Analysis of Antibiotic Exposure and Early-Onset Neonatal Sepsis in Europe, North America, and Australia

Abstract

ImportanceAppropriate use of antibiotics is life-saving in neonatal early-onset sepsis (EOS), but overuse of antibiotics is associated with antimicrobial resistance and long-term adverse outcomes. Large international studies quantifying early-life antibiotic exposure along with EOS incidence are needed to provide a basis for future interventions aimed at safely reducing neonatal antibiotic exposure.ObjectiveTo compare early postnatal exposure to antibiotics, incidence of EOS, and mortality among different networks in high-income countries.Design, Setting, and ParticipantsThis is a retrospective, cross-sectional study of late-preterm and full-term neonates born between January 1, 2014, and December 31, 2018, in 13 hospital-based or population-based networks from 11 countries in Europe and North America and Australia. The study included all infants born alive at a gestational age greater than or equal to 34 weeks in the participating networks. Data were analyzed from October 2021 to March 2022.ExposuresExposure to antibiotics started in the first postnatal week.Main Outcomes and MeasuresThe main outcomes were the proportion of late-preterm and full-term neonates receiving intravenous antibiotics, the duration of antibiotic treatment, the incidence of culture-proven EOS, and all-cause and EOS-associated mortality.ResultsA total of 757 979 late-preterm and full-term neonates were born in the participating networks during the study period; 21 703 neonates (2.86%; 95% CI, 2.83%-2.90%), including 12 886 boys (59.4%) with a median (IQR) gestational age of 39 (36-40) weeks and median (IQR) birth weight of 3250 (2750-3750) g, received intravenous antibiotics during the first postnatal week. The proportion of neonates started on antibiotics ranged from 1.18% to 12.45% among networks. The median (IQR) duration of treatment was 9 (7-14) days for neonates with EOS and 4 (3-6) days for those without EOS. This led to an antibiotic exposure of 135 days per 1000 live births (range across networks, 54-491 days per 1000 live births). The incidence of EOS was 0.49 cases per 1000 live births (range, 0.18-1.45 cases per 1000 live births). EOS-associated mortality was 3.20% (12 of 375 neonates; range, 0.00%-12.00%). For each case of EOS, 58 neonates were started on antibiotics and 273 antibiotic days were administered.Conclusions and RelevanceThe findings of this study suggest that antibiotic exposure during the first postnatal week is disproportionate compared with the burden of EOS and that there are wide (up to 9-fold) variations internationally. This study defined a set of indicators reporting on both dimensions to facilitate benchmarking and future interventions aimed at safely reducing antibiotic exposure in early life.

Reducing Antibiotic Exposure at the Beginning of Life

“…This study 4 highlights important considerations that can improve our understanding of both antibiotic exposure and EOS burden in early life and can stipulate future benchmarking and quality improvement initiatives. The findings indicate a large potential for antimicrobial stewardship in this area and should pave the way toward incentives designed to safely shift the balance toward more judicious use of antibiotics.…”

“…4 The strikingly wide range of antimicrobial usage across networks was not sufficiently explained by local EOS incidence, indicating that variability in practices to screen and treat neonates for EOS remains a key contributor to unrestrained antimicrobial usage in this age group. Intriguingly, although the authors 4 unfortunately do not provide a quantitative analysis of this aspect, this discrepancy seems not to relate to the types of protocols and strategies officially used in each of the networks. Furthermore, the median duration of treatment even in those without EOS was 4 days, 4 which is substantially longer than the time required to assess microbiological cultures and clinical response, implying that there were ample opportunities to shorten not only antibiotic duration, but also length of stay, with sizeable cost-saving potential.…”

“…Although the study design does not allow us to draw firm conclusions on population-based estimates, the report demonstrates that approximately 1 in 35 (2.86%) late preterm and term newborns were started on antibiotics, which contrasts with the observed incidence of proven EOS (0.49 cases per 1000 live births) and with the low sepsis-related mortality (12 potential EOS-associated deaths in 21 703 newborns treated for EOS). 4 The strikingly wide range of antimicrobial usage across networks was not sufficiently explained by local EOS incidence, indicating that variability in practices to screen and treat neonates for EOS remains a key contributor to unrestrained antimicrobial usage in this age group. Intriguingly, although the authors 4 unfortunately do not provide a quantitative analysis of this aspect, this discrepancy seems not to relate to the types of protocols and strategies officially used in each of the networks.…”

“…8 Implementation studies are warranted to explore clinicians' beliefs and practices in relation to guideline compliance. Importantly, the findings of Giannoni et al4 raise the question of whether antimicrobial stewardship initiatives, commonly focused on tertiary centers with in-house pediatric infectious disease specialists and pharmacists, fail to reach smaller clinics where the majority of newborns with suspected EOS are born and treated.Finally, careful titration of measures to reduce antimicrobial usage against the background incidence of EOS as proposed by Giannoni et al4 should be specific to the resource level of the respective health care setting and need to evaluate robust safety measures because of the low incidence of mortality.Giannoni et al4 conclude that in view of contemporary epidemiology, the exposure of neonates to antibiotics for suspected EOS is disproportionate. The proposed minimal set of indicators to benchmark antimicrobial usage against EOS incidence should serve as the basis for prospective studies designed to tackle the excessive use of antibiotics early in life.November 23, 2022. doi:10.1001/jamanetworkopen.2022.43705 7.…”

“…3,8 Implementation studies are warranted to explore clinicians' beliefs and practices in relation to guideline compliance. Importantly, the findings of Giannoni et al 4 raise the question of whether antimicrobial stewardship initiatives, commonly focused on tertiary centers with in-house pediatric infectious disease specialists and pharmacists, fail to reach smaller clinics where the majority of newborns with suspected EOS are born and treated.…”