Dynamical systems of self-organized segregation

We re-consider Schelling’s (1971) bounded neighbourhood model as put into the form of a dynamical system by Haw and Hogan (2018). In the case of a single neighbourhood we explain the occurring bifurcation set, thereby correcting a minor scaling error. In the case of two neighbourhoods we correct a major error and derive a dynamical system that does satisfy the modeling assumptions made by Haw and Hogan (2020), staying as close as possible to their construction. We find that stable integration then is only possible if the populations in the two neighbourhoods have the option to be in neither neighbourhood. In the absence of direct movement between the neighbourhoods the problem is furthermore equivalent to independent single neighbourhood problems.

Antimicrobial Susceptibility Testing for Staphylococcus lugdunensis

Evaluation of penicillin and oxacillin susceptibility testing was conducted on two hundred Staphylococcus lugdunensis isolates. Disc diffusion with penicillin 1 IU (P1, EUCAST) and penicillin 10 IU (P10, CLSI) was compared with nitrocefin discs (Cefinase®) and automated broth microdilution (Vitek2®). Oxacillin susceptibility was extrapolated from cefoxitin 30μg disc diffusion (FOX) and compared with Vitek2®. Reference methods were blaZ and mecA PCR. Penicillin zone diameter and zone edge correlated with blaZ in all except two P10 susceptible isolates (VME; very major error) and one P1 resistant isolate (ME). One hundred and forty-eight isolates were blaZ -negative of which one hundred and forty-six and one hundred and forty-nine isolates were susceptible by P1 and P10 respectively. One hundred and twenty-seven isolates were penicillin susceptible by Vitek2®. Vitek2® overcalled resistance in twenty-one blaZ -negative, twenty P1 and twenty-two P10 susceptible isolates (Vitek2® ME rate, 14.2%). Two mecA -positive isolates were oxacillin resistant by FOX and Vitek2® (categorical agreement). However, eighteen FOX susceptible, mecA -negative isolates tested resistant by Vitek2®. In conclusion, Vitek2® over-estimated penicillin and oxacillin resistance compared with disc diffusion and PCR. Disc diffusion with zone edge interpretation was more accurate and specific than automated broth microdilution for S. lugdunensis in our study.

Performance Evaluation of the Gradient Diffusion Strip Method and Disk Diffusion Method for Ceftazidime–Avibactam Against Enterobacterales and Pseudomonas aeruginosa: A Dual-Center Study

Objectives: Ceftazidime–avibactam is a novel synthetic beta-lactam + beta-lactamase inhibitor combination. We evaluated the performance of the gradient diffusion strip method and the disk diffusion method for the determination of ceftazidime–avibactam against Enterobacterales and Pseudomonas aeruginosa.Methods: Antimicrobial susceptibility testing of 302 clinical Enterobacterales and Pseudomonas aeruginosa isolates from two centers were conducted by broth microdilution (BMD), gradient diffusion strip method, and disk diffusion method for ceftazidime–avibactam. Using BMD as a gold standard, essential agreement (EA), categorical agreement (CA), major error (ME), and very major error (VME) were determined according to CLSI guidelines. CA and EA rate > 90%, ME rate < 3%, and VME rate < 1.5% were considered as acceptable criteria. Polymerase chain reaction and Sanger sequencing were performed to determine the carbapenem resistance genes of all 302 isolates.Results: A total of 302 strains were enrolled, among which 182 strains were from center 1 and 120 strains were from center 2. A percentage of 18.21% (55/302) of the enrolled isolates were resistant to ceftazidime–avibactam. The CA rates of the gradient diffusion strip method for Enterobacterales and P. aeruginosa were 100% and 98.65% (73/74), respectively, and the EA rates were 97.37% (222/228) and 98.65% (73/74), respectively. The CA rates of the disk diffusion method for Enterobacterales and P. aeruginosa were 100% and 95.95% (71/74), respectively. No VMEs were found by using the gradient diffusion strip method, while the ME rate was 0.40% (1/247). No MEs were found by using the disk diffusion method, but the VME rate was 5.45% (3/55). Therefore, all the parameters of the gradient diffusion strip method were in line with acceptable criteria. For 31 blaKPC, 33 blaNDM, 7 blaIMP, and 2 blaVIM positive isolates, both CA and EA rates were 100%; no MEs or VMEs were detected by either method. For 15 carbapenemase-non-producing resistant isolates, the CA and EA rates of the gradient diffusion strips method were 100%. Whereas the CA rate of the disk diffusion method was 80.00% (12/15), the VME rate was 20.00% (3/15).Conclusion: The gradient diffusion strip method can meet the needs of clinical microbiological laboratories for testing the susceptibility of ceftazidime–avibactam drugs. However, the VME rate > 1.5% (5.45%) by the disk diffusion method. By comparison, the performance of the gradient diffusion strip method was better than that of the disk diffusion method.

