Surgical site infection (SSI) following orthopedic surgery is a complication associated with morbidity and economic burden. Transmission of airborne bacteria that settle into surgical wounds constitutes a risk factor for SSIs. However, monitoring microbial contamination inside operating rooms with conventional methods is resource and time-consuming.

This study aimed to assess correlation between a biofluorescent particle counter (BFPC) and conventional air sampling, to enable real-time monitoring of airborne contamination. Additionally, the study aimed to analyze correlation between particles near the surgical site and particles 1 meter away, to evaluate the feasibility of distance-based measurements.

Correlation analysis was conducted to compare colony-forming units (CFU) collected using a Sartorius MD8 air sampler with biofluorescent viable particles detected by BioTrak 9510-BD, both positioned near the surgical site. Additionally, correlation between particle counts measured by AeroTrak 6510, positioned 1 meter away, and total particle counts measured by the BioTrak near the surgical site was evaluated. Sampling took place in two operating rooms: one with turbulent mixed airflow (TMA) and one with unidirectional airflow (UDAF).

Negligible to low correlation between biofluorescent particles and CFU was observed, both in UDAF (n = 100) and TMA (n = 22). However, strong correlation was found between BFPC and particle counter measurements of total numbers of particles (Rp = 0.634–0.769, P < .001).

While BFPCs offer real-time monitoring of airborne contamination, their predictive ability for CFU levels remains uncertain. Yet, the strong correlation between particles in the surgical site and particles measured 1 meter away suggests feasibility to conduct future studies with larger cohorts.

Artificial intelligence (AI) has the potential to enhance clinical decision-making, including in infectious diseases. By improving antimicrobial resistance prediction and optimizing antibiotic prescriptions, these technologies may support treatment strategies and address critical gaps in healthcare. This study evaluates the effectiveness of AI in guiding appropriate antibiotic prescriptions for infectious diseases through a systematic literature review.

We conducted a systematic review of studies evaluating AI (machine learning or large language models) used for guidance on prescribing appropriate antibiotics in infectious disease cases. Searches were performed in PubMed, CINAHL, Embase, Scopus, Web of Science, and Google Scholar for articles published up to October 25, 2024. Inclusion criteria focused on studies assessing the performance of AI in clinical practice, with outcomes related to antimicrobial management and decision-making.

Seventeen studies used machine learning as part of clinical decision support systems (CDSS). They improved prediction of antimicrobial resistance and optimized antimicrobial use. Six studies focused on large language models to guide antimicrobial therapy; they had higher prescribing error rates, patient safety risks, and needed precise prompts to ensure accurate responses.

AI, particularly machine learning integrated into CDSS, holds promise in enhancing clinical decision-making and improving antimicrobial management. However, large language models currently lack the reliability required for complex clinical applications. The indispensable role of infectious disease specialists remains critical for ensuring accurate, personalized, and safe treatment strategies. Rigorous validation and regular updates are essential before the successful integration of AI into clinical practice.

Antipseudomonal antibiotics are commonly prescribed for diabetic foot infections (DFI) at our institution despite a low local prevalence of Pseudomonas aeruginosa. A multidisciplinary team implemented a DFI electronic health record (EHR)-embedded treatment algorithm and order set.

This multi-center, quasi-experimental study evaluated adults on antibiotics admitted for DFI to vascular surgery or medical units pre- and post-implementation of an EHR-embedded treatment algorithm and order set. Exclusion criteria: duplicate patients, concomitant infection, transfer from an outside hospital. Primary endpoint: antipseudomonal use among included patients (DOT/1000 DFI days present). Secondary outcomes: empiric antipseudomonal use, length of stay, 30-day readmission, mortality, amputation, and Clostridioides difficile infection. Patient demographics, diagnostics, treatments, and outcomes were evaluated.

