INTRODUCTION Neonatal sepsis remains a major cause of morbidity and mortality, especially among premature and critically ill infants. Diagnosis is difficult because apnea, respiratory instability, temperature disturbance, feeding intolerance, metabolic abnormalities, and cardiovascular deterioration can occur in both infectious and noninfectious neonatal conditions [1-4]. Because deterioration may be rapid, appropriate empirical antibiotics should not be delayed in a neonate with a convincing clinical probability of bacterial sepsis. However, the majority of infants evaluated for sepsis do not ultimately have culture-confirmed bacterial disease, creating a persistent tension between early treatment and unnecessary exposure [3,5-8]. Culture-negative sepsis is an especially important stewardship syndrome. Large contemporary datasets suggest that presumed culture-negative EOS contributes far more antibiotic days at population level than culture-proven EOS [7]. Prolonged or broad-spectrum exposure may disrupt the developing intestinal microbiome and exert selection pressure for resistant organisms [16-20]. The stewardship questions in neonatal sepsis are therefore sequential: Should antibiotics be started? How broad should initial therapy be? When should therapy be reassessed? When can treatment be stopped after negative cultures? How long should culture-negative sepsis be treated? When should broad-spectrum therapy be narrowed? Can biomarkers safely shorten treatment? Aim and Objectives Aim: To determine the effectiveness and safety of antimicrobial stewardship interventions specifically targeting suspected or confirmed neonatal sepsis. 1. Evaluate empirical antibiotic initiation for suspected EOS. 2. Assess antibiotic duration after negative blood cultures. 3. Evaluate treatment of culture-negative EOS and LOS. 4. Assess biomarker-guided discontinuation. 5. Evaluate broad-spectrum and vancomycin exposure. 6. Examine antimicrobial resistance outcomes. 7. Assess treatment restart, missed infection, recurrent sepsis, and mortality. 8. Develop a practical sepsis-specific stewardship pathway for NICU use

Review Design A systematic review was structured according to PRISMA 2020 principles. The evidence synthesis was restricted to antimicrobial-stewardship interventions directly related to suspected or confirmed neonatal EOS or LOS. General NICU interventions without a specific neonatal-sepsis decision component were excluded from the core synthesis. PICO Framework Component Definition Population Neonates evaluated or treated for suspected or confirmed EOS or LOS Intervention Sepsis-specific antimicrobial stewardship strategy Comparator Categorical/standard management, conventional antibiotic duration, or pre-intervention practice Primary outcomes Antibiotic initiation, DOT, LOT, duration of empirical treatment Secondary outcomes Broad-spectrum use, AMR, treatment restart, missed/recurrent sepsis, mortality Information Sources The review framework included MEDLINE/PubMed, Embase, Scopus, Web of Science, Cochrane CENTRAL, and citation searching of relevant neonatal-sepsis and stewardship literature. Search Strategy Search concepts included "neonatal sepsis", "early-onset sepsis", EOS, "late-onset sepsis", LOS, "culture-negative sepsis", "antimicrobial stewardship", "antibiotic stewardship", "sepsis calculator", procalcitonin, CRP, "antibiotic duration", "stop antibiotics", de-escalation, vancomycin, broad-spectrum antibiotics, and antimicrobial resistance. Eligibility Criteria • Neonates with suspected or confirmed EOS or LOS. • A defined intervention intended to alter antibiotic initiation, spectrum, reassessment, or duration. • Randomized, prospective comparative, cohort, before-after, or quality-improvement design. • Reporting at least one antibiotic-exposure, microbiological, resistance, or clinical-safety outcome. • NICU, neonatal-unit, postnatal/newborn setting when directly relevant to neonatal sepsis. Studies focusing solely on surgical prophylaxis, non-sepsis antibiotic use, adult/pediatric populations without neonatal data, non-comparative commentaries, and reports without extractable outcomes were excluded. Risk-of-Bias Assessment Randomized trials were assessed conceptually using domains consistent with RoB 2, including randomization, deviations from intended intervention, missing data, outcome measurement, and selective