Improving Treatment and Outcomes for Melioidosis in Children, Northern Cambodia, 2009-2018.

We report trends in manifestations, treatment, and outcomes of 355 children with culture-confirmed melioidosis over 10 years at a pediatric hospital in northern Cambodia. Bacteremia and presentation with pneumonia were risk factors for death. A total of 39 children recovered after being given only oral antimicrobial drug treatment.

Impact of 13-Valent Pneumococcal Conjugate Vaccine on Colonization and Invasive Disease in Cambodian Children.

BACKGROUND: Cambodia introduced the 13-valent pneumococcal conjugate vaccine (PCV13) in January 2015 using a 3 + 0 dosing schedule and no catch-up campaign. We investigated the effects of this introduction on pneumococcal colonization and invasive disease in children aged <5 years. METHODS: There were 6 colonization surveys done between January 2014 and January 2018 in children attending the outpatient department of a nongovernmental pediatric hospital in Siem Reap. Nasopharyngeal swabs were analyzed by phenotypic and genotypic methods to detect pneumococcal serotypes and antimicrobial resistance. Invasive pneumococcal disease (IPD) data for January 2012-December 2018 were retrieved from hospital databases. Pre-PCV IPD data and pre-/post-PCV colonization data were modelled to estimate vaccine effectiveness (VE). RESULTS: Comparing 2014 with 2016-2018, and using adjusted prevalence ratios, VE estimates for colonization were 16.6% (95% confidence interval [CI] 10.6-21.8) for all pneumococci and 39.2% (95% CI 26.7-46.1) for vaccine serotype (VT) pneumococci. There was a 26.0% (95% CI 17.7-33.0) decrease in multidrug-resistant pneumococcal colonization. The IPD incidence was estimated to have declined by 26.4% (95% CI 14.4-35.8) by 2018, with a decrease of 36.3% (95% CI 23.8-46.9) for VT IPD and an increase of 101.4% (95% CI 62.0-145.4) for non-VT IPD. CONCLUSIONS: Following PCV13 introduction into the Cambodian immunization schedule, there have been declines in VT pneumococcal colonization and disease in children aged <5 years. Modelling of dominant serotype colonization data produced plausible VE estimates.

Elizabethkingia anophelis Infection in Infants, Cambodia, 2012-2018.

We describe 6 clinical isolates of Elizabethkingia anophelis from a pediatric referral hospital in Cambodia, along with 1 isolate reported from Thailand. Improving diagnostic microbiological methods in resource-limited settings will increase the frequency of reporting for this pathogen. Consensus on therapeutic options is needed, especially for resource-limited settings.

Automating the Generation of Antimicrobial Resistance Surveillance Reports: Proof-of-Concept Study Involving Seven Hospitals in Seven Countries.

BACKGROUND: Reporting cumulative antimicrobial susceptibility testing data on a regular basis is crucial to inform antimicrobial resistance (AMR) action plans at local, national, and global levels. However, analyzing data and generating a report are time consuming and often require trained personnel. OBJECTIVE: This study aimed to develop and test an application that can support a local hospital to analyze routinely collected electronic data independently and generate AMR surveillance reports rapidly. METHODS: An offline application to generate standardized AMR surveillance reports from routinely available microbiology and hospital data files was written in the R programming language (R Project for Statistical Computing). The application can be run by double clicking on the application file without any further user input. The data analysis procedure and report content were developed based on the recommendations of the World Health Organization Global Antimicrobial Resistance Surveillance System (WHO GLASS). The application was tested on Microsoft Windows 10 and 7 using open access example data sets. We then independently tested the application in seven hospitals in Cambodia, Lao People's Democratic Republic, Myanmar, Nepal, Thailand, the United Kingdom, and Vietnam. RESULTS: We developed the AutoMated tool for Antimicrobial resistance Surveillance System (AMASS), which can support clinical microbiology laboratories to analyze their microbiology and hospital data files (in CSV or Excel format) onsite and promptly generate AMR surveillance reports (in PDF and CSV formats). The data files could be those exported from WHONET or other laboratory information systems. The automatically generated reports contain only summary data without patient identifiers. The AMASS application is downloadable from https://www.amass.website/. The participating hospitals tested the application and deposited their AMR surveillance reports in an open access data repository. CONCLUSIONS: The AMASS is a useful tool to support the generation and sharing of AMR surveillance reports.