Two hundred ten patients were included with 70 patients included in each group. The post-algorithm group had lower antipseudomonal DOT/1000 DFI days present compared to the pre-intervention group (360 vs 503, P < 0.001). The post-order set group had the lowest antipseudomonal use (347 vs 503, P < 0.001). Empiric antipseudomonal use decreased from 85.7% pre-intervention to 72% post-algorithm and 68.5% post-order set. Collectively, 30-day mortality was < 5%. Amputation during and within 30 days of hospitalization was similar in the pre-intervention (48.6%), post-algorithm (30%), and post-order set (41.4%) groups. Methicillin-susceptible Staphylococcus aureus and Streptococcus spp. were most frequently isolated. Wound cultures were not collected in 24.3%, 22.9%, and 40% of the pre-intervention, post-algorithm, and post-order set group.

EHR-embedded clinical decision-making tools reduce antipseudomonal use for DFI treatment without increasing 30-day mortality or amputation.

In our study, we aim to compare the resistance profiles of Acinetobacter baumannii and Pseudomonas aeruginosa isolates from intensive care unit (ICU) patients before and during the COVID-19 pandemic.

The study involved adult patients monitored in the ICUs of a secondary-level hospital from January 2019 to December 2022. Isolates of A. baumannii and P. aeruginosa were obtained from blood, urine, and respiratory samples. Identification and antibiotic susceptibility tests were conducted using the disk diffusion method and the VITEK 2 system.

The average age of the patients was 61.3 ± 21.9 years (range: 18–95), with a majority of 1306 (51.6%) being male. During the pandemic, A. baumannii isolates showed a significant increase in resistance rates for several antibiotics compared to the pre-pandemic period: imipenem (96% vs 35.1%), amikacin (84.1% vs 14.4%), ciprofloxacin (96.9% vs 36.9%), trimethoprim-sulfamethoxazole (66.4% vs 27%), and ceftazidime (96.5% vs 33.3%) (all with P < .001). However, there was no significant change in colistin resistance rates in these isolates (0.9% vs 0%; P = .307). Similarly, Pseudomonas aeruginosa isolates exhibited significant increases in resistance rates during the pandemic compared to the pre-pandemic period: imipenem (51.5% vs 18.8%; P < .001), colistin (4.9% vs 0.6%; P = .009), amikacin (23.5% vs 4.4%; P < .001), ciprofloxacin (53.3% vs 13.8%; P < .001), and ceftazidime (39.2% vs 12.7%; P < .001).

Our results demonstrate a significant increase in antibiotic resistance levels in Acinetobacter and Pseudomonas strains associated with hospital-acquired infections or colonization during the COVID-19 pandemic.

Outbreaks of Pseudomonas aeruginosa infections have been linked to water-related sources. We describe the investigation of a suspected outbreak of five P. aeruginosa infections in pediatric oncology patients in 2021 that triggered a retrospective review and prospective monitoring of additional cases, environmental sampling, and bacterial genomic analysis.

Pediatric oncology center.

Medical records of patients with P. aeruginosa were reviewed and staff were interviewed to identify common exposures. Environmental samples were cultured for P. aeruginosa. Patient and environmental isolates underwent whole genome sequencing and core genome multi-locus sequence typing (cgMLST) and sequences were added to a previously existing library of P. aeruginosa clinical isolates collected in 2017 and onwards to determine strain relatedness.

During 2019–2022, 82 patients with 110 episodes of P. aeruginosa infections were identified and 132 isolates of P. aeruginosa were sequenced. Twenty-three environmental samples were collected, of which two grew P. aeruginosa in culture. CgMLST demonstrated four multi-patient isolate clusters but no genetic relatedness among the isolates from the patients in the suspected outbreak. Two sink-derived isolates from 2021 were genetically related to patient-derived isolates from 2018 and 2017.

Sequencing revealed there is no common source or linkage between the isolates of the suspected P. aeruginosa outbreak in 2021. However, it revealed genetic relatedness of previous patient strains to later strains collected from hospital sinks, suggesting persistent colonization of a reservoir with P. aeruginosa.

Despite many reports of similar effectiveness between oral and intravenous antibiotics for bone and joint infections, prescribing practice has been slow to change in the United States. We sought to determine if implementing an intravenous-to-oral treatment guideline could increase prescribing of oral antibiotic regimens at our center.

Retrospective, quasi-experimental study.

Single US academic medical center.

