Advertisement
  • Loading metrics

Gram-negative neonatal sepsis in low- and lower-middle-income countries and WHO empirical antibiotic recommendations: A systematic review and meta-analysis

  • Sophie C. H. Wen ,

    Roles Conceptualization, Data curation, Investigation, Methodology, Project administration, Validation, Writing – original draft, Writing – review & editing

    Sophie.wen@health.qld.gov.au

    Affiliations Centre for Clinical Research, University of Queensland, Brisbane, Queensland, Australia, Infection Management Prevention Service, Queensland Children’s Hospital, Brisbane, Queensland, Australia

  • Yukiko Ezure,

    Roles Data curation, Formal analysis, Investigation, Methodology, Validation, Writing – review & editing

    Affiliations Centre for Clinical Research, University of Queensland, Brisbane, Queensland, Australia, School of Public Health, University of Queensland, Brisbane, Queensland, Australia

  • Lauren Rolley,

    Roles Data curation, Validation, Writing – review & editing

    Affiliation Infection Management Prevention Service, Queensland Children’s Hospital, Brisbane, Queensland, Australia

  • Geoff Spurling,

    Roles Methodology, Supervision, Writing – review & editing

    Affiliation Primary Care Clinical Unit, University of Queensland, Brisbane, Queensland, Australia

  • Colleen L. Lau,

    Roles Methodology, Supervision, Writing – review & editing

    Affiliations School of Public Health, University of Queensland, Brisbane, Queensland, Australia, Research School of Population Health, Australian National University, Canberra, Australian Capital Territory, Australia

  • Saba Riaz,

    Roles Supervision, Writing – review & editing

    Affiliation Institute of Microbiology and Molecular Genetics, University of the Punjab, Lahore, Pakistan

  • David L. Paterson,

    Roles Methodology, Supervision, Writing – review & editing

    Affiliations Centre for Clinical Research, University of Queensland, Brisbane, Queensland, Australia, Royal Brisbane and Women’s Hospital, Brisbane, Queensland, Australia

  • Adam D. Irwin

    Roles Conceptualization, Formal analysis, Methodology, Software, Supervision, Visualization, Writing – review & editing

    Affiliations Centre for Clinical Research, University of Queensland, Brisbane, Queensland, Australia, Infection Management Prevention Service, Queensland Children’s Hospital, Brisbane, Queensland, Australia

Gram-negative neonatal sepsis in low- and lower-middle-income countries and WHO empirical antibiotic recommendations: A systematic review and meta-analysis

  • Sophie C. H. Wen, 
  • Yukiko Ezure, 
  • Lauren Rolley, 
  • Geoff Spurling, 
  • Colleen L. Lau, 
  • Saba Riaz, 
  • David L. Paterson, 
  • Adam D. Irwin
PLOS
x

Abstract

Background

Neonatal sepsis is a significant global health issue associated with marked regional disparities in mortality. Antimicrobial resistance (AMR) is a growing concern in Gram-negative organisms, which increasingly predominate in neonatal sepsis, and existing WHO empirical antibiotic recommendations may no longer be appropriate. Previous systematic reviews have been limited to specific low- and middle-income countries. We therefore completed a systematic review and meta-analysis of available data from all low- and lower-middle-income countries (LLMICs) since 2010, with a focus on regional differences in Gram-negative infections and AMR.

Methods and findings

All studies published from 1 January 2010 to 21 April 2021 about microbiologically confirmed bloodstream infections or meningitis in neonates and AMR in LLMICs were assessed for eligibility. Small case series, studies with a small number of Gram-negative isolates (<10), and studies with a majority of isolates prior to 2010 were excluded. Main outcomes were pooled proportions of Escherichia coli, Klebsiella, Enterobacter, Pseudomonas, Acinetobacter and AMR. We included 88 studies (4 cohort studies, 3 randomised controlled studies, and 81 cross-sectional studies) comprising 10,458 Gram-negative isolates from 19 LLMICs. No studies were identified outside of Africa and Asia. The estimated pooled proportion of neonatal sepsis caused by Gram-negative organisms was 60% (95% CI 55% to 65%). Klebsiella spp. was the most common, with a pooled proportion of 38% of Gram-negative sepsis (95% CI 33% to 43%). Regional differences were observed, with higher proportions of Acinetobacter spp. in Asia and Klebsiella spp. in Africa. Resistance to aminoglycosides and third-generation cephalosporins ranged from 42% to 69% and from 59% to 84%, respectively. Study limitations include significant heterogeneity among included studies, exclusion of upper-middle-income countries, and potential sampling bias, with the majority of studies from tertiary hospital settings, which may overestimate the burden caused by Gram-negative bacteria.

Conclusions

Gram-negative bacteria are an important cause of neonatal sepsis in LLMICs and are associated with significant rates of resistance to WHO-recommended first- and second-line empirical antibiotics. AMR surveillance should underpin region-specific empirical treatment recommendations. Meanwhile, a significant global commitment to accessible and effective antimicrobials for neonates is required.

Author summary

Why was this study done?

  • Neonatal sepsis is a significant cause of childhood mortality with marked disparities across world regions and countries.
  • Gram-negative bacteria are becoming increasingly predominant in neonatal sepsis, particularly in low- and middle-income countries, along with growing concerns of multidrug resistance.
  • It is likely that the World Health Organization (WHO) empirical neonatal sepsis antibiotic recommendations are no longer adequate. We analysed data on neonatal sepsis from low- and lower-middle-income countries (LLMICs) to assess the burden of Gram-negative infections and associated antimicrobial resistance (AMR).

What did the researchers do and find?

  • We pooled data from 88 published studies from 19 LLMICs on neonatal sepsis (with a total of 10,458 Gram-negative isolates) and summarised the proportions of Gram-negative sepsis caused by E. coli, Klebsiella, Enterobacter, Pseudomonas, and Acinetobacter. We also examined the rates of AMR against key empirical antibiotics used in neonatal sepsis.
  • Gram-negative bacteria accounted for 60% of neonatal sepsis in LLMICs. Regional variations in types of Gram-negative bacteria were observed, along with high rates of resistance against the WHO-recommended empirical antibiotics for neonatal sepsis.

What do these findings mean?

  • The WHO empirical antibiotic recommendations for neonatal sepsis are likely inadequate in many LLMICs.
  • Robust AMR surveillance and reporting in LLMICs is necessary to develop region-specific empirical antibiotic recommendations for neonatal sepsis.

Introduction

Neonatal sepsis is a major cause of mortality and morbidity, accounting for approximately 22% of global annual neonatal deaths [1]. Improvements in neonatal mortality over the last 30 years have occurred at a slower rate than those observed for post-neonatal mortality, and the neonatal period contributes greater than 40% of all mortality in children under 5 years of age. The Sustainable Development Goals (SDGs) target a reduction of neonatal mortality in all countries to less than 12 deaths per 1,000 live births by 2030 [1,2]. There is significant variation in the reported incidence of neonatal sepsis worldwide, with a paucity of data particularly from low-income countries. In high- and middle-income countries, it has been estimated that neonatal sepsis occurs in 2,200 neonates per 100,000 live births, equating to 3 million cases of neonatal sepsis annually, with a mortality rate of 11% to 19% [3]. The incidence of neonatal sepsis in middle-income countries has been reported to be up to 40 times higher than in high-income countries [3].

Neonatal sepsis is historically categorised as either early-onset sepsis (EOS) or late-onset sepsis (LOS), with EOS variably defined as sepsis within 72 hours or up to 7 days of birth. EOS is traditionally thought to be caused by organisms such as group B streptococcus and enteric Gram-negative bacteria, acquired peripartum from the maternal genital tract. LOS, on the other hand, is considered to arise due to the acquisition of pathogens during hospitalisation, with very low birth weight and early gestational age being strong risk factors [4]. Increasingly, this distinction is called into question by reports, particularly from low- and middle-income countries (LMICs), of a predominance (up to 64% [5]) of Gram-negative and hospital-associated infections in both EOS and LOS [46]. There has been a worldwide increase in the prevalence of Gram-negative neonatal sepsis, with an alarming upward trend in multidrug-resistant (MDR) infections [47]. It has been estimated that globally 214,000 neonatal sepsis deaths are attributable to resistant pathogens each year [8]. Access to antimicrobials remains a significant barrier for many neonates and children in LMICs [8] and has resulted in an increase in neonatal mortality [9].

