Journal of Antimicrobial Chemotherapy最新文献

Background: Development of clinical decision support systems (CDSS) has been ongoing for over 60 years, more recently leveraging technologies such as artificial intelligence (AI) and machine learning (ML). Intelligent CDSS addressing different stages of the infection management process offer potential advantages in interpreting complex data and guiding clinical decision-making.

Objectives: We outline the current applications of AI-driven CDSS across the continuum of bacterial infection management, from prevention and diagnosis to antibiotic prescribing and treatment individualization. We discuss the main limitations hindering their translation into clinical practice, as well as opportunities to improve their development to better meet clinical needs.

Methods: References for this review were identified through searches of PubMed, Google Scholar, bioRxiv and arXiv up to March 2025 by use of a combination of ML, decision-making and bacterial infection keywords.

Key findings: AI-CDSS studies increasingly leverage multimodal electronic health record (EHR) data, with most adopting lower-complexity models that perform well on structured data, particularly when supported by effective feature engineering. Despite efforts to develop accurate AI-driven systems, some of which achieve clinician-level accuracy in solving diagnostic and prescribing tasks, AI-CDSS have largely failed to integrate into clinical settings. Their adoption faces challenges related to the narrow scope of the defined medical task, failure to consider stakeholder workflow and lack of proper evaluation frameworks.

Conclusion: There is a need to shift CDSS development towards a more adaptive and holistic approach that recognizes the continuous nature of the decision-making process in infection management. Comprehensive AI-powered platforms that can model infection dynamics could improve antibiotic stewardship and help tackle the global health emergency of antimicrobial resistance.

Background: The knowledge about benznidazole pharmacokinetics (PK) in patients with chronic Chagas disease (CD) is limited.

Objective: To evaluate the benznidazole PK parameters in patients with chronic CD, stratified by age, sex, and adverse drug reactions (ADRs) occurrence.

Methods: Single-centre, open-label, clinical trial that included adult patients with chronic CD who received benznidazole 300 mg/day for 60-80 days. Blood samples were collected on Days 1, 7, 15, 30, and 60 of treatment. Benznidazole was quantified by HPLC-MS/MS and PK analysis used non-compartmental analysis.

Results: Twenty-nine participants (16 females; 55.2%) were included. Five (17.2%) participants interrupted the treatment definitely due to ADRs. PK parameters after a single benznidazole dose were as follows: Cmax = 2.48(0.53) µg/mL, AUC0-6 = 10.80(2.59) h*µg/mL, and Tmax = 2.7(1.3) h. The steady-state PK parameters on Day 15 were as follows: Cmax,ss = 7.86(2.06) µg/mL, AUC0-6,ss = 42.05(10.87) h*µg/mL, and Tmax,ss = 2.3 (1.2 h). Tmax was longer among those ≥60 years old (P = 0.045), and Cmax (P = 0.0004) and AUC0-6 (P = 0.003) were higher in females, but PK parameters normalized by weight were similar between sexes. The most frequent ADRs were skin reactions (44.8%), gastrointestinal (37.9%), haematological (20.7%), and neurological (27.6%). Sex was associated with gastrointestinal ADRs, while weight was associated with skin reactions. Higher benznidazole plasma levels on Days 1, 7, and 15 of treatment were associated with skin reactions even adjusting for age and sex.

Conclusions: Benznidazole PK presented little variation according to sex and age, but differences in sex appeared linked to females lower weight. Higher benznidazole plasma levels were associated with skin reactions, indicating a potential dose-dependent relationship of this ADR.

Trial registration: This study was registered on the Brazilian Clinical Trials Database-REBEC (RBR-5vg8p36). Registered on 11 August 2021. https://ensaiosclinicos.gov.br/rg/RBR-5vg8p36.

Background and objective: Benign intracranial hypertension, i.e. increased intracranial pressure, is a well-known rare adverse effect of tetracyclines, but its association with glycylcyclines (tetracycline-like antibiotics) was not confirmed. To investigate whether there are signals in an adverse events database for increased intracranial pressure after using tigecycline and eravacycline, and to compare it with probably existing signals for tetracyclines.

Methods: This secondary research was based on the FDA Adverse Event Reporting System (FAERS) Database. Disproportionality analysis was conducted and the following disproportionality indices were calculated: reporting odds ratios (ROR), Yule's Q, proportional reporting ratio (PRR) and the information component (IC) for 'intracranial pressure increased' and 'pseudotumor cerebri' adverse events.