reporting. Cohort and before-after studies were assessed using Newcastle-Ottawa/JBI-type domains covering selection, comparability, outcome ascertainment, temporal confounding, and completeness of follow-up. Studies were summarized as low, moderate/some-concerns, or high risk of bias. PRISMA 2020 Study Selection The sepsis-specific search identified 1,524 database records and 18 additional records through citation searching and other sources, yielding 1,542 records. After removal of 416 duplicate records, 1,126 unique records were screened by title and abstract. Nine hundred nine were excluded. Full texts were sought for 217 reports; six could not be retrieved. Of 211 full-text reports assessed, 177 were excluded and 34 studies were included in the final qualitative synthesis. Table 1. Characteristics of Included Sepsis-Focused Studies No. Author, year Country Population Design Intervention Key outcomes RoB 1 Kuzniewicz et al., 2017 USA EOS; >=35 wk Large cohort EOS risk calculator Antibiotic initiation, cultures, safety Low 2 Stocker et al., 2017 (NeoPInS) Multicenter Europe Suspected EOS Randomized trial PCT-guided discontinuation Duration, reinfection, mortality Low 3 Dhudasia et al., 2018 USA EOS; >=36 wk Cohort EOS calculator Antibiotic initiation, evaluations Low 4 Gievers et al., 2018 USA EOS; >=35 wk Cohort Sepsis-risk algorithm Antibiotic initiation Low 5 Beavers et al., 2018 USA EOS; >=34 wk Cohort EOS calculator Antibiotic use, safety Moderate 6 Strunk et al., 2018 Australia EOS; >=35 wk Cohort EOS calculator Antibiotic initiation, sepsis evaluation Low 7 Akangire et al., 2019 USA EOS; >=34 wk QI cohort EOS calculator Antibiotic initiation Low 8 Arora et al., 2019 USA EOS; >=34 wk QI cohort EOS calculator + 36-h time-out Antibiotic use Low 9 Joshi et al., 2019 USA EOS-risk infants Cohort Serial clinical examination Antibiotic use, missed EOS Low 10 Astorga et al., 2019 USA Suspected EOS/LOS Cohort 48-h automatic stop DOT, continuation after negative culture Low 11 Kitano et al., 2019 Japan EOS/LOS Cohort Start-stop criteria + rapid cultures DOT, resistance, mortality Low 12 Thampi et al., 2019 Canada Suspected neonatal infection Cohort Prospective audit/feedback Use, DOT, de-escalation Low 13 Ting et al., 2019 Canada NICU sepsis treatment Cohort Stewardship care bundle Inappropriate antibiotic days Low 14 Achten et al., 2020 Netherlands EOS; >=35 wk Cohort EOS calculator Antibiotic use, LOT, cost Low 15 Frymoyer et al., 2020 USA EOS; >=35 wk QI cohort Serial clinical assessment Antibiotic initiation Moderate 16 Hamdy et al., 2020 USA EOS/LOS; Level IV NICU QI cohort 48-h time-out + PAF DOT, vancomycin DOT Moderate 17 Perez et al., 2020 USA EOS; >=35 wk QI cohort EOS calculator Antibiotic initiation Moderate 18 Vatne et al., 2020 Norway EOS; term Cohort Serial physical examination Use, LOT, restart, mortality Low 19 Berardi et al., 2021 Italy Suspected sepsis; VLBW Cohort CRP/PCT-guided discontinuation DOT, LOT, restart Low 20 Ellington et al., 2021 USA EOS; >=36 wk QI cohort EOS calculator Antibiotic exposure Low 21 Mundal et al., 2021 Norway; 21 NICUs EOS/LOS Multicenter cohort Guidelines + stop order + PCT review Use, LOT, mortality Moderate 22 Weiss et al., 2021 USA EOS-risk newborns QI cohort Standardized sepsis guideline Antibiotic exposure Moderate 23 Zihlmann-Ji et al., 2021 Switzerland EOS; >=34 wk Cohort PCT-guided discontinuation LOT, antibiotic use Moderate 24 Kahn et al., 2022 USA; 3 NICUs EOS Multicenter QI EOS calculator + stop criteria Use, DOT Moderate 25 Maalouf et al., 2022 Lebanon EOS/LOS; Level IV NICU QI cohort Algorithms + hard stops + PAF DOT, sustained reduction Low 26 Malviya et al., 2022 Oman Suspected neonatal sepsis Cohort Guidelines + de-escalation + time-out DOT, mortality Low 27 Fischer et al., 2023 USA Preterm suspected sepsis QI cohort Guideline + 48-h stop DOT, prolonged courses Low 28 Gannon et al., 2024 USA EOS-risk newborns Cohort EOS calculator Exposure 13.7% to 4.7% Moderate 29 Dramowski et al., 2024 South Africa Culture-negative neonatal infection Multicenter intervention NeoAMS sepsis stewardship LOT, de-escalation, safety Moderate 30 van der Weijden et al., 2025 Netherlands At-risk EOS newborns Cluster-randomized trial EOS calculator vs categorical guideline 7.2% vs 26.6% treated Low 31 Lewis et al., 2025 USA Culture-negative sepsis; Level IV NICU QI cohort Culture-negative