Patients with bone or joint infections managed by Infectious Disease providers from September 2020 to December 2022.

An intravenous-to-oral treatment guideline for patients with bone and joint infections.

The prescribing rates of fully oral antibiotic regimens before and after implementation of the guideline were compared. Additionally, variables independently associated with oral antibiotic prescribing were identified by logistic regression.

There were 450 patients included: 213 before and 237 after implementation of the guideline. Oral antibiotic prescribing significantly increased following implementation of the treatment guideline to 59% from 33% of patients (difference 25.8%, 95% CI [16.7%, 34.4%]. In multivariable analysis, the post-intervention phase was associated with a significantly greater likelihood of oral antibiotic prescribing (aOR 2.89 [1.90, 4.45]). Other variables independently associated with oral antibiotic prescribing included male sex (aOR 1.88 [1.20, 2.98]), prosthetic joint infection (aOR 0.29 [0.17, 0.47]), and infection with Enterobacterales (aOR 2.86 [1.45, 5.92]), methicillin-sensitive Staphylococcus aureus [aOR 0.41 [0.26, 0.65]), or coagulase-negative staphylococci (aOR 0.34 [0.18, 0.62]).

Implementation of a treatment guideline resulted in a significant increase in oral antibiotic prescribing. Antimicrobial stewardship programs should implement similar interventions to improve outpatient antibiotic utilization.

Assess the efficacy of staged interventions aimed to reduce inappropriate Clostridioides difficile testing and hospital-onset C. difficile infection (HO-CDI) rates.

Interrupted time series.

Community-based.

National Healthcare Safety Network (NHSN) C. difficile metrics from January 2019 to November 2022 were analyzed after three interventions at a community-based healthcare system. Interventions included: (1) an electronic medical record (EMR) based hard stop requiring confirming ≥3 loose or liquid stools over 24 h, (2) an infectious diseases (ID) review and approval of testing >3 days of hospital admission, and (3) an infection control practitioner (ICP) reviews combined with switching to a reverse two-tiered clinical testing algorithm.

After all interventions, the number of C. difficile tests per 1,000 patient-days (PD) and HO-CDI cases per 10,000 PD decreased from 20.53 to 6.92 and 9.80 to 0.20, respectively. The EMR hard stop resulted in a (28%) reduction in the CDI testing rate (adjusted incidence rate ratio ((aIRR): 0.72; 95% confidence interval [CI], 0.53 to 0.96)) and ID review resulted in a (42%) reduction in the CDI testing rate (aIRR: 0.58; 95% CI, 0.42–0.79). Changing to the reverse testing algorithm reduced reported HO-CDI rate by (95%) (cIRR: 0.05; 95% CI; 0.01–0.40).

Staged interventions aimed at improving diagnostic stewardship were effective in overall reducing CDI testing in a community healthcare system.

Outbreaks of Group A Streptococcal (GAS) infection are difficult to detect in community healthcare settings and present unique challenges for infection prevention and control (IPC). We describe investigation of a cluster of GAS among individuals receiving wound care from the same community integrated care team (CIT) and associated complexities.

Prospective and retrospective surveillance for cases of invasive and noninvasive GAS infection linked to the CIT was undertaken with the local NHS trust IPC team. Screening samples were requested from staff working in the CIT (n = 191) and from staff and residents (n = 73) in care home A where several cases resided. Clinical isolates were sent to the UKHSA reference laboratory for emm typing and whole genome sequencing (WGS).

Twenty-two cases were identified over a five-month period. Eighteen had isolates available for typing, 11 of which were emm type 108.1 and 0-2SNPs apart on WGS. Six were different emm types and one emm type 108.1 but 9-13SNPs apart from other isolates and so excluded from the investigation. No staff infected or colonized with emm 108.1 were identified, and no single healthcare worker had attended all cases. GAS was isolated in the room of a case resident in care home A and found to be closely genetically related to clinical isolates.

WGS was integral in identifying outbreak cases and a multiagency approach essential to the investigation. Unfortunately, despite this no clear source or route of transmission was identified. Further research is required to determine the most effective IPC strategies for community healthcare.