Antimicrobial Resistance in Invasive Bacterial Infections in Hospitalized Children, Cambodia, 2007-2016.

To determine trends, mortality rates, and costs of antimicrobial resistance in invasive bacterial infections in hospitalized children, we analyzed data from Angkor Hospital for Children, Siem Reap, Cambodia, for 2007-2016. A total of 39,050 cultures yielded 1,341 target pathogens. Resistance rates were high; 82% each of Escherichia coli and Klebsiella pneumoniae isolates were multidrug resistant. Hospital-acquired isolates were more often resistant than community-acquired isolates; resistance trends over time were heterogeneous. K. pneumoniae isolates from neonates were more likely than those from nonneonates to be resistant to ampicillin-gentamicin and third-generation cephalosporins. In patients with community-acquired gram-negative bacteremia, third-generation cephalosporin resistance was associated with increased mortality rates, increased intensive care unit admissions, and 2.26-fold increased healthcare costs among survivors. High antimicrobial resistance in this setting is a threat to human life and the economy. In similar low-resource settings, our methods could be reproduced as a robust surveillance model for antimicrobial resistance.

Utilization of a clinical microbiology service at a Cambodian paediatric hospital and its impact on appropriate antimicrobial prescribing.

Background: Antimicrobial resistance threatens human health worldwide. Antimicrobial misuse is a major driver of resistance. Promoting appropriate antimicrobial use requires an understanding of how clinical microbiology services are utilized, particularly in resource-limited settings. Objectives: To assess the appropriateness of antimicrobial prescribing and the factors affecting utilization of the established clinical microbiology service (CMS). The CMS comprises the microbiology laboratory, clinical microbiologists (infection doctors) and antimicrobial treatment guidelines. Methods: This mixed-methods study was conducted at a non-governmental Cambodian paediatric hospital. Empirical and post-culture antimicrobial prescriptions were reviewed from medical records. The random sample included 10 outpatients per week in 2016 (retrospective) and 20 inpatients per week for 4 weeks in the medical, neonatal and intensive care wards (prospective). Post-culture prescriptions were assessed in patients with positive blood and cerebrospinal fluid cultures from 1 January 2014 to 31 December 2016. Focus group discussions and semi-structured interviews with clinicians explored barriers and facilitators to use of the CMS. Results: Only 31% of outpatients were prescribed empirical antimicrobials. Post-culture prescriptions (394/443, 89%) were more likely to be appropriate than empirical prescriptions (447/535, 84%), based on treatment guidelines, microbiology advice and antimicrobial susceptibility test results (P = 0.015). Being comprehensive, accessible and trusted enabled CMS utilization. Clinical microbiologists provided a crucial human interface between the CMS and physicians. The main barriers were a strong clinical hierarchy and occasional communication difficulties. Conclusions: Antimicrobial prescribing in this hospital was largely appropriate. A culturally appropriate human interface linking the laboratory and physicians is essential in providing effective microbiology services and ensuring appropriate antimicrobial prescribing in resource-limited settings.

The aetiologies of central nervous system infections in hospitalised Cambodian children.

BACKGROUND: Central nervous system (CNS) infections are an important cause of childhood morbidity and mortality. The aetiologies of these potentially vaccine-preventable infections have not been well established in Cambodia. METHODS: We did a one year prospective study of children hospitalised with suspected CNS infection at Angkor Hospital for Children, Siem Reap. Cerebrospinal fluid specimens (CSF) samples underwent culture, multiplex PCR and serological analysis to identify a range of bacterial and viral pathogens. Viral metagenomics was performed on a subset of pathogen negative specimens. RESULTS: Between 1st October 2014 and 30th September 2015, 284 analysable patients were enrolled. The median patient age was 2.6 years; 62.0% were aged <5 years. CSF white blood cell count was ≥10 cells/µL in 116/272 (42.6%) cases. CNS infection was microbiologically confirmed in 55 children (19.3%). Enteroviruses (21/55), Japanese encephalitis virus (17/55), and Streptococcus pneumoniae (7/55) accounted for 45 (81.8%) of all pathogens identified. Of the pathogens detected, 74.5% (41/55) were viruses and 23.6% (13/55) were bacteria. The majority of patients were treated with ceftriaxone empirically. The case fatality rate was 2.5%. CONCLUSIONS: Enteroviruses, JEV and S. pneumoniae are the most frequently detected causes of CNS infection in hospitalised Cambodian children.

