Antimicrobial stewardship in NICUs of low- and lower-middle-income countries – A narrative review

Introduction

Neonatal sepsis remains a leading cause of neonatal morbidity and mortality in low- and middle-income countries (LMICs), driving widespread empiric use of antibiotics in neonatal intensive care units (NICUs).^1–3 Surveys show that on any given day, c. 26% of NICU infants globally are being treated with antibiotics, with even higher proportions in LMIC settings.^4,5 Antibiotic usage rates in NICUs vary enormously (from <3% to ∼97% of patient-days) without corresponding differences in patient outcomes. ⁶ This suggests substantial overuse in many units, helping to fuel antimicrobial resistance (AMR), with proven increased risks of invasive candidiasis, necrotising enterocolitis and even mortality in preterm infants. LMIC NICUs face high rates of perinatal sepsis and often limited diagnostic resources, leading clinicians to err on the side of antibiotic over-treatment.

Antimicrobial stewardship programmes (ASPs) comprise coordinated interventions to optimise antibiotic use: using the right agent, dose and duration only when necessary, thereby improving patient outcomes and slowing resistance.^7,8 In high-income countries, NICU-focused ASPs have demonstrated reduced antibiotic utilisation without compromising safety. ⁹ Recognising the global AMR threat, the World Health Organization (WHO)'s action plan calls for AMS implementation in all countries, and a WHO toolkit now exists for LMICs. However, LMICs face unique challenges, such as limited laboratory capacity, high neonatal sepsis rates and often inconsistent drug supply or infection control.

In this narrative review, we examine the implementation and outcomes of NICU antimicrobial stewardship interventions in LMICs over the last ten years. We focussed on peer-reviewed studies (2015–2025) of stewardship within NICUs (excluding general paediatric or adult settings) reporting on intervention strategies (e.g., policies, audits, guidelines, antibiograms, stop orders) and their impacts on antibiotic use and clinical outcomes. We aimed to highlight practical stewardship approaches that have proven effective and discuss their relevance to NICUs in low-resource, high-burden settings, such as those served by Tropical Doctor readers.

Methods

We conducted a comprehensive literature search (2015–2025) in PubMed, Web of Science and Scopus for English-language articles on antimicrobial/antibiotic stewardship in NICUs of countries classified by the World Bank as low-income or lower-middle-income. Search terms included combinations of ‘neonatal’, ‘NICU’, ‘antimicrobial stewardship’, ‘antibiotic policy’, ‘low-income’ and ‘developing country’. We included interventional studies (e.g., quality improvement projects, before-and-after studies, trials) and relevant reviews that focused on NICU-level AMS interventions and reported outcomes on antibiotic use or clinical measures (e.g., infection rates, mortality). Studies from upper-middle or high-income countries, conference abstracts and grey literature were excluded.

In total, we identified 19 relevant studies from Asia, Africa and the Middle East that met our criteria. Given the heterogeneity of study designs and outcomes, we synthesised findings in a narrative format rather than a formal meta-analysis. We followed a broad narrative approach to ensure inclusion of diverse interventions. Key data (setting, intervention components and results) were extracted from each study for tabular presentation.

Results and thematic discussion of AMS implementation strategies in resource-limited NICUs

1. Developing Standard Treatment Guidelines and Policies. Unit-specific antibiotic guidelines were a common entry point for NICU stewardship in LMIC settings. In India, the introduction of an antibiotic policy increased the use of first-line ampicillin plus gentamicin regimens from 66% to 84% and reduced third-generation cephalosporin use. In comparison, another NICU reported a fall in overall antibiotic exposure from 58% to 46% after policy implementation with cephalosporin restriction.^10–12 In Egypt, an antibiogram-informed sepsis protocol eliminated inappropriate empiric therapy and was associated with lower mortality in late-onset sepsis. ¹³ These studies suggest that locally adapted guidelines can reduce unnecessary broad-spectrum therapy while improving appropriateness and, in some settings, clinical outcomes.

