Pooled prevalence of third-generation cephalosporin resistance among Escherichia coli bloodstream isolates in Southeast Asia: protocol for systematic review and meta-analysis

Abstract

Introduction

Bloodstream infections remain a major contributor to global infectious disease mortality, and antimicrobial resistance increasingly compromises effective empirical and definitive treatment. Escherichia coli is among the leading bacterial causes of bloodstream infections and a major pathogen associated with antimicrobial resistance-related mortality. Resistance to third-generation cephalosporins, including ceftriaxone, cefotaxime and ceftazidime, is clinically important because it may limit standard treatment options and increase reliance on broader-spectrum agents. In Southeast Asia, available data on third-generation cephalosporin-resistant E. coli bloodstream isolates remain fragmented across countries, hospitals, populations and study designs. This systematic review and meta-analysis aims to estimate the pooled prevalence of third-generation cephalosporin resistance (3GCR) among E. coli bloodstream isolates in Southeast Asia.

Methods and analysis

This protocol will follow the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocols (PRISMA-P) guideline and has been registered in PROSPERO under registration number CRD420261381302. Searches will be conducted in PubMed, Scopus, ScienceDirect and Google Scholar from database inception to 1 October 2026, supplemented by citation searching and reference list screening. Eligible studies will include cross-sectional, surveillance, cohort, retrospective laboratory database and prospective observational studies conducted in Southeast Asian countries that report blood culture-specific data on third-generation cephalosporin resistance among E. coli isolates. Studies reporting ceftriaxone, cefotaxime or ceftazidime resistance, or ESBL phenotype clearly representing third-generation cephalosporin resistance, will be considered. Title and abstract screening will be performed independently by three reviewers. Full-text assessment, data extraction and quality assessment will be performed independently by two reviewers, with disagreements resolved through discussion with a third reviewer. Methodological quality will be assessed using a modified version of the Newcastle-Ottawa Scale. The primary outcome will be the pooled prevalence of third-generation cephalosporin resistance among E. coli bloodstream isolates. A random-effects meta-analysis will be used, with planned subgroup analyses by country, age group and infection-acquisition setting. Temporal trends will be explored using the year of data collection or publication.

Ethics and dissemination

Ethical approval is not required because this study will synthesise data from previously published studies and will not involve direct recruitment of human participants or collection of individual patient data. The findings will be disseminated through peer-reviewed publications and conference presentations.

PROSPERO registration number

CRD420261381302.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• This protocol follows Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P) guidance and has been prospectively registered in PROSPERO, supporting transparency and reproducibility of the planned review methods.

• The search strategy will use multiple databases, including PubMed, Scopus, ScienceDirect and Google Scholar, with additional citation and reference list screening to improve retrieval of relevant studies.

• The eligibility criteria are specifically restricted to Escherichia coli bloodstream isolates with extractable third-generation cephalosporin resistance data, reducing clinical and microbiological misclassification from mixed-organism or mixed-specimen studies.

• Study screening, data extraction and methodological quality assessment will be performed independently by multiple reviewers, with disagreements resolved through discussion or consultation with an additional reviewer.

• The review may be limited by heterogeneity in study design, laboratory methods, resistance definitions, surveillance systems and reporting quality across Southeast Asian countries; therefore, random-effects meta-analysis, subgroup analyses and narrative synthesis will be used where appropriate.

Introduction

Bloodstream infections are a major contributor to global infectious-disease mortality. In the Global Burden of Disease Study (2019), bloodstream infection was reported as one of the leading bacterial infectious syndromes worldwide.^{1 2} Antimicrobial resistance further increases the clinical burden of bacterial infections as it reduces the effectiveness of empirical antimicrobial therapy. In 2019, bacterial antimicrobial resistance was estimated to be associated with 4.95 million deaths and directly attributable to 1.27 million deaths worldwide.³ Among resistant pathogens, Escherichia coli was one of the leading contributors to antimicrobial resistance-related mortality, underscoring its clinical and public health importance in severe infections such as bacteremia.¹

Third-generation cephalosporins, including ceftriaxone, cefotaxime and ceftazidime, are widely used for empirical and definitive treatment of serious Gram-negative infections. Resistance to these agents in E. coli often reflects the production of extended-spectrum beta-lactamases (ESBLs) or related beta-lactam resistance mechanisms.⁴ The WHO uses the proportion of bloodstream infections caused by E. coli resistant to third-generation cephalosporins as an antimicrobial resistance indicator, defined by resistance to ceftriaxone, cefotaxime or ceftazidime using internationally recognised clinical breakpoints.^{5 6} Recent WHO-reported values for this indicator have been approximately 36–42%, indicating that third-generation cephalosporin resistance among E. coli bloodstream infections is already a substantial global problem.^{5 6}