Assess the impact of a multifaceted discharge antimicrobial stewardship initiative by comparing proportion of appropriate antimicrobial regimens before and after implementation.

Cohort study.

Non-teaching, urban, community medical center.

Adult patients prescribed an oral antimicrobial regimen at discharge were included. Patients were randomized irrespective of encounter type or discharge disposition. Pregnant and post-partum patients were excluded.

A discharge antimicrobial stewardship program was implemented at our facility. Components of the initiative included development of a comprehensive, institution-specific, inpatient and outpatient prescribing guideline, extensive face-to-face clinician education, and real-time, pharmacist prospective audit and feedback at discharge. The validated National Antimicrobial Prescribing Survey tool was used to then categorize one hundred randomized discharge antimicrobial prescriptions as appropriate (optimal or adequate), inappropriate (suboptimal or inadequate), or not assessable. Hospital-specific treatment guidelines, literature references, and patient-specific factors were used to determine appropriateness.

One hundred antimicrobial regimens selected via random sampling were analyzed in each cohort. The proportion of appropriate antimicrobial regimens increased by 15% after program implementation (47% vs 62%, P = .03).

Study results highlight the positive impact of a multidisciplinary, multipronged approach in improving discharge antimicrobial prescribing.

Hospital employees are at risk of severe acute respiratory coronavirus 2 (SARS-CoV-2) infection from patient, coworker, and community interactions. Understanding employees’ perspectives on transmission risks may inform hospital pandemic management strategies.

Qualitative interviews were conducted with 23 employees to assess factors contributing to perceived transmission risks during patient, coworker, and community interactions and to elicit recommendations. Using a deductive approach, transcripts were coded to identify recurring themes.

Tertiary hospital in Boston, Massachusetts.

Employees with a positive SARS-CoV-2 PCR test between March 2020 and January 2021, a period before widespread vaccine availability.

Employees generally reported low concern about transmission risks during patient care. Most patient-related risks, including limited inpatient testing and personal protective equipment availability, were only reported during the early weeks of the pandemic, except for suboptimal masking adherence by patients. Participants reported greater perceived transmission risks from coworkers, due to limited breakroom space, suboptimal coworker masking, and perceptions of inadequate contact tracing. Perceived community risks were related to social gatherings and to household members who also had high SARS-CoV-2 infection risk because they were essential workers. Recommendations included increasing well-ventilated workspaces and breakrooms, increasing support for sick employees, and stronger hospital communication about risks from non-patient-care activities, including the importance of masking adherence with coworkers and in the community.

To reduce transmission during future pandemics, hospitals may consider improving communication on risk reduction during coworker and community interactions. Societal investments are needed to improve hospital infrastructure (eg, better ventilation and breakroom space) and increase support for sick employees.

To compare rates of clinical response in children with Clostridioides difficile infection (CDI) treated with metronidazole vs vancomycin.

Retrospective cohort study was performed as a secondary analysis of a previously established prospective cohort of hospitalized children with CDI. For 187 participants 2–17 years of age who were treated with metronidazole and/or vancomycin, the primary outcome of clinical response (defined as resolution of diarrhea within 5 days of treatment initiation) was identified retrospectively. Baseline variables associated with the primary outcome were included in a logistic regression propensity score model estimating the likelihood of receiving metronidazole vs vancomycin. Logistic regression using inverse probability of treatment weighting (IPTW) was used to estimate the effect of treatment on clinical response.

One hundred seven subjects received metronidazole and 80 subjects received vancomycin as primary treatment. There was no univariable association between treatment group and clinical response; 78.30% (N = 83) of the metronidazole treatment group and 78.75% (N = 63) of the vancomycin group achieved clinical response (P = 0.941). After adjustment using propensity scores with IPTW, the odds of a clinical response for participants who received metronidazole was 0.554 (95% CI: 0.272, 1.131) times the odds of those who received vancomycin (P = 0.105).

In this observational cohort study of pediatric inpatients with CDI, the rate of resolution of diarrhea after 5 days of treatment did not differ among children who received metronidazole vs vancomycin.