Acquisition of extended-spectrum cephalosporin-resistant Gram-negative bacteria: epidemiology and risk factors in a 6-year cohort of 507 severe trauma patients

The Prevalence of Alert Pathogens and Microbial Resistance Mechanisms: A Three-Year Retrospective Study in a General Hospital in Poland

The development of antibiotic resistance mechanisms hinders the treatment process. So far, there is limited data on the problem of bacterial resistance in hospitals in Central and Eastern Europe. Therefore, this study aimed to assess the prevalence of resistance mechanisms and alert pathogens based on reports regarding cultures of samples collected from general hospital patients in Poland in the period 2019-2021. This study examined the prevalence of resistance mechanisms and alert pathogens and the structure of microorganisms, including the type of diagnostic material in the hospital department. The frequency of occurrence and the trends were analysed based on Cochran's Q-test, relative change and the average annual rate of change (AARC). Of all 14,471 cultures, 3875 were positive for bacteria, and 737 were characterised by resistance mechanisms (19.0%). Alert pathogens were identified in 983 cases (24.6%), including pathogens isolated from blood samples. The most commonlyisolated bacteria were Escherichia coli (>20% of positive cultures), Enterococcus faecalis (7%), and Klebsiella pneumoniae (6%) increasing over time; Staphylococcus aureus (13%) was also found, but its proportion was decreasing over time. Extended-spectrum β-lactamase (ESBL) was the most frequent resistance mechanism in Internal Medicine (IM) (p < 0.001) and the Intensive Care Unit (ICU) (p < 0.01), as well as in ICU-COVID; this increased over the study period (AARC ↑34.9%). Methicillin-resistant Staphylococcus aureus (MRSA) (AARC ↓50.82%) and AmpC beta-lactamase (AARC ↓24.77%) prevalence fell over time. Also, the number of alert pathogens was dominant in the IM (p < 0.01) and ICU (p < 0.001). The most common alert pathogen was ESBL-positive E. coli. Gram-negative rods constitute a significant epidemiological problem in hospitals, especially the growing trend of ESBL in IM and ICU, which calls for increased control of sanitary procedures.

DeepBiomarker2: Prediction of alcohol and substance use disorder risk in post-traumatic stress disorder patients using electronic medical records and multiple social determinants of health

Introduction

Prediction of high-risk events amongst patients with mental disorders is critical for personalized interventions. In our previous study, we developed a deep learning-based model, DeepBiomarker by utilizing electronic medical records (EMR) to predict the outcomes of patients with suicide-related events in post-traumatic stress disorder (PTSD) patients.

Methods

We improved our deep learning model to develop DeepBiomarker2 through data integration of multimodal information: lab tests, medication use, diagnosis, and social determinants of health (SDoH) parameters (both individual and neighborhood level) from EMR data for outcome prediction. We further refined our contribution analysis for identifying key factors. We applied DeepBiomarker2 to analyze EMR data of 38,807 patients from University of Pittsburgh Medical Center diagnosed with PTSD to determine their risk of developing alcohol and substance use disorder (ASUD).

Results

DeepBiomarker2 predicted whether a PTSD patient will have a diagnosis of ASUD within the following 3 months with a c-statistic (receiver operating characteristic AUC) of 0·93. We used contribution analysis technology to identify key lab tests, medication use and diagnosis for ASUD prediction. These identified factors imply that the regulation of the energy metabolism, blood circulation, inflammation, and microbiome is involved in shaping the pathophysiological pathways promoting ASUD risks in PTSD patients. Our study found protective medications such as oxybutynin, magnesium oxide, clindamycin, cetirizine, montelukast and venlafaxine all have a potential to reduce risk of ASUDs.

Discussion

DeepBiomarker2 can predict ASUD risk with high accuracy and can further identify potential risk factors along with medications with beneficial effects. We believe that our approach will help in personalized interventions of PTSD for a variety of clinical scenarios.