In Southeast Asia, this issue is particularly important because antimicrobial use, healthcare infrastructure, microbiology laboratory capacity and surveillance systems vary considerably across countries. WHO GLASS emphasises that comparable AMR estimates require standardised surveillance systems, national coordination, reference laboratory capacity, and defined surveillance sites; however, surveillance capacity remains uneven, particularly in low- and middle-income settings, where weak laboratory infrastructure, limited resources, and variable health information systems can affect AMR data completeness and representativeness.^{7 8} These differences may contribute to substantial heterogeneity in resistance estimates across countries, hospitals, patient populations and study periods. Regional reports also highlight that bloodstream infection indicators, including third-generation cephalosporin-resistant E. coli, are increasingly monitored in South-East Asia, but data availability and reporting remain variable across countries.^{9 10}

Existing studies on antimicrobial resistance in bloodstream infections in Southeast Asia vary in country coverage, hospital type, patient background, study period, laboratory method, and reporting format. Some studies report AMR among all Gram-negative bacteria or Enterobacterales, while other studies combine E. coli with Klebsiella spp. or other organisms. Several studies also combine the specimens such as blood, urine, respiratory, or stool, making it difficult to isolate the specific burden among blood culture-confirmed E. coli isolates. These differences may obscure the bloodstream-specific burden of third-generation cephalosporin-resistant E. coli in the region. Previous reviews may have discussed antimicrobial resistance or bloodstream infections more broadly. However, a focused quantitative synthesis restricted to the blood culture-confirmed E. coli isolates and third-generation cephalosporin resistance in Southeast Asia remains limited.

A focused systematic review and meta-analysis are therefore needed to provide a clearer regional estimate. This review will identify, appraise and quantitatively synthesise available evidence on third-generation cephalosporin resistance among E. coli bloodstream isolates in Southeast Asia. The primary objective is to estimate the pooled prevalence of this resistance phenotype, with planned subgroup analyses by country, age group, infection-acquisition setting and study period where data are available. These findings may support empirical antibiotic policy, antimicrobial stewardship, microbiology surveillance and future research priorities in the region.

Methods and analysis

Protocol and registration

This systematic review is designed to evaluate the prevalence of third-generation cephalosporin resistance among E. coli bloodstream isolates in Southeast Asia. This protocol will be conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocols guideline (PRISMA-P).¹¹ This review has been registered in the International Prospective Register of Systematic Reviews (PROSPERO) under registration number CRD420261381302 on 4 June 2026. Any important protocol amendments will be documented with the date and reason for the change, updated in PROSPERO where applicable, and described in the final systematic review manuscript.

Patient and public involvement

Patients and the public were not involved in the design, conduct, reporting or dissemination planning of this protocol, as this systematic review will use data from previously published studies.

Data source and search strategy

A comprehensive search will be conducted to identify all relevant studies on the E. coli bloodstream isolates in Southeast Asia. PubMed, Scopus, ScienceDirect and Google Scholar will be searched from database inception to 1 October 2026, with no restriction on publication year. An updated search will be conducted before completion of the systematic review to identify any newly published eligible studies. Additional searches may be conducted through citation searching and reference list screening. The search strategy will be structured using the Condition, Context, Population (CoCoPop) framework to define research questions and search terms clearly.¹² The condition is third-generation cephalosporin resistance and/or ESBL production. The context is Southeast Asia, defined according to the United Nations M49 “South-eastern Asia” geoscheme as Brunei Darussalam, Cambodia, Indonesia, Lao People’s Democratic Republic, Malaysia, Myanmar, Philippines, Singapore, Thailand, Timor-Leste and Viet Nam. The population is patients with blood culture-confirmed E. coli isolates. Search terms will include combinations of terms related to (“Escherichia coli”), (“bloodstream infection”), (“bacteraemia”), (“blood culture”), (“third-generation cephalosporin resistance”), (“ceftriaxone”), (“cefotaxime”), (“ceftazidime”), (“ESBL”), (“extended-spectrum beta-lactamase”) and (“Southeast Asian”). The search strategy will be adapted for each database using appropriate Boolean operators, Medical Subject Headings and database-specific syntax. Search results will be exported to a systematic review platform for deduplication and screening. The details of the search strategy are presented in table 1, while the flowchart for this systematic review is illustrated in figure 1.