A retrospective analysis of melioidosis in Cambodian children, 2009-2013.

BACKGROUND: Melioidiosis, infection by Burkholderia pseudomallei, is an important but frequently under-recognised cause of morbidity and mortality in Southeast Asia and elsewhere in the tropics. Data on the epidemiology of paediatric melioidosis in Cambodia are extremely limited. METHODS: Culture-positive melioidosis cases presenting to Angkor Hospital for Children, a non-governmental paediatric hospital located in Siem Reap, Northern Cambodia, between 1st January 2009 and 31st December 2013 were identified by searches of hospital and laboratory databases and logbooks. RESULTS: One hundred seventy-three evaluable cases were identified, presenting from eight provinces. For Siem Reap province, the median commune level incidence was estimated to be 28-35 cases per 100,000 children <15 years per year. Most cases presented during the wet season, May to October. The median age at presentation was 5.7 years (range 8 days-15.9 years). Apart from undernutrition, co-morbidities were rare. Three quarters (131/173) of the children had localised infection, most commonly skin/soft tissue infection (60 cases) or suppurative parotitis (51 cases). There were 39 children with B. pseudomallei bacteraemia: 29 (74.4%) of these had clinical and/or radiological evidence of pneumonia. Overall mortality was 16.8% (29/173) with mortality in bacteraemic cases of 71.8% (28/39). At least seven children did not receive an antimicrobial with activity against B. pseudomallei prior to death. CONCLUSIONS: This retrospective study demonstrated a considerable burden of melioidosis in Cambodian children. Given the high mortality associated with bacteraemic infection, there is an urgent need for greater awareness amongst healthcare professionals in Cambodia and other countries where melioidosis is known or suspected to be endemic. Empiric treatment guidelines should ensure suspected cases are treated early with appropriate antimicrobials.

Pneumococcal Infection among Children before Introduction of 13-Valent Pneumococcal Conjugate Vaccine, Cambodia.

Vaccination of children with pneumococcal conjugate vaccine (PCV13) was initiated in Cambodia in 2015. To determine baseline data, we collected samples from children in 2013 and 2014. PCV13 serotypes accounted for 62.7% of colonizing organisms in outpatients and 88.4% of invasive pneumococci overall; multidrug resistance was common. Thus, effectiveness of vaccination should be high.

Exploring the pediatric nasopharyngeal bacterial microbiota with culture-based MALDI-TOF mass spectrometry and targeted metagenomic sequencing.

The nasopharynx is an important reservoir of disease-associated and antimicrobial-resistant bacterial species. This proof-of-concept study assessed the utility of a combined culture, matrix-assisted laser desorption/ionization-time of flight mass spectrometry (MALDI-TOF MS), and targeted metagenomic sequencing workflow for the study of the pediatric nasopharyngeal bacterial microbiota. Nasopharyngeal swabs and clinical metadata were collected from Cambodian children during a hospital outpatient visit and then biweekly for 12 weeks. Swabs were cultured on chocolate and blood-gentamicin agar, and all colony morphotypes were identified by MALDI-TOF MS. Metagenomic sequencing was done on a scrape of all colonies from a chocolate agar culture and processed using the mSWEEP pipeline. One hundred one children were enrolled, yielding 620 swabs. MALDI-TOF MS identified 106 bacterial species/40 genera: 20 species accounted for 88.5% (2,190/2,474) of isolates. Colonization by Moraxella catarrhalis (92.1% of children on ≥1 swab), Haemophilus influenzae (87.1%), and Streptococcus pneumoniae (83.2%) was particularly common. In S. pneumoniae-colonized children, a median of two serotypes [inter-quartile range (IQR) 1-2, range 1-4] was detected. For the 21 bacterial species included in the mSWEEP database and identifiable by MALDI-TOF, detection by culture + MALDI-TOF MS and culture + mSWEEP was highly concordant with a median species-level agreement of 96.9% (IQR 86.8%-98.8%). mSWEEP revealed highly dynamic lineage-level colonization patterns for S. pneumoniae which were quite different to those for S. aureus. A combined culture, MALDI-TOF MS, targeted metagenomic sequencing approach for the exploration of the young child nasopharyngeal microbiome was technically feasible, and each component yielded complementary data. IMPORTANCE: The human upper respiratory tract is an important source of disease-causing and antibiotic-resistant bacteria. However, understanding the interactions and stability of these bacterial populations is technically challenging. We used a combination of approaches to determine colonization patterns over a 3-month period in 101 Cambodian children. The combined approach was feasible to implement, and each component gave complementary data to enable a better understanding of the complex patterns of bacterial colonization.