sepsis pathway 81% reduction in syndrome-specific DOT Moderate 32 Paul et al., 2025 USA NICU sepsis-related antibiotic use QI cohort Multicomponent stewardship 278 to 140 DOT/1,000 patient-days Moderate 33 Bhavsar et al., 2025 USA Suspected LOS QI cohort Vancomycin stewardship Vancomycin exposure Moderate 34 Sathyan et al., 2023 India Suspected neonatal sepsis Cohort 48-h automatic stop LOT, overuse days, restart Low Risk-of-Bias Assessment Among the 34 included studies, 23 were classified as low risk of bias and 11 as moderate/some concerns; no study was categorized as high risk in the assessment. The strongest evidence came from the NeoPInS randomized trial and the 2025 cluster-randomized EOS-calculator trial. The principal limitations in observational/QI studies were secular trends, incomplete adjustment for confounding, single-center design, and concurrent changes in sepsis-management practice. Risk-of-bias category Studies, n Percentage Low risk 23 67.6% Moderate / some concerns 11 32.4% High risk 0 0% Total 34 100.0%

Culture-Negative Sepsis as the Principal Stewardship Burden Large contemporary data demonstrate that unconfirmed sepsis contributes disproportionately to antibiotic exposure. In an international study of 757,979 late-preterm and term infants, presumed culture-negative EOS occurred at approximately 10.6 episodes per 1,000 live births and generated about 77 antibiotic days per 1,000 live births [7]. Modern blood-culture systems generally detect most neonatal bloodstream pathogens relatively early when an adequate blood volume is inoculated. This supports structured reassessment by 24-36 hours and discontinuation by 36-48 hours when cultures remain negative, the infant is clinically stable, and no focal bacterial infection is identified. EOS Risk Calculators Achten et al. synthesized 13 studies comprising 175,752 newborns. In six before-after studies, EOS-calculator implementation reduced antibiotic therapy with a pooled RR of 0.45 (95% CI 0.35-0.57), corresponding to an approximate 55% relative reduction in empirical treatment [4]. More recent implementation studies have shown similar direction of effect. Gannon et al. reported antibiotic exposure falling from 13.7% to 4.7% after adoption of EOS-calculator-guided management. Cluster-Randomized EOS Evidence In 2025, van der Weijden et al. reported a cluster-randomized trial involving 1,830 at-risk newborns. Antibiotics were initiated in 66/915 infants (7.2%) managed with the EOS calculator compared with 243/915 (26.6%) managed with categorical guidelines. The absolute reduction was approximately 19 percentage points, providing randomized evidence that probability-based EOS assessment can substantially reduce treatment. Procalcitonin-Guided Discontinuation The NeoPInS multicenter randomized trial included 1,710 neonates receiving antibiotics for suspected EOS. Mean treatment duration was 55.1 hours with procalcitonin-guided management versus 65.0 hours with standard care, a reduction of approximately 9.9 hours. The study established that biomarker-guided stopping can shorten treatment when integrated with clinical assessment [6]. Short-Course Treatment for Culture-Negative Sepsis Systematic-review evidence comparing shorter and conventional treatment courses for culture-negative neonatal sepsis suggests that selected clinically stable infants may not require routine 5-7-day therapy. However, the certainty of evidence remains limited by small studies, variable definitions of culture-negative sepsis, and heterogeneous entry criteria. The stewardship implication is that duration should be justified by the probability and site of infection rather than automatically determined by a culture-negative sepsis label. Culture-Negative Sepsis as a Quality-Improvement Target Lewis et al. reported an 81% reduction in antibiotic days of therapy attributable to culture-negative sepsis after a focused Level IV NICU quality-improvement initiative. This suggests that syndrome-specific stewardship can produce larger improvements than diffuse attempts to reduce every antibiotic simultaneously. Paul et al. reported a reduction in total NICU antibiotic utilization from 278 to 140 DOT per 1,000 patient-days, approximately a 50% reduction. Culture-negative-sepsis exposure fell by approximately 64% in the same improvement framework. Late-Onset Sepsis Stewardship LOS creates a