Background: Neonatal sepsis, a severe infection in newborns, remains one of the leading causes of morbidity and mortality among preterm infants. This study aimed to investigate the distribution of pathogens responsible for early-onset sepsis (EOS) and late-onset sepsis (LOS), the annual variability of pathogens responsible for each type of infection, and potential trends in their profiles in preterm infants from a tertiary care neonatal intensive care unit over a ten-year period. Methods: We analyzed 177 episodes of confirmed bloodstream infection between 1 January 2014 and 31 December 2023. An episode of confirmed bloodstream infection was defined by the growth of a single potential pathogen in the blood of an infant who met four criteria: showing clinical symptoms of infection, having abnormal hematologic parameters, receiving appropriate antibiotics for less than 5 days, and being born before 37 weeks of gestation. Pathogens were isolated from aseptically collected blood samples, processed within 2 h, incubated using the BACTEC system, and identified by Gram stains and selective media cultures. Pathogen identification was performed using standardized biochemical tests or MALDI-TOF MS. Sepsis was classified as EOS if it occurred within the first 72 h of life and as LOS if it occurred after 72 h. Results: Among the confirmed bloodstream infections, EOS accounted for 31 cases, while LOS accounted for 146 cases. Escherichia coli was identified as the primary pathogen responsible for early-onset sepsis (EOS), while Coagulase-negative Staphylococcus (CoNS) was most commonly associated with late-onset sepsis (LOS). The differences in the prevalence of these bacteria between EOS and LOS were statistically significant. However, no significant differences were found in the distribution of pathogens across different years, nor were there significant trends in their frequency over the study period. Our results demonstrate significant stability in the distribution of pathogens causing sepsis over the ten-year observation period, even during the COVID-19 pandemic. Conclusions: Understanding the temporal distribution of pathogens in neonatal sepsis can help prevent the overuse of antibiotics and support the implementation of screening programs, empiric therapy, and strategies to prevent healthcare-associated infections.

Background: The goal of this study was to characterize the microbial etiology, antimicrobial susceptibility, and temporal resistance trends of early-onset neonatal sepsis (EOS) pathogens in a tertiary neonatal intensive care unit over 10 years (2014–2023), assessing empirical therapy adequacy and mortality associations. Methods: Retrospective analysis was performed on the positive blood cultures of neonates with confirmed EOS, born between 1 January 2014 and 31 December 2023. Blood was aseptically collected into PEDS Plus/BC bottles, incubated using the BACTEC system, with pathogen identification by biochemical assays or MALDI-TOF MS. Susceptibility testing followed EUCAST disk-diffusion standards, with additional resistance assays. Results: Among 6631 NICU admissions, 39 neonates met EOS criteria (31 preterm, 8 term). In preterm infants, Gram-negative Enterobacterales—mainly E. coli (n = 20)—predominated, while GBS was most common in term infants. All GBS isolates (n = 7) were susceptible to benzylpenicillin and vancomycin. Although 90% of E. coli were ampicillin-resistant, 90–95% remained susceptible to third-generation cephalosporins, piperacillin–tazobactam, and aminoglycosides. Two E. coli isolates produced ESBL but remained susceptible to aminoglycosides and carbapenems. Mortality was higher in E. coli EOS (50%) than in GBS (0%) or other pathogens (25%), with borderline significance (p = 0.0547; adjusted RR 1.55, 95% CI 0.54–4.41). Ampicillin resistance was not associated with increased mortality. No annual resistance trends were observed. Conclusions: In this 10-year NICU cohort, the etiology of EOS differed markedly between preterm and term neonates. Recommended empirical ampicillin–aminoglycoside therapy demonstrated in vitro efficacy against most neonatal bloodstream isolates pending pathogen identification. However, the widespread ampicillin resistance, particularly among E. coli strains, supports consideration of cephalosporin–aminoglycoside combinations or meropenem monotherapy when rapid beta-lactam bactericidal activity is clinically essential. Mortality was higher in E. coli EOS, though not statistically significant, and unrelated to ampicillin resistance.