Results: Significant signals for adverse event 'intracranial pressure increased' were found for doxycycline (ROR ± 95%CI = 27.19 ± 4.95), tetracycline (ROR ± 95%CI = 79.29 ± 24.64) and minocycline (ROR ± 95%CI = 59.53 ± 16.39). Signals for omadacycline and sarecycline were not found, primarily due to the small number of reports. In the database, there were no reports of increased intracranial pressure, pseudotumor cerebri or intracranial hypertension for both tigecycline and eravacycline.

Conclusions: Although structurally related to tetracyclines, glycylcyclines are not associated with increased intracranial pressure based on current pharmacovigilance data. This may warrant re-evaluation of SmPC warnings and supports safer use in patients with suspected intracranial hypertension when no better alternatives are available.

Background: Vaccines have an important role in reducing antibiotic prescribing. Population-based estimates of antibiotic prescribing for acute viral respiratory illness among high-risk populations are needed to inform vaccine policy-making. Our objective was to derive population-based estimates of seasonal outpatient antibiotic prescriptions among high-risk adults with respiratory syncytial virus (RSV), influenza, or human metapneumovirus (hMPV) illness.

Methods: We included adults with laboratory-confirmed medically attended RSV, influenza, or hMPV illness and ≥1 high-risk condition enrolled in the US Flu VE Network in central and northern Wisconsin, 2015-16 to 2019-20. We extracted data on outpatient antibiotics prescribed ≤7 days of enrolment. We estimated the seasonal number of antibiotic prescriptions associated with each viral illness per 10 000 high-risk adults using Poisson regression and stratification and weighting. We incorporated adjustment factors to account for incomplete overlap between influenza and RSV or hMPV seasons using state respiratory infection surveillance data.

Results: 3601 respiratory specimens from high-risk adults were included. 303 (8.4%), 1011 (28%), and 289 (8.0%) tested positive for RSV, influenza, or hMPV, respectively. 106/303 (35%), 174/1011 (17%), and 107/289 (37%) patients received ≥1 antibiotic prescription, respectively. After stratification and weighting, seasonal population-based estimates for the number of antibiotic prescriptions where 29 (95% confidence interval 22-39), 52 (41-65), and 31 (24-40) per 10 000 high-risk adults, respectively.

Conclusion: Our results suggest that RSV, influenza, and hMPV illness contribute substantially to antibiotic prescribing among high-risk adults in rural Wisconsin. Understanding the impact of vaccines on antibiotic use and resistance is imperative for assessing their full public health value.

Background: Infective endocarditis (IE) is a severe infection requiring efficient, sometimes combined, antibiotic therapy. For β-lactam-susceptible Enteroccus spp., particularly Enterococcus faecalis (EFS), the combination of amoxicillin and ceftriaxone is currently recommended. However, limited data are available for ampicillin-susceptible E. faecium strains (EFM-S).

Objectives: To characterize the in vitro synergy of amoxicillin/ceftriaxone combination against clinical isolates of EFM-S and evaluate its clinical relevance.

Methods: Twenty-six EFM-S and 10 EFS clinical isolates from the Henri Mondor Hospital (Créteil, France) were tested. EFM-S clades were previously determined by whole-genome sequencing. Checkerboard assays were performed to assess the MICs and synergy across a range of amoxicillin and ceftriaxone concentrations. The fractional inhibitory concentration (FIC) index was computed for each combination, with synergy defined as ΣFIC ≤ 0.5.

Results: Amoxicillin and ceftriaxone MIC distributions were comparable between EFM-S (amoxicillin, 0.06-2 mg/L; ceftriaxone, 1 to ≥1024 mg/L) and EFS (amoxicillin, 0.5-2 mg/L; ceftriaxone, 64 to ≥1024 mg/L). Synergy was observed in 21/26 EFM-S and all EFS strains. The five non-synergistic EFM-S strains had low ceftriaxone MICs (1-16 mg/L), probably limiting synergy detection. No difference was observed according to EFM clades. The minimal synergistic concentrations were consistently higher for EFM-S (median amoxicillin, 0.125 mg/L; ceftriaxone, 32 mg/L) compared to EFS (amoxicillin, 0.03 mg/L; ceftriaxone, 2 mg/L).

