Journal of Hospital Infection最新文献

Background: The relationship between colorectal and hysterectomy surgical site infections (SSI) and air pressure variability inside operating rooms (OR) during procedures is understudied.

Methods: Our retrospective cohort study featured patients who underwent procedures assigned National Healthcare Safety Network (NHSN) colorectal or abdominal hysterectomy codes. The primary exposure was percentage of procedure time out of pressure range. NHSN criteria were used to identify SSIs. We used multivariable binary logistic regression modelling to analyze the association between the exposure and SSIs. Sensitivity models of more homogeneous patients (e.g., only colorectal patients, only patients with at least one instance of pressure < 1 Pascal) were also conducted.

Results: Of 1,663 patients, 933 (56%) underwent colorectal surgery and 730 (44%) had hysterectomies. After multivariable adjustment, no association between OR pressure deviations and SSIs was identified (odds-ratio for 1% increase of procedure time pressure out of range [95% confidence interval]: 1.05 [0.46, 2.24]); this null association remained for the aforementioned sensitivity models.

Conclusions: No significant association observed between OR air pressure variability and development of colorectal and hysterectomy SSIs.

Aseptic technique is integral to infection prevention and control but there is no agreement concerning its aims, how it is undertaken outside operating rooms, related terminology and lack of concordance in international guidelines. We undertook a critical review to explore breaches of asepsis that could occur during aseptic procedures outside of the operating room environment. There was little direct research evidence, leading to low levels of certainty in outcomes. We established that low levels of contamination are exhibited by some materials not marketed as sterile. Contents of sterile procedure packs rapidly became contaminated after opening. Large, unwieldy procedure packs and method of opening result in contamination. Surface contamination occurs as soon as colonised and infected wounds are exposed and is persistent. Few studies compared different approaches to conducting aseptic technique. Non-sterile disposable gloves might reduce risk of microbial acquisition and transmission but there are no data on how quickly sterile gloves become contaminated. In conclusion: there is lack of evidence to underpin practice. Many non-sterile products are functionally sterile; conversely sterile products become contaminated quickly. Relatively simple methods such as use of non-sterile disposable gloves and non-touch technique could significantly reduce risk of microbial transmission. Further research is needed to corroborate these findings.

Background: Surgical site infections (SSI) is a major healthcare-associated complication, yet early detection remains challenging.

Objective: To develop and validate a machine learning-based predictive model for the detection of SSI in Chinese surgical patients.

Methods: A multicenter cohort study was conducted at two tertiary hospitals in China. Data from 118,314 patients who underwent surgery between June 2023 and December 2024 were used for model development and validation. Clinical, microbiological, and demographic variables were considered as predictors. Multiple machine learning algorithms were applied, with hyperparameter tuning and cross-validation. Model performance was evaluated using area under the curve (AUC), sensitivity, specificity, and accuracy.

Results: The decision tree model demonstrated robust predictive performance, achieving an AUC of 0.92 in the development cohort and 0.90 in the validation cohort. Sensitivity and specificity values confirmed its effectiveness in identifying patients at elevated risk of SSIs. Performance remained stable across subgroups.

Conclusions: This study established and externally validated a machine learning-based model for SSI prediction in surgical patients. The model showed strong and consistent performance and may support clinical decision-making by enabling real-time, automated risk assessment.

Objectives: Carbapenem-resistant P. aeruginosa (CRPA) causes challenging healthcare-associated infections, many of which arise from superficial sites. Here, we aim to develop an effective phage cocktail targeting CRPA and to evaluate its potential as a topical hydrogel to prevent wound infections using an ex vivo porcine skin model.

Methods: Phage EKU1, MHN1, and Bobae, isolated from wastewater were characterized. The cocktails of these phages were formulated against a CRPA strain. Phage-resistant isolates emerged after exposure to the cocktail were also analysed for fitness costs. Hydroxyethyl cellulose (HEC)-phage cocktail hydrogel was tested against a CRPA on ex vivo porcine skins. Shelf life of the gel was observed under storage in different temperatures RESULTS: EKU1 (Yuavirus), MHN1 (Bruynoghevirus), and Bobae (Chimallivirus) belonged to distinct evolutionary lineages. Their cocktail exhibited synergistic activity against CRPA over 5-log (CFU/mL) reduction after 24 hours at an MOI of 1000. Among 12 phage-resistant isolates, 11 exhibited slower growth rates than wild type, and 5 out of 12 isolates produced significantly less biofilm. On porcine skins, phage hydrogel significantly reduced bacterial load by 1.5 log CFU/mL after 12 hours compared to control. Phage titres in the gel stored at 4°C decreased by 1-1.5 log PFU/mL over 8 weeks.

Conclusions: The cocktail from distinct phages shows strong synergy against CRPA with resistance-associated trade-offs. Incorporated into HEC hydrogel, the cocktail significantly suppressed CRPA growth on porcine skins and remained stable under refrigeration for months, supporting its potential as a promising topical prototype for wound infections.

Objective: This study aimed to identify high-risk items (HRIs) in the fomite route transmission of hospital associated infections (HAIs) in an intensive care unit (ICU) environment and assess their bacterial contamination levels.

Method: A 15-day structured hand-to-surface contact observation was conducted in a 13-bed ICU at a tertiary hospital in Hangzhou, China. Directed weighted networks were constructed from contact data to identify HRIs based on centrality of the networks. Aerobic bacteria was sampled from environmental surfaces including high-touch, low-touch, high-risk and low-risk surfaces at 9:00, 13:00, 17:00, and 20:00 over four separate days.

Results: While there was some overlap between high-touch items (HTIs) and HRIs, HTIs were not consistently identified as HRIs. The bacterial contamination level on HRIs was found to be significantly higher than on low-risk items, but no significant difference was observed between HTIs and low-touch items on the bacterial contamination levels.

Conclusion: The identification of HRIs through network analysis offers valuable insights for guiding more precise surface cleaning and disinfection strategies in the ICU environments. Although HTIs are commonly considered important, they are not always more contaminated.

At a cancer center, among 304 hospital-onset COVID-19 cases from 2021-24, 194 patients shared a room with 299 other patients during their infectious period. Secondary infection occurred in 17% (51/299) of roommates; 50% were asymptomatic; 12% developed severe infection. Attack rates did not differ in the pre- and post-omicron period.