The World Health Organization (WHO) recommends the use of gentamicin with either ampicillin or benzylpenicillin as first-line treatment for neonatal and paediatric sepsis in resource-limited settings, with ceftriaxone as recommended second-line therapy [10]. Recent systematic reviews of antimicrobial resistance (AMR) in neonates and children in sub-Saharan Africa have highlighted the increasing prevalence of resistance to these antibiotics, particularly in Gram-negative bacteria [11,12]. As a consequence, antibiotic prescribing practices in neonatal and paediatric sepsis have been shown to significantly diverge from the WHO recommendations [13]. In light of the increasing evidence of multidrug resistance in Gram-negative neonatal sepsis, we performed a systematic review of published data on Gram-negative neonatal sepsis from all low- and lower-middle-income countries (LLMICs) from 1 January 2010 to 21 April 2021 to assess the appropriateness of current WHO first- and second-line antimicrobial recommendations.

Methods

Search strategy and selection criteria

This systematic literature search and review was performed using a predesigned study protocol (published in PROSPERO, CRD42020181110) and adheres to the PRISMA guideline for reporting systematic reviews and meta-analyses (see S1 Table). We searched Ovid MEDLINE, Embase, PubMed, CENTRAL (Cochrane Central Register of Controlled Trials), Web of Science, LILACS (Latin American and Caribbean Health Sciences Literature), WHOLIS, MedCarib, African Journals Online, African Index Medicus, IMSEAR (Index Medicus for South-East Asia Region), IMEMR (Index Medicus for the Eastern Mediterranean Region), WPRIM (Western Pacific Region Index Medicus), IndMED, Google, and OpenGrey to identify studies published from 1 January 2010 to 21 April 2021 (date of last search) that reported aetiology of neonatal sepsis (bacteraemia, sepsis, septicaemia, or meningitis) in LLMICs. Bibliographies of published systematic reviews were also assessed for eligibility (snowball method). Included studies needed to specifically address neonatal data or provide neonatal data that were clearly distinguishable from other age groups. To be inclusive, we defined neonates as infants up to 3 months of age, and neonatal sepsis was defined as neonates with clinical signs and/or laboratory evidence of sepsis. Countries were defined as low income and lower middle income according to the World Bank in June 2019 [14], and this categorisation was applied across the entire review period. Search terms were developed in accordance with PICOS (population, intervention, control, outcome, and setting) domains, and each database was searched using various combinations of the following terms: ‘infant, newborn’, ‘sepsis’, ‘meningitis’, ‘Gram-negative aerobic bacteria’, ‘Gram-negative bacteria’, ‘Gram-negative bacterial infections’, ‘microbial sensitivity tests’, ‘drug resistance’, and ‘developing countries’. These terms were applied to the title and abstract of publications. LLMICs per the World Bank 2019 list were also searched individually in Google with.org and.gov domains to include relevant materials from local shelves. The full search strategy and details of the quality assessment performed on each article can be found in S1 Text and S2 Table.

Studies were excluded if they presented aggregated data from which country-specific data could not be clearly identified, or from which neonates could not clearly be distinguished from older children or adults. Studies were also excluded if they presented insufficient information on the type of Gram-negative organisms or antimicrobial susceptibility, or if they reported infections of non-sterile sites. To ensure data were relevant to the current epidemiology of neonatal sepsis in LLMICs, a decision was made to exclude studies prior to 2010 and studies reporting a majority of isolates prior to 2010. Case reports or series (studies of fewer than 10 patients or with fewer than 10 Gram-negative isolates in total) were excluded due to study design, as they were likely to have significant biased selection of participants. Abstracts and titles were compiled in Endnote, and duplicates were removed. Two investigators (SCHW and YE) individually reviewed the identified articles to determine eligibility. All eligible articles were retrieved in full text. For references where we were unable to retrieve the full text and those with results that required clarification to assess eligibility, direct email contacts were sent to the corresponding authors. Non-English articles were included if data were able to be reliably extracted using Google translate. Disagreement over inclusion was resolved by consensus.

Data extraction

A data extraction checklist was developed based on the PICOS domains. Population variables included demographics of the neonatal population (sex, gestational age, and median age) and definition of neonatal sepsis (clinical or laboratory based). Outcome variables included timing of neonatal sepsis (early or late, and proportion with positive cultures for each category), total number of neonatal sepsis cases, mortality rate, number of blood or cerebrospinal fluid (CSF) cultures performed and number of positive cultures, microbiological methods, number of Gram-negative organisms identified (specifically, number of E. coli, Klebsiella species, Enterobacter species, Pseudomonas species, and Acinetobacter species), results of antimicrobial susceptibility testing (ampicillin, gentamicin, amikacin, third-generation cephalosporins [3GC], ciprofloxacin, and carbapenem), and burden of extended-spectrum beta-lactamases. Setting variables included country, city, setting (community, hospital-based [including neonatal intensive care], special care baby unit, or paediatric ward), study design, publication year, and study years. Two investigators (SCHW and LR) independently extracted the above variables into an Excel spreadsheet.

Quality assessment

The quality of each article was assessed independently by 2 investigators (SCHW and LR) using either the Newcastle–Ottawa quality assessment tool (case–control, cross-sectional, and cohort studies) or Cochrane risk of bias tool randomised controlled trials. Consensus was reached by panel discussion between 3 investigators (SCHW, LR, and YE). The results of quality assessment are summarised in S2 Table.

Statistical analysis

Meta-analysis was conducted to calculate pooled prevalence of positive blood or CSF culture, and of the 5 major Gram-negative bacterial species using the ‘metaprop’ command of the ‘metan’ package in Stata 16 [123]. Pooled prevalence of resistance against 6 key antimicrobials for each major Gram-negative bacterial species was also calculated with this method. Ampicillin resistance was not reported for Klebsiella, Enterobacter, Pseudomonas, and Acinetobacter spp. as these bacteria are intrinsically resistant. Stratification was done by continent of study (Asia versus Africa). Pooled prevalence was calculated as effect size with 95% confidence intervals (CIs) using logistic-normal random-effect models. Given the variability of the patient characteristics within the studies, the random-effect model was applied irrespective of the I2 statistics.

Publication bias and study heterogeneity

Sensitivity and subgroup analyses as well as meta-regression models were used to investigate sources of heterogeneity and the factors that affect the magnitudes of estimates, where data were available. The sensitivity analyses were conducted by excluding 1 study each time and recalculating the pooled prevalence. Funnel plots and Egger’s meta-regression test were used to assess small study effects. Study heterogeneity was reported using the I2 measure of inconsistency.

All statistical analyses were done using Stata 16 (StataCorp, 2015). The results are illustrated on a world map, using data from the public domain map dataset Natural Earth (https://www.naturalearthdata.com), through the ‘rnaturalearth’ package (version 0.1.0) [16] in R 3.6.0 [17].