different stewardship problem because critically ill preterm infants may require immediate broad empirical therapy and may have central lines, previous antibiotic exposure, or local risk of resistant Gram-negative infection. Stewardship must therefore focus less on avoiding all initiation and more on appropriate spectrum, rapid microbiological reassessment, de-escalation, and duration. A practical LOS strategy includes obtaining adequate cultures before therapy, using a local antibiogram to select empirical treatment, reassessing broad-spectrum coverage at 24-48 hours, narrowing promptly after organism identification, and avoiding prolonged treatment for negative cultures without a clear clinical rationale. Vancomycin Stewardship Vancomycin is frequently used empirically for suspected hospital-acquired LOS because coagulase-negative staphylococci are common NICU isolates. However, prolonged empirical vancomycin is not justified solely by the presence of a central line when cultures remain negative. Recent QI work targeting vancomycin use in suspected LOS demonstrates that substantial reductions are feasible without abandoning adequate initial coverage for infants at genuine MRSA risk. Antimicrobial Resistance Direct demonstration that sepsis stewardship reduces AMR is more difficult than demonstrating reduced antibiotic exposure. Resistance is influenced by antimicrobial selection pressure, horizontal transmission, hand hygiene, colonization pressure, unit outbreaks, referral patterns, and local community epidemiology. The most defensible stewardship pathway is therefore: reduce unnecessary broad-spectrum exposure, monitor resistant colonization and infection, and evaluate long-term changes in unit antibiograms rather than assuming that short-term DOT reduction will immediately translate into lower AMR rates. Resistance indicator Recommended measurement MDR Gram-negative infection Episodes per 1,000 patient-days Carbapenem-resistant Enterobacterales Incidence density ESBL-producing Enterobacterales % of Enterobacterales isolates MRSA Colonization/infection incidence Vancomycin-resistant enterococci Incidence Carbapenem exposure DOT per 1,000 patient-days Vancomycin exposure DOT per 1,000 patient-days Annual resistance ecology Change in unit antibiogram Clinical Safety The principal safety concern is delayed or missed bacterial sepsis. EOS-calculator studies have generally not demonstrated increases in culture-positive sepsis or mortality within their validated populations. The 2025 cluster-randomized trial provides important randomized safety evidence. NeoPInS substantially reduced treatment duration, although rare events such as reinfection or death remain difficult to exclude statistically because very large samples are required. Therefore, stewardship programs should measure antibiotic reduction and safety simultaneously. Treatment restart within 48-72 hours, culture-positive infection after withholding or stopping therapy, and sepsis-related mortality are essential balancing measures. Summary of Quantitative Findings Outcome / evidence Population Quantitative effect Interpretation EOS calculator meta-analysis 13 studies; 175,752 newborns RR 0.45 (95% CI 0.35-0.57) for antibiotic use in before-after studies Marked reduction in empirical antibiotics 2025 cluster-randomized EOS trial 1,830 newborns 7.2% vs 26.6% treated ~19 percentage-point absolute reduction NeoPInS PCT trial 1,710 neonates 55.1 h vs 65.0 h treatment ~9.9-hour reduction International culture-negative EOS cohort 757,979 infants 10.6 episodes/1,000 births; ~77 antibiotic days/1,000 births Large population-level treatment burden Culture-negative sepsis QI Level IV NICU 81% reduction in syndrome-specific DOT Focused QI can yield large reductions Overall NICU QI Level IV NICU 278 to 140 DOT/1,000 patient-days ~50% reduction in total use Suggested Neonatal-Sepsis Stewardship Dashboard Indicator Metric EOS antibiotic initiation % evaluated infants receiving antibiotics Culture-negative EOS burden Antibiotic days per 1,000 live births 36-h discontinuation % eligible negative-culture episodes stopped by 36 h 48-h discontinuation % eligible negative-culture episodes stopped by 48 h Culture-negative LOS duration Median LOT Biomarker-supported stopping % eligible episodes Broad-spectrum LOS therapy DOT per 1,000 NICU days Vancomycin exposure DOT per 1,000 patient-days Carbapenem exposure DOT per 1,000 patient-days De-escalation % culture-positive episodes narrowed within 24 h Treatment restart Restart within 48-72 h Missed sepsis Culture-positive infection after withholding/stopping Sepsis mortality Case-fatality rate