Use of Fosfomycin Etest® to Determine In Vitro Susceptibility of Clinical Isolates of Enterobacterales Other Than Escherichia coli , Non-Fermenting Gram-Negative Bacilli, and Gram-positive Cocci

Clinical isolates of Enterobacterales other than Escherichia coli (EOTEC), non-fermenting Gram-negative bacilli, and Gram-positive cocci were tested for susceptibility to fosfomycin using Etest ® and reference agar dilution. Applying EUCAST (v. 11.0, 2021) intravenous fosfomycin breakpoints, Etest ® MICs for EOTEC showed essential agreement (EA), categorical agreement (CA), major error (ME), and very major error (VME) rates of 70.4%, 88.4%, 4.1%, and 32.1%, respectively. No species of EOTEC tested with acceptable rates for all of EA (≥90%), CA (≥90%), ME (≤3%), and VME (≤3%). Etest ® MICs for Enterococcus faecalis , interpreted using CLSI oral/urine criteria (M100, 2021), showed EA, CA, minor error, ME, and VME rates of 98.5%, 81.2%, 18.8%, 0%, and 0%. Against Staphylococcus aureus , EA, CA, and ME rates were 84.1%, 98.7%, and 1.3% (EUCAST intravenous criteria). S. aureus isolates with fosfomycin MICs >32 μg/ml (resistant) were not identified by agar dilution. We conclude performing fosfomycin Etest ® on isolates of S. aureus will reliably identify fosfomycin-susceptible isolates with low, acceptable rates of MEs and VMEs. Testing of urinary isolates of E. faecalis by Etest ® is associated with an unacceptably high rate of minor errors (18.8%) but low, acceptable rates of MEs and VMEs when results are interpreted using CLSI criteria. Isolates of EOTEC tested by Etest ® with resulting MICs interpreted by EUCAST criteria were associated with an unacceptably high VME rate (32.1%). In vitro testing of clinical isolates beyond E. coli , E. faecalis , and S. aureus to determine susceptibility to fosfomycin is problematic with current methods and breakpoints.

Neural network-based predictions of antimicrobial resistance in Salmonella spp. using k-mers counting from whole-genome sequences

Artificial intelligence-based predictions have emerged as a friendly and reliable tool for the surveillance of the antimicrobial resistance (AMR) worldwide. In this regard, genome databases typically include whole-genome sequencing (WGS) data containing AMR metadata that can be used to train machine learning (ML) models, in order to predict phenotype features from genome samples. In this study, using a Neural Network (NN) architecture and the SGD-ADAM algorithm, we build ML antibiotic resistance models that can predict Minimum Inhibitory Concentrations (MICs) and antimicrobial susceptibility profiles of Salmonella spp. Data analysis was based on 7,268 genomes publicly available in PATRIC database, containing about 75,000 AMR annotations. ML models were built using reference-free k-mer analysis of whole-genome sequences, MIC measurements and susceptibility categories, obtaining robust and accurate results for 9 antibiotics belonging to beta-lactam, fluoroquinolone, phenicol, aminoglycoside, tetracycline and sulphonamide classes. Although the accuracy of predicting the actual MIC reaches modest levels, the within +/- 1 2-fold dilution accuracy per antibiotic reaches significant levels with values that varies from 85% to 95%, with narrow 95% CIs of about 5% and individual accuracies per MIC ≥ 80%. For differentiation between ''susceptible'' and ''resistant'' values, by measuring the accuracy and error of model's susceptibility predictions to different antibiotics, the accuracy is the same as before and ranges from 85% to 95%, with 95% CIs of about 5%, the recall extends from 75% to 85%, the precision from 60% to 90%, whereas the very major error is ≤ 20%. In summary, these results show that NN-based models are able to learn and predict the AMR phenotype from bacterial genomes based on a gene-free k-mer analysis.