Results

Our search yielded 766 results, of which 446 studies were eligible for full-text screening; 358 studies were excluded after full-text screening, and 88 studies were included in the full synthesis (Fig 1). Two studies were published in Bahasa Indonesian and 1 in French, with the remainder in English. There was no evidence of asymmetry by Egger’s meta-regression test (S1 Data). There were 34 studies from the Africa region—Angola (1) [18], Congo (1) [19], Egypt (5) [2024], Ethiopia (4) [2528], Ghana (1) [29], Guinea (1) [30], Madagascar (1) [31], Malawi (1) [32], Nigeria (12) [28,3343], Rwanda (1) [28], Sudan (1) [15], Tanzania (3) [4446], Uganda (1) [47], and Zambia (1) [48]—and 59 studies from the Asia region—Bangladesh (4) [28,4951], India (37) [5,28,5286], Indonesia (2) [87,88], Nepal (9) [8997], and Pakistan (7) [28,98103]—as depicted in Fig 2. One publication included data from 6 countries [28]. There were no studies identified from LLMICs outside the Africa or Asia regions. Most studies (n = 81) used a cross-sectional design. Study settings were reported in 84 studies, with a majority of the studies undertaken in a hospital setting. Forty-nine studies reported data from neonatal intensive care units (NICUs), 10 studies from special care baby units, 22 studies from a paediatric ward or unspecified hospital setting, and only 3 studies from the community. The definition of EOS was documented in 58 studies (66%): sepsis occurring within 48 hours of birth (1 study), within 72 hours of birth (43 studies), and within 1 week of birth (14 studies). There was also variation in the definition of the neonatal period (reported in 45 studies): 0–4 days (1 study), 0–28 days (38 studies), 0–30 days (2 study), and 0–60 days (4 studies). Bacteriological identification and antimicrobial susceptibility testing methods were reported in 70 studies (80%) and 76 studies (86%), respectively. Disk diffusion was the most commonly reported antimicrobial susceptibility testing method (70/72 studies). Most studies included data on all major Gram-negative species. Five studies reported on a subgroup of Gram-negative bacteria (3 studies on Acinetobacter spp. [n = 223] and 2 studies on Enterobacterales [n = 273]). Significant heterogeneity was observed across all meta-analyses and subgroup analyses. We could not identify explicit sources of heterogeneity due to the limitations of the available data. The characteristics of all included studies are summarised in S3 Table.

thumbnail
Fig 1. Study selection.

AJO, African Journals Online; IMEMR, Index Medicus for the Eastern Mediterranean Region; IMSEAR, Index Medicus for South-East Asia Region; LILACS, Latin American and Caribbean Health Sciences Literature; LMIC, low- or middle-income country; WPRIM, Western Pacific Region Index Medicus.

https://doi.org/10.1371/journal.pmed.1003787.g001

thumbnail
Fig 2. Distribution of included studies in low- and lower-middle-income countries.

The world map was created using data from the public domain map dataset Natural Earth (https://www.naturalearthdata.com).

https://doi.org/10.1371/journal.pmed.1003787.g002

The included studies of neonatal sepsis in LLMICs documented 20,828 positive blood/CSF cultures. The culture positivity prevalence ranged from 3% to 88% across 82 studies (denominator data missing for 6 studies). In 43 studies, a median of 60% (range 26% to 95%) of positive blood/CSF cultures were reported to have been taken in the study’s defined period of EOS. In 21 studies, a median of 62% (range 13% to 82%) of positive blood/CSF cultures were from premature neonates. The estimated pooled proportion of neonatal sepsis caused by Gram-negative organisms was 60% (95% CI 55% to 65%, I2 97%), and was 58% (95% CI 51% to 64%, I2 97%) and 61% (95% CI 53% to 66%, I2 98%) for Africa and Asia, respectively (Fig 3). Klebsiella spp. accounted for 38% (95% CI 33% to 43%, I2 96%) of Gram-negative neonatal sepsis, followed by 15% E. coli (95% CI 12% to 18%, I2 95%), 7% Pseudomonas spp. (95% CI 5% to 9%, I2 89%), 6% Acinetobacter (95% CI 4% to 10%, I2 96%), and 3% Enterobacter spp. (95% CI 2% to 5%, I2 86%). We observed a higher proportion of neonatal sepsis caused by Klebsiella spp. in Africa than Asia (44% versus 35%, with I2 90% and 96%, respectively), while Acinetobacter was more commonly reported in Asia than Africa (10% versus 3%, I2 98% and 79%, respectively). Significant heterogeneity was noted with these findings (S2 Data).

thumbnail
Fig 3. Pooled prevalence of Gram-negative neonatal sepsis.

ES, effect size; FE, fixed effect; LR, likelihood ratio; RE, random effect.

https://doi.org/10.1371/journal.pmed.1003787.g003

The pooled prevalence estimates of resistance to ampicillin, gentamicin, amikacin, 3GC, ciprofloxacin, and carbapenem are shown in Fig 4. Substantial resistance to gentamicin (from 42% to 70%) was observed in each of the specified Gram-negative species. Similarly, high levels of resistance to ceftriaxone were noted in E. coli, Klebsiella spp., Enterobacter spp., and Acinetobacter spp. (57% to 81%). We observed a higher prevalence of 3GC resistance in Africa compared to Asia, particularly with Klebsiella and Pseudomonas spp.; however, there was significant heterogeneity for these findings. Pooled prevalence of 3GC resistance in Africa for Klebsiella spp. was 88% (95% CI 72% to 96%, I2 92%) versus 77% in Asia (95% CI 65% to 87%, I2 90%). For Pseudomonas spp., 3GC resistance was 59% (95% CI 34% to 80%, I2 9%) in Africa versus 46% in Asia (95% CI 28% to 65%, I2 45%). The prevalence of ciprofloxacin resistance was higher in Asia across all 5 key groups of Gram-negative bacteria compared to Africa (37% to 76% versus 20% to 44%; S4 Table). The overall prevalence of extended-spectrum beta-lactamases was reported in 10 studies and ranged widely, from 14% to 95%. The overall pooled estimate of carbapenem resistance was 10% for E. coli (95% CI 4% to 21%, I2 75%), 10% for Klebsiella spp. (95% CI 2% to 36%, I2 88%), 15% for Pseudomonas spp. (95% CI 9% to 23%, I2 41%), and 42% for Acinetobacter spp. (95% CI 28% to 57%, I2 80%; S4 Table).

thumbnail
Fig 4. Pooled prevalence of antimicrobial resistance.

3GC, third-generation cephalosporin; ES, effect size; GN, Gram-negative.

https://doi.org/10.1371/journal.pmed.1003787.g004

Discussion

Our systematic review and meta-analysis identified that in LLMICs, approximately 60% of cases of neonatal sepsis were caused by Gram-negative bacteria. The prevalence of resistance to the WHO-recommended first-line antimicrobials ampicillin and gentamicin in 5 common groups of Gram-negative neonatal sepsis organisms is over 90% and 40%, respectively. Resistance to 3GC is also highly prevalent and of substantial concern.

This review has several limitations. To facilitate meaningful data interpretation about the current epidemiology of neonatal sepsis in LLMICs, a decision was made to exclude studies prior to 2010. There may be potential sampling bias as most of the studies in this review are from tertiary-level urban hospitals, which could lead to an overestimate of the burden caused by Gram-negative bacteria. Studies of AMR in children [12,104,105] have highlighted the lack of differentiation between community- and hospital-acquired infections as an issue, which we have similarly found. For most of the studies included, the sources of positive blood cultures were not reported, and therefore we were unable to determine the relative proportions of community- and hospital-acquired infections. This could be significant as there may be important epidemiological and AMR profile differences, depending on setting. Efforts to improve maternal and neonatal mortality in LMICs by increasing the numbers of infants delivered in healthcare facilities may increase exposure to antimicrobial-resistant organisms [106]. The focus of this study is to provide a snapshot on AMR in the least resourced countries. We have limited our search to LLMICs and acknowledge that this fails to address the issue of AMR in other middle-income countries. We applied the World Bank 2019 income categories across the entire review period and may have inadvertently missed studies reporting data from LLMICs that subsequently transitioned to a higher income category. Our findings may have less relevance for the empirical approach to ‘possible serious bacterial infection’ as defined by the WHO. This much broader group of acute infections, in which pathogens are rarely cultured even when referral to hospital is achieved, exhibit much lower mortality using narrow-spectrum penicillins and gentamicin [107,108]. Despite the focus on tertiary-level facilities, previous studies have noted a variable quality of microbiological data from studies in LMICs [109]. In this review, we report the microbiological methods used where these are documented, alongside the reported methods used to measure antimicrobial susceptibility, largely disc diffusion methods, as per the Clinical and Laboratory Standards Institute. The gold standard of laboratory accreditation by the International Organization for Standardisation may be challenging to achieve for laboratories in resource-limited facilities in LMICs [110]. Our findings further stress the importance of quality improvement initiatives such as Strengthening Laboratory Management Toward Accreditation and WHO’s Laboratory Quality Stepwise Implementation tool.