Conclusions: Despite in vitro synergy, the amoxicillin/ceftriaxone combination is clinically irrelevant for EFM-S IE due to unachievable drug concentrations in vivo. These findings support clearer guidelines explicitly excluding this regimen for E. faecium, regardless of β-lactam susceptibility.

Clofazimine is a WHO-recommended therapy for multibacillary leprosy and an emerging option for drug-resistant tuberculosis and non-tuberculous mycobacterial infections. Yet in high-income countries like the USA, access remains restricted through expanded access pathways and other limitations, despite decades of safe use. This review examines clofazimine's regulatory trajectory in the USA, contrasts global access models, and highlights the consequences of relying on a single manufacturer and informal donation programmes. We argue that existing systems can be ill-suited to ensure equitable, sustainable access to essential but low-profit medicines like clofazimine. Policy options include simplified access pathways, WHO-based procurement for leprosy, and public manufacturing to stabilize supply. Clofazimine's case illustrates the need for proactive stewardship and access models that align with public health goals.

Background and objectives: The MIC, determined in vitro using a fixed tazobactam concentration, is used as basis for the pharmacokinetic/pharmacodynamic (PK/PD) index of piperacillin/tazobactam. The β-lactamase inhibitor (BLI) concentration-dependent MIC (MICcBLI) represents a promising alternative PD metric accounting for varying tazobactam concentrations observed in vivo. The aim was to investigate how the effect of varying in vivo tazobactam concentrations can be incorporated through the in vitro MICcBLI as a PD metric to assess the probability of three piperacillin/tazobactam dosing regimens to achieve predefined target values.

Methods: The MICcBLI for six isolates was evaluated as function of tazobactam concentrations. Two nonlinear mixed-effects PK models for piperacillin and tazobactam were developed based on data from 44 critically ill patients. Probability of target attainment (PTA) analysis was performed, and PTA values for the PK/PD targets 100% fT > MIC and 100% fT > MICcBLI were assessed.

Results: The susceptibility against piperacillin was dependent on the tazobactam concentration for all isolates. A two-compartment PK model with inter-occasion variability on clearance (CL) and inter-individual variability on CL and both volumes, with creatinine clearance as a covariate on CL, described the PK of piperacillin and tazobactam best. The PTA analysis revealed differences between the isolates, three infusion types and the two metrics.

Conclusions: By adopting this framework, dosing optimization can be approached to provide support for the rational design of piperacillin/tazobactam dosing regimens and other β-lactam/β-lactamase combinations. The shift from the conventional MIC to a β-lactamase inhibitor concentration-dependent metric clearly improved the knowledge of the susceptibility of a pathogen against piperacillin/tazobactam.

Objectives: In our hospital, patients are treated with continuous vancomycin infusion following standard dosing guidelines. Recent literature suggests that obese and neutropenic patients may require increased vancomycin dosages. This study aimed to reassess our dosing guidelines by evaluating how the time and dose required to achieve vancomycin target concentrations differ across specific patient groups compared to all vancomycin patients.

Methods: A retrospective cohort study was conducted using hospital records from patients treated with continuous vancomycin infusion between 1 January and 31 December 2023. Adult patients treated according to the antibiotic guidelines, with at least one vancomycin target concentration measurement, were included. We compared the time and maintenance dose required to reach target concentrations in neutropenic patients (neutrophils <1.5 × 109/L), patients weighing >100 kg, patients with augmented renal clearance (ARC; glomerular filtration rate >130 mL/min/1.73 m2) and all vancomycin patients.

Results: The median time to reach target was 2.59 days for all vancomycin patients (n = 244), and 2.73, 2.74, and 3.39 days for ARC patients (n = 5), neutropenic patients (n = 15) and patients weighing >100 kg (n = 50), respectively. The median maintenance dose was 2000 mg/day among all vancomycin patients, compared to 3000 mg/day for patients weighing >100 kg and neutropenic patients and 4000 mg/day for ARC patients.

Conclusions: According to current guidelines, there is no significant difference in the time required to achieve vancomycin target concentrations between neutropenic patients, ARC patients and all vancomycin patients. However, patients weighing >100 kg took nearly a day longer to reach target concentrations (P < 0.001) and required higher maintenance doses, as did neutropenic and ARC patients.