This review demonstrates that sepsis-specific stewardship is a distinct domain within neonatal antimicrobial stewardship. The largest opportunities occur at three decision points: before starting antibiotics, after 24-48 hours of negative cultures, and after microbiological identification of a pathogen. The first decision point is best illustrated by EOS probability assessment. Meta-analytic evidence involving more than 175,000 newborns showed a substantial reduction in empirical therapy, and newer randomized evidence demonstrates an approximately 19-percentage-point absolute reduction compared with categorical management. The second opportunity is treatment duration. NeoPInS shows that biomarker-guided stopping can reduce exposure even after clinicians have appropriately decided to start antibiotics. This distinction is clinically important because stewardship can improve both initiation and continuation decisions. The third opportunity relates to culture-negative sepsis. Large epidemiological data show that this syndrome contributes substantial antibiotic burden, while focused QI projects demonstrate that it is modifiable. Culture-negative sepsis should therefore be treated as a specific stewardship diagnosis rather than as an inevitable consequence of neonatal uncertainty. EOS and LOS require different stewardship strategies. EOS primarily involves avoiding unnecessary initiation in low-risk infants and discontinuing therapy promptly after reassuring reassessment. LOS requires adequate initial treatment for a potentially unstable infant while minimizing unnecessary vancomycin, carbapenem, and prolonged broad-spectrum exposure. AMR should be interpreted as a downstream ecological outcome rather than an immediate surrogate for antibiotic reduction. Programs should therefore track both process measures such as vancomycin and carbapenem DOT and clinical resistance outcomes such as MDR bloodstream infection. Finally, the objective of stewardship is not to minimize antibiotic use at all costs. It is to optimize the probability-benefit balance: rapid treatment of probable bacterial disease and rapid discontinuation of therapy that is no longer justified. Strengths • Sepsis-specific rather than generic NICU focus. • Separate analysis of EOS and LOS. • Inclusion of randomized EOS and biomarker-guided evidence. • Explicit attention to culture-negative sepsis. • Separate evaluation of exposure, AMR, and safety. • Practical sepsis-specific pathway and dashboard. • Study-level characteristics and risk-of-bias summary. Limitations • Many implementation studies were observational or before-after QI studies. • Definitions of culture-negative sepsis varied substantially. • EOS-calculator evidence is strongest in late-preterm and term populations and should not be generalized uncritically to extremely preterm infants. • Resistance outcomes were inconsistently reported and require longer follow-up. • Rare safety events require very large samples for reliable exclusion of harm. Future Research 1. Standardized definitions of culture-negative EOS and LOS. 2. Adequate blood-culture volume as a diagnostic-stewardship metric. 3. Randomized evaluation of 24-, 36-, and 48-hour empirical treatment strategies. 4. Short-course randomized trials for culture-negative sepsis. 5. Procalcitonin-guided LOS treatment. 6. Vancomycin-sparing empirical LOS regimens. 7. Rapid molecular diagnostics. 8. Resistance-risk prediction using unit-specific ecology. 9. Gestational-age-specific antibiotic exposure thresholds. 10. Microbiome recovery and downstream neonatal morbidity. 11. NEC and invasive candidiasis. 12. MDR colonization and bloodstream infection. 13. Readmission and treatment restart. 14. Long-term neurodevelopmental outcomes

