Machine Learning for Antibiotic Resistance Prediction: A Prototype Using Off-the-Shelf Techniques and Entry-Level Data to Guide Empiric Antimicrobial Therapy.

OBJECTIVE: In the era of increasing antimicrobial resistance, the need for early identification and prompt treatment of multi-drug-resistant infections is crucial for achieving favorable outcomes in critically ill patients. As traditional microbiological susceptibility testing requires at least 24 hours, automated machine learning (AutoML) techniques could be used as clinical decision support tools to predict antimicrobial resistance and select appropriate empirical antibiotic treatment. METHODS: An antimicrobial susceptibility dataset of 11,496 instances from 499 patients admitted to the internal medicine wards of a public hospital in Greece was processed by using Microsoft Azure AutoML to evaluate antibiotic susceptibility predictions using patients' simple demographic characteristics, as well as previous antibiotic susceptibility testing, without any concomitant clinical data. Furthermore, the balanced dataset was also processed using the same procedure. The datasets contained the attributes of sex, age, sample type, Gram stain, 44 antimicrobial substances, and the antibiotic susceptibility results. RESULTS: The stack ensemble technique achieved the best results in the original and balanced dataset with an area under the curve-weighted metric of 0.822 and 0.850, respectively. CONCLUSIONS: Implementation of AutoML for antimicrobial susceptibility data can provide clinicians useful information regarding possible antibiotic resistance and aid them in selecting appropriate empirical antibiotic therapy by taking into consideration the local antimicrobial resistance ecosystem.

Inactivation of mgrB gene regulator and resistance to colistin is becoming endemic in carbapenem-resistant Klebsiella pneumoniae in Greece: A nationwide study from 2014 to 2017.

INTRODUCTION: In Greece, the spread of carbapenem-resistant Enterobacteriaceae in humans has led to the reintroduction of colistin as a therapeutic agent. Unfortunately, colistin resistance with different mechanisms has emerged. The present work aims to determine the prevalence of carbapenem and colistin resistance and the corresponding mechanisms in Klebsiella pneumoniae clinical isolates from Greece. METHODS: From 2014 to 2017, 288 carbapenem-resistant K. pneumoniae clinical strains were gathered from a collection of 973 isolates from eight different hospitals in Greece. Antibiotic susceptibility testing was performed using three different methods. Screening of carbapenem and colistin resistance genes was conducted using polymerase chain reaction (PCR) amplification and sequencing. RESULTS: Among the 288 (29.6 %) carbapenem-resistant isolates, 213 (73.9%) were colistin-resistant (minimum inhibitory concentration [MIC] >2 mg/L). The KPC type was the most common carbapenemase gene (116; 40.3%), followed by VIM (41; 14.2%), NDM (33; 11.5%) and OXA-48 (22; 7.6%). Moreover, 44 (15.3%) strains co-produced two types of carbapenemases. No mcr genes were detected for colistin resistance but mutations in chromosomal genes were found. These included inactivation of the mgrB gene for 148 (69.5%) strains, including insertion sequences for 94 (44.1%), nonsense mutations for 4 (1.9%) and missense mutations for 24 (11.3%). Moreover, PCR amplification of mgrB gene was negative for 26 (12.2%) strains. Finally, 65 (30.5%) colistin-resistant strains exhibited a wild-type mgrB, the mechanisms of which remain to be elucidated. CONCLUSION: This study shows that K. pneumoniae clinical strains in Greece are resistant to both carbapenems and colistin and this is endemic and is likely chromosomally encoded.

Using Machine Learning Techniques to Aid Empirical Antibiotic Therapy Decisions in the Intensive Care Unit of a General Hospital in Greece.

Hospital-acquired infections, particularly in the critical care setting, have become increasingly common during the last decade, with Gram-negative bacterial infections presenting the highest incidence among them. Multi-drug-resistant (MDR) Gram-negative infections are associated with high morbidity and mortality with significant direct and indirect costs resulting from long hospitalization due to antibiotic failure. Time is critical to identifying bacteria and their resistance to antibiotics due to the critical health status of patients in the intensive care unit (ICU). As common antibiotic resistance tests require more than 24 h after the sample is collected to determine sensitivity in specific antibiotics, we suggest applying machine learning (ML) techniques to assist the clinician in determining whether bacteria are resistant to individual antimicrobials by knowing only a sample's Gram stain, site of infection, and patient demographics. In our single center study, we compared the performance of eight machine learning algorithms to assess antibiotic susceptibility predictions. The demographic characteristics of the patients are considered for this study, as well as data from cultures and susceptibility testing. Applying machine learning algorithms to patient antimicrobial susceptibility data, readily available, solely from the Microbiology Laboratory without any of the patient's clinical data, even in resource-limited hospital settings, can provide informative antibiotic susceptibility predictions to aid clinicians in selecting appropriate empirical antibiotic therapy. These strategies, when used as a decision support tool, have the potential to improve empiric therapy selection and reduce the antimicrobial resistance burden.