High-risk Escherichia coli clones that cause neonatal meningitis and association with recrudescent infection.

Neonatal meningitis is a devastating disease associated with high mortality and neurological sequelae. Escherichia coli is the second most common cause of neonatal meningitis in full-term infants (herein NMEC) and the most common cause of meningitis in preterm neonates. Here, we investigated the genomic relatedness of a collection of 58 NMEC isolates spanning 1974-2020 and isolated from seven different geographic regions. We show NMEC are comprised of diverse sequence types (STs), with ST95 (34.5%) and ST1193 (15.5%) the most common. No single virulence gene profile was conserved in all isolates; however, genes encoding fimbrial adhesins, iron acquisition systems, the K1 capsule, and O antigen types O18, O75, and O2 were most prevalent. Antibiotic resistance genes occurred infrequently in our collection. We also monitored the infection dynamics in three patients that suffered recrudescent invasive infection caused by the original infecting isolate despite appropriate antibiotic treatment based on antibiogram profile and resistance genotype. These patients exhibited severe gut dysbiosis. In one patient, the causative NMEC isolate was also detected in the fecal flora at the time of the second infection episode and after treatment. Thus, although antibiotics are the standard of care for NMEC treatment, our data suggest that failure to eliminate the causative NMEC that resides intestinally can lead to the existence of a refractory reservoir that may seed recrudescent infection.

Pneumococcal colonization and severity of pneumonia in hospitalized Cambodian children following introduction of the 13-valent pneumococcal conjugate vaccine.

Objectives: This study sought to characterize pneumococcal colonization and clinical/radiological features in Cambodian children admitted to hospital with an illness compatible with pneumonia following national introduction of the 13-valent pneumococcal conjugate vaccine (PCV13). Methods: Children aged 0-59 months admitted to Angkor Hospital for Children who met the World Health Organization (WHO) case definition for clinical pneumonia were enrolled over a 3-year period. Clinical, radiological and vaccination data were collected at enrolment. A nasopharyngeal swab was collected for detection of pneumococcal colonization using the WHO standard culture method. Results: Between 1 September 2015 and 31August 2018, 2209 analysable illness episodes were enrolled. Pneumococci were detected in 943/2209 (42.7%) children. PCV13 serotypes were detected less frequently in children who had been vaccinated appropriately for their age compared with undervaccinated children: 309/567 (53.6%) vs 216/342 (63.2%) (P=0.006). Age-appropriate PCV13 vaccination was negatively associated with hypoxic presentation [adjusted odds ratio (aOR) 0.72, 95% confidence interval (CI) 0.60-0.87; P=0.0006] and primary endpoint pneumonia on chest x ray (aOR 0.69, 95% CI 0.54-0.90; P=0.006). Conclusions: The introduction of PCV13 in Cambodia was associated with a decline in vaccine serotype nasopharyngeal colonization, and clinical and radiological severity in children hospitalized with clinical pneumonia.

Antimicrobial resistance patterns in bacteria causing febrile illness in Africa, South Asia, and Southeast Asia: a systematic review of published etiological studies from 1980-2015.

OBJECTIVE: In this study, we aimed to conduct a systematic review to characterize antimicrobial resistance (AMR) patterns for bacterial causes of febrile illness in Africa and Asia. METHODS: We included published literature from 1980-2015 based on data extracted from two recent systematic reviews of nonmalarial febrile illness from Africa, South Asia, and Southeast Asia. Selection criteria included articles with full bacterial identification and antimicrobial susceptibility testing (AST) results for key normally sterile site pathogen-drug combinations. Pooled proportions of resistant isolates were combined using random effects meta-analysis. Study data quality was graded using the Microbiology Investigation Criteria for Reporting Objectively (MICRO) framework. RESULTS: Of 3475 unique articles included in the previous reviews, 371 included the target pathogen-drug combinations. Salmonella enterica tested against ceftriaxone and ciprofloxacin were the two highest reported combinations (30,509 and 22,056 isolates, respectively). Pooled proportions of resistant isolates were high for third-generation cephalosporins for Klebsiella pneumoniae and Escherichia coli in all regions. The MICRO grading showed an overall lack of standardization. CONCLUSION: This review highlights a general increase in AMR reporting and in resistance over time. However, there were substantial problems with diagnostic microbiological data quality. Urgent strengthening of laboratory capacity, standardized testing, and reporting of AST results is required to improve AMR surveillance.