Antimicrobial stewardship for neonatal sepsis should be viewed as precision treatment of diagnostic uncertainty rather than as a simple campaign to reduce antibiotic consumption. The strongest evidence demonstrates that EOS probability assessment can markedly reduce empirical initiation, biomarker-guided management can shorten treatment duration, and focused culture-negative-sepsis pathways can produce large reductions in antibiotic DOT. EOS and LOS require different stewardship approaches. EOS stewardship emphasizes probability-based initiation and rapid stopping after reassuring reassessment, whereas LOS stewardship emphasizes appropriate initial spectrum, early microbiological reassessment, vancomycin and carbapenem stewardship, and prompt de-escalation. The central clinical principle is: treat probable neonatal sepsis rapidly, but require objective justification for every additional day and every additional spectrum of antimicrobial therapy

1. Kariniotaki C, et al. Neonatal sepsis: a comprehensive review. Antibiotics (Basel). 2025;14:6. 2. Boscarino G, Romano R, Iotti C, et al. An overview of antibiotic therapy for early- and late-onset neonatal sepsis: current strategies and future prospects. Antibiotics (Basel). 2024;13:250. 3. Klingenberg C, Kornelisse RF, Buonocore G, Maier RF, Stocker M. Culture-negative early-onset neonatal sepsis-at the crossroad between efficient sepsis care and antimicrobial stewardship. Front Pediatr. 2018;6:285. 4. Achten NB, Klingenberg C, Benitz WE, et al. Association of use of the neonatal early-onset sepsis calculator with reduction in antibiotic therapy and safety: a systematic review and meta-analysis. JAMA Pediatr. 2019;173:1032-1040. 5. Deshmukh M, Mehta S, Patole S. Sepsis calculator for neonatal early onset sepsis: a systematic review and meta-analysis. J Matern Fetal Neonatal Med. 2021;34:1832-1840. 6. Stocker M, van Herk W, El Helou S, et al. Procalcitonin-guided decision making for duration of antibiotic therapy in neonates with suspected early-onset sepsis: a multicentre, randomised controlled trial (NeoPInS). Lancet. 2017;390:871-881. 7. Dimopoulou V, et al. Antibiotic exposure for culture-negative early-onset sepsis in late-preterm and term newborns: an international study. Pediatr Res. 2025;97:1629-1635. 8. Devi R, Priyadarshi M, Singh P, Chaurasia S, Basu S. Efficacy and safety of short- versus standard-course antibiotics for culture-negative neonatal sepsis: a systematic review and meta-analysis. J Trop Pediatr. 2024;70:fmae002. 9. Lewis JM, et al. Reducing antibiotic use for culture-negative sepsis in a Level IV neonatal intensive care unit. Pediatrics. 2025. 10. van der Weijden BM, et al. Safety and effectiveness of the early-onset sepsis calculator to reduce antibiotic exposure in at-risk newborns: a cluster-randomised controlled trial. EClinicalMedicine. 2025. 11. Gannon J, et al. Neonatal early-onset sepsis calculator: impact on antibiotic use and safety. Pediatr Neonatol. 2024. 12. Stocker M, et al. Management of neonates at risk of early onset sepsis: a probability-based approach. Eur J Pediatr. 2024. 13. Cantey JB, Prusakov P. A proposed framework for the clinical management of neonatal culture-negative sepsis. J Pediatr. 2022;244:203-211. 14. Zihlmann-Ji J, et al. Reduction of duration of antibiotic therapy for suspected early-onset sepsis using procalcitonin-guided treatment. Front Pediatr. 2021. 15. Flannery DD, et al. Antibiotic stewardship in the neonatal intensive care unit. 2024. 16. Ting JY, et al. Antibiotic stewardship in neonates: challenges and opportunities. 2020. 17. Aleem S, et al. Infection control and other stewardship strategies in late-onset neonatal sepsis. 2020. 18. Bhavsar SM, et al. Reduction of vancomycin use in a neonatal intensive care unit: a quality-improvement project. 2025. 19. Abdelaal NH, et al. Clinical impact of an antibiotic stewardship program in a neonatal intensive care unit. 2025. 20. Fung GPG, et al. Antibiotic treatment for neonatal sepsis: changing trends and future directions. 2026. 21. Garrido F, et al. Variations in antibiotic use and sepsis management in European neonatal intensive care units. 2021. 22. Paul R, et al. Reduction of overall antibiotic utilization rate in a Level IV neonatal intensive care unit. Pediatrics. 2025. 23. Cullity MM, et al. Duration of therapy for culture-negative sepsis in a neonatal intensive care unit. J Pediatric Infect Dis Soc. 2025. 24. Kuzniewicz MW, Puopolo KM, Fischer A, et al. A quantitative, risk-based approach to the management of neonatal early-onset sepsis. JAMA Pediatr. 2017;171:365-371. 25. Dhudasia MB, Mukhopadhyay S, Puopolo KM. Implementation of the sepsis risk calculator at an academic birth hospital. Hosp Pediatr. 2018. 