There was significant heterogeneity among the included studies, reflecting the differences in geography, case ascertainment, and microbiological and data collection methods. Some pooled estimates of AMR were derived from studies with small numbers of isolates and highly variable blood culture positivity rates, resulting in residual uncertainty about the precision of these estimates. These limitations are important when interpreting our findings. Whilst we did not detect any significant effect from small studies using a conventional funnel plot, there was clear evidence of a geographical publication bias. There are 78 countries categorised as LLMICs by the World Bank, and we identified data from only 19 countries (all within Asia and Africa). There was a predominance of studies from South Asia in the Asian continent, and care must be taken not to extrapolate these results to other countries in the region where there were no data available. Significant regional gaps were also noted from the Americas and the Western Pacific region, with no studies available for inclusion from the Pacific Island countries and territories. It has been estimated that a newborn infant dies every 2 minutes in the Western Pacific region, with infection being an important cause of death [111]. Yet there are no published data available from many countries in this region to help us understand the infectious aetiology of neonatal sepsis. Recent reviews from Pacific Island countries and territories confirm the presence of MDR organisms in the region [112], but the capacity for structured AMR surveillance and reporting is limited, which leads to challenges in interpretation of the findings [113]. It has been previously noted that systematic reviews aiming to include evidence from LMICs face challenges in accessing non-English literature and require searches of less well-known regional databases, particularly for grey literature [114]. We made extensive efforts to identify eligible studies in regional databases. We identified a single potentially eligible study from the Americas but were unable to obtain data on request from the corresponding author.

Our findings echo similar reports from systematic reviews focussing on sub-Saharan Africa and South Asia [11,12,115]. Klebsiella spp. was the most common causative Gram-negative bacteria, accounting for 38% of Gram-negative neonatal sepsis. This predominance appears more pronounced in studies from LMICs in Africa, which is also consistent with previous reports [11,12]. Our finding of a higher prevalence of Acinetobacter spp. in Asia is both interesting and of substantial concern. This observation may reflect the NICU setting of these studies (over 60% of studies in Asia were from NICUs), with increased likelihood of early invasive interventions, particularly in premature infants. Similarly, there is a high rate of cesarean delivery reported in South Asia [116], which may impact on maternal–neonatal acquisition of MDR Acinetobacter during hospitalisation. Of note, the Aetiology of Neonatal Infections in South Asia (ANISA) study [107], one of the very few community-based studies of neonatal sepsis, also reported Gram-negative organisms as the predominant cause of serious neonatal bacterial infections. Similar studies in other LMICs will help to assess whether there are any significant differences between hospital- and community-acquired neonatal sepsis.

Inappropriate empirical antibiotic therapy has recently been shown to be associated with increased mortality in young children and neonates, highlighting the importance of appropriate empirical antibiotic recommendations [117]. WHO-recommended first-line antibiotics for neonatal and paediatric sepsis are ampicillin and gentamicin, with ceftriaxone being second-line. Our finding of a high level of resistance against gentamicin across all key groups of Gram-negative bacteria raises the question of the appropriateness of its inclusion in empirical neonatal sepsis treatment regimens for LMICs, particularly in the hospital-based setting. Rates of resistance to ceftriaxone are similarly concerning. The most recent report from the BARNARDS observational cohort study of neonatal sepsis and AMR in 6 LMICs also found that only 28.5% of Gram-negative isolates were susceptible to at least 1 antibiotic in the combination of ampicillin and gentamicin, and declared that ampicillin is now redundant for treating neonatal sepsis in LMICs, with 97% of Gram-negative isolates resistant to ampicillin [118]. The prescribing practices of clinicians may reflect these findings of high levels of gentamicin and ceftriaxone resistance. A recent global point prevalence survey of antimicrobial prescribing in neonatal and paediatric sepsis identified that less than a quarter of neonates received WHO-recommended first- or second-line empirical antibiotics for sepsis [13]. In LMICs, meropenem was the most common empirical antibiotic prescribed for sepsis in hospitalised neonates and children (15.9% of antibiotic regimens prescribed) [13]. This may be appropriate given the local epidemiology, as suggested by the findings in this systematic review and meta-analysis. Region- or country-specific empirical antibiotic regimens for neonatal sepsis are indicated, which further highlights the need for structured AMR surveillance and reporting in LMICs, as these data are required to inform the most appropriate local recommendations. It is worthwhile noting that in some LMICs, AMR surveillance and reporting are impossible due to the lack of access to blood cultures [119]. Rapid diagnostics for infections are not novel to LMICs, with point-of-care testing readily available for conditions such as HIV and malaria. Rapid diagnostics including culture-independent methods for bloodstream infections and AMR may have an important role in LMICs. The development of low-cost tests that do not require significant laboratory infrastructure should be prioritised [120]. Tests that facilitate timely identification of causative pathogens and antibiotic resistance mechanisms may guide empirical antibiotic choice for neonatal sepsis, and improve antimicrobial stewardship by reducing empirical broad-spectrum antibiotic use.

Our findings provide important insight into the role of Gram-negative pathogens in neonatal sepsis in LLMICs, the burden of AMR in this context, and the appropriateness of existing recommendations for antimicrobial therapy. With limited access to third-line therapies (such as carbapenems) and the development of resistance even to these, robust surveillance of infection and AMR, along with infection prevention and antimicrobial stewardship strategies, is critical to address this global health threat [9]. An increase in the rate of facility births and neonatal interventions across many LMICs further highlights the importance of infection control and prevention. Dedicated cleaning interventions can improve cleanliness and reduce the burden of contaminated surfaces in low-resource NICUs [121]. The types of Gram-negative pathogens and associated AMR patterns are likely to evolve over time, and consideration should be given to the development of platforms that provide these data in a useful format and timely fashion. This will facilitate dynamic review of existing recommendations and their appropriateness. New antimicrobial strategies against MDR Gram-negative organisms appropriate for LLMICs need to be prioritised in parallel. The optimal dosing and duration of treatment with repurposed and new antimicrobials effective against MDR Gram-negative bacterial infections are frequently unknown for neonates [122]. There is a clear ethical mandate to prioritise trials of antimicrobials in neonates born in countries with the highest burden of AMR, and to ensure that antimicrobials are made accessible to clinicians and families in LLMICs.

Conclusion

Neonatal sepsis is increasingly caused by Gram-negative bacteria with alarming rates of multidrug resistance. Mortality is increased in neonatal sepsis caused by MDR organisms. The development of robust AMR surveillance and reporting in LLMICs should be prioritised to underpin region-specific empirical antimicrobial recommendations. There is an urgent need for high-quality antimicrobial trials in neonates and ensuring equitable access to new and effective antimicrobials.

Supporting information

S2 Data. Prevalence of key Gram-negative bacteria species.

https://doi.org/10.1371/journal.pmed.1003787.s002

(DOCX)

S2 Table. Risk of bias assessment (individual studies).

https://doi.org/10.1371/journal.pmed.1003787.s004

(DOCX)

S4 Table. Pooled antimicrobial resistance rates.

https://doi.org/10.1371/journal.pmed.1003787.s006

(DOCX)

Acknowledgments

The authors gratefully acknowledge Lars Eriksson, Mark Chatfield, and Caitlin Falconer for their input and expertise.