A Clinically Oriented antimicrobial Resistance surveillance Network (ACORN): pilot implementation in three countries in Southeast Asia, 2019-2020.

Background: Case-based surveillance of antimicrobial resistance (AMR) provides more actionable data than isolate- or sample-based surveillance. We developed A Clinically Oriented antimicrobial Resistance surveillance Network (ACORN) as a lightweight but comprehensive platform, in which we combine clinical data collection with diagnostic stewardship, microbiological data collection and visualisation of the linked clinical-microbiology dataset. Data are compatible with WHO GLASS surveillance and can be stratified by syndrome and other metadata. Summary metrics can be visualised and fed back directly for clinical decision-making and to inform local treatment guidelines and national policy. Methods: An ACORN pilot was implemented in three hospitals in Southeast Asia (1 paediatric, 2 general) to collect clinical and microbiological data from patients with community- or hospital-acquired pneumonia, sepsis, or meningitis. The implementation package included tools to capture site and laboratory capacity information, guidelines on diagnostic stewardship, and a web-based data visualisation and analysis platform. Results: Between December 2019 and October 2020, 2294 patients were enrolled with 2464 discrete infection episodes (1786 community-acquired, 518 healthcare-associated and 160 hospital-acquired). Overall, 28-day mortality was 8.7%. Third generation cephalosporin resistance was identified in 54.2% (39/72) of E. coli and 38.7% (12/31) of K. pneumoniae isolates . Almost a quarter of S. aureus isolates were methicillin resistant (23.0%, 14/61). 290/2464 episodes could be linked to a pathogen, highlighting the level of enrolment required to achieve an acceptable volume of isolate data. However, the combination with clinical metadata allowed for more nuanced interpretation and immediate feedback of results. Conclusions: ACORN was technically feasible to implement and acceptable at site level. With minor changes from lessons learned during the pilot ACORN is now being scaled up and implemented in 15 hospitals in 9 low- and middle-income countries to generate sufficient case-based data to determine incidence, outcomes, and susceptibility of target pathogens among patients with infectious syndromes.

ACORN (A Clinically-Oriented Antimicrobial Resistance Surveillance Network): a pilot protocol for case based antimicrobial resistance surveillance.

Background: Antimicrobial resistance (AMR) / drug resistant infections (DRIs) are a major global health priority. Surveillance data is critical to inform infection treatment guidelines, monitor trends, and to assess interventions. However, most existing AMR / DRI surveillance systems are passive and pathogen-based with many potential biases. Addition of clinical and patient outcome data would provide considerable added value to pathogen-based surveillance. Methods: The aim of the ACORN project is to develop an efficient clinically-oriented AMR surveillance system, implemented alongside routine clinical care in hospitals in low- and middle-income country settings. In an initial pilot phase, clinical and microbiology data will be collected from patients presenting with clinically suspected meningitis, pneumonia, or sepsis. Community-acquired infections will be identified by daily review of new admissions, and hospital-acquired infections will be enrolled during weekly point prevalence surveys, on surveillance wards. Clinical variables will be collected at enrolment, hospital discharge, and at day 28 post-enrolment using an electronic questionnaire on a mobile device. These data will be merged with laboratory data onsite using a flexible automated computer script. Specific target pathogens will be Streptococcus pneumoniae, Staphylococcus aureus, Salmonella spp ., Klebsiella pneumoniae, Escherichia coli, and Acinetobacter baumannii. A bespoke browser-based app will provide sites with fully interactive data visualisation, analysis, and reporting tools. Discussion: ACORN will generate data on the burden of DRI which can be used to inform local treatment guidelines / national policy and serve as indicators to measure the impact of interventions. Following development, testing and iteration of the surveillance tools during an initial six-month pilot phase, a wider rollout is planned.

Genomic surveillance for hypervirulence and multi-drug resistance in invasive Klebsiella pneumoniae from South and Southeast Asia.

