14 NICU babies (not enrolled) who were diagnosed with infection by strict criteria during the period from 6 months before to 6 months after the colonization study. An additional 14 babies had blood culture isolates that were deemed to be contaminants.
Results: Over 90% of infants became colonized with S. epidermidis by the second week of life and carried 1-6 (mean 2.8) different types of S. epidermidis during the two-week study period. Three predominant clones (designated types 1, 3, and 4) were common both from skin of healthy enrolled babies and among blood culture isolates. Types 1 and 3 were also found in HCWs and mothers. Type 4 was frequently found on HCWs’ hands (but not their nares) and rarely found in mothers. Type 7 was not associated with infection but was found 3 times as often in mothers as babies, suggesting little mother to baby spread.
Objective: We sought to determine the sources of S. epidermidis colonization in newborns admitted to a level III NICU to see if they became colonized by particular S. epidermidis clones that might be associated with infections. We also wanted to see if infants acquired any of their S. epidermidis skin flora from their mothers, or if their strains were more often shared with HCWs and other babies. To accomplish this we developed a rapid DNA fingerprinting method to screen large numbers of staphylococcal isolates.
Methods: In a prospective 6-month colonization study, we swabbed the skin (chest and axillae) of 54 infants enrolled at birth and cultured twice weekly until a total of 4 swabs were collected. Nineteen additional infants of various ages were swabbed one time each. Samples from 65 of their mothers (9 had twins) and 42 HCWs were collected once from nares, dominant hand, and side of the neck. Samples were plated on Columbia CNA blood agar, and up to 8 typical (mannitol-negative) colonies were picked for DNA fingerprinting, using a novel highly variable single locus method. About 1500 isolates were screened, of which 1230 were S. epidermidis. Selected strains were further characterized by Multiple-Locus Variable number of tandem repeat Analysis (MLVA). We also examined 72 blood culture isolates from 14 NICU babies (not enrolled) who were diagnosed with infection by strict criteria during the period from 6 months before to 6 months after the colonization study. An additional 14 babies had blood culture isolates that were deemed to be contaminants.
Results: Over 90% of infants became colonized with S. epidermidis by the second week of life and carried 1-6 (mean 2.8) different types of S. epidermidis during the two-week study period. Three predominant clones (designated types 1, 3, and 4) were common both from skin of healthy enrolled babies and among blood culture isolates. Types 1 and 3 were also found in HCWs and mothers. Type 4 was frequently found on HCWs’ hands (but not their nares) and rarely found in mothers. Type 7 was not associated with infection but was found 3 times as often in mothers as babies, suggesting little mother to baby spread.
Conclusions: Premature infants in the NICU carry multiple S. epidermidis types, some of which appear to be common to the NICU environment and are associated with infection. Colonization patterns suggest that HCWs may be the main source of spread among babies in the NICU.