Prevention of Healthcare-associated Infections in Intensive Care Unit Patients

Development and validation of an artificial intelligence mobile application for predicting 30-day mortality in critically ill patients with orthopaedic trauma

BACKGROUND

Given the intricate and grave nature of trauma-related injuries in ICU settings, it is imperative to develop and deploy reliable predictive tools that can aid in the early identification of high-risk patients who are at risk of early death. The objective of this study is to create and validate an artificial intelligence (AI) model that can accurately predict early mortality among critical fracture patients.

METHODS

A total of 2662 critically ill patients with orthopaedic trauma were included from the MIMIC III database. Early mortality was defined as death within 30 days in this study. The patients were randomly divided into a model training cohort and a model validation cohort. Various algorithms, including logistic regression (LR), extreme gradient boosting machine (eXGBM), decision tree (DT), support vector machine (SVM), random forest (RF), and neural network (NN), were employed. Evaluation metrics, including discrimination and calibration, were used to develop a comprehensive scoring system ranging from 0 to 60, with higher scores indicating better prediction performance. Furthermore, external validation was carried out using 131 patients. The optimal model was deployed as an internet-based AI tool.

RESULTS

Among all models, the eXGBM demonstrated the highest area under the curve (AUC) value (0.974, 95%CI: 0.959-0.983), followed by the RF model (0.951, 95%CI: 0.935-0.967) and the NN model (0.922, 95%CI: 0.905-0.941). Additionally, the eXGBM model outperformed other models in terms of accuracy (0.915), precision (0.906), recall (0.926), F1 score (0.916), Brier score (0.062), log loss (0.210), and discrimination slope (0.767). Based on the scoring system, the eXGBM model achieved the highest score (53), followed by RF (42) and NN (39). The LR, DT, and SVM models obtained scores of 28, 18, and 32, respectively. Decision curve analysis further confirmed the superior clinical net benefits of the eXGBM model. External validation of the model achieved an AUC value of 0.913 (95%CI: 0.878-0.948). Consequently, the model was deployed on the Internet at https://30-daymortalityincriticallyillpatients-fnfsynbpbp6rgineaspuim.streamlit.app/, allowing users to input patient features and obtain predicted risks of early mortality among critical fracture patients. Furthermore, the AI model successfully stratified patients into low or high risk of early mortality based on a predefined threshold and provided recommendations for appropriate therapeutic interventions.

CONCLUSION

This study successfully develops and validates an AI model, with the eXGBM algorithm demonstrating the highest predictive performance for early mortality in critical fracture patients. By deploying the model as a web-based AI application, healthcare professionals can easily access the tool, enabling them to predict 30-day mortality and aiding in the identification and management of high-risk patients among those critically ill with orthopedic trauma.

The survival of multi-drug resistant bacteria on raw Douglas fir material

In today's age of ecological transition, the use of materials such as renewable wood in construction is particularly relevant, but also a challenge in the healthcare sector where the hygiene dimension also comes into play. In this study we have investigated the survival of multi-resistant bacteria commonly responsible for healthcare-associated infections (HAIs) (ESBL-positive Klebsiella pneumoniae and glycopeptide-resistant Enterococcus faecalis) on two different types of wood (Douglas fir : Pseudotsuga menziesii and Maritime Pine : Pinus pinaster) compared to other materials (smooth: stainless steel and rough: pumice stone) and the effect of a disinfection protocol on the bacterial survival on Pseudotsuga menziesii. Approximately 108 bacteria were inoculated on each material and bacterial survival was observed over several days (D0, D1, D2, D3, D6, D7 and D15). Each analysis was performed in triplicate for each time and material. The results show an important reduction of the bacterial inoculum for Klebsiella pneumoniae and Enterococcus faecalis on Douglas fir, in contrast with the results obtained on maritime pine, stainless steel and pumice stone. No bacterial survival was detected on Douglas fir after application of a hospital disinfection protocol. These different results show that wood may have a place in the future of healthcare construction. Further studies would be interesting to better understand the different properties of wood.

Predictors of bacteremia and death, including immune status, in a large single-center cohort of unvaccinated ICU patients with COVID-19 pneumonia

BACKGROUND

We investigated the possible role of the immune profile at ICU admission, among other well characterized clinical and laboratory predictors of unfavorable outcome in COVID-19 patients assisted in ICU.

METHODS

Retrospective analysis of clinical and laboratory data collected for all consecutive patients admitted to the ICUs of the General Hospital of Pescara (Abruzzo, Italy), between 1^st March 2020 and 30^th April 2021, with a confirmed diagnosis of COVID-19 respiratory failure. Logistic regressions were used to identify independent predictors of bacteremia and mortality.

RESULTS

Out of 431 patients included in the study, bacteremia was present in N = 191 (44.3%) and death occurred in N = 210 (48.7%). After multivariate analysis, increased risk of bacteremia was found for viral reactivation (OR = 3.28; 95% CI:1.83-6.08), pronation (3.36; 2.12-5.37) and orotracheal intubation (2.51; 1.58-4.02). Increased mortality was found for bacteremia (2.05; 1.31-3.22), viral reactivation (2.29; 1.29-4.19) and lymphocytes < 0.6 × 10³c/µL (2.32; 1.49-3.64).