BACKGROUND: Klebsiella pneumoniae is a leading cause of bloodstream infection (BSI). Strains producing extended-spectrum beta-lactamases (ESBLs) or carbapenemases are considered global priority pathogens for which new treatment and prevention strategies are urgently required, due to severely limited therapeutic options. South and Southeast Asia are major hubs for antimicrobial-resistant (AMR) K. pneumoniae and also for the characteristically antimicrobial-sensitive, community-acquired "hypervirulent" strains. The emergence of hypervirulent AMR strains and lack of data on exopolysaccharide diversity pose a challenge for K. pneumoniae BSI control strategies worldwide. METHODS: We conducted a retrospective genomic epidemiology study of 365 BSI K. pneumoniae from seven major healthcare facilities across South and Southeast Asia, extracting clinically relevant information (AMR, virulence, K and O antigen loci) using Kleborate, a K. pneumoniae-specific genomic typing tool. RESULTS: K. pneumoniae BSI isolates were highly diverse, comprising 120 multi-locus sequence types (STs) and 63 K-loci. ESBL and carbapenemase gene frequencies were 47% and 17%, respectively. The aerobactin synthesis locus (iuc), associated with hypervirulence, was detected in 28% of isolates. Importantly, 7% of isolates harboured iuc plus ESBL and/or carbapenemase genes. The latter represent genotypic AMR-virulence convergence, which is generally considered a rare phenomenon but was particularly common among South Asian BSI (17%). Of greatest concern, we identified seven novel plasmids carrying both iuc and AMR genes, raising the prospect of co-transfer of these phenotypes among K. pneumoniae. CONCLUSIONS: K. pneumoniae BSI in South and Southeast Asia are caused by different STs from those predominating in other regions, and with higher frequency of acquired virulence determinants. K. pneumoniae carrying both iuc and AMR genes were also detected at higher rates than have been reported elsewhere. The study demonstrates how genomics-based surveillance-reporting full molecular profiles including STs, AMR, virulence and serotype locus information-can help standardise comparisons between sites and identify regional differences in pathogen populations.

Antimicrobial resistance in Cambodia: a review.

OBJECTIVES: Following the launch of the Global Antimicrobial Resistance Surveillance System (GLASS), antimicrobial resistance (AMR) rates in many countries remain poorly described. This review provides an overview of published AMR data from Cambodia in the context of recently initiated national human and food-animal surveillance. METHODS: PubMed and the Cochrane Database of Systematic Reviews were searched for articles published from 2000 to 2018, which reported antimicrobial susceptibility testing (AST) data for GLASS specific organisms isolated from Cambodia. Articles were screened using strict inclusion/exclusion criteria. AST data was extracted, with medians and ranges of resistance rates calculated for specific bug-drug combinations. RESULTS: Twenty-four papers were included for final analysis, with 20 describing isolates from human populations. Escherichia coli was the most commonly described organism, with median resistance rates from human isolates of 92.8% (n=6 articles), 46.4% (n=4), 55.4% (n=8), and 46.4% (n=5) to ampicillin, 3rd generation cephalosporins, fluoroquinolones, and gentamicin respectively. CONCLUSIONS: Whilst resistance rates are high for several GLASS organisms, there were insufficient data to draw robust conclusions about the AMR situation in Cambodia. The recently implemented national AMR surveillance systems will begin to address this data gap.

Using machine learning to guide targeted and locally-tailored empiric antibiotic prescribing in a children's hospital in Cambodia.

Background: Early and appropriate empiric antibiotic treatment of patients suspected of having sepsis is associated with reduced mortality. The increasing prevalence of antimicrobial resistance reduces the efficacy of empiric therapy guidelines derived from population data. This problem is particularly severe for children in developing country settings. We hypothesized that by applying machine learning approaches to readily collect patient data, it would be possible to obtain individualized predictions for targeted empiric antibiotic choices. Methods and Findings: We analysed blood culture data collected from a 100-bed children's hospital in North-West Cambodia between February 2013 and January 2016. Clinical, demographic and living condition information was captured with 35 independent variables. Using these variables, we used a suite of machine learning algorithms to predict Gram stains and whether bacterial pathogens could be treated with common empiric antibiotic regimens: i) ampicillin and gentamicin; ii) ceftriaxone; iii) none of the above. 243 patients with bloodstream infections were available for analysis. We found that the random forest method had the best predictive performance overall as assessed by the area under the receiver operating characteristic curve (AUC). The random forest method gave an AUC of 0.80 (95%CI 0.66-0.94) for predicting susceptibility to ceftriaxone, 0.74 (0.59-0.89) for susceptibility to ampicillin and gentamicin, 0.85 (0.70-1.00) for susceptibility to neither, and 0.71 (0.57-0.86) for Gram stain result. Most important variables for predicting susceptibility were time from admission to blood culture, patient age, hospital versus community-acquired infection, and age-adjusted weight score. Conclusions: Applying machine learning algorithms to patient data that are readily available even in resource-limited hospital settings can provide highly informative predictions on antibiotic susceptibilities to guide appropriate empiric antibiotic therapy. When used as a decision support tool, such approaches have the potential to improve targeting of empiric therapy, patient outcomes and reduce the burden of antimicrobial resistance.