References

  1. 1. Lawn JE, Blencowe H, Oza S, You D, Lee AC, Waiswa P, et al. Every Newborn: progress, priorities, and potential beyond survival. Lancet. 2014;384(9938):189–205. pmid:24853593
  2. 2. World Health Organization. Reaching the Every Newborn national 2020 milestones: country progress, plans and moving forward. Geneva: World Health Organization; 2017.
  3. 3. Fleischmann-Struzek C, Goldfarb DM, Schlattmann P, Schlapbach LJ, Reinhart K, Kissoon N. The global burden of paediatric and neonatal sepsis: a systematic review. Lancet Respir Med. 2018;6(3):223–30. pmid:29508706
  4. 4. Tsai MH, Wu IH, Lee CW, Chu SM, Lien R, Huang HR, et al. Neonatal gram-negative bacillary late-onset sepsis: a case-control-control study on a prospectively collected database of 5,233 admissions. Am J Infect Control. 2016;44(2):146–53. pmid:26559734
  5. 5. Investigators of the Delhi Neonatal Infection Study (DeNIS) collaboration. Characterisation and antimicrobial resistance of sepsis pathogens in neonates born in tertiary care centres in Delhi, India: a cohort study. Lancet Glob Health. 2016;4(10):e752–e60. pmid:27633433
  6. 6. Dharmapalan D, Shet A, Yewale V, Sharland M. High reported rates of antimicrobial resistance in indian neonatal and pediatric blood stream infections. J Pediatric Infect Dis Soc. 2017;6(3):e62–8. pmid:28339675
  7. 7. Schlapbach LJ, Straney L, Alexander J, MacLaren G, Festa M, Schibler A, et al. Mortality related to invasive infections, sepsis, and septic shock in critically ill children in Australia and New Zealand, 2002–13: a multicentre retrospective cohort study. Lancet Infect Dis. 2015;15(1):46–54. pmid:25471555
  8. 8. Laxminarayan R, Matsoso P, Pant S, Brower C, Røttingen JA, Klugman K, et al. Access to effective antimicrobials: a worldwide challenge. Lancet. 2016;387(10014):168–75. pmid:26603918
  9. 9. Laxminarayan R, Bhutta ZA. Antimicrobial resistance—a threat to neonate survival. Lancet Glob Health. 2016;4(10):e676–7. pmid:27633421
  10. 10. World Health Organization. Pocket book of hospital care for children. 2nd edition. Geneva: World Health Organization; 2013.
  11. 11. Okomo U, Akpalu ENK, Le Doare K, Roca A, Cousens S, Jarde A, et al. Aetiology of invasive bacterial infection and antimicrobial resistance in neonates in sub-Saharan Africa: a systematic review and meta-analysis in line with the STROBE-NI reporting guidelines. Lancet Infect Dis. 2019;19(11):1219–34. pmid:31522858
  12. 12. Williams PCM, Isaacs D, Berkley JA. Antimicrobial resistance among children in sub-Saharan Africa. Lancet Infect Dis. 2018;18(2):e33–44. pmid:29033034
  13. 13. Jackson C, Hsia Y, Basmaci R, Bielicki J, Heath PT, Versporten A, et al. Global divergence from World Health Organization treatment guidelines for neonatal and pediatric sepsis. Pediatr Infect Dis J. 2019;38(11):1104–6. pmid:31425329
  14. 14. World Bank. World Bank list of economies. Washington (DC): World Bank; 2019.
  15. 15. Abdelaziz M, Hamadalnil Y, Hashim O, Bashir T, Mahjoub E. Microbiological profile of neonatal sepsis at a maternity hospital in Omdurman, Sudan. Sudan J Med Sci. 2019;14:45.
  16. 16. South A. rnaturalearth. Comprehensive R Archive Network; 2017.
  17. 17. R Core Team. R: a language and environment for statistical computing. Vienna: R Foundation for Statistical Computing; 2019.
  18. 18. Pelkonen T, Urtti S, dos Anjos E, Cardoso O, de Gouveia L, Roine I, et al. Aetiology of bacterial meningitis in infants aged < 90 days: prospective surveillance in Luanda, Angola. Int J Infect Dis. 2020;97:251–7. pmid:32534141
  19. 19. Ekouya Bowassa G, Ontsira-Ngoyi EN, Okoko AR, Kimpolo Tsiba HG, Oko AP, Moyen E, et al. Bacteriology of early neonatal infection in Brazzaville (Congo). Arch Pediatr. 2015;22(10):1099–101. pmid:26299910
  20. 20. Almohammady MN, Eltahlawy EM, Reda NM. Pattern of bacterial profile and antibiotic susceptibility among neonatal sepsis cases at Cairo University Children Hospital. J Taibah Univ Med Sci. 2020;15(1):39–47. pmid:32110181
  21. 21. El-Amir MI, El-Feky MA, Elwafa DAA, Abd-Elmawgood EA. Rapid diagnosis of neonatal sepsis by PCR for detection of 16S rRNA gene, while blood culture and PCR results were similar in E.coli-predominant EOS cases. Infect Drug Resist. 2019;12:2703–10. pmid:31564919
  22. 22. Shehab El-Din EM, El-Sokkary MM, Bassiouny MR, Hassan R. Epidemiology of neonatal sepsis and implicated pathogens: a study from Egypt. Biomed Res Int. 2015;2015:509484. pmid:26146621
  23. 23. EL-Morsi RM, El-Masry SM, Hamad EA. Neonatal bloodstream infections. Int J Curr Microbiol App Sci. 2020;9(2):1700–10.
  24. 24. Hassan DM, Madkour LA, Abuelhamd WA. Epidemiology of neonatal septicemia in the era of extended spectrum beta-lactamase producing bacteria: a prospective study in a tertiary referral hospital. J Pure Appl Microbiol. 2020;14(3):2189–202.
  25. 25. Gebrehiwot A, Belay WL, Moges F, Moges B, Anagaw B, Yismaw G, et al. Bacterial profile and drug susceptibility pattern of neonatal sepsis in Gondar University Hospital, Gondar northwest Ethiopia. Der Pharmacia Lettre. 2012;4:1811–6.
  26. 26. G/eyesus T, Moges F, Eshetie S, Yeshitela B, Abate E. Bacterial etiologic agents causing neonatal sepsis and associated risk factors in Gondar, Northwest Ethiopia. BMC Pediatr. 2017;17(1):137. pmid:28587631
  27. 27. Sorsa A, Früh J, Stötter L, Abdissa S. Blood culture result profile and antimicrobial resistance pattern: a report from neonatal intensive care unit (NICU), Asella teaching and referral hospital, Asella, south east Ethiopia. Antimicrob Resist Infect Control. 2019;8:42. pmid:30828446
  28. 28. Sands K, Carvalho MJ, Portal E, Thomson K, Dyer C, Akpulu C, et al. Characterization of antimicrobial-resistant Gram-negative bacteria that cause neonatal sepsis in seven low- and middle-income countries. Nat Microbiol. 2021;6(4):512–23. pmid:33782558
  29. 29. Labi AK, Obeng-Nkrumah N, Bjerrum S, Enweronu-Laryea C, Newman MJ. Neonatal bloodstream infections in a Ghanaian Tertiary Hospital: are the current antibiotic recommendations adequate? BMC Infect Dis. 2016;16(1):598. pmid:27776490
  30. 30. Shatalov A, Awwad F, Mangue P, Foqahaa R. Predominance of multi-drug resistant Klebsiella pneumonia and other Gram negative bacteria in neonatal sepsis in Equatorial Guinea. Open J Med Microbiol. 2015;05:254–8.
  31. 31. Huynh BT, Kermorvant-Duchemin E, Herindrainy P, Padget M, Rakotoarimanana FMJ, Feno H, et al. Bacterial infections in neonates, Madagascar, 2012–2014. Emerg Infect Dis. 2018;24(4):710–7. pmid:29553312
  32. 32. Iroh Tam PY, Musicha P, Kawaza K, Cornick J, Denis B, Freyne B, et al. Emerging resistance to empiric antimicrobial regimens for pediatric bloodstream infections in Malawi (1998–2017). Clin Infect Dis. 2019;69(1):61–8. pmid:30277505