2. Prospective Audit-and-Feedback was one of the most consistently effective strategies. In the multicentre South African NeoAMS initiative, stewardship teams made 437 recommendations, 77% of which were accepted, and the mean antibiotic therapy duration fell from 9.1 to 6.9 days, with the largest reductions in culture-negative sepsis. ¹⁴ In northern India, daily audits embedded within PDSA cycles reduced antibiotic use by 32%, ¹² while a neonatal-specific programme in Lebanon achieved a 35% fall in antibiotic utilisation within 3 months and a sustained 63% reduction over 5 years. ¹¹ Together, these data show that routine review and feedback can shorten antibiotic exposure and help embed lasting changes in prescribing.

3. Education and Multidisciplinary Team Engagement supported implementation across settings. In Neonatal AMS, structured training and recurring case discussions increased clinician confidence in stopping antibiotics when infection was unconfirmed. ¹⁴ Pharmacist involvement also strengthened stewardship; in the Egyptian NICU, pharmacist interventions increased significantly during the programme, thereby optimising antimicrobial choice and dosing. ¹³ Reviews from India and other LMIC paediatric settings similarly conclude that multifaceted interventions combining guidelines, education, and feedback are more effective than passive policy dissemination alone.^15,16

4. Antibiotic Time-Outs and Automatic Stop Orders addressed a major source of NICU overuse: prolonged empiric treatment of culture-negative sepsis. Several programmes mandated reassessment at 48–72 h and used expected-duration benchmarks or stop orders at seven days unless continuation was justified.^11,14 These approaches reduced antibiotic duration without evidence of increased relapse, treatment failure or safety concerns, findings that are consistent with interrupted time-series and audit-based NICU studies showing no increase in mortality or readmissions despite lower antibiotic exposure.^17,18

5. Infection Control and Other Supplemental Strategies combined stewardship with infection prevention measures, recognising that lower infection burden reduces antibiotic pressure. In Indonesia, a bundled intervention involving hand hygiene, clean care practices, and prescribing policy reduced hospital-acquired infection rates from 22.6% to 8.6% while also halving inappropriate antibiotic use. ¹⁹ Other reports noted reductions in resistant organism burden after kerbing broad-spectrum prescribing, including lower ESBL infection rates in India and reduced multidrug-resistant colonisation in China.^12,17 These combined gains highlight the close relationship between stewardship, infection prevention and microbiological outcomes.

Overall, NICU stewardship interventions in LIC/LMIC settings consistently reduced antibiotic use, particularly prolonged empiric and broad-spectrum therapy, without worsening neonatal outcomes.^11–17,19 In some units, stewardship was also associated with improved survival, lower resistant-organism burden and wider hospital-level benefits including reduced drug expenditure and spread of stewardship practices beyond the NICU.^13,15

Several studies explicitly looked for evidence of missed infections or harm from reduced antibiotic use and found none.^11,17,18 This suggests much of antibiotic use was unnecessary to begin with.

Instead, mortality in neonates with late-onset sepsis decreased where a more targeted sepsis protocol was introduced, ¹³ underlining that choosing more effective empiric coverage based on local data translates into lives saved – a critical consideration in high-mortality NICUs, and several studies have shown reductions in hospital-acquired infection rates or colonisation by multidrug-resistant organisms.^12,16,17,19

Stewardship interventions in NICUs across LIC/LMIC contexts consistently reduced antibiotic use, whether by shortening empiric courses, kerbing broad-spectrum antibiotics or increasing adherence to narrow-spectrum protocols. Crucially, these reductions were attained without increasing neonatal morbidity or mortality. Our review of LMIC NICU studies concurs. ¹⁶ Also, by avoiding antibiotic overuse, NICUs may avert antibiotic-related complications (such as drug toxicity, fungal superinfections or necrotising enterocolitis associated with microbiome disruption).

Beyond the NICU, successful stewardship often had ripple effects in the wider hospital context. Stewardship committees went on to champion hospital-wide AMS initiatives. There are also obvious cost savings, an essential consideration for resource-limited hospitals. ¹⁵ Shorter hospital stays and fewer complications also relate to reduced costs.

Conclusion

The main message is that effective NICU stewardship depends less on sophisticated infrastructure than on clear protocols, regular review and sustained multi-disciplinary engagement.