A metagenomics method for the quantitative detection of bacterial pathogens causing hospital-associated and ventilator-associated pneumonia.

IMPORTANCE: The management of ventilator-associated pneumonia and hospital-acquired pneumonia requires rapid and accurate quantitative detection of the infecting pathogen. To this end, we propose a metagenomic sequencing assay that includes the use of an internal sample processing control for the quantitative detection of 20 relevant bacterial species from bronchoalveolar lavage samples.

Machine learning in the clinical microbiology laboratory: has the time come for routine practice?

BACKGROUND: Machine learning (ML) allows the analysis of complex and large data sets and has the potential to improve health care. The clinical microbiology laboratory, at the interface of clinical practice and diagnostics, is of special interest for the development of ML systems. AIMS: This narrative review aims to explore the current use of ML In clinical microbiology. SOURCES: References for this review were identified through searches of MEDLINE/PubMed, EMBASE, Google Scholar, biorXiv, arXiV, ACM Digital Library and IEEE Xplore Digital Library up to November 2019. CONTENT: We found 97 ML systems aiming to assist clinical microbiologists. Overall, 82 ML systems (85%) targeted bacterial infections, 11 (11%) parasitic infections, nine (9%) viral infections and three (3%) fungal infections. Forty ML systems (41%) focused on microorganism detection, identification and quantification, 36 (37%) evaluated antimicrobial susceptibility, and 21 (22%) targeted the diagnosis, disease classification and prediction of clinical outcomes. The ML systems used very diverse data sources: 21 (22%) used genomic data of microorganisms, 19 (20%) microbiota data obtained by metagenomic sequencing, 19 (20%) analysed microscopic images, 17 (18%) spectroscopy data, eight (8%) targeted gene sequencing, six (6%) volatile organic compounds, four (4%) photographs of bacterial colonies, four (4%) transcriptome data, three (3%) protein structure, and three (3%) clinical data. Most systems used data from high-income countries (n = 71, 73%) but a significant number used data from low- and middle-income countries (n = 36, 37%). Performance measures were reported for the 97 ML systems, but no article described their use in clinical practice or reported impact on processes or clinical outcomes. IMPLICATIONS: In clinical microbiology, ML has been used with various data sources and diverse practical applications. The evaluation and implementation processes represent the main gap in existing ML systems, requiring a focus on their interpretability and potential integration into real-world settings.

From genotype to antibiotic susceptibility phenotype in the order Enterobacterales: a clinical perspective.

OBJECTIVES: Predicting the antibiotic susceptibility phenotype from genomic data is challenging, especially for some specific antibiotics in the order Enterobacterales. Here we aimed to assess the performance of whole genomic sequencing (WGS) for predicting the antibiotic susceptibility in various Enterobacterales species using the detection of antibiotic resistance genes (ARGs), specific mutations and a knowledge-based decision algorithm. METHODS: We sequenced (Illumina MiSeq, 2×250 bp) 187 clinical isolates from species possessing (n = 98) or not (n = 89) an intrinsic AmpC-type cephalosporinase. Phenotypic antibiotic susceptibility was performed by the disc diffusion method. Reads were assembled by A5-miseq and ARGs were identified from the ResFinder database using Diamond. Mutations on GyrA and ParC topoisomerases were studied. Piperacillin, piperacillin-tazobactam, ceftazidime, cefepime, meropenem, amikacin, gentamicin and ciprofloxacin were considered for prediction. RESULTS: A total of 1496 isolate/antibiotic combinations (187 isolates × 8 antibiotics) were considered. In 230 cases (15.4%), no attempt of prediction was made because it could not be supported by current knowledge. Among the 1266 attempts, 1220 (96.4%) were correct (963 for predicting susceptibility and 257 for predicting resistance), 24 (1.9%) were major errors (MEs) and 22 (1.7%) were very major errors (VMEs). Concordance were similar between non-AmpC and AmpC-producing Enterobacterales (754/784 (96.2%) vs 466/482 (96.7%), chi-square test p 0.15), but more VMEs were observed in non-AmpC producing strains than in those producing an AmpC (19/784 (2.4%) vs 3/466 (0.6%), chi-square test p 0.02). The majority of VMEs were putatively due to the overexpression of chromosomal genes. CONCLUSIONS: In conclusion, the inference of antibiotic susceptibility from genomic data showed good performances for non-AmpC and AmpC-producing Enterobacterales species. However, more knowledge about the mechanisms underlying the derepression of AmpC are needed.