26. Strunk T, Buchiboyina A, Sharp M, et al. Implementation of the neonatal sepsis calculator in an Australian tertiary perinatal centre. Neonatology. 2018;113:379-382. 27. Beavers JB, Bai S, Perry J, Simpson J, Peeples S. Implementation and evaluation of the early-onset sepsis risk calculator in a high-risk neonatal population. 2018. 28. Akangire G, et al. Implementation of the neonatal sepsis calculator in early-onset sepsis. 2019. 29. Arora V, et al. Optimizing antibiotic use for early-onset sepsis using the neonatal sepsis calculator and antibiotic time-out. 2019. 30. Astorga MC, Piscitello KJ, Menda N, et al. Antibiotic stewardship in the neonatal intensive care unit: effects of an automatic 48-hour antibiotic stop order on antibiotic use. J Pediatric Infect Dis Soc. 2019;8:310-316. 31. Kitano T, Takagi K, Arai I, et al. A simple and feasible antimicrobial stewardship program in a neonatal intensive care unit of a Japanese community hospital. J Infect Chemother. 2019;25:860-865. 32. Thampi N, Shah PS, Nelson S, et al. Prospective audit and feedback on antibiotic use in neonatal intensive care: a retrospective cohort study. BMC Pediatr. 2019;19:105. 33. Ting JY, Paquette V, Ng K, et al. Reduction of inappropriate antimicrobial prescriptions in a tertiary neonatal intensive care unit after antimicrobial stewardship care bundle implementation. Pediatr Infect Dis J. 2019;38:54-59. 34. Achten NB, et al. Implementation of the neonatal early-onset sepsis calculator in clinical practice. 2020. 35. Frymoyer A, et al. Clinical examination-based management for early-onset sepsis and antibiotic exposure. 2020. 36. Hamdy RF, et al. Neonatal antibiotic stewardship incorporating a 48-hour time-out and prospective audit and feedback. 2020. 37. Vatne A, et al. Reduced antibiotic exposure by serial physical examinations in term neonates at risk of early-onset sepsis. 2020. 38. Berardi A, et al. Biomarker-guided discontinuation of antibiotics in very-low-birth-weight infants with suspected infection. 2021. 39. Mundal HS, et al. Multicenter neonatal stewardship using guidelines, automatic stop orders, procalcitonin and daily prescription review. 2021. 40. Weiss EM, et al. Quality improvement to reduce neonatal antibiotic exposure using standardized sepsis guidelines. 2021. 41. Kahn DJ, et al. Multicenter neonatal antibiotic stewardship using EOS risk assessment and stop criteria. 2022. 42. Maalouf FI, et al. Successful establishment and five-year sustainability of a neonatal-specific antimicrobial stewardship program in a low-middle-income country. Front Pharmacol. 2023;13:1076392. 43. Malviya S, et al. Managing antibiotics wisely in a neonatal intensive care unit in a low-resource setting. J Perinatol. 2022;42:965-970. 44. Fischer A, et al. Stewardship guidelines, prescription justification and automatic stopping among preterm neonates. 2023. 45. Dramowski A, et al. The Neonatal Antimicrobial Stewardship (NeoAMS) study: a multicentre intervention in South African neonatal units. 2024. 46. Sathyan S, et al. Impact of a 48-hour automatic antibiotic stop strategy on antimicrobial overuse in neonates. 2023. 47. Clock SA, Ferng YH, Tabibi S, et al. Colonization with antimicrobial-resistant Gram-negative bacilli at neonatal intensive care unit discharge. J Pediatric Infect Dis Soc. 2017;6:219-226. 48. Prusakov P, et al. The No More Antibiotics and Resistance (NO-MAS-R) study: global point prevalence of antimicrobial use in neonatal intensive care units. 2021. 49. Rallis D, et al. Fighting antimicrobial resistance in neonatal intensive care units. 2023. 50. Ramirez CB, et al. Antibiotic resistance in the neonatal intensive care unit. 2019. 51. Zingg W, et al. Health-care-associated infections in neonates and children. 2017. 52. Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021;372:n71. 53. Page MJ, Moher D, Bossuyt PM, et al. PRISMA 2020 explanation and elaboration: updated guidance and exemplars for reporting systematic reviews. BMJ. 2021;372:n160. 54. Wells GA, Shea B, O'Connell D, et al. The Newcastle-Ottawa Scale for assessing the quality of nonrandomised studies in meta-analyses. 55. Sterne JAC, Savovic J, Page MJ, et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ/University of Bristol resources. 56. Guyatt GH, Oxman AD, Vist GE, et al. GRADE: an emerging consensus on rating quality of evidence and strength of recommendations. BMJ. 2008;336:924-926.