  33. 33. Arowosegbe AO, Ojo DA, Dedeke IO, Shittu OB, Akingbade OA. Neonatal sepsis in a Nigerian tertiary hospital: clinical features, clinical outcome, aetiology and antibiotic susceptibility pattern. S Afr J Infect Dis. 2017;32(4):127–31.
  34. 34. Iregbu KC, Zubair K, Modibbo I, Aigbe A, Sonibare S, Ayoola O. Neonatal infections caused by Escherichia coli at the National Hospital, Abuja: a three-year retrospective study. Afr J Clin Exp Microbiol. 2013;14:95–100.
  35. 35. Medugu N. Trends in profiles of bacteria causing neonatal sepsis in Central Nigeria Hospital. Afr J Clin Exp Microbiol. 2017;18:49–52.
  36. 36. Ojide CK, Onwuezobe IA, Asuquo EE, Obiagwu CS. Bacteriologic profile and antibiotic susceptibility pattern of suspected septicaemic patients in Uyo, Nigeria. Res J Med Sci. 2013;7:35–9.
  37. 37. Omoregie R, Egbe CA, Dirisu J, Ogefere HO. Microbiology of neonatal septicemia in a tertiary hospital in Benin City, Nigeria. Biomarkers Genomic Med. 2013;5(4):142–6.
  38. 38. Onyedibe K, Okolo M, Toma B, Afolaranmi T. The necessity of full sepsis screen in neonatal sepsis: experience in a resource-limited setting. Sahel Med J. 2016;19(2):89–93.
  39. 39. Onyedibe KI, Bode-Thomas F, Afolaranmi TO, Okolo MO, Banwat EB, Egah DZ. Bacteriologic profile, antibiotic regimen and clinical outcome of neonatal sepsis in a university teaching hospital in north central Nigeria. J Adv Med Med Res. 2015;7(7):567–79.
  40. 40. Peterside O, Pondei K, Akinbami FO. Bacteriological profile and antibiotic susceptibility pattern of neonatal sepsis at a teaching hospital in Bayelsa State, Nigeria. Trop Med Health. 2015;43(3):183–90. pmid:26543394
  41. 41. Pius S, Bello M, Galadima GB, Ibrahim HA, Yerima ST, Ambe JP. Neonatal septicaemia, bacterial isolates and antibiogram sensitivity in Maiduguri North-Eastern Nigeria. Niger Postgrad Med J. 2016;23(3):146–51. pmid:27623727
  42. 42. Shittu M, Orisadare O, Jikeme O, Shittu B, Bello L, Oluremi A. antibiotic susceptibility pattern of bacteria isolates in neonates at a children hospital, Nigeria. J Med Sci Clin Res. 2014;2(10):2576–83.
  43. 43. Shobowale EO, Solarin AU, Elikwu CJ, Onyedibe KI, Akinola IJ, Faniran AA. Neonatal sepsis in a Nigerian private tertiary hospital: bacterial isolates, risk factors, and antibiotic susceptibility patterns. Ann Afr Med. 2017;16(2):52–8. pmid:28469117
  44. 44. Marando R, Seni J, Mirambo MM, Falgenhauer L, Moremi N, Mushi MF, et al. Predictors of the extended-spectrum-beta lactamases producing Enterobacteriaceae neonatal sepsis at a tertiary hospital, Tanzania. Int J Med Microbiol. 2018;308(7):803–11. pmid:29980372
  45. 45. Onken A, Said AK, Jørstad M, Jenum PA, Blomberg B. prevalence and antimicrobial resistance of microbes causing bloodstream infections in Unguja, Zanzibar. PLoS ONE. 2015;10(12):e0145632. pmid:26700032
  46. 46. Silago V, Kovacs D, Msanga DR, Seni J, Matthews L, Oravcova K, et al. Bacteremia in critical care units at Bugando Medical Centre, Mwanza, Tanzania: the role of colonization and contaminated cots and mothers’ hands in cross-transmission of multidrug resistant Gram-negative bacteria. Antimicrob Resist Infect Control. 2020;9:14. pmid:31956403
  47. 47. Mkony MF, Mizinduko MM, Massawe A, Matee M. Management of neonatal sepsis at Muhimbili National Hospital in Dar es Salaam: diagnostic accuracy of C-reactive protein and newborn scale of sepsis and antimicrobial resistance pattern of etiological bacteria. BMC Pediatr. 2014;14:293. pmid:25475836
  48. 48. Kabwe M, Tembo J, Chilukutu L, Chilufya M, Ngulube F, Lukwesa C, et al. Etiology, antibiotic resistance and risk factors for neonatal sepsis in a large referral center in Zambia. Pediatr Infect Dis J. 2016;35(7):e191–8. pmid:27031259
  49. 49. Afrin M, Siddique M, Ahmed A, Islam M, Sarker P, Showkath M, et al. Neonatal septicemia: isolation, identification and antibiotic sensitivity pattern of bacteria in a tertiary hospital in Bangladesh. Faridpur Med Coll J. 2016;11(2):58–61.
  50. 50. Ahmed F, Raj A, Nahar L, Hasan Z. Antimicrobial resistance of bacterial pathogens in a neonatal intensive care unit. Bangabandhu Sheikh Mujib Med Univ J. 2018;11:25.
  51. 51. Haque SM, Jahan N, Mannan MA, Hasan M, Begum M, Rob S, et al. Identification of bacterial isolates in neonatal sepsis and their antimicrobial susceptibility. Mymensingh Med J. 2014;23(4):709–14. pmid:25481589
  52. 52. Agarwal A, Bhat S. Clinico-microbiological study of neonatal sepsis. J Int Med Dent. 2015;2(1):22–19.
  53. 53. Annapurna D, Ramu P, Rakesh P. Profile of bacterial isolates in neonatal blood culture and their antibiotic susceptibility patterns (antibiogram) in neonatal sepsis at a tertiary care hospital NICU in South India. J Evol Med Dent Sci. 2018;7(2):249–55.
  54. 54. Bandyopadhyay T, Kumar A, Saili A, Randhawa VS. Distribution, antimicrobial resistance and predictors of mortality in neonatal sepsis. J Neonatal Perinatal Med. 2018;11(2):145–53. pmid:29991144
  55. 55. Bhat BV, Prasad P, Ravi Kumar VB, Harish BN, Krishnakumari K, Rekha A, et al. Syndrome evaluation system (SES) versus blood culture (BACTEC) in the diagnosis and management of neonatal sepsis—a randomized controlled trial. Indian J Pediatr. 2016;83(5):370–9. pmid:26732807
  56. 56. Chatterjee S, Datta S, Roy S, Ramanan L, Saha A, Viswanathan R, et al. Carbapenem resistance in Acinetobacter baumannii and other Acinetobacter spp. causing neonatal sepsis: focus on NDM-1 and its linkage to ISAba125. Front Microbiol. 2016;7:1126. pmid:27551277
  57. 57. Chelliah A, Ravinder T, Katragadda R, Leela KV, Babu RN. Isolation of MRSA, ESBL and AmpC-beta-lactamases from neonatal sepsis at a tertiary care hospital. J Clin Diagn Res. 2014;8(6):DC24–7. pmid:25120982
  58. 58. Dalal P, Gathwala G, Gupta M, Singh J. Bacteriological profile and antimicrobial sensitivity pattern in neonatal sepsis: a study from North India. Int J Res Med Sci. 2017;5:1541.
  59. 59. Dhanawade S, Tagare A, Gadre K, Shah S. Pattern and antimicrobial susceptibility of neonatal sepsis at a tertiary care center in western India. J Pediatr Infect Dis. 2015;10(3):76–81.
  60. 60. Dhaneria M, Jain S, Singh P, Mathur A, Lundborg CS, Pathak A. Incidence and determinants of health care-associated blood stream infection at a neonatal intensive care unit in Ujjain, India: a prospective cohort study. Diseases. 2018;6(1):14. pmid:29385762
  61. 61. Garg P, Usha MG. A study of neonatal septicaemia in a tertiary care hospital. J Pure Appl Microbiol. 2018;12(2):369–74.
  62. 62. Haider F, Dar ZP, Gupta A, Yaqoob S, Singh M. Multidrug resistance pattern in bacteriological isolates of neonatal septicemia in NICU of a tertiary care center. Indian J Microbiol Res. 2018;5(3):307–12.
  63. 63. Jajoo M, Manchanda V, Chaurasia S, Sankar MJ, Gautam H, Agarwal R, et al. Alarming rates of antimicrobial resistance and fungal sepsis in outborn neonates in North India. PLoS ONE. 2018;13(6):e0180705. pmid:29953451
  64. 64. Keerthi AM, Keerthi BJ, Mamatha PS, Anitha G. Spectrum of bacteria in neonatal sepsis at the district hospital NICU attached to Mandya Institute of Medical Sciences, Karnataka. J Evol Med Dent Sci. 2014;3(45):11059–63.
  65. 65. Kumar A, Randhawa V, Nirupam N, Rai Y, Saili A. Risk factors for carbapenem-resistant Acinetobacter baumanii blood stream infections in a neonatal intensive care unit, Delhi, India. J Infect Dev Ctries. 2014;8:1049–54. pmid:25116673
  66. 66. Kumar D, Mohan J, Rakesh PS, Prasad J, Joseph L. Bacteriological profile of neonatal sepsis in a secondary care hospital in rural Tamil Nadu, Southern India. J Family Med Prim Care. 2017;6:735. pmid:29564254
  67. 67. Lamba M, Sharma R, Sharma D, Choudhary M, Maheshwari RK. Bacteriological spectrum and antimicrobial susceptibility pattern of neonatal septicaemia in a tertiary care hospital of North India. J Matern Fetal Neonatal Med. 2016;29(24):3993–8. pmid:26858036
  68. 68. Mehar V, Yadav D, Somani P, Bhatambare G, Mulye S, Singh K. Neonatal sepsis in a tertiary care center in central India: microbiological profile, antimicrobial sensitivity pattern and outcome. J Neonatal Perinatal Med. 2013;6(2):165–72. pmid:24246519
  69. 69. Mishra PP, Bisht D, Prakash V, Kumar A, Goyal V. Extended spectrum β lactamase producing lactose fermenters causing neonatal septicaemia in a tertiary care center in Uttar Pradesh. Int J Contemp Med Res. 2018;5(4):27.
  70. 70. Mitra S, Mukherjee S, Naha S, Chattopadhyay P, Dutta S, Basu S. Evaluation of co-transfer of plasmid-mediated fluoroquinolone resistance genes and bla (NDM) gene in Enterobacteriaceae causing neonatal septicaemia. Antimicrob Resist Infect Control. 2019;8:46. pmid:30858970
  71. 71. Mittal S, Sharma M, Yadav A, Bala K, Chaudhary U. Acinetobacter lwoffii an emerging pathogen in neonatal ICU. Infect Disord Drug Targets. 2015;15(3):184–8. pmid:26307173
  72. 72. Mukherjee P, Biswas P, Satpathi S, Satpathi PS. Neonatal sepsis—trends in a peripheral tertiary health care facility of eastern India. J Evol Med Dent Sci. 2019;8(14):1089–93.
  73. 73. Mukherjee S, Bhattacharjee A, Naha S, Majumdar T, Debbarma SK, Kaur H, et al. Molecular characterization of NDM-1-producing Klebsiella pneumoniae ST29, ST347, ST1224, and ST2558 causing sepsis in neonates in a tertiary care hospital of north-east India. Infect Genet Evol. 2019;69:166–75. pmid:30677535
  74. 74. Pandita N, Wasim S, Bhat NK, Chandra V, Kakati B. Identification of the bacterial isolates in neonatal septicaemia and their antimicrobial susceptibility in a tertiary care hospital in Uttarakhand, India: a retrospective study. Int J Contemp Pediatr. 2016;3(1):200–5.
  75. 75. Patel D, Nimbalkar A, Sethi A, Kungwani A, Nimbalkar S. Blood culture isolates in neonatal sepsis and their sensitivity in Anand District of India. Indian J Pediatr. 2014;81(8):785–90. pmid:24408399
  76. 76. Rohatgi S, Dewan P, Faridi MMA, Kumar A, Malhotra RK, Batra P. Seven versus 10 days antibiotic therapy for culture-proven neonatal sepsis: a randomised controlled trial. J Paediatr Child Health. 2017;53(6):556–62. pmid:28398692
  77. 77. Roy MP, Gaind R, Aggarwal KC, Chellani HK, Biswal I. Pattern of pediatric bacterial infection and antibiotic resistance in New Delhi. Indian Pediatr. 2017;54(2):153–4. pmid:28285291
  78. 78. Shah A, Mulla S, Rajdev S. Neonatal sepsis: high antibiotic resistance of the bacterial pathogens in a neonatal intensive care unit of a tertiary care hospital. J Clin Neonatol. 2012;1:72–5. pmid:24027694
  79. 79. Sharma CM, Agrawal RP, Sharan H, Kumar B, Sharma D, Bhatia SS. “Neonatal sepsis”: bacteria & their susceptibility pattern towards antibiotics in neonatal intensive care unit. J Clin Diagn Res. 2013;7(11):2511–3. pmid:24392386
  80. 80. Srivastava R, Agarwal J, Srivastava S, Kumar M, Singh M. Multidrug resistant Gram-negative bacilli from neonatal septicaemia at a tertiary care centre in north India: a phenotypic and genotypic study. Indian J Med Microbiol. 2014;32(1):97–8. pmid:24399406
  81. 81. Thakur S, Thakur K, Sood A, Chaudhary S. Bacteriological profile and antibiotic sensitivity pattern of neonatal septicaemia in a rural tertiary care hospital in north India. Indian J Med Microbiol. 2016;34(1):67–71. pmid:26776121
  82. 82. Dhanalakshmi V, Suguna Sivakumar E. Comparative study in early neonates with septicemia by blood culture, staining techniques and C-reactive protein (CRP). J Clin Diagn Res. 2015;9:DC12–5. pmid:25954618
  83. 83. Ghosh T, Singh B, Bisai S, Satpathi PS, Ghosh A, Pal R. Neonatal sepsis and their microbial spectrum and susceptibility in a tertiary care hospital of eastern India. J Dent Med Sci. 2020;19(12):46–52.
  84. 84. Govindaraju G, Arumugam V, Mageswari TU, Rajaiah B, Ramakrishnan S. Sepsis in neonates: prevalence of micro-organisms and their susceptibility pattern in neonatal intensive care unit of a tertiary care hospital–a retrospective study. J Basic Clin Pharm. 2020;11(5):12.
  85. 85. Saha N, Sengupta M, Sarkar S, Sengupta M. Clinical and microbiological profile of neonatal septicemia in a tertiary care hospital in Kolkata. J Pure Appl Microbiol. 2020;14(2):1537–43.
  86. 86. Vaniya HV, Patel NM, Agrawal JM, Trivedi HR, Dhanani JV, Balat JD. Antimicrobial culture sensitivity pattern in neonatal sepsis in a tertiary-care hospital. Int J Med Sci Public Health. 2016;5:661–5.
  87. 87. Noor T, Nurhayana S, Rusli B. Antimicrobial sensitivity of blood culture in neonatal sepsis. Indones J Clin Pathol Med Laboratory. 2012;19(1):24–9.
  88. 88. Rukmono P, Zuraida R. Uji kepekaan antibiotik terhadap Pseudomonas aeroginosa penyebab sepsis neonatorum. Sari Pediatri. 2016;14(5):332–6.
  89. 89. Adhikari N, Shah PK, Acharya G, Vaidya KM. Bacteriological profile and associated risk factors of neonatal sepsis in Paropakar Maternity and Women’s Hospital Thapathali, Kathmandu. Nepal Med Coll J. 2014;16(2–4):161–4. pmid:26930737
  90. 90. Ansari S, Nepal HP, Gautam R, Shrestha S, Neopane P, Chapagain ML. Neonatal septicemia in Nepal: early-onset versus late-onset. Int J Pediatr. 2015;2015:379806. pmid:26649057
  91. 91. Gyawali N, Sanjana RK. Bacteriological profile and antibiogram of neonatal septicemia. Indian J Pediatr. 2013;80(5):371–4. pmid:23180407
  92. 92. Kanodia P, Yadav S, Singh R, Bhatta N. bacteriological profile of blood culture positive sepsis in newborn at BPKIHS, Dharan Nepal. J Coll Med Sci Nepal. 2017;13:193.
  93. 93. Pokhrel B, Koirala T, Shah G, Joshi S, Baral P. Bacteriological profile and antibiotic susceptibility of neonatal sepsis in neonatal intensive care unit of a tertiary hospital in Nepal. BMC Pediatr. 2018;18(1):208. pmid:29950162
  94. 94. Shrestha S, Shrestha NC, Dongol Singh S, Shrestha RPB, Kayestha S, Shrestha M, et al. Bacterial isolates and its antibiotic susceptibility pattern in NICU. Kathmandu Univ Med J. 2013;41(1):66–70. pmid:23774417
  95. 95. Thapa B, Thapa A, Aryal D, Thapa K, Pun A, Khanal S. Neonatal sepsis as a major cause of morbidity in a tertiary center in Kathmandu. J Nepal Med Assoc. 2013;52:549–56.
  96. 96. Yadav NS, Sharma S, Chaudhary DK, Panthi P, Pokhrel P, Shrestha A, et al. Bacteriological profile of neonatal sepsis and antibiotic susceptibility pattern of isolates admitted at Kanti Children’s Hospital, Kathmandu, Nepal. BMC Res Notes. 2018;11(1):301. pmid:29764503
  97. 97. Bhargava D, Kumari A, Paudyal R. Bacteriological profile of neonatal sepsis and antibiogram pattern of the isolates in national medical college and teaching hospital, Birgunj. Int J Health Med Curr Res. 2020;5(2):1694–702.
  98. 98. Ahmed M, Yasrab M, Khushdil A, Qamar K, Ahmed Z. Neonatal sepsis in a tertiary care hospital: bacteriological profile and its antimicrobial sensitivity. Pak Armed Forces Med J. 2018;68(6):1654–58.
  99. 99. Mir F, Nisar I, Tikmani SS, Baloch B, Shakoor S, Jehan F, et al. Simplified antibiotic regimens for treatment of clinical severe infection in the outpatient setting when referral is not possible for young infants in Pakistan (Simplified Antibiotic Therapy Trial): a randomised, open-label, equivalence trial. Lancet Glob Health. 2017;5(2):e177–85. pmid:27988146
  100. 100. Sana F, Satti L, Zaman G, Gardezi A, Imtiaz A, Ahmed A, et al. Pattern of Gram-negative bloodstream infections and their antibiotic susceptibility profiles in a neonatal intensive care unit. J Hosp Infect. 2018;98(3):243–4. pmid:29128348
  101. 101. Sheikh AN, Sajjad A, Hanif S. Neonatal sepsis: an evaluation of bacteriological spectrum, antibiotic susceptibilities and prognostic predictors at Civil Hospital, Karachi. Pak Paediatr J. 2014;38(3):143–55.
  102. 102. Ullah O, Khan A, Ambreen A, Ahmad I, Akhtar T, Gandapor AJ, et al. antibiotic sensitivity pattern of bacterial isolates of neonatal septicemia in Peshawar, Pakistan. Arch Iran Med. 2016;19(12):866–9. pmid:27998162
  103. 103. Shaikh M, Hanif M, Gul R, Hussain W, Hemandas H, Memon A. Spectrum and antimicrobial susceptibility pattern of micro-organisms associated with neonatal sepsis in a hospital in Karachi, Pakistan. Cureus. 2020;12(10):5. pmid:33194490
  104. 104. Waters D, Jawad I, Ahmad A, Lukšić I, Nair H, Zgaga L, et al. Aetiology of community-acquired neonatal sepsis in low and middle income countries. J Glob Health. 2011;1(2):154–70. pmid:23198116
  105. 105. Hamer DH, Darmstadt GL, Carlin JB, Zaidi AK, Yeboah-Antwi K, Saha SK, et al. Etiology of bacteremia in young infants in six countries. Pediatr Infect Dis J. 2015;34(1):e1–8. pmid:25389919
  106. 106. Bhutta ZA, Das JK, Bahl R, Lawn JE, Salam RA, Paul VK, et al. Can available interventions end preventable deaths in mothers, newborn babies, and stillbirths, and at what cost? Lancet. 2014;384(9940):347–70. pmid:24853604
  107. 107. Saha SK, Schrag SJ, El Arifeen S, Mullany LC, Shahidul Islam M, Shang N, et al. Causes and incidence of community-acquired serious infections among young children in south Asia (ANISA): an observational cohort study. Lancet. 2018;392(10142):145–59. pmid:30025808
  108. 108. African Neonatal Sepsis Trial (AFRINEST) group, Tshefu A, Lokangaka A, Ngaima S, Engmann C, Esamai F, et al. Simplified antibiotic regimens compared with injectable procaine benzylpenicillin plus gentamicin for treatment of neonates and young infants with clinical signs of possible serious bacterial infection when referral is not possible: a randomised, open-label, equivalence trial. Lancet. 2015;385(9979):1767–76. pmid:25842221
  109. 109. Ashley E, Dance D, Turner P. Grading antimicrobial susceptibility data quality: room for improvement. Lancet Infect Dis. 2018;18:603–4.
  110. 110. Zellweger RM, Carrique-Mas J, Limmathurotsakul D, Day NPJ, Thwaites GE, Baker S. A current perspective on antimicrobial resistance in Southeast Asia. J Antimicrob Chemother. 2017;72(11):2963–72. pmid:28961709
  111. 111. WHO Regional Office for the Western Pacific. Action plan for healthy newborn infants in the Western Pacific Region (2014–2020). Manila: WHO Regional Office for the Western Pacific; 2014.
  112. 112. Foxlee ND, Townell N, McIver L, Lau CL. Antibiotic resistance in Pacific Island countries and territories: a systematic scoping review. Antibiotics (Basel). 2019;8(1):29. pmid:30893880
  113. 113. Loftus M, Stewardson A, Naidu R, Coghlan B, Jenney A, Kepas J, et al. Antimicrobial resistance in the Pacific Island countries and territories. BMJ Glob Health. 2020;5(4):e002418. pmid:32349993
  114. 114. Shenderovich Y, Eisner M, Mikton C, Gardner F, Liu J, Murray J. Methods for conducting systematic reviews of risk factors in low- and middle-income countries. BMC Med Res Methodol. 2016;16(1):32. pmid:26979282
  115. 115. Chaurasia S, Sivanandan S, Agarwal R, Ellis S, Sharland M, Sankar MJ. Neonatal sepsis in South Asia: huge burden and spiralling antimicrobial resistance. BMJ. 2019;364:k5314. pmid:30670451
  116. 116. Verma V, Vishwakarma RK, Nath DC, Khan HTA, Prakash R, Abid O. Prevalence and determinants of caesarean section in South and South-East Asian women. PLoS ONE. 2020;15(3):e0229906. pmid:32163440
  117. 117. Cook A, Hsia Y, Russell N, Sharland M, Cheung K, Grimwood K, et al. Association of empiric antibiotic regimen discordance with 30-day mortality in neonatal and pediatric bloodstream infection—a global retrospective cohort study. Pediatr Infect Dis J. 2021;40(2):137–43. pmid:33395208
  118. 118. Thomson KM, Dyer C, Liu F, Sands K, Portal E, Carvalho MJ, et al. Effects of antibiotic resistance, drug target attainment, bacterial pathogenicity and virulence, and antibiotic access and affordability on outcomes in neonatal sepsis: an international microbiology and drug evaluation prospective substudy (BARNARDS). Lancet Infect Dis. 2021 Aug 9. (21)00050–5
  119. 119. Kawaza K, Kinshella MW, Hiwa T, Njirammadzi J, Banda M, Vidler M, et al. Assessing quality of newborn care at district facilities in Malawi. BMC Health Serv Res. 2020;20(1):227. pmid:32183795
  120. 120. Ayfan AKS, Macdonald J, Harris PNA, Heney C, Paterson DL, Trembizki E, et al. Rapid detection of NDM and VIM carbapenemase encoding genes by recombinase polymerase amplification and lateral flow-based detection. Eur J Clin Microbiol Infect Dis. 2021 May 11. pmid:33974185
  121. 121. Dramowski A, Aucamp M, Bekker A, Pillay S, Moloto K, Whitelaw AC, et al. NeoCLEAN: a multimodal strategy to enhance environmental cleaning in a resource-limited neonatal unit. Antimicrob Resist Infect Control. 2021;10(1):35. pmid:33579364
  122. 122. Folgori L, Bielicki J, Heath PT, Sharland M. Antimicrobial-resistant Gram-negative infections in neonates: burden of disease and challenges in treatment. Curr Opin Infect Dis. 2017;30(3):281–8. pmid:28306563
  123. 123. Nyaga VN, Arbyn M, Aerts M. Metaprop: a Stata command to perform meta-analysis of binomial data. Arch Public Health. 2014; 72(1):39–.