Travel and acquisition of multidrug-resistant Enterobacteriaceae.

Bacterial resistance to antibiotics is considered a major threat to health. Enterobacteriaceae have increasingly become resistant to antibiotics through the acquisition and dissemination of extended-spectrum beta-lactamases (ESBL) that confer resistance to most beta-lactams. While ESBL-producing Enterobacteriaceae were formerly restricted to hospitals, they have now spread to community settings, especially in developing countries. The tremendous expansion of international travels contributed to the importation of multidrug-resistant Enterobacteriaceae (MRE) to low prevalence countries. Several studies reported that 21 to 51% of healthy travelers acquire a MRE when travelling abroad, depending on the visited region (Asia, and especially South Asia being associated with the highest risk - up to 85%). Traveling to Africa or the Middle East is associated with lower but still disturbing rates (13-44%). In addition, the occurrence of digestive disorders and/or diarrhea and antibiotic intake increase the risk of MRE acquisition by 2-3 folds. After traveling though, the length of MRE carriage seems to be short (<1 month) and the risk of transmission within the household appears to be low. Nonetheless and beyond the intestinal carriage of MRE, traveling to endemic areas has also been pointed as a risk factor for infections involving MRE, mainly urinary tract infections.

In vivo selection of a multidrug-resistant Aeromonas salmonicida during medicinal leech therapy.

We report the selection in a 15-year-old boy of a multidrug-resistant, extended-spectrum ß-lactamase (ESBL)-producing Aeromonas salmonicida after medicinal leech therapy that required an antibiotic prophylaxis based on piperacillin/tazobactam and cotrimoxazole. Whole genome sequencing of the strain indeed revealed 13 antibiotic resistance genes, including the ESBL CTX-M-3 and the unusual ß-lactamase SCO-1.

Erratum to: Antimicrobials: a global alliance for optimizing their rational use in intra-abdominal infections (AGORA).

[This corrects the article DOI: 10.1186/s13017-016-0089-y.].

Clonal or not clonal? Investigating hospital outbreaks of KPC-producing Klebsiella pneumoniae with whole-genome sequencing.

OBJECTIVES: Whole-genome sequencing (WGS) is a promising tool for identifying transmission pathways in outbreaks caused by multidrug-resistant bacteria. However, it is uncertain how the data produced by WGS can be best integrated into epidemiologic investigations. METHODS: We tested various genomic analyses to identify clonal groups in two distinct outbreaks of Klebsiella pneumoniae carbapenemase-producing K. pneumoniae that occurred in Switzerland in 2013 and 2015. In blinded fashion, we sequenced 12 strains involved in the two outbreaks, respectively, and six that were epidemiologically unrelated. We analysed genomic commonalities from conserved genes to plasmid-borne antibiotic resistance genes (ARGs) and contrasted these results with available epidemiologic evidence. RESULTS: Using WGS, blinded analysts correctly identified the two clusters of strains from the two outbreaks. Nonetheless, the 2015 index strain was found to be slightly different (1-3 single nucleotide variants) from the strains recovered from secondary cases, likely because prior long-term carriage (3 years) by the index patient allowed for genetic mutations over time. Also, we observed occasional loss of ARG-bearing plasmidic fragments in outbreak-causing strains. CONCLUSIONS: Retrospective WGS analysis was successful in identifying clonal groups in both outbreaks. Still, data should be analysed with caution in cases of previous long-term carriage of the studied bacteria.

Impact of a short exposure to levofloxacin on faecal densities and relative abundance of total and quinolone-resistant Enterobacteriaceae.