CONCLUSIONS

We found that viral reactivation, mostly due to Herpesviridae, was associated with increased risk of both bacteremia and mortality. In addition, pronation and intubation are strong predictors of bacteremia, which in turn together with severe lymphocytopenia due to SARS-CoV2 was associated with increased mortality. Most episodes of bacteremia, even due to Acinetobacter spp, were not predicted by microbiological evidence of colonization.

Trending prevalence of healthcare-associated infections in a tertiary hospital in China during the COVID-19 pandemic

BACKGROUND

The purpose of this study was to demonstrate both the four-year prevalence trend of healthcare-associated infections (HAIs) in a large tertiary hospital and the trend regarding the prevalence of HAIs following the outbreak of coronavirus disease 2019 (COVID-19) in order to provide evidence of hospital infection management during the COVID-19 pandemic.

METHODS

Based on the hospital's electronic nosocomial infection databases related to HAIs, we retrospectively identified the HAI cases to assess the epidemiological characteristics of HAIs from January 1, 2018, to December 31, 2021, in a large tertiary hospital in China. Similarly, the trends of HAIs after the COVID-19 outbreak and the seasonal variation of HAIs were further analyzed.

RESULTS

The HAI cases (n = 7833) were identified from the inpatients (n = 483,258) during the 4 years. The most frequently occurring underlying cause of HAIs was respiratory tract infections (44.47%), followed by bloodstream infections (11.59%), and urinary tract infections (8.69%). The annual prevalence of HAIs decreased from 2.39% in 2018 to 1.41% in 2021 (P = 0.032), with the overall prevalence of HAIs significantly decreasing since the outbreak of COVID-19 (2.20% in 2018-2019 vs. 1.44% in 2020-2021, P < 0.001). The prevalence of respiratory tract infections decreased most significantly; whereas, overall, the prevalence of HAIs was significantly greater during the winter compared with the rest of the year.

CONCLUSIONS

Not only did the annual prevalence of HAIs decrease from 2018 to 2021, but it also significantly decreased since the start of the COVID-19 pandemic, particularly respiratory tract infections. These results provide evidence for the need to prevent HAIs, especially during the winter season.

Effectiveness of Infection Control Teams in Reducing Healthcare-Associated Infections: A Systematic Review and Meta-Analysis

The infection control team (ICT) ensures the implementation of infection control guidelines in healthcare facilities. This systematic review aims to evaluate the effectiveness of ICT, with or without an infection control link nurse (ICLN) system, in reducing healthcare-associated infections (HCAIs). We searched four databases to identify randomised controlled trials (RCTs) in inpatient, outpatient and long-term care facilities. We judged the quality of the studies, conducted meta-analyses whenever interventions and outcome measures were comparable in at least two studies, and assessed the certainty of evidence. Nine RCTs were included; all were rated as being low quality. Overall, ICT, with or without an ICLN system, did not reduce the incidence rate of HCAIs [risk ratio (RR) = 0.65, 95% confidence interval (CI): 0.45-1.07], death due to HCAIs (RR = 0.32, 95% CI: 0.04-2.69) and length of hospital stay (42 days vs. 45 days, p = 0.52). However, ICT with an ICLN system improved nurses' compliance with infection control practices (RR = 1.17, 95% CI: 1.00-1.38). Due to the high level of bias, inconsistency and imprecision, these findings should be considered with caution. High-quality studies using similar outcome measures are needed to demonstrate the effectiveness and cost-effectiveness of ICT.

Tissue Penetration of Antimicrobials in Intensive Care Unit Patients: A Systematic Review—Part I

The challenging severity of some infections, especially in critically ill patients, makes the diffusion of antimicrobial drugs within tissues one of the cornerstones of chemotherapy. The knowledge of how antibacterial agents penetrate tissues may come from different sources: preclinical studies in animal models, phase I–III clinical trials and post-registration studies. However, the particular physiopathology of critically ill patients may significantly alter drug pharmacokinetics. Indeed, changes in interstitial volumes (the third space) and/or in glomerular filtration ratio may influence the achievement of bactericidal concentrations in peripheral compartments, while inflammation can alter the systemic distribution of some drugs. On the contrary, other antibacterial agents may reach high and effective concentrations thanks to the increased tissue accumulation of macrophages and neutrophils. Therefore, the present review explores the tissue distribution of beta-lactams and other antimicrobials acting on the cell wall and cytoplasmic membrane of bacteria in critically ill patients. A systematic search of articles was performed according to PRISMA guidelines, and tissue/plasma penetration ratios were collected. Results showed a highly variable passage of drugs within tissues, while large interindividual variability may represent a hurdle which must be overcome to achieve therapeutic concentrations in some compartments. To solve that issue, off-label dosing regimens could represent an effective solution in particular conditions.

Emerging Paradigms in the Prevention of Surgical Site Infection: The Patient Microbiome and Antimicrobial Resistance

This article summarizes new scientific evidence on the pathogenesis of surgical site infection, including the roles of the patient microbiome and antimicrobial resistance, and reviews changes in guidelines and clinical practices for prevention.