A 2-Year Single-Centre Audit on Antibiotic Resistance of Pseudomonas aeruginosa, Acinetobacter baumannii and Klebsiella pneumoniae Strains from an Intensive Care Unit and Other Wards in a General Public Hospital in Greece.

Hospital-acquired infections, particularly in the critical care setting, are becoming increasingly common during the last decade, with Gram-negative bacterial infections presenting the highest incidence among them. Multi-drug-resistant (MDR) Gram-negative infections are associated with high morbidity and mortality, with significant direct and indirect costs resulting from long hospitalization due to antibiotic failure. As treatment options become limited, antimicrobial stewardship programs aim to optimize the appropriate use of currently available antimicrobial agents and decrease hospital costs. Pseudomonas aeruginosa, Acinetobacter baumannii and Klebsiella pneumoniae are the most common resistant bacteria encountered in intensive care units (ICUs) and other wards. To establish preventive measures, it is important to know the prevalence of Gram-negative isolated bacteria and antibiotic resistance profiles in each ward separately, compared with ICUs. In our single centre study, we compared the resistance levels per antibiotic of P. aeruginosa, A. baumannii and K.pneumoniae clinical strains between the ICU and other facilities during a 2-year period in one of the largest public tertiary hospitals in Greece. The analysis revealed a statistically significant higher antibiotic resistance of the three bacteria in the ICU isolates compared with those from other wards. ICU strains of P. aeruginosa presented the highest resistance rates to gentamycin (57.97%) and cefepime (56.67%), followed by fluoroquinolones (55.11%) and carbapenems (55.02%), while a sensitivity rate of 97.41% was reported to colistin. A high resistance rate of over 80% of A. baumannii isolates to most classes of antibiotics was identified in both the ICU environment and regular wards, with the lowest resistance rates reported to colistin (53.37% in ICU versus an average value of 31.40% in the wards). Statistically significant higher levels of resistance to most antibiotics were noted in ICU isolates of K. pneumoniae compared with non-ICU isolates, with the highest difference-up to 48.86%-reported to carbapenems. The maximum overall antibiotic resistance in our ICU was reported for Acinetobacter spp. (93.00%), followed by Klebsiella spp. (72.30%) and Pseudomonas spp. (49.03%).

A prospective study on bacterial and atypical etiology of acute exacerbation in chronic obstructive pulmonary disease.

AIM: The bacterial and atypical etiology of acute exacerbations of chronic obstructive pulmonary disease was investigated and the diagnostic techniques used were compared among 92 hospitalized patients. MATERIALS & METHODS: Sputum specimens were investigated using culture and PCR, serological status evaluation was performed and the inflammatory profile was associated with the microbiological results. RESULTS & CONCLUSION: The majority of the patients (65.2%) had very severe airway obstruction. The most common bacteria were Haemophilus influenzae and Pseudomonas aeruginosa (23.9 and 14.1%, respectively). Acinetobacter baumannii- and P. aeruginosa-positive cultures were associated with prolonged hospitalization and severe airway obstruction (p = 0.03 and 0.031, respectively). Chlamydia pneumoniae or Mycoplasma pneumoniae infection was diagnosed in four and two patients, respectively. Discrepant results were detected between PCR and serology, especially regarding C. pneumoniae.

Health benefits of probiotics: a review.

Probiotic bacteria have become increasingly popular during the last two decades as a result of the continuously expanding scientific evidence pointing to their beneficial effects on human health. As a result they have been applied as various products with the food industry having been very active in studying and promoting them. Within this market the probiotics have been incorporated in various products, mainly fermented dairy foods. In light of this ongoing trend and despite the strong scientific evidence associating these microorganisms to various health benefits, further research is needed in order to establish them and evaluate their safety as well as their nutritional aspects. The purpose of this paper is to review the current documentation on the concept and the possible beneficial properties of probiotic bacteria in the literature, focusing on those available in food.

Rare Case of Strongyloides stercoralis Hyperinfection in a Greek Patient with Chronic Eosinophilia.

Strongyloidiasis is a disease characterized by a diverse spectrum of unspecific manifestations that complicate its diagnosis. Although, the course of its chronic form is usually benign, in cases of immunosuppression, iatrogenic or not, it can evolve to a hyperifection syndrome with even fatal complications. Herein, we report a case of Strongyloides stercoralis hyperinfection in a Greek patient receiving corticosteroid treatment for chronic eosinophilia and angioedema. The case represents an extremely rare case of autochthonous strongyloidiasis in Greece and underlines the importance of the early diagnosis of the disease's uncomplicated forms in order to prevent its severe sequelae.