Emergence of resistant Enterobacteriaceae in the intestinal microbiota during antibiotic treatment is well documented but its early dynamic is not. Here, we compared the densities of total Enterobacteriaceae and relative abundance (RA) of quinolone-resistant Enterobacteriaceae (QRE) in the first stool passed by patients who had a short exposure to levofloxacin (levofloxacin, n=12) or not (control, n=8). Mean densities (SD) (log CFU/g stool) of total Enterobacteriaceae were lower in the levofloxacin group than in the control group-3.4 (1.6) versus 6.7 (1.7), respectively, p <0.001. Conversely, mean RA (SD) of QRE was significantly higher in the levofloxacin group than in the control group-49.7% (23.4) versus 0.1% (3.2), respectively, p <0.05). In conclusion, even a short exposure to levofloxacin has a profound impact on the densities of total Enterobacteriaceae and the QRE-RA.

Elaboration of a consensual definition of de-escalation allowing a ranking of ß-lactams.

Empirical broad spectrum antimicrobial therapy prescribed in life-threatening situations should be de-escalated to mitigate the risk of resistance emergence. Definitions of de-escalation (DE) vary among studies, thereby biasing their results. The aim of this study was to provide a consensus definition of DE and to establish a ranking of ß-lactam according to both their spectra and their ecological consequences. Twenty-eight experts from intensive care, infectious disease and clinical microbiology were consulted using the Delphi method (four successive questionnaires) from July to November 2013. More than 70% of similar answers to a question were necessary to reach a consensus. According to our consensus definition, DE purpose was to reduce both the spectrum of antimicrobial therapy and the selective pressure on microbiota. DE included switching from combination to monotherapy. A six-rank consensual classification of ß-lactams allowing gradation of DE was established. The group was unable to differentiate ecological consequences of molecules included in group 4, i.e. piperacillin/tazobactam, ticarcillin/clavulanic acid, fourth-generation cephalosporin and antipseudomonal third-generation cephalosporin. Furthermore, no consensus was reached on the delay within which DE should be performed and on whether or not the shortening of antibiotic therapy duration should be included in DE definition. This study provides a consensual ranking of ß-lactams according to their global ecological consequences that may be helpful in future studies on DE. However, this work also underlines the difficulties of reaching a consensus on the relative ecological impact of each individual drug and on the timing of DE.

Performance and impact of a rapid method combining mass spectrometry and direct antimicrobial susceptibility testing on treatment adequacy of patients with ventilator-associated pneumonia.

Inappropriate antibiotic therapy in ventilator-associated pneumonia (VAP) is associated with increased mortality. Using broad-spectrum antibiotics for 48 h until the results of conventional cultures and antimicrobial susceptibility testing (AST) are available, may promote the emergence of drug-resistant bacteria. Performing AST directly on clinical respiratory samples would hasten the process by at least 24 h. Here, we analysed the diagnostic performance of a rapid method combining mass spectrometry and direct AST (DAST), and compared it with the conventional method (mass spectrometry with conventional AST (CAST)). Additionally, we assessed its potential impact on antimicrobial use in patients. Over a period of 18 months, the two methods were performed on 85 bronchoalveolar lavages obtained from intensive care unit patients with suspected VAP, and in which Gram-negative bacilli were observed on direct examination. Only the CAST results were reported to the clinicians. DAST produced useable results in 85.9% of the patients. The sensitivity and negative predictive values of DAST were 100% for all antibiotics tested, except gentamicin (97.1%, (95% CI 93.3-101) and 97.4% (93.7-101), respectively) and amikacin (88.9% (81.7-96.1) and 96.4% (92.1-100.7), respectively), compared with CAST. Specificity and positive predictive values ranged from 82.9 (74.2-91.5) to 100%, and from 86.4 (78.5-94.2) to 100%, respectively. If the DAST results had been reported to the clinicians, treatment could have been optimized 24 h earlier in 35/85 (41.2%) patients, with 17 carbapenem patient-days saved. Overall, routine use of the DAST method could help optimize earlier antibiotic treatment in patients with suspected VAP.