Table 1. Search strategy based on Condition, Context, Population (CoCoPop) framework.

Category	Description	Keywords
Condition	The specific antimicrobial resistance being studied: third-generation cephalosporin resistance (3GCR) and/or extended-spectrum beta-lactamase (ESBL) production.	(“third-generation cephalosporin resistance” OR “3GC resistance” OR “ceftriaxone-resistant” OR “cefotaxime-resistant” OR “ceftazidime-resistant” OR “ESBL” OR “extended-spectrum beta-lactamase”)
Context	The geographical setting, focusing on the eleven countries categorised within Southeast Asia.	(“Southeast Asia” OR “Brunei” OR “Cambodia” OR “Indonesia” OR “Laos” OR “Malaysia” OR “Myanmar” OR “Philippines” OR “Singapore” OR “Thailand” OR “Timor-Leste” OR “Vietnam”)
Population	The specific pathogen and source of infection: Escherichia coli bloodstream isolates (from bacteremia or bloodstream infections).	(“Escherichia coli” OR “E. coli”) AND (“bacteremia” OR “bacteraemia” OR “bloodstream infection” OR “blood culture” OR “septicemia”)

Figure 1. Planned Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow chart of the study.

Eligibility criteria

This review will include studies focusing on individuals of any age group (neonates, children and adults), any gender and any ethnicity with confirmed third-generation cephalosporin-resistant E. coli bloodstream isolates. Eligible study designs will include cross-sectional studies, surveillance studies, cohort studies, retrospective laboratory database studies and prospective observational studies. Studies must be conducted in one or more Southeast Asian countries. Studies must also report blood culture or bloodstream-specific data only, or have separate blood-specific data. Eligible studies must report third-generation cephalosporin resistance based on antimicrobial susceptibility testing, including ceftriaxone, cefotaxime or ceftazidime. Studies reporting ESBL-producing E. coli will be included only if the ESBL phenotype clearly represents third-generation cephalosporin resistance among E. coli bloodstream isolates. Studies must report prevalence, proportion or raw data that allow calculation of the number of resistant isolates and the total number of E. coli bloodstream isolates tested.

Studies will be excluded if they report only non-blood specimens, such as urine, sputum, stool, rectal swabs, wound swabs or environmental samples. Mixed-specimen studies will be excluded if bloodstream-specific data cannot be separated. Studies pooling Enterobacterales or Gram-negative bacteria without separate E. coli data will also be excluded. Studies that do not report third-generation cephalosporin resistance, ceftriaxone resistance, cefotaxime resistance, ceftazidime resistance or clearly related ESBL data will be excluded. Studies conducted outside Southeast Asia or multicountry studies without extractable Southeast Asia-specific data will be excluded. Reviews, systematic reviews, meta-analyses, editorials, commentaries, case reports, animal studies, in vitro-only studies and conference abstracts without usable data will not be included.

Data management and study selection

The management of articles will be facilitated using Rayyan AI to streamline the process of organising search results, removing duplicates, screening titles and abstracts, reviewing full texts, extracting data and performing quality assessments. A centralised master database will also be maintained to document study identification, screening decisions, full-text eligibility assessment, reasons for exclusion and final inclusion status.

The study selection process will be conducted in several stages. First, titles and abstracts of all remaining records after deduplication will be independently screened by three reviewers against predefined inclusion and exclusion criteria. Studies considered potentially eligible by at least two reviewers will proceed to full-text assessment. Full-text articles will then be retrieved and reviewed independently by two reviewers to determine final eligibility for inclusion in the systematic review and meta-analysis. Any disagreements between reviewers during title and abstract screening or full-text assessment will be resolved through discussion. If consensus cannot be reached, a third reviewer will be consulted to ensure an unbiased selection process. Reasons for exclusion at the full-text stage will be recorded. Final inclusion decisions will be made following a thorough review of full texts and quality assessment. The entire selection process will be visualised using a PRISMA flow diagram in the final review. A structured and transparent methodology will be maintained to ensure that the conclusions are based on high-quality, relevant and valid evidence.