Evaluating the performance characteristics of different antimicrobial susceptibility testing methodologies for testing susceptibility of gram-negative bacteria to tigecycline

Background The current emergence of multi-drug resistance among nosocomial pathogens has led to increased use of last-resort agents including Tigecycline (TGC). Availability of reliable methods for testing TGC susceptibility is crucial to accurately predict clinical outcomes. We evaluated the influence of different methodologies and type of media on TGC susceptibility of different gram-negative bacteria of clinical origin. Methods The TGC susceptibility of 84 clinical isolates of Klebsiella pneumoniae (n = 29), Escherichia coli (n = 30), and Acinetobacter baumannii (n = 25) was tested by broth microdilution (BMD), Etest, agar dilution (AD) and disk diffusion (DD) methods using Mueller Hinton agar from Difco and Mueller Hinton broth (MHB) from two different manufacturers (Difco and Condalab). FDA TGC susceptibility breakpoints issued for Enterobacteriaceae were used for interpretation of the results. Results MICs determined by BMD using MHB from two suppliers showed a good correlation with overall essential agreement (EA) and categorical agreement (CA) being 100% and 95% respectively. However, a twofold rise in BMD-Condalab MICs which was detected in 50% of the isolates, resulted in changes in susceptibility categories of few isolates with MICs close to susceptibility breakpoints leading to an overall minor error (MI) rate of 4.7%. Among the tested methods, Etest showed the best correlation with BMD, being characterized with the lowest error rates (only 1% MI) and highest overall EA (100%) and CA (98.8%) for all subsets of isolates. AD yielded the lowest overall agreement (EA 77%, CA 81%) with BMD in a species dependent manner, with the highest apparent discordance being found among the A. baumannii isolates. While the performance of DD for determination of TGC susceptibility among Enterobacteriaceae was excellent, (CA:100% with no errors), the CA was lower (84%) when it was used for A. baumannii where an unacceptably high minor-error rate was noted (16%). No major error or very major error was detected for any of the tested methods. Conclusions Etest can be reliably used for TGC susceptibility testing in the three groups of studied bacteria. For the isolates with close-to-breakpoint MICs, testing susceptibility using the reference method is recommended.

E-Test or Agar Dilution for Metronidazole Susceptibility Testing of Helicobacter Pylori: Importance of the Prevalence of Metronidazole Resistance

Background: A number of studies have shown that E-test overestimated the presence of Helicobacter pylori (H. pylori) resistance compared to agar dilution.Objective: The purpose of this study was to explore whether E-test could be an alternative for agar dilution to detect the metronidazole susceptibility of H. pylori.Method: E-test and agar dilution were used to assess susceptibility of H. pylori to metronidazole, clarithromycin and levofloxacin in 281 clinical isolates obtained from China where resistance was high. Cohen kappa analysis, McNemar test, essential and categorical agreement analysis were performed for these two methods. Results: Overall, the result of E-test showed similar prevalence of resistance rate to all antibiotics compared with agar dilution. The essential agreement (EA) of E-test method and agar dilution in the evaluation susceptibility of H. pylori to clarithromycin and levofloxacin were moderate, with 89.0% and 79.7% respectively, but only 45.9% for metronidazole. Results showed categorical agreement (CA) between E-test and agar dilution were 100% for both clarithromycin and levofloxacin. As for metronidazole, the CA was 98.7%, no major error was identified, and rate of very major error was 1.8%.Conclusion: E-test can be an alternative method to detect the metronidazole susceptibility of H. pylori in regions where high-level resistance is common.

Azole resistant Aspergillus fumigatus clinical isolate screening in azole-containing agar plates (EUCAST E.Def 10.1): low impact of plastic trays used and poor performance in cryptic species

Azole-containing agar is used in routine Aspergillus fumigatus azole resistance screening. We evaluated the impact of the type of plastic used to prepare in-house agar plates on the procedurés performance against A. fumigatus sensu stricto and cryptic species. A. fumigatus sensu stricto (n=91) and cryptic species (n=52) were classified as susceptible or resistant (EUCAST E.Def 9.3.2; clinical breakpoints v10). In-house azole-containing agar plates were prepared following EUCAST E.Def 10.1 on three types of multi-dish plates. We assessed the sensitivity, specificity, and agreement values of the agar plates to screen for azole resistance. Overall, sensitivity and specificity values of the agar screening method were 100% and 93.3%, respectively. The type of tray used did not affect these values. All isolates harbouring TR 34 -L98H substitutions were classified as resistant to itraconazole and voriconazole by the agar method; however, false susceptibility (very major error) to posaconazole was not uncommon and happened in isolates with posaconazole MICs of 0.25 mg/L. Isolates harbouring G54R and TR 46 -Y121F-T289A substitutions were correctly classified by the agar method as itraconazole/posaconazole resistant and voriconazole-resistant, respectively. False resistance (major error) occurred in isolates showing tiny fungal growth. Finally, agreements between both procedures against cryptic species were much lower. Azole-containing agar plates are a convenient and reliable tool to screen for resistance in A. fumigatus sensu stricto ; the type of plastic tray used minimally affects the method. On the contrary, the performance against cryptic species is rather poor.