Activity of temocillin in a murine model of urinary tract infection due to Escherichia coli producing or not producing the ESBL CTX-M-15.

OBJECTIVES: Temocillin is a 6α-methoxy derivative of ticarcillin that is resilient to ESBLs. Prospective data about its in vivo activity remain scarce. Our aims were: (i) to evaluate the activity of temocillin in a urinary tract infection (UTI) model due to ESBL-producing Escherichia coli and compare it with that of imipenem; and (ii) to define in vivo susceptibility breakpoints. METHODS: Mice were infected with a susceptible E. coli CFT073-RR or its transconjugant (CFT073-RR Tc) harbouring a blaCTX-M-15-carrying plasmid, using an ascending UTI model. Therapeutic regimens were chosen in order to reproduce percentage of time of free drug concentrations above MIC (fT>MIC) obtained in humans with standard regimens of temocillin (200 mg/kg every 2 h for 2 g every 12 h) or imipenem (100 mg/kg every 2 h for 1 g every 8 h). Additional regimens of temocillin (200 mg/kg every 4 and 6 h) with reduced fT>MIC were studied. RESULTS: MICs of temocillin and imipenem were 4/8 and 0.5/0.5 mg/L, for CFT073-RR and CFT073-RR Tc, respectively. In vivo, when given every 2 h (fT>MIC = 82% and 70%), temocillin was bactericidal and as effective as imipenem in kidneys against both strains without selecting resistant mutants. Temocillin remained active even when given every 4 h, generating an fT>MIC of 41% and 35%, which corresponded to a breakpoint of 16 mg/L in humans with the standard regimen. CONCLUSIONS: Our observations support the consideration of a standard regimen of temocillin as an alternative to carbapenems for the treatment of UTI due to CTX-M-producing E. coli strains with an MIC of 16 mg/L or less.

Acquisition of carbapenemase-producing Enterobacteriaceae by healthy travellers to India, France, February 2012 to March 2013.

Healthy travellers to countries where carbapenemases-producing Enterobacteriaceae (CPE) are endemic might be at risk for their acquisition, even without contact with the local healthcare system. Here, we report the acquisition of CPE (two OXA-181, one New Delhi metallo-beta-lactamase 1 (NDM-1)) in three healthy travellers returning from India. The duration of CPE intestinal carriage was less than one month. The results indicate that healthy travellers recently returning from India might be considered as at risk for CPE carriage.

Enterobacteriaceae bacteremia: risk factors for ESBLPE.

OBJECTIVES: The increasing prevalence of extended spectrum beta-lactamase producing enterobacteriaceae (ESBLPE) requires defining the use of carbapenems in first intention. We analyzed the associations between enterobacteriaceae bacteremia (EbBact) and ESBLPE carriage during 10 years in a 950-bed teaching hospital. METHODS: We analyzed a 10-year (July 2001 to June 2011) prospective collection of bacteremia cases including 2 databases: (1) EbBact and (2) a computerized database of patients carrying EBLSE. Only one episode of EbBact was analyzed per patient and hospital stay. Factors associated with ESBLPE bacteremia were assessed by univariate and multivariate logistic regression analysis. RESULTS: Overall, 2355 cases of EbBact were identified, among which 135 (5.7%) were ESBLPE (2001-05: 1.4%, 2006-09: 7.6%, 2010-11: 14.2%). ESBLPE bacteremia was observed in 52 of the 88 (59%) patients carrying ESBLPE and in 83/2267 (3.7%) patients not known to be colonized with ESBLPE. Factors associated with ESBLPE bacteremia in patients not known to be colonized were: female gender (ORa=0.56, CI95% [0.34-0.91]), hospitalization in the ICU (ORa=2.51 [1.27-5.05]) or medical/surgical wards (ORa=1.83 [1.04-3.38]), the period (2006-09, ORa=4.08 [2.21-8.16]; 2010-11, ORa=8.17 [4.14-17.06] compared to 2001-05), and history of EbBact (ORa=2.29 [0.97-4.79]). CONCLUSION: In case of EbBact, patients known to be colonized with ESBLPE present with ESBLPE bacteremia in more than half of the cases, requiring carbapenems as empirical antibiotic treatment. The global prevalence of ESBLPE among patients presenting with EbBact not known to be colonized with ESBLPE was 3.7%.