Climatic drivers of melioidosis in Laos and Cambodia: a 16-year case series analysis.

BACKGROUND: Burkholderia pseudomallei is the cause of melioidosis, a serious and difficult to treat infection that is endemic throughout the tropics. Melioidosis incidence is highly seasonal. We aimed to identify the climatic drivers of infection and to shed light on modes of transmission and potential preventive strategies. METHODS: We examined the records of patients diagnosed with melioidosis at the Microbiology Laboratory of Mahosot Hospital in Vientiane, Laos, between October, 1999, and August, 2015, and all patients with culture-confirmed melioidosis presenting to the Angkor Hospital for Children in Siem Reap, Cambodia, between February, 2009, and December, 2013. We also examined local temperature, humidity, precipitation, visibility, and wind data for the corresponding time periods. We estimated the B pseudomallei incubation period by examining profile likelihoods for hypothetical exposure-to-presentation delays. FINDINGS: 870 patients were diagnosed with melioidosis in Laos and 173 patients were diagnosed with melioidosis in Cambodia during the study periods. Melioidosis cases were significantly associated with humidity (p<0·0001), low visibility (p<0·0001), and maximum wind speeds (p<0·0001) in Laos, and humidity (p=0·010), rainy days (p=0·015), and maximum wind speed (p=0·0070) in Cambodia. Compared with adults, children were at significantly higher odds of infection during highly humid months (odds ratio 2·79, 95% CI 1·83-4·26). Lung and disseminated infections were more common during windy months. The maximum likelihood estimate of the incubation period was 1 week (95% CI 0-2). INTERPRETATION: The results of this study demonstrate a significant seasonal burden of melioidosis among adults and children in Laos and Cambodia. Our findings highlight the risks of infection during highly humid and windy conditions, and suggest a need for increased awareness among at-risk individuals, such as children. FUNDING: Wellcome Trust.

Prospective surveillance of healthcare associated infections in a Cambodian pediatric hospital.

BACKGROUND: Healthcare associated infections (HAI) are the most common preventable adverse events following admission to healthcare facilities. Data from low-income countries are scarce. We sought to prospectively define HAI incidence at Angkor Hospital for Children (AHC), a Cambodian pediatric referral hospital. METHODS: Prospective HAI surveillance was introduced for medical admissions to AHC. Cases were identified on daily ward rounds and confirmed using locally adapted Centers for Disease Control and Prevention (CDC) definitions. During the surveillance period, established infection prevention and control (IPC) activities continued, including hand hygiene surveillance. In addition, antimicrobial stewardship practices such as the creation of an antimicrobial guideline smartphone app were introduced. RESULTS: Between 1st January and 31st December 2015 there were 3,263 medical admissions and 102 HAI cases. The incidence of HAI was 4.6/1,000 patient-days (95% confidence interval 3.8-5.6) and rates were highest amongst neonates. Median length of stay was significantly longer in HAI cases: 25 days versus 5 days for non-HAI cases (p < 0.0001). All-cause in-hospital mortality increased from 2.0 to 16.1% with HAI (p < 0.0001). Respiratory infections were the most common HAI (54/102; 52.9%). Amongst culture positive infections, Gram-negative organisms predominated (13/16; 81.3%). Resistance to third generation cephalosporins was common, supporting the use of more expensive carbapenem drugs empirically in HAI cases. The total cost of treatment for all 102 HCAI cases combined, based on additional inpatient days, was estimated to be $299,608. CONCLUSIONS: Prospective HAI surveillance can form part of routine practice in low-income healthcare settings. HAI incidence at AHC was relatively low, but human and financial costs remained high due to increased carbapenem use, prolonged admissions and higher mortality rates.