c Coagulase-negative staphylococci (CoNS) isolated in neonatal late-onset sepsis are often antibiotic resistant. We analyzed CoNS from skin and feces of neonates during hospitalization. Antibiotic resistance of skin isolates increased during hospitalization, especially in Staphylococcus haemolyticus. Staphylococcus warneri showed low antibiotic resistance. Our data suggest that different CoNS species may play distinct roles in colonization. C oagulase-negative staphylococci (CoNS) are the most frequent cause of late-onset sepsis among newborn infants in neonatal intensive care units (NICU) worldwide. Bloodstream isolates were frequently antibiotic resistant, similar to CoNS isolates from NICU personnel and from NICU sites (1, 2). Previously, it was shown that the majority of CoNS causing sepsis among neonates can be found on the hands of NICU personnel (1). Since the incidence of antibiotic-resistant CoNS in the nonmedical population is low, it is generally assumed that neonates become colonized with antibiotic-sensitive CoNS after birth. It is, however, unknown how skin and gut colonization with resistant CoNS develops during NICU hospitalization. A better understanding of CoNS colonization dynamics may assist the development of preventive strategies, for example, improvement of hygienic measures. We therefore studied CoNS colonization dynamics in neonates, focusing on the development of antibiotic resistance. We investigated skin and intestinal colonization, as well as maternal CoNS colonization after birth.This study was performed from mid November 2006 to mid March 2007 at the NICU of Erasmus MC-Sophia Children's Hospital, Rotterdam, The Netherlands. All infants born at this hospital with a presumed hospitalization time of at least 7 days (gestational age of Յ30 weeks, birth weight of Յ1,500 g or other reason) were included in this study. Children who were discharged within 72 h were excluded.We performed a longitudinal study of skin and intestinal carriage of CoNS among neonates and their mothers. Samples were taken from all infants 24, 48, and 72 h (Ϯ4 h) and, if still hospitalized, 7, 14, and 21 days after birth. Samples were taken from the mothers only once in the first 3 days and 7, 14, and 21 days after delivery. Skin samples from infants were obtained by gently pressing the bottom of a foot on a phenol-mannitol agar (PMA) plate (5% NaCl). Intestinal samples were obtained by culturing feces on PMA plates. Samples from mothers were obtained by culture of the thumb, by pressing it on a PMA plate (1). Selection, culturing, storage, species determination by internal transcribed spacer PCR, antibiotic susceptibility determination, and statistical analysis were performed as previously described (1), with the exception that in view of the multiple tests performed, we set the limit of significance at P ϭ 0.01 (two-sided), instead of the conventional P ϭ 0.05.Forty-one infants were initially included in the study. One infant was then excluded after being discharged within 72 h. General characteristics are summa...