Clinical predictive values of extended-spectrum beta-lactamase carriage in patients admitted to medical wards.

We aimed to reassess, through clinical items, populations at risk for extended-spectrum beta-lactamase-producing Enterobacteriaceae (ESBL-E) carriage at admission to hospital and to assess the risk of further positive clinical culture of ESBL-E among carriers. We performed a 5-month cohort study in a medicine ward of a 500-bed university teaching hospital in the Parisian area of France. All admitted patients were prospectively enrolled for rectal swabbing and clinical data collection, including bacterial infection at admission and during stay. Variables associated with ESBL-E carriage were identified by univariate and multivariate analysis. Five hundred patients were included. The prevalence of ESBL-E was 6.6% (33/500) upon admission. Only previous carriage of multidrug-resistant bacteria (MDRB) was associated with carriage (odds ratio [OR]: 17.7, 95% confidence interval (CI) 5.8-54.2, p < 0.001), yet, the positive predictive value (PPV) was not higher than 50%. When prior MDRB carriage was not considered in the multivariate analysis, only prior antibiotic consumption was found to be associated with carriage at admission (OR: 2.2 [1.1-4.5], p = 0.02). Two patients had ESBL-E infection at admission, yet, no patient became infected with ESBL-E during their stay. The clinical prediction of ESBL carriage at admission in our wards was found to be poorly efficient for assessing the at-risk population.

Genesis of a KPC-producing Klebsiella pneumoniae after in vivo transfer from an imported Greek strain.

We document here the in vivo transfer of bla(KPC-2) between intensive care unit-acquired and a commensal strain of K. pneumoniae in a French patient after his repatriation from Greece. This first report of an emerging KPC-producing strain in France raises further concerns about the spread of carbapenem resistance among Enterobacteriaceae.

Secular trends and dynamics of hospital associated methicillin-resistant Staphylococcus aureus.

We performed an 11-year retrospective analysis of consecutive nonduplicate methicillin-resistant Staphylococcus aureus (MRSA) clinical isolates in two neighbouring hospitals in the Paris area. MRSA isolates were classified according to resistance (R) to fluoroquinolones (Fq), kanamycin (K), tobramycin (T) and gentamicin (G). The yearly number of MRSA isolates (3446 in total) decreased, from approximately 350 in 1997­2002 to 212 in 2007. Four patterns (P) were found: P1 (KTGFq R, n = 776), P2 [KTFq R; G susceptible (S), n = 1630], P3 (Fq R; KTG S, n = 397) and P4 (Fq S; any KTG susceptibility, n = 201). P1 predominated in 1997 (183 isolates) then dropped sharply (nine in 2007); P2 and P4 remained stable over time; and P3 increased from 13 isolates in 1997 to 72 in 2007. Patterns were significantly and positively associated with several variables, independently of the year of collection: P1, age < 80 years, male gender, intensive care unit stay, and hospital onset; P3, age > 80 years and stay in intermediate or long-term care wards; and P4, age < 40 years, stay in an obstetric ward, and imported cases. Molecular typing of 79 isolates in 2005 and 2007 using multilocus sequence typing, spa type, and SCCmec showed that P1, P2 and P3 isolates were mainly clonal, whereas P4 isolates were more diverse. P1 comprised mainly ST247-I isolates, P2 mainly ST8-IVc, and P3 mainly ST8-IVc and ST5-VI. In conclusion, the epidemiology of MRSA in Paris is changing rapidly at the local level, with phenotypically defined clones being substituted by others, with associations existing between changes in specific patient populations or circumstances.