Detection of Colistin Resistance in Carbapenem Resistant Enterobacteriaceae by Reference Broth Microdilution and Comparative Evaluation of Three Other Methods

Objective Challenges in susceptibility testing of colistin along with increase in the prevalence of colistin-resistant carbapenemase-producing Enterobacteriaceae (CRE) pathogens needs addressal. Evaluation of user-friendly methods is necessary as an alternative to broth microdilution (BMD), the reference susceptibility testing method, for routine implementation in diagnostic clinical microbiology laboratories. Genotypic detection of the plasmid-mediated colistin resistance is also needed for infection control purposes. Materials and Methods Colistin susceptibility of 200 nonduplicate clinical CRE isolates from December 2017 to June 2019 was determined by BMD, agar dilution (AD), E test, and rapid polymyxin NP test and interpreted as per the European Committee on Antimicrobial Susceptibility Testing. The results of AD, E test, and NP test were compared with that of BMD, considering minimal inhibitory concentration (MIC) ≤ 2 µg/mL as susceptible and > 2 µg/mL as resistant. Presence of any plasmid-mediated colistin resistance (mcr-1 and 2) was evaluated in 27 colistin-resistant CRE isolates by polymerase chain reaction. Statistical Analysis Performance of different phenotypic methods was analyzed by comparing MIC results of AD and E test with that of reference BMD method. Agreement between BMD and the other two methods was expressed in terms of categorical agreement and essential agreement. Errors were expressed as very major error (VME: false-susceptible) and major error (ME: false-resistance) by AD/E test. VME and ME of 3% disagreement were considered unacceptable. Results Colistin resistance was found in 27 (13.5%) isolates by BMD method. The VME rates of both AD (11%) and E test (37%) could not meet the Clinical and Laboratory Standards Institute recommendation (< 3% VME rate is acceptable) as alternative tests to the reference BMD. Colistin NP test showed sensitivity and specificity of 85% and 98%, respectively. The percentage discordant result in NP test was highest in Enterobacter spp. (17%). None of the 27 colistin resistant isolates showed presence of mcr-1 and mcr-2 genes. Conclusion High VME rate in AD and E tests precludes their use as alternatives to BMD for colistin susceptibility testing. NP test with moderate sensitivity but excellent specificity can be a good alternative for testing colistin susceptibility in CRE isolates, except in Enterobacter spp. Absence of mcr-1 and mcr-2 gene necessitates the exploration of other mechanisms of colistin resistance.

Implementing EUCAST rapid antimicrobial susceptibility testing method for sepsis: lessons learned in a tertiary care center

Introduction: We prospectively evaluated EUCAST rapid antimicrobial susceptibility testing methodology for susceptibility testing directly from blood culture bottles in comparison to CLSI disk-diffusion method. Methodology: During May-November 2019, positively flagged blood culture bottles showing Gram-negative micro-organisms were simultaneously processed by rapid antimicrobial susceptibility testing and CLSI methodology. Antibiotics tested were cefotaxime, ceftazidime, piperacillin-tazobactam, imipenem, meropenem, gentamicin, tobramycin and trimethoprim-sulphamethoxazole. Results: Overall, 80 isolates identified as Escherichia coli (n = 24, 30%), Klebsiella pneumoniae (n = 15, 18.7%), Pseudomonas aeruginosa (n = 16, 20%) and Acinetobacter baumannii (n = 25, 31.2%) were included. Categorical agreements of rapid antimicrobial susceptibility testing at 4-, 6- and 8-hour reading times were 88.1% (304/345), 90.8% (425/468) and 92.3% (467/506), respectively. Major Error rates were 14% (21/150), 4.9% (10/206) and 4/236 (1.7%), whereas Very Major Error rates were 1.1% (2/177), 1.3% (3/232) and 3.3% (8/243), respectively. Results categorized as “Area of Technical Uncertainty” were significantly lower at 8-hour {10.2% (39/384) vs 5.2% (28/534), 4- vs 8-hour, p = 0.003, Fischer’s exact test}. Conclusions: Except for a slightly higher Very major error rate, rapid antimicrobial susceptibility testing at 8-hour is equivalent to Disk-diffusion method (CLSI-M100) using CLSI-M52 criteria for equivalence: (Categorical agreement ≥ 90%, Very major error ≤ 1.5% and Major error ≤ 3%). Poor Categorical agreements at all reading times were noted for piperacillin-tazobactam, ciprofloxacin and E. coli. Performance of rapid antimicrobial susceptibility testing methodology in resource limited settings brings unique challenge of identifying micro-organisms within 8 hours. We suggest reading and reporting of results at a single time point using rapid antimicrobial susceptibility testing method i.e. at 8-hour.