Data extraction

Data extraction will be conducted using a standardised data extraction form. Two reviewers will independently extract relevant data from the included studies. Extracted data will include first author, year of publication, article title, journal, DOI, database source, country, study location, hospital or surveillance site, study design, study period, year of data collection, type of healthcare facility, study population, age group, patient setting, infection acquisition setting, source of bacteraemia, blood culture confirmation and antimicrobial susceptibility testing method. Microbiological data will include the total number of E. coli bloodstream isolates tested, the number of third-generation cephalosporin-resistant E. coli bloodstream isolates, the antimicrobial agent used to define resistance and the resistance definition used in each study. Resistance will be defined by resistance to ceftriaxone, cefotaxime, ceftazidime or another clearly stated third-generation cephalosporin. ESBL phenotype will be extracted when used as a proxy for third-generation cephalosporin resistance. Reported resistance mechanisms, including ESBL production, AmpC beta-lactamase, CTX-M, TEM, SHV, porin alteration, efflux pump activity or other mechanisms, will be extracted where reported.

Quality assessment

The methodological quality of the included studies will be assessed independently by two reviewers using a modified version of the Newcastle-Ottawa Scale for observational studies reporting prevalence data.¹³ This tool evaluates methodological quality across domains related to the selection of the study population, the representativeness of the sample, the ascertainment of the outcome, the adequacy of reporting of numerator and denominator data, and comparability, where applicable. Studies that receive between zero and three stars will be classified as low quality, those with four to five stars as moderate quality and those with six to nine stars as high quality. Quality assessment findings will be used to support interpretation of results and sensitivity analyses.

Outcome variables

The primary outcome is the pooled prevalence of third-generation cephalosporin resistance among E. coli bloodstream isolates in Southeast Asia. The prevalence will be calculated by dividing the number of third-generation cephalosporin-resistant E. coli bloodstream isolates by the total number of E. coli bloodstream isolates tested for third-generation cephalosporin resistance.

Additional outcomes will include country-specific pooled prevalence, pooled prevalence by infection-acquisition setting, pooled prevalence by age group, and temporal trends in prevalence based on year of data collection or publication. Infection acquisition settings will include hospital-acquired, healthcare-associated and community-acquired bloodstream infection where these categories are reported by included studies. Age groups will include adult and paediatric populations where extractable data are available.

Data synthesis and statistical analysis

As heterogeneity is expected across countries, hospitals, study populations, laboratory methods and study periods, a random-effects meta-analysis model will be used. The DerSimonian-Laird method will be applied to estimate pooled prevalence. Pooled estimates will be reported with 95% CIs. Statistical heterogeneity will be assessed through visual inspection of forest plots, Cochran’s Q test, I² statistic and τ² where appropriate. Heterogeneity will be considered substantial when I² is greater than 50% or when the Cochran’s Q test indicates statistically significant heterogeneity. The Meta-Essentials software, which integrates with Microsoft Excel, will be used for the statistical analysis.¹⁴ The results will be presented in tables and visually summarised with forest plots. Publication bias or small-study effects will be assessed using funnel plot symmetry, where sufficient studies are available. If asymmetry is detected, the trim-and-fill method may be used to estimate the potential bias introduced by missing studies in the pooled prevalence estimate.

Planned subgroup analyses include country, infection-acquisition setting and age group. Infection-acquisition settings will include hospital-acquired, healthcare-associated and community-acquired bloodstream infections, where data are available. Temporal trends will be assessed using year of data collection or year of publication when year of data collection is unavailable. Studies that cannot be included in a quantitative synthesis due to incomplete or non-extractable data will be summarised narratively.

In condition where sufficient studies are available, sensitivity analysis will be performed by repeating the meta-analysis after excluding studies classified as low quality based on the modified Newcastle-Ottawa Scale. This analysis will assess whether the pooled prevalence estimate is robust to the methodological quality of the included studies.

The certainty of evidence for the primary outcome will be assessed narratively as this review focuses on prevalence estimates from observational and surveillance studies rather than intervention effects. The interpretation of confidence in the cumulative evidence will consider methodological quality, consistency of prevalence estimates, precision of pooled estimates, directness of study population and outcome definition, and potential publication bias or small-study effects. A formal GRADE rating will not be applied as standard GRADE domains are primarily designed for intervention effect estimates.

Ethics and dissemination

As this systematic review and meta-analysis will use data from previously published studies and will not involve direct recruitment of human participants or collection of individual patient data, ethical approval is not required.

The findings will be disseminated through publication in a peer-reviewed journal and presentation at academic or scientific meetings.