BACKGROUND: Late-onset sepsis (LOS) in preterm infants is a leading cause of mortality and morbidity. Timely recognition and initiation of antibiotics are important factors for improved outcomes. Identification of risk factors could allow selection of infants at an increased risk for LOS. OBJECTIVES: The aim was to identify risk factors for LOS. METHODS: In this multicenter case-control study, preterm infants born at ≤30 weeks of gestation were included at 9 neonatal intensive care units. Detailed demographical and clinical data were collected daily up to day 28 postnatally. Clinical and demographic risk factors were identified using univariate and multivariate regression analyses in a 1: 1 matched case-control cohort. RESULTS: In total, 755 infants were included, including 194 LOS cases (41 gram-negative cases, 152 gram-positive cases, and 1 fungus). In the case-control cohort, every additional day of parenteral feeding increased the risk for LOS (adjusted OR = 1.29; 95% CI 1.07-1.55; p = 0.006), whereas antibiotics administration decreased this risk (OR = 0.08; 95% CI 0.01-0.88; p = 0.039). These findings could largely be attributed to specific LOS-causative pathogens, since these predictive factors could be identified for gram-positive, but not for gram-negative, LOS cases. Specifically cephalosporins administration prior to clinical onset was inversely related to coagulase-negative staphylococcus LOS (CoNS-LOS) development. Formula feeding was an independent risk factor for development of CoNS-LOS (OR = 3.779; 95% CI 1.257-11.363; p = 0.018). CONCLUSION: The length of parenteral feeding was associated with LOS, whereas breastmilk administration was protective against CoNS-LOS. A rapid advancement of enteral feeding, preferably with breastmilk, may proportionally reduce the number of parenteral feeding